Beenrol ECOLOGY

A Foundation for Sustainable Forest Management and

Environmental Ethics in Forestry
fei) bd1 OWN
‘
 ‘
tha
iePR a ee
 . Fi

c (
= ;

a] <
e by . ¥
: r r

_ : > i ;
. *

| q -

we _- reg *
° <

\
A
.

, « w

’ P 2
,— ’
=
“y ‘ “
‘ * " r a
i "
, . , i ‘ ‘
_—* .
FOREST ECOLOGY
A Foundation for Sustainable Forest Management
and Environmental Ethics in Forestry

Third Edition

J.P. KIMMINS
The University of British Columbia

Prentice
Hall
ee
Te PP ee

Upper Saddle River, New Jersey 07458

 Dedication

This book is dedicated to my family—Ann, Mark, Shaun, Katherine, Henry, and Charlotte—particular-
ly Ann, who has patiently borne the many hours that we did not spend together when I was working on
the book, and the memory of Adrian Weber, PhD, graduate student and friend, who was making a major
contribution to our science when his untimely death robbed his family and friends and forest ecology of
a future shining light.

Library in Congress Cataloging-in-Publication Data

Kimmens,J. P.
Forest ecology: a foundation for sustainable forest management and environmental ethics
in forestry/J.P. Kimmens.—3rd ed.
.cm.
Includes bibliographical references and index.
ISBN 0-13-066258-5
1. Forest ecology. I. Title.
QH541.5.F6K55 2004
577.3—de21 2003054862
Assistant Editor: Colleen Lee
Executive Editor: Teresa Chung
Editor-in-Chief, Science: John Challice
Editorial Assistant: Mary Kuhl
Executive Managing Editor: Kathleen Schiaparelli
Production Editor: Kevin Bradley
Art Director: Jayne Conte
Cover Designer: Bruce Kenselaar
Manufacturing Manager: Trudy Pisciotti
Manufacturing Buyer: Lynda Castillo
© 2004, 1997 Pearson Education, Inc.
Pearson Prentice Hall
Pearson Education, Inc.
Upper Saddle River, NJ 07458
Earlier edition, © 1987, Macmillan Publishing Company, a division of Macmillan, Inc.
All rights reserved. No part of this book may be reproduced in any form or by any means, without permission in writing from
the publisher.
Printed in the United States of America

Pearson Education Ltd., London

Pearson Education Australia Pty. Ltd., Sydney
Pearson Education Singapore, Pte. Ltd.
Pearson Education North Asia Ltd., Hong Kong
Pearson Education Canada, Inc., Toronto
Pearson Educacion de Mexico, S.A. de C.V.
Pearson Education—Japan, Tokyo
Pearson Education Malaysia, Pte. Ltd.
Pearson Education, Inc., Upper Saddle River, New Jersey

Prentice
Hall
oS eee 10°93 726. 4a
ISBN O-13-0bb258-5
 Brief Contents

Preface to the Third Edition xl

Introduction XV
About the Author XVIli

PART I People and Forests: Why the Science of

Forest Ecology Developed 1
Sustainability of Forest Ecosystems: The Problem of
Human Population Growth 3
Development of Forestry and Forest Ecology 11

Ecosystem as the Basic Unit of Forest Ecology 23

Ecology and the Ecosystem Concept 25
Production Ecology: The Transfer and Storage of Energy in Ecosystems 35
Biogeochemistry: Cycling of Nutrients in Ecosystems 172

PART UI Ecological Diversity and the Ecological Stage: Physical Determinants of

the Ecological Play 139
Ecosystem Classification: The Ecological Foundation for Sustainable Forest
Management 141
Ecological Role of Solar Radiation 169
Temperature as an Ecological Factor 201
Wind: Ecological Effects of Atmospheric Movements 236
Water: The Material That Makes Life Possible 255
Soil: The Least Renewable Physical Component of the Ecosystem 284
Fire: A Pervasive and Powerful Environmental Factor 329
Patterns of Biotic Communities Along Environmental Gradients 349

Biological Diversity: Population, Community, and Genetic Determinants

of the Ecological Play 365
Population Ecology: Study of the Abundance and Dynamics of Species
Populations 367
Community Ecology 405
Genetic and Evolutionary Aspects of Ecosystems:
Adaptation and Evolution 441
il
IV BRIEF CONTENTS

PART V Temporal Diversity: The Ecological Play in Which All Aspects of

Ecosystems Change Over Time 461
17 Ecological Succession: Processes of Change in Ecosystems 463
18 Understanding and Emulating Natural Forest Disturbance: A Template for the
Management of Succession and Temporal Diversity? 516

PART VI Spatial Diversity: The Landscape Revisited 531

19 Ecosystem Management and Landscape Ecology: The Ultimate Focus In Forest
Ecology 533

PART VII Application of Ecological Knowledge in the Management of Forest

Ecosystems 543
20 Environmental Issues in Forestry: The Role of Ecology in the Management of the
Ecological Play 545
21 Models and Their Role in Ecology and Resource Management 565
22 Sustainability and Renewability of Natural Resources, and Implications for Forest
Management 597
Zo Ecology and Environmental Ethics in Forestry 609

Glossary G-1
Appendix A: Scientific and Common Names ofPlants AP-1
Appendix B: Scientific and Common Names of Animals AP-5
References Cited R-1
Index I-1
 Contents

PREFACE TO THE THIRD EDITION xi PART II

INTRODUCTION xv
ABOUT THE AUTHOR xviii Ecosystem as the Basic Unit of Forest
Ecology 23
PgkAD Ret ein CHAPTER 3
Ecology and the Ecosystem Concept 25
People and Forests: Why the Science of
Forest Ecology Developed 1 3.1 The Science of Ecology and Its Historical
Development 25
CHAPTER 1 3.2 Subdivisions of Ecology and the Concept Levels
Sustainability of Forest Ecosystems: The of Biological Organization 27
Problem of Human Population Growth 3 3.3. The Ecosystem Concept 28
3.4 The Concept Levels of Biological Integration 29
ig Introduction 3
3.5 Ecosystem Change: A Key Attribute 31
12 Past and Future Growth in Human Numbers 4
ES) Implications of Population Growth for Global Summary 34
‘Take-Home Message 34
Resources 5
Study Questions 34
Summary 9
‘Take-Home Message 10
Study Questions 10 CHAPTER 4
Production Ecology: The Transfer and
CHAPTER 2 Storage of Energy in Ecosystems 35
Development of Forestry and Forest Ecology 11 4.1 Introduction 35
Jan| Introduction 11 4.2 Sources of Energy for Living Organisms 36
aed Human Evolutionary Dependence on 4.3. Trophic Chains and Webs, Ecological Pyramids,
Forests 11 and Energy Flow Diagrams 37
ce History of Human Impacts on Forests 13 4.4 ‘Terminology Used in Production Ecology 42
Zt Development of Forestry 14 4.5 Production Ecology at the Primary Producer
Ecology, Environmentalism, or Green Level 43
ao
Religion: Which Will Guide the Future 4.6 Production Ecology at Consumer Trophic
Evolution of Forestry? 16 Mevelsm 5.t
2.6 Forest Ecology: A Practical Means of Dealing 4.7 Detritus Trophic Chains 54
with the Problems of Complexity 17 4.8 Leaf Area, Foliage Production Efficiency, and
Growth of Forests 57
Summary 21
Take-Home Message 21 4.9 Carbon Allocation and Carbon Storage in
Study Questions 21 Forests 60
vi CONTENTS

4.10 Energy Benefit/Cost Relationships of Forest CHAPTER 6

Production 64 Ecosystem Classification: The Ecological Foun-
4.11 Comparison of the Energy Requirements of dation for Sustainable Forest Management 141
Wood and Alternative Materials 64
6.1 Introduction 141
4.12 Effects of Forest Management on Energy in the
Forest Ecosystem 66 6.2 Climatic Classification 143
6.3 Landform, or Physiographic, Classification 146
Summary 69
Take-Home Message 70 6.4. Vegetation Classification 147
Study Questions 71 6.5 Ecosystem Classification 158

CHAPTER 5 Summary 166

Take-Home Message 167
Biogeochemistry: Cycling of Nutrients in
Study Questions 168
Ecosystems 72
5.1 Introduction 72 GHAR TERE
Bile Plant Nutrition: Energy and Nutrients at the Ecological Role of Solar Radiation 169
Primary Producer Trophic Level 74 7.1 Introduction 169
3.3 Animal Nutrition: Energy and Nutrients at 7.2 Physical Nature of Solar Radiation and its
Consumer Trophic Levels 79 Variations in Time and Space 169
The Geochemical Cycle: Nutrient Cycling
7.3. Ecological Effects of Variations in the Spectral
Between Ecosystems 81
Quality of Solar Radiation 174
The Biogeochemical Cycle: Nutrient Cycling
7.4 Ecological Effects of Variations in the Intensity
Within an Ecosystem 86
of Solar Radiation 175
The Biochemical Cycle: Nutrient Cycling
7.5 Ecological Effects of Temporal Variations in
Within Plants 105
Soiar Radiation 189
Nitrogen Cycling: The Key to Productivity in
7.6 Importance of the Light Factor in Forestry 196
Many Forests 110
Biogeochemical Efficiency of Forest Summary 198
Ecosystems 113 ‘Take-Home Message 199
Study Questions 200
Nutrient Cycling in Tropical Forest
Ecosystems 115
CHAPTER 8
QO Effects of Humans on Biogeochemistry and the
‘Temperature as an Ecological Factor 201
SI — oa

Cycling of Toxic Substances 120

Effects of Forest Management on Forest 8.1 Introduction 201
Biogeochemistry 124 8.2 Geographical and Temporal Variations In
Variations in Contributions of the Three ‘Temperature 202
Main Cycles Over the “Stand Cycle” and 8.3 General Ecological Concepts Concerning
Succession 134 Temperature 209
3 Long-Term Declines in Forest Productivity— 8.4 Effects of Temperature on Plants 211
Are They Fertility Related? 135 8.5 Animal Adaptations to Temperature 220
Summary 137 8.6 Role of Temperature in Limiting the
Take-Home Message 138
Distribution of Organisms 226
Study Questions 138
8.7 Importance of the Temperature Factor in
Forestry 228
|
aeaes Bee Fg 8.8 Human Effects on the World Radiation Budget
and Global Temperatures 232
Ecological Diversity and the Ecological
Summary 234
Stage: Physical Determinants of the Take-Home Message 234
Ecological Play 139 Study Questions 235
 CONTENTS vil

CHAPTER 9 11.8 Soil Development 315

Wind: Ecological Effects of Atmospheric Move- 11.9 Ecological Significance of Soil 321
ments 236 11.10 Importance of Soils for Forest
9.1 Introduction 236 Management 323
9.2 ‘Temporal and Spatial Variations in Wind 236 Summary 327
‘Take-Home Message 328
9.3 Effects of Wind on Vegetation 239
Study Questions 328
9.4 Effects of Wind on Animals 246
9.5 Effects of Wind on Ecosystems 248 CHAPTER 12
9.6 Effects of Vegetation on Wind 249 Fire: A Pervasive and Powerful Environmental
9.7 Significance of Wind for Forestry 251 Factor 329

Summary 253 12.1 Introduction 329

‘Take-Home Message 253 12.2 Types and Occurrence of Fires 330
Study Questions 254 12.3 Effects of Fire on Soil 331
12.4 Effects of Fire on Plants 338
CHAPTER 10
12.5 Effects of Fire on Animals 341
Water: The Material That Makes
12.6 Effects of Fire on Ecosystems and Ecosystem
Life Possible 255
Processes =343
10.1 Introduction 255 12.7. Effects of Fire Exclusion 346
10.2 The Water Cycle 256 12.8 Fire and Forest Management 347
10.3 Availability of Water to Plants and Animals 269 Summary 347
10.4 Adaptations of Plants to Excess or Deficit of ‘Take-Home Message 347
Water 270 Study Questions 348
10.5 Adaptations of Animals to Excess or Deficit of
Water 275 CHAPTER 13
10.6 Effects of Moisture on Plant Distribution and
Patterns of Biotic Communities Along Environ-
Production 276 mental Gradients 349
10.7 Interaction Between Moisture and 13.1 Introduction 349
Nutrients 278 13.2 Possible Patterns of Species Distribution 349
10.8 Competition for Water Between Trees and 13.3. Major Forest Ecotones: The Interface Between
Minor Vegetation 280 Forests and Communities of Different
10.9 Forestry and the Water Factor 281 Physiognomy 354
Summary 282 13.4 Forestry and Environmental Gradients—The
‘Take-Home Message 283 Question of Site 363
Study Questions 283 Summary 364
‘Take-Home Message 364
CHAPTER 11 Study Questions 364
Soil: The Least Renewable Physical Component
of the Ecosystem 284 PAS RSS Teel Vi
Ete Introduction 284
Biological Diversity: Population, Community,
Mey! Physical Properties of Soil 285
and Genetic Determinants of the Ecological
Lies Soil Water 290
Play 365
LG: Chemical Properties of Soil 293
115 Soil Organic Matter 298 CHAPTER 14
L156 Soil Biology 299 Population Ecology: Study of the Abundance
ARig! Soil Fertility 310 and Dynamics of Species Populations 367
Vill CONTENTS

14.1 Introduction 367 Summary 458

Take-Home Message 459
14.2 Population Growth and Demographic Study Questions 459
Characteristics of Populations 372
14.3 Major Determinants of Population Size 380
14.4 Theories About the Natural Regulation of Pi AvR way:
Population Size 385
14.5 Cyclical Population Fluctuations 389 Temporal Diversity: The Ecological Play in
14.6 Role of Predation in Population Which All Aspects of Ecosystems Change
Regulation 390 Over Time 461
14.7 Plant Population Ecology 394
CHAP TE Rely
14.8 Importance of Population Ecology in Forest
Management 402 Ecological Succession: Processes of Change in
Summary 403
Ecosystems 463
‘Take-Home Message 403 17.1 Introduction 463
Study Questions 404 17.2. Terminology and Concepts 464
17.3 Classical Concepts and Models of
CHAPTER 15 Succession 467
Community Ecology 405 17.4 Review of Recent Theories and Models of
15.1 Introduction 405 Succession 471
ope. Structure and Growth Forms of Plant 17.5 Mechanisms of Successional Change 476
Communities 407 17.6 Rates of Successional Change 488
Interactions Between Species in a 17.7. The Three Major Types of Sere 491
Community 411 17.8 Linear and Cyclical Succession: The Problem
The Competitive Exclusion Principle and the with the Concept of Climax 495
Ecological Niche Concept 424 17.9 “Vital Attributes” Model of Succession 501
Biological Diversity 429 17.10 Changes in Ecosystem Function During
Community Ecology and Forest Succession 502
Management 438 17.11 Relationship Among Ecosystem Stability,
Summary 439 Biodiversity, and Successional Stage 507
‘Take-Home Message 439 17.12 Succession in Animal Communities 510
Study Questions 440
17.13 Effect of Forest Management on Succession 511
Summary 514
CHAPTER 16
Take-Home Message 515
Genetic and Evolutionary Aspects of Ecosys- Study Questions 515
tems: Adaptation and Evolution 441
16.1 Introduction 441 CHARTER LS
16.2 Genetic Variation in Populations: One of the
Understanding and Emulating Natural Forest
Important Measures of Biological Diversity and Disturbance: A Template for the Management of
the Raw Material for Natural Selection 442 Succession and Temporal Diversity? 516
16.3 Examples of Selection, Adaptation, and 18.1 Introduction 516
Evolution 448
18.2 Paradigms for a New Approach to Forestry 517
16.4 Coevolution and the Biotic Environment 454 18.3 Definitions of Disturbance and ENFD 520
16.5 Evolutionary Time Scales 456
18.4 Management Strategies with Respect to
16.6 Implications of Genetic Variability, Evolution, Succession 521
and Adaptation for Forestry 456
18.5 Ecological Rotation Concept as a Framework for
16.7 Evolution of Ecosystems 458 ENFD 523
 CONTENTS 1x

18.6 Small-Scale or Gap Disturbance 528 21.1 Introduction 565

Summary 530 21.2 Limitations of Models 566
Take-Home Message 530 21.3 Why People Model and the Survivorship of
Study Questions 530 Models 567
21.4 ‘Types of Model: Computer Models Are Only
One of Many ‘Types of Model 568
PRAT REE Vel 21.5 Stages in the Development of a Computer
Simulation Model 571
Spatial Diversity: The Landscape 21.6 Application of Different Types of Model to the
Revisited 531 Study of Site Nutrient Depletion Through
Harvesting of Forest Biomass 572
CHAPTER 19 21.7. The Three Major Approaches to Modeling in
Ecosystem Management and Landscape Forestry 577
Ecology: The Ultimate Focus in Forest 21.8 Examples of Hybrid Simulation Models 580
Ecology 533 21.9 Landscape Models 587
19.1 Introduction 533 21.10 How Complex Should a Model Be? Occam’s
19.2 Ecosystem Management 534 Razor and the Scientific Method 589
19.3 Landscape Ecology 536 21.11 Example of the Need for Sufficient Complexity
in Successional Models: The Two Edges of
19.4 Fragmentation, Connectivity, and Edges 537 Occam’s Razor 591
19.5 Patch Size and Event Size Frequency
Distributions 539 Summary 594
‘Take-Home Message 595
19.6 ‘Tools in Landscape Ecology: Landscape Models, Study Questions 595
Remote Sensing, and Gis 541
Summary 541 CHAPTER 22
‘Take-Home Message 541 Sustainability and Renewability of Natural
Study Questions 542
Resources, and Implications for Forest
Management 597
POAT R21 6V TSI 22.1 Introduction 597

Application of Ecological Knowledge in the 22.2 Renewability of Resources 597

Management of Forest Ecosystems 543 2 3 The Nature of a Material Resource 599
4 How Does Our Use of Material Resources Affect
CHAPTER 20 Their Renewability? 601
Environmental Issues in Forestry: The Role of 22.5 Timber Mining Versus Sustained Yield
Ecology in the Management of the Ecological Management Versus Sustainable Ecosystem
Management 601
Playas
22.6 Sustainability and the Concept of Ecological
20.1 Introduction 545 Rotation 602
20.2 Environmental Issues in Forestry 545 22.7 Nonrenewable Aspects of the Forest
20.3 Forest and Ecosystem Health 547 Ecosystem 605
20.4 Ecosystem Integrity 550 Summary 607
20.5 Role of Forests and Forestry in Global Carbon ‘Take-Home Message 607
Cycles and Climate Change 553 Study Questions 607

CHAPTER 21 CHAPTER 23
Models and Their Role in Ecology and Ecology and Environmental Ethics in
Resource Management 565 Forestry 609
x CONTENTS

GLOSSARY G-I APPENDIX B: SCIENTIFIC AND COMMON

APPENDIX A: SCIENTIFIC AND COMMON NAMES OF ANIMALS AP-5
NAMES OF PLANTS AP-1 REFERENCES CITED Ry
INDEX I-1
 Preface to the
Third Edition

alfway through the life of the second edition, the sires of today’s generation against the needs and de-
human population passed the 6 billion mark. sires that we anticipate for our grandchildren, their
The world’s population has more than doubled since I grandchildren, and generations beyond. Forestry is
took my first university-level ecology course: an in- defined as the art (skill), practice, science, and business
crease of 3.2 billion people. It is expected to increase of managing forest stands and forested landscapes to
by another 3 to 5 billion in the time it takes to grow a sustain the balance of values and environmental ser-
tree crop in many sustainably managed, multivalue, vices desired by society. This definition requires that
temperate or northern forests: 60 to 90 years, or ap- forest policies and practices change as society changes
proximately the life expectancy of my generation’s the balance of values that it desires, and responding to
grandchildren. The good news for the world’s forests this change is the first responsibility of forestry. How-
is that the rate of human population growth has ever, a second and equal responsibility is to reject cur-
slowed and may stabilize or even start to decline with- rent practices and oppose suggested changes that are
in this century. not consistent with the ecology and sociology of the
Human pressure on the world’s forests has fol- new desired balance of values and that will not provide
lowed—but lagged somewhat behind—population our grandchildren and theirs with the range of values
growth. The increasing replacement of forest ex- that we think they will want and need.
ploitation by sustainable forest management and the Fulfilling these two key stewardship responsibili-
combined results of forest protection, product substi- ties of forestry requires that this profession have a
tution, technology, increasing public concern, and the sound understanding of the ecology of the values that
recycling of wood products have slowed the rate of in- are to be sustained. Although forestry is first and fore-
crease in human impacts on forests to less than the most about people and their needs, preferences, and
increase in human numbers. However, as people con- desires and only second about ecology, biodiversity,
tinue to increase the range of values that they desire and other currently politically correct topics, without
from forests, as the slowing of population growth a biological and ecological foundation, it is very un-
leads to increased wealth, and as it is increasingly rec- likely that either the present or the future generations
ognized that wood from sustainably managed forests is will have access to the resource values, nonresource
a socially and environmentally “green” product, I ex- values, and environmental services that they will want.
pect that total human demand on forests for wood and This is the rationale for the science of forest ecology.
nonwood values will continue to rise even after popu- Our science can certainly exist outside this utilitarian
lation levels have stabilized. focus, but the prospect of another 3 to 5 billion hu-
Forestry faces an ethical dilemma: how to balance mans in the time it takes for most trees that we plant
the changing social and environmental needs and de- today at temperate and northern latitudes to reach

Xl
xi PREACE TO THE THIRD EDITION

economic maturity (or provide the other values of- settings or stages, and a particular setting or stage lim-
fered by trees of that age) requires that forest ecology its the plays that can be performed on it. Particular
also serve society’s needs for an ecological foundation biotic communities and temporal sequences of com-
for a sustainable relationship between humans and munities can occur only in certain climatic, topo-
forests. graphic, and soil conditions. Particular physical
Since the first edition of the book (1987), great environments limit the species and species sequences
progress has been made in recognizing the importance that can occur there, and there is generally a broad
of diversity: cultural diversity, economic diversity, bio- similarity in ecosystem development over time after
logical diversity, and diversity in the way that we man- successive disturbances of a particular type and severi-
age forests. Increasingly, forestry around the world is ty in a particular type of ecosystem.
recognizing the importance of respecting the spatial A play has a story line that describes events and re-
diversity of forests. Forests differ in different climates, lationships and how they change over time. Like a
on different geologies and soils, and with different bi- play, ecosystems do not exist in a single, fixed con-
ological circumstances. Diversity in management dition. There is periodic disturbance that initiates an
practice from place to place, from ecosystem type to internal sequence of changes in ecosystem characteris-
ecosystem type, is required if we are to respect this tics. During this sequence of changes, there will be
spatial diversity. Recognition of the equal importance different species and different processes and different
of the temporal diversity of forests has developed rates of particular processes, just as in a multiact play
more slowly. Most people dislike change, and there is there are different groups of actors on stage acting out
widespread antipathy toward the visual, spiritual, and different parts of the story in the different phases
other changes that accompany natural or human- (acts) of the play. Some actors appear throughout the
caused disturbance in forests. It is one of the duties of play; some are specific to a particular act. During
ecologists and foresters to make available to the inter- ecosystem change, some species may persist through-
ested public sufficient knowledge about the processes, out, but many appear only at a particular part of the
products, and ecological role of natural and human- “play.” However, whereas the story line of a play is re-
induced change in forests that they can comprehend peated relatively unchanged every time the play is per-
and respond appropriately toward forest disturbance formed, ecosystems exhibit considerable variation
and change. between successive renditions of the overall “ecologi-
A metaphor for forest ecology and for forestry it- cal play.” Thus, differences between ecosystems and
self is ecological theater. There are ecological stages, or the metaphor of theater emerge as soon as one goes
settings, that define the ecological plays that can be from the broad concept to the details.
acted out. The ecological plays (ecosystem disturbance- Thankfully, society has become preoccupied with
recovery sequences) in turn define which ecological ac- biodiversity—nature’s insurance policy in the face of
tors (species) will appear on the ecological stage and in disturbance and environmental change. It is a rich
what order over time. Ecological stages, or settings, inheritance for humans and a legacy to be passed undi-
are defined by ecological diversity—the diversity of cli- minished to future generations. However, a preoccu-
mate, geology, topography, soil, and physical distur- pation with biodiversity before or outside the context
bance agents such as fire, wind, frost, flood, drought, of a consideration of ecological and temporal diversity
and landslide. Ecological plays are the sequences of is unlikely to be a successful foundation for biodiversi-
changes in the species composition, structure, func- ty conservation and sustainable forest management. It
tion, complexity, and internal organization that occur is analogous to being preoccupied with the actors and
in forests as a result of human and nonhuman distur- ignoring the play and the stage setting. As in theater,
bances. These sequences are a result of external dis- all three must be considered in assessing the ecology
turbance events and the internal ecosystem processes and designing the management of forest ecosystems.
of recovery toward predisturbance conditions or some This third edition of Forest Ecology is structured
new condition. around the ecological theater metaphor to emphasize
All metaphors are incomplete as a description of the necessity of a comprehensive approach to estab-
reality. There are many inconsistencies between the lishing an ecological foundation for sustainable forest
concept of theater and ecology, but there are also management. It deals with the spatial diversity of forest
some useful parallels. Different plays require different ecosystems in terms of ecological classification and the
 PREACE TO THE THIRD EDITION Xill

physical factors of climate (explored as solar radiation, the mechanism by which to spatially stratify ecological
temperature, precipitation, wind, and fire) and site and biological diversity. Chapters 7 to 10 then address
(explored in terms of soil and water). These factors the major physical factors of climate: light (solar radi-
collectively define the ecological stage, and their varia- ation), temperature, wind, and water. Chapter 11 ad-
tion from place to place constitutes ecological diversity. dresses the major nonclimatic, physical component of
The book then examines biological diversity in terms of site: the soil. Chapter 12 examines what is the major
the processes and products of individual species popu- physical disturbance factor in many of the world’s
lations and of communities of diverse species. It then forests: fire. These chapters collectively define ecolog-
considers temporal diversity by examining change over ical diversity and set the stage for the ecological play:
time in the components and processes of the forest the disturbance-recovery sequence of ecosystem
ecosystem: the ecological play. Ecological attributes of change. Chapter 13 examines the consequences of this
individual species—the ecological actors—are touched physical diversity for the diversity of biotic communi-
on throughout the book, but this subject area lies ties along environmental gradients.
more in the realm of autecology, ecophysiology, den- Having established in Part III the major features of
drology, behavioral ecology, and so on, a detailed the ecological stage and the physical causes of new eco-
treatment of which is beyond the scope of this book logical plays, the book turns in Part IV to the biotic de-
and of most courses in forest ecology. terminants thereof. Chapter 14 examines the biotic
The book commences in Part I, “People and processes that operate within or directly affect single-
Forests,” with a trio of contextual chapters. These deal species populations: population ecology. This chapter
with the historical relationship between humans and discusses many of the population-level processes that
forests (Chapter 1) and why and how forestry and the contribute to various measures of biological diversity
science of forest ecology developed (Chapter 2). Part and the biotic change over time that accompanies stand
II introduces the ecosystem as the basic unit of forest development and ecological succession. Chapter 15 in-
ecology (Chapter 3). Because the forest is an ecosys- vestigates the interactions between different species in
tem and all subsequent discussions of spatial and tem- biotic communities and how interactions contribute to
poral diversity need to be in the context of the forest the composition of communities and how this changes
as a system, Part II then examines the key functional over time. The genetics of the individual actors—their
processes that ultimately drive organic production and genetically controlled “ecological personalities”—is
change in ecosystems: the capture, storage, and flow of then briefly discussed because all biological processes
energy through and out of forest ecosystems, thereby and the response of all species to ecological diversity
maintaining ecosystem structure and order (Chapter are mediated through the genetics of the organisms
4), and the inputs of nutrients into, their circulation concerned (Chapter 16).
within, and the loss of nutrients from forest ecosys- Having identified many of the key components of
tems (Chapter 5). the ecological stage and some physical and biotic de-
Having established the ecosystem as the basic unit terminants of the ecological play, attention switches to
of forest ecology, with five key attributes—structure temporal diversity in Part V. Chapter 17 examines the
(including species composition), function, complexity, overall processes, pathways, and patterns of both ex-
interactions and linkage between its component parts ternally driven (allogenic and biogenic) ecosystem
and processes, and change over time—the book turns change and change mediated by processes internal to
in Part III to the spatial diversity of forests. In previ- the ecosystem (autogenic). Chapter 18 considers the
ous editions, classification of this diversity was delayed ecological role of disturbance and the current interest
to Chapter 16, but here it is addressed earlier because in emulating natural forest disturbance as a template
it deals with the ecological framework for biological for management of ecosystem succession and sustain-
and temporal diversity and with the spatial ecological ing temporal diversity.
foundation for sustainable forest management. Unless In Part VI, we return to the spatial diversity of
you can recognize the spatial diversity of forests, it is ecosystems and landscapes that was introduced in
difficult to apply your ecological knowledge to explain Chapter 6 as ecological classification was discussed.
their differences from place to place. It is also difficult Ecosystem management and landscape ecology have
to design and apply ecosystem-specific management. become the current focus and are the ultimate issues
Chapter 6 deals with ecological site classification as in forest ecology (Chapter 19) because this is the spa-
XIV PREACE TO THE THIRD EDITION

tial scale at which many of the issues of sustainability often persist because they are complex and people
and stewardship must be assessed. However, some offer simple solutions that do not solve them. Ecology
ecosystem values and many aspects of the ecology and other biophysical sciences on their own cannot
of individual species relate to spatial scales that are solve problems in forestry. Neither can social sciences
smaller than forest stands. Scaling up from our largely and considerations alone. There must be a marriage—
stand-level knowledge of forest ecology to forest land- a partnership—between these two great bodies of
scapes at various scales and scaling down to individual knowledge, guided by ethics, if we are to be successful
trees, groups of trees, and canopy gaps represents one in achieving a sustainable relationship between the
of the major challenges in our science today. human species and the world’s forests and broader en-
The final part of the book, Part VII, examines the vironment.
application of ecological knowledge in forestry. It be-
gins with Chapter 20 by examining some current is-
sues in the management and conservation of forests: Acknowledgments
“old growth”; aesthetics versus ecology; ecosystem
health, integrity, and stability and their relationship to Iam grateful to many people who contributed to the
measures of diversity; and the role of forests in carbon production of this edition: Maxine Horner, who did
cycles and climate change. Foresters have important so much to help organize the revision and provided
responsibilities for these issues and need to be able to excellent secretarial support whenever it was need-
harness the knowledge of ecosystem ecology pre- ed; Colleen Lee of Prentice Hall for her patience
sented in earlier chapters to design management and constant encouragement; Kevin Bradley, who
strategies that will achieve objectives with respect to helped me through the copyediting stage; and
these issues. Each of these topics is worthy of a sepa- Christine Chourmouzis for several new figures. My
rate book; only a brief introduction is provided here. associates John Worrall, Robert Guy, Michael
Chapter 21 then examines how such knowledge Feller, and Tom Sullivan reviewed specific chapters;
can be harnessed in quantitative forecasting and sce- and Daniel Mailly, Eliot McIntire, Robin Duches-
nario analysis tools: multivalue ecosystem manage- neau, and Marco Albani made helpful contributions.
ment simulators as well as single value predictive My thanks to the University of British Columbia for
models. ‘This chapter reviews types of models and making it possible to do the revision; to all those
modeling and their role in helping foresters and the who helped with previous editions; and to 35 years
public apply ecological knowledge to answer questions of my graduate students, from whom I learned more
of sustainability and stewardship. A major goal of than I taught them.
these modeling tools is to facilitate the definition of I extend a particular thanks to the reviewers of the
sustainability, renewability, and stewardship, topics second edition for their contributions to the improve-
that are discussed in Chapter 22. ment of the book. Time and space did not permit me
The book ends on a philosophical note, seeking to act on all their suggestions, which I regret. I was
wisdom from one of the great thinkers and writers very honored that they would give of their valuable
about issues of environmental ethics: Aldo Leopold time to do this: Jiquan Chen, Michigan Technological
and other, more recent commentators. There is an University; John Davis, Southern Vermont College;
ethical responsibility for stewarding the world’s forests Laura DeWald, Northern Arizona University; Allan
that rests with the practitioners who actually manage Drew, State University of New York College of Envi-
the forests, with governments who legislate and regu- ronmental Science and Forestry; Steve Hallgren, Ok-
late their management, and with public ~interest lahoma State University; Mark MacKenzie, Auburn
groups who lobby for change in the way forests are University; Michael Messina, Texas A&M University;
managed. Issues of intergenerational equity, the ethics Paul Mou, University of North Carolina, Greensboro;
of forest management, and stewardship are beyond the and Thomas Veblen, University of Colorado, Boulder.
scope of this book but require a sound foundation in I also thank Bert Weller for identifying errors in the
ecological sciences if they are to be solved. second edition.
A problem is an issue that does not get solved. An
issue that gets solved is not a problem. Problem issues J.P. KIMMINS
 Introduction
Sustainability: The Rationale for Ecology as the
Foundation for Forest Resource Management

he 1992 World Summit on the Environment in cause they all are talking about the sustainability of
Rio de Janeiro adopted sustainable development the same human-—forest system.
as the template for a new relationship between hu- It is unlikely that agreement will ever be reached
mans and the global environment. Although there are on a single operational definition of sustainable
energetic debates over the meaning—and even the de- forestry that is acceptable and applicable every-
sirability—of sustainable development, most people where, because the issue is so complex. It varies be-
seem to like the idea that resources should be developed tween different forest regions, between different
and used to meet the needs of the present generation in a forest ecosystem types within a region, and accord-
way that does not compromise the ability offuture genera- ing to the values that are to be sustained and the
tions to satisfy their needs.” temporal and spatial scales at which the assessment
Fine words, good intentions, and warm feelings is made. Any definition that will be true every-
about sustainability are easy to come by. Actually where is likely to be so general as to be trivial and
achieving our sustainability goals in forestry is another of little value in designing site-specific systems of
matter. There are many reasons for this difficulty: sustainable forest management.
Frequent lack of an adequate ecological classifica-
1. Lack of an accepted definition of what we mean by tion of forested landscapes that identifies different
sustainability of forest ecosystems and their man- ecological “stages” or settings, which in turn de-
agement. Numerous national and international fine the ecological “plays” that can occur in each
groups (e.g., the Forest Stewardship Council, the type of ecosystem. Without such a system, it is dif-
Canadian Standards Association, the US Sustain- ficult to devise and apply ecosystem-specific forest
able Forestry Initiative) claim to provide the “best” management practices that will sustain the forest
definition. These organizations provide a method values about which we are concerned in the many
by which to certify that forests are being managed different types of forest in the world. Without such
in a way that is consistent with their definitions. a classification, we cannot get away from the over-
However, to date, no agreement has been reached generalizations and oversimplified approaches that
on a common definition, although there are signs have been the hallmark of nonsustainable forestry
of convergence between the different approaches. in the past.
This convergence is to be expected, to the degree . Lack of an adequate understanding about forest
that is possible, when there are different philo- ecosystems and the ecological “plays” and ecologi-
sophical foundations for the different systems, be- cal “actors” therein by many of the politicians who
pass laws and the forest policy makers who estab-
*Adapted from WCED (1987) lish regulations about how to manage forests, by
XV
xvi INTRODUCTION

many of the public who lobby the politicians to ecology of our forests. Without such definitions and
change the laws and regulations, and by many of knowledge, we do not know how to practice sustain-
the foresters or forest engineers who are responsi- able forestry, and we cannot establish international
ble for putting these laws and regulations into controls on trade in forest products produced using
practice or who help to design new regulations for nonsustainable practices. We also cannot predict the
the politicians to impose. long-term consequences of our actions. Clearly, all re-
source managers and resource policy makers must ac-
4. Lack of ability to predict the long-term ecological
quire and use this knowledge if they are to satisfy the
consequences of alternative ways of managing a
public’s vision for the future of our forests.
particular forest ecosystem and the failure to use
ecologically based computer models to make eco- This book is intended merely as an introduction to
logically based rankings of the relative sustainabil- life-long learning about forests. Understanding forest
ity of various new but untested systems of ecosystems is always achieved best by observing reality
management. In the absence of this ability or use in the forest itself, not by reading textbooks or sitting
of appropriate ecologically based planning tools, it in a classroom, lecture hall, or laboratory. “If in doubt,
is difficult to be certain that we are indeed practic- ask a tree,” is a wise maxim. However, the complexity
ing sustainable forestry. It is equally difficult to of forests is so great that to go to the forest without a
judge whether the changes suggested in the name basic understanding of ecosystem components and
of stewardship and sustainability really will consti- processes is to invite confusion, frustration, and a case
tute an improvement and will achieve the changing of “not seeing the forests for the trees.” Herein lies the
objectives of management. justification for both classroom or laboratory prepara-
tion and learning aids such as textbooks and heuristic
Many of the other impediments to the achieve- ecosystem management models. The most effective
ment of sustainable forestry are beyond the scope of way to learn a foreign language is always to go to the
ecological sciences. These include inadequate forest country concerned and speak and read the language as
inventories, inappropriate forest tenure systems, lack a medium of communication. ‘Io do so without first
of adequate control of forest harvesting and manage- learning vocabulary and grammatical rules makes the
ment, and economic constraints. Forestry is primarily task much harder. The basics of forest ecology con-
about people—their preferences, needs, and desires. tained in this book are analogous to the vocabulary
Ecology is simply a tool with which to help design and grammatical rules of the “language of the forest.”
policies and practices that will satisfy society’s chang- It is the objective of this book to help you learn to be
ing balance of desired forest values. an “ecological detective” so that when you go to the
Development of a sustainable system of forestry is forest, you can read the clues, analyze the information,
analogous to a house and its foundation: the former is and reach appropriate conclusions and interpretations.
the system of forest management, and the latter is the The first ecological challenge in forestry is to rec-
ecological and related sciences on which the manage- ognize the spatial diversity in the ecological character-
ment is based. A good foundation does not guarantee a istics of forests. The second is to understand their
good house, but an otherwise good house structure that processes well enough to be able to forecast their fu-
is built on a weak and inadequate foundation will fail to ture development (temporal diversity) and response to
function acceptably over the long term and may even natural disturbance and management. Diversity is an
collapse. A sound ecological understanding does not important and popular concept today, but with diver-
guarantee sustainable forestry but is a necessary prereq- sity comes complexity, and with complexity comes dif-
uisite for it. Ecological science alone cannot ensure that ficulty in understanding and in making predictions of
our forests will be managed sustainably, but without a future states of complex systems. To face successfully
sound ecological foundation, it is unlikely that we can the ecological challenge, foresters must appreciate the
build a sustainable forest management system. ecological dimensions of diversity, not merely the bio-
The urgent need to be able to define, for any par- logical, social, and cultural dimensions.
ticular type of forest, the stand-level and landscape- The book is based on the metaphor of forest
level forest practices that will or will not sustain a ecosystems as “ecological theater.” This theater has
given set of ecosystem conditions or values is the ecological “stages,” or settings; ecological “plays”; and
major justification for a sound understanding of the ecological “actors.” After reviewing the historical rela-
 INTRODUCTION XVII

tionship between humans and forests and the resulting topical issues in forestry and to apply it in predictive
development of forestry (Part I), you will be intro- ecological models so that we can explore the long-
duced to the concept of ecosystem and its basic func- term consequences of different ways of modifying for-
tions of energy flow and nutrient cycling (Part ID). est policy and practice to address issues.
Part If focuses on the ecological “stage.” This is de- As noted in the preface but worth repeating is that
fined in terms of ecological diversity—the diversity of a problem is an issue that does not get solved quickly.
physical climatic and site factors and disturbance An issue that gets solved quickly is not a problem. The
agents that collectively define which species can exist reason that many problems persist is because they are
where, what disturbance regimes are to be expected, complex or are part of a complex system, and the solu-
and the potential range of ecosystem change after dis- tions offered are too simple.
turbance: the ecological “play.” Part IV examines bio- A fundamental tenet in science is the principle of
logical diversity, the population and community biotic parsimony—usually referred to as Occam’s razor. This
processes that determine the various different mea- says that explanations or hypotheses should not be
sures of biodiversity, and the biotic controls on the more complex than they need to be complexity should
“ecological play.” not be posited without necessity. The interpretation of
Part V describes the “play” itself—the disturbance this principle in science has often been “the simpler
factors that initiate a sequence of change and the the better.” However, in ecology, Occam’s razor has
biotic community changes that occur by internal two edges: as simple as possible, but as complex as neces-
ecosystem processes in response to the disturbance. It sary. Repeatedly in ecology, we find that simple theo-
also examines how we can use “natural” patterns of ries and explanations and solutions to problem based
change as a template for the design of management- thereon are ineffective because they do not deal ade-
induced change. quately with complexity. Science also seeks unifying
The single chapter in Part VI examines the emerg- theory and general principles. In ecology, experience
ing fields of ecosystem management and landscape teaches us that theories and principles that are based
ecology—potentially the most important aspects of on simple generalizations that are true everywhere are
the science of forest ecology because they synthesize generally trivial in terms of understanding, explaining,
ecological knowledge up to the spatial scales at which and managing the differences between forest ecosys-
many issues of stewardship and sustainability exist. tems in different places and the differences in any par-
However, this chapter also emphasizes the need to de- ticular ecosystem over time.
velop our understanding at a smaller spatial scale—to In forest ecology, it seems that there is only one
understand forest gaps, stand edges, and the small generalization that is always true:
scale habitat features of what make up the species rich- IT DEPENDS. The trick is to identify what “it”
ness measure of biological diversity. It emphasizes that depends on in a particular circumstance. It is my hope
landscapes consist of stands and that stands consist of that this book will help you to learn and organize the
individual trees, other plants, and other organisms. basic concepts about forests ecosystems—their spatial
The science of forest ecology must become much bet- and temporal diversity—so that you can accurately de-
ter at crossing these spatial scales. fine for every “it depends” just on what “it” depends.
Part VII explores how we can apply the knowledge Good luck in your quest to define the questions and
gained in Parts I to VI to achieve an understanding of then seek possible answers.
 About the Author

ollowing an undergraduate degree in Forestry

from the University of Wales and an M.Sc. in For-
est Entomology from the University of California at
Berkeley, James Peter Kimmins, generally known as
Hamish, received his PhD in Forest Ecology at Yale
University, focusing on the relationship between
ecosystem function and herbivore population dynam-
ics. Since 1969 he has been a Professor of Forest Ecol-
ogy at the University of British Columbia, where he
has presented courses on forest ecology, ecosystem
classification, ecological aspects of silviculture, envi-
ronmental issues in forestry, ecosystem function and
response to disturbance, and modelling forest ecosys-
tems to undergraduates, graduate students and profes-
sional foresters. He has developed and currently tutors
a UBC web-based Distance Education course in For-
est Ecology. The fifty-three graduate students he has
supervised over the past thirty-five years have studied
nearly all aspects of forest ecology from ecophysiology
to nutrient cycling and production ecology, to succes-
sion, landscape ecology and modelling.
For the past twenty-six years, Dr. Kimmins has
worked to develop ecologically-based forest ecosystem
management models, from the spatially-explicit, indi-
vidual tree stand model FORCEE, to the a-spatial assessment tools. He is currently a Senior Canada Re-
stand model FORECAST, to the spatial, local land- search Chair in Modelling the Sustainability of Forest
scape, complex disturbance patch model LLEMS, and Ecosystems, and is Director of the Forest Ecosystem
the spatial watershed ecosystem management model Management Modelling Group in the Department of
POSSIBLE FOREST FUTURES. These models Forest Sciences. Dr. Kimmins is a member of
range from a high school education forest manage- UNESCO's World Commission on the Ethics of Sci-
ment game (FORTOON), to decision support and re- ence and Technology (COMEST). and serves on
search tools, to scenario analysis and value tradeoff many science advisory boards.

XVili
 People and Forests
Why the Science of Forest
Ecology Developed

ype were the evolutionary vessel in which were distilled the

origins of the most remarkable of all animals: Homo sapiens.
Forests were the habitat of our earliest evolutionary ancestors and
have remained an important part of the environment of most
branches of the human family tree. The rise and fall of empires,
the conquest of nations, and the political, economic, and military
power of human societies have been intimately related to the ac-
cessibility of forests and/or forest products for most of our recent
history. Although modern humans are no longer truly forest ani-
mals, we still are—and probably always will remain—dependent on
forests for a wide variety of the necessities and pleasures of life as
we know it.
Part I examines the current human predicament and reviews
the evolution and population growth that led to the present cir-
cumstances. It examines the history of our relationship to the for-
est and the role that this relationship has played in the
development of forestry and forest ecology.
 Sustainability of Forest
Ecosystems
The Problem of Human Population Growth

1.1 Introduction siderable capacity of planet earth to sustain humans,

other species, forests, and the earth’s environmental
In the minds of many people, the concept of sustain- conditions. Gloomy forecasts of famines of food and
ability is closely linked to the concept of lack of wood, water and air pollution, loss of soil, depletion of
change. Their notion of sustainability of the world’s the world’s marine biota, and, currently, alteration of
forests involves the idea of no reduction in forest area, the global climate have been a major preoccupation of
forest condition, or the multitude of resource and environmental organizations, many scientists, and the
non-resource values that forests provide for humans. media for the past few decades. These environmental
In the past, this has been illogical. As the human pop- “fire alarms” have had a salutary effect in galvanizing
ulation has grown from an estimated 1 to 10 million public opinion and government policy to address these
12,000 years ago to approximately | billion 200 years issues. However, as long as population growth and his-
ago to approximately 6.5 billion in 2004, as much as torical patterns of consumption continue, it is difficult
40% of the world’s forest cover has been removed for many people to see how historical trends in im-
(Salim and Ullsten, 1999). Agriculture, cities, roads, pacts on the world’s forest can be changed.
power lines, reservoirs, and other land uses that serve Recent trends are more encouraging. Evidence of
the expanding human population have replaced forest- a major slowing of world population growth and of the
related land uses. Sustainability of forests at the global ability of agricultural and forest ecosystems to supply
level has been an attractive but unattainable ideal as human demands for food and fiber from a relatively
long as human populations and their per capita impact small proportion of the land have raised doubts about
on forests have continued. the continuing validity of these earlier concerns. The
The Club of Rome (Limits to Growth, Meadows agricultural “green revolution” of the past half century
et al., 1972; Beyond the Limits, Meadows et al., 1992), and the prediction that the world’s wood fiber supply
the World Commission on Environment and Devel- could be satisfied sustainably from only 10 to 20% of
opment (Our Common Future, WCED, 1987), the the world’s forest land (Binkley, 1997; Sedjo and
United Nations Conference on Environment and De- Botkin, 1997) suggest that the neo-Malthusian “doom
velopment (Agenda 21—the UN Action Plan on Envi- and gloom” scenarios have fulfilled their role of alert-
ronment and Development into the 21st Century, ing the public and their politicians to the environmen-
UNCED, 1992), and the World Commission on tal predicament that humans were creating. It is now
Forests (Our Forests, Our Future, Salim and Ullsten, time to design ways to overcome the social and politi-
1999) are examples of many warnings during the past cal impediments to achieving the ecologically possible
30 years that human population growth and per capita sustainable relationship between people and forests. A
impact on the environment are already taxing the con- growing number of people are challenging the extrap-
~
3
4+ CHAPTER1 = Sustainability of Forest Ecosystems

olation of past trends as a way of predicting the future The increase in hunting success and the concomi-
and the validity of claims that things continue to get tant improvement in diet that accompanied the devel-
worse. In some areas of environmental concern, evi- opment of weapons are thought to have stimulated the
dence suggests that things are getting better or were first of three major changes that have occurred in the
never as bad as has been suggested (Lomberg, 2001), growth rate of the human population. In the first of
although many disagree with this conclusion. Other these, there was probably only a marginal increase in
aspects of the environment continue to deteriorate. In the rate of population growth, which previously had
yet others the good news is limited to a slowing of the been very close to zero. However, this small increase,
rate at which things are getting worse. combined with the improvement in hunting ability,
The debate over the merits of competing claims seems to have resulted in the virtual decimation of
with respect to the environmental situation will con- much of the world’s large game populations between
tinue. Some of the disagreement arises from the over- 30,000 and 10,000 B.c., known as the Pleistocene
extrapolation from real local problems and_ the overkill. Although there is some dispute as to whether
ignoring of broader trends on the one hand and the all of the Pleistocene extinction of larger mammals
focus on averages and global trends while ignoring can be explained by human predation alone, the evi-
real local and regional problems on the other. Much of dence of rapid declines in abundance of large mam-
the debate arises from the unreliability and incom- mals in several localities after the first invasion of
pleteness of the data on which it is based and from dif- human hunters is compelling. For example, 71% of
fering assumptions about whether past trends will the genera of large animals in North America became
continue in the future. There can be little doubt, how- extinct shortly after the arrival of technologically ad-
ever, that the very rapid growth of the human popula- vanced human predators (Martin, 1966, 1967). Habi-
tion during the past 200 years has been the major tat changes accompanying climate change, as well as
driver of our environmental impact and that the future other biological interactions, may also have con-
growth of the population will determine to a consider- tributed.
able extent the future of the world’s forests. Early human success at hunting provided the
protein-rich diet necessary to support population
growth, but it was not without its price. It created food
1.2 Past and Future Growth shortages that initiated an important new evolutionary
in Human Numbers development. By 15,000 to 10,000 years ago, many
populations of primitive humans had been forced back
As our ancestors evolved from tree-dwelling primates to the earlier ground-ape habit of food gathering.
into successful ground-dwelling group hunters, there Grains of wild grasses became a staple food, and it was
was a gradual increase in numbers. From what is not long (on an evolutionary time scale) before do-
thought to have been an initial population of approxi- mestication and cultivation of both wheat and rice oc-
mately 125,000 hominids 1 or 2 million years ago, curred. The nomadic hunters became territorial
numbers increased slowly to reach an estimated 1 mil- farmers, and the next two millennia saw the develop-
lion approximately 150,000 years ago, when Homo ment of the great “hydraulic societies” based on irriga-
sapiens evolved as an identifiable species.' By the time tion agriculture in the Tigris-Euphrates, the Indus,
of the Cro-Magnon and Neanderthal types of primi- and the Nile river valleys. This proved to be a highly
tive human (30,000 to 40,000 years ago), the popula- successful development, which resulted in a large ex-
tion had increased substantially to approximately 3 cess of food production over immediate needs and
million; by 10,000 years ago, it is thought to have stimulated the second major increase in the rate of
reached approximately 5 million (with a variation in human population growth. From an estimated 5 mil-
estimates from 1 to 10 million). lion in approximately 8000 B.c. (when agriculture was
in its infancy), human numbers grew to approximately
250 million by A.D. 1 and to approximately 350 million
, ‘ : : by the late 13th century.
All population estimates before approximately A.D. 1800 are approximate
guesses made by demographers and archaeologists. After 1800, the esti- In contrast to the agriculture of the eastern hy-
mates became increasingly reliable, but even today we do not know exactly draulic societies, which was very successful, early Eu-
how many people there are on Earth. ropean agriculture was much less productive. Because
 Implications ofPopulation Growth for Global Resources 5

of the lack of methods for working the heavier, more about future mortality and fertility (birth rate). Fi-
fertile valley bottom soils, food production was largely nally, Figure 1-1F presents the 1990 world age class
restricted to areas with lighter, less fertile soils in distribution and the predicted 2030 distributions for
valley-slope and ridge-top forests. These lower- two of the scenarios in Figure 1-1E. The 100-year
productivity soils lacked the fertility to sustain population forecast from the International Institute
continued cropping, and it was necessary to leave large for Applied Systems Analysis in Austria (Lutz, 2000)
areas in fallow each year. The invention in approxi- suggests that the population will peak in approximate-
mately A.D. 1000 of a new agricultural implement that ly 2070 at slightly less than 9 billion and then begin a
was capable of tilling heavy soils rich in clay and silt slow decline. In contrast, the United Nations forecast
(the moldboard plow) led to utilization of the richer suggests continuing growth to the end of the century
and more productive valley-bottom soils and signifi- to a level of approximately 9.5 billion.
cant increases in agricultural production. However, These forecasts have very significant implications.
truly sustained-yield agriculture in Europe awaited the If they are correct—if social, cultural, and political
introduction of wheat-turnip-barley-clover rotation considerations permit the application of sustainable,
cropping in 1650. This system resulted in a dramatic ecologically based forestry, agriculture, and fishing and
increase in agricultural productivity and marked the if humanity can switch to renewable energy sources
beginning of the third and most spectacular increase that do not contribute to changes in atmospheric
in the rate of human population growth. chemistry that result in ecologically disruptive climate
The explosive increase in agricultural production change—then our great grandchildren will inherit a
that followed the introduction of rotation cropping planet in which much of the undesirable environmen-
freed both labor and capital for the pursuit of scien- tal alterations of the past century will have been recti-
tific, industrial, and medical activities that were to fied. The contribution of the science of forest ecology
transform Europe’s population from 22 to 35% of the is to provide the foundation for sustainable forestry. If
world’s population in the short period from 1800 to these population forecasts are wrong and population
1930 (Russell, 1969). From a world population of ap- growth were to continue, then talk of sustaining the
proximately 545 million in 1650, there was an increase world’s forests is probably an academic exercise.
to approximately 728 million in 1750, to approxi- Table 1-1 summarizes the history of population
mately 1171 million in 1850, to approximately 2515 growth and presents some forecasts.
million in 1950 (Hauser, 1971), and to approximately
4000 million in 1975 to 1976. The growth of the
human population over the past 2 to 5 million years is 1.3. Implications of Population Growth
shown in Figure I-1A. for Global Resources
Figure 1-1B shows the U.S. Census Bureau data
on population growth from 1950 to 2000, with a pro- An animal population is limited ultimately by aggres-
jection to 2050, showing the doubling from 1960 to sion, disease, or lack of vital resources (food, water,
2000 and the slowing in growth from 2000 to 2050. shelter, space, or other material needs). These ulti-
Figure 1-1C shows how the annual growth rate (in mate limits to growth are generally not reached be-
percentage) varied over this period, with peak rates in cause other, less violent contro] mechanisms maintain
the mid 1960s—exactly the time when concern over the population at much lower levels. Unfortunately,
human population growth peaked. Since then, there populations of some countries still do not seem to be
has been a steady decline, and this is predicted to con- exercising such restraints, and if past trends were to
tinue. The number of people added annually to the continue, it seems very likely that war, pestilence, star-
population peaked 25 years later, in 1990 (Figure vation or shortage of some essential resource could be
1-1D). Even though the percentage growth rate was the means by which their population will level off and
declining, the number to which this percentage ap- possibly decline to lower levels. As to whether and
plied was increasing faster, so the number of people when this might occur, opinions vary greatly. Most au-
added to the population each year continued to in- thorities judge food to be the most important, ulti-
crease. All of these projections are fraught with uncer- mate limiting factor, but there is a wide range of
tainty, however. Figure 1-1E shows the range of opinion as to how many people the world can feed.
projections that result from different assumptions Food production in turn will be ultimately limited by
 6 CHAPTER1 — Sustainability ofForest Ecosystems
10
‘ 3S 2 ie
Sige aiengd
og <2 -
=o
¢ 8
oo Z
oleae 5 < Py < & =>2 8
= 2 of o &
a
5 6 | g&
3 Seema
¢ 2)
Pigs
= S
$7
=
5
24
<
wer)
L z Sp
ae Z
sy
a = a6
a { &
cS ms nr ttoo ee
e on
S 3 Black death— =
= 4
¥ the plague ead
s
S Christian =
era begins = 2
Ps <
|
l

8000 7000 6000 5000 4000 3000 2000 1000 0 1000 2000 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050
A.D. Year
B.C.
(d)
(a)

High fertility, low mortality

High fertility, high mortality
Central projection
Low fertility, low mortality
3 billion Low fertility, high mortality
more people

(billions)
Population (billions)
Population
Population doubled
40 years

1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050
Year Year
(b) (e)
2.5 -- Age Class
OM 80-85 BCT Z Actual distribution 1990
0 75 2- 80 ZO Annee
Distribution pe» 2030 for
in
O70 ls cro DO jow mortality, low fertility
Cs 60 - 65 ET oO Distribution in 2030 for
(TSS - 60 Ba) — high mortality, high fertility

Growth
(percent)
rate
Oo sn a ' iS)oO
: Ss
. t

1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 500 400 300 200 100 O 0 100 200 300 400 500
Year Numbers in each age classs (millions)
(c) (f)

Figure 1-1 Growth of the human population. (A) The historical pattern of population increase over
the past 2-5 million years. The majority of the growth in numbers occurred in the past 50 years. (B) Popu-
lation growth from 1950 to 2000, with a projection to 2050. (C) World population growth rate, 1950 to
2050. (D) Annual world population change, 1950 to 2050. (B—D are based on U.S. Bureau of the Census
data, 2000.) (E) Estimates of future population growth to year 2050 based on various assumptions. [The va-
riety of projections illustrates the uncertainty involved in projecting the growth of the human population.
(F) Distribution ofthe world population in different age classes for 2030 according to the central projections and the
low mortalityow fertility and high mortality/high fertility assumptions shown in E (data from Lutz et al., 1994),
(Redrawn after Van der Tak et al., 1979, and Lutz. et al., 1994, and from the U.S. Census Bureau Data for 2000).
 Implications ofPopulation Growth for Global Resources 7

Table 1-1 Growth of the Human Population

Estimated Range Estimated Time Estimated Time
in Population Required to Double Required to Attain
Date (Millions) the Population (Years) Successive Billions Date Attained

10,000 B.c. 1-10 35,000 1st billion: 2-5 million years A.D. 1800
5,000 B.c. 2-20 5,500 1st billion: 2-5 million years A.D. 1800
B.C./A.D. 170-400 1,700 1st billion: 2-5 million years A.D. 1800
A.D. 1000 254-45 1,700 1st billion: 2-5 million years A.D. 1800
A.D. 1400 350-374 230 1st billion: 2-5 million years A.D. 1800
A.D. 1800 813-1,125 139 1st billion: 2-5 million years A.D. 1800

A.D. 1900 1,550-1,762 116 2nd billion: 130 years A.D. 1930

A.D. 1950 2,400-2,556 44

A.D. 1970 3,707 35 8rd billion: 30 years A.D. 1960

A.D. 1980 4,457 30 4th billion: 15 years A.D. 1975

A.D. 1990 5,284 42 5th billion: 12 years A.D. 1987

A.D. 2000 6,080 — 6th billion: 12 years A.D. 2000

Projections

A.D. 2010 6,820 =

A.D. 2025 7,840 - 7th billion: 13 years A.D. 2013

A.D. 2050 9,100 — 8th billion: 15 years A.D. 2028

A.D. 2070? 9,000 — 9th billion: 17-20 years A.D. 2045

A.D. 2100 8,400*-9,500 —

All data not so indicated are taken from the U.S. Census Bureau web page: Historical Estimates of World Population.
*Data from Lutz, 2001.

the availability of water. Earlier optimistic estimates solve pollution or energy problems, which would then
suggested that by applying state-of-the-art agriculture limit population growth. Solution of the pollution
to all suitable land, we could feed 38 to 48 billion peo- problem would limit the capital for agricultural expan-
ple (Revelle, 1974). However, such estimates of tech- sion. Use of land to solve the food production problem
nological possibilities tend to ignore the large capital would require the removal of large areas of forest and
investment required for their attainment and other result in “timber famine.” Political and social problems
limiting factors. also limit the application of technology and capital:
One of the conclusions of a study by the Club of many countries spend far more on defense budgets than
Rome (Meadows et al., 1972) was that if present trends on trying to solve population and resource problems.
of population growth and resource depletion were to More recently, it has been pointed out that even if vast
continue, then the limits to population growth will be increases in food production were achieved, many of the
reached in less than 100 years. They concluded that al- world’s people would have no fuel with which to cook
though it is technologically possible to solve individual the food unless alternatives to traditional energy sources
problems (e.g., food, minerals, pollution, water sup- are developed. Forest clearing to grow the food would
plies) by application of existing and new technology, have eliminated the supply of domestic fuel wood.
competition for economic resources will prevent this The Club of Rome study concluded that the
from happening. The solution of the food problem human predicament is not hopeless—that we have
would require the preemption of capital required to much of the knowledge required and still have the
8 CHAPTER1 = Sustainability ofForest Ecosystems

time in which to alter these growth trends and to es- Table 1-2 Estimates of Percentage Annual Defor-
tablish a condition of ecological and economic stabili- estation during the 1980s in Various Parts of the
World*
ty that is sustainable far into the future. The sooner a
start is made, the greater the chance that planned Average Forest Area
rather than cataclysmic controls will limit the world’s Annual Lost per
population. However, unless major steps are taken in Location Deforestation (ha) Year (%)
the next few decades, the chance of an innocuous solu-
World 11,385,000 0.6
tion to the population predicament (i.e., control by re-
ducing the birth rate) will become very small, and Africa 3,714,000 0.5
solutions involving an increased death rate will in- Asia/Pacific 2,020,000 0.7
evitably prevail.
Central/South America 5,650,000 0.6
What is the meaning of all this for forestry? Even
with the prediction that the world’s population will North America —° —
now grow only to 9 to 10 billion and then decline, the Europe = =
additional 3 to 5 billion people will place tremendous-
ly increased demands on the world’s forest resources, Jamaica 2,000 3.0

demands that cannot easily be met because of the Haiti 2,000 3.8
long-time scale of forest crops in comparison with the Costa Rica 65,000 3.6
rate of population growth. Some of the tree crops that
we regenerate in the first decade of the 21st century Sri Lanka 58,000 3.5
will serve a world population 50 to 66% larger than in Nepal 84,000 4.0
2000. At the same time, the growing competition for Thailand 379,000 2.4
land between forestry and agriculture and between
forestry and other land uses will mean that in many Malaysia 255,000 AZ

parts of the world, total forest area or the area of forest Indonesia 620,000 0.5
available for the harvest of wood products will be Malawi 150,000 3:5)
greatly reduced. In addition to forest clearance for
agriculture, there are significant losses of commercial Cédte d’lvoire 510,000 Oe
forest land to hydroelectric developments, powerline Kenya 39,000 let
rights of way, parks, and environmental protection
areas. Forest land is also being lost because of erosion From Sharma, 1992.
°No data given. In many parts of the temperate latitudes, the area
accompanying poor-quality forest management. Much and volume of forests are increasing because of reduced forest fires
of the developing world is currently facing a serious and reforestation of abandoned agricultural land and denuded areas.
timber famine, which is likely to get substantially
worse. The effects of this will have serious implica-
tions for human welfare. rate of loss of forests. In 2000, the world’s forests were
In 1988, the total world area of closed forest was estimated by the United Nations (U.N.; State of the
estimated to be 2.9 billion ha. It was thought that this World’s Forests, 2001) to have lost 94 million hectares
would be reduced by 170 to 200 million ha by the year since 1990, an average annual loss rate of 0.2%, one
2000, mostly in the tropics, and that by the year 2025, third of the rate reported for the 1980s by Sharma et
the loss would rise to 600 to 700 million ha, a reduc- al. (1992) and 60% of the rate estimated for this peri-
tion in the area of tropical forest of approximately od by the U.N. (note the substantial difference
30% (Sharma et al., 1992). This is a rate of deforesta- between these two estimates for 1980 to 1990, em-
tion of approximately 37 to 43 ha per minute. A some- phasizing the difficulty in determining true rates of
what lower estimate of 22 ha per minute is given in forest loss).
Table 1-2. Between 1980 and 1990, the average annu- In 2000, the U.N. estimated that forests covered
al loss of forest area was estimated to be 15.5 million 3.9 billion ha, or approximately 30% of the world’s 13
ha per year. billion ha of land: 187 million ha were listed as planta-
Estimates from 12 years later are different, indicat- tion forests. As noted in Table 1-1, forest losses are
ing that some progress is being made to reduce the not equal by region or country within a region. Be-
 Summary 9

tween 1990 and 2000, Africa lost 52 million ha (an an- support in general terms the warnings posed by the
nual loss of 0.8%); Asia lost 3.6 million ha. This less Club of Rome. The continuing although slower
dramatic estimate, which conflicts with what one sees growth in human numbers demands that the world’s
while visiting many Asian countries, is the result of a remaining forests be managed so that they are sustain-
reported 18 million ha gain in China’s forests (1.2% able, and that depleted forests be restored wherever
per year) and a 2.4 million ha gain in Kazakhstan possible. The present momentum of human popula-
(2.2% per year). These offset losses of 13 million ha in tion growth requires that all resource managers have a
Indonesia (1.2% per year), 5 million ha in Myanmar good working knowledge of ecology, and that forest
(1.4% per year), and 2.4 million ha in Malaysia. Eu- management be based on ecosystem ecology, the sci-
rope gained 8.8 million ha (0.1%). The United States ence dealing with the relationships between living or-
gained 3.9 million ha (0.2% per year), and Canada re- ganisms and their total environment. Anything less is
mained unchanged. Mexico lost 6.3 million ha (1.1% unacceptable.
per year), Nicaragua lost 1.2 million ha (3.0%), South
America lost 23 million ha (0.4% per year), and six
other South American countries lost 0.4% to 0.8% of SUMMARY
their forest per year. The human species is a product of biological evolution,
These U.N. figures suggest a substantial reduction and we share with other living organisms the fate of
in deforestation around the world (1990 to 2000 annu- being dependent on the environment. Cultural evolution
al loss of 0.2%, vs. 1970 to 1980 loss of 0.6%) and in has certainly made us seem to be different from other
Asia (~0.1 vs. ~0.7%). However, annual losses in Africa animals and has conferred on us a certain degree of in-
were greater (~0.8 vs. 0.5%), and losses in South dependence from our environment. However, this inde-
pendence is fragile, and there is no justifiable basis for
America of 0.4% were less than the earlier decade rate
claiming that humans and our impact on the environ-
of 0.6%. These comparisons are complicated by dif-
ment are any less natural than other species and their ef-
ferences in definition of “forest,” by differing areas in- fects on the environment. The most useful view of our
cluded in some of the summaries, and by often place in the world is to regard people as an integral and
questionable data. Nevertheless, they do suggest sig- dependent part of the total environment.
nificant progress in reducing the rate of deforestation Recent human evolutionary success has resulted in the
and in reforesting land stripped by agriculture or attainment of apopulation level that is taxing the ability of
other land uses during the past century. the earth to support it. Local populations have experi-
The Club of Rome study noted above rang an en- enced shortages throughout much of human history, but
vironmental “fire alarm.” It predicted disastrous con- there has nearly always been somewhere else to which
sequences of failure to change the patterns of people could turn where resources were still in abundance.
This situation has changed. The world’s population has
economic development and population growth that
reached such a large size and is still growing so rapidly that
characterized the 1950s and 1960s. ‘The study has there are few if any unexploited areas and uncommitted
been criticized for the assumptions that it made and resources. The decline in the rate of population growth
methods of prediction. Both involved considerable and the forecast that the size of the human population will
simplification of very complex issues, and many things level off this century is good news. However, the addition
were not foreseen by the Club of Rome study team. of another 3 to 4 billion people—an increase from 2000
For example, they did not foresee the extent to which levels of 50 to 66%—before stabilization and the increas-
recycling, improved efficiency of wood and energy ing wealth and per capita resource demands of the majori-
use, and product substitution would occur before the ty of the population that lives in underdeveloped countries
end of this century. They also did not foresee the suggest a moderation of optimism. ‘The 21st century will
see serious shortages of several basic necessities for life.
slowing of population growth that has occurred. Nev-
There can be little doubt that the application of knowl-
ertheless, there are limits to growth, and the Club of edge and technology can go a long way toward alleviating
Rome report played a vital role in alerting the world to these shortages, but the evidence that acceptable techno-
the problems that we face. logical solutions will be developed as rapidly as the popu-
The continuing loss of tropical forests, alterations lation increases is not very reassuring. Neither is the
of atmospheric chemistry, pollution of many of the evidence that major economic, cultural, social, and politi-
world’s fresh water and coastal marine environments, cal impediments will constrain what knowledge and tech-
damage to soil, and loss of wildlife and plant species do nology could potentially achieve.
10 CHAPTER 1 Sustainability ofForest Ecosystems

Managers of renewable natural resources face a will have to establish a sustainable relationship with the
tremendous challenge: that of greatly increasing produc- planet’s forests and other ecosystems during the final 100
tion to satisfy the needs of the growing human popula- years of population growth.
tion while reducing undesirable effects on other
resources. In many cases, the increased production must
come from a smaller area. Faced with this challenge, it
should be obvious that only the highest-quality manage- STUDY QUESTIONS
ment will be acceptable. Such management can occur
1. How many people might there be in the world when
only if it is based on a sound understanding of the ecolo-
you harvest a 60-year-old forest that was established
gy of the resource.
in 2000? What will this mean for the world’s forests?
2. What will happen if per capita consumption of forest
TAKE-HOME MESSAGE products and use of forest values increase?
The ultimate threat to the world’s forests and the global 3. From where do you think that the extra wood that will
environment is human population growth. If the human be needed by this population will come?
species wants to persist indefinitely on this planet, then it
 P

2
Development of Forestry
and Forest Ecology

2.1 Introduction forestry as in any other resource field. At a global level,

a shrinking resource base as a result of poor land man-
In the developed countries of the world, the 1960s was agement practices and competition from alternative
a decade of concern over pesticides, local air and water land uses almost certainly requires that increasing de-
pollution, and, toward the end of the decade, popula- mands for a wide variety of forest products, non-
tion. The 1970s became a decade of concern over re- consumptive values, and environmental services in the
source depletion (especially energy), population future will have to be satisfied from a smaller land area.
growth, acid rain, and economic instability. In the Our dependence and impact on the forests of the world
1980s, the danger of resource depletion was widely from our early evolutionary history until the present
recognized and environmental concerns expanded and the historical pattern of development of forestry
from a local and regional focus to global issues such as show that the past 6,000 years of cultural evolution have
climate change, ozone depletion, and the health of the been molded to a considerable extent by the availability
biosphere. Concern over clearance of tropical forest of wood and that self-induced shortages of forest re-
and threats to tropical biodiversity grew steadily dur- sources have been our frequent companion. Such short-
ing the decade to become a dominant conservation ages have often played a critical role in the unfolding of
issue. The 1990s became the decade in which the world history and were instrumental in the develop-
world confronted issues of sustainability, biodiversity, ment of the art, science, and business of forestry. The
and old-growth forest. The early focus on the survival possibility that our dependence on the forest is going to
of the tropical rain forest became overshadowed by increase again makes it necessary to review the history
the international debate about logging temperate rain of this dependence and its effects on human cultural
forest and the increasing harvest of the boreal forest. and political evolution. Much of the following discus-
The need for conservation and wise management of sion is based on that of Winters (1974); for an excellent
resources is gaining wide acceptance, and an apprecia- review of human impacts on Mediterranean forests, the
tion of the importance of renewable or “inexhaustible” reader should consult Thirgood (1981).
resources is growing. Actual or incipient shortages of
energy, raw materials, food, and water are influencing
both the lives of individuals and the politics and poli- 2.2 Human Evolutionary
cies of nations—as they always have done. Dependence on Forests
Local and regional resource depletion and resource
scarcity are as real in forestry as in any other resource Humans almost always have been—and probably al-
endeavor, and awareness of the need for improved re- ways will be—to some extent dependent on forests.
source management is probably as well developed in Having evolved from a primitive species that lived in

11
12 CHAPTER 2 Development ofForestry and Forest Ecology

and depended on ancient tropical forests, human an- and the development of metal saws and axes permitted
cestors became totally tree-dependent, arboreal pri- the clearing of the much larger trees associated with
mates. Trees were their habitat, their environment, these more productive sites.
their source of food, and their protection from ene- The major agricultural advances accompanying
mies. It was the reduction in area of ancestral tropical these developments resulted in an excess of food pro-
forests that generated the evolutionary pressures to duction over current needs. This stimulated trade,
leave the safety of the tree, to enter the more danger- which in turn gave rise to transportation needs involv-
ous environment of the grassland, and to develop the ing wooden boats and animal-drawn wooden carts.
physical and mental attributes that ultimately led to The development of a merchant marine led to the
our current dominance over the plants and animals of need for a military marine to protect trading routes.
the world. As our ancestors evolved from tree- As the growth of trade and commerce led to the
dwelling primates into ground apes, they still depend- growth of wealth and power, the earlier skirmishes and
ed on trees for escape cover, and the forest— looting raids between rival tribes evolved into orga-
grassland interface was probably a good hunting area. nized wars between nations competing for control
Development of stone tools gave primitive humans over lands, resources, and trade routes. These wars
a period of relative independence from the forest, in- were based for a period of approximately 5,000 years
creased their hunting ability, and stimulated the first on a largely wooden technology, and preparations for
of three major increases in population growth rate; war placed heavy demands on local forest resources.
this occurred between 1 million and 100,000 years The great periods of naval warfare saw the supremacy
ago. With the onset of the last ice age, this growth of those nations that were able to secure ample sup-
probably would have been reduced had the use of fire, plies of shipbuilding timber, and they resulted in the
using wood as fuel, not developed. ‘Together with the denudation of substantial areas of forest and subse-
use of furs and shelters, fire permitted in cold areas the quent timber famines. Sometimes it was the quest for
existence of an animal that was thermally designed for alternative supplies of timber to compensate for tim-
a warm grassland environment. The development of ber shortages that led to a subsequent war.
wooden stockades and shelters gave increasing inde- The importance of forests as commercial and
pendence from cave dwellings and increasing protec- strategic assets was recognized very early in the devel-
tion from wild animals, and the construction of dry opment of human culture. The ownership of forests
food storage facilities improved humans’ ability to sur- played a major role in the history of human ex-
vive long, cold winters. As fire and wooden shelters ploration, colonization, and international conflict.
became important, the earlier dependence on the for- Several major events in human history of the last mil-
est began to return. lennium have turned on the supply of timber for trade
The second major increase in the rate of human and warfare, and earlier history has many similar ex-
population growth was associated with the develop- amples of the important role of forests. For example, it
ment of agriculture approximately 8,000 years ago. has been suggested that a significant factor in the
This development also marked the start of one of the British defeat in the American War of Independence
major processes of deforestation: forest clearing for was her inability to build and maintain a sufficient
food production. Because the lighter soils used for number of ships to attend to both her European and
early arable agriculture were frequently infertile, pat- her North American military commitments. Indeed,
terns of shifting agriculture developed that depended the reservation of extensive areas of New England
on the reinvasion of the forest to restore the fertility of forests for British naval use was probably a contributo-
the land. This method of food production has little ry factor leading to the war. The development of
long-term effect on forests if the forest fallow period is British colonial interests such as India in the 18th and
long enough. Major improvements in agriculture— 19th centuries was to a considerable extent related to
and the concomitant deforestation—awaited the dis- shortages of home-grown ships’ timbers to support
covery of how to produce copper and iron by smelt- British naval supremacy. The critically important sup-
ing, which was dependent on the use of wood charcoal ply of mining timbers (pit props) to Britain’s coal in-
as a fuel. The development of metal and wooden dustry was greatly reduced during World War I by the
plows and the harnessing of domestic livestock power German naval blockade of the north Atlantic and
permitted the utilization of heavier, more fertile soils, other strategic supply routes, and pit prop reserves
 History ofHuman Impacts on Forests 13

were at one point reduced to a 6-week supply. With- pend on forests for many basic resources, for renew-
out coal, Britain’s war machine could not have contin- able energy, and for the maintenance of what is com-
ued, and European history during the past half monly referred to as a “high-quality environment.”
century might have been substantially different.
The Industrial Revolution, now more than two
centuries old, could not have occurred without ample 2.3 History of Human
supplies of wood for buildings, machinery, and power. Impacts on Forests
Although coal increasingly supplanted wood and char-
coal during this period (largely because of a shortage Although human populations have developed at differ-
of the latter), the coal could not have been obtained ent times in different parts of the world, it is possible to
without adequate supplies of mining timbers, wooden make a general statement about the characteristic pat-
carts, and other wooden paraphernalia that accompa- tern of growth of their impact on the forest. Of course,
nied early coal-mining activities. Without the devel- as was the case for population growth, it must be real-
opment of inexpensive paper derived from wood fiber, ized that this basic pattern has varied considerably in
it is unlikely that the rapid accumulation and spread of different areas. Following this general statement, the
knowledge and information that accompanied this pe- development of human impacts on the forest in some
riod of technological revolution would have occurred. specific areas is examined.
Although paper made from animal or grass fiber had The effects of primitive humans on forests before
been available from much earlier times, its use would 11,000 B.c. were very minor. Being hunters and gath-
have been too expensive or the supply too limited to erers, humans had little reason to clear the forest, and
support the knowledge and education explosion of the with only crude stone tools, there was not the technol-
past century. ogy to do it. Undoubtedly, fire was used on some occa-
In addition to a direct dependence on trees, hu- sions to drive animals during hunting and possibly to
mans have enjoyed and depended on many of the indi- improve grazing. This may have affected significant
rect benefits of trees. Regulation of stream discharge, areas of forest locally, but human numbers were too
maintenance of water quality, stabilization of steep few for this to have been a major force influencing the
hillsides, protection from high winds and avalanches, forests of the time. Their modest needs for fuel, tools,
protection from marauding armies, provision of and shelter would easily have been met by the current
wildlife habitat for food and sport hunting, and provi- growth of the forest, as is the case in some of the sur-
sion of food for domestic livestock (beech and oak viving examples of hunter-gatherer societies. Humans
mast) have long been recognized as valuable forest- at this stage of development had little more impact on
related assets. Deforestation eliminates these benefits. the forest environment than did other inhabitants of
It should be apparent from this brief review that the forest.
almost all human evolution, from primeval, primate The first major change in the world’s forest that is
stock to ground ape and then through agricultural, attributable to humans accompanied the development
technical, and cultural revolutions to the present time, of agriculture. Clearings were made in the forest, and
has been intimately associated with, and conditioned reinvasion by trees was sometimes prevented manual-
by, forests. Whereas the enormous technical advances ly, by fire or by grazing. For a long time, this patch-
of the past 50 years seemed to offer alternatives to or clearing agriculture was on a temporary, shifting basis,
replacements for most forest products, many of these but as metal tools were evolved and heavier, richer
require the use of more energy, create more problems soils were used, agricultural clearing accelerated and
of pollution, or are inferior to wood products. Claims became permanent. The production of metal tools
made during the past four decades that we were about and weapons created a demand for fuel, and the com-
to escape from our historic dependence on the forest bined effects of improved agriculture and metal tech-
have proved to be premature. Because of their renewa- nology produced further population increases and
bility, forest products are likely to become more, concomitant demands for more fuel and timber.
rather than less, important in the future, and the indi- As humans evolved, both their numbers and their
rect and non-timber benefits of forests are receiving per capita use of forest products increased. As a result,
increasing recognition. There is good reason to be- the rate of deforestation outstripped the simple
lieve that most human societies will continue to de- growth in human numbers, larger areas were cleared,
14 CHAPTER 2 Development ofForestry and Forest Ecology

and forest regeneration was increasingly prevented by logging was often followed by grazing and fire, so
agriculture. The two processes of population growth many areas were slowly denuded of tree vegetation
and forest exploitation became linked in a positive after the initial selective harvesting of ships’ timbers.
feedback system: a process in which the size of A de- Of course, this process of deforestation has not contin-
pends on the size of B, which in turn depends on the ued uninterrupted throughout recent human history.
size of A. Any increase in A increases B, which in turn The decline of civilizations and periodic reductions of
increases A, and so on. Forest clearing for agriculture populations by disease and war permitted some of the
led to increased agricultural production, which led to cleared areas to revert to forest. However, many areas
an increase in population. This resulted in an in- were so devastated by loss of soil that prompt return of
creased demand for food, which led to further forest the former forest vegetation was impossible, and early
clearing for agriculture. This positive feedback records indicate that large areas of present-day desert
process has continued with only temporary interrup- and semi-desert were once forested. In some regions,
tions throughout most of the past 10,000 years, with the conversion from forest to desert has probably re-
increases in agricultural and other forest clearing sur- sulted from climatic change, but this cannot account
passing the increase in human numbers. Only recently for all such conversions. It is apparent that human de-
has the rate of forest clearance for agriculture slowed forestation has also played an important part.
down. With great increases in agricultural produc- In addition to the direct military uses of wood, the
tivity, increased food production in many countries is use of wood in the production of metal for the manu-
now being achieved from a constant or even shrinking facture of weapons and tools has been a major factor
land base, and recent forest conservation and replant- contributing to deforestation. Starting as early as 3200
ing have reversed the process of deforestation in B.C., the development of copper smelting, together
some countries. However, the continuing population with the discovery of iron, placed considerable de-
growth in some underdeveloped countries threatens mands on the forest in addition to those associated
to increase the extent of deforestation greatly during with the housing, feeding, and warming of the grow-
the next 50 years. ing population.
Despite the tremendous inroads that agriculture
has made, the greatest human impact on some of the
world’s forests in the past was associated with warfare. 2.4 Development of Forestry
Although our current military activities may have little
dependence on forests, this is a very recent develop- Forestry is the science, art, business, and practice of
ment. During the past 6,000 years, a substantial part of conserving and managing forests and forest lands to
the world’s forests has been denuded in support of ag- provide a sustained supply of forest products, forest
gressive human activities. As early as 3500 B.c., the conditions, or other forest values desired by the forest
Egyptians were building large boats, and if Homer's owner (cf. Ford-Robertson, 1971). It is a branch of
mythological accounts of Agamemnon’s ships at the human endeavor that has developed at various times in
siege of Troy in 1100 B.c. are to be believed, then the history and at various places in the world. It has always
Greeks and the Romans had vast fleets of wooden developed in response to the loss or anticipated loss of
boats. Considering that these boats are thought to forest values caused by unregulated forest exploita-
have lasted only 10 to 15 years, the demands of timber tion. Although the details of the evolution of forestry
for the construction and maintenance of such fleets has varied from century to century, there is generally a
must have been substantial. predictable sequence of stages (Table 2-1).
Naval warfare based on wooden ships was paral- In the early stages of the development of a human
leled by “wood and metal” military activity on land in- society, the forest is simply a part of the environment:
volving wooden siege equipment, wooden defense the habitat of both prey and enemies and the provider
works, and wooden and metal weapons and transport of some of the necessities of life. Trees are cut as human
vehicles. The tradition of wooden warfare was contin- use of the forest develops, but low numbers of people
ued until as recently as a century ago, and the vast Eu- and the lack of technology limit the demands on the
ropean navies of the 16th to 19th centuries account for forest. Unmanaged nature can easily replace what is
the denudation of large areas of the world’s forests. taken. As human numbers and the power of their tech-
Logging on its own did not eliminate the forest, but nology increase, unregulated forest removal begins to
 Development of Forestry 15

Table 2-1 Stages in the Development of Forestry

Stage of Development Result

Preforestry Exploitation Resource depletion

Forestry stage 1 Administrative forestry Failure to achieve conservation and sustainability goals
Forestry stage 2 Ecologically based forestry Sustained production of timber and other conventional
forest products
Forestry stage 3 Social forestry Ecologically based forestry that sustains a wide range
of forest conditions and values desired by society

exceed the regenerative powers of the forest. Local de- Precisely this sequence or parts thereof can be seen
forestation occurs. With time, this leads to regional de- throughout human cultural evolution. The Egyptians
forestation, which may be solved by seeking supplies in got as far as creating local wood shortages, which were
more distant areas, either by trade or by military con- solved by either trade or invasion of forested areas
trol. However, this generally results in the same unreg- such as Cyprus and Phoenicia. When these sources of
ulated exploitation and forest depletion in these new timber were cut off by rival military powers, Egypt
areas. The depletion of forest resources that results has failed to proceed to the subsequent phase of establish-
always been the stimulus to develop forestry. ing its own forest ordinances, possibly because by that
The first stage of forestry is typically based on a set time the population had grown to the point at which
of laws, regulations and rules that lack any sensitivity almost all suitable tree-growing land was being used
to the ecological differences between different forest for agriculture.
ecosystem types across the landscape. Failure to re- The Greeks and Romans progressed through sim-
flect the spatially variable and ever-changing character ilar stages, but both got as far as creating reservations
of forests almost always results in the failure of the ad- overseas to ensure continued supplies. Reliance on
ministrative stage of forestry to achieve its objectives. overseas supplies relieved the need for the develop-
This stage has generally focused on industrial and/or ment of forestry in the home country, although plant-
military supplies of forest products, but sometimes it ing of trees along rivers and canals was encouraged.
has also been concerned with wildlife and protection Extensive deforestation of Greece led to severe soil
forests. Lacking a sound ecological foundation, ad- erosion and land degradation, which in turn led to a
ministrative forestry eventually gives way to an eco- switch from horses and cows to sheep and goats, ani-
logically based stage. mals that are able to survive on the nutritionally more
When well developed and implemented, ecologi- marginal vegetation that grows on the severely eroded
cally based forestry is usually successful in sustaining Greek landscape. This has had lasting consequences
the functional process of ecosystems and their produc- for the national diet. The soil erosion and land degra-
tivity of conventional forest products. However, it dation was so serious that it led to large-scale emigra-
may not necessarily sustain all of the values desired by tions from Greece between 750 and 550 B.c. and
a wealthy, post-industrial society. Silviculturally sus- aroused interest in the biological nature and role of
tainable forests may not sustain the full range of biodi- the forest. In 400 B.c., Plato wrote about the denuda-
versity, aesthetics, and spiritual values of unmanaged tion of previously forested mountains, the drying up of
forests.' Consequently, this is not the final stage in the springs, and the loss of soil. In the same period,
evolution of forestry. This final stage is social forestry, Theophrastus (who is looked on as the father of
which is both ecologically based and biologically sus- botany) developed an interest in plants and forests. He
tainable but also sustains a wide variety of social and named and classified 500 plants and wrote about the
environmental values in our forested landscapes. ecology, technological characteristics, and utilization
potential of different tree species and how the latter
varied according to the type of site in which the trees
‘Some ofthese values are also very important in the cultures ofpeople who were growing. Had the greatness of the Greek empire
are still at the pretechnology exploitive stage. survived for another millennium, history might have
16 CHAPTER 2 Development of Forestry and Forest Ecology

recorded the development of ecology and modern beginnings of modern forestry with the establishment
forestry in Greece. of state forestry commissions, forest reserves, and a
Knowledge of botany, of soil, and of methods of national forest service in the United States. In the past
forest regeneration and culture was developed in Eu- half century, U.S. forestry has continued to evolve,
rope by monks in the 11th century A.D. Control of with a slowly but steadily increasing emphasis on eco-
substantial areas of forest by monasteries prevented logical principles and the beginnings of a move into
wholesale deforestation and ensured a supply of char- social forestry.
coal and small wood, but it did nothing to prevent the Despite—or perhaps because of—the extent and
shortage of large ships’ timbers when the modern peri- importance of its forest resources, the development of
od of wooden navies started in the 17th century. Ordi- forestry in Canada postdates even that in the United
nances that stipulated cutting rotations, the retention States. In some areas, active forest management in
of large trees, and the extent of clearcuts were intro- Canada has started only relatively recently.
duced. However, these early attempts at forest man- It is encouraging to note that in many countries
agement were not entirely successful because they around the world, forestry is now in transition from
were excessively rigid. They did not allow for natural the administrative stage to the ecologically based
biological variations or disasters and did not adequate- stage. Unfortunately, this optimistic statement must
ly recognize differences in the ecology of different be balanced by the fact that large-scale deforestation is
forests and forest species. The subsequent four cen- not merely of historical interest. It is a present-day
turies have witnessed the successive modification of fact and probably an inevitable component of the
these early regulations and methods in the light of ex- human future. Every year, several million hectares of
perience and new biological knowledge, resulting in an Asian forests are lost and millions more are degraded
increasingly ecologically based approach to European by improper use (Ranjitsinh, 1979). For example, the
forestry. Vestiges of “administrative” forestry can still total area of forest in Thailand was reduced from
be found in both Europe and Scandinavia, but there is 58.3% of the country in 1952 to 39% by 1973 and to
now strong public pressure to move rapidly into the 33% by 1978. Nepal is likely to be largely denuded of
ecologically based and social stages of forestry. forest by the end of the 21st century unless there is a
The development of national forest ordinances in reversal of recent trends. ‘The attention of politicians
France predated the similar event in Germany. For has been focused on this serious state of affairs, and
political reasons, Germany did not develop national large programs of afforestation have been initiated.
ordinances until near the end of the 19th century, al- However, there are many ecological and institutional
though there were local and regional ordinances well impediments to these programs, and it remains to be
before this, including a ban on clearcutting in the seen which of the two forces, afforestation/reforesta-
Salzburg area as early as 1237. Lack of national coor- tion or deforestation, will win.
dination permitted extensive deforestation despite
these regulations, and with the exception of the less
accessible mountain forests, much of the country had 2.5 Ecology, Environmentalism,
been deforested by the time a national organization or Green Religion: Which Will Guide
was developed. the Future Evolution of Forestry?
The history of forestry in North America follows
the same general pattern, although the knowledge and The development of forestry is one of the inevitable
experience gained in the development of European outcomes of unregulated forest exploitation. Pre-
forestry had a restraining influence. The conservative dictably, administrative forestry fails to achieve its
attitude of the first settlers toward the forest reflected conservation and sustainability goals, and although
the more advanced stage in the evolution of forestry in ecologically based forestry may sustain ecosystems, it
their home countries. However, profligate use of the does not necessarily sustain all of the values desired by
forest soon developed, returning the country to near a wealthy, post-industrial society. The forces that drive
the beginning of the development sequence. The de- the evolution of forestry thus lead inexorably toward
nudation of North American forests by fire and log- the “social” stage. However, the public pressure to
ging appalled travelers returning from timber-short complete the transition from administrative forestry
and frugal Europe, and they helped to stimulate public to social forestry has outstripped the actual change
interest in conservation. The late 19th century saw the that has occurred in most areas and threatens to by-
 Forest Ecology: A Practical Means of Dealing with the Problems of Complexity 17

pass the ecologically based stage of forestry. In the ab- components, functional processes, and change over
sence of experience gained from practicing ecological- time (Chapters 3 to 5 and 7 to 18); (2) systems by
ly based forestry, what will be the paradigm that which to describe the ecological variability of forested
guides us into sustainable social forestry? landscape (ecological site classification; Chapter 6);
One product of the “environmental movement” and (3) systems by which we can predict the response
has been the development of “green religion”: a belief of ecosystems to natural and human-caused distur-
system about the environment that is based more on bance (Chapters 17 to 22). Equipped with this knowl-
faith than on the currently available facts about what edge and these tools, forest managers and_ policy
forest ecosystems are, how they function, and how makers can establish the ecologically defined opportu-
they respond to disturbance. Belief systems based on nities for and constraints on the management of the
faith play an important role in human society: we call world’s forests. Once society has established the con-
them religions. They impart notions of morality, ethi- dition in which it wants its forested landscapes to be at
cal responsibilities, and intergenerational equity—the various times in the future (e.g., 20, 50, 100, 200 or
obligation of the present generation to consider the more years from now), foresters can determine which
needs and options of future generations as well as their type of management and which disturbance regimens
own. Green religion, which has become manifested in must be applied now and at various times in the future
a movement called “deep ecology,” took over part of to achieve these desired future conditions.
the environmental movement in the 1980s and 1990s.
However, the public enthusiasm for the claims of the
adherents of this philosophy are beginning to be tem- 2.6 Forest Ecology: A Practical
pered by the growing body of scientific knowledge Means of Dealing with the
and empirical evidence that does not support some of Problems of Complexity
its claims and assertions.
Science by itself cannot provide the paradigm for Forestry has always developed as the result of the need
future forestry. Unlike Father Christmas, ecology can- to ensure an adequate supply of certain desired goods
not say which forester has been good and which has and services from the forest. The development of for-
been bad. Such value judgments can be made only in est ecology has been for similar reasons. Although it
the context of a society’s prevailing value system, has also evolved into a branch of basic ecological sci-
which is based in part on belief and value systems ence, the growth of forest ecology as an applied sci-
about the environment. ence can, in part, be related to the failure of early
Similarly, green religion cannot be the sole basis forestry to achieve its objectives. The development
for the design of sustainable forest management. It ig- and acceptance of forest ecology by foresters occurred
nores too much of the ecology of the values that it as- because it provided a means of recognizing, under-
serts we should sustain. standing, classifying, and mapping the natural varia-
What, then, should the paradigm be? It should be tion of forests and of predicting the consequences of
environmentalism. Not that branch of today’s envi- alternative methods of management where there was a
ronmental movement that has been taken over by lack of practical experience of these consequences:
green religion, but environmentalism defined in terms forest ecology provides the essential ecological basis
of the origins of forestry and conservation. In this con- for successful forest management.
text, environmentalism is the act of managing the en- It has been a consistent experience throughout the
vironment to sustain that balance of environmental history of forestry that only by understanding the eco-
and social values desired by society and that will main- logical characteristics of forests can they be managed
tain a long-term, stable relationship between humans successfully and sustainably. The empirical approach
and the environment. Environmentalism is the bal- of “try it and see” has made an important contribution
ancing of social demands on the environment with the to the development of forestry and will probably con-
need to sustain ecosystem function and integrity so tinue to do so. However, as the pressures of population
that the environment will continue to be the life- growth and resource demands continue to increase,
support system for the earth’s diversity of life forms this approach will become less and less useful. The
(Kimmins, 1993). time required to produce an answer is too long, and
The role of forest ecology in the evolution of there is a decreasing acceptance of “mistakes.” ‘Today it
forestry is to provide (1) knowledge of ecosystem is necessary to be able to make accurate predictions
 18 CHAPTER 2 Development of Forestry and Forest Ecology

about the forest and its management without waiting NUMBER OF ANTECEDENT DETERMINANTS

decades or centuries for evidence on the success or

failure of our management methods. Such predictions
cannot be made without an understanding of forest Several Many Very Many

AN a
ae
ecology and the allied scientific disciplines.
Forests are difficult to manage because they are
Causal ee ra CSca st =e Determinism
very complex and vary greatly from one place to an-
other. Forest ecology strives to reduce the complica- endantans OR bane
tions posed by this complexity and variability by Increasing complexity of the determinism of a condition or event
recognizing and understanding it. Perhaps the best Ss
>
Decreasing predictability of the occurrence of a condition or event
way to explain this is by discussing the nature of bio-
logical causality and the principle of determinism.
It is a tenet of science that all things are deter- Figure 2-1 Diagram of the principle of determinism,
mined. There is an explanation for all events or condi- which states that all conditions and/or events are associ-
ated with a set of antecedent determinants. There is large
tions (hereinafter referred to simply as events) in the
variation in the number of antecedent determinants that can
form of a set of antecedent conditions, factors, or
be associated with a condition or event, from a single an-
determinants that collectively determine or “cause” the tecedent determinant (the cause) to some very large number.
event. For example, the sun does not rise every morn- This results in three types of events/conditions: those that are
ing in the east and set every evening in the west merely causally determined, those that are multiply determined, and
by chance. If this were so, then it might periodically set those that are statistically determined. The importance of this
in the north or the south. It sets in the west because (im- principle for resource management lies in the inverse relation-
plying determinism or causation) of the way in which ship between the complexity of the determinism and the pre-
the earth moves in relation to the sun. The antecedent dictability of the condition or event.
determinants of sunrise and sunset thus are all of the
factors of celestial mechanics that determine how the
earth moves around its own axis and that of the sun. occurs in biological systems. Any biological event in-
Similarly, if atree dies, then there will be a number of volves at the very least two and most often several
environmental and physiological factors that can be major antecedents. Events such as the death of a tree
identified as antecedent determinants (hereinafter re- thus is an example of what is referred to as multiple de-
ferred to simply as antecedents) of this pathological terminism. When the number of significant determi-
event. This scientific principle is called the principle of nants is large (more than 30, a condition rarely if ever
determinism.’ It covers all categories of determinism observed in biological systems), one approaches the
from the simplest, in which there is a single antecedent other extreme of determinism: statistical determinism.
(the cause) for each event, to the most complex, in The importance of the principle of determinism to
which there is an infinite number of antecedents for science and forestry lies in the relationship between
each event. This is represented in Figure 2-1. the predictability of an event and the complexity of the
The simplest case of determinism, the cause— determination of that event. The fewer the an-
event relationship, is referred to as causal determinism. tecedents, the more predictable the event. In causal
A cause is defined as an antecedent that is both neces- determinism, one has perfect prediction because of
sary and sufficient to determine a particular event. In the one-to-one relationship between antecedent and
other words, the event occurs om/y when the cause is effect. In statistical determinism, one has very low pre-
present, and every time the cause is present the event dictive ability because the very large number of an-
will occur. This is a very rigid definition, and such a tecedents combine to produce a seemingly random,
relationship is largely theoretical. It probably never unpredictable occurrence of the event. In multiple de-
terminism, one has events that vary in their pre-
dictability from high (Figure 2-1, left) to low (Figure
2-1, right). The following serials illustrate the rela-
“Antecedent: Happening before the event. demshig between complexity and predictability.
‘The terminology and concepts in this discussion are based on Bunge First, consider a simple case of determinism: a
(1959). glass rod being broken by dropping it onto a stone
 Forest Ecology: A Practical Means of Dealing with the Problems of Complexity 19

floor. This is a reasonable approximation of causal de- know that soils derived from volcanic rocks are more
terminism. Although there are several antecedents to prone to landslides than are soils derived from granitic
the event of the rod’s being broken (e.g., the action of rocks. Thus, if we were planning clearcuts in a region
gravity, the fragility of the rod, the hardness of the that contains an unmapped patchwork of steep and flat
floor, the weight of the rod, the lifting and subsequent land with both volcanic and granitic soils (50:50), then
release of the rod at a height that will result in break- our ability to predict a clearcut/landslide event would
age), they all are considered to be known and invari- be only 25% on average. Soil parent material is not
able. One could say that the rod is broken because it is the only soil factor determining landslides, however.
released and allowed to fall (the antecedent). The The wetness of the soil and the rooting patterns of the
causal nature of the event results in its predictability trees and minor vegetation are two of several other
being 100%, and 100 repetitions of the antecedent factors that contribute significantly to the determina-
(the action of dropping the rod on the floor) will al- tion of a landslide event. As we add additional un-
ways produce 100 broken rods. known antecedents to the determination of landslides
Consider now the success of a blindfolded observ- following clearcutting, our success at predicting these
er in predicting the fate of 100 glass rods dropped at events in the absence of knowledge about these an-
random onto a stone floor, half of which is covered tecedents drops rapidly, as shown in Figure 2-2. Land-
with small squares of rubber matting distributed at slides are unpredictable events in areas about which
random. On average, 50 rods will land on stone and we have no information on topography, soils, rooting
break, whereas the other 50 will land on rubber and patterns, climate, and so on.
remain intact. The blindfolded observer would be A forest is not merely a stand of trees. It is the total
only 50% successful on average in predicting the assemblage of
breakage of the rods. Thus, by introducing one un-
known antecedent, we have halved the predictability 1. ‘Trees
of that event. If, unknown to our blindfolded observer,
we were to add a second unknown antecedent by re- 2. The substrate on which trees depend for support,
nutrition, and moisture
placing half of the breakable glass rods by unbreakable
plastic rods, then the observer’s predictive success 3. The other plants with which trees interact in terms of
would be reduced to 25%. If we were to add a third mutual shelter, competition, benefit, or antagonism
unknown antecedent by dropping the rods alternately 4. The animals that feed on, shelter under, or bene-
from 2 cm (generally, too small a drop for glass rods fit the plants
falling onto stone to break) and 2 m, then the observ- 5. The microorganisms that exert direct or indirect
er’s average success in predicting the breaking of the beneficial or antagonistic effects on the trees and
rods would be further halved to 12.5%. This sequence other living organisms
of declining predictability is shown in Figure 2-2. It
6. The atmosphere and climate, including fire and
should be obvious that there is little hope of accu-
wind, that influence the distribution, abundance,
rately predicting events that have more than three or
and productivity of all of the organisms in the forest
four unknown antecedents.
For a second example, let us consider a hypotheti-
cal relationship between clearcut logging and the The forest is a complex biological and physical sys-
occurrence of landslides. [f there was a causal relation- tem in which there is an enormous variety of interac-
ship between clearcutting steep hillsides and land- tion and interdependence among the different parts.
slides, then we could state categorically and Because of this great complexity, interaction, and in-
unequivocally that a landslide will occur every time we terdependence, most events or conditions in a forest
clearcut a steep slope. If we were planning a series of are determined by many antecedents: it is multiply de-
clearcuts in a region that had 50% flat land and 50% termined. This has been expressed as follows (Major,
steep land but we had no information as to the loca- 1951): vegetation = f(soil, climate, parent material,
tion of the cuts relative to the topography, then our
success in predicting landslides would, on average, be
50%. However, we know that the determination of
landslides is more complex than this. For example, we * f: a function of.
20 CHAPTER2 _ Development ofForestry and Forest Ecology

100

Example of Breaking Rods

s
B
=

Bs}
2io)
2
a3)
100

2 Stone Stone and Stone and Stone and rubber

& surface rubber — rubber surface
| surface surface Fragile and
3 Fragile and unbreakable rods
r unbreakable Small and large drop
2 rods

“is (b)
2
a
A=}
Y 100
5
Ss
.
“A
s Example of Clearcutting
3

0 I
0123 45 67 8 9 1011 12130 8
Number of unknown antecedents to the event &
a—-
Increasing complexity of determinism

(a)
0
Steep land Steep and Steep and Steep and Steep and flat
flatland flatland flat land land
Volcanic Volcanic Volcanic
and and till soil and till soil
till soil Wet and Wet and
dry climate dry climate
Deep and
shallow roots
(c)

Figure 2-2 Hypothetical examples of the relationship between the predictability of a condi-
tion/event and the complexity of its determinism. (A) General case. (B) Breaking rods. (C)
Clearcutting and landslides. See the text for explanation.

topography, biota, time) and soil = f(vegetation, cli- recognized and its role is understood, we move one
mate, parent material, topography, biota, time). Be- place to the left in Figure 2.2. By understanding all of
cause of this complexity, attempts to predict veg- the antecedents in a complex, multiply determined
etation or soil conditions in a region about which we event, we can effectively reduce it to a case of causal
have no knowledge of the factors on the right side of determinism as far as prediction is concerned. When
the equations will have a low probability of success. we understand the contribution of soil parent materi-
Knowledge of the contribution of individual an- al, soil moisture, vegetation rooting patterns, and all
tecedents to the cause of a particular event effectively of the other significant antecedents of landslides, we
simplifies that cause. For example, in the case of the can predict them accurately if we have mapped spatial
glass rod, the observer’s predictive success improves variations in these factors and if we know how they
when he or she knows the fragility of each rod and the vary seasonally and annually.
height from which it is dropped. It is improved still Herein lie both the justification and the necessity
further when a map of the location of the rubber pads for all foresters to have a working knowledge of both
is available and the observer knows where on this map basic forest ecology and the ecological aspects of
each rod is to be dropped. Each time an antecedent is forestry. A forester’s job is to contribute to the estab-
 Study Questions pak

lishment of a set of management objectives and then adequate basis on which to develop successful sustained-
find ways to achieve them. In helping to set objectives, yield forest management. The result of these failures was
the forester must be able to predict the biological ca- an effort to learn more about the ecological nature of the
pabilities and tolerances of the ecological system resource so that more successful management objectives
under management. In deciding on the methods to be and methods could be developed. The failure of early
used to achieve these objectives, the forester must be forest management was the practical impetus behind the
able to select from among a number of alternatives the development of forest ecology.
Successful forest management requires the correct
one best suited to the situation at hand. This requires
choice from among alternative methods of attaining
the ability to predict the consequences of the various management objectives. Such a choice requires the abili-
alternatives. Only in this way can there be a reasonable ty to predict the outcomes of the various alternatives so
expectation of choosing the best alternative. However, that the best can be selected. Accurate prediction is diffi-
to be able to make these predictions successfully in a cult in a forest because of the complexity of factors
highly complex system such as a forest, one must first responsible for determining events and conditions. Accu-
have a good knowledge of the structure, function, and rate prediction in complex systems is possible, however, if
spatial variability of the forest. By understanding how the components and functional processes of the system
soil, climate, microorganisms, plants, and animals in- are understood. ‘The major role of forest ecology is to
provide forest managers with a basic understanding of
fluence each other, a forester can predict the growth
the way in which the forest ecosystem works: to familiar-
and productivity of crop trees, because this knowledge
ize them with the major determinants in the system.
effectively establishes causal growth relationships. By
understanding how such practices as timber harvest-
ing or slash-burning affect the structure and function
TAKE-HOME MESSAGE
of forests and by knowing the local conditions of soil,
climate, and biota, we can predict the consequences of By definition, forestry must change as the set of forest
these treatments on a particular site. Through knowl- ecosystem values that are to be sustained changes. As
edge and understanding of forest ecology and of the forestry evolves through its various stages, the manage-
local forest conditions comes predictability of those ment objectives change and, therefore, so must the man-
agement methods used. It is a forester’s responsibility to
forests. Through predictability comes the selection of
make those changes, but the forester also has the respon-
ecologically and economically rational objectives and sibility to resist pressures from the public for changes
the best methods of attaining them. that will not result in sustainable management of all the
values that the public wants.
SUMMARY Because of the complex, multiple determined charac-
ter of forest ecosystems, prediction of the short- and long-
Throughout most of their evolution, humans have had a term consequences of alternative forest management
close relationship with forests. Wood and other forest practices is difficult. Success requires considerable knowl-
products have played an important role in molding cul- edge about forests. Foresters have the responsibility to en-
tural evolution, and humans have radically changed the sure that they both have and use the necessary knowledge.
extent and composition of the world’s forests. Unlike
most other biological relationships, the interaction be-
tween humans and forests developed a positive feedback.
As numbers grew, more timber was used and more forest
STUDY QUESTIONS
was cleared for agriculture, and this led to further in- 1. Why has the human species been so dependent on
creases in the population. The process culminated at forests during its evolution?
various times in various different countries in wood 2. Have we escaped from this historical dependence?
shortages that were of great political and economic sig-
nificance and that led ultimately to the development of 3. At what stage of its evolution is the forestry that is
forestry. Early attempts to conserve and sustain wood presently being practiced in the region where you live?
supplies were largely unsuccessful. Regulations and ordi- 4. List some examples of events in forests that are multi-
nances based on military or short-term economic and/or ply determined. How could we learn to predict these
political rather than on biological principles are not an events?
r
a
7
;
|

|
.

)
'

meng it toad 8A
tay
~ f uit)
yeas
ne
ipey
se
> eee

7
ny
'
’

:
‘ ~~
Jxtesngigah ae? ‘
ea 2 Th ir) TP ary itites at
| | whawefe ehce Tibet é may spelling
at Pt rahe rea we eT Sr :

“S10 Gav) Parrsite 76 sve

rine? e Lined mm ST
ony) 4 om } : aie) 7 ie ” A= op ee .
@ i
~
i 2 _ ' a \ a ats Aue:

pitty, Cutt is (GT -Saphgstlh 1hpt sl

lpn — _—_
Par oe hey . ‘aaa ry :
~
Fy ee ;
ol ( an )-o9) ;

-
Me “ oh JOG IT . Sie ar aay om

Rapier bs ~
dui
r
; hy 7
i
, {
:
' rbas tfMyi Vel Gg &
uf
‘i
' ’ bis
{ |

oes ular ev ove

ha — =

-
a)

8 > }

> f
rpitend
- a

ae
ro . pede ) +

aes a hints ys inst act gprs

iM eaatin
» v :

es ah ps rit * am¥V \ee ome 7 yy ’

=i m Cama * e
—_—— ’
 Ecosystem as the Basic
Unit of Forest Ecology

H.... presented the case for forest ecology as an essential

ingredient in the management of the world’s forests, we turn now
to the subject itself. This part of the book examines the forest as a
complex ecological system, an ecosystem, and its major functional
processes.
Many people think of a forest as a stand of trees, but it is far more
than that. A forest is a complex, functional system of interacting and
often interdependent biological, physical, and chemical components,
the biological part of which has evolved to perpetuate itself through
the production of new organic matter. From the earliest stages of
their evolution, humans have been interested in their environment
as much for its functional characteristics as for any other attributes.
Continuing this tradition, contemporary renewable-resource man-
agement is largely concerned with organic production and its ma-
nipulation through ecosystem modification. However, forestry is,
once again, in transition. As it evolves from the administrative stage
to the ecologically based stage or directly to the social forestry stage,
there is a change of focus from sustained yield of timber products to
sustaining the function of the entire ecosystem.
Because of the importance of viewing any renewable resource
as a system and because of the preoccupation of renewable-
resource management with ecosystem function, we consider first
the nature of ecosystems and then examine the two major aspects
of ecosystem function: (1) energy capture, storage, and dynamics
and (2) the input into, their circulation within, and their loss of
nutrient chemicals from ecosystems.
i) an
7
i
‘
egy =
Se
- >

Lf .
' t2 *

Ty)! T 325

_—> a

a, ©

193 i ten
be 2
;
=

D =

é
: ;

a
|
= >
i
P .° yo eee

iMiSpils «Awd pol

uh
PY yous onal ‘With
: P mt OM? ot ‘take
<7 <s

if
“
>
-
Hat 1

=
; |

{ : i+ Otley ri ae ;
a ; wy art 1 aieaN ute
- — ige' ivf : ;
: ze oa ~~ angeryi a8} 18 9
sede weanmeelt Yaelae
ro ney ae eRe Weer ty
rts Aiea Reamer eas
Liiry it
| ui
in ber
wut), vf
7

De ache ‘il“sluse
Oo
 ro

Ecology and the

Ecosystem Concept

3.1 The Science of Ecology and Its and of environmental tolerances and requirements
Historical Development would have followed the development of farming and
the domestication of plants and animals. This knowl-
Ecology is the branch of biological science concerned edge was purely experience based, of course. The de-
with the distribution, abundance, and productivity of velopment of scientific observation and the recording
living organisms and their interactions with each other of relationships among animals, plants, and the physi-
and with their physical environment. As a scientific cal environment awaited the development of Egyptian
discipline, it is relatively young and lacks the large and Greek cultures.
body of generally accepted principles and theories that The Egyptians left evidence in their sculpture and
characterize older disciplines such as physics and wall paintings of a well-developed appreciation of ani-
chemistry. It has been given various definitions, in- mals and plants, and the Greeks gave us the first for-
cluding scientific natural history (Elton, 1927), the study mal record of botany and plant ecology. In the fourth
of the structure and function of nature (Odum, 1971), and century B.C., Aristotle wrote about plagues of field
the scientific study of the interactions that determine the mice and locusts, echoing earlier concerns of Egyptian
distribution and abundance of organisms (Krebs, 1978). and Babylonian societies about outbreaks of animal
The choice of definition is not critical as long as it is pests. There is evidence in these early writings of a
remembered that the focus of ecology is on the inter- recognition of a “balance of nature” and a harmony
relationships between living organisms and both their among plants, animals, and humans. In approximately
biotic (living) and abiotic (nonliving) environment. 300 B.c., Theophrastus, a pupil of Aristotle, demon-
The history of ecology dates from approximately strated a clear understanding of habitat selection by
the turn of the 20th century, but the historical roots of different plants, the effects that different habitats have
the subject are much older. All animals, if they are to on the growth and morphology of plants, and the im-
survive, must have an operational “knowledge” of plications of these effects for the utilization of trees by
those ecological relationships that affect them. The people (Kormondy, 1965):
hunting success of early humans suggests that they
The differences between trees of the same kind have
must have known a lot about the behavior, food re-
already been considered. Now all grow fairer and are
quirements, and habitat of both their prey and animals
that preyed on them. After the Pleistocene extinction, more vigorous in their proper positions; for wild, no
knowledge of plant ecology would have increased con- less than cultivated trees, have each their own posi-
siderably as they changed from hunters to gatherers. tions: some love wet and marshy ground, as black
Further improvements in knowledge of life histories poplar, abele, willow and in general those that grow

i) wm
26 CHAPTER3 — Ecology and the Ecosystem Concept

by rivers; some love exposed and sunny positions; providential ecology (a concept in which nature is
some prefer a shady place. The fir is fairest and thought to be designed to benefit and preserve each
tallest in a sunny position, and does not grow at all in species) and the balance of nature by the ideas of natural
a shady one; the silver-fir on the contrary isfairest in selection and the struggle for existence (Egerton, 1968).
a shady place, and not so vigorous in a sunny one. The development of ecological thought was further
stimulated by problems of agricultural pests and dis-
Thus, there is in Arcadia, near the place called
ease vectors, and by the end of the 19th century, the
Krane, a low-lying district sheltered from wind,
recognition and description of plant and animal com-
into which they say that the sun never strikes; and munities had gained considerable momentum.
in this district the silver-firs excel greatly in height A growing realization of the interrelatedness of liv-
and stoutness, though they have not such close grain ing organisms and their physical environment led to a
nor such comely wood, but quite the reverse—like general recognition of the need for a new branch of
the fir when it grows in a shady place. Wherefore science. The development of this recognition oc-
men do not use these for expensive work, such as curred earlier among soil scientists and silviculturists
doors or other choice articles, but rather for ship- (foresters) than among other natural scientists
building and house-building. For excellent rafters, (Sukachev and Dylis, 1964), but by approximately
beams and yard-arms are made from these, and also 1900, it had become widespread and led to the incep-
masts ofgreat length which are not however equal- tion of the science of ecology. The name for the new
science was coined originally as oecology by two Ger-
ly strong; while masts made of trees grown in a
man zoologists, Reither and Haekle, in 1869 (Kor-
sunny place are necessarily short but of closer grain
mondy, 1965). However, the word did not appear
and stronger than the others.
again until 1895, when a report on ecological plant ge-
ography was published by Warming, a Danish
These early expressions of ecological awareness re- botanist. The word is derived from the Greek words
mained largely undeveloped for many centuries until oikos, meaning “house,” and /ogos, meaning “the study
the growing sophistication of agriculture led to an in- of.” Development of ecology gathered momentum in
crease in practical knowledge of the relationships be- the first two decades of the 20th century, but most of
tween plants and animals and_ between living the major advances have been achieved in the past 60
organisms and their physical environment. European years. Only in the past 30 years has it been recognized
monastic settlements in the Middle Ages contributed as a major branch of biological science.
to the growth of biological knowledge of forests, and Use of the word ecology was rare outside scientific
this period saw the development of an interest in circles as recently as 30 years ago. Today it is a house-
human populations (the beginnings of the science of hold word. This semantic revolution has unfortunate-
demography) and natural history. The 18th century ly been accompanied by frequent misuse and
witnessed a general rejection of the view that the misinterpretation of the word. It is commonly and
human situation could be perfected by social institu- wrongly used as a synonym for preservation and as a
tions alone. This view was replaced by a recognition banner for the “environmental movement.” As a sci-
that humans, animals, and plants all are held in check ence, ecology is amoral; it makes no moral or value
by the same general processes. Thomas Malthus in his judgments about the desirability or lack of desirability
famous Essay on Population (1798) told the world ex- of any ecological condition or event. It merely de-
plicitly that humans, like any other organism, are ulti- scribes the ecological characteristics or consequences
mately resource limited. From this essay, one can trace of things. Judgments as to whether the characteristics
the development of Darwin’s theories of evolution and of a particular ecosystem or ecosystem condition are
natural selection and much of the resulting conceptual any better or worse than those of any other ecosystem
base of biology. or condition must be made by society according to the
Malthus’s essay, which has been referred to as “cer- prevailing value systems. Similarly, judgments as to
tainly one of the most influential in the history of whether the ecological consequences. of particular
western thought” (McNaughton and Wolf, 1973), events are good or bad, desirable or undesirable, can
stimulated interest in the mathematical aspects of bi- be made only using the prevailing measures of socio-
ology. It also led to the replacement of the ideas of economic value. Ecological information is—or should
 Subdivisions ofEcology and the Concept ofLevels of Biological Organization 27

be—used as an aid to the formulation of such value Level-of-biological-

judgments, but most often they are made on the basis organization hierarchy Related scientific disc

of social rather than ecological criteria. Biosphere Landscape ecology

Ecosystem* Ecosystem ecology
Community | > Ecology Community ecology
Population Population ecology
3.2 Subdivisions of Ecology me g >, | Organism* | }
p=
Autecology, behaviora
and the Concept of Levels of =
2)
5 &
a=
Organ system
Ore Anatomy
ecology

Biological Organization
g gl} 2) 2 EO AT ee ge ee
5 2//-2/= Tissue Morphology
ZS) Sls cei 1 Physiolog
|| 2||3] | 2-5,
Z| 2 Subcellular component — Cellular biologsy
There is frequently some confusion over just what
= = & S ~=Molecules in biological ---— Biochemistry
ecology covers that is not already covered by other, M\e};}<} systems
more traditional disciplines such as botany, zoology,
soil science, meteorology, and microbiology. To gain Atoms and molecules Chemistry
an understanding of the unique contribution of ecol- Energy and matter =e Ehysics
Numbers and space Mathematics
ogy, it is necessary to understand the concept of /evels
of biological organization and the related concept of *True levels of biological integration

levels of biological integration (Rowe, 1961) that is dis- Figure 3-1 Levels of biological organization and the re-
cussed in section 3.4. lated biological disciplines. The diagram shows three hierar-
When all the knowledge of biology and closely re- chies of knowledge that lead from the fundamental subjects of
lated sciences is considered, we find that it can be mathematics, physics, and chemistry. These three hierarchies
arranged in a hierarchy of increasing biological com- coalesce at the ecosystem level. Details are given only for the
plexity. In this hierarchy, each level incorporates biological hierarchy. True levels of biological integration (ac-
knowledge of the level below. The lowest level is that cording to Rowe, 1961) are marked with an asterisk.
of molecules in biological systems, which is succes-
sively based on the nonbiological levels of atoms and
In other words, ecology is concerned with levels of
molecules (chemistry), energy and matter (physics),
organization from the organism upward. In common
and space and numbers (mathematics). Various writers
with other biological disciplines, ecology knows no
have identified successively more complex biological
taxonomic boundaries. Just as the sciences of physiol-
levels of organization: subcellular components, cells,
ogy, anatomy, and genetics can be applied to ferns and
tissues, organs, organ systems, and individual organ-
trees, insects and mammals, so one can study the ecol-
isms. These are shown, together with the traditional
ogy of mosses and shrubs, bacteria and birds. Ecology
scientific discipline that is most commonly associated
is a functional and organizational division rather than
with them, in Figure 3-1. Similar hierarchies could
a taxonomic division of scientific knowledge.
presumably be drawn up for the atmospheric and earth
The science of ecology has several subdivisions,
sciences, which emanate, together with the biological
each of which is associated with a different level of bi-
hierarchies, from the level of atoms and molecules.
ological organization.
It is apparent from Figure 3-1 that almost all of
the traditional subdivisions of biology are concerned
1. Study of the life history and the response to its envi-
with levels of organization at or below the individual
ronment of a single individual or species is frequently
organism. Ecology, however, is concerned with
referred to as autecology, for example, the life history
of an eagle, the food requirements of a deer, or the
1. individual organisms in relation to other organ- temperature tolerance of Douglas-fir seedlings.
isms and the nonliving environment
2. Study of the abundance, distribution, productivity,
2. groups of organisms of the same species (popula- and/or dynamics of a group of organisms of the
tions) same type (a single-species population) would be
3. natural assemblages of populations of different classified as population ecology, for example, an in-
species (communities) vestigation of competition for light and nutrients
4. entire natural systems composed of communities in a pine plantation, the role of disease in control-
and their physical environment (ecosystems) ling the numbers ofinsects on a tree, or the rate of
28 CHAPTER 3 Ecology and the Ecosystem Concept

growth and mortality of the individuals in a salmon Whittaker suggested that “an ecosystem is a func-
population. tional system that includes an assemblage of interact-
ing organisms (plants, animals and saprobes) and their
3. Studies involving the description and quantifica-
tion of some aspect of a natural assemblage of dif- environment, which acts on them and on which they
ferent species of organisms are classified as act” (1975). Odum (1971) proposed a longer but more
community ecology, for example, the study, classifica- explicit definition:
tion, and mapping of forest plant associations or
“Any unit that includes all of the organisms (i.e.,
forest types; the description of the animal commu-
the community) in a given area interacting with
nity in a small lake; or a study of the change in
plant and animal communities in an area over the physical environment so that a flow of energy
time. Population ecology and community ecology leads to a clearly defined trophic structure, biotic di-
are sometimes referred to collectively as synecology. versity, and material cycles (i.e., exchange of mate-
4. Studies involving both the biotic community and rials between living and non-living parts within
its abiotic environment are classed as ecosystem ecol- the system) is an ecological system or ecosystem”.
ogy. Such studies may be primarily descriptive, as
It is clear from these definitions that the term
in the classification and mapping of different types
ecosystem is more of a concept than a specific physical
of ecosystems. They can also be functional, such as
entity—a concept with five major attributes.
a study of the interrelationships between the plant
community and the soil or the way in which en- 1. The attribute of structure (Figure 3—2). Ecosystems
ergy and nutrients are distributed in and move are made up of biotic and abiotic subcomponents.
through an ecosystem. At the very least, a terrestrial ecosystem must have
green plants, a substrate, and an atmosphere, and
in most ecosystems there must be an appropriate
3.3. The Ecosystem Concept' mixture of plants, animals, and microbes if the
ecosystem is to function. ‘Terrestrial ecosystems
The word ecosystem has been used several times al-
normally consist of a complex biotic community,
ready, but its importance in forest ecology warrants
together with soil and atmosphere, a source of en-
further discussion. It is the central concept of this
ergy (generally the sun), and a supply of water.
book, of ecology, and of the biologically rational man-
agement of forest resources. Our long-term survival 2. The attribute of function, the constant exchange of
on this planet will depend in no small part on our ac- matter and energy between the physical environ-
ceptance of the concept of ecosystem as a guide to our
future conduct. Plant community
The term ecosystem was suggested by an English
ecologist, Tansley. He defined it as including “not only
the organism-complex, but the whole complex of
Biotic Animal community
physical factors forming what we call the environ- community
ment” (Tansley, 1935). There are several alternative
definitions, all proposed by American ecologists. Lin-
deman proposed that “an ecosystem is any system Ecosystem Microbial community
composed of physical, chemical and_ biological
processes active within any space-time unit” (1942).
Atmosphere

‘One ofthe important earlier collections ofpapers concerned with the appli- Abiotic
cation of the ecosystem concept in natural resource management can be environment

found in van Dyne (1969). The chapters by A Lajor, Bormann, and Likens; Soil or geological
substrate
Schultz; and Bakuzis and Cooper are recommended. For a history of the
ecosystem concept, see Golley (1993); see also Vogt et al. 1997. Figure 3-2 The structural components of an ecosystem.
 The Concept ofLevels of Biological Integration 29

ment and the living community. Because living and ties, although their size may sometimes be difficult to
nonliving things both are composed of energy and define. A flock of birds or a school of fish constitute
matter and because it is often difficult to define easily identifiable populations, but their spatial bound-
when organic material is alive and when it is dead, aries may be difficult to establish because the space
there are some merits in looking at an ecosystem in that they occupy may change periodically. Similarly,
terms of a physical—chemical system. Within this identification of a population of spruce trees in the
system, there is a constant exchange of matter and northern boreal spruce forest or of a biotic community
energy between different components, some of in the open ocean may require a somewhat arbitrary
which have the characteristics of life and some of definition of spatial boundaries. However, despite
which do not. This way of looking at ecosystems in these problems, the focus of attention in the terms
no way denies the importance of the more tradi- population and community is clearly on a real physical
tional biological and genetic view of life; it is com- entity that often can be defined easily. The biotic com-
plementary to it. munity of a clearcut or of a wet valley bottom can be
3. The attribute of complexity, which results from the readily observed, and its spatial limits can be described.
high level of biological integration that is inherent The term ecosystem, however, focuses on the structure,
in an ecosystem. All events and conditions in the complexity of organization, the interaction and in-
ecosystems are multiply determined. They are terdependence, and the functioning of the system, not
therefore difficult to predict without a consider- on the geographical boundaries of the system.
able knowledge of the structure and functional
processes of the system.
4. The attribute of interaction and interdependence. So
3.4 The Concept of Levels
complete is the interconnectedness of the various of Biological Integration
living and nonliving components of the ecosystem
Before leaving the ecosystem concept, let us refer
that a change in any one will result in a subsequent
change in almost all the others. The extent and briefly to Figure 3.1, the levels of biological organiza-
tion. Hierarchies that include such terms as tissues,
completeness of this interaction and interdepen-
organs, populations, and communities have been criti-
dence led some of the earlier ecologists to think of
cized (Rowe, 1961) on the grounds that although such
physical examples of the ecosystem concept (e.g., a
levels of biological organization have merit as objects
hectare of forest, an agricultural field, a small pond)
of study and focus for the associated biological subdis-
as a sort of superorganism. This view has been re-
cipline, they do not represent true /evels of biological in-
jected because although there are certain parallels
tegration. The only true levels in biology are cells,
between an individual and an ecosystem, the differ-
individuals, and ecosystems.
ences are too great to make the concept useful.
Rowe (1961) defined a true level of biological inte-
5. The attribute of temporal change. Ecosystems are gration as one that is “the total environment of all the
not static, unchanging systems. In addition to the levels of biological organization below, and a structur-
continuous exchanges of matter and energy, the al and functional component of the next level above.”
entire structure and function of an ecosystem un- He then noted that accurate prediction of events or
dergoes change over time. conditions at any one level of biological organization
can be made only on the basis of knowledge of the
The importance of the ecosystem concept lies in its next true level of biological integration above. For ex-
explicit recognition of complexity, interaction, func- ample, the future development of a cell cannot be pre-
tional processes, and change over time. Its weakness dicted merely from knowledge of the tissue in which it
lies in the difficulty of using the concept for the identi- is found or of the organ in which the tissue is located.
fication, mapping, description, and study of specific Only by knowing the physiological condition of the
ecosystems because of its failure to define their physi- entire organism can a reliable prediction be made con-
cal boundaries. An individual organism is a tangible cerning all aspects of any one cell in that organism. An
entity. It has a clearly defined physical size. Popula- organism is therefore the next true level of integration
tions and communities are also spatially defined enti- above a cell. The fate of an individual organism cannot
30 CHAPTER 3 Ecology and the Ecosystem Concept

be predicted on the basis of a knowledge of the popu- and exchanges of matter and energy between the com-
lation to which it belongs or from an understanding of ponents of the ecosystem and of describing the tempo-
the biotic components of the community to which that ral and spatial variations in ecosystems: a focus on the
population belongs. Only on the basis of knowledge of structure, functioning, and variability of the system.
the ecosystem will all the relevant antecedents that af- By applying the ecosystem concept at all levels of
fect that individual be identified and considered and a ecology, Rowe’s objections to the population and com-
reliable prediction concerning that individual be ob- munity levels are entirely overcome and population
tained. Ecosystem is the only true level of biological and community ecology attain their full potential. By
integration above the individual. considering individuals, populations, or communities
The importance of this conclusion for ecology and within an ecosystem context, we identify the total set
forest management is that it points out the danger of of factors that determine the abundance, distribution,
trying to predict population or community events and and productivity of the biological object or system of
conditions on the basis of knowledge about these lev- interest. Natural phenomena exist and/or occur in
els of organization alone. This in no way denies the natural landscapes: three-dimensional systems that
value of population and community ecology. It is vital have climatic, geological, and biological components
that processes that occur at the population and com- and that vary in a fourth dimension—time. Conse-
munity levels be understood. However, multiply de- quently, we are unlikely to be able to understand and
termined events in ecosystems cannot be accurately predict these phenomena unless we consider them as
predicted on the basis of causal mechanisms hypothe- components of landscapes and of ecosystems.
sized from studies conducted exclusively at the popu- Climate sets the overall framework for the biotic
lation or community level. For example, predictions as potential of an area, for example, tropical rain forest,
to the future growth of a Douglas-fir plantation will desert, boreal forest, or tundra. Within a climatic area,
not be reliable if they are based only on a knowledge the landform (the shape of the landscape, e.g., ridge
of the inherent growth abilities of this species, how the top, midslope, lower slope, valley bottom), the origin
trees compete with each other, and the trees’ suscepti- of the surface materials (e.g., glacial deposits, water-
bility to natural diseases and enemies. A knowledge of sorted materials, wind-transported soils, organic mate-
the other plants, animals, and microbes that enhance rials), and the mineralogy of the inorganic soil matrix
or impede the growth of the plantation will improve (e.g., slowly weathering, nutrient-poor geological ma-
the prediction somewhat, but only a recognition of the terials; rapidly weathering, nutrient-rich materials)
entire variety of biotic, climatic, hydrologic, and modify the climatically set potential by determining
edaphic (soil) factors that affect that population of the suitability of the substrate and the availability of
Douglas-fir trees is sufficient to give reliable predic- moisture and nutrients to plants and animals.
tions as to their future. This does not suggest that the The attainment of the physically determined biotic
silviculturist has to switch the focus of attention from potential depends on the types of organisms available to
the population of trees to every other factor of the colonize and occupy the area: their genetic constitution
ecosystem. It merely suggests that the population be (which defines their adaptations to, requirements for,
examined with an explicit recognition of the effects of and tolerances of the physical features of the environ-
the other factors of the ecosystem on the population. ments) and how they interact with the other organisms
It was suggested nearly 50 years ago (Evans, 1956) of the area. The occurrence, distribution, abundance,
that the basic unit of ecology should be the ecosystem and productivity of any one organism thus are deter-
rather than the individual, the population, or the com- mined by the past and present climatic, geomorpholog-
munity. If this is accepted, then autecology becomes the ical (landform), mineralogical, edaphic (type of soil),
study of the interactions of a single individual with its and biotic characteristics of the area in question.
biotic and abiotic environment: an ecosystem with the The ecosystem concept is the single most impor-
organism being studied at its center. Synecology be- tant concept in ecology. It is the essential basis for sus-
comes the study of a group or groups of organisms re- tainable forest management and is a prerequisite for
acting with their biotic and abiotic environment: an the long-term survival of the human species on this
ecosystem with the population or community at its cen- planet. Humankind must learn that we are but one
ter. Ecosystem ecology is then left with the important species in the world ecosystem; that our activities alter
jobs of studying the interactions, interdependencies, the world ecosystem that is our “life-support system.”
 Ecosytem Change: A Key Attribute 31

Large-scale alterations of atmospheric chemistry (e.g., or self-caused, mechanisms. Other mechanisms that
acid rain, atmospheric CO,) will not only affect plants contribute to change involve disturbances that are either
and other animals but also affect us because we are a not related to or not initiated by the plant community
part of the ecosystem. Foresters must recognize that (e.g., fire, wind, flooding, erosion, climate change).
forest management affects the forest ecosystem, not These are allogenic mechanisms. A third category of
just the tree crops, which are merely a component of mechanism is biogenic, these include insect epidemics,
the ecosystem. Ecologists who are concerned with diseases, and invasions of nonnative species.
subsets of ecosystems must always remember the The relative importance of autogenic, allogenic,
ecosystem and landscape setting of the phenomena in and biogenic mechanisms of succession changes from
which they are interested. one type of ecosystem to another and from one time to
These ideas are illustrated in Figure 3-3, which shows another. Allogenic or biogenic mechanisms are gener-
both the object of interest and the degree to which other ally responsible for disrupting the sequence of commu-
factors are considered in “traditional” approaches and the nity replacements and the environmental alterations
ecosystem approach to the subdisciplines of ecology. that are the result of autogenic processes of change;
they are therefore referred to as disturbance mecha-
nisms, or simply disturbances. Fire, wind, and insect
3.5 Ecosystem Change: A Key Attribute epidemics, for example, frequently kill trees, moving
the ecosystem condition to that of an earlier stage of
In common with all living systems, ecosystems are not ecosystem development and initiating a new autogeni-
static, unchanging, steady-state systems (Botkin, cally driven succession of plant communities and their
1990). There is continual change in their structure, associated animal and microbial communities.
function, degree of complexity, and the interactions Understanding forest ecosystems and managing
between their components. In some types of forest, the them to sustain a desired seral condition, or a desired
rate of change is slow; in others, it is rapid. In some, mosaic of seral stages across the landscape, requires
the change is fairly continuous and small scale. In oth- that we understand how and why ecosystems change.
ers, it is large scale and either frequent or infrequent. Understanding the changes in the physical and chem-
The topic of ecosystem change is explored in Chapter ical characteristics of the ecosystem requires that we
17, but it is important to introduce some of the con- understand the ecological role of light (solar radia-
cepts at this stage to help you appreciate the signifi- tion), temperature, wind, soil and water, and the
cance of the subjects discussed in Chapters 4 to 16. processes of nutrient cycling and nutrient and organic
Ecosystems change for a variety of reasons: change matter accumulation and how these are influenced by
in climate, change in soil conditions, change as a result the growth of plants and the action of animals and mi-
of interactions between the members of the living crobes. Understanding how and why plant, animal,
community, and change caused by disturbances such as and microbial communities change over time requires
wind and fire. The process of change after disturbance knowledge of the processes of population ecology and
is called ecological succession. Over time, there is a series the species interactions defined by community ecol-
of living communities—plants, animals, and mi- ogy. Understanding community and _ population
crobes—that successively occupy and are replaced on a processes and how these alter the physical and chemi-
particular area of land, with accompanying changes in cal environment requires knowledge of production
soil and microclimatic conditions. The entire sequence ecology and biogeochemistry. The purpose of Chap-
of community and ecosystem change is called a sere, ters 4 to 16 thus is to prepare you for Chapter 17.
each stage in the sequence characterized by a different Understanding ecosystem change is essential for
community being called a sera/ stage (Figure 3-4). the design of sustainable forest management systems.
Some of the mechanisms of successional change are Figure 3-4C shows some alternative successional ob-
caused by the plants that participate in the succession: jectives by which a forest might be managed to main-
invasion of new species, alteration of the physical envi- tain any one seral stage or any desired sequence of
ronment and the availability of resources (light, mois- seral stages.
ture, nutrients), and displacement of the species that
were occupying the site by one or more of a variety of 1. Large-scale disturbance, exposed mineral soil: re-
between-species interactions. These are called autogenic, placement of climax hemlock forest by an alder stand
 SUBDIVISIONS ; ECOSYSTEM VIEW _
OF ECOLOGY Po
oe
gS =
=
——
Vp ae a
Ecosystem LHI
ea ey WY
Ry
ecology (e.g., RI: es RS /
study of biomass
production and
nutrient cycling)

ZN AMSA I So RE WTA AUSSI Sh NS TG RN

Ecosystem

Community
ecology (e.g.,
study of the
pee effects of animals
on competition ‘
between different
. /
plant species) /

rey HY ‘Zz —~ 4 — A Pp ed
WATS re WAAR a Kv Fe NS PA TGRe A
Community

Population
ecology (e.g., study,
of the competition jy de
between members Wf
ofa plant
population)

ci
ci: ! dae ~ c
Wir - at "HAG 4) TOS a Oram
WANTS Se WIN ESS ie LE WSS PAYS Saw EN
Population

<==
A ANN <
‘TTY (SSS Autecology (e.g.,
; ° i vf
aN < EN studyof of the light é Sth
eo / LIME ,
requirements
m z
of a
VEU {fy
if, YoU lean
d/l
he
ie
a species of plant Var: /

i:
fh
Individual GMS ay: CE, WON PG MGS AN
Individual

Figure 3-3 “Traditional” and ecosystem views of the subdivisions of ecology. In the traditional
view, the focus is on the entity(ies) identified at the particular subdivision (e.g., the population at
the population level). In the ecosystem view, the major focus is the same, but at every level the relation-
ship between the object of study and the other components of the ecosystem is given explicit considera-
tion. Note that the two views are identical at the ecosystem level but differ at the other levels. Organisms,
physical factors, and interactions that are shaded in the ecosystem view on the right are explicitly studied.
Those that are not shaded are explicitly recognized as potentially important determinants but are not
necessarily studied. Their role in modifying the relationship being studied is frequently investigated.
 Ecosytem Change: A Key Attribute 33

Late seral
s Early seral Pioneer Mid seral Late seral ,; Climax herb/
conifer herb/shrub hardwood conifer conifer ishrub/byrophyte
| woodland

Previous forest floor New forest floor

Cc |
|
Douglas-fir |
|
|
| |
Herb/shrub Western Hemlock
1

|
Allogenic/biogenic
disturbance
Autogenic
succession

Figure 3-4 Diagram of the pattern of autogenic ecosystem change, which can be seen in
some types of ecosystem at low elevation in coastal British Columbia. The sequence rarely
continues uninterrupted from start to finish because of natural or human-related disturbance. If it
did, then it might take as much as 1,000 years to complete. (A) The possible pattern of plant com-
munities that successively occupy the site over time. (B and C) Diagram of the accumulation of for-
est floor mass (B) and rates of above-ground biomass production (C). (D) Alternative successional
management strategies (see text for explanations).

. Medium to large-scale disturbance, less distur- 6. Medium-scale disturbance of a Douglas-fir stand

bance of mineral soil: replacement of climax hem- with only moderate soil disturbance: replacement
lock forest by a Douglas-fir stand of Douglas-fir by Douglas-fir. Repeated rotations
Small-scale disturbance, little soil disturbance: re- of Douglas-fir may lead to an increase in root-
placement of mature hemlock stand by a young decay fungi and a decline in nitrogen availability in
hemlock stand comparison with a succession that includes a red
alder stage
Successional recovery over the subsequent, for ex-
ample, 80 to 100 years . Medium- to large-scale disturbance of a Douglas-fir
stand with mineral soil exposure, leading to an
Large-scale disturbance of alder stand with min-
alder stand
eral soil exposure: replacement of alder by alder
(the second and subsequent generations of alder Successional recovery (followed by disturbance
may grow poorly because of soil acidification pathway 7) that permits an alder stage followed by
caused by the alder) a Douglas-fir stage. This reduces the problems of
34 CHAPTER 3 Ecology and the Ecosystem Concept

root pathogens and decline in alder growth as a re- no basis for value judgments and is only one, albeit of
sult of soil acidification; the nitrogen fixation by preeminent importance, of several sources of information
contributing to socioeconomic decisions.
the alder improves the growth of the Douglas-fir;
Within a hierarchy of systems of increasing biological
the alder may be harvested as a crop
complexity, ecology begins where the more traditional
9, Successional recovery (followed by disturbance branches of biology end: at the individual organism. Sub-
pathway 1) that allows alder, Douglas-fir, and divisions of the science deal with individuals, popula-
hemlock stages in normal successional sequence; tions, communities, and ecosystems, but the full potential
the alder and the Douglas-fir can be harvested in of ecology is attained only when the ecosystem concept is
this sequence, as well as the hemlock used, irrespective of the complexity of the ecological sys-
tem in question. The term ecosystem provides the concep-
The choice of which strategy to apply is not a tual basis for ecology, but its shortcomings have led to the
use of the term ecosystem type when identifying and map-
purely scientific question. It involves value judgment
ping individual ecosystems in the field.
about which ecosystem values and conditions are to be Forest ecology is merely the application of general
sustained. However, scientific understanding of suc- ecology to a specific type of ecosystem: the forest. Forest
cession is essential for the design of forest manage- ecology involves studies at the individual, population,
ment systems that will achieve society’s successional community, and ecosystem levels, but such studies should
objectives. This knowledge is also vital to our under- always involve the ecosystem concept. Although forest ecol-
standing that some of the methods of forest harvesting ogy has traditionally been thought of as being concerned
and management suggested by the public may not in largely with the community level of organization, its full
fact sustain desired values in some types of forest contribution can be realized only when all levels of biologi-
ecosystem. This is because the suggested management cal organization within the forest ecosystem are studied.
will not produce the seral conditions required by some
of the values that are to be sustained. As we will see
throughout the book, the productivity, biodiversity, TAKE-HOME MESSAGE
and other characteristics of most forests are the result Forests are ecosystems, not populations of commercial
of periodic disturbance and the subsequent sequence tree species or even communities of living organisms.
of seral stages. Management that attempts to maintain They are complex biological, physical, and chemical sys-
a forest in any one seral condition over many tree gen- tems that are constantly changing in structure, species
erations will change the character of a forest that has composition, and productivity over time.
historically developed through several seral stages Prediction in complex systems is difficult. Accurate
under the influence of natural disturbance regimes. prediction of the consequences of forest management re-
quires that we understand how ecosystems function and
This may result in undesirable ecosystem conditions
change over time in the absence of management and how
and events in such forests. It would not do this in a
management affects the components, functional process-
forest ecosystem that is adapted to being held in one es, and change of forest ecosystems. This requires that
seral condition by natural disturbance. we adopt an ecosystem approach to both forest manage-
ment and the ecological science that supports it.
SUMMARY
Ecology is one of the more recent major developments in STUDY QUESTIONS
the biological sciences, although its subject matter has 1. Why must ecosystem management replace timber
been of vital interest to humans throughout their evolu- management?
tion. There has been a remarkable increase in scientific,
bh . Why is the ecosystem concept so important in both
political, and public interest in ecology during the past
two decades. This has been accompanied by some confu- ecology and resource management?
sion over the role of ecology in socioeconomic decision 3. What is the significance of the concept of “true levels
making. Ecology has been advanced by some as the basis of biological integration”?
on which socioeconomic decisions should be made. This 4. Why is “change over time” such an important at-
is incorrect because the science of ecology itself provides tribute of ecosystems?

34
 Production Ecology
The Transfer and Storage of Energy
in Ecosystems

4.1 Introduction The origins of life on earth are thought to have re-
sulted from the action of solar energy on the simple
Energy is what life is all about, and everything about inorganic chemicals that were present on the sterile
life is associated with energy. Organisms are accumu- surface of the young planet. Organic molecules were
lations of energy. Without a continuing supply of en- produced that grew and replicated themselves by coa-
ergy, they die, and the energy accumulation is lescing with and using some of the chemical bond en-
dispersed as their remains decompose. All organisms ergy contained in other such molecules. Different
require a source of energy. ‘Io reproduce, they require combinations of molecules resulted, and eventually
growth; to grow, they require new energy; to acquire pigments evolved that had the characteristic of ab-
new energy, they must do work; and to do work, they sorbing solar radiation. This was stored as chemical
must expend energy. Merely to stay alive requires en- bond energy by synthesizing additional complex or-
ergy. The study of the energy relationships (inputs, ganic molecules from simple inorganic substances in
storage, transfer, and outputs) of ecosystems is called the environment. The following 2 billion years of evo-
production ecology. lution has elaborated a myriad of variations on this
Both evolution and ecology can be viewed from an basic theme, but fundamentally the nature of life has
energy perspective. The evolutionary “struggle for not changed. Living organisms such as black spruce,
existence” between organisms is basically a struggle to bears, or bacteria are nothing more than highly orga-
obtain sufficient energy in usable form to sustain and nized associations between energy (largely solar ener-
reproduce life. The abundance, productivity, and dis- gy) and inorganic chemicals from the soil and
tribution of organisms (three central themes in popu- atmosphere. This viewpoint is substantiated by the ex-
lation and community ecology) all are ultimately ample of a forest fire, in which the solar energy ab-
determined by the availability of energy. However, a sorbed and stored by the plants is released as the heat
large number of other factors are involved in deter- and light of the flames and the chemicals involved in
mining this availability, and in many ecosystems, fac- capturing and storing the energy are released, often in
tors other than energy play the primary role in their original form, as atmospheric gases or chemicals
determining the distribution and abundance of or- in the ash.
ganisms. Thus, energy availability is often not the Life in general—and the forest in particular—is
proximate factor, and the science of ecology involves merely a complex physical-chemical organization
far more than merely a consideration of energy. that has evolved to become self-perpetuating. Al-
These points notwithstanding, energy remains the ul- though such a fundamental, functional view of life
timate determinant of ecosystems and the central lacks the richness and appeal of the more traditional
thread of ecology. taxonomic approach (which considers life as a diverse

3
36 CHAPTER4 — Production Ecology

assemblage of different organisms), it is essential if of organisms into a variety of trophic categories. The
we are to gain a complete understanding of ecosys- major division of this classification is between organ-
tems. The manager of an electrical power station isms that utilize abiotic energy sources (autotrophs) and
would be foolish to try to run the factory without organisms that depend on biotic energy sources
knowledge of electricity. Similarly, ecologists and (heterotrophs).
ecosystem managers (e.g., foresters, agriculturalists, Autotrophic (self-feeding) organisms utilize energy
wildlife and fisheries managers) are unlikely to sources that are independent of the activities of other
achieve their desired management objectives unless organisms, energy that is available in completely abi-
they are familiar with the distribution and move- otic environments. Autotrophs are referred to as
ments of energy that are responsible for the charac- producers (and sometimes as primary producers) because
ter and productivity of the ecosystems under their they produce the high-energy organic molecules that
management. Maximization of forest growth re- provide the energy source for nonautotrophic organ-
quires the maximization of energy capture and the isms. There are two subtypes, photo- and chemoau-
minimization of energy losses from trees. High pro- totrophic organisms.
duction of fish requires knowledge of where they get Photoautotrophs utilize a portion of the electromag-
their energy from, what affects the supply and its uti- netic energy from the sun (sunlight) in the process
lization, and what limits the storage of this energy in of photosynthesis. This process depends on chloro-
fish biomass. Wildlife management requires a de- phyll pigments, which are capable of absorbing cer-
tailed understanding of energy flow in wildlife popu- tain wavelengths of solar radiation and converting
lations. All three of these ecosystem components are them into the chemical bond energy of glucose using
the result of the processes of production ecology. carbon dioxide and water. All green plants are photo-
Their management inevitably involves the manipula- autotrophs.
tion of these processes. Chemoautotrophs obtain their energy from simple
Energy is the driving force of ecosystems, and the inorganic chemicals, for example, by oxidizing sulfide
physiology, anatomy, stature, abundance, behavior, ions to free sulfur, sulfur to sulfate ions, am-
distribution, and ecological role of individual organ- monium ions to free nitrogen gas (N>) or nitrite ions,
isms are largely determined by the manner in which and nitrite ions to nitrate ions, each oxidation being
they satisfy their energy requirements. Understand- accompanied by a release of energy. The contribution
ing an ecosystem requires (1) recognition that func- of chemoautotrophs to total forest ecosystem energy
tional organization is largely a matter of energy flow is generally minor. However, their ecological
transfers and storage, (2) appreciation of the pathways role goes beyond their energy relationships because
and magnitude of energy transfers, and (3) identifica- some of the oxidation products have important envi-
tion of the factors that determine the storage and ronmental effects. For example, sulfur-oxidizing bac-
dynamics of energy within and between various com- teria can lead to the production of sulfuric acid and
ponents of the system. The first two of these are dis- hence acidity in the water draining from piles of min-
cussed in this chapter; the third is the focus of ing waste that contain sulfur, with subsequent leach-
Chapters 5 and 7 to 17. ing of toxic heavy metals into aquatic ecosystems.
Conversion of ammonium ions to nitrate ions can in-
crease the leaching of nitrogen from the soil into
4.2 Sources of Energy water bodies.
for Living Organisms Heterotrophic (other-feeding, i.e., feeding on oth-
ers) organisms are unable to utilize either sunlight or
The relentless quest for energy to maintain and repro- inorganic chemical bond energy. They are dependent
duce life has continued for billions of years. It has re- on energy obtained by oxidizing high-energy organic
sulted in the evolution of a remarkable diversity of molecules such as carbohydrates, fats, and proteins
species of plants, animals, and microbes (approximate- synthesized by autotrophs. Heterotrophs are referred
ly 2 million different species) that differ, among other to as consumers, reflecting their dependence on the
things, in the source of their energy supply. This dif- producers. The four subtypes are identified by their
ference is the basis of a broad functional classification energy source:
 Trophic Chains and Webs, Ecological Pyramids, and Energy Flow Diagrams 37

1. Herbivores (primary consumers) utilize the energy

4.3 “Trophic Chains and Webs,
contained in the organic matter of plants; that is,
they are consumers of green plants.
Ecological Pyramids, and Energy
2. Carnivores (secondary, tertiary, or higher-order con-
Flow Diagrams
sumers) utilize the energy contained in the organic ‘Trophic Chains
matter of animals. Primary carnivores (secondary
consumers) solve their energy needs by eating her- On first inspection by an untrained observer, an
bivores, secondary carnivores (tertiary consumers) ecosystem might appear to be a relatively random as-
by eating primary carnivores, and so on. semblage of organisms. First appearances are often
deceptive, however. There are, in fact, highly ordered
3. Omnivores utilize the energy contained in both
sequences of energy or trophic (e.g., food) dependen-
plants and animals. Their diet will vary according to
cies within an ecosystem. Each organism obtains the
circumstance, and at any one time, an omnivore
energy that it requires for survival, growth, and repro-
may behave as either an herbivore or a carnivore.
duction in a manner characteristic of that species. The
However, omnivores generally do best on a mixed
supply will be from a particular physical source or
diet of plant and animal material. Humans are om-
from a particular type of organism, and each organism
nivores.
is in turn the energy supply for some other organism.
4. Saprotrophs (decomposers, detrivores, or detritus These characteristic sequences of energy transfers are
feeders) utilize the energy contained in dead or- referred to as food chains or trophic chains. Sequences
ganic matter of either plant or animal origin. that involve autotroph — herbivore — carnivore en-
Those that depend on dead plant materials or the ergy transfers are referred to as grazing trophic chains.
feces of herbivores (including organisms that live Trophic chains that commence with dead organic
in the alimentary canal of herbivores) are analo- matter and involve saprotrophs are referred to as
gous to herbivores, whereas those that depend on decomposer or detritus trophic chains. ‘Vhe successive
dead animals or the feces of carnivores (including stages along a trophic chain are called trophic levels.
organisms that live in the alimentary canal of car-
nivores) are analogous to carnivores. Trophic Webs
Ecosystems are complex (the importance of the fact jus-
The assignment of a particular organism to a tifies the repetition of this phrase almost ad nauseam),
single trophic category is not always easy, because and although it is frequently both convenient and useful
trophic dependencies may vary during a life history or to examine the movement of energy along a single
with circumstances. Carnivores may be forced tem- trophic chain, such knowledge generally does not pro-
porarily onto a plant diet at times of shortage of their vide an adequate understanding of the energy relation-
prey, and animals normally thought of as herbivores ships of the whole ecosystem. Most plants are fed on by
may become carnivorous if starving. Tadpoles are a variety of herbivores, and most herbivores feed on a
mainly herbivorous, whereas frogs are mainly carniv- variety of plants (except in cases in which coevolution
orous. Most green plants are entirely autotrophic, but has resulted in very high herbivore—plant specificity).
certain species of algae require certain complex sub- Most herbivores are also prey to a variety of carnivores.
stances that they cannot synthesize themselves and Thus, trophic chains are but components of an intricate
must get from the activities of other organisms; this network, or web, of trophic dependencies. ‘The distinc-
makes them partly heterotrophic. Some plants trap tion between the grazing and detritus trophic webs and
and digest insects and other small animals and are the different trophic levels tends to become less and less
therefore also partly heterotrophic. It has been report- clear the further one proceeds through the web.
ed that some phytoplankton (microscopic aquatic
plants) in far northern lakes are autotrophic during the Ecological Pyramids
summer but heterotrophic during the long arctic
night, when they utilize organic substances dissolved in The full complexity of most of the trophic webs that
the water as a supplementary energy source (Rodhe, have been described has escaped complete identifica-
LORS) tion. However, even such simplified representations
38 CHAPTER4 = Production Ecology

can be confusing unless they are studied carefully. the function of different ecosystems. The shape of the
They often contain too much taxonomic information numbers pyramid will change progressively from the
to permit easy comparisons between the trophic webs early years of a forest plantation through to the ma-
of different ecosystems and yet not enough functional ture tree crop and thus can provide some useful infor-
information to permit an accurate comparison of mation to the forester. However, this type of pyramid
functional processes. This problem can be overcome gives no indication as to the total biomass and its rate
by eliminating the taxonomic information and repre- of accumulation and therefore gives no guidance as to
senting the trophic web as a series of stages or trophic the expected crop yield.
levels in the transfer of energy through the system.
This is done in the form an ecological pyramid. Three Pyramid of Biomass. The pyramid of biomass dif-
types of ecological pyramids can be used. fers from the pyramid of numbers in that the area of
the rectangles is proportional to the biomass or weight
Pyramid of Numbers. ‘The simplest type of eco- of organisms occupying each trophic level. By ac-
logical pyramid is a pile of rectangles representing counting for size, the biomass pyramid gives a more
successive trophic levels in the trophic web, with the accurate picture of the distribution of energy through
area of each rectangle proportional to the number of the ecosystem. This facilitates comparison of the bio-
organisms in the corresponding trophic level. For mass present in different trophic levels and in different
some ecosystems, the result is a pyramid-shaped fig- types of ecosystem (Figure 4-2). However, the bio-
ure, with fewer and fewer organisms as one proceeds mass pyramid is not much better than the pyramid of
through the trophic web. However, this is true only numbers as a representation of ecosystem function
where the size of the organisms either remains con- and the dynamics of energy in the system, because no
stant or increases up the pyramid. In a forest ecosys- allowance is made for the longevity (life span) of or-
tem, where the producers (trees) are very large relative ganisms. One hectare of fast-growing but short-lived
to the consumers (e.g., defoliating insects), or in the grass will have a much smaller biomass of producers
situation in which primary and secondary consumers than | ha of slow-growing but long-lived trees, yet the
are successively smaller (e.g., a plant—parasite—hyper- grassland ecosystem can support a larger biomass of
parasite trophic web), the numbers pyramid may take consumers than the forest.
on a different shape (Figure 4-1). The biomass pyramid of a forest will change
Although the numbers pyramid is useful in de- through the life of the tree crop and will provide use-
scribing the relative abundance of different types of ful information on the size of the crop that can be har-
organisms, it tells us little about overall ecosystem vested at any particular time. It tells us nothing,
function and is therefore of little help in comparing however, about how rapidly biomass is accumulating
in each trophic level and therefore provides no guid-
ance as to the long-term yield that could be harvested.
Trophic level

2° carnivores
1° carnivores
Herbivores
Producers

(a) (b) (c)

(d)

Figure 4-1 Examples of the pyramid of numbers (after Figure 4-2 Examples of the pyramid of biomass. The num-
Odum, 1971). The shape of the pyramid varies according to the bers refer to grams (dry weight) of organisms per square meter
type of ecosystem and can change with time, both seasonally (after Odum, 1971). These pyramids are much less variable in
and annually. (A) Temperate grassland in summer (numbers per shape than the numbers pyramids. For most ecosystems, they
0.1 ha); producers are small. (B) Temperate forest in summer are upright, although they can be inverted in aquatic ecosys-
(numbers per. 0.1 ha); producers are large. (C) Plant/parasite tems. (A) Tropical rain forest. (B) Wisconsin Lake. (C) Aban-
trophic web (hypothetical). (From Fundamentals of Ecology, doned agricultural field. (D) English Channel. (From
Third Edition, by Eugene P. Odum. Copyright © 1971 by Fundamentals of Ecology, Third Edition, by Eugene P. Odum.
W. B. Saunders Company. Reprinted by permission of CBS Copyright © 1971 by W. B. Saunders Company. Reprinted by
College Publications.) permission of CBS College Publishing.)
 Trophic Chains and Webs, Ecological Pyramids, and Energy Flow Diagrams 39

Although the biomass pyramid tells us a lot about the combination of all three types of pyramids gives an even
mventory of an ecosystem, it tells us little about its more complete representation of the ecosystem. Unfor-
productivity. It provides information that is of more tunately, there have been few studies in which all three
value for short-term ecosystem exploitation than for types of information have been obtained.
long-term ecosystem management. Finally, although
the biomass pyramid retains its upright form better Variations in Ecological Pyramids
than does the numbers pyramid, it can still be inverted
in certain types of ecosystem; for example, there is a Grazing Versus Detritus Trophic Webs. The first
greater weight of herbivorous plankton than of phyto- essential modification of simple ecological pyramids is
plankton in oceanic ecosystems (Figure 4—2). the explicit identification of the energy flow to sapro-
trophs: the separation of energy flow into grazing and
detritus trophic webs. In many ecosystems, there is a
Energy-Flow Pyramid. The problems with the
greater flow of energy from producers to the detritus
numbers and biomass pyramids can be overcome if we
trophic web than from producers to the grazing
make the area of each rectangle proportional to the
trophic web. For example, in a densely stocked, inten-
flow of energy through that trophic level. This type of
sively managed spruce plantation, there will be very
pyramid should always be upright because of the sec-
little herbivore biomass and grazing trophic web en-
ond law of thermodynamics.' The loss of energy as it
ergy flow, but there will be a large flow to the detritus
is transferred from one trophic level to the next means
trophic web in the form of litterfall. Figure 4-4
that there can never be more energy flow at any troph-
ic level than at the next lower level (except on a tem-
porary basis), with the one exception of energy
Biomass pyramid, kcal m~ Energy flow pyramid, kcal m7 yr“!
imports, discussed below.
1. Pine Forest
Figure 4-3 gives some examples of energy-flow
Detritus Grazing
pyramids. Such pyramids provide information from food web food web
which rates of biomass accumulation in different
trophic levels can be deduced, but even the energy-flow
pyramid fails to give a complete description of ecologi-
cal energetics, as we shall see shortly. The combination
of energy and biomass pyramids is obviously essential
for an understanding of ecosystem energetics, and the

. . =?
Biomass Pyramids, kcal m~
Detritus Grazing
web web

Detritus food web accounts for 99%

VLA
Detritus food web accounts for 10%
Figure 4-3 Examples of the pyramid of energy flow (kcal of heterotrophic energy flow. Very _ of heterotrophic energy flow. Large
m ? yr ') (after Odum, 1971; Kozlovsky, 1968). These pyra- small grazing food web biomass. grazing food web biomass.

mids always retain their upright shape as long as there is no net (b)
import or export of organic matter into or out of the ecosystem.
(A) Silver Springs (fresh water spring in Florida). (B) Salt marsh Figure 44 (A) Comparison of biomass and energy-flow pyra-
in Georgia. (C) Lake Mendota, Wisconsin. (From Fundamentals mids for two different types of ecosystem. Grazing and detritus
of Ecology, Third Edition, by Eugene P. Odum. Copyright © food (trophic) webs are shown separately (after Odum, 1957, 1962).
1971 by W. B. Saunders Company. Reprinted by permission of The relative contribution of the detritus food web to heterotrophic
CBS College Publishing.) energy flow varies between the two ecosystems and cannot be
judged by the biomass of detrivores. (B) Effect on the biomass
pyramid of variation in the proportion of heterotrophic energy
‘All transformations of energy (other than heat) are incomplete because flow accounted for by the detritus food web. (Reprinted by per-
some ofthe energy is converted from nonrandom to random (i.e., heat). mission of Eugene P. Odum and Japanese Journal ofEcology.)
40 CHAPTER4 — Production Ecology

provides a comparison between the biomass and en- Tuble 4-1 Energy Flow, per Unit of Biomass, in Six
for two Primary Consumer Populations Differing in the
ergy flow in detritus and grazing trophic webs
relative Size of Individuals
different types of ecosystem. It shows that the
importance of the two webs varies greatly and that the Energy Flow,
energy flow in each web cannot be predicted from the Organism kcal day ‘g (biomass)'
biomass of that web. It also shows the variation in the
Soil bacteria 1000
biomass of detritus and grazing food webs when the
detritus web accounts for either 99% or only 10% of Marine copepods 125
post-producer energy flow. Saltmarsh grasshoppers 0.4

Intertidal snails 0.1

Effect of Organism Size. As already noted, the
Meadow mice 1.16
functional importance of specific organisms within the
ecosystem cannot be judged on the basis of biomass Deer 0.49
alone. The small biomass of producers in a marine bay
ecosystem can have the same order of magnitude of Data from Odum, 1968.

energy flow as a forest with a biomass several orders of

magnitude greater (Figure 44A). A very small bio-
mass of detritus organisms in a freshwater spring in to a very flat right-side-up pyramid (where size de-
Florida was found to account for five times as much creases sharply through the web).
energy flow as a detritus biomass 22 times larger in a A consequence of the relationship between size
marine bay ecosystem. and metabolic rate is that for a given biomass, a com-
The unreliability of a biomass pyramid as an indi- munity of smaller organisms will have a higher rate of
cator of ecosystem function is related to the size of the energy flow than will a community of larger organ-
organisms (Odum, 1956). It is known that there is a isms, and for a given energy flow, a community of
broad general tendency for the metabolism per unit smaller organisms will have a smaller biomass than
weight of heterotrophic organisms (i.e., consumers) to will a community of larger organisms. This conse-
be inversely proportional to body size (see Kleiber, quence has very important implications for ecosystem
1961, or Gordon et al., 1968): management. Because the flow of energy through an
ecosystem is ultimately limited by the availability of
1 solar energy and by photosynthetic efficiency (see
Metabolism per gram a —~——_3;7
body size ~” below) and because neither of these two parameters
or varies greatly in any particular climatic area, commu-
nities of large plants, such as trees, will support a
y = ax’ larger biomass than will small plants, even though the
where y is the respiration rate, x is the animal’s weight
in grams, and a@ and J are empirically determined con-
Grasshoppers
stants (b is generally approximately 0.7). Spiders
Insect parasites
Table 4-1 shows the relationship between the size Insects
of an organism and its contribution to energy flow.
There is a general trend toward increasing energy Protozoa
flow per unit of biomass as size decreases. Inclusion of (a) (b)
vertebrates, invertebrates, and microbes-in the com-
Figure 4-5 Influence of the size of organisms on the shape
parison results in there being no linear relationships,
of biomass pyramids (after Odum, 1956). The diagrams as-
but within each group the trend can be seen. Figure
sume no imports or exports of organic matter into or out of the
4-5 examines the effects that variation in organism ecosystem (photosynthesis = total ecosystem respiration) and
size has on the shape of the biomass pyramid. Assum- 50% of the heterotrophic energy flow in the grazing food web.
ing that 50% of the total energy flow is in detritus (A) Size increases up the pyramid. (B) Little change in size up
food webs, the shape of the pyramid varies from in- the pyramid. (C) Size decreases up the pyramid. (Used by per-
verted (for increasing organism size through the web) mission of H. T. Odum and the Ecological Society of America.)
 Trophic Chains and Webs, Ecological Pyramids, and Energy Flow Diagrams 41

smaller plants may actually produce new biomass at a energy in organic matter will have a total community
faster rate. respiration that is greater than photosynthetic produc-
Where it is impractical or uneconomical to harvest tion, and ecosystems that experience exports of organ-
plants very frequently, large plants will be the most ef- ic energy (out of the ecosystem or into long-term
ficient way of harvesting sunlight energy because of storage) will have a total community respiration that is
the large biomass that they sustain. Conversely, if the less than their photosynthetic production.
production of plants can be harvested frequently and
energy-efficiently (e.g., by domestic herbivores), then
the use of small plants such as grasses, which have high Energy-Flow Diagrams
metabolic and biomass production rates but small bio- The ecological pyramids discussed so far are useful as
mass, may be the most useful way for us to harvest an overall summary of ecosystem function, but they
sunlight energy. Which size of producer is best obvi- omit a great deal of important information. They give
ously depends on whether one wants the energy in the no information on significant flows of energy between
form of timber or beef. However, the important point grazing and detritus trophic webs, and they fail to
to note is that it takes energy to harvest a plant crop, draw adequate attention to the consequence of the
and the more frequent the harvest, the less favorable second law of thermodynamics: the loss of heat energy
the energy benefit/cost ratio is likely to be. By using via the process of respiration each time there is an en-
animals to perform the harvest of rapidly growing but ergy transformation. They also give no information
short-lived plants, we can largely avoid the direct en- on the relative magnitude of the several pathways of
ergy costs of harvesting such plants. energy transfer that determine the biomass at any one
trophic level. These deficiencies can be overcome by
Effects of Energy Imports or Exports. A third using energy-flow diagrams that depict the biomass of
factor that can complicate the interpretation of the trophic levels by boxes that are joined by pipes that
basic ecological pyramid is imports or exports of ener- represent energy transfers between trophic levels and
gy into or out of the ecosystem. In apparent contradic- between trophic webs. Figure 4-6 presents an energy-
tion of the second law of thermodynamics, the flow diagram for the pine forest pyramids presented in
biomass and energy flow in heterotrophic trophic lev- Figure 4-4. Only the major energy transfers are
els is sometimes larger than that of producer biomass shown, but it is clear that there is a considerable
and producer energy flow. In fact, the law is not violat- amount of energy interchange between grazing and
ed: such anomalies result from imports of organic detritus trophic webs and that their distinction be-
matter from another ecosystem. Similarly, het- comes increasingly arbitrary as one proceeds through
erotrophic energy flow and biomass may be smaller the trophic web. Individual organisms may change
than predicted because of exports of autotrophic ener- their trophic relationships according to the availability
gy out of the ecosystem. For example, inputs of dead of food so that they may alternately feed primarily in
insects and leaves into stream ecosystems from the detritus trophic web and primarily in the grazing
streambank vegetation can support a level of biomass trophic web.
and productivity in populations of aquatic het- Although energy-flow diagrams present a static
erotrophs far in excess of that possible on the basis of picture of what is in reality a dynamic and ever-chang-
aquatic primary production. Conversely, downstream ing network, they do provide a useful summary of the
drift of animals in well-illuminated but fast-flowing functional character of an ecosystem and are more
streams may result in heterotrophic biomass and ener- than merely an academic form of “doodling.” Success-
gy-flow pyramids that are much smaller than would be ful management of ecosystems for material products
expected from the energy flow in the aquatic producer requires an understanding of the balance of energy in-
trophic level. puts and outputs that determine the size of the crop.
In a closed ecosystem, the only input of energy is The relationship between energy inputs and outputs
sunlight and the output is heat energy. The energy of and the size of biomass compartments can be likened
photosynthesis equals the total loss of energy through to that between the amount of water in a tank and the
respiration (heat loss through radiation from warm flows into and out of that tank. For a given rate of out-
objects is ignored here but is discussed in Chapters 7 flow, the volume of water will increase as the rate of
and 8). Conversely, ecosystems that receive imports of inflow increases. For a given rate ofinflow, the volume
42 CHAPTER 4 — Production Ecology

Heat lost eventually

to space

Grazing trophic

available
to plants
3000

y,}Detritus trophic
Sara web
storage eae ealera,
(current annual
increment)

Figure 4-6 Energy-flow diagram for the pine forest ecosystem shown in Figure 4-5. Boxes
represent the energy in biomass (kcal m~’); pipes represent energy transfers (kcal m~* day
'). (After
Odum, 1957, 1962. Copyright 1957, 1962 by the Ecological Society of America. Used by permis-
sion.)

of water will decrease as outflow increases (Figure tion is limited to logs larger than 10 cm in diameter or
4-7). For a given photosynthetic input of energy to whether the complete tree, including branches, fo-
trees, biomass in the trees will depend on the outflows liage, entire stems, stumps, and roots, is harvested.
of energy to respiration, grazing losses, or litterfall. The standing crop is the weight or volume of mate-
These are discussed next. rial that can be sampled or harvested by a particular
method from a given area at a particular time. Crop
and standing crop are the same at the time of harvest.
4.4 ‘Terminology Used in As with a crop, the size of a standing crop depends
Production Ecology greatly on the method of harvesting.
Yield is the average rate of accumulation of har-
Before beginning a detailed review of energy transfers vestable material: crop divided by the period of crop
and storage at the different trophic levels of the ecosys- production. Because the time required to produce
tem, we must establish the meaning of various terms crops of different species varies, it is sometimes difficult
used by production ecologists and resource managers. to make comparisons between crop data. This problem
The crop is the total volume or weight of material is overcome by calculating yield over a year, although
that can be removed from a given area over a given pe- agriculturalists sometimes use yield per month during
riod, for example, the weight or volume of Douglas- the growing season to overcome the problems of vari-
fir wood harvested from a hectare during a 60-year ro- able growing-season length. In terms of timber man-
tation. The magnitude of a given crop refers only to agement, yield is the factor of greatest long-term
the material removed from the ecosystem and there- interest because it determines the return on invest-
fore depends greatly on the proportion of the individ- ment—essential information for production forestry.
ual organism that is harvested: the harvest index. The The terms crop, standing crop, and yield are widely
crop from a hectare of pine forest might vary by as used in resource management. They are of less value
much as 30% or more, according to whether utiliza- to ecology, however, because they are not based on
 Production Ecology at the Primary Producer Level 43

duction at higher trophic levels is referred to as

secondary production. Real photosynthesis and apparent
WA) photosynthesis are terms used as alternatives to gross
DNS es Bee
es N
primary production and net primary production
Big inflow, small Small inflow, big “
outflow, medium volume outflow, big volume outflow, small volume (NPP), respectively, because autotrophic production is
(a) the result of photosynthesis.
Biomass (analogous to standing crop) refers to the
total weight of organic material in a population, a
trophic level, or an ecosystem in a given area at any
Respiration Increased
respiratory loss
point in time. It is equal to the standing crop plus all
Solar D,
adem
pu) ; -
the unharvested and unharvestable organic material
and is the term used by ecologists to describe the
Biomass
of Increased quantity of biotic components of ecosystems, hence its
trees razing loss use in biomass pyramids.
Productivity is the production per unit area per unit
time and is therefore analogous to yield. As with yield,
disease, or decay the period over which production is calculated must
loss
be carefully defined. Productivity often varies consid-
2. Increase in any of the three erably over the season, and if it is expressed over a
outflows reduces the biomass of trees
(b)
short time period (e.g., a day), then the value usually
represents the average daily rate over the season. As
Figure 4-7 ‘The relationship between inputs, outputs, and with production, one can refer to gross, net, primary,
the size of a biomass compartment. (A) The volume of water or secondary productivity.
in a tank depends on the rate of inflow and the rate of outflow.
(B) Biomass in a trophic level depends on energy inflows and
outflows (e.g., tree biomass in a 100-year-old stand).
4.5 Production Ecology at
the Primary Producer Level
stable definitions and are subject to change according
to technological and economic changes. Ecologists
Fundamentals of Production Ecology
use the following three similar terms to overcome this Figure 4-8 summarizes the conversion of solar radia-
problem: production, biomass, and productivity. tion into harvested biomass (economic production)
Production (analogous to crop) describes the in- and the major determinants of this process. First, we
crease in total weight (biomass) or quantity of organic describe the overall process and then examine the in-
material on a given area over a defined period. It is dividual factors that control it.
usually used to refer to a specific population or troph- The photosynthetically active portion of sunlight
ic level, but it can refer to the whole ecosystem. It in- (PAR) is captured by plants in the process of photo-
cludes not only the increased quantity of material synthesis. The proportion of PAR that is captured de-
remaining on the area at the end of the time period pends first on the surface area of leaves and second on
but also material that was produced and subsequently the efficiency with which these leaves process PAR
lost before the end of the period. Thus, production is into high-energy organic molecules.
equal to the crop plus the nonharvested and nonhar- Both leaf area and the efficiency of that leaf area
vestable organic material produced over the period, are determined by the availability of the site resources:
plus losses to other populations, trophic levels, or water, nutrients, and light. Leaf area is also affected by
ecosystems, as appropriate. the way in which plants allocate their NPP between all
Production can be subdivided into gross production the different parts of the plant. Like leaf area, this car-
(increase in organic material plus losses to respiration bon allocation is influenced by the availability of
over the period) and net production (increases in organ- water, light, and nutrients. NPP is allocated to the
ic material after losses to respiration). Production by root system or to leaf and stem growth in response to
plants is referred to as primary production, whereas pro- which resources are in shortest supply: soil nutrients
cet CHAPTER4 — Production Ecology

Leaf area and photosynthetic

efficiency

Solar radiation

Water
Light Net photosynthesis
Nutrients

Respiration

ues Net primary production

allocation
Litterfall
Root death
Plant death
Herbivory

Net biomass accumulation

Unharvested biomass

Harvested biomass
(economic production or yield)

Figure 4-8 The major determinants of economic production (yield) in a forest ecosystem
within a particular climatic regime. Clearly, the availability of site resources (water, light, nutri-
ents) is critically important, both directly and indirectly, through their effect on the allocation of net
photosynthate. (Adapted from Kimmins, 1990.)

and water versus energy and carbon. The allocation is some extent, on carbon allocation, which in turn de-
thought to represent the plant’s attempt to maintain pends on the availability of the primary site resources.
optimum ratios between carbon and other nutrients Net photosynthesis minus respiration loss defines
and between leaf area and the supply of light, mois- NPP. Some of this is allocated to tissues and organs
ture, and nutrients to that leaf area. Leaf area seems to that are not long lived. They are ephemeral and die
be more important to a plant than photosynthetic effi- after one or more months (e.g., fine roots, tropical
ciency, although both are obviously important. If a tree leaves), one or more years (e.g., leaves of temper-
plant community that has low leaf area and low photo- ate and northern trees), and one or more decades (e.g.,
synthetic efficiency because of lack of resources is pro- branches, bark). The allocation to ephemeral rather
vided with additional resources, then it will normally than long-lived tissues depends to a considerable ex-
expand its leaf area to intercept as much of the avail- tent on resource availability.
able sunlight as possible, rather than invest most of Some of the individual plants in the community
these new resources into making its present leaf area die because of competition or other causes. This usu-
more efficient. However, there will obviously be a bal- ally results from competition for light, but on dry sites
ance between leaf area and efficiency. or in dry climates, death of plants can also result from
The interaction between foliage and solar radia- competition for water. Plants do not usually die as a
tion results in net photosynthesis. Some of the photo- direct result of lack of nutrients, but such a lack may
synthate produced is used by the plant as an energy prevent them from competing effectively for water
source, and the energy and carbon are lost via respira- and light. An additional proportion of NPP is lost to
tion. Different tissues have somewhat different respi- herbivory. During an insect outbreak, this loss can ex-
ration rates, so the magnitude of this loss depends, to ceed the NPP of perennial plants, but, except during
 Production Ecology at the Primary Producer Level 45

insect epidemics, loss to herbivory is generally modest duction, respectively, and incident radiant energy may
in most forests. be based on the total solar energy falling on the area,
After individual plants, populations, or communi- the visible solar energy, or the energy of those visible
ties have lost NPP to herbivory, litterfall, and death, wavelengths that are utilized in photosynthesis (see
we are left with net biomass accumulation. Depending Chapter 7). The latter amounts to approximately
on the management objectives, the prevailing market 25% of total radiation (Gates, 1971). An additional
demand for plant biomass, and the economics of har- complication arises from the distinction by some re-
vesting, only a portion of the net biomass accumula- searchers between the radiant energy falling per
tion (the harvest index) will be harvested. The harvest square meter of the earth’s surface and the energy ac-
index depends greatly on the plant’s NPP allocation tually falling on photosynthetic organs. Some have
strategy. The unharvested portion of the biomass re- even estimated the energy that is actually absorbed by
mains in the ecosystem. It participates in the detritus plants and use this to calculate photosynthetic ability.
food web, contributes to future soil organic matter, The most commonly used measure of photosynthetic
and provides habitat and food for the animals and mi- efficiency is net production divided by the radiant en-
crobes that depend on this material. ergy reaching the surface of the vegetation in the
Throughout this discussion, you should have wavelength range 0.4 to 0.7 ps (visible light). However,
noted the overriding importance of carbon allocation, even this efficiency will vary according to whether it is
which ultimately depends on the availability of the site based on radiation for the whole year, for the leafy sea-
resources of light, moisture, and nutrients. Foresters son, or for the growing season.
have the ability to manipulate the availability of nutri- Photosynthetic efficiencies have been calculated
ents to plants and can modify the availability of light for a wide variety of ecosystems. Values are generally
and the competition for light and nutrients. They have small (1 to 5%) and surprisingly invariable in ecosys-
less influence on the availability of moisture, because tems that have a continuous plant community (Phillip-
this is largely determined by the climate, soil, and son, 1966). In a large number of studies conducted
topography. However, loss of soil organic matter can since the 1940s, efficiencies (based on total incident
reduce soil moisture storage, and competition for radiation) of more than 3% have seldom been found,
water can be influenced by management. Thus, the and 1% is typical, although when recalculated on the
fundamental determinants of production ecology at basis of energy in the visible range, these figures must
the primary producer level are, to a considerable ex- be adjusted to 6 and 2%, respectively (Kormondy,
tent, under the influence of forest managers. 1969). Much higher efficiencies are attained over
The following sections review the major inputs short periods when optimum conditions prevail, and
and outputs referred to in Figures 4-6 and 4-8. efficiency increases as light intensity drops, approach-
ing 20% at very low intensities under laboratory con-
Inputs: Photosynthesis. We have already noted ditions (Collier et al., 1973). Efficiency also varies with
that with the relatively minor exception of chemoau- leaf morphology and orientation; leaves that have de-
totrophs, systems are organized and driven by inputs veloped in the shade are generally more efficient at
of solar energy to photoautotrophs, the green plants, lower light intensities than leaves that have developed
or producers, of the ecosystem. The rate of primary in full sunlight.
production is therefore of great interest to both ecolo- Photosynthetic efficiencies calculated in the field
gists and ecosystem managers. The efficiency with over long periods are lower than those obtained in
which solar energy enters ecosystems is referred to as short-term studies. This is because for at least part of
the photosynthetic efficiency (expressed as a_ percentage). the growing season, much of the light falls on non-
It is defined as photosynthetic organs or on bare ground and because
production conditions for optimum photosynthesis are present
x 100 only part of the time. It has been calculated that if op-
timum conditions existed in the field throughout the
Great care must be taken when comparing photosyn- growing season, then net photosynthetic efficiency
thetic efficiency figures because they are calculated in should be approximately 5% of the total incident solar
many different ways. One arrives at either gross or net radiation and 12% of the total visible solar radiation
photosynthetic efficiency by using gross or net pro- (Gates, 1971). The modest photosynthetic efficiencies
46 CHAPTER 4 _ Production Ecology

vs) Period of full photosynthetic efficiency Figure 4-9 that a significant proportion of available
Photosynthetically
* (leaf area index greater than 4)
Leaf area
solar radiation during the growing season is not used
active radiation, 30 See Solar radiation by these two annual crops.
:
5ier
- ha | 25 -

20
15
Losses
50
Respiration. According to the second law of
40 thermodynamics, some of the energy of photosynthe-
— area, 30 sis is used up during growth and maintenance, a
10° m* ha 10
process of energy loss referred to as respiration. Respi-
10
ratory losses in plant communities vary from as little
May June July August as 15% to more than 90% (Table 4-2), but the results
of a wide range of studies suggest that, in general, au-
Figure 4-9 Development of leaf area in relation to the
availability of solar energy for photosynthesis during the
totrophic respiration accounts for approximately 30 to
growing season for wheat and maize (after Niciporovic, 70% of autotrophic gross production (refs. in Sprugel
1968). Wheat is better able to exploit available solar radiation et al., 1995).
than maize because its maximum development of leaf area coin- The temperature of the environment (especially
cides with maximum solar radiation. Leaf area index is the ratio nighttime temperature) and the size, rate of growth, and
of horizontal leaf surface area to horizontal ground surface area. physiological condition of plants (i.e., metabolic rates)
(Courtesy UNESCO.) infiuence plant respiratory losses. The relative size of
this loss is important in determining the productivity of
the ecosystem (Figure 4-8B). The rate of growth and
achieved by many agricultural crops reflect incom- biomass accumulation of individual plants are in turn
plete coverage of the ground by leaves for much of the dependent on respiration losses, which will affect the
growing season. Figure 4-9 shows the development of competitive status of the plant in the community.
the leaf area of wheat and maize during a growing sea- Respiration losses at the primary producer level
son in relation to photosynthetically active solar radia- seem to increase as one moves from polar toward
tion. Maximum net use of sunlight is achieved at a /eaf tropical latitudes, presumably because of increasing
area index* of approximately 4, and it is apparent from temperatures (especially during the warmer nights).
As a result, NPP in the tropics may not be much
greater than in humid temperate areas despite the
"Leaf area index: the projected leaf surface area per unit ofground surface. higher gross production in the tropics. Lower gross
For a more complete discussion of leaf area in forests, see Section 4.6. production in higher elevations or more northerly

Table 4-2 Loss of Gross Primary Production to Respiration in Various Different Terrestrial Ecosystems
Loss of Gross
Primary Production
to Respiration (%) NPE? Reference

Abandoned field, Mich. 15 85 Golley, 1960

Cornfield, Ohio 23 77 Transeau, 1926
Alfalfa field, U.S. 38 62 Odum, 1971
Scotch pine plantation, England 39 61 Odum, 1971
Oak-—pine forest near New York City 95 45 Woodwell, 1970b
Tropical rain forest, Puerto Rico 71 29 Odum and Pigeon, 1970
450-year-old Douglas-fir forest, Ore. 93 if Grier and Logan, 1977

“NPE is defined as (net production/gross production) x 100.

 Production Ecology at the Primary Producer Level 47

areas is partly offset by reduced respiration losses, par- sole cause of it. For a review of respiration, see
ticularly during the cool nights that are characteristic Sprugel et al. (1995).
of these areas (Iranquillini and Schutz, 1971). High
nighttime respiration because of high nocturnal tem- Consumption by Herbivores. ‘Vhe extent to which
peratures plays an important role in limiting the plants are consumed by herbivores varies greatly
southerly or lower altitudinal extension of the range of between different types of ecosystem (Table 4-3). In
some plants. grasslands, for example, a very high percentage of the
Aboveground productivity of forests increases to a aboveground plant biomass and the NPP is palatable,
peak at intermediate ages and then declines as the age digestible, and within the reach of herbivores. In
of the forest increases (Figure 4-10). This was previ- forests, however, the foliage is a much smaller propor-
ously attributed to increasing respiration losses (e.¢., tion of NPP, the rest being woody tissues that are little
Kira-and Shidei, 1967), but this explanation has been used by herbivores (with the exception of small roots
challenged. It is now thought that the decline in net and young twigs). Forest foliage tends to be less palat-
production has many determinants, including one or able and digestible than grassland foliage because of its
more of the following: reduced leaf area, reduced pho- physical and chemical characteristics, and much of it is
tosynthetic efficiency of the foliage, reduced nutrient physically out of reach of many herbivores. Grassland
availability, greater internal moisture stress in very tall plants experience between 28 and 60% loss of NPP to
trees, physiological old age, increased tree mortality, herbivores, whereas in forests, only approximately 5
pathogens, and reduced ability to repair wood and to 10% (an average of 8% is given by Bray, 1961,
snow damage as a result of declining vigor (Ryan et 1964) of the foliage is consumed. This average level
al., 1997). Respiration losses as a percentage of photo- represents only 1.5 to 2.5% of NPP except during pe-
synthetic production continue to increase as biomass riods of high insect herbivore abundance. Phytoplank-
accumulates and the ratio of photosynthetic to non- ton communities in aquatic ecosystems experience
photosynthetic respiring tissue declines, and this con- even heavier exploitation by herbivores (60 to 99% of
tributes to the loss of net production, but it is not the NPP) than grassland communities because of the ac-
cessibility of phytoplankton and their low proportion
of nondigestible supporting structures.
The importance of herbivores in the dynamics of
Photosynthetic
production energy is illustrated by estimates of the consumption
of NPP by termites and its conversion to CH,
Aboveground (methane), CO, and hydrogen gas (Zimmerman et al.,
production 1982). Termites are exceedingly abundant in the trop-
ical and subtropical areas of the world, where they
Respiration & consume an estimated 37% of the NPP (which is
belowground
production equal to 28% of the world’s terrestrial NPP). The final
product of this consumption is thought to be approxi-
mately 152 million tons of methane, 50 billion tons of
CO), and 200 million tons of hydrogen per year.
The numbers given in ‘Table 4-3 for forest ecosys-
tems are examples of the exploitation of forest plants
by herbivores under conditions of normal, stable her-
bivore populations. Anyone who has been in a forest
during an outbreak of defoliating insects knows that
INSSOS SENG ae although these figures may be an accurate estimate of
the utilization of trees by herbivores under “normal”
Figure 4-10 Conceptual model of the relationship between
circumstances (periods of herbivore scarcity), they
stand age and photosynthetic production, respiration, and
probably underestimate the long-term utilization.
below ground production (mainly fine-root production).
(Adapted from Ryan et al., 1997.) The difference between the two
Forest insects periodically increase to epidemic pro-
lines is aboveground production, which peaks at relatively young portions (e.g., the spruce budworm, the hemlock
ages and then declines. See text for explanation. looper, the Douglas-fir tussock moth, the gypsy moth,
48 CHAPTER 4 Production Ecology

Table 4-3 Extent of Consumption of NPP by Herbivores in Various Types of Ecosystem

Consumption of NPP
Type of Primary Producer by Herbivores, %
Type of Ecosystem

Old agricultural field,

30 yr after abandonment Perennial forbs and grasses 1.1

Mature deciduous forest Trees t= 2s0

Desert scrub Annual and perennial herbs and shrubs 5.5

Salt marsh Perennial herbs 8.0

Old agricultural field,

1-7 yr after abandonment Annual herbs 12.0

Conifer plantation Pine trees 12.5

Lake Mendota, Wis. Phytoplankton (algae) and macrophytes 14.7

Cedar Bog Lake, Minn. Phytoplankton (algae) and macrophytes 21.8

Managed rangeland Perennial grasses 30-45

African grasslands Perennial grasses 28-60

Marine bay Phytoplankton 15

Ocean Phytoplankton 60-99

Consumption of Leaf Production, %

By Insects By Birds and Frogs

Pine forest Trees 9.2 1S

Pine-oak forest Trees 3.4 2.7

Pine—oak—alder forest edge Trees 4.0 Sal

Data from Kozlovsky, 1968; Odum, 1962; Ricklefs, 1973; Wiegert and Owen, 1971.

species of sawfly or tent caterpillar), at which time remarkably well. Oak trees are often defoliated with
they can consume all of a deciduous tree’s foliage or only modest effects on growth because they quickly re-
several years of evergreen foliage production in a sin- foliate. Under some circumstances, Douglas-fir can
gle summer. Such massive transfers of energy to graz- recover from heavy defoliation by the Douglas-fir tus-
ing and detritus food webs via insect frass (feces), dead sock moth (Orgyia pseudotsugata) with little growth re-
insects, and dead trees result in reductions in NPP for duction the following year (Majawa, 1977). This may
several years and can drastically reduce the biomass of be partly because tussock moth outbreaks often occur
primary producers in the ecosystem (e.g., Morris, during a drought-induced period of slow tree growth.
1963; Rafes, 1970, 1971; Varley et al., 1973). Such If the tree is partially defoliated during the period of
pulses of energy flow from producer to primary con- moisture stress, then the photosynthetic efficiency of
sumer level may be reflected by temporary expansions the remaining foliage may be temporarily increased,
in the magnitude of energy flow and biomass all the largely compensating for the reduced foliar biomass.
way up the trophic web. Foresters and farmers are concerned with harvest-
It has generally been assumed that heavy defolia- ing NPP, either directly or via some desired species of
tion kills or at least drastically reduces the growth of herbivore. The existence of the pesticide industry re-
trees. This is undoubtedly true in many species, but flects our desire to prevent the energy of primary pro-
some trees seem to be able to tolerate such defoliation duction from flowing to members of the secondary
 Production Ecology at the Primary Producer Level 49

trophic levels other than ourselves or domestic herbi- tion of the total forest biomass invested in photosyn-
vores. However, although transfer of energy from thetic organs decreases as one moves from boreal to
plants to herbivores may be considered undesirable in temperate regions but increases again somewhat in
some ecosystems, in others it may be the objective of tropical forests. This pattern reflects the evergreen
management. Pasture and stock farmers spend their character of northern forests, the deciduous character
working lives trying to promote such energy flow by of temperate forests, and the evergreen character of
growing plants in a manner that optimizes energy flow tropical rain forests.
into and accumulation of energy in herbivore popula- The annual quantity of aboveground litterfall ob-
tions (e.g., cows, sheep, goats). viously depends heavily on the proportion of the fo-
Herbivory in forests that are managed for timber liage biomass that dies each year; the longer the
production might seem to be unequivocally bad from foliage retention, the less the quantity of leaf litterfall.
the standpoint of timber management, but considered For example, in coastal British Columbia, Pacific sil-
in terms of overall, long-term ecosystem energetics, a ver fir (Abies amabilis) retains approximately 6 to 9
stable, low level of herbivory might in fact promote years of foliage at the lower end of its altitudinal range
ecosystem productivity by stimulating litter decompo- (approximately 300 m near Vancouver, BC), approxi-
sition and the circulation of nutrients. Also, it has been mately 13 to 19 years at midrange (approximately
suggested that over a long period of time, defoliating 1000 m), and approximately 22 to 26 years of foliage
insects may actually increase the productivity of un- (oldest green needles on one branch were 31 years old)
managed, unharvested (by foresters) ecosystems. It has at the upper altitudinal limit of continuous forest (ap-
been noted that insect outbreaks that result in exten- proximately 1700 m) (Kimmins, unpublished data).
sive tree mortality are frequently associated with With relatively little variation in stand foliage biomass
mature or overmature stands in which primary pro- over this range, leaf litterfall will vary by a factor of
duction has begun to decline. The insects remove approximately 3. Foliage retention can vary latitudi-
these older, less productive stands, which are then re- nally as well as with elevation. The ecological signifi-
placed by younger, more productive stands, thus in- cance of evergreenness is discussed in greater detail in
creasing energy flow through the ecosystem (Mattson Chapter 5. A detailed review of this topic can be found
and Addy, 1975). Further studies of the overall effect in Chabot and Hicks (1982), Sprugel (1989), and
of herbivores on ecosystem productivity are needed Reich et al. (1995).
(e.g., Zlotin and Khodashova, 1980) to elucidate the Aboveground litterfall varies with a variety of fac-
overall role of herbivory in ecosystem function. tors, including climate, soil moisture and soil fertility
(greatest on moist and fertile sites), and altitude (often
Aboveground Litterfall. With the exception of greatest at intermediate altitudes). Stand density, how-
heavy exploitation by herbivores during periods of ever, seems to make little difference. For example, lit-
high population, the greatest loss of forest NPP is by tle variation in either total or nonwoody aboveground
litterfall: the regular annual transfer of living plant litterfall was observed in four stands in Alaska ranging
material to the nonliving organic matter of the forest from a 12- to 30-year-old hardwood-—conifer stand of
floor and mineral soil. Plant litterfall is obviously of 10,000 stems per hectare to a 132- to 163-year-old
less significance in ecosystems such as oceans and spruce-hemlock stand of 660 stems per hectare (Hurd,
grasslands, in which much of the NPP is consumed by 1971). Several other studies have reached the same
herbivores, but in many terrestrial ecosystems, it is the conclusion, including one that showed constant litter-
major pathway of energy flow beyond the producer fall in eucalyptus forests varying in age from 55 to 200
level. Leaves are the major component of above- years and varying in density by more than fourfold.
ground litterfall in most ecosystems, although other Understory vegetation contributed 25% of the above-
components, such as bark, can be important in some ground litterfall in the oldest stand but only approxi-
areas (e.g., eucalyptus forests). mately 2.5% in the 55-year-old stand (Bray and
Both total aboveground litterfall and leaf litterfall Gorham, 1964).
of forests increase from polar regions toward the Litterfall occurs either seasonally or continuously.
equator. This parallels a similar variation in biomass Almost all aboveground litterfall in temperate decidu-
and NPP. because aboveground litterfall generally re- ous forests occurs in the autumn, and several species of
flects aboveground forest productivity. The propor- temperate conifers also shed most of their old foliage
50 CHAPTER4 Production Ecology

in the fall. In some climatic regions, deciduous leaf-fall supply of energy to the detritus food web. In Chapter
may occur just before a dry season. Aboveground lit- 5, we shall see that the quantity and qualitative charac-
terfall in tropical evergreen hardwood forests is dis- teristics of litterfall are important in the overall func-
tributed throughout the year, but minor peaks do tioning of the ecosystem. The annual mortality of fine
occur in slightly drier months and in association with roots is so important that it will be referred to again
storms. In some species of pine, most needlefall occurs several times.
in the summer either before or during hot, dry peri-
ods. In species such as spruce, leaf-fall is continuous Net Productivity and Biomass. Although energy
and random, occurring in response to random varia- gains by photosynthesis and energy losses by respira-
tions in such factors as weather and insect attack. Re- tion, herbivory, and litterfall are individually of con-
views of litterfall can be found in Meentmeyer et al. siderable interest, it is the combined outcome of these
(1989) and Vogt et al. (1986). processes that is of greatest concern to ecosystem
managers and ecologists: net biomass accumulation.
Belowground “Litterfall”: Death of Fine Roots. Traditionally, foresters have been interested in the
Until recently, most of the data on litterfall was for standing crop rather than biomass, but with the devel-
aboveground litter, which was assumed to be the opment of more complete utilization of trees, biomass
major contribution of organic matter to the soil. This is becoming of increasing interest. Productivity is gen-
assumption is now known to be wrong. In many types erally reported in terms of the weight, volume, or en-
of forest, shrub, and herbaceous communities, there is ergy content of the NPP per hectare per year.
a greater annual turnover of organic matter below Net primary productivity of a given type of vegeta-
ground than above ground. A large proportion of a tion generally increases as one moves from temperate
plant’s net photosynthate is invested each year in rela- to tropical areas; within a given major climatic region,
tively short-lived fine roots and mycorrhizal fungi (see productivity increases as one moves from drier habi-
Chapter 5), an apparently necessary strategy in the tats to wetter habitats and from less fertile to more fer-
quest for the soil moisture and nutrients required to tile sites. Maximum productivity may be attained in
sustain photosynthesis. In closed-canopy forests in semiterrestrial ecosystems (shallow water, swamps,
which trees account for most or all of the NPP, alloca- marshes, or terrestrial sites associated with abundant
tion to fine roots has been reported to be between 8 supplies of water), although in many forest ecosys-
and 67% of NPP (Comeau and Kimmins, 1989; San- tems, productivity declines if the soil becomes exces-
tantonio and Grace, 1987; Vogt et al., 1986). In some sively wet, and excessively wet sites may have low
types of forest, such as in high-elevation subalpine nutrient availability. Productivity also declines as one
areas, the majority of the annual input of organic mat- moves into larger bodies of water away from the
ter to the forest floor comes from belowground “lit- land/water interface, such as the middle of large lakes
terfall”—the annual, monthly, or weekly death of fine or oceans. Forests generally exhibit productivities
roots and associated micro-organisms. In a study of comparable to, if not slightly lower than, agriculture.
Douglas-fir growing on poor (dry, low fertility) and This is surprising considering that forests tend to be
good (moist, fertile) sites, Keyes and Grier (1981) re- located on less fertile sites in more severe climates
ported a difference in total NPP of only 13%, com- than agricultural crops and receive far fewer aids to
pared with a net aboveground production difference production (fertilizers, insecticides, etc.).
of 88% (13.7 t ha! on the good site versus 7.3 t ha! Most published estimates of forest productivity are
on the poor site). This was explained by the difference based on aboveground data, or, if they include roots,
in investment in the production of fine roots (2 mm); they refer only to the large roots. Recent studies have
1.4 tha ' on the good site and 5.6 t ha! on the poor shown that a different picture may emerge if one in-
site. On the good site, 8% of total stand dry matter cludes the production of fine roots. As noted above,
production was in fine roots, compared with 36% on Keyes and Grier (1981) studied productivity in two
the poor site. These results were confirmed by Kurz stands of Douglas-fir that varied widely in stemwood
(1989). production (site indices of 24 and 40 [m at 50 yr] for
Litterfall is an important parameter of energy flow. the poor and good stands, respectively). They re-
Its magnitude influences the proportion of NPP that ported that aboveground and coarse-root biomass incre-
is stored as permanent biomass. It is also the major ment on the poor site (18.4 t ha ') was approximately
 Production Ecology at Consumer Trophic Levels 51

half that of the good site (15.3 t ha~!). In contrast, the of only the larger parts of the stem to the utilization of
estimated total biomass increment for the poor site the “whole tree” (all aboveground biomass) or even
(15.4 t ha”) was approximately 87% of that for the the “complete tree” (all biomass, including roots), the
good site (17.8 t ha '). This is because 36% of the distribution among crowns, stems, and roots may be
NPP on the poor site was allocated to fine-root pro- less important. Nevertheless, because of different end
duction (5.6 t ha~') in comparison with 8% on the uses for and value of bark, wood, branches, foliage,
good site (1.4 t ha !). and roots, it is still important to know the distribution
A similar study in subalpine stands of Pacific silver of biomass among these components in a completely
fir revealed a marked variation in allocation of NPP to harvested crop. Such knowledge is also needed for the
fine roots as a stand ages: 36% in a 23-year-old stand, design of efficient harvesting equipment and process-
in comparison with 66.4% in a mature, 180-year-old ing techniques and the advanced planning for post-
stand (Grier et al., 1981). Such age-related variation in harvest slash disposal.
the allocation of NPP between perennial and A study of the relative distribution of biomass com-
ephemeral tissues makes it difficult to compare the ponents in both forest and agricultural crops on good-
productivity of stands in which fine roots have not and medium-quality sites revealed that the proportion
been studied. This study also examined the biomass of of stemwood in the total biomass of a spruce crop
mycorrhizae.’ These accounted for a mere 0.7 and dropped from 42% on a good site to 27% on a medi-
0.3% of the ecosystem biomass but for 13.9 and 15% um site. Similarly, the proportion of a Solanum (potato)
of the NPP in the 23- and the 180-year-old stands, re- crop that was a harvestable product dropped from
spectively. Fine roots and mycorrhizae together ac- 75% ona good agricultural site to 68% on a medium
counted for 45 and 75%, respectively, of NPP in these site. For beech, there was less change, the proportion
two stands (Vogt et al., 1982). Other studies suggested of stemwood dropping from 41 to 39%; for pine, there
that from 15 to 24% of NPP may be allocated to my- was essentially no change: 34 to 35% (Duvigneaud,
corrhizal fungi (refs. in Sdderstr6m, 1991; see also 1971). Thus, in many species, there seems to be a
Fogel and Hunt, 1979). This is an excellent example of change in the allocation of primary production to dif-
the importance of mycorrhizae in ecosystem function ferent components of biomass as the quality of the en-
and of how a small component of an ecosystem can vironment changes. The investment in stems is
play a vital role. reduced whereas root biomass is increased on nutri-
For a detailed analysis of biomass and of productivi- ent-poor sites because a larger root system is required
ty in different types of forest ecosystem, the reader to satisfy the moisture and nutrient demands of the
should consult Art and Marks (1971), Ajtay et al. (1979), aboveground parts (Keyes and Grier, 1981). Nutrient-
O’Neill and De Angelis (1981), Gardner and Mankin demanding species show this change more sensitively
(1981), De Angelis et al. (1981), Edwards et al. (1981), than do non—nutrient-demanding species. Knowl-
Satoo and Madgwick (1982), and Cannell (1982). edge of this type of variation will become increasingly
important in forest management in the future.
Distribution of Biomass Between Different Parts
of Plants. The relative distribution of biomass be-
tween belowground and aboveground organs and be- 4.6 Production Ecology at Consumer
tween stem and reproductive organs is of great Trophic Levels
concern to farmers who grow such crops as potatoes
and corn. Similarly, the relative distribution of tree Herbivores
biomass among roots, stems, branches, and foliage is
of more than academic interest to foresters. High pro- Energy flow above the primary producer level is influ-
ductivity has little economic value if it is allocated to a enced by several factors. The efficiency with which
nonmarketable part of the plant. Where forest har- food resources are exploited, the efficiency of diges-
vesting is changing from the conventional utilization tion, and the loss of energy in respiration all influence
the rate at which energy is transferred from plants to
herbivores. Each of these parameters of energy flow
‘Symbiotic relationship between the fine roots ofplants and soil fungi. The are discussed in turn, referring to Figure 4-11 for clar-
mycorrhizal relationship is discussed in Chapters 5, 11, and 15. ification. The examples cited are from Odum (1957,
52 CHAPTER4 Production Ecology

3 o/b TL) Lee Ea ee

/OT LTE Tal ae Ty a a \
4
| pS
Sey aM
‘ ae 3 a
/ [hg Ap 2a Ape sass
/ i / Vee,
Energy Gross Net
/ / /f assimilated consumer: ooo)
consumer = the next sneutee for
trophic level
// / by consumers \ production production P

| / i Energy lost Energy from previous

/ ‘ony through = trophic level used in
/ 1 / / respiration creating net production
Ce ee a ae ne EM ey ee =~

\
Energy in the Energy available Energy ingested Energy not
previous trophic to consumers by consumers assimilated
level by consumers
Energy from previous
(net production)
trophic level that is
Unavailable Unused De Energy Excreted _ not used in this trophic
energy is energy * eliminated + energy ~ Chain, but may be used in
in feces other chains of the same
trophic web or in other
trophic webs

Figure 4-11 General pattern of energy flow from one trophic level to another. The energy of
net production in one trophic level proceeds through various stages to become incorporated in the
net production of the next trophic level of that trophic chain, to be transferred to another chain or
web, or to be lost as heat during respiration. Various ratios that have been used to describe the effi-
ciency of energy transfer at the various stages between two trophic levels are as follows:

1. Utilization (consumption or exploitation) efficiency

ingestion

if e : ef assimilation
2. Assimilation (digestion) efficiency = ——————
% ingestion

a , Si net production
3. Tissue growth (net production) efficiency = ————_—
: assimilation

net production
4. Ecological growth (gross production) efficiency = : 5
: ingestion

net production
wm . Trophic (ecological efficiency) =

For a comparison ofdifferent terms used to describe the efficiency of energy transfer, see Kozlovsky,
1968. (Used by permission of the Ecological Society of America.)

1962), Kozlovsky (1968), Golley (1960), Cowan tion that are sufficiently high to significantly reduce
(1962), Ricklefs (1973), and Wiegert and Evans (1967). subsequent primary productivity are unusual. They
We have already seen that the proportion of pri- are normally associated with herbivore populations
mary producer energy that is exploited by the primary that are in temporary outbreak condition.
consumer trophic level (the utilization efficiency) is The ratio of plant net production to plant biomass
highly variable but generally low for forest ecosys- is often used as a measure of the availability of plant
tems. The utilization of vegetation by herbivores is production to herbivores, because it reflects the rela-
determined by the palatability, nutritional quality, and tive allocation of production between woody and non-
physical availability of the plants to the herbivores and woody tissues. Table 4-4 shows production/biomass
by the abundance of the herbivores. Levels of utiliza- ratios for a number of ecosystems, and a comparison
 Production Ecology at Consumer Trophic Levels 53

Table 4-4 Net Production/Biomass Ratios for Several elephants and grasshoppers, which eat woody or ma-
Types of Ecosystem ture leaf materials that are much less digestible.
Ecosystem Net Production/Biomass High assimilation efficiency does not necessarily
imply high net production. Tisswe growth efficiency (or
Ocean 42 net production efficiency) expresses the efficiency
Coastal waters 35 with which assimilated energy is converted into net
production. It takes into account losses of assimilated
Lakes and streams 25
energy to respiration and excretion. Tissue growth ef-
Attached algae and estuaries 2 ficiencies are generally low, with a marked difference
Agricultural land 0.65 between the efficiency of invertebrates and verte-
brates. Terrestrial mammals have very low efficiencies;
Temperate grassland 0.33
values for terrestrial invertebrates are much higher,
Tundra and alpine 0.23 and aquatic invertebrates have the highest values. A
Savanna 0.18 similar term, ecological growth efficiency, expresses the
efficiency with which ingested energy is converted
Swamps and marshes 0.17
into net production; it accounts for both assimilation
Woodland and shrubland 0.10 efficiency and losses to respiration and excretion.
Temperate forest 0.04 The difference between assimilation efficiency and
ecological growth efficiency is accounted for by losses
Tropical forest 0.04
to respiration and excretion. Animals with a_ high as-
Boreal forest 0.04 similation efficiency, such as a sparrow or a field
Overmature old-growth temperate 0.01 mouse, often have lower ecological growth efficiencies
conifer forest, northwestern U.S. because of their high respiratory losses. More than
98% of the gross production of these two species is
From Grier and Logan, 1977; Whittaker and Woodwell, 1971. lost through respiration. Animals such as grasshoppers
and leafhoppers, however, have significantly higher
ecological growth efficiencies, despite lower assimila-
with Table 4-3 reveals that there is a close parallel be- tion efficiencies, because they are losing less than 70%
tween production/biomass ratios and consumption of of gross production to respiration.
NPP by herbivores. Agricultural ecosystems have the An overall figure for energy transfer through con-
highest production/biomass value for terrestrial sumer trophic levels or populations can be obtained by
ecosystems, a tribute to the success of agricultural calculating the trophic efficiency: the ratio of net produc-
management in promoting energy flow from plants to tion in one trophic level to net production in the previ-
consumers (human beings or domestic animals). ous trophic level. This efficiency can be calculated by
Ingestion of plants by herbivores does not neces- multiplying the net production of the previous trophic
sarily constitute a transfer of energy from producers to level or prey population by the percentage efficiencies
primary consumers. Not all ingested plant biomass of utilization, assimilation, and tissue growth of the
can be digested and assimilated by herbivores. Much level being studied to arrive at its net production.
of it passes through and out of the alimentary canal ‘Trophic efficiency represents the overall efficiency
undigested, although digestion is aided in many herbi- with which energy is passed from one trophic level to
vores by detritus organisms (bacteria) that live in the the next and is similar to the expression of energy flow
alimentary tract. The proportion of ingested energy used in the construction of energy-flow pyramids.
that is digested and assimilated into the consumer's The efficiency of transfer of plant net biomass pro-
body is called the assimilation efficiency. It varies be- duction into herbivore biomass in terrestrial ecosys-
tween different species and different ecosystems. High tems is very low, less than 1% in most cases. The
values are associated with foods such as seeds (con- explanation for this low trophic efficiency varies from
sumed by birds and small mammals) and aquatic algae one species to another. For some, it is primarily the re-
(consumed by aquatic herbivores) as these materials sult of a low utilization efficiency, whereas in others it
have a high proportion of digestible material. Low val- is more a function of high metabolic rates and respira-
ues are associated with browsers and grazers such as tory losses. The efficiency in aquatic ecosystems is
54 CHAPTER4 — Production Ecology

much higher because of the lower investment of ener- predators), whereas herbivores may cover long dis-
gy in supporting tissues, which results in greater uti- tances searching for food. With all these complications
lization and assimilation efficiencies. Because energy and with the relative scarcity of reliable data, one
is passed through aquatic trophic webs more efficient- should be cautious in drawing firm conclusions about
ly than through terrestrial webs, the former often have the relative magnitude of respiratory costs at the pri-
more trophic levels than the latter. mary and secondary consumer trophic levels.
The assimilation efficiency of carnivores is gener-
ally higher than that of herbivores. Much of the plant
Carnivores material ingested by the latter is lignin and cellulose,
The amount of energy diminishes rapidly as one pro- which are often inefficiently used. Conversely, only a
ceeds along a trophic chain; so does our knowledge of relatively small proportion of herbivore biomass is in-
that energy flow. We know a lot about energy at the digestible to carnivores. The degree of chemical re-
primary producer level and a lot about the herbivore arrangement between trophic levels is far greater
level, but relatively little is known about energy flow at between plants and herbivores than between herbi-
the secondary consumer level and above. This is part- vores and carnivores, and the change is correspond-
ly because agriculture and forestry have not consid- ingly less complete and more expensive in terms of
ered carnivores to be commercial species. It is also metabolic energy.
because there are fewer carnivores than herbivores,
and they are frequently difficult to study.
The available data suggest that loss of gross pro- 4.7 Detritus Trophic Chains
duction to respiration at the carnivore level is high, al-
though for terrestrial carnivores, it is not substantially We have already seen that in many terrestrial ecosys-
greater than for herbivores of comparable size and tems, especially forests, herbivores normally utilize
physiology. For aquatic carnivores, there does seem to only a modest proportion of the NPP. Because much
be a higher respiratory loss than for aquatic herbivores. of this unused net production is invested in short-lived
Intuitively, one might expect greater respiration tissues such as leaves and reproductive structures,
losses at the secondary consumer level. After all, preda- forests are characterized by substantial quantities of lit-
tors that must search for, pursue, overpower, and kill terfall. This provides the major energy source for the
the cautious, swift, and strong herbivores that are their saprotrophic organisms of the detritus trophic web,
prey will have a large metabolic bill to pay for all that which also receive inputs from higher up the grazing
work. Herbivores, conversely, are generally moving trophic web. Because of low assimilation efficiencies,
through continuous communities of food plants that much of the plant material ingested by herbivores is
neither run away nor put up any physical resistance to promptly redirected to the detritus web in the form of
being eaten. The situation requires more detailed con- feces, and additional transfers from the grazing to de-
sideration, however. Once a mammalian predator has tritus webs result from the death of herbivores and car-
captured an ungulate herbivore, it may not need to nivores and from carnivore defecation. The result is
catch another for many days. Large carnivores often that for many ecosystems, the detritus trophic web ac-
feed on the young, the old, the sick, or the maimed and counts for most of the postproducer energy flow.
thus avoid the high-energy cost of chasing and attack- Saprotrophic organisms such as seagulls, crabs,
ing fully grown, healthy adults. In contrast, the ungu- hyenas, and vultures are large and well known, but
late herbivore must spend almost all daylight hours most of the decomposition of dead organic matter is
searching for enough plant material of acceptable nu- accomplished by very small organisms: fungi and bac-
tritional quality and palatability to satisfy its energy teria. The biomass of these decomposer organisms is
needs; a large proportion of the plants in a particular often very small relative to the large quantity of ener-
community will not satisfy these requirements. The gy passing through this trophic level—the result of the
metabolic costs to herbivores of processing large quan- small size and short life span of many saprotrophic or-
tities of low-quality plant food is also higher than the ganisms. However, in some ecosystems, the biomass of
cost to carnivores of processing smaller quantities of saprotrophs may be substantial.
nutritious herbivore flesh. Some predators simply wait Soil fauna (e.g., worms, beetles, millipedes,
for their prey to come to them (spiders and ambush springtails) are important in the process of decompo-
 Detritus Trophic Chains — 55

sition. The soft tissues of entire leaves are protected be rapid when it is primarily bacterial, and slow when
from decomposition by the surface cuticle, and even if it is primarily fungal (Table 4-5). Bacteria tend to be
this is removed, the rate of decomposition is often dominant in the decomposition of deciduous an-
slow because the surface area of leaf fragments may be giosperm leaves, whereas fungi frequently dominate in
relatively small. Soil mesofauna break up leaves into the decomposition of the acidic litterfall of evergreen
large numbers of tiny pieces (a process called com- gymnosperms, but the mixing of forest floor materials
minution), which greatly increases the surface area by soil animals can alter the ratio of bacteria to fungi.
available to microorganisms, exposes the more easily Fungal hyphae can grow rapidly through the forest
decomposed tissues, and may chemically alter the ma- floor to find moist, fresh litter; they do not need to
terial in a way that improves it as a substrate for bacte- have fresh litter “delivered” to them by the mixing of
ria and fungi. Without this pretreatment, the fungi the forest floor by soil animals. In contrast, soil bacte-
and bacteria that actually do most of the decomposi- ria either are unable to move through the forest floor
tion are much less efficient. This can easily be demon- or do so slowly; they are greatly benefited by mixing,
strated by enclosing leaves in mesh bags of different which continually exposes them to new comminuted
mesh size and burying them in the forest floor. Leaves fragments of fresh litter. In addition, mixing by soil
in bags that exclude the mesofauna decompose much animals breaks up the fungal mycelium, reducing the
more slowly than do leaves in bags that permit their growth potential of the fungi and reducing the possi-
entry (Figure 4-12). bility of an antibiotic effect of the fungi on the bacter-
Soil fauna do more than comminute litterfall. ial community. Small soil animals are relatively
They mix the fragments of litter with mineral soil, ineffective in this regard, although they may affect
thereby enriching the surface mineral soil layers with fungal/bacterial ratios by feeding differentially on the
organic matter, and with a supply of nutrients when two groups of microbes. It is the medium-sized soil
the litter decomposes. Litter fragments in the mineral animals (mesofauna, e.g., worms, beetles, millipedes)
soil are less likely to dry out in the summer than litter that are most effective. As a result of these processes,
at the surface of the forest floor. Consequently, this forest floors with abundant mesofauna tend to have
mixing promotes decomposition. rapid, bacteria-dominated decomposition and high
Decomposition of litterfall varies in rate according fertility, whereas sites with few mesofauna because of
to the organisms responsible. Decomposition tends to summer droughts, very acidic conditions, or litter that
has very low nutritional value tend to have fungus-
dominated decomposition and low fertility. This is
100 discussed further in Chapter 11.
Litter decomposition is influenced by many fac-
~ oa . .
\_ Leaf pieces in 0.5 mm mesh bags
—

w~ ©6680 .
tors, including moisture, temperature, pH, the avail-
ob eR ability of O, the abundance of soil animals, the
2 UW Ave @) os =
aS 77 chemical and physical character of the litter, and the
s 60
|vo relative abundance of fungi and bacteria. These fac-
4 tors vary predictably between different geographical
soO 40 f areas, and it is possible to characterize different re-
a Leaf pieces in 7.0 mm mesh bags
S gions in terms of their rates of litter decomposition. A
o
120: simple way to express this rate is the ratio of the bio-
mass of litterfall to the biomass of organic forest floor
(Olson, 1963). There is an overall trend of increasing
wee
Se OMNI D a aboveground litterfall from the poles toward the equa-
Month tor (Vogt et al., 1986), whereas in many places, the
weight of forest floor shows the reverse trend. Where
Figure 4-12 Decomposition of pieces of oak leaves con-
it occurs, this pattern results partly because conditions
tained in mesh bags that either exclude or permit entry of
of temperature and moisture and the chemical compo-
medium- and large-soil animals into the bags, where they
can influence the rate of decomposition. (After Edwards and
sition of the litter become less favorable for the organ-
Heath, 1963. Used by permission of Dr. C. A. Edwards and isms responsible for decomposition as one travels
North-Holland Publishing Company, Amsterdam.) poleward from the equator. As a result, where these
56 CHAPTER 4 — Production Ecology

Bacteria and Fungi in

Table 4-5 Relationship Between Rate of Leaf Litter Decomposition and the Balance Between
‘Tennessee
Number of Number of
Bacterial Fungal
Colonies, Colonies,
Weight Loss Millions g ' Thousands g '
in 12 Months, Carbon/Nitrogen Dry Weight Dry Weight Bacteria/
% Ratio of Litter of Litter Fungi
Plant Species
ee
ei
Red mulberry 90 25 698 2650 264

Redbud 70 26 286 1870 148

White oak Bey 34 32 1880 ive

Loblolly pine 40 43 15 360 42

Data from Witkamp, 1966.

two trends occur, litterfall and NPP both show an in- tively. Residence times based on tota/ litterfall, includ-
verse relationship to the biomass of the forest floor, ing the annual production and death of fine roots and
whereas the ratio of forest floor to litterfall (which is associated mycorrhizal fungi, were reduced to 10 and
an index of resistance of litterfall to decomposition) 20 years, respectively (Vogt et al., 1983). Obviously,
increases as the weight of forest floor increases (see we will have to reexamine our understanding of forest
Figure 5—7). There are many exceptions to this attrac- ecosystem function as more data on fine-root produc-
tively simple relationship. Forest floors at high lati- tion and turnover become available, and simple com-
tudes may be thin because of frequent fires, whereas parisons between the easily observed aboveground
thick forest floors and peat deposits can be found at litterfall and forest floor mass may give us an incom-
tropical latitudes where nutrient-poor conditions re- plete picture of the true energy dynamics in a forest
sult in litterfall of very low chemical quality and slow ecosystem.
decomposition. Similarly, aboveground litterfall can Research on decomposition has focused on the im-
be appreciable in high-latitude forests on moist, fertile portance of litter quality (Aber et al., 1990; Prescott et
sites and can be very low in tropical forests on very in- al., 1993; ‘Taylor et al., 1991). The litter of different
fertile soils (Mailly and Margolis, 1992). The relation- plant species decomposes at different rates that reflect
ship is also confused because aboveground litterfall is the different physical and chemical characteristics
often not the major input of organic matter to the for- of these plants. Decomposition is accelerated by
est floor (Vogt et al., 1986). disturbances to the forest, such as clearcutting. This
The ratio of mass of forest floor to mass of annual improves the microclimatic conditions for decomposi-
litterfall has been called the organic matter residence tion, and it varies somewhat on sites of different nutri-
time (Gosz, 1976). Generally, residence times have ent and moisture status. Variations in decomposition
been based on aboveground litterfall. This can result between different climatic regions have been shown to
in overestimates of residence time (underestimates of reflect the variations in macroclimate (Meentemeyer,
decomposition rates) in ecosystems where there is a 1978). However, the chemistry and physical properties
large contribution of fine-root “litterfall.” In Pacific of litter seem to have an overriding influence on
silver fir stands in Washington, Grier et al. (1981) re- decomposition (Prescott, 1994). Differences in litter
ported annual aboveground and belowground litterfall decomposition rates between different forests are
in a 23-year-old stand as 6.5 and 11.8 t ha' respective- generally explained more by the differences in species
ly, with values of 4.6 and 12.2 in a 180-year-old stand. composition and their characteristic litter quality
Organic matter residence times in the forest floor between these forests than by other factors. In partic-
based on aboveground litterfall only were estimated as ular, it has been shown that there is a relation-
30 and 70 years in the young and old stands, respec- ship between both the lignin content and_ the
 Leaf Area, Foliage Production Efficiency, and Growth of Forests 57

lignin: nitrogen ratio of litter and the rate of weight area—sapwood basal area relationship, which reflects
loss (Aber et al., 1990; Aber and Melillo, 1982b; Berg the type of climate in which it grows (Table 4-6).
and McClaugherty, 1989; Harmon et al., 1990; ‘Taylor Leaf area exhibits some interesting relationships to
et al., 1991). stemwood biomass on both an individual tree and a
There is a series of trophic levels beyond the stand basis. Stemwood biomass shows a linear rela-
saprotrophs analogous to secondary and tertiary con- tionship to the projected leaf area of the tree; many
sumers (primary and secondary carnivores) of the species exhibit a similar relationship (Figure 4-15).
grazing food web. These levels involve larger organ- The potential maximum stemwood biomass accumu-
isms and therefore support a larger biomass than do lation in a stand also seems to be proportional to the
the saprotrophs. The latter are generally analogous to development of stand leaf area (Waring, 1979).
herbivores because they depend primarily on energy Although high values for accumulated biomass are
in plant products, although some saprotrophs obtain associated with high leaf areas, the efficiency with
their energy from the waste products and remains of which the biomass is produced does not show a linear
consumers and are therefore analogous to carnivores. relationship to leaf area. As leaf area increases, an in-
Secondary consumers in detritus food webs may feed creasing proportion of the phytomass is operating in
exclusively in this web, or they may feed in both detri- successively deeper shade (Figure 4-16) and therefore
tus and grazing food webs. Micro-invertebrates and contributes less and less to the net production of pho-
earthworms are restricted to feeding on the bacteria tosynthate. This lowers the overall efficiency of
and fungi decomposing soil organic matter, whereas growth per unit of leaf area as leaf area increases. The
an herbivore may consume the fruiting bodies of fungi result is that, in environments in which very high leaf
as well as the green parts of plants. Generally, the far- areas can accumulate (humid forests), basal area
ther up food webs one proceeds, the harder it is to cat- growth for the stand may peak at less than maximum
egorize a particular organism as a member of either a stand leaf area.
grazing or a detritus food web. In northern and temperate forests in which nitro-
gen availability is the major factor limiting growth, the
quantity of photosynthate produced per kilogram of
foliage per year—the foliage production efficiency—has
4.8 Leaf Area, Foliage Production been shown to be closely related to the quantity of ni-
Efficiency, and Growth of Forests trogen in the foliage. The nitrogen productivity (Agren,
1983a, 1985), or the foliage nitrogen efficiency (Comeau
Leaf Area and Kimmins, 1986, 1989; Kimmins, 1993)—the kilo-
Research in the Pacific Northwest has revealed some grams of NPP produced per kilogram of foliar nitro-
interesting relationships between the leaf area of forests gen—has been shown to be a useful predictor of NPP,
and their growth (Gholz et al., 1976; Grier and Run- and these relationships have been used in the modeling
ning, 1977; Waring et al., 1978, 1980). ‘Total leaf area is of forest ecosystems (Agren, 1983b; Kimmins, 1993;
reported to vary from 5 to 52 m’ per square meter of Kimmins et al., 1999; see also Chapter 21). These rela-
ground surface, the highest values being found at mid- tionships have also led to interest in the water use effi-
dle elevations, where winter snowpack accumulates and ciency and the nutrient use efficiency of plants: the
growing season temperatures are cool. Shrub and herb amount of NPP produced per unit of water transpired
leaf area varies from 3 to 14% of the total. Variation in and per unit of nutrients taken up, respectively. Exam-
leaf area has been shown to be closely correlated with ples of nutrient use efficiency studies can be found in
the water balance of the site (Figure 4-13); the drier the Birk and Vitousek (1986), Bockheim and Leide (1991),
site, the lower the leaf area. and Elliot and White (1993). Resource use efficiency
Leaf area is closely related to the basal area of sap- in general is discussed by Sheriff et al. (1995).
wood (Grier and Waring, 1974), and, once the rela- Estimates of the production efficiency of foliage
tionship for a species has been established (Figure depends on whether they are based on total produc-
4-14), foliage surface area per tree can easily be tion or aboveground production (Comeau and Kim-
estimated simply by boring the tree to obtain an in mins, 1986). Production efficiency is also affected by
crement core. Each tree species has a specific leaf the position of a tree’s crown in the canopy and by the
 60 FHT fir-hemlock transition
IV interior valley
J juniper
=
MC mixed conifer
MH mountain hemlock
> P-B ponderosa pine-bitterbrush
bo PP ponderosa pine
PSF Pacific silver fir
S Sagebrush
SS Sitka spruce
WH Western hemlock

20
leaf
Total
index,
area
m
m=

yi = 0.120x + 6.70

=)

E 200 a
j=

6 =

100 oa
)

Fe

0
0 100 200 300
2000 Precipitation, cm
(b)

i
e
“40
=
y= 0:289x-- 33184
© 63 r2 = 0.99
g 1000 6
IV

0, gS 0
West East Fe +10 ~20 —40 ~60 ~80
Vegetation zones Water balance index (cm)
(a) (c)

Figure 4-13 Total leaf area index (leaf surface area per square meter of ground) and its rela-
tionship to elevation and precipitation in 12 major vegetation zones in Oregon (after Grier and
Running, 1977). The vegetation zones lie along a west-to-east transect from the coast, over two
mountain ranges, to a dry interior plateau. (A) Leaf area of major vegetation zones. (B) Relationship
between leaf area and precipitation. (C) Relationship between leaf area and a water balance index for
five forest zones. The water balance index is computed by adding soil water storage to measured
growing season precipitation and then subtracting open-pan evaporation. (Used by permission of the
Ecological Society of America.)

58
 200

Sitka spruce
[e)

Oo

100 Noble fir Rae Douglas-fir

NS
Lodgepole pine oF Mountain hemlock
m2
leaf
Projected
area, kg
biomass,
Stemwood
S 5 Ponderosa pine
yo Juniper

0 50 100 150
0 100 200 300
: Projected leaf area, m7
Sapwood basal area, cm~
Figure 4-15 Relationship between biomass of stemwood
Figure 4-14 Leaf area of lodgepole pine (Pinus contorta) in and projected leaf area for individual specimens of seven
relation to sapwood basal area. Projected leaf area is the hor- tree species referred to in Tables 4-7 and 4-8. All of the
izontally projected area of foliage, as opposed to the total leaf trees had a dbh (diameter at breast height (1.3m)) of 20 cm and
surface area. The data are for individual trees. (After Waring, 314 cm? of sapwood basal area. (After Waring, 1980. Used by
1980. Used by permission of Dr. R. H. Waring.) permission of the New Zealand Forest Service and author.)

i>)

2
Table 4-6 Ratios of Projected Leaf Area to Sapwood
Basal Area for Selected Conifer Species Found along 5

a West-East Transect from the Humid Coast over

Two Mountain Ranges to an Arid Interior Plateau Fagus forest
in Oregon
Ratio of
Projected \ Castanopsis
Leaf Area \.__ forest
N
to Sapwood NN
Basal Area, Rain forest (Thailand) \
leaf
Cumulative
index,
area
m
m- x
Species Environment m? cm 7
10

Sitka spruce Humid 0.44

Douglas-fir Moderate 0.32

100 50 10 5) l
Noble fir Humid subalpine 0.27
Relative light intensity, %
Lodgepole pine Continental ONS
Figure 4-16 The intensity of light penetrating through a
Mountain hemlock Continental 0.16
forest decreases exponentially as the leaf area increases.
Ponderosa pine Semiarid 0.17 Each additional unit of leaf area index receives a rapidly dimin-
ishing quantity of solar energy and consequently contributes
Western juniper Arid 0.07
less and less to net growth. Note that the horizontal scale is log-
arithmic. (After Kira et al., 1969. Used by permission of Dr.
Source: Waring, 1980b. Used with permission of the New Zealand
Forest Service and author. ‘Tatuo Kira.)
60 CHAPTER 4 — Production Ecology

ratio of photosynthesis to respiration (Kimmins, 1993; Chapter 5, but the question of how much carbon is
Roberts and Long, 1992). stored in forest organic matter is of interest here.
These relationships have not yet been studied Some of the carbon in forests is accumulated in stand-
under a wide range of conditions, but because they ing dead trees (snags) and decaying logs (coarse woody
have important implications for the management of debris [CWD]) on the ground. These ecosystem
commercial tree crops, they will undoubtedly receive components also provide a habitat for certain animal
more attention in the future. species or for the insects and other animals that they
eat. Where there is a significant amount of CWD ina
forest, such as in humid and productive forests of the
4.9 Carbon Allocation and Carbon western United States and Canada, CWD plays an
Storage in Forests important role in the ecology of the forest (see reviews
by Caza, 1993; Harmon et al., 1986).
Carbon Allocation in Forests Forests constitute a major storage of carbon, al-
The importance of carbon allocation in production though there is still considerable uncertainty in all the
ecology has been mentioned several times. Although data on global carbon uptake by, storage in, and re-
our understanding of the controls of allocation to fine lease from forests. Globally, they account for approx-
roots is still far from complete, it has been clearly es- imately 1146 billion t of carbon: 37% of this is at low
tablished that nutrient availability plays a major role latitudes (tropical and subtropical forests), 14% in
(Ingestad and Kahr, 1985). Although there is an im- mid-latitudes, and 49% at high latitudes. More than
portant interaction between moisture and nutrients two thirds is in soils and peat deposits. Tropical defor-
(Grier et al., 1990; Khanna and Raison, 1990), in estation is believed to be causing a net release of ap-
many forests, nutrients seem to be more important proximately 1.6 billion t of carbon per year, compared
than moisture (Axelsson and Axelsson, 1986; Grier et with fossil fuel release of approximately 5.4 billion t
al., 1990; Linder, 1987) (Figure 4-17). However, there (1980 to 1984 data: refs. in Dixon et al., 1994).
is a complex interaction between moisture and nutri- The mass of decaying logs and snags in coniferous
ents. Poor tree nutrition asa result of lack of moisture forests is reported to vary from 1 to 490 t ha! and from
will reduce the quality and quantity of litterfall, reduc- 0 to 157 tha’, respectively. Deciduous forests general-
ing nutrient availability (Gosz, 1984). Lack of soil ly have much less CWD, with reported values of 0 to
moisture reduces soil animal abundance and activity, 38 and 0 to 12 t ha' for logs and snags, respectively
reducing litter comminution and soil mixing, promot- (Harmon et al., 1986). In some forests, the combina-
ing slowly decomposing, acidic, fungus-dominated tion of decaying logs and snags considerably exceeds
forest floors with lower rates of nutrient mineraliza- the mass and carbon storage in live trees.
tion, and leading to poor tree nutrition. All these fac- The carbon stored in U.S. forests increased by
tors, which are ultimately related to soil moisture 38% between 1952 and 1992 (Birdsey et al., 1993). A
deficits, combine to reduce nutrient availability, which similar pattern of increase has been reported from
in turn results in greater allocation to fine roots. De- Scandinavia and Europe. Canadian forests are simi-
tails of carbon allocation in conifers are reviewed by larly increasing in biomass and carbon storage (Kurz et
Luxmoore et al. (1995) and Gower et al. (1995). al., 1995; Pollard, 1991). Reforestation of abandoned
farmland is one reason for the increase, and reduction
of forest loss to wildfire is another. An increase in car-
Carbon Storage in Forests
bon storage in northern forests is believed to account
Many scientists are concerned about the danger of cli- for much of the substantial difference between carbon
mate change caused by the release of “greenhouse released to the atmosphere by fossil fuel burning (ap-
gases”: gases that absorb long-wavelength radiation proximately 5.4 billion t per year in the 1980s) and the
and contribute to the warming of the earth’s atmos- observed increase in atmospheric carbon (approxi-
phere. CO, is a greenhouse gas, and the balance mately 3.4 billion per year) (MacKenzie, 1994).
between releases of CO, from forests by fire, decom- There is a substantial release of carbon when car-
position, and respiration and uptake by photosynthesis bon-rich, old-growth forests that contain a lot of snags
is one of the factors that determine levels of CO, in and CWD are logged. Using a computer simula-
the atmosphere. This is discussed in more detail in tion model, Harmon et al. (1990) concluded that a
 Carbon Allocation and Carbon Storage in Forests 61

Treatment

Control Irrigation Fertilization Fertilization and

irrigation
10
(Above
ground) _] Stem and stump
__| Branches
8 = Foliage
i Fine roots (<2 mm)
LJ Coarse roots (>2 mm)

iL
Ss
Les 4
KS
of

}
a
o
5 2
5)
Si
3
E 0

o
s |

Ds 1/ 6.2 11.8 12.4

Total Production (Mg ha! yr_!)

Figure 4-17 Changes in carbon allocation by 10-year-old Scots pine trees in response to 6
years of irrigation and fertilization. Nitrogen was added annually, and phosphorus and potassium
were added every third year. Boron was added in year 5 to correct a boron deficiency induced by the
nitrogen additions. Annual irrigation resulted in little change in allocation or total annual produc-
tion. Fertilization more than doubled total production. Fertilization plus irrigation provided little
extra total production but increased aboveground allocation. Total allocation to fine roots was similar
for all treatments, but the percentage allocation to belowground was greatly reduced by the fertilizer
and fertilizer + irrigation treatments (Axelsson and Axelsson, 1986).

second-growth forest in western Oregon and Wash- total removal of carbon from the atmosphere over a
ington will not have the same carbon content as the 200-year period plus the carbon content of the second-
old growth that it replaced until it reaches at least 200 growth forest would exceed the carbon content of the
years of age. However, if the carbon content of har- old-growth forest. Under these conditions, the man-
vested wood (old-growth harvest plus at least two sec- aged forest would be a net carbon sink, even though
ond-growth harvests) is put into long-term storage the carbon stored in the second-growth forest at any
(e.g., long-lived houses, furniture, etc., or is buried in one time would be substantially less than in the old-
landfills where it decomposes very slowly), then the growth forest. However, Harmon et al. (1990) pointed
62 CHAPTER 4 — Production Ecology

out that if the carbon in the harvested wood is re- and land use change (see Can.J.For. Res. 1991, 21:87,
turned rapidly to the atmosphere by burning or de- 142. NZ.J;For: Sei. 1993, 232339, 412),
composition of short-lived wood products, then
converting the old-growth to second-growth forest Role of CWD in Forest Ecosystems
would result in a net release of carbon to the atmos-
phere. They claim that the proportion of tree biomass Public and scientific interest in mature and “old-
harvested, the proportion of the harvested products growth” forests in the Pacific Northwest United
manufactured into long-lived wood products, and the States in the late 1970s and 1980s identified large ac-
longevity of these products all would have to increase cumulations of decaying woody material and large
significantly from the levels of the 1980s to avoid a net standing dead trees (snags) as distinctive features of
release of carbon to the atmosphere caused by logging these forests. The high NPP, large size achieved by
West Coast old-growth forests. This has in fact hap- the trees, and the slow decomposition of large woody
pened. The fate of the harvested carbon thus deter- material result in accumulations of CWD in these
mines whether logging old-growth forests increases or forests that can exceed the living tree biomass of many
reduces atmospheric carbon levels. This topic is dis- other types of forest.
cussed again in Chapter 20. CWD, defined as snags, logs (dead tree stems on the
There seems to be fairly general agreement about ground), chunks of wood, large branches and roots, and
the significant release of carbon to the atmosphere stumps fulfill a variety of roles in the forest ecosystem.
caused by deforestation in the tropics. However, as
with much of what has been written about tropical Habitat for Animals. Snags provide important
forests, the more closely one examines the generaliza- habitat for birds and mammals that nest in cavities in
tions that are made, the more one finds great variabil- tree stems. Snags provide habitat and food for a vari-
ity in the ecological characteristics of tropical ety of insects, which in turn support birds, such as
ecosystems and how they respond to disturbance. For woodpeckers, that feed on them. Snags may also be
example, Desjardins et al. (1994) examined the carbon important as perches and nesting sites for large birds,
content of the soil beneath an undisturbed tropical such as eagles.
forest in the eastern Amazonian forest of Brazil and in Decaying logs provide habitat for a variety of in-
an adjacent 10-year-old pasture. As has been widely sects and insect-feeding animals. Because they tend to
reported, there was a loss of 30 to 55% of the original remain moist throughout the summer, large, well-
forest-stored soil carbon, depending on soil layer and decayed logs may be important habitats for moisture-
calculation method, but this was largely replaced by requiring amphibians, such as salamanders, in forests
soil carbon derived from the pasture grasses. The car- with dry summer climates and soils. Logs also play a
bon content of the 10-year-old pasture was 45 t ha! in role in the behavior of birds, such as grouse, which use
the upper 40 cm compared with 47 t ha! in the forest. them for territorial displays, and may provide protec-
Clearly, the total carbon of the pasture is much less tion and nest sites for small mammals and small birds,
than that of the forest because of the lack of trees and such as wrens.
CWD, but in this study, there was much less differ-
ence in the soil carbon than has often been suggested. Habitat for Plants. Large decaying logs provide a
Johnson (1992) also noted that earlier estimates and substrate for young tree seedlings in some old-growth
models of the effects of forest harvesting on soil or- forests (e.g., Harmon and Franklin, 1989). Dense un-
ganic matter, which assumed losses of 35 to 50%, are derstory vegetation and/or thick forest floors that dry
not supported by the recent literature. He concluded out in the summer may make it impossible for tree
that harvesting followed by reforestation has little ef- seedlings to get established on the ground in undis-
fect on soil carbon reserves on a global scale (Johnson, turbed mature stands in these forests. The moisture
1994). However, managed forests generally have much retained in a decayed log, the elevation of seedlings
less carbon stored in the forest floor, in standing dead above herbs and shrubs that are competing for light,
trees, and in CWD than unmanaged forests. A lot the reduction in the barrier to seedling establishment
more research is obviously needed to gain an accurate posed by a thick layer of moss, and the reduced com-
understanding of how the carbon storage in forests is petition for moisture and nutrients may render these
affected by natural disturbance, forest management, “nurse logs” the most favorable location for regenera-
 Carbon Allocation and Carbon Storage in Forests 63

tion in undisturbed old-growth forest. However, be- organic atmospheric nitrogen into organic forms that
cause of the low levels of limiting nutrients provided can be used by plants. Nitrogen fixation in logs is nor-
by decaying wood, seedlings whose roots are restricted mally low (often less than 1 or 2 kg"! ha! yr_') Jur-
entirely to nurse logs frequently grow very slowly and gensen et al., 1984, 1987, 1990), but in unmanaged,
are often much older than one might think. For exam- humid, west coast forests of Canada and northwestern
ple, 1-m-tall western redcedar and western hemlock United States that have mild winters and are subject to
saplings growing on logs in a northern Vancouver Is- massive windthrow events, nitrogen fixation in logs
land old-growth forest were found to be as much as 57 may nevertheless be a significant nitrogen input to the
years old and to average approximately 25 years (S. system. Much remains to be learned about the signifi-
Ignesti, personal communication, 1992). Thus, al- cance of asymbiotic nitrogen fixation in forest ecosys-
though large nurse logs play an important role in the tems, and it is too early to draw firm conclusions about
regeneration of trees in old-growth forests, this the importance of CWD in this regard.
process is generally too slow to make a significant con-
tribution to the regeneration of forests that have been Input of Logs to Streams. Streamside trees that
disturbed by natural processes or forest harvesting. fall across streams and small rivers play an important
CWD clearly plays a different role in tree regenera- role in creating a diversity of stream habitats as they
tion in undisturbed old-growth in comparison with become incorporated into the bed of the stream or
disturbed forests. river. In particular, large logs are often responsible for
the creation of alternating pool and riffle sections in
Contribution to Long-Term Soil Organic Matter streams. The diversity of stream organisms and the
and Soil Nutrients. Wood of many tree species de- total stream carrying capacity are often closely related
cays slowly, especially the heartwood. Wood is gener- to this physical stream diversity (Maser et al., 1988).
ally much more persistent than fine litter, such as
leaves and small branches, and therefore tends to ac- Regulation of Surface Runoff. The presence of
cumulate in the forest floor of unmanaged forests decaying logs on a slope can reduce the surface flow of
unless removed periodically by fire. Wood also con- water and promote infiltration. By trapping water on
tributes longer lived and more stable humus inputs to their upslope side, logs can significantly reduce surface
the soil than most types of fine litter. This may confer runoff and the accompanying risk of erosion.
on ecosystems that have large quantities of this mater- Much of the focus in studies of CWD has been on
ial a greater ability to resist ecosystem change in the decomposing logs and snags. The contribution of
face of some types of disturbance than ecosystems that stumps and roots has sometimes been overlooked, but
lack abundant CWD. they can account for a significant proportion of the
Decaying logs do not generally contain large total CWD on a site, and because they are largely
quantities of nitrogen and phosphorus—the two nutri- below ground, they tend not to dry out as much in the
ents that are most commonly limiting in forest ecosys- summer as do surface logs and snags. As a result, they
tems. Consequently, compared with the supply of may be more important in some ways and in some
nutrients from decomposing leaves and other fine lit- ecosystems than the surface CWD.
ter, logs generally make a small annual contribution to Considering all these roles, it will generally be de-
the nutrient supply for forest trees. However, this sup- sirable to retain some quantity of CWD in a managed
ply becomes much more important in a burned forest forest. This quantity will vary according to the type of
in which the fine litter and forest floor have been de- ecosystem, the tree species involved, the quantities of
stroyed or in a forest from which all the branches and CWD that are characteristic of that forest, the specif-
foliage have been harvested (e.g., by “whole-tree” har- ic CWD-related values that are to be sustained, and
vesting or litter raking). The annual contribution of the economics of leaving material that might other-
nutrients from the remaining fine litter in these situa- wise be harvested. Too much CWD can be a problem
tions lasts only a few years, whereas the small annual (Prescott and Weetman, 1994). Snags may be a safety
contributions from decaying logs may persist for many hazard for the public and forest workers. CWD may
decades or, in the case of very large logs, even cen- cause a fire hazard. Large quantities of CWD can lead
turies. In addition, some of the microorganisms that to soil acidification because of organic acids released
accomplish the decay oflogs have the ability to fix in- during decomposition and the acidic humus that is
64 CHAPTER4 — Production Ecology

produced when the wood of many tree species decom- The use of wood products for insulation reduces the
poses. Excessive quantities of CWD may inhibit forest use of fossil fuels for heating. The use of foliage and
regeneration. Clearly, the question of CWD manage- branches as cattle feed or of wood as a source of chem-
ment is very site and situation specific. As with most icals can reduce the energy cost of cattle rearing and
things about forests, generalizations are rarely useful chemical production. Wood can also be converted to
and should normally be avoided. combustible gases or liquids that can themselves be
used as fuel. Obviously, forests have an important role
to play in the human energy budget, although this is
4.10 Energy Benefit/Cost not without potentially serious problems, some of
Relationships of Forest Production which will be discussed in later chapters.
Energy benefit/cost analyses for contemporary
Intensive forest management, especially forest harvest- forestry are harder to obtain than comparable figures
ing, requires considerable amounts of energy. Road for agriculture. There is increasing interest in such in-
building, tree falling and bucking, log extraction, and formation, and in the next few years, we should see an
transportation to the mill all are energy-intensive ac- increase in such analyses. Tables 4-7 and4-8 present
tivities. As forestry becomes increasingly mechanized, some data used by Weyerhaeuser Company to evalu-
the energy requirements of harvesting can, in some ate the energy benefit/cost ratio for forestry. Douglas-
cases, increase by as much as 200%, although gen- fir stands managed to first commercial thinning age
erally the switch from manual to fully mechanized har- yielded a benefit/cost ratio of 554 for average manage-
vesting would result in an increased use of petroleum ment intensity and 40 for intensive management, ex-
products of only 15 to 30% (Bent et al., 1978). In- cluding the energy costs of harvest and transportation.
creased use of fertilization, the introduction of mecha- Taken to a full rotation (50 years) with intensive man-
nized thinning, and mechanized site preparation are agement (five fertilizations with urea nitrogen ferti-
trends that are likely to continue, all of which will in- lizer during the 20 years after commercial thinning),
crease the energy demands of forest practice. the ratio drops to 31. By comparison, average manage-
The processing of wood requires large amounts of ment to age 25 of a loblolly pine site would yield a ratio
energy, and in a forest-rich country such as Canada, of 264, compared with 13 for intensive management.
wood processing is the largest single industrial user of These ratios decline significantly when the energy
power. Until the past decade, disposal of wood waste costs of harvesting and delivery of logs to the mill, both
was a major problem for this industry, but the energy energy-intensive activities, are included. However,
crisis has changed this. Use of mill wood waste to gen- final delivered-to-the-mill benefit/cost ratios varying
erate electricity is increasing, and many sawmills are from 22 to 8 still compare favorably with agricultural
now both energy self-sufficient and sellers ofelectrical values. Forestry is undoubtedly a rela-tively energy-ef-
power to the local grid. ficient means of converting solar energy into biomass,
The energy crises of the 1970s stimulated an exam- harvesting the biomass, and converting it to a usable
ination of the concept of “the energy forest”: a forest product. Table 4-9 presents some energy benefit/cost
grown specifically to provide the fuel for a power sta- data for forest harvesting in Canada and Scandinavia.
tion. At the present price of oil, the energy forest is not Steinhart and Steinhart (1974) compared the energy
an economically viable prospect for most developed benefit/cost ratios of different food and fiber crop pro-
countries except in remote situations. However, fossil duction systems in agriculture and fisheries. Their
fuel is becoming more expensive and will eventually analysis confirms that forestry is generally a more
run out, and it is probable that energy forests will be- energy-efficient production system.
come a reality in some developed countries in the fu-
ture. In countries such as Brazil, this has already
happened. Large areas of fast-growing Eucalyptus have 4.11 Comparison of the Energy
been established there for the production of charcoal to Requirements of Wood and
be used to replace imported coal in the steel industry. Alternative Materials
Forests can also contribute to the solution of the
energy problem in other ways. Use of wood in build- Approximately half of all the wood consumed in the
ings requires less energy than other materials, such as world is used as fuel wood, and it is likely that we will
bricks, cement, or metal and plastic (see Chapter 20). see an increase in the use of wood waste and bioenergy
 Comparison of the Energy Requirements of Wood and Alternative Materials 65

Table 4-7 Energy Benefit/Cost Analysis of Forest Production

Average Fuel Energy Costs of Silviculture Energy Use by Silvicultural Activity

kcal Activity kcal Per

Tree propagation (per kg seed) Seed production 567,328 kg

Seed collection Nursery seedlings 78,624 1000
Wild 126,413 Site preparation
Seed orchard 137,083 Slashburn: manual ignition 84,686 ha
Seed processing 388,770 Slashburn: helicopter ignition 323,799 ha

Regeneration Mechanical piling 823,821 ha

Site preparation Roll and chop 564,159 ha
Southern U.S. (mechanical) Tree crush 564,159 ha
Piling 823,696 Ripping 564,159 ha
Crushing 564,159 Bedding 270,248 ha
Western U.S. (burning) 323,923 Planting
Nursery production
(per 1,000 seedlings) 78,624 Hand 388,560 ha
Greenhouse production
(per 1,000 seedlings) 41,198 Machine 306,987 ha
Planting (per ha) Herbicide (aerial spray) 435,884 ha
Hand 388,560 Spacing 210,470 ha
Machine 306,987 Fertilizer (urea-N) 9; 461,182 ha

Stand treatment (per ha)

Brush control 498,752
Spacing 210,470
Fertilization 9,461,182

Table 4-8 Plantation Benefit/Cost Analysis

Costs, kcal ha~! 10°
Biomas: s Energyin Benefit/Cost
Management Brush Fire Total Wood, Biomass, Ratios
ob
Species Age Intensity Preparation Planting Contribution Spacing Fertilization Protection Cost m* ha! kcal ha if

Douglas-fir 30° Average 324 388 ~ = 74 786 301 4.35 ¥ 10° 554 22

25° High 324 529 210 9,462 62 11,147 308 4.45 ¥ 10°

50° High 324 210 56,766 121 58,513 1,239 17.8 ¥ 10°

Loblolly pine 25° Average 887 = a 62 1,337 244 3.4 ¥ 10°

25° High 887 561 529 29,683° 112,633 62 37,903 339 4.9 ¥ 10°

Source: Courtesy Weyerhaeuser Co., Tacoma, Wash. 239 kcal = 1 MJ.

Energy content of wood produced/silvicultural energy costs.
Energy content of wood produced/silvicultural + harvest + transportation-to-mill energy costs.
“Age at first commercial thinning.
4Age at final harvest.
®Spacing and a commercial thinning.
66 CHAPTER4 — Production Ecology

Table 4-9 Comparison of Energy Benefit/Cost of Harvesting in Scandinavia and Canada

3
Energy 0 ofeee
egy Log Production, MJ m~
Location and Management Benefit /Cost
Intensity Harvest Transport Total Ratio*
Forest Type

Sweden High 144 218 362 19

pine-spruce Medium 37 41 78 90

Eastern Canada High 243 264 507 14

spruce-fir Medium 90 287 377 19

Western Canada High 262 142 404 17

Douglas-fir Medium 90 90 180 39

Western hemlock

Data from Ash et al., 1980.

“Energy equivalent of 1 m* of softwood - 6994 MJ; 1 MJ - 239 kcal.

tree plantations as a source of renewable energy in the with energy flow and storage. Many management
future. By displacing the use of fossil fuels, fuel wood practices influence energy in the forest, but this dis-
can make a useful contribution to reducing the threat cussion is limited to a consideration of the effects of
of global climate change caused by release of fossil clearcutting.
carbon to the atmosphere. This is not the only energy Clearcutting is a method of forest harvesting that
benefit that wood can confer, however. has been used throughout the world, particularly in
Energy is required to produce and market wood the harvest of previously unmanaged old-growth
products, but wood compares favorably to alternative forests. The method has been variously defined as the
materials. Replacing wood studs by steel studs in house removal of the entire standing crop, complete cutting,
frame construction increases the fossil fuel energy and a silvicultural system in which the old crop is
use by a factor of 9. Replacing a wooden floor by con- cleared over a considerable area at one time (Ford-
crete increases energy use 21 times. A brick veneer wall Robertson, 1971). Basically, clearcutting is the re-
replacing a plywood siding to a house increases the moval of all of the trees on the logged area at one
energy requirements by 30 times (Koch, 1992). Consid- harvest over an area large enough to remove the “for-
ering the extent of wood use in home construction, the est influence” from more than 50% of the harvested
anticipated future demand for construction timber, and area (Keenan and Kimmins 1993; Kimmins 1997).
the energy use implications of failing to satisfy that de-
mand, it is vital that we manage the world’s forests sus- Redistribution of Forest Biomass
tainably for timber as well as other values. Failure to
supply timber needs would lead to the use of other, The uncut mature forest represents an enormous ac-
cumulation of energy. Maximum aboveground bio-
more fossil fuel energy-demanding materials, and this
mass of 422, 575, and 415 t ha ! has been reported for
would have negative consequences for the global envi-
ronment (see Figures 20-5 and 20-6). temperate deciduous, temperate evergreen hardwood,
and tropical forests, respectively, whereas the biomass
of cool temperate coniferous forests in Japan and the
northeastern United States can exceed 600 t ha”! (Art
4.12 Effects of Forest Management on and Marks, 1971). An old-growth subalpine forest in
Energy in the Forest Ecosystem southwestern British Columbia was reported to have
an aboveground biomass of 731 t ha! (Krumlik,
Forest management generally involves the production 1979), whereas aboveground biomass in Douglas-fir,
of a crop, whether this is timber, wildlife, mushrooms, western hemlock, and noble fir forests in the Oregon
berries, or other materials. It is therefore concerned Cascade Mountains was reported to vary from 734 to
 Effects of Forest Management on Energy in the Forest Ecosystem 67

Table 4-10 Forest Floor Weights* for Various Forest Types?

Forest Floor
Dominant Weight
Location Vegetation Age, yr tha’ Reference

Oregon, U.S. Abies, Tsuga Old growth 61 Williams and Dyrness, 1967

(46 sites)
Alaska, U.S. Betula (7 sites) 25-120 41 van Cleve and Noonan, 1971
Washington, U.S. Conifers Old growth 103 Gessel and Balci, 1965
Ontario, Canada Pinus contorta 55 26 Foster and Morrison, 1976
Arizona, U.S. Pinus ponderosa 49 47 Klemmedson, 1975

California, U.S. Pinus radiata 30 60 Kittredge, 1940

New York, U.S. Pinus resinosa 37 37 Stutzbach et al., 1972

Massachusetts, U.S. Pinus strobus 34-96 43 Mader and Lull, 1968

Alaska, U.S. Populus (7 stands) 20-120 42 van Cleve and Noonan, 1971

New Hampshire, U.S. Acer, Fagus 200 89 Covington, 1976

New Mexico, U.S. Abies = 80 Wollom, 1973

Colorado, U.S. Abies lasiocarpa = 103 Snell et al., 1979

Germany Fagus 80 39 Ulrich in Cole and Rapp, 1981

Alaska, U.S. Picea mariana 55 ies} van Cleve in Cole and Rapp, 1981

New Brunswick, Canada Pinus banksiana 49 129 MacLean, 1978°

*Forest floor weight data are complicated by the failure to establish a common definition as to which root sizes are included in the esti-
mate and which are excluded as being plant rather than forest floor materials.
>The examples were chosen to represent upper limits. Many studies have reported lower values.
"Datum from a fire-killed stand. The fallen dead trees, most of which were not yet incorporated into the forest floor, were included in the
estimate. This provides a second illustration of the difficulties in comparing data from different studies.

1773 tha | (Zobel et al., 1976). Even higher values are crease in the energy flow through the detritus food
reported for coast redwood forests in northern Cali- web. Depending on the type of harvesting, variable
fornia, where a stand that exceeded 1,000 years in age quantities of tree biomass (s/ash) are deposited on the
had a stem biomass alone of 3461 t ha_'(Waring and ground. In whole-tree harvesting (removal of all above-
Franklin, 1979). Converted to energy at 4,000 cal g', ground biomass), very little slash is added to the forest
this would amount to 1.38 ¥ 10'° kcal ha ', or the floor (apart from the stumps and roots of the cut
equivalent of approximately 2,000 tons of coal. In ad- trees), whereas after clearcutting in some old-growth
dition to this is the energy contained in the forest forests in the Douglas-fir region of western Oregon
floor. Forest floor biomass before logging in the Ore- and Washington, the slash is reported to weigh be-
gon coast range has been reported to vary from 22 to tween 67 and 516 tha | (Dell and Ward, 1971).
85 tha |(Youngberg, 1966). Table 4-10 shows forest
floor biomass values for some other areas. Changes in Detritus Food
When a forest is clearcut, there are several impor- Web Energy Flow
tant changes in the distribution of energy and the
pathways of energy flow. First, by removing the over- After clearcutting, there is characteristically a period of
story canopy, the entry of energy into the ecosystem relatively rapid reduction in the thickness and biomass
by photosynthesis is temporarily more or less elimi- of the forest floor. For example, removal of tree cover
nated. Second, there is the potential for a large in- in an eastern deciduous hardwood forest resulted in a
68 CHAPTER4 Production Ecology

loss of 23% of the biomass and 3 cm of the depth of the 2. Temperature. Extreme temperatures limit decom-
forest floor over the first three post-devegetation years poser activity, and in areas with cold soils, clearcut-
(Dominski, 1971). This loss is due partly to the termi- ting will increase detritus energy flow by raising
nation of litterfall inputs and partly to the accelerated soil temperatures. Conversely, in areas with hot
decomposition that results from an increase in the flow summers, clearcutting can decrease decomposition
of energy through the detritus food web. by raising surface and/or slash temperatures to
Although there are reports of a marked decrease in lethal levels. Minimum and maximum tempera-
the forest floor after logging in both hardwoods (e.g., tures for decomposition are usually 2 and 40°C,
Drobikov, 1969) and conifers (Cole and Gessel, 1963; respectively. However, in some environments (¢.g.,
Shibota et al., 1951), there are also reports of either no below the snow in subalpine forests), decomposi-
loss (Carmean, 1959; Suchting and Christmann, 1935) tion may be active at approximately 0°C.
or only very small reductions even after 10 to 20 years 3. Aeration. Maximum decomposer activity requires
(e.g., Diebold, 1942; Johnson, 1994). good aeration to supply O, and prevent toxic accu-
Increased detritus energy flow after clearcutting is mulations of CO. Where the activity of nitrogen-
the combined result of both increases in the quantity fixing bacteria plays a role in decomposition,
of decomposable organic matter and changes in the exchange of N, may also be important. Where
condition of the forest floor. Summer daytime temper- harvesting causes soil compaction and/or water-
atures in the forest floor are warmer than before log- logging, anaerobic conditions that reduce decom-
ging because of the removal of tree shade, and there is position may be created.
no longer any phenolic-compound-rich throughfall
4. pH. Low pH encourages fungal activity, whereas bac-
from the overstory that can inhibit decomposer organ-
teria are important in decomposition under less
isms (King and Heath, 1967; Kowal, 1969). Cutting of
acidic conditions. Extreme pH values limit all de-
trees results in the breakdown of mycorrhizal relation-
composer activity; minimum and maximum values of
ships, a reduction in the mycorrhizal fungi, and an in-
pH for biological decomposition are approximately 4
crease in the activity of free-living saprotrophs. A
and 10, respectively. Forest floor pH usually increas-
study in New Zealand suggested that the fungal sym-
es after clearcutting as the result of the release of di-
bionts of Pinus radiata roots actually suppress the de-
valent cations by the process of mineralization.
composition of the pine needle litter. By cutting a
trench around small plots in a pine stand and thereby 5. Litter quality. Utilization of the energy in the forest
eliminating living mycorrhizal roots and their fungal floor requires that the microbes have access to an
symbionts, needle decomposition and loss of forest adequate supply of nutrients. If the carbon/nitro-
floor biomass were accelerated. A laboratory study gen ratio is too high, then the energy cannot
showed that this was probably due to the suppression be passed along the food web. Optimum rates of
of free-living, litter-decomposing microbes by mycor- decomposition occur at a ratio of readily avail-
rhizal symbionts (Gadgil and Gadgil, 1971, 1975). able carbon to nitrogen of approximately 25:1.
The extent to which energy flow increases in detri- Douglas-fir sapwood, heartwood, bark, and nee-
tus food webs after clearcutting depends on the effects dles have ratios of 548:1, 429:1, 491:1, and 58:1,
of the clearcutting on several factors (Bollen, 1974). respectively. These components decompose slowly
until the ratio approaches the optimum by loss of
1. Water. Optimum conditions for microbial activity carbon or gain of nitrogen. In comparison, rapidly
occur at 50% soil water-holding capacity. Excess decomposing alfalfa hay has a value of 18:1
moisture limits activity because of poor aeration, (Bollen, 1974). The carbon/nitrogen ratio is not
whereas decomposer organisms become inactive at always a reliable predictor of decomposition rates,
low moisture levels. Clearcutting can reduce the however. The presence or absence of various or-
rate of decomposition in hot, dry climates or ganic chemicals such as lignin and tannin can alter
where the clearcutting leads to soil waterlogging. the relationship between the carbon/nitrogen ratio
Minimum and maximum water contents for active and decomposition rate.
decomposition are 5 and 80%, respectively. When
logs and forest floors become dry, decomposition The acceleration in forest floor decomposition
becomes very slow. that characteristically follows clearcutting has been
 Summary 69

known for a long time (references in Lutz and Chan- speed with which these changes occur and the type of
dler, 1946) but has not been quantified for many forest litter inputs to the forest floor.
types. Generally, there is a period after harvesting
when forest floor biomass declines, after which it re-
builds toward predisturbance levels. Whether this Changes in Grazing Food Web Energy Flow
steady state is achieved depends on whether another Accompanying the increase in detritus food web en-
disturbance intervenes and again decreases forest floor ergy flow, there is a reduction in grazing food web en-
biomass. Studies in the Adirondack Mountains of New ergy flow. The degree of reduction is proportional to
York suggest that a reasonably steady-state forest floor the reduction in usable and accessible living plant bio-
biomass is reached under yellow birch—red spruce mass. Where all living plants are eliminated, so is the
stands at approximately 300 years of age with a value grazing food web. However, clearcuts often leave ad-
of approximately 265 tha’ '(McFee and Stone, 1965). vanced regeneration intact, and the shrubs and herbs
Attainment of steady state is believed to take longer in in the understory often respond rapidly to the in-
some western North American forests, if indeed it creased availability of light, moisture, and nutrients.
ever occurs. Fire, insects, disease, wind, or humans Coupled with the invasion of pioneer herbs and shrubs
generally intercede before steady state is reached. A of early succession," this rapidly reestablishes the graz-
value of 158 t ha | has been reported for old-growth ing food web, often at a higher level of energy flow
Douglas-fir (Gessel and Balci, 1965), whereas old- than prelogging. The greater physical accessibility,
growth subalpine forests in southwestern British Co- palatability, and nutritive value of the early seral vege-
lumbia have accumulations of up to 162 t ha! tation can result in more diverse, abundant, and pro-
(Kimmins, unpublished data). In some cool, humid ductive animal communities in clearcuts than in the
old-growth forests in coastal British Columbia that original uncut forest, as long as the animals’ require-
have been free of fire for many centuries, the depth of ments for shelter and winter range are satisfied.
the forest floor on some sites can exceed 1 m, which
would weigh in excess of 1,200 t ha' if this depth is
achieved over significant areas (based on a bulk den- SUMMARY
sity value of 0.12 g cm °). Energy is the single most essential ingredient of life, and
Loss of forest floor biomass and depth after all organisms are energy dependent. The quest for en-
clearcutting can have several effects on the ecosys- ergy pervades the lives of all living things; as a result,
tem. The water and nutrient storage capacity of the ecosystems are organized into sequences of energy de-
forest floor may be reduced, which can be undesir- pendencies that give an internal structure to the system.
able in hot, dry climates and for infertile sites but de- This structure is in the form of a trophic web, which can
sirable for cold mineral soils. On sites where the be divided into a number of major stages (trophic levels)
forest floor has adverse chemical properties or where of energy transfer.
deep layers of low bulk density material pose prob- The efficiency of energy transfer from one trophic
lems for regeneration (because it dries rapidly in the level to the next is generally low, but it tends to be greater
in aquatic than in terrestrial ecosystems. As a conse-
summer), the reduction in depth may favor the
quence, aquatic trophic webs tend to have more stages
reestablishment of a tree crop. (levels) than terrestrial trophic webs. The efficiency of
As the pioneer plant community reestablishes a fo- energy transfer at specific points along a trophic web de-
liage canopy, the entry of energy into the ecosystem is pends more on the size, metabolism, physiology, diet,
restored. This reestablishes litterfall inputs to the for- and habitat of the organism involved than on the particu-
est floor, much of which decomposes rapidly. The de- lar trophic level at which the transfer is taking place.
velopment of summer shading increases the moisture The biomass of any particular trophic level or popu-
levels in the slash and forest floor, which promotes de- lation depends on the balance of inputs and outputs of
composition, especially in hot, dry environments. energy. The size and longevity of organisms are impor-
However, as a tree canopy develops and succession tant determinants of biomass, but the contribution of a
particular group of organisms to energy flow is not pro-
proceeds, the type of litterfall changes, mycorrhizal
fungi begin to dominate the soil microflora, and de-
composition rates slowly return to prelogging values.
Replanting the area obviously influences both the *The terms “succession” and “seral” are discussed in Chapter 16.
70 CHAPTER 4 Production Ecology

portional to their biomass. Generally, the smaller the or- Forestry practices have become increasingly energy
ganisms, the greater their utilization of energy per unit intensive over the past three decades. Energy limitations
of biomass. will probably spur the development of “small tech-
Detritus trophic webs (which generally involve nology” energy self-sufficiency and biological solutions
smaller organisms than grazing trophic webs) account for to problems in forestry.
most of the post-producer energy flow in terrestrial Forest management has a significant impact on the
ecosystems, whereas grazing trophic webs tend to domi- quantity and distribution of energy in and the subsequent
nate in aquatic ecosystems. energy flows through the forest ecosystem. Production
Forests constitute one of the most efficient types of forestry is largely concerned with manipulating energy
terrestrial vegetation for harvesting solar energy, espe- flows, and the forester should therefore understand the
cially in inhospitable environments. They contribute to energy effects of management. In particular, care should
world NPP disproportionately to their area and make be taken to conserve soil organic matter and nutrients be-
up the majority of the world’s living plant biomass. For- cause they are so fundamental to all aspects of production
est productivity is closely related to leaf area, and the ecology.
high leaf area of forests is one of the reasons for their
surprisingly high NPP. Leaf area is ultimately deter-
mined by site moisture balance, but it seems that be- TAKE-HOME MESSAGE
cause most forests are growing under conditions of
nutrient limitations, the actual leaf area achieved is ac- Forest ecosystems are complex systems of living organ-
tually determined by nutrient availability. This empha- isms and atmospheric and mineral components that are
sizes the need to understand what determines nutrient maintained by inputs of solar energy. The internal orga-
availability and how it is affected by forest management nization and structure of an ecosystem are largely gov-
(Chapter 5). erned by the flow of energy through food webs and the
‘Total NPP in forests varies much less than conven- storage of energy in various living and nonliving ecosys-
tional measures of aboveground productivity would sug- tem components. To understand ecosystems, we have to
gest. Much of the aboveground variation across a local understand the ecology of organic matter production
landscape is due to changes in the allocation of NPP be- and storage at various different trophic levels and the
tween above- and belowground biomass as soil nutrient ecological determinants of energy transfers between
and moisture availability change. As our understanding trophic levels.
of the importance of resource allocation improves, we see Forests provide a variety of values to society, among
the critical importance of conserving soil organic matter which are timber products, employment, renewable
and soil nutrients for the achievement of sustainable for- bioenergy, and carbon sequestering and storage. These
est management. values all share one thing in common: their renewability
Sustainable management of terrestrial wildlife and depends on maintaining NPP and its allocation to long-
fish is determined just as much by our understanding of lived or harvestable biomass. Other forest values, such as
the determinants of energy flow through grazing and de- terrestrial wildlife and aquatic organisms, depend on pri-
tritus food webs as our knowledge of other determinants mary production because it provides food and habitat. In
of animal population ecology. Management of energy short, forest managers are in charge of a natural power
transfers from primary producers to consumer popula- station that harvests and stores solar energy in a form
tions lies at the heart of forest wildlife management and that is available for the economies of humans and other
management of fish during their period of residence in forest species.
forest streams. The key to sustaining NPP is the maintenance of ef-
By the beginning of the 1980s, it became apparent ficient leaf area, and the key to sustaining economic pro-
that after a brief absence (of approximately 100 years) duction of tree biomass is the management of production
from the energy scene, biomass is about to make a come- allocation. Both leaf area and its efficiency and produc-
back as a significant contributor to national energy bud- tion allocation are closely related to soil nutrient avail-
gets in developed countries. Developing nations never ability, so conservation of soil organic matter and
escaped from this dependency, and wood seems to be nutrients is fundamental to the successful management of
about to regain some ofits historic importance as a major production ecology.
factor in human cultural evolution. With increasing con- Management of secondary production and mainte-
cern about nuclear power and the threat to global climate nance of various measures of biodiversity are closely re-
caused by the combustion of fossil fuels, forests offer a lated to the supply of appropriate biomass energy to
source of bioenergy that is renewable and environmen- grazing and detritus food webs. Although the abundance
tally friendly. and diversity of animal and microbial species are affected
 Study Questions 71

by many factors, access to adequate supplies of organic i) . The laws of physics and our knowledge of physiol-
matter in the appropriate physical and chemicai forms is ogy suggest that the energy pyramid should always
certainly a fundamental determinant. be upright. Explain how the energy pyramid in the
Because of the relationship between some species of aquatic community of a forest stream could be in-
animals and microbes and the presence of woody debris verted.
in the forest, together with society’s preoccupation with
. Why are forests such productive ecosystems, and why
biodiversity, it is important to consider the role of CWD do they store so much biomass?
in ecosystems. The quantity of CWD varies greatly in
different forests, and decaying wood plays different roles . Why is leaf area so important in determining
in different ecosystem types. The need for and desirable ecosystem NPP?
quantity of standing dead trees and downed logs should . What determines carbon allocation in forests?
be assessed on an ecosystem-by-ecosystem basis. A single . What is the major loss pathway of NPP in many of
policy with respect to CWD that is applied to all forests, the world’s forest ecosystems? What are the other im-
irrespective of the ecological variation across the land- portant loss pathways?
scape, should be avoided.
Forest management can have a major influence on . What are the ultimate and proximal determinants
the amount of organic matter in the ecosystem and on of primary production ecology in most of the
forest leaf area. As a result, it can have a major effect, world’s forests?
either positive or negative, as far as our management . Which of the energy transfer efficiencies in consumer
objectives are concerned, on ecosystem productivity, trophic levels might a forester be able to influence?
resilience, and biodiversity. Forest management must What are some of the factors that determine the con-
consider organic matter conservation and forest leaf version of plant biomass into herbivore biomass?
area as key issues in the sustainable management of . What are the factors that determine litter decomposi-
forests. tion?
The forest ecosystem is a solar-powered biophysical
10. Why are estimates of rates of aboveground litter de-
system in which the availability of nutrients plays the
composition obtained from residence-time studies
dominant regulating role in the humid forests of the
inaccurate for many forests?
world and a coequal role in dry forests. Clearly, the next
stop on your journey through the forest ecosystem must 1 How might the character of the forest floor change if
be nutrients and nutrient cycling. suddenly there were no soil animals?
122 What type of land use has the highest energy bene-
fit/cost ratio?
STUDY QUESTIONS
13; How does forest management affect energy in the
1. Why are the biomass and energy-flow pyramids al- ecosystem?
ways the right way up in forests, whereas the num-
bers pyramid is often inverted? Why are the 14: Why is CWD believed to be important in forest
pyramids of numbers and biomass inverted in some ecosystems?
other types of ecosystem? 13: What is meant by foliage production efficiency?
 Biogeochemistry
Cycling of Nutrients in Ecosystems

5.1 Introduction recycling of the chemical elements associated with this

energy. Understanding one requires knowledge of the
Chapter 4 examined the entry of energy into, energy’s other. The study of the distribution and dynamics of
distribution and transfer within, and the ultimate loss these chemical elements is the focus of the ecological
of energy from ecosystems. This way of looking at subdiscipline known as biogeochemistry.
ecosystems may be strange to some people. Our sen- Energy enters, flows through, and is ultimately
ses of sight, hearing, touch, smell, and taste recognize lost from an ecosystem. It does not cycle within the
trees, shrubs, mosses, animals, fish, and birds rather ecosystem because it is not reused once it has been
than accumulations of solar energy at different troph- converted to heat. The chemical elements involved in
ic levels, yet this is what they are. We are aware of an- this energy flow behave differently. Once they are re-
imals eating plants and other animals and of fungi leased from their association with energy in an organ-
decomposing dead plant and animal material rather ic molecule, they are returned to the nonliving part of
than energy flowing through grazing or detritus the ecosystem, where they may again become available
trophic webs, yet this is what is happening. Living or- for uptake by plants. Once in plants, they are reunited
ganisms are fundamentally nothing more than accu- with solar energy in the form of a new organic mole-
mulations of solar energy in association with certain cule. Alternatively, they may be moved to another
chemical elements that have been sequestered from ecosystem or may go into long-term storage. Chemi-
the soil and atmosphere. When we express concern cals associated with energy flow (hereinafter referred
about the productivity of a forest or a farm, we are re- to as nutrients) are therefore cycled: they are reused
ally concerned about the flow of energy and the dy- within the ecosystem indefinitely, unless they are
namics of the associated nutrient chemicals. transferred to the cycle of another ecosystem or are
The beginnings oflife are thought to have been the converted to a long-term immobile form.
synthesis of organic molecules through the combina- The cycling of nutrients in ecosystems is complex.
tion of solar energy (or chemical energy from the inor- Some elements cycle predominantly between living
ganic environment) with appropriate assemblages of organisms and the atmosphere, whereas others gener-
atoms (e.g., Dickerson, 1978). Without energy, these ally cycle between organisms and the soil. Some ele-
atoms could not have been assembled into complex, ments follow both pathways. There is also an internal
high-energy molecules of living matter. Conversely, cycle within plants and animals that acts to conserve
without the appropriate atoms, the energy necessary nutrients within individual organisms. On the basis of
for life could not have been captured and stored. these differences, the cyclic movements of elements in
Movement and storage of energy in ecosystems are in- ecosystems can be assigned to one or more of three
separable from the accumulation, storage, transfer, and major types of cycles: the geochemical cycle, the bio-

759]
Th
 Introduction 73

ase \

Slope seepage
Soil erosion ——>>
Weathering
Soil leaching —3

ap Br
Y Y
Vy,

(a)

Cycle Nitrogen Phosphorus Potassium Calcium

Geochemical 16% 5% 12% 31%
Biogeochemical 45% 35% 66% 69%
Biochemical 39% 60% 22% Trace
(b)

Figure 5-1 (A) The three major types of nutrient cycle: geochemical (between ecosystems),
biogeochemical (within an ecosystem), and biochemical (within an organism; also referred to
as internal cycling). (B) Percentage of the nutrient dynamics in a forest ecosystem accounted
for by the three cycles. The relative importance of the three cycles varies for different nutrients.
Data for a 20-year-old loblolly pine plantation. (Data from Switzer and Nelson, 1972.)

geochemical cycle, and the biochemical or internal tances that vary from as little as 100 m to as much
cycle (Figure 5-1). as thousands of kilometers. Streamwater trans-
ports nutrients from forests to oceans, and water
1. Geochemical cycles: exchanges of chemicals between moving through the soil can carry nutrients from
ecosystems. Wind transports nutrients in dust and upslope to downslope ecosystems. Carbon dioxide
rain from one biogeocoenose to another over dis- (CO,) released from a respiring tree in one valley
74 CHAPTER 5 Biogeochemistry

may be blown over a mountain range to be reab- although animals exhibit similar physiological
sorbed by a photosynthesizing tree in the valley on functions. Nutrients are conserved within plants
the other side. The spatial scale of geological cy- by removing them from short-lived tissues such as
cles is generally large (greater than hundreds of leaves before they are shed and translocating them
meters), and the cycle generally does not follow to younger, actively growing tissues or to storage
the same spatial pathway repeatedly. Once a nutri- sites. Such transfers do not occur only at the time
ent has left a particular ecosystem, it will probably of tissue senescence. Plants are apparently contin-
never return. The time scale is generally long (mil- ually moving nutrients to where they are needed
lions of years in the case of nutrients deposited in most. Animals regulate the chemical composition
oceanic sediments), although it can be short, as in of excreta in a similar manner by removing re-
the case of CO;, which may enter a forest ecosys- quired nutrients from and adding unwanted or ex-
tem and leave again in a matter of hours. Alterna- cess chemicals to waste material before it is
tively, the CO, may be combined in organic matter eliminated from the body. Both the time and size
that remains undecomposed in the same ecosystem scales of biochemical cycles are much smaller than
for hundreds of years. the other two types of cycles because they occur
Biogeochemical cycles: exchanges of chemicals within an within individual organisms as a part of active
ecosystem. Nitrogen absorbed by tree roots from metabolic processes.
decomposing litter on the forest floor may be
translocated to the young developing leaves and In this chapter, we examine these three cycles. We
returned to the forest floor when these leaves be- then examine the biogeochemistry of some human ac-
come leaf litterfall. Potassium in the foliage of tivities and how biogeochemical information can be
shrubs may enter a grazing food chain when the useful in the management of the world’s forests. Be-
shrubs are browsed by a deer and may then be re- fore embarking on the discussion of cycles, it is neces-
turned to the forest floor in the urine of a moun- sary to consider the biological mechanisms by which
tain lion that catches and eats the deer in the same the nutrients in these cycles help determine productiv-
ecosystem. Once in the forest floor, the potassium ity and energy flow in ecosystems. ‘These mechanisms
is generally recovered efficiently by the shrubs or are referred to as plant and animal nutrition.
other vegetation. The spatial scale of biogeochem-
ical cycles is generally small, involving uptake from
the soil beneath an individual plant and return to 5.2 Plant Nutrition: Energy and
the same area. Animals, wind, and water may redis- Nutrients at the Primary Producer
tribute nutrients over longer distances within the Trophic Level
local ecosystem and can also transfer nutrients
from biogeochemical to geochemical cycles by Life may have had its beginnings in shallow bodies of
transporting the nutrients out of the local ecosys- water as the chance combination of certain atoms in
tem. The time scale of biogeochemical cycles is certain ratios. It is thought that under the influence of
generally shorter than geochemical cycles. It can energy from ultraviolet solar radiation or lightning,
be as short as a few hours, as in the case of potas- this could have led to the formation of organic mole-
stum uptake and subsequent loss by foliar leaching, cules. Such chance combinations would have occurred
or as long as thousands of years, as in the case of rarely, but it is thought that over millions of years,
calcium stored in the woody tissues of long-lived shallow lakes and coastal waters probably became di-
trees. Perhaps the major characteristic of biogeo- lute solutions of these organic molecules. Life itself is
chemical cycles, especially in forest ecosystems, is thought to have begun when certain of these complex
that most of the nutrients in the cycle normally re- molecules began to replicate themselves by acquiring
main within a particular ecosystem. They are effi- the energy and atoms stored in other organic mole-
ciently retained and accumulated with only modest cules. The earliest forms of life were probably similar
losses to geochemical cycles. to the anaerobic bacteria that we know today: sapro-
Ww Biochemical or internal cycles: redistribution of chemi- phytic heterotrophs utilizing fermentation as the
cals within individual organisms. The term internal means of liberating energy from organic molecules in
cycle has generally been used in reference to plants, their environment.
 Plant Nutrition: Energy and Nutrients at the Primary Producer Trophic Level ie)

Evolution has elaborated magnificently on this essential nutrient cannot be guessed from its concen-
original theme, but the basic mechanisms of energy tration. Some elements that are present at concentra-
acquisition and storage are still present in all living or- tions of only a few ppm (micronutrients) are as
ganisms. Before life, biosynthesis, or energy storage essential to the nutrition of the plant as elements that
and transfer can occur, certain critical combinations of are present at concentrations of several percent
atoms at the appropriate concentrations must be pre- (macronutrients).
sent. Life and energy flow are totally dependent on Characteristic concentrations of the 16 elements
the adequate nutrition of individual organisms. that are considered essential for the growth of most
plants are presented in Table 5-1. These figures repre-
sent average values for a wide variety of plants and will
Inorganic Chemistry of Plants therefore be inaccurate for many individual species.
Careful analysis will reveal that a large number of For example, there are some plant species for which
chemical elements can be found in plants. Many or calcium is a micronutrient rather than a macronutri-
even most of these elements are present at low con- ent. Some species require sodium or silicon as
centrations: as low as one part by weight of the ele- macronutrients. Most plants have no known physio-
ment to 1 billion parts by weight of the plant (one logical requirement for these elements even though
part per billion [ppb], or 10-’) or even one part per they are found at some level in all plants. Selenium is
trillion (ppt; 107’). Only a few are found at concen- toxic to most plants, but species that have evolved to
trations greater than one part per million (ppm; 10°), tolerate selenium-rich soils may require it. The rela-
and very few are found at concentrations that can be tive concentrations of different elements vary between
expressed as percentages by weight of plant substance different parts of the same plant, at different times of
(parts per hundred [%]). Many of the elements found the year, and with variation in the age and physiologi-
in plants have no known role in plant metabolism. cal condition of the plant. However, research has
Others must be present before the plant can grow and shown that in a greenhouse under conditions of opti-
function normally and are therefore referred to as mum nutrition, the nutrient requirements of young
essential nutrients. Whether a particular element is an trees varies little (Ericsson, 1994).

Table 5-1 Average Concentrations of Essential Nutrient Elements in Adequately Nourished Plants
Concentration Relative Abundance Average Concentration
Chemical Atomic in Dry Matter of Atoms in Dry in the Earth’s Crust,
Element Symbol Weight ppm % Matter ppm

Macronutrients
Hydrogen H 1.01 — 6 60,000,000 1,400
Carbon C 12.01 _ 45 40,000,000 200
Oxygen O 16.00 =_ 45 30,000,000 466,000
Nitrogen N 14.01 _ ko 1,000,000 20
Potassium K 39.10 _ 1.0 250,000 25,900
Calcium Ca 40.08 _ OS 125,000 36,300
Magnesium Mg 24.32 _ 0.3 80,000 20,900
Phosphorus 2 30.98 — 0.2 60,000 1,050
Sulfur S 32.07 — 0.1 30,000 260

Micronutrients
Chlorine Cl 35.46 100 — 3,000 130
Boron B 10.82 20 — 2,000 10
Iron Fe 55.85 100 = 2,000 50,000
Manganese Mn 54.94 50 — 1,000 950
Zinc Zn 65.38 20 _ 300 70
Copper Cu 63.54 6 — 100 Se)
Molybdenum Mo 95.95 0.1 = 1 Wes

After Epstein, 1972; Mason, 1966.

76 CHAPTER 5 Biogeochemistry

Acquisition of Nutrients Nuszdorfeér (1982) found that the overstory had a root
surface area of 7.3 to 11.9 m? m™ and a root length of
Approximately 96% of the dry weight of plants is ac- 3.8 to 6.5 km m™; understory values were 0.5 to 3.2 and
counted for by atoms of hydrogen, oxygen, carbon, 0.7 to 4.7, for ecosystem totals of 9.7 to 15.1 and 5.8 to
and nitrogen, the major chemical constituents of bio- 11.3 for area and length, respectively. On a per hectare
mass. The remaining 4% is divided unequally among basis, this length is equal to between one and two times
the other 12 essential nutrients and numerous the circumference of the earth! It is not surprising that
nonessential elements. Plants must obtain their nutri- plants have been described as “solar-powered, chemical
ents in the required proportions from sources in their machines that mine the soils for minerals.”
environment that have a very different composition, In addition to the investment in roots, most plants
and selective uptake from the lithosphere, the atmos- invest a significant proportion (as much as 15%; Séder-
phere, and the hydrosphere is an essential part of plant strém, 1991) of their net primary production (NPP) in
physiology. The source of different nutrients for up- supporting a mycorrhizal partnership with fungi to as-
take varies according to their availability to plants sist in soil-nutrient acquisition. It has been suggested
rather than according to their absolute abundance. that these symbiotic partnerships may have been a pre-
For example, although hydrogen and oxygen are requisite for the colonization of land by plants some
abundant in the lithosphere, they are largely in an un- 400 million years ago (Pyronzinski and Mallock, 1975).
available form, so plants must depend on the atmos- The acquisition of nutrients by fine roots and mycor-
phere and the hydrosphere for these two nutrients. rhizae and the costs and benefits to the plant of differ-
The earliest forms of plant life were aquatic. They
ent fine-root allocation strategies are discussed by
were bathed in dilute solutions of nutrients, and be- Oren and Sheriff (1995) and Yanai et al. (1995).
cause they were small, none of the cells was far from
the nutrient solution. Consequently, these plants did
not require any special morphological adaptations for Nutritional Deficiencies
nutrient acquisition. The situation on land is different, The definition of an adequate supply of a particular
and the development of land plants could not occur nutrient is difficult. It depends on the relative abun-
until the problem of nutrient acquisition on “dry” land dance and availability of other elements, the nutrition-
had been solved. The evolutionary answer to this al demands of the plant species, and the environmental
problem was not to develop different biochemical conditions under which they are growing. Nutritional
mechanisms of nutrient uptake but to make morpho- adequacy or deficiency has usually been defined em-
logical adaptations that permitted uptake from solu- pirically by observing the growth response of a plant
tion in the terrestrial environment. Roots exploited to additions of the nutrient in question (Figure 5-2),
the soil solution, and modified leaves permitted the Deficiency is defined as a condition in which an in-
entry of gases into intracellular spaces, where they crease in the availability of a nutrient results in an in-
were dissolved in thin films of water bathing the cells crease in growth or reproductive performance of the
and then absorbed. plant. As the availability of a nutrient to a deficient
The acquisition of an adequate, balanced supply of plant increases, growth will increase rapidly, but the
nutrients from the soil requires a major effort by terres- concentration of the nutrient in the plant will show
trial plants: the investment of a large proportion of their little change because of dilution of the newly absorbed
annual net production in the growth and maintenance nutrient by the newly captured carbon, which is pre-
of roots. Nutrient uptake from the thin layers of mois- sent in increased abundance because the improved nu-
ture surrounding soil particles requires intimate contact trition has increased the rate of photosynthesis. As the
between roots and the soil volume; to achieve this, condition of adequate nutrition is reached or if some
plants produce an extraordinarily large length and sur- other nutrient or environmental factor becomes limit-
face area of roots. A single rye plant grown for 4+ months ing to growth, uptake may continue and tissue con-
in approximately 0.05 mi} (2 ft’) of soil was found to have centrations will start to rise without any further
a root system with a total surface area of 639 m? (6875 increase in the rate of plant growth or biomass. If up-
ft) and a total length of 623 km (387 mi) (Dittmer, take continues without any increases in growth or re-
1937). In a study of small and fine roots (<5 mm diame- production, then the plant is said to be experiencing
ter) in subalpine forests of coastal British Columbia, “luxury consumption.” If high rates of luxury con-
 Plant Nutrition: Energy and Nutrients at the Primary Producer Trophic Level 77

Adequate nutrition Luxury consumption

100
a
RN
\
E 80 -4...-/ 4 it zone
Transition \.
5 | \ Toxic accumulation
a \ \
& | \
Arco
s
:
> ! \
s ae \
z eel
(tas)
\
m 4047/8 | \
z
pce
aw
=oe. Ml
depos
Bayi
2055 |
ak Critical concentration (concentration when growth
equals 90% of maximum)
0) ay ae ee |ae | ae

Concentration of nutrient in tissue

Figure 5-2 Relationship between the growth of a plant and the concentration of a nutrient
in its tissue. (Modified after Ulrich and Hill, 1967.) If an addition of the nutrient increases plant
growth but has little effect on the concentration of the nutrient in the plant, then the plant is nutri-
ent deficient. If such an addition results in little change in growth but an increase in concentration,
then the plant is adequately nourished. For an alternative method of assessing plant nutrition, see
Figure 11-13. (Reproduced by permission of Soil Science of America and A. Ulrich.)

sumption continue, then the element can accumulate trate nor ammonium ions are available to the plants.
to concentrations that are inimical to the plant, and In the presence of abundant inorganic nitrogen, the
growth may decline. cobalt requirement disappears. Sodium and strontium
The concentration of nutrients in foliage is used are not normally used in plant metabolism but can be
by both foresters and agriculturalists to gauge the nu- used to a limited extent as a substitute for potassium
tritional status of their crops. However, the concentra- and calcium, respectively, when these elements are in
tion of a particular nutrient in a particular tissue short supply. Manganese can be toxic in the absence of
cannot be used on its own as a direct indicator of plant silicon, so although silicon may not be an essential
nutrient status. “nutrient,” its presence may be necessary to permit
The absolute level at which a nutrient is deficient plant growth under conditions of high manganese
often depends on the level of other nutrients. For ex- availability.
ample, plants can be grown at very low levels of calci- It should be obvious from these examples that the
um when other divalent cations (ions with two positive relative abundance of different elements is often as
charges) are absent. As the availability of these other important as their absolute abundance. Recognition of
cations increases, the calcium requirement increases the importance of nutrient ratios in plant nutrition has
greatly, and unless the calcium supply improves, been attributed to the work of Ingestad (1979, 1981b,
growth ceases. Apparently, calcium acts not only as a 1982), Ingestad and Lund (1986), and Ingestad and
nutrient but also in the reduction of toxic effects of Agren (1988). The use of Ingestad ratios is discussed in
other cations. High levels of calcium in soil can induce Chapter 11, as is the use of foliar analysis in the diag-
phosphorus deficiencies, because at the high pH asso- nosis of the nutritional status of trees.
ciated with calcium-rich soils, much of the phospho- Deficiencies of essential nutrients can induce a va-
rus is tied up as insoluble calcium phosphate. riety of biochemical responses that result in abnormal
Similarly, the requirement for cobalt by plants with ni- growth and metabolic performance. Deficiency of
trogen-fixing root nodules is highest when neither ni- nitrogen, which is vitally involved in protein synthesis,
78 CHAPTERS — Biogeochemistry

affects the production of enzymes, amino acids, and the nutritional requirements of new growth the fol-
nonenzymatic proteins, thus affecting all aspects of lowing spring.
plant metabolism. The synthesis of chlorophyll (the Many of the world’s forests are growing on soils
plant pigment responsible for photosynthesis) is in- derived from rock materials that weather very slowly
hibited, and one of the first symptoms of nitrogen de- and/or release very few nutrients. Where the relative
ficiency is yellowing of the leaves. Shortage of abundance of the essential nutrients is favorable, this
phosphorus restricts the synthesis of ATP, which is in- need not necessarily result in low forest productivity.
volved in all metabolic energy transformations. This Forests have an impressive ability to accumulate and
immediately disrupts plant metabolism. Deficiencies recycle nutrients if left undisturbed, and, given time,
of iron and magnesium interfere with chlorophyll syn- they are frequently able to create a nutrient supply
thesis and cause yellowing of foliage (as do a variety of that will satisfy the requirements of at least moderate
other nutritional deficiencies). Calcium deficiency may growth rates. However, growth in many northern and
induce toxicity from other nutrients, and it results in high-elevation forests is limited by the low availability
weaker cell walls because calcium is present in the of nitrogen, and phosphorus deficiencies in various
middle lamellae of cell walls in the form of calcium parts of the world can severely limit growth. This is
pectate. Lack of calcium pectate reduces plant resis- both because of the lack of phosphorus and because
tance to moisture stress, and the soil acidity that is this lack inhibits the utilization of nitrogen (Beadle,
often associated with low soil calcium renders soluble 1966; Loveless, 1961). When the phosphorus defi-
several metals that may be toxic to plants at high con- ciency is alleviated, some of these forests subsequently
centration (e.g., iron, aluminum). exhibit symptoms of nitrogen deficiency.
The importance ofthe balance ofdifferent nutrients The nutrient requirements of different tree species
has not always been given adequate attention in forest vary greatly. Pines, for example, can photosynthesize
fertilization. Application of nitrogen fertilizers some- and grow efficiently on soils with very low levels of
times induces deficiencies of other nutrients, which lim- available nutrients. Conversely, many hardwood trees
its the tree growth response. The nutrient status of a will grow well only under conditions of moderate to
forest must be assessed before treatment to ensure that good nutrition. This variation in the nutritional toler-
an appropriate balance of nutrients is available. ances and requirements of plants plays an important
role in determining their spatial distribution. Pines are
frequently located on infertile, ridge-top soils, whereas
Forest Nutrition
hardwoods occupy richer valley soils. This does not
Plants are found growing in a wide variety of environ- necessarily mean that pines reach their greatest growth
ments, which vary from those that have abundant sup- on poor soils or that hardwoods could not grow on in-
plies of nutrients to those that have major deficits. fertile soils. The actual distribution of plants is fre-
Many forests fall into the latter category, and the sup- quently determined by plant competition, with
ply of nutrients for uptake by trees during the active nutrition acting indirectly by influencing the relative
period of spring growth may be particularly inade- competitive abilities of the various species on the dif-
quate. Plants are able to solve this problem to some ferent sites. However, there are cases in which plants do
extent by internal redistribution of nutrients from exhibit specific relationships to soil chemistry. Perhaps
storage sites or from older tissues to the sites of active some of the best examples of the ecological importance
growth and synthesis. This can be important for ele- of nutrition is seen in the difference in plant communi-
ments such as potassium, nitrogen, and phosphorus, ties that grow on calcium-rich and calcium-poor soils
which are relatively mobile within the plant, but not and on magnesium-rich (serpentine) and magnesium-
for elements such as boron and calcium, which are im- poor soils (see references in Epstein, 1972).
mobile. Deficiencies of the last two nutrients are par- Forest fertilization, a development of agricultural
ticularly marked in the actively growing parts of the fertilization, has proved to be highly successful in
plant because of the low mobility of these elements. many parts of the world, particularly in the case of mi-
Plants with low nutrient uptake in the spring will not cronutrient deficiencies in trees. Shortages of mi-
necessarily experience deficiencies if they are success- cronutrients such as zinc, boron, and molybdenum
ful in absorbing sufficient nutrients during the rest of have seriously limited forest growth in areas such as
the year to provide, by means of retranslocation, for New Zealand and Australia, and growth has increased
 Animal Nutrition: Energy and Nutrients at Consumer Trophic Levels 79

dramatically after additions of the deficient ele- the species, the part of the plant, the age of the plant,
ment(s). Spectacular increases in tree growth have fol- the time of year, and the physiological condition and
lowed the addition of phosphorus to areas such as the nutritional status of the plant. For example, amino
phosphorus-deficient Brazilian savanna. Nutrient de- acid composition of conifer foliage varies between
ficiencies may be harder to correct where the deficien- species, with the age of the tree and of the foliage,
cy is the result of low availability rather than low total with the flowering condition of the plant, and with the
quantity, as is the case with nitrogen. However, nitro- fertility of the site (Gagnon, 1966; Kimmins, 1971).
gen is the most widely used macronutrient fertilizer in The amino acid composition of several tree species
forestry, and although the economics of forest fertil- has been shown to vary according to whether the trees
ization raises questions about its future, the use of fer- obtain their nitrogen as nitrate or ammonium ions,
tilizers in forestry will probably grow if the value of and mineral nutrition is known to affect the total
forest products increases. amount and relative proportions of different soluble
Much of our present understanding of plant nutri- nitrogen compounds in trees (Durzan and Steward,
tion comes from studies of herbaceous agricultural 1967; Ebell and McMullen, 1970; Steinberg, 1951).
plants that were bred for high productivity and growth The inorganic nutrient composition of plants also
under conditions of high nutrient availability. Much varies greatly. For example, it has been found that the
less is known about nutritional strategies of wild plants concentrations of mineral elements in vegetables
growing under conditions of nutritional stress. Such grown in different states in the United States can vary
knowledge is essential, however, for forest manage- by a factor of 8(Hopkins and Eisen, 1959).
ment. A useful review of the nutrition of wild plants is The mineral content of tree foliage is known to re-
given by Chapin (1980), and Ericsson (1994) discusses flect the nutrient status of the soil, and the chemical
the nutrient requirements of forest crops. analysis of foliage has been widely used as a means of
assessing soil fertility and the levels of plant nutrition.
Wells and Metz (1963) found a twofold difference in
5.3 Animal Nutrition: Energy and foliar calcium and a threefold difference in magnesium
Nutrients at Consumer Trophic Levels concentrations in the foliage of loblolly pine growing
on different soils; there was also a threefold difference
Nutrition is just as important as a factor regulating the in foliar phosphorous levels in young foliage. The
production of consumers as it is for producers. Being mineral content of plants that grow on Scottish
heterotrophs, animals require more than a handful of heather moors was found to reflect the nutrient con-
mineral elements in certain critical proportions. They tent of the underlying soil parent materials. Heather
also require a wide variety of organic substances, in- that grew on soils derived from dioritic rocks con-
cluding amino acids, proteins, vitamins, fats, and car- tained higher levels of protein, phosphorus, and other
bohydrates. Some animals are more tolerant of varying minerals than did heather that grew nearby on soils
nutrition than others because they have populations of derived from granite (Jenkins et al., 1964; Robertson
saprophytic microorganisms in their digestive tracts. and Davis, 1965). Because of this dependence of plant
These microorganisms are able to synthesize some of chemistry on soil chemistry, herbivores that live on
the essential organic nutrients, which are then avail- areas underlain by different soil types or geological
able to the host animal. However, these symbiotic materials may experience widely different levels of nu-
(mutually beneficial) saprotrophs also have nutritional tritional value in their food supply.
tolerances and can synthesize these essential organic Despite the considerable variation in the organic
nutrients only when the host animal eats plant materi- and inorganic chemistry of the diets of herbivores, it is
als with an acceptable chemical composition. much less than the variation in the chemistry of the
plants in a community. This difference is the result of
selective feeding by animals. Insect herbivores have
Herbivores been shown to select the parts of the plants best suited
The nutritional quality of food material varies consid- to their nutritional needs and to vary their diet during
erably between herbivores and carnivores. Herbivores the year to maintain the best available diet in the face
eat plant materials that vary widely in their organic of changing plant chemistry (Beck, 1956; Blais, 1958).
and inorganic chemical composition, depending on Insects have also been shown to select from a number
80 CHAPTERS Biogeochemistry

of different artificial diets the one that best satisfies ing to the content of these elements in the plant food.
their nutritional needs (House, 1967). Sheep, cattle, Herbivores that feed entirely on old grass can become
and deer exhibit great discrimination in the selection deficient in phosphorus, especially in areas with low
of their food, and nutrition seems to be a major factor soil phosphorus levels. Female mammals can become
in this behavior (Albrecht, 1958). Experiments have depleted in calcium while nursing their young because
demonstrated that sheep can select plants with an av- of the high calcium demand of milk production, and
erage of 17% crude protein from a plant community the calcium requirements of antler production places
with an average of 7% crude protein, and deer that extra calcium demands on the males of antler-produc-
feed in a nutritionally marginal salal (Gau/theria shal- ing species. The ratio of fat to protein will also vary
Jon) plant community are able to locate and select the according to the nutrient status of the animal. Para-
2% of the shrubs that are nutritionally adequate meters such as age and health, time of year, and repro-
(Laukhart, 1962). ductive condition of the animal all influence body
The nitrogen content of plants is only one of many chemistry somewhat.
nutritional attributes that are important to herbivores.
However, because of its critical role in animal metabo-
Micronutrients
lism and because the nitrogen content of plants is
often low, herbivores often respond particularly to the Micronutrient deficiencies are as important in animals
nitrogen content of their food. A review of herbivory as they are in plants. Iodine, copper, selenium, and
in relation to plant nitrogen content can be found in cobalt are important micronutrients known to pro-
Mattson (1980). duce deficiency diseases when in low supply. Animals
Herbivores do not choose food plants solely on the have a fairly general requirement for sodium, but it is
basis of their inorganic and organic nutrient content. frequently lacking in the plant diets of herbivores. Be-
They also make their feeding selection on the basis of cause sodium is not an essential nutrient for plants, it
secondary plant chemicals. Plants that grow in nutri- is not conserved or accumulated by them, and sodium
tionally deficient (e.g., the subarctic and parts of the deficiency is common in herbivores (Orians and Pitel-
tropics) and other environments in which it is difficult ka, 1960). Farmers and wildlife managers solve this
to replace losses to herbivory have evolved chemical problem by providing mineral salt “licks” (lumps of
defenses against herbivores, which avoid plants with salt-rich mineral material), and wildlife frequently use
high concentrations of these secondary chemicals soil or rocks rich in sodium as natural salt licks. Wild
(Bryant and Kuropat, 1980; and Bryant et al., 1983; animals are widely known to have an appetite for salt,
Chapin, 1980). Physical defenses such as leaf toughness and animals that live in sodium-deficient environ-
and hairiness may also be important (Coley, 1983). ments have been known to chew the wooden handles
of tools or the canvas straps of backpacks that have
been soaked with salty human sweat (Braestrup, 1940;
Carnivores
Blair-West et al., 1968). Potassium is also important in
The chemical composition of the food of carnivores is animal nutrition, and potassium deficiencies can occur
much less variable than that of herbivores. Despite the in areas of high rainfall because this element is readily
considerable variation in dietary items consumed by lost from plants by leaching.
herbivores and the large variations in the chemistry of
those items, the average chemical composition of dif-
ferent species of large mammal differs surprisingly Importance of Nutrition in Animal Ecology
little (Davis, 1968). In elephants, sheep, humans, hip- Animal nutrition has long been recognized by agricul-
popotamuses, and horses, mineral elements account tural zoologists as a critical determinant of animal
for approximately 5% of the total biomass. Of this productivity. In contrast, some animal ecologists have
5%, more than two thirds is made up of calcium and been slow to recognize the fundamental role of nutri-
phosphorus, the two mineral nutrients of prime con- tion in animal ecology in unmanaged or nonagricul-
cern in the nutrition of domestic livestock and wild tural ecosystems. For example, Hairston et al. (1960)
mammalian herbivores. stated that because herbivore populations are nearly
The chemical composition of herbivores can vary, always limited in size despite the presence of an abun-
however. Phosphorus and calcium levels differ accord- dant, intact, green plant community, herbivores could
 The Geochemical Cycle: Nutrient Cycling Between Ecosystems 81

not be food limited. However, from the discussion ecosystem and their deposition in another in which
above, it should be clear that the productivity of herbi- they may remain indefinitely or from which they may
vore populations can be limited in the presence of be transferred to yet other ecosystems. These cycles
abundant quantities of plant biomass if the chemical constitute the inputs to and the losses from biogeo-
quality is inadequate; both the utilization efficiency chemical cycles, and they play an important role in de-
and the assimilation efficiency of herbivores are influ- termining the quantity of nutrients cycling within an
enced by plant chemistry. It has been shown in experi- ecosystem and therefore in ecosystem energy flow.
ments with an insect herbivore that a 50% dilution of For purposes of discussion, it is convenient to catego-
an optimum diet with indigestible materials can result rize geochemical cycles as either gaseous or sedimentary.
in a 20% increase in food intake but a lower total as-
similation of protein, and a similar response has been
demonstrated for mice (Dalton, 1963; House, 1965). Gaseous Cycles
Sheep are able to digest and assimilate 20 to 30% of Carbon, hydrogen, oxygen, nitrogen, and sulfur all
the protein in their ingested food when it contains 3 to can enter or leave ecosystems as gases or vapors, as
5% protein. When the protein content of ingested solids or in solution. However, for nitrogen, carbon,
food rises to 20 to 25%, the assimilation efficiency in- and oxygen, the gaseous state is the predominant form
creases to 70 to 80% (Orians and Pitelka, 1960). of entry. Most types of rock contain little or no nitro-
There have been many reports of animals and birds gen (some sedimentary deposits may contain some),
dying with stomachs and crops full of indigestible (nu- the oxygen in rocks is chemically bound, and carbon
tritionally inadequate) plant materials, and the ravages either is present at low levels or is released too slowly
of malnutrition in people who subsist exclusively on to satisfy the carbon requirements of plants (release of
white rice or white flour are known all too well. The carbon from limestone deposits may be an exception).
chemical form in which energy is presented to a con- For sulfur, however, there is a substantial input from
sumer ultimately dictates whether that energy is meta- rock weathering. Sulfur does enter ecosystems as a
bolically available to that consumer. gas, but in many ecosystems this is much less impor-
There is growing recognition of the importance of tant than its entry as sulfates in solution. Plants in
plant chemistry, nutritional and otherwise, as a regula- urban or industrial areas where there is abundant sul-
tor of the distribution and abundance of animals fur dioxide (SO,) in the atmosphere may take up a lot
(Bryant and Kuropat, 1980; Mattson, 1980; Price et of this gas through their leaves. Much of the sulfur
al., 1980), and it is likely that future theories about the that leaves ecosystems is in the form of ions dissolved
regulation of animal numbers will reflect this recogni- in stream water, but there may also be a significant
tion (Bryant et al., 1983; Haukioja, 1980; Kimmins, proportion in the form of gas.
1970; White, 1969, 1974, 1976). The uptake of CO, from the air by plants is well
known. Somewhat less well known is the uptake of at-
mospheric SO), and direct uptake of gaseous nitrogen
5.4 The Geochemical Cycle: in the form of ammonia (NH;) has been demon-
Nutrient Cycling Between Ecosystems strated. Most nitrogen enters ecosystems by the mi-
crobial fixation of nitrogen gas (N>), but uptake of
The geochemical cycle (or geological cycle as it is atmospheric NH; may provide up to 10% of the plant
sometimes called) involves exchanges of chemicals be- community’s nitrogen requirements and amount to as
tween different ecosystems. These may be as close to- much as 20 kg ha? yr! (Hutchinson et al., 1972). CO,
gether as a mountain slope above a valley floor or as and SO, are emitted by plants, whereas animals emit
far apart as an ocean and the center of a continent. CO, and the reduced gases hydrogen sulfide (H,S)
The term cycle implies a repeated movement through a and methane (CH,). The contribution of animals to
cyclic pathway—the removal of nutrients from one the carbon cycle has sometimes been underestimated
ecosystem to another and the subsequent return to the by the failure to recognize the considerable production
original ecosystem. Although this can occur for some of methane. For example, it has been estimated (with-
of the elements that participate in the geochemical out the trace of a smile) that large herbivorous mam-
cycle, for most it does not. Geochemical cycling gen- mals release between 45 and 73 million tons a year of
erally involves the removal of chemicals from one methane as flatulence (Ehhalt, 1973)! This figure
82 CHAPTER 5 Biogeochemistry

seems small, however, compared with the estimate mann et al., 1993; van der Kamp, 1986). As a result,
that termites in the tropical and subtropical regions forests may be able to accumulate significant quanti-
annually convert 37% of net primary production into ties of nitrogen in the absence of symbiotic nitrogen
50 billion tons of CO, and 152 million tons of CH, fixation and major inputs in precipitation. Gaseous
(Zimmerman, 1982). This release of CO, exceeds the losses of nitrogen are also important. Until recently, it
annual release from the combustion of fossil fuels. was believed that denitrification was relatively unim-
Gaseous cycles have attracted a great deal of atten- portant in well-drained, acidic forest floors and soils of
tion over the past 20 years. Not only do they account humid forests. It is now recognized that gaseous losses
for some of the major inputs and losses of macronutri- of nitrogen from forests are an important geochemical
ents to and from ecosystems, but they also are the pathway in many forests (Davidson et al., 1990; Mar-
pathway by which society’s gaseous pollutants move tin, 1985; Struwe and Kjoller, 1989), especially during
through the global environment. Very large quantities the “assart period” after ecosystem disturbance (see
of carbon monoxide, carbon dioxide, and the oxides of the section “Increased Availability of Nutrients: The
sulfur and nitrogen enter the atmosphere daily as the Assart Effect” and Figure 5-16 and 5-17).
result of human activities, and a wide variety of or-
ganic chemicals and pesticides spend some time mov-
Sedimentary Cycles
ing through these gaseous cycles. As we shall see later,
the consequences of such large-scale additions to Although only a few chemicals are involved in gaseous
gaseous cycles and the fate of the added chemicals are cycles, all chemicals participate in the sedimentary
of urgent concern. In particular, the addition of large type of geochemical cycle. For those elements that do
quantities of the oxides of nitrogen and sulfur to the have a gaseous phase, the relative importance of the
atmosphere has resulted in the phenomenon of acid two geochemical pathways depends on the physical
rain, which was recognized during the 1970s to be one and chemical character of the element and its biologi-
of the most serious forms of pollution in heavily in- cal role and the nature of the environment. For exam-
dustrialized and urbanized parts of the world. Acid ple, in a dry area, much of the carbon and sulfur that
rain is discussed in more detail in the section “Acid leave an ecosystem will be in the form of gases. In a
Rain and Its Effects on Forests and Lakes.” wet area, a lot of the gaseous oxides of carbon and sul-
For many, perhaps most, of the forests of the fur will be taken into solution and removed in
world, nitrogen is the nutrient that is in most limiting streamwater, and substantial amounts of dissolved or-
supply to plants. Nitrogen is the main nutrient applied ganic carbon (DOC) may leave in drainage waters.
in most forest fertilization treatments, although phos- The Rio Negro (black river) of southern Venezuela
phorus and micronutrient fertilization is certainly very and northern Brazil (the northern Amazon) is black
important where these nutrients are deficient (Madg- because of large amounts of the tropical forest NPP
wick, 1990). A great deal of research has been done on leaving the ecosystem as DOC. The brown color of
nitrogen availability in soils and nitrogen cycling in water draining peatlands around the world reflects the
ecosystems, but there is a growing realization that our importance of DOC in the global carbon cycle.
knowledge of the biogeochemistry of nitrogen is in- Sedimentary cycles involve several different mech-
complete because of our inadequate understanding of anisms of movement: meteorological, biological, and
the gaseous exchanges of nitrogen in the geochemical geological/hydrological.
cycle. Symbiotic fixation of atmospheric nitrogen is
usually the major nitrogen input in forests that have Meteorological Mechanisms. Meteorological mech-
symbiotic nitrogen-fixing plants, but in forests that anisms include inputs in dust and precipitation (rain
lack such species, asymbiotic fixation may play a major and snow) and outputs as the result of wind erosion
role. Asymbiotic fixation is known to occur in decay- and transportation. Dust and pollen from land and salt
ing wood (Harmon et al., 1986; Heath et al., 1988; spray from oceans are carried by wind to be deposited
Hendrickson, 1991; Jurgensen et al., 1987), but evi- in some distant ecosystem during periods of precipita-
dence is accumulating that it also occurs in the forest tion or calm weather. Extreme examples of the poten-
floor (Nohrstedt, 1988), in the rhizosphere and by tial of the meteorological pathway are dust and
bacteria associated with mycorrhizal fungi (Niu et al., sandstorms and the deep layers of /oess soil or the sand
1989), and even in the stemwood of living trees (Bor- dunes that are produced. Ash from the 1980 eruption
 The Geochemical Cycle: Nutrient Cycling Between Ecosystems 83

Table 5—2 Fallout of Chemicals in Various Parts of the World (Listed in Order of Nitrogen Inputs),*” kg ha™! yr

Location NH,* z NO,” P K Ca Na Mg tS) Cl

Europe 0.9-14.3 0.3-7.3

England 8.2-14.0 0.3-0.9 3.1-4.8 9.0-13.1 25.5-35.1 3.2-5.4 12
Mississippi 11.2 0.3 4.0 5.0 1.0 3.5
Germany 4.8-5.7 2.9-4.1 0.1 1.7-2.1 gy.
Denmark 4.6 2.3 3.4 6.5 16.1 12.9 26
New Hampshire 2.2 (5/4) 0.7 2.6 15) 0.6 14.8
Scotland 0.9-4.7 0.4 -2.0 1.4-2.5 5.9-6.7 11.5-47.2 1.2-4.6 7.5-18.3 20-82
Belgium 1.8 4.2 2.9 9.1 2.3
U.S.S.R. 5.6 0.5 Ua 15.4 2.5
Sweden 0.9-3.7 0.5-1.5 0.6-3.7 2.6-13.9
New Zealand 5.0 0.4 7.0 9.0 68 12 10 134
Haney, B.C.,
Canada 18 2.4 183 4.6 5.4 allt 6.6 122
Vancouver, B.C.,
Canada 0.6 ileal 0.4 0.9 8} 13.2 2.2 6.7 23.1
Washington lol 0.8 2.8
Oregon O:9=120 0.3 0.1-0.2 2.3-5.0 1.8 1.0
Coweta, U.S. 3.2 6.0 5.4 ise
Fire Island, N.Y. U8 9.8 141.5 19.1

*Data from Art et al., 1974; Crisp, 1966; Feller, 1974; Frederiksen, 1972; Gore, 1968; Grier et al., 1974; Holden, 1966; Jensen, 1962; Likens et al.,
1970; Miller, 1979; Switzer and Nelson, 1972; Weetman and Webber, 1972; and Zeman (1978).
Values are probably underestimations because of failure to measure impact of aerosols on vegetation (see the text).

of Mt. St. Helens in Washington state was carried er) and wet fallout (dust, aerosols, and dissolved chem-
around the globe, and the dirty color of old snow and icals in rainfall, mist, or snow) results in a continuous
of some glaciers demonstrates the considerable depo- input of nutrients into ecosystems that is small in
sition of dust from the atmosphere even in periods some ecosystems and large in others. The quantity of
when there is little volcanic activity. On a less dramat- nutrient elements in fallout varies at different times of
ic scale, the dust from logging roads can result in an the year, in different years, and in different places. It is
appreciable deposition of chemicals on nearby forests. greatly affected by the climate, the type of weather
In a single year, 1000 kg ha! of windborne material (which determines the source of air masses), and loca-
was deposited 80 m from the edge of a wood in Den- tion relative to the sources of the fallout chemicals
mark (Holstener-Jorgensen, 1960). A total of 0.7 kg such as the ocean, areas of active wind erosion of soil,
ha! of calcium and 0.1 kg ha"! of potassium was de- and areas of industrial air pollution.
posited 20 m from a forest road in Sweden over a 2- The input of various chemicals in fallout has been
week period in the spring (Iamm and Troedsson, measured in many parts of the world. Table 5-2! gives
1955). The annual deposition of nutrients in dust from some example values for terrestrial ecosystems. Al-
North American logging roads has not been mea- though this is not an exhaustive list and higher and
sured, but it must constitute an appreciable input to
roadside ecosystems in some areas.
The combination of dry fallout (dust and aerosols 1A much more complete set of data on many aspects offorest biogeochem-
settling out of the atmosphere during calm, dry weath- istry from the earlier literature can befound in Kimmuins et al., 1985.
84 CHAPTERS — Biogeochemistry

lower values occur, it gives some idea of the variability The data in Table 5-2 are for annual chemical in-
of this part of the geochemical cycle. There is great puts. If these rates of input are sustained, then forest
variability even within small countries such as Scot- ecosystems will receive substantial quantities of chemi-
land, reflecting the variable distance of sampling sites cals over the life of a tree crop (a rotation, which is nor-
downwind (prevailing wind direction) from the ocean, mally approximately 50 to 120 years). Where the
industrial centers, and areas of arable farming: three nutrients are efficiently retained, these inputs may be
major sources of fallout chemicals. Nitrogen inputs sufficient to supply much of the annual requirements
range from less than 1 kg ha! yr“ in areas immed- of the forest biota for certain nutrients. Where this is
iately downwind of large oceans to more than 21 kg the case, it is possible for plant and animal communi-
ha’! yr! in areas downwind of large industrial com- ties to develop in relative nutritional independence of
plexes. Annual atmospheric deposits of nitrogen in the soil. The soil is still vitally important, of course, be-
west and central European forests in the early 1980s cause it determines in part the extent to which the nu-
were reported to be 10 to 70 kg ha! (Moseholm et al., trients added in rainfall and dust are retained and made
1986). Phosphorus inputs are generally much smaller available to plants. It is also important in water rela-
and less variable than for the other macronutrients, tionships and in providing anchorage for plants. How-
varying between 0.1 and 0.9 kg ha"! yr“!. Potassium in- ever, the phenomenon of chemical fallout is one of the
puts vary between 0.1 and 7.7 kg hat yr‘. Calcium in- factors that make possible the growth of productive
puts tend to be larger at between 2.3 and 5.2 kg ha! forest on mineral soils with a very low nutrient status.
yr', whereas magnesium inputs tend to be much
lower (0.6 to 5.4 kg ha"! yr“). Values for sulfur input Biological Mechanisms. Redistribution of nutri-
lie between 3.5 and 18.3 kg ha! yr‘. Chloride and ents between ecosystems can occur as the result of an-
sodium inputs reach extremely high values close to imal migrations. Many of the animals that regularly
coastlines but are much lower inland. participate in biogeochemical cycles also become in-
Many of these values will actually be underesti- volved in geochemical cycles because they feed in one
mates of atmospheric nutrient inputs to forests. They ecosystem and defecate in another. Many species of
are based on precipitation collected in rain gauges, flocking birds feed in agricultural areas during the day
which do not collect fine rain, windblown rain, or dust but return to woodlands to roost at night. In one
as effectively as does a tree crown. For example, a woodland in England, this resulted in an importation
study of a very windy exposed site in the White Moun- of approximately 6.1, 9.5, and 89.2 kg ha"! of sodium,
tains of the northeastern United States revealed that potassium, and calcium by rooks over an 8-week
positioning a plastic artificial foliage structure above period, compared with an annual input in rain of ap-
the precipitation collector increased the catch of proximately 11, 4, and 24 kg ha'', respectively (Weir,
water, calcium, magnesium, sodium, and potassium by 1969). The nighttime droppings of starling flocks in
factors of 4.5, 8.3, 6.0, 4.9, and 5.7, respectively such woodlands can sometimes accumulate to a depth
(Schlesinger and Reiners, 1974). A similar study in a of several centimeters and kill the minor vegetation, or
much less windy location at low elevation in coastal even the trees. However, in most terrestrial ecosys-
British Columbia showed that artificial foliage posi- tems, biological exports from one ecosystem will be
tioned above a precipitation collector increased the balanced (or nearly so) by biological imports from an-
collection of water by 8% and nutrient quantities by other. Migratory birds that feed temporarily in an area
25 to 150% (DeCatanzaro and Binkley, 1981). Al- will remove some nutrients but may defecate similar
though one cannot use these factors to estimate the amounts that they imported from elsewhere. A study
true atmospheric inputs to a forested area, the results in a northern temperate hardwood forest ecosystem in
of these and similar studies indicate that-the data in the United States found that feeding by migratory
Table 5-2 are minimum estimates of atmospheric in- birds resulted in an annual net removal of only approx-
puts to forested ecosystems. imately 3 g ha” of calcium and nitrogen and less than
The values in Table 5-2 are for areas with ample 2 g ha of sulfur and phosphorus (Sturges et al., 1974).
precipitation. Lower inputs can be expected in lower- Snowy owls that feed on lemmings in the arctic
rainfall areas, although dry fallout (i.e., dust) may cause a marked nutrient enrichment of the area
compensate to a variable extent. around their nest because of the importation of lem-
 The Geochemical Cycle: Nutrient Cycling Between Ecosystems 85

ming carcasses (and the nutrients contained therein) patterns in those ecosystems, and in many cases these
from the surrounding tundra. Salmon that have fed alterations are not desirable.
and grown in the ocean migrate back up rivers to the
stream from which they originated, carrying with Geological/Hydrological Mechanisms. Geological/
them nutrients from the ocean. Much of this nutrient hydrological mechanisms involve inputs of nutrients to
input enters local terrestrial biogeochemical cycles an ecosystem by chemical weathering of rock and soil
when bears and eagles feed on the spawned-out minerals or as nutrients dissolved in soil water or
salmon. Alternatively, the nutrients may temporarily streamwater that is moving into the ecosystem. These
enter stream biogeochemical cycles before being car- mechanisms also involve outputs from an ecosystem of
ried slowly back to the ocean. Some of them may be nutrients dissolved in soil water or surface water or car-
recovered by roots of streambank plants that grow ried as particles in the form of eroded soil and organic
into the stream. matter.
Perhaps the best example of the biotic contribution Despite the sometimes considerable contributions
to a geochemical cycle is the formation of phosphate of meteorological and biological inputs, the major
deposits on the so-called gwano islands off the west supply of many nutrients for the biogeochemical cycle
coast of South America. Phosphate arrives in the vicin- of many ecosystems is the geological process of weath-
ity of these islands in nutrient-rich seawater upwelling ering, erosion, and solution. Soil is formed by the
from deep in the Pacific Ocean. The phosphorus en- physical and chemical breakdown of rock materials
ters a productive grazing trophic web that leads even- under the combined influence of climatic and biologi-
tually to carnivorous seabirds. These catch fish at sea cal processes. Nutrients that are released into solution
but roost and nest on the islands. Fish carcasses, bird during this process either enter a biogeochemical
droppings, and dead birds accumulate on the islands in cycle or are removed by erosion (wind or water) or so-
the almost rainless climate at a rate that has been esti- lution and water drainage. The natural rate of weath-
mated to be 190,000 metric tons per year, containing ering of rock or soil minerals and the subsequent
8,800 metric tons of phosphorus (Hutchinson, 1950). release of nutrients is difficult to measure, and there
Thus, seabirds are the critical link in building up a vast are few reliable direct estimates (Table 5-3).
deposit of phosphorus, a process analogous to the bio- The magnitude of this input to the geochemical
logical events that are thought to have been involved cycle has usually been estimated indirectly as the dif-
in creating many of the world’s sedimentary deposits of ference between the quantity of chemicals entering
phosphate, carbon, and calcium. the ecosystem by meteorological and_ biological
Agricultural and forest management activities pro- processes and the quantity of chemicals leaving in
vide another example of biotic contributions to geo- water, assuming no change in the quantity of chemi-
chemical cycles. Fertilizers are dug up or manufactured cals stored in the system. By basing the estimate on a
in one ecosystem and distributed in another. Some of small watershed with a watertight bedrock, the differ-
the accumulated capital of nutrients in a forest or agri- ence is equal to the net export of chemicals from the
cultural crop is removed at the time of harvest. Eventu- area in streamwater. Estimates derived in this way can
ally, these nutrients find their way via combustion or underestimate the rate of weathering because the as-
waste disposal into the soil of some remote ecosystem sumption of a watertight bedrock is often incorrect,
or via sewers into remote water bodies. although the error is partially offset by underestimates
Humans differ from other animals in that we not of atmospheric inputs (discussed above). The estimate
only redistribute nutrients but also concentrate, redis- also assumes that the biomass (and its nutrient con-
tribute, and disperse a lot of nonnutrient elements and tent) of the watershed is neither increasing nor de-
chemical compounds. Geological cycles form an intri- creasing. Over the short period during which the
cate and amazingly well-balanced system that provides estimate is made (a few years), this may well be in
both terrestrial and aquatic ecosystems with sufficient error. Losses of water to deep seepage and short-term
nutrient inputs and outflows to keep their biogeo- gains or losses of nutrients seem to occur in most
chemical cycles running smoothly. However, we are ecosystems (Curry, 1972). A final shortcoming of this
altering the geochemical inputs to, and losses from, method of estimating weathering rates results from bi-
many ecosystems in a way that alters the energy flow ological and chemical processes within streams that
86 CHAPTER 5 Biogeochemustry

Tuble 5-3 Estimates of the Rate of Release of Nutrients from Primary Minerals by Weathering, kg ha™ yr“
Forest Type Bedrock Type Location P K Ca Mg Reference

Northern hardwoods Moraine/gneiss Hubbard Brook, N.H. = at 2A 3.5 Likens et al., 1977

Mixed hardwoods Outwash sands Long Island, N.Y. — dialer 24.2 8.4 Woodwell and
Whittaker, 1967

Mixed hardwoods Schists Maryland _ Pass) 1.3 Til Cleaves et al., 1970
Serpentine Maryland -— Trace Trace 34.1 Cleaves et al., 1974

Coastal hardwoods Quartz sand Fire Island, N.Y. — 0.01 0.04 0.01 Art et al., 1974

Aspen-mixed Glacial till Wisconsin 0.9 3.6 cil — Boyle et al., 1973
hardwoods

West coast conifers Plutonic rocks Southwestern British 0.33 har 34.4 6.6 Zeman, 1973
Columbia

Cascade conifers Dolomite White Mountains, Calif. — 4 86 52 Marchand, 1971

Adamellite — 8 it 2

Cascade conifers Tuffs/brecias Oregon = 1.6 47 11.6 Fredriksen, 1972

Tropical rain forest Holocene alluvial Venezuela

deposits Soil age: 1,000 yr _ 5.3-6.7 0.6-1.4 1.3-1.7 Hase and Foelster,
Soil age: 5,000 yr — 1.1-1.3 0.1-0.3 0.3 1983

take some of the nutrients leached from terrestrial (see the section “Nutrient Cycling in Tropical Forest
into aquatic ecosystems out of solution (Cummins, Ecosystems”). Losses of silicate and sodium, which
1980; Perrin, 1981; Triska and Cromak, 1980). These tend to reflect chemical weathering of rock and soil
problems can cumulatively result in a significant error more accurately than do macronutrients such as calci-
in estimates of geochemical weathering obtained by um and phosphorus (Johnson, 1971), are variable.
studying the balance between atmospheric inputs and This is the result of differences in both bedrock chem-
outputs in streamwater. Mineral weathering inputs are istry and rates of weathering.
reviewed by van Miegroet et al. (1994) and Zabrowski
et al. (1994).
Table 5—4 presents some examples of net losses of
several chemicals from undisturbed, humid, temper-
5.5 “The Biogeochemical Cycle:
ate, forested watersheds. Losses of inorganic nitrogen Nutrient Cycling Within an Ecosystem
dissolved in streamwater that drains from undisturbed
forest ecosystems are nearly always less than precipita- Biogeochemical cycling involves a continuing cyclical
tion inputs, especially when the forests are relatively exchange of chemicals between the biota and the
young, resulting in negative net output values. This physical environment within an ecosystem (Figure
reflects the much greater importance of gaseous cycles 5-1). It is the sequence of transfers of chemicals that
for inorganic nitrogen. However, concentrations of occurs between the time an element enters a particular
dissolved organic nitrogen in the soil solution that ecosystem and the time it leaves. Because nutrients in
drains from humid forests are often much higher than this cycle follow a continuous pathway, we could start
the inorganic nitrogen concentrations. Neither this the discussion at any point, but we shall begin with the
loss nor the fate of this nitrogen has been quantified uptake of nutrients by plants from the soil.
extensively. Very little loss of phosphorus or potassium
occurs, and in several ecosystems, there is net accumu-
lation of these elements (negative net output). Calci- Nutrient Uptake by Plants
um and magnesium are lost in considerable quantities In common with all other life forms, plants satisfy
from most temperate ecosystems, which are normally much of their nutrient requirements by direct absorp-
well supplied with these elements, but they are effi- tion of nutrients from solution in water. However, up-
ciently retained in calcitum-deficient tropical forests take of nutrients by roots from soil solution is not the
 The Biogeochemical Cycle: Nutrient Cycling Within an Ecosystem 87

Table >—4 Net Losses of Several Chemicals from Undisturbed Forest Watersheds in Streamwater (Total Stream
Chemical Content Minus Precipitation and Dry Fallout Chemical Content) in Various Parts of the Humid
Temperate Regions of the World (Listed in Order of Calcium Loss)*”

Preston ot Hydrological Balance, mm N

Study Area Precipitation Discharge NH,* NO, P K Ca Mg Na _ S&S Cl Si

Oak Ridge, Tenn. 1280 529 - — -1.6 586 46.7 -1.1 — = =

Oregon 2330 1525 -0.5 0.3 We 2/0) Wilde, 27) = — 99:3
Oregon — — -0.5 0.3 2.1 48.0 — = = = =
Great Britain® 2130 1731 -5.3 -0.1t003 59 448 — 19.7 — = =
Coastal B.C.,
Canada 4541 3668 -0.03 -0.3 0.1 If S445 6:6 12.47 0:85) 15.0 9128
Coastal B.C.,
Canada 2288 1080 -1.2 -19 0.0 OG “Ke 29 G2 af Of =
Finland? — - -4.0 -0.2 Pray S040) eH0) EY
New Hampshire 1372 850 -2.0 2.9 — et Gs BGS SH) SLO) Ue
Japan _ _ -4.9 -0.4 0.4 38 10 — _ - -
Coweta, U.S. 2035 1072 — = Poi Oe ss, ABB} - =-

“Data from Cleaves et al., 1970; Crisp, 1966; Fredrickson, 1972; Grier et al., 1974; lwatsubo and Tsutsumi, 1968; Johnson and Swank, 1973;
Likens et al., 1977; Miller and Williams, 1968; Swank and Elwood, 1971; Viro, 1953; and Zeman, 1973.
Values may be underestimations if there is appreciable undetected loss to deep seepage. They may be overestimations if precipitation inputs are
significantly underestimated (see the text). Negative values indicate less output in streams than input in precipitation and dry fallout, indicating
net accumulation in the ecosystem.
°Nonforested ecosystem.
4Estimate for the entire country.

only mechanism of plant nutrition or necessarily the and NH, dissolve in the film of moisture that covers
most important. In many soils, water contains low the mesophyll cells inside the leaf and then move into
concentrations of macronutrients, and there may be the cells in solution. Oren and Sheriff (1995) reviewed
relatively little direct uptake from solution. Direct up- nutrient acquisition by tree roots and canopies.
take can occur from soil minerals that are in intimate
contact with roots (actually a root-mediated geochem- From Soil Solution. A wide variety of physical and
ical input), and in some ecosystems, roots that are in chemical characteristics of the soil will influence the
contact with weathering rock surfaces are able to ob- concentration of nutrients in and their uptake from
tain much of their mineral nutrients directly from the soil water, but three major factors determine nutrient
rock (lithoponics; see Chapter 11). Uptake can also uptake by roots from soil solution.
occur through leaves if they are in contact with a nutri-
ent in solution, and this ability is used to advantage in 1. The rate at which nutrients diffuse from the sur-
some types of agricultural and horticultural fertiliza- rounding soil to the root (diffusion transfer)
tion practice in which nutrient solutions are sprayed 2. The rate at which water that contains nutrients
onto the foliage. Uptake of carbon is almost entirely moves from the surrounding soil to the root
through the leaves, and foliar absorption of gaseous (mass transfer)
sulfur and nitrogen also occurs (these are geochemical
or biogeochemical pathways). Atmospheric CO}, SO), 3. The rate at which new roots grow into unoccupied
soil that contains unutilized pools of nutrients

This term was suggested to me by Dr. Harold Young, University of The relative importance of diffusion, mass trans-
Maine, Orono, Me. fer, and dispersion (diffusion enhanced by the mixing
88 CHAPTERS Biogeochemistry

that is associated with massflow) varies for different less branched and consequently less efficient on their
nutrients. Different ions vary considerably in the rate own at filtering nutrients out of soil solution than are
at which they diffuse through soil. For nutrients that the root systems of most angiosperm plants. It has
diffuse slowly, such as phosphorus and potassium, the been suggested that this difference in root systems may
supply to the root will be more dependent on the root- relate to differences in the availability of nutrients in
ing pattern of the plant (which in turn controls the soil solution during the evolutionary periods when
volume of soil in close contact with the root) than on gymnosperms and angiosperm plants evolved (Voigt,
the total amount of the nutrient in the entire soil vol- 1968). Henderson et al. (1990) proposed that the pat-
ume. However, the supply of nutrients that diffuse tern of vertical distribution of fine roots in the soil
rapidly, such as nitrate, depends on the total quantity could be used as an index of site productivity.
of nitrate available in the soil (Baldwin, 1975). The
mass flow of nutrients to the root will depend on the Via Mycotrophy. The characteristically low con-
amount of water in the soil and the rate at which it is centrations of dissolved ions in forest soil water, to-
moving toward the root. This in turn will be influ- gether with the form and distribution of their root
enced by the rate at which the plant is transpiring systems, render direct uptake of nutrients from soil so-
water. In a study of Douglas-fir growing on a coarse, lution by roots an inadequate mode of nutrition for
infertile soil, it was estimated that somewhat less than many forest plants. It is generally recognized that
22, 37, and 80% of the nitrogen, potassium, and cal- total, direct nutritional dependence on the soil solu-
cium taken up by the trees was provided by mass flow. tion is abnormal. Most forest plants depend either
Diffusion and dispersion, although limited by the low completely or to a considerable extent on a mutually
soil water content, were apparently very important in advantageous relationship between their roots and soil
the transport of nitrogen to roots in this soil (Ballard microorganisms, the most common of which is a
and Cole, 1974). root—fungus relationship called a mycorrhiza.
Plants vary greatly in the quantity and spatial dis- A mycorrhiza is an intimate association between a
tribution of their small absorbing roots. Plants such as root tip of a higher plant and one of many species of
grasses and many species of deciduous tree (an- fungus. The form of the relationship varies somewhat
giosperms) have finely divided root systems with an (see Chapters 11 and 15 for a more detailed treat-
enormous total surface area and length. These grow ment), but in one important type (ectotrophic mycor-
through a substantial proportion of the soil volume, rhiza), the fungus forms a mantle of strands (hyphae)
and such root systems are efficient at exploiting the around the root tip and penetrates into the outer lay-
soil solution. In contrast, many conifers (gym- ers of the rootlet. Here the hyphae grow between the
nosperms) have root systems in which the root tips and cells to form a network (the Hartig net).
fine absorbing rootlets are restricted almost entirely to The hyphae within the root are in continuous con-
the surface layers of the soil, especially when growing tact with the fungal mantle, which in turn is in contin-
an infertile soil (Figure 5—3). The roots are coarser and uous contact with a dense network of hyphae growing
throughout much of the surrounding soil. The mycor-
rhizal association seems to confer many advantages on
% of root biomass or root tips to measured depth of soil
the plant. It improves nutrition under infertile soil
0 2598 0 iSO 2a Onn ee Js eth IS
> a 4 i. 4 — J i = | conditions and confers greater resistance against dis-
ease organisms, drought, high temperatures, toxic
o20 Forest floor
substances in the soil, and extremes of soil acidity. The
40
mycorrhizal relationship is probably one of the more
60 : ; z q ; : important ecological relationships in forest ecosys-
Number of Tree fine root biomass} Minor vegetation
80 tree root tips (<2 mm) total root biomass tems, and reviews by Hacskaylo (1971), Harley (1969),
in
Depth
profile
soil
fromof
surface
floor,
forest
cm Marks and Kozlowski (1973), and Séderstrém (1991)
100
are recommended to the interested reader.
Figure 5-3 Vertical distribution of tree root tips, tree The nutritional significance of the mycorrhizal re-
fine-root biomass, and minor vegetation root biomass in a lationship lies in the frequent observation that trees
100- to 150-year-old white spruce-subalpine fir stand that have mycorrhizal roots generally grow larger and
growing on deep, infertile, moderately drained sand. (After absorb greater quantities of nutrients than trees with-
Kimmins and Hawkes, 1978.) out mycorrhizal roots. Uptake of nitrogen, phospho-
 The Biogeochemical Cycle: Nutrient Cycling Within an Ecosystem 89

rus, and potassium is particularly enhanced. This is is still high after the removal of the previous stand.
thought to be due to increased uptake of soil water Uptake then declines somewhat as the stand becomes
(Duddridge et al., 1980), to an enlargement in the vol- more dependent on internal retranslocation of nutri-
ume of the soil occupied by the root system, and to its ents from old to younger tissues, thereby reducing the
greatly increased effective absorbing surface. The dependence of primary production on uptake. It may
fungi throughout the soil are essentially an extension also decline because of (1) a decline in soil nutrient
of the roots with which they are associated, and it has availability that results from a reduction in litter de-
been shown that this fungal mycelium (network of hy- composition rates and (2) the immobilization of an in-
phae) is efficient at absorbing nutrients from solution creasing proportion of the site nutrient capital in
and holding them against leaching (Stark, 1972). The heartwood and/or slowly decomposing litter (see Heal
hyphae also increase the solubility of nutrients in the et al., 1982, and references therein).
soil. Organic acids secreted by the hyphae and carbon- An accurate determination of the annual uptake of
ic acid that results from the respiration of the fungi, nutrients by a forest is a difficult task, and although
many estimates have been made, most of these must
CO, + H,O > H,CO; pH dependent H* + HCO, be taken as approximate. Uptake can be measured
only indirectly, by measuring the net change in plant
attack undecomposed soil minerals and organic matter, nutrient content and all losses of nutrients from the
releasing previously unavailable nutrients that are then plants for the period being considered. Uptake of
absorbed by the hyphae. Some of the nutrients that nutrients = net change in plant nutrient content + re-
enter hyphae in the soil subsequently become available placement of nutrient losses from the plant. Alterna-
to roots in mycorrhizal association with the fungus. tively, it can be approximated by measuring the
Mycotrophy, as plant nutrition with the aid of myc- nutrient content of all the new biomass and the inter-
orrhizal associations is called, seems to be the normal nal redistribution of nutrients within the plant (see
mode of nutrition for most forest plants. For some below). Uptake = nutrient content of new biomass-
plants, the relationship is obligatory; they will die if contribution of internal cycling.
grown in soils devoid of the fungal partner. For other Although the work of calculating nutrient uptake
plants, it seems to be obligatory for survival on infer- manually is laborious, it is reasonably easy to do this
tile soils but unnecessary on fertile soil. The mycor- for the aboveground parts of forest plants. The nutri-
rhizal association has been described as the keystone ent exchanges of the root system, which may equal or
of the forest biogeochemical cycle; in many forest en- exceed the aboveground exchanges for some nutrients,
vironments, active biogeochemical cycling and the ex- are much harder to measure. The work is much more
isting vegetation could not occur without it. laborious, many of the measurements cannot be made
The efficiency of nutrient acquisition by fine roots directly, and there are few studies that have quantified
and mycorrhizae is discussed by Bowen (1984) and is the nutrient dynamics of root systems. Consequently,
explored through modeling (see Chapter 21) by Yanai most of the available estimates of uptake must be con-
et al. (1995). The role of mycorrhizae in moving nitro- sidered as just that—estimates.
gen and carbon between spatially separated plants and Figure 5-4 shows the nutrient uptake (ignoring
between overstory and understory trees of the same the root dynamics) by several different forests in com-
and different species is reported by Simard et al. parison with three agricultural crops. Both the magni-
(1997). Simard et al. (2002) give a detailed review of tude of the uptake and the relative importance of the
carbon and nutrient fluxes within and between mycor- different nutrients vary considerably. Hardwoods have
rhizal plants. the greatest overall uptake, with calcium being the
The quantity of nutrients taken up annually by dominant element. Nitrogen dominates the annual
plants varies greatly. It is influenced by the availability uptake of most conifers, although some genera of
of nutrients to the plants (i.e., site fertility) and their conifers (e.g., Chamaecyparis, Thuja) have an uptake
nutritional requirements, which in turn depend on the pattern more like hardwoods. The uptake patterns in
species and their physiological maturity. The uptake Figure 5—4 reflect site as well as species differences.
varies during the development of a stand. It is very The higher uptake by the mixed-oak forest compared
high during the early stages, when there is a rapid ac- with the oak-ash forest reflects higher levels of soil
cumulation of biomass of foliage and live woody tis- calcium on the former site. The differences also reflect
sues, and when the availability of nutrients in the soil variation in age; both the magnitude and the relative
90 CHAPTERS Biogeochemistry

Scots pine Loblolly pine mee pine Wheat Sugar beet _—_Potatoes >A0
240 Mixed oak forest Oak-ash forest Chamaecyparis 70
(MTs 115-160 yr obtusaforest plantation plantation plantation
16 yr 55 yr 200
+ etd 45 yr »
|, 200
= 180
22 160
160
te
Minor vegetation uptake ee
g 140
100
.
2. res (CD Tree uptake
a 100
60

4
60 @ 60
2=

: LAE
6
< 40

g KCaMg N PKCaMg NPKCa NP KCaMg N P KCaMg NP KCa NP KCa N P KCa

(a)

Ca 100 Birch 50 Pine

7 500 Oak
vo
vu

-oh a N 80
70 wri Ca 40
¥ 300 60 R
50 K ee x
3
Ca
: id 30
20 m 10 rs
2 100 Pp
P
Sc
; 0
< 0
50 100 0 50 100 0 50 100
Age of stand, yr Age of stand, yr Age of stand, yr
(b)

Figure 5-4 Nutrient uptake by forest crops. (A) Variation in the annual uptake of the major
macronutrients by several different forest and agricultural plant crops. (B) Variation in uptake with
age for three tree species. Some of these uptake estimates are based on studies of the aboveground
organs alone (uptake = net increment of nutrients in aboveground parts + replacement oflosses from
aboveground parts). Inclusion of belowground biomass production and turnover (cf. Keyes and
Grier, 1981) could result in a considerable increase in the estimates. (Data from Duvigneaud and
Denaeyer-De Smet, 1970; Malkonen, 1974; Remezov and Pogrebnyak, 1969; Tsutsumi, 1971; Wells
and Jorgensen, 1975.)

importance of the different nutrients vary with the age For an example of the importance of the error, see
of the stand, making simple comparisons among the Vogt et al. (1983).
various estimates problematic. The magnitude of the Nutrient uptake estimates are not reliable as esti-
annual uptake by agricultural crops is generally higher mates of the minimum nutrient requirements for pri-
than that of most conifer crops. Bred for rapid growth mary production. Plants may absorb abundantly
and lacking significant internal cycling, annual agri- available nutrients in excess of requirements (luxury
cultural crops exert a strong demand on soil nutrient consumption) and may also absorb nonessential ele-
resources and generally require the addition of fertil- ments (e.g., sodium), and, conversely, much of the re-
izers if soil fertility and productivity are to be sus- quirement for net production of long-lived perennial
tained. Conifer and agricultural crops differ in the plants may be met by internal redistribution from
generally lower requirement for calcium and the older tissues (see below). Requirements also cannot be
higher requirement for potassium in the latter. gauged accurately from the levels of nutrients in the
Most nutrient uptake estimates were made before plant, just as energy dynamics cannot be gauged from
the magnitude of the annual production and turnover the biomass pyramid. For example, a plant may have a
of tree fine roots was appreciated (c.f. Keyes and low total content of an element that is readily lost
Grier, 1981). The nutrient uptake requirements of (e.g., potassium) but a high uptake requirement to re-
fine-root production have not yet been quantified for place the losses. An accurate assessment of the nutri-
many forests, and the reader is advised to interpret ent requirements of a forest must ultimately come
most nutrient uptake data with appropriate caution. from a comparison of uptake and productivity. Suffi-
 The Biogeochemical Cycle: Nutrient Cycling Within an Ecosystem 91

cient quantitative information is now available to per- very different rates. For example, dogwood trees have
mit comparisons of uptake and requirement for net long been recognized as selective absorbers of calcium
production in different forest types (e.g., Cole and (Thomas, 1969), which accounts for 2 to 4% of the
Rapp, 1981). Increasingly, this is done using ecosys- leaf weight. The manganese content of plants in a
tem management models that explicitly represent the Wisconsin forest was found to vary from as low as 20
multiple pathways of nutrient movement in ecosys- ppm for a species of Pyrus to as high as 2,999 ppm for
tems and the feedback between nutrient cycling and a species of Gaultheria (Gerloff et al., 1966).
tree growth (e.g., FORECAST; see Chapter 21). The relative distribution of nutrients varies ac-
Comparisons are sometimes made between nutri- cording to the age of the plants. For example, the dis-
ent uptake and the total capital of nutrients in the soil. tribution of nitrogen among foliage, branches, and
Such comparisons may be useful, especially for mobile stems in a loblolly pine stand was found to vary as the
nutrient ions, but for many elements, the comparison stand aged. Most of the nitrogen in the very young
should be made with a knowledge of the distribution stand was accumulated in the foliage, but by the time
of the roots. This will strongly influence how much of the plantation reached 60 years, the stems contained
the soil nutrient capital is readily available to the trees. more than four times as much nitrogen as the foliage.
In some soils, roots of trees can penetrate deeply, but This reflected the continued accumulation of stem
most of the deeper roots are coarser, anchoring roots. and branch biomass throughout the 60 years, whereas
As a general rule, the majority of the fine feeding roots foliage biomass peaked at approximately 25 years and
of trees are restricted to the forest floor and the sur- then declined slowly over the rest of the period
face layers of the mineral soil. Results from one study (Figure 5-6).
showed that 82% of the fine roots of a 140-year-old Within the crown, the chemical content of foliage
Scots pine forest in the Soviet Union were in the top varies according to the age of the foliage and its posi-
20 cm of soil and 76% were in the top 10 cm. Ninety- tion in the crown. Nitrogen, phosphorus, and potas-
three percent of the roots of minor forest vegetation sium concentrations of a given annual cohort of
in a forest glade were in the top 20 cm of soil; 76% needles often decline with age,’ whereas calcium and
were in the top 10 cm (Vasil’eva, 1968). In a study in magnesium concentrations tend to increase. The de-
western Canada, it was found that 71% of the conifer clines in concentration may be either because the nee-
root tips in a white spruce/subalpine fir forest were in dles continue to get heavier or because of an absolute
the forest floor, whereas 91% were in the forest floor loss of the chemicals. Because the chemistry of foliage
and upper 10 cm of mineral soil (Figure 5—3). All these is used as a diagnostic tool in assessing the nutritional
stands were on infertile sandy soils. On richer soils, a status and fertilizer needs of forest trees, a knowledge
greater proportion of the fine-root biomass and root of how foliage chemistry varies within tree crowns is
tips are found at greater depths. However, many of the of considerable practical concern. When foliage is
world’s forests are growing on infertile soils, and for sampled carefully, it can be a useful index of the fertil-
these forests, the forest floor and superficial layers of ity of the site. This is discussed further in Chapter 11.
mineral soil provide the majority of the nongaseous
nutrients. The question as to the volume of soil that is
Nutrient Losses from Plants
being tapped by root systems is considered again in
Chapter 11. All plants and animals are faced with continual losses
of nutrients, and the majority of the annual uptake by
Nutrient Distribution in Plants plants simply replaces losses. A much smaller percent-
age is retained as the nutrient content of permanent
Once nutrients have been absorbed, they are trans- new plant biomass. A certain amount of loss is un-
ported to various parts of the plant for use in metabol- doubtedly necessary for both plants and animals to
ic processes or storage. Just as uptake varies between prevent toxic accumulations of chemicals, and either
species and sites, the relative distribution of nutrients
between the various parts of forest plants varies con-
siderably (Figure 5-5). This variation is a result of dif- ’This pattern is not necessarily found in the various age classes ofneedles on
ferences in both the distribution of biomass and the a single branch. Because ofyear-to-year variation in nutrient availability
concentration of chemicals in the various tissues. Dif- and growth conditions, the chemistry of new foliage may vary considerably
ferent plant species absorb and accumulate elements at from year to year, and this variation may persist until the needles are shed.
92 CHAPTER 5 = Biogeochemustry

Mixed oak forest, Belgium White spruce-subalpine Douglas-fir plantation, U.S. Lodgepole pine forest, Canada
70—75 yr (1) fir forest, Canada 36 yr (3) 125 yr (2)
110-350 yr (2)
1029

663

482 ap
360 375 305
291 247
198
= 148
96 75 32 56
38 67

Ni bee Ga N P K Ca Mg

Foliage

Branches

Stemwood

Stembark

Roots

Scale: t pee
0 100%

Figure 5-5 ‘The tree nutrient content and its distribution in four different forest types. (A)
The total content of five macronutrients in tree biomass (kg ha’'). (B) Percentage distribution of this
content between the five major biomass components. (Data from Cole et al., 1967; Duvigneaud and
Denaeyer-De Smet, 1970; Kimmins, unpublished.)

active secretion or passive loss enables the organism to Losses of nutrients from plants can occur in many
regulate its internal chemistry within acceptable lim- ways: (1) leaching of above- or belowground organs by
its. House or greenhouse plants protected from rain rainfall or soil water; (2) defoliation by herbivores, in-
may experience leaf damage because of toxic accumu- cluding the physical removal of foliar biomass, the ac-
lations of chemicals that would have been lost by celeration of leaching losses from damaged foliage,
leaching in a natural environment. Similarly, agricul- and the death of fine roots and mycorrhizae as a result
tural crops in areas deficient in precipitation may ex- of defoliation; (3) losses associated with reproduction;
perience disorders because of these accumulations. and (4) litterfall of leaves, branches, stem bark, and
Plant species that are adapted to dry areas frequently roots, including the fall of dead tissues, the sloughing
have specialized morphological adaptations for the of living tissues (seasonal leaf and fine-root drop), and
disposal of excess chemicals. the physical removal of living tissues by mechanical
Under certain circumstances, losses of nutrients force (e.g., wind and snow break).
may be so great relative to uptake that the metabolism Each of these is discussed in turn.
of the organisms may be adversely affected. This can
occur in both plants (e.g., in areas of heavy rainfall or Leaching by Rainwater. All plants experience loss-
at times of heavy defoliation by herbivores) and ani- es of a wide variety of chemicals from leaves, bark, and
mals (e.g., excessive sodium loss in humans through roots by the leaching action of water. In some climates
sweating). Thus, average rates of nutrient loss are the and for some nutrients, this is the major pathway of
normal condition, whereas extremely high or very low loss (Parker, 1983); it may be so great as to interfere
rates of loss may interfere with normal metabolism. with the growth and physiology of the plant (Stenlid,
 The Biogeochemical Cycle: Nutrient Cycling Within an Ecosystem 93

growth-regulating substances (hormones), phenols,

a =)
and many other plant chemicals are regularly washed
off and out of plants when it rains. Of the inorganic
nutrients, sulfur, potassium, calcium, magnesium,
= i)j=) phosphorus, and manganese are usually leached in the
greatest quantities, the amount depending on the
species of plant and the nature of the rainfall event (for
\oo a discussion of different types of rainfall events, see
Chapter 10). Sodium, a nonessential chemical, is also
oe
Rea
See
readily leached. Table 5—5 gives some examples of the
60 7 Branches variation in the inorganic chemistry of the water that
has leached the aboveground biomass of forests. The
nutrient quantities in this table include both materials
30 leached out of vegetation and materials, such as dust,
washed off the surface of the vegetation.
Although considerable quantities of inorganic nu-
ha!
Nitrogen
aboveground
in
kg
biomass,
content
tree
trients are leached, organic substances account for
most of the chemicals in water reaching the forest
floor (Tukey, 1970). Losses of up to 30% have oc-
Stand age, yr
curred in agricultural crops as the result of heavy
Figure 3-6 Variation with stand age in the relative contri- leaching of sugars frem flowers and developing seeds.
bution of stems, foliage, and branches to the aboveground Palatability of soft fruits is reduced by heavy rains just
nitrogen content of a loblolly pine stand. (After Switzer et before maturity and harvest because of the reduction
al., 1968. Redrawn by permission of TVA and G.L. Switzer.) in sugar content. The vitamin and nutritive value of
forage crops is reduced by rain just before harvest, and
1958). Inorganic micro- and macronutrients, other in- the flowering of plants is affected by the leaching of
organic chemicals, amino acids, sugars, vitamins, plant growth hormones. Up to 800 kg ha"! yr of car-

Table 5-5 Comparison of Some Macronutrients in Stemflow and Throughfall in Various Forests,” kg ha™ yr“!
N P K Ca Mg
Forest Type Thfl Stf Thfl Stf Thifl Stf Thfl Stf Thfl Stf References

Douglas-fir, U.S., ues Os} CS OR 10.7 1.6 SS) 1100- — Cole et al., 1967
36 yr old
Oakwood, Great Britain Crp Ol OM meRON 8.6 1.6 6.4 2.0 Oro) 0.7 +Carlisle et al., 1967

Oakwood, Belgium Of O08 Ga OW 16.0 0.8 G2 OY 5.6 0.6 Duvigneaud and Denaeyer-De
Smet, 1970

Hardwoods, U.S. = = = = 6.7 3.3 6.2 ONG 4.8 Torrenneva, 1975

Loblolly pine, U.S. — — — = 4.1 2.0 Onl 0.4 1.7.‘ Torrenneva, 1975

Scots pine, Finland Us ee Oa 0.4 0.0 5.8 7.4 04 — _— Malkonen, 1974

Douglas-fir, U.S., Sy) = 2/ — PXI 44 — 21 — Abee and Lavender, 1972

450 yr old
Tropical forest, Ghana 11.0 — 3.3 — 196 26 — 16 _ Nye, 1961

Hardwoods, U.S.° 18 de) Gi Of 26.9 3.5 7.0 0.6 2.0 0.2 Eaton etal., 1973

aMuch of the throughfall and stemflow nutrient content originates as “aerosols” intercepted by vegetation rather than materials leached out of
vegetation. Thfl, throughfall: water falling through and dripping from tree crowns and minor vegetation foliage. Stf, stemflow: water reaching the
soil by flowing down plant stems.
>Summer period only.
94 CHAPTERS — Biogeochemistry

bohydrate can be leached from apple trees compared throughfall* and stemflow, 53, 70, and 41% originated
with 20 to 30 kg of potassium, 10 kg of calcium, and 9 from intercepted aerosols rather than true foliar leach-
kg of sodium. Leaching of inorganic nutrients is nev- ing (Miller et al., 1976). Estimates of nitrogen leach-
ertheless considerable; up to 50% of the calcium and ing, however, may be too low. Most studies measured
more than 80% of the potassium content of apple only increased inorganic nitrogen in leachate, whereas
leaves can be leached in 24 hours (Tukey, 1970). the majority of the nitrogen may be in organic form.
Deciduous hardwood trees generally lose more The relative contributions of throughfall and
nutrients by leaching than conifers during the summer stemflow also vary. They are largely determined by
leafy period, but conifers may lose a greater total over the morphology of the crown, which in turn is deter-
the year because of leaching during a wet winter mined by the species, the age, and the structure of the
period. The greatest leaching occurs in the tropics, stand. For example, the contribution of stemflow to
where hardwoods hold their foliage all year round in a the leaching of cations over a 6-month period in a 17-
climate with abundant rainfall throughout the year. year-old Douglas-fir plantation varied according to
Leaching losses are influenced by the age of the fo- the spacing of the trees, which in turn varied the mor-
liage, and less leaching may occur from younger than phology of the crown. At 3 X 3-ft spacing, the stem-
from older foliage. However, leaching of nutrients flow contributions were 44, 41, and 48% _ for
such as nitrogen and phosphorus is greatest when potassium, calcium, and magnesium, respectively,
these nutrients are in their most mobile state, which whereas at 12 X 12-ft spacing, they were 9, 9, and 3%
occurs in very young foliage during rapid early growth (Kimmins, unpublished data; cf. Figure 10-2).
and in senescent foliage before leaf fall.Consequently, The leaching of foliage and other tissues is an im-
there is no simple linear relationship between foliage portant ecological process. By constantly returning a
age and leaching. Leaching is greater from damaged supply of nutrients to the rooting zone in a highly
than from intact foliage, the undamaged cuticle acting available condition, foliar leaching contributes to the
as an efficient barrier to leaching. The extent of leach- nutrition of plants that grow on infertile sites. For ex-
ing during any particular rainstorm is greatly influ- ample, plants that grow on calcium-deficient soils may
enced by the duration and intensity of the rainfall and experience calcium deficiency at the growing tips be-
the time since the last rainfall (Attiwill, 1966; Miller, cause, unlike nitrogen, phosphorus, and potassium,
1966). Annual leaching losses are determined by the calcium cannot be easily retranslocated from old tis-
quantity, timing, and character of precipitation. These sues to growing tissues. However, this lack of mobility
vary from year to year and therefore so does the total can be partially compensated for by calcium leached
leaching loss. out of the older foliage, absorbed by roots in the forest
Published values for the removal of nutrients from floor, and translocated directly to the growing tissues.
tree crowns by rain vary enormously: 0 to 12 kg ha"! The leaching of plant hormones, phenols, and other
yr! for nitrogen, a trace to 8 kg ha? yer for phospho- organic molecules affects rates of litter decomposition,
rus, 1 to 320 kg ha yr! for potassium, 0.2 to 194 kg the germination and survival of the seeds of other
ha! yr‘ for calcium, and 0.9 to 16 kg ha?! yr! for species, and the chemistry of the soil beneath a plant
magnesium (Attiwill, 1966; Madgwick and Ovington, (Gersper and Holowaychuck, 1970; Gosz, 1984; Ma-
1959; Ovington, 1962). The accuracy of these data as hendrappa 1974). The supply of soluble carbohydrates
estimates of annual leaching totals is suspect. In many to the forest floor in throughfall provides a readily
cases, they are probably overestimates of vegetation available energy source for free-living microbes, and
leaching. They are calculated as the difference be- this may be important for the activity of free-living
tween the nutrient content of precipitation collected nitrogen-fixing microflora. We consider the ecologi-
in the open and under the canopy, and the latter in- cal significance of some of these effects in Chapter 15.
cludes chemicals washed off the surface as well as A useful example of the effect of throughfall and
leached out of vegetation. Also, vegetation (especially stemflow is the alteration of soil chemistry and under-
trees) is efficient at intercepting fine rain, mist, and story vegetation near the stems of large, old, deep-
dust (“aerosols”), which are not sampled efficiently by
traditional precipitation collectors. In a study of nutri-
ent cycling in Corsican pine in Scotland, it was calcu-
lated that of the potassium, calcium, and magnesium in ‘Defined in Table 5-5.
 The Biogeochemical Cycle: Nutrient Cycling Within an Ecosystem = 95

rooted oaks in southwestern Sweden. Growing on nu- of the total stand capital of these nutrients, respec-
trient-poor surface soils, these deep-rooted trees take tively. By comparison, the combined litterfall and fo-
up calcium from underlying calcium-rich layers, much liar leaching in this stand was 15 kg ha‘ of potassium,
of which is added to the surface soil by stemflow and 16 kg ha! of calcium, 15 kg ha"! of nitrogen, and 0.6
throughfall. This promotes the growth of more nutri- kg ha"! of phosphorus (Cole et al., 1967). Frass from
ent-demanding herbs near the stems of oaks large Tortrix viridiana \arvae defoliating an oak stand in
enough to produce appreciable stemflow (Andersson, England made up only 15% of the carbon and energy
1991). For a detailed examination of the role of (i.e., the biomass) of the annual litterfall but con-
throughfall and stemflow in forest nutrient cycling, tained 24% of the nitrogen and 42% of the phospho-
see Parker (1983). rus in the litterfall (Carlisle et al., 1966). This
reflected the much higher concentrations of nitrogen
Defoliation by Herbivores. The loss of chemicals and phosphorus in young oak leaves than in autumnal
from forest plants as the result of feeding by herbi- oak leaf litterfall.
vores has received relatively little attention. This The effects of defoliation on nutrient cycling are
probably reflects the normally low levels of utilization not limited to the increase in the litterfall pathway. If
of forest plants by herbivores (Table 4-3), but defolia- there is a lot of rainfall during the period of defolia-
tion can be an important pathway of loss at times of tion, leaching losses will be increased because the her-
herbivore epidemics (Gosz, 1984; Kimmins, 1972; bivore feeding impairs the ability of the foliage to
Mattson and Addy, 1975; Parker, 1983; Schowalter et resist leaching (Kimmins, 1972; Parker, 1985; Tukey
al51 991). and Morgan, 1963). Defoliation-induced increases in
Table 5-6 shows the amount of nutrients removed leaching will be greater in the spring, when nutrients
from an oak stand in litterfall during total defoliation are being actively transported to the foliage, than in
by the gypsy moth. Substantial quantities of nutrients the summer, when this transport is greatly reduced.
were removed in the insect feces (frass) and pieces of In addition to its aboveground effects, defoliation is
uneaten leaf, and the defoliation resulted in consider- known to induce substantial root mortality. As much as
able increases in the loss of phosphorus, potassium, 75% of the fine roots of balsam fir trees were found to
calcium, and magnesium. Heavy defoliation of a be killed after the removal of 100% of the new foliage
conifer stand will result in even larger losses because by spruce budworm (Redmond, 1959). This represents
the accumulation of several years of foliage growth both a substantial loss of nutrients from the root bio-
may be removed. A single complete defoliation of the mass and a substantial temporary reduction in uptake
36-year-old Douglas-fir stand in Figure 5-5 would to replace the losses. Similarly, severe or frequent graz-
remove approximately 70 kg ha"! of potassium, 82 kg ing has been found to reduce the size and growth of
ha! of calcium, 115 kg ha! of nitrogen, and 32 kg roots in agricultural pasture plants. The loss of fine
ha! of phosphorus, representing 28, 22, 32, and 43% roots, together with the accompanying reduction in

Table 5-6 Contribution of Defoliation by the Gypsy Moth to the Loss of Nutrients from 59-Year-Old Oak Trees,
kg hat yr!
Litter Type Biomass N P K Ca Mg

Frass (insect droppings) 750 24.4 15 24.1 14.0 5.7

Dead insects 35 3.4 0.5 ‘fell 0.1 0.1

Remains of eaten leaves 2176 37.4 3.6 32.1 44.0 eS

Second flush of leaves 490 7.8 0.6 6.9 8.8 ler

Total litter from defoliated area 3451 73.0 6.2 64.2 66.9 14.8

Total litter from undefoliated area 3480 68.7 3.6 47.4 55.6 9.4

Percentage of increase in litterfall losses - 6 72 35 20 57

attributable to defoliation

Data from Rafes, 1971.

96 CHAPTERS _ Biogeochemistry

soluble carbohydrates, reduces the plant’s ability to ab- and 6% of the magnesium in the total litterfall
sorb nutrients (Alcock, 1964) either directly or with the (Carlisle et al., 1966). The nutrient content of heavy
assistance of mycorrhizae, an association that is very crops of pollen and seeds has not been adequately
sensitive to the supply of carbohydrates to the roots. quantified, but it is probably considerable. Singh et al.
Reduced supplies of carbohydrates to the roots also re- (1990) found that an exceptionally heavy seed crop ina
sult in the loss of nutrients from living roots back to Himalayan oak forest significantly altered litterfall
the soil. Furthermore, the reduction in water uptake patterns and doubled the retranslocating of nitrogen
that accompanies heavy defoliation will reduce the rate from senescing leaves the following year.
of mass transfer of nutrients from the soil to the roots.
The effect of defoliation on the plant obviously Losses in Litterfall. Losses of nutrients by defolia-
goes beyond the mere removal of leaf material, and tion and reproduction can be substantial, but they are
defoliation has biogeochemical impacts other than the characterized by great year-to-year variability. In a
removal of nutrients from plants. It can result in large year with low herbivore populations or in a year with-
transfers of frass and nutrients to streams and lakes, out a heavy flower and seed crop, these losses will be
leading to an increase in aquatic productivity (Turner, very small. Conversely, in a year with an outbreak of
1963). Defoliation of overstory species is frequently insect defoliators or with a bumper seed crop, these
accompanied by rapid growth of understory plants in losses will be high. Nutrient losses in leaf litterfall,
response to the increased light, soil temperature, and conversely, tend to be somewhat more regular, be-
nutrient availability. Because frass contains higher cause they are largely tied to the more predictable
concentrations of nutrients and lower concentrations event of leaf shedding. Of all the pathways of nutrient
of various decomposition-inhibiting organic chemi- loss from plants, aboveground litterfall has received
cals (e.g., tannins) than does leaf litter (e.g., Feeny, the greatest attention. Conversely, belowground lit-
1968, 1970; Feeny and Bostock, 1968), heavy inputs of terfall (the annual death of large quantities of fine
frass to the forest floor may stimulate decomposition roots) is one of the least studied pathways of nutrient
and mineralization of the litter. removal from plants, despite that in some forests it ex-
ceeds aboveground litterfall by several times (Vogt et
Losses Associated with Reproduction. Horticul- al., 1986).
turalists and foresters have recognized for many years ‘Table 5—7 presents a summary of the quantities of
that the production of abundant crops of fruit and certain macronutrients transferred from trees to soil
seeds tends to deplete the carbohydrate and nutrient by aboveground litterfall. The quantity is a function of
reserves of plants (Gessell, 1962; Kramer and Koz- the biomass, the type (leaves, branches, bark, etc.), and
lowski, 1979; Stenlid, 1958). This temporarily reduces the nutrient concentrations in the litterfall, all of
their growth and the uptake of nutrients (Kozlowski, which vary from site to site. Litterfall losses are gener-
1962). The initiation and development of flowers and ally greatest on moist, warm, fertile, and other high-
seeds requires higher levels of nutrients than vegeta- productivity sites and least on dry, cold, infertile, and
tive growth (Matthews, 1963; Mustanoja and Leaf, other low-productivity sites. Table 5-8 shows how the
1965), and plants in many areas are unable to produce biomass and litter chemical content varies in a Japan-
a subsequent seed crop until they have rebuilt their re- ese hardwood forest along a topographic gradient
serves. It has been suggested that this is one of the de- from a moist valley bottom to a dry ridgetop.
terminants of the infrequent production of good seed As was noted in Chapter 4, most of the existing lit-
years in northern forests and tundra regions (Kalela, terfall data refer to aboveground litterfall only, which
1962). The stimulation of flower and seed crops in in many kinds of ecosystems may involve a much
trees by fertilization reflects the heavy nutrient de- smaller turnover of biomass than belowground “litter-
mands of reproduction (Ebell, 1972). fall” (it does not actually fall anywhere), which results
As was the case with defoliation, there have been from the annual production and mortality of fine
relatively few studies of the nutrient losses associated roots. Although this aspect of litterfall has received lit-
with reproduction. In one study, the male flowers of tle study, recent investigations have suggested that it is
sessile oak were found to make up 4% ofthe litter bio- a major pathway of nutrient loss from plants. Mea-
mass but to contain approximately 11% of the nitro- surements in an oak—hickory hardwood forest in Ten-
gen, 14% of the phosphorus, 12% of the potassium, nessee revealed that the death of fine roots accounted
 The Biogeochemical Cycle: Nutrient Cycling Within an Ecosystem 97

Table 5—7 Quantities of Certain Macronutrients in Aboveground Litterfall in Various Forests, Arranged in Order of
Increasing Nitrogen Content, kg ha™! yr“!
Location N Pp K Ca Mg Reference

Scots pine, Finland 1 1.0 25 7.8 = Malkonen, 1974

Douglas-fir, Washington 13.6 0.2 Axi lala = Cole et al., 1967
Jack pine, Ontario, Canada 16.6 - 4.8 10.4 — Foster and Gessel, 1972
Douglas-fir, Oregon Seah 5.6 9.8 63.1 1.4 Abee and Lavender, 1972
Nothofagus forest, New Zealand 37 2.6 30 74 11 Ovington, 1962
Oak forest, England 41.0 2.2 10.5 23.8 3.4 Carlisle et al., 1966
Oak forest, Belgium 50 2.4 21.0 110 5.6 Duvigneaud and Denaeyer-De Smet, 1970
Spruce, U.S.S.R. 52 2.6 12 48 i Ovington, 1962
Hardwood forest, N.H. 54.2 4.0 18.3 40.7 5.9 Gosz et al., 1972
Loblolly pine, N.C. 58.2 7.8 16.0 29.2 6.9 Ovington, 1962
Oak forest, U.S.S.R. 59 3.0 62 86 13 Ovington, 1962
Birch forest, U.S.S.R. 66 5.0 13 54 19 Ovington, 1962
Beech, Sweden 69 5.0 14.4 Cie 4.3 Nihlgard, 1972
Red alder, B.C., Canada 137 5.4 16 51 10 Binkley, 1982
Tropical forest, Ghana 199.5 U8 68.4 206 44.8 Nye, 1961

for a loss of 67.5 kg ha! yr! of nitrogen from the trees The importance of fine root litterfall varies ac-
in comparison with 34 kg ha"! yr of nitrogen in leaf cording to stand age and site. Vogt et al. (1983) found
litterfall and 4.4 kg ha’ yr! of nitrogen in above- that belowground litterfall was approximately double
ground leachates (Henderson and Harris, 1975). aboveground litterfall in a 23-year-old Pacific silver fir
Death of fine roots in a 16-year-old loblolly pine stand stand but more than four times as great in a 180-year-
in North Carolina accounted for a loss of 48.7 kg ha” old stand. The difference in turnover of macronutri-
yr! of nitrogen in comparison with 58.2 kg ha! yr! of ents between above- and belowground litter varied
nitrogen in aboveground litterfall and 9.6 kg ha! yr“ from zero to approximately four times according to
in leachates (Wells and Jorgensen, 1975). nutrient and stand age. Fine roots cycled 60, 10, 20,
30, and 10 kg ha™ of nitrogen, phosphorus, potassium,
calcium, and magnesium in the 23-year-old stand and
Table 5-8 Variation in the Macronutrient Content of 110, 20, 20, 30, and 10 kg ha" in the 180-year-old
Aboveground Litterfall Along a Topographic Gradient stand (Vogt et al., 1982). Keyes and Grier (1981)
from a Valley Bottom to a Ridge Top in a Japanese reported that fine-root litterfall in a dry ridgetop
Hardwood Forest, kg ha™ yr™ Douglas-fir stand was four times that of a stand on a
Bottom Ridge moist lower slope site (5.6 versus 1.4 t ha? yr“).
of Slope Top It is probable that the annual death of fine roots
Nutrient (Moist Site) Midslope (Dry Site) constitutes a major proportion of the total litterfall of
all forests. If this proves to be the case, then total nu-
N 56.8 64.5 S257, Saul 35.4
trient uptake by most forests has been significantly un-
P 4.2 4.3 23 PLS 2.6 derestimated, and many of the values in Figure 5—4
K 14.6 14.7 Tol 7.8 9.9 may be much too small.
Ca 41.0 43.8 19.9 25.9 CIS)7/ Litterfall generally accounts for the majority of the
Mg 10.0 9.4 4.4 Bf vata
nitrogen, calcium, and magnesium loss from vegeta-
tion, and leaching generally accounts for the majority
Biomass 4300 4728 2293 2640 3114
of the potassium loss. The major pathway of phospho-
Data from Katagiri and Tsutsumi, 1973. rus loss is sometimes litterfall and sometimes leaching.
98 CHAPTERS — Biogeochemistry

Litter Decomposition position requires 40 years or more. The rate of litter

decomposition generally increases as the quantity of
Decomposition of litter and the release of nutrients are litterfall increases, with the result that there is an in-
often the critical link in the forest biogeochemical verse relationship between litterfall biomass and nutri-
cycle. If decomposition is too slow, then most of the ent content, and the biomass and nutrient content of
nutrients returned to the forest floor are removed the forest floor (Figure 5—7). The very high above-
from active circulation for long periods of time, and ground litterfall of moist tropical forests on fertile
both nutrient cycling and forest productivity are re- soils is generally associated with little or no forest
duced. Excessive accumulation of undecomposed lit- floor, whereas the small aboveground litterfall of cold
terfall can also lead to the development of forest floors forests is generally associated with deep forest floors,
that have undesirable physical effects on the soil. Deep unless fire is frequent.
forest floors can be excessively wet, be excessively acid, Most estimates of litter decomposition are based
and remain cold throughout most of the growing sea- on short-term studies (1 or 2 years) of litter confined
son. This results in poor root development, poor tree in litterbags or on the mean residence time of litter
nutrition, and slow tree growth. Excessively rapid lit- based on a comparison of the forest floor biomass and
ter decomposition, conversely, can release nutrients the annual litterfall.° The latter may underestimate de-
faster than the plants and soil are able to retain them, composition rates because of the general failure to in-
and they may be leached out of the rooting zone. Ni- clude root death in the litter input rate. Inclusion of
trogen may be lost by denitrification (see For. Ecol. root litter input data reduced litter residence time esti-
Manage. 1991, 44:1, 92). The loss of soil organic mat-
mates from 32 to 8 years in a 23-year-old stand and
ter (SOM) that will accompany excessively rapid rates from 69 to 16 in a 180-year-old Pacific silver fir stand
of decomposition can also lead to the development of
in Washington. Mean residence times for some broad-
undesirable physical and chemical soil conditions, with
leaved species were reduced from 7.6 to 9.3 years to
serious consequences for site fertility, soil moisture
4.7 to 6.6 years by the inclusion of fine-root turnover
status, and resistance to erosion and other forms of soil
(references in Vogt et al., 1986).
damage. Many ofthe major problems offorest tree nu-
Litter decomposition is accomplished by the com-
trition and soil fertility are related to the amount and
bined action of soil animals of various sizes (soil meso-
nature of the organic debris reaching the forest floor
fauna and soil microfauna) and soil microorganisms
and its rate of decomposition (Waksman, 1932).
(bacteria and fungi). The soil animals (worms, mites,
The rate of litter decomposition’ varies enor-
springtails, beetles, insect larvae, etc.) break down the
mously. It is often very rapid in humid tropical forests
litter into small pieces. This increases the surface area
(between 6 and 10 times faster than in temperate
and ruptures the outer surface of the litter (the waxy
forests [Madge, 1965]), where a leaf may decompose
cuticle), which otherwise restricts the access of the mi-
completely within weeks of falling to the forest floor.
croflora to the more decomposable tissues inside the
However, tropical decomposition rates can also be
leaf. The soil animals bury the fresh litter, which places
very slow (e.g., on highly weathered tropical white
it in a moisture—temperature environment that is gen-
sands in Amazonia), variable (where there are seasonal
erally more conducive to decomposition, and by
dry periods), or moderate (in drier tropical forest)
churning the soil, they are constantly bringing nonmo-
(e.g., Mailly and Margolis, 1992). Litter decomposi-
bile microflora (bacteria) into contact with decompos-
tion is generally rapid (leaves decomposing in | to 3
able materials. Some changes in the chemical
years) in temperate hardwood forests and slow to very
composition of the litter material occur after it has
slow in temperate and boreal conifer forests (4 to 30
passed through the gut of a soil animal, and this may
years for needle decomposition). Arctic, alpine, and
result in faster microbial decomposition. However, it is
dry land forests have the slowest rates; needle decom-

‘The rate of litter decomposition can be calculated using the equation x, = *The mean residence time for litterfall organic matter in the forest floor is
xe, where x, = amount ofsubstrate at time t; x = initial amount ofsub- given by T = H/L where T = organic matter residence time; H = forest
strate; k = fractional loss rate per unit time; k is called decay or decomposi- floor weight minus the biomass of live roots; L = annual litterfal (Gosz et
tion constant (Olson, 1963). al., 1976).
 The Biogeochemical Cycle: Nutrient Cycling Within an Ecosystem 99

: 40 (0.1) (1) ij
ee| &10,0) | a eS
E |
=
Very rapid
: /
U/
a E x Mineral content f
iS |decomposition /
is 7 Tropical ° Biomass o (2)
2 1 +30 a 30

:
=
=
1000
S
ae
=|
a | Ze
a
(3)

}20
= Temperate
apa
a
t| Subarctic,subalpine,
r
z 500 i hardwood r g | and semidesert forests
S a forests Boreal Subarctic, subalpine, A ‘gan /
° ie wy, i
3
Fr)
Zutelwil
J forests and semidesert
forests
10
|tha!
of
Biomass
page
litterfall,
yr
2
| Le
ay) fe
0 0 Sass laa
0 1000 2000 3000 4000 40 \ 80
Mineral content of forest floor, kg ha ! Very slow
[S59
eee 81s © T= eee Biomass of forest floor, tha! ais
decomposition
20 40 60 80
Biomass of forest floor, t ha’!
(a) (b)

Figure 5-7 (A) Generalized relationship between litterfall biomass and mineral content and
between forest floor biomass and mineral content in different forest regions of the world. (B)
Litter decomposition rates (numbers in brackets) expressed as the ratio of litterfall biomass
to forest floor biomass. A ratio of 2 indicates that on average it take 2 years for the complete de-
composition of | year’s litterfall. (Data from Rodin and Bazilevich, 1967.) The litterfall data are for
aboveground litter only, whereas forest floor biomass is derived from both above- and belowground
litterfall. Consequently, decomposition rates shown in the diagram are underestimations. See also
Vogt et al. (1986). There are numerous exceptions to this generalized pattern with high forest floor
mass values and low decomposition rates in some tropical forests (e.g., Mailly and Margolis, 1992),
low forest floor mass in boreal forests subject to frequent fire, and low mass with moderate to rapid
decomposition in some boreal broadleaf forests.

generally agreed that the major contribution of soil an- when the soil fauna were present. The study showed
imals to litter decomposition is the physical comminu- that by enhancing litter decomposition and nutrient
tion of the material. These factors all combine to result mobilization and by improving soil structure, the
in a much more rapid rate of mineralization of fresh lit- fauna play an important role in nutrient cycling and
ter when there is an active community of soil animals. plant growth (Satalaé and Huhta, 1991).
The importance of these animals in decomposition After it has passed through a soil animal, litter is
has been demonstrated by comparing the rate of litter eliminated as fecal pellets, which are rapidly invaded
decomposition in the presence and absence of soil by fungi and bacteria. Intense activity by soil animals
fauna. In one study, it was found that the elimination is normally a prerequisite for the high populations of
of soil fauna by applying naphthalene to the forest soil microflora that are responsible for most of the ac-
floor reduced the normal loss of oak leaf material after tual decomposition. The relationship between rate of
145 days by 50%, and in another it was reduced by decomposition and abundance of soil microflora in
25% after 1 year (cf. Figure 4-13). various types of litterfall in Tennessee is shown in
In another study, birch seedlings were grown with Table 4-5. Obviously, any explanation of the variation
and without soil animals. Leaf, stem, and root growth in rate of litter decomposition must be based on an
were 70, 53, and 38% greater, respectively, and the ni- understanding of the factors that control the abun-
trogen and phosphorus content of the leaves was ap- dance, species composition, and activity of the soil mi-
proximately3 and 1.5 times greater, respectively, when croflora. These include the activity of the soil fauna,
the fauna were present. Despite higher transpiration the species of tree producing the litter, the chemical
from the larger seedlings, soil moisture was higher composition of the litter, the acidity (pH) of the litter
100 CHAPTERS — Biogeochemistry

and of the forest floor that it produces, the soil micro- Romell, 1935; Stone, 1975), and a recent review (Bink-
climate, and the fertility status of the soil. ley, 1995) concluded that the issue is much too com-
plex to be described by simple generalizations. The
Activity of the Soil Fauna. Soil fauna include evidence in support of the earlier conclusions is weak,
earthworms, enchytraeid worms, diplopods, isopods, and carefully designed experiments have generally
dipteran larvae, collembolans, and orobatid mites (Ed- failed to support the generalizations that have charac-
wards et al., 1970). Soil fauna are discussed in more terized the debate. Prescott et al. (2000) have also chal-
detail in Chapter 11. lenged the traditional idea that the accumulation of
humus in forests is always undesirable. They note that
Species of Tree Producing the Litter. Species while excessive accumulations lower nitrogen avail-
vary in the physical and chemical properties of their ability and soil temperature, may produce undesirable
litter and in the soil microclimate and soil chemistry soil moisture conditions and thereby reduce soil pro-
that they induce. ductivity, humus is important in nutrient and moisture
There is growing recognition of the importance of storage, and can confer resilience to forest ecosystems
the species of plant producing litterfall. Much or even in the face of disturbance (the topic of forest floor
most of the difference in decomposition rates between types is considered again in Chapter 11). Although
forests with different climates and soils is the result of conifers often do occur on poor soils, careful experi-
differences in the species in these different forests. ments have not supported the contention that conifers
Wedin and Tilman (1990) found a 10-fold difference cause soil impoverishment, and nitrogen availability
in nitrogen-mineralization rates under monocultures is often as high or higher under conifers than un-
of five different perennial grass species. Old growth der hardwoods on the same site. Binkley (1995) con-
forests of western redcedar—-western hemlock on cluded that the effect of species on soils is related to
northern Vancouver Island have lower levels of total the quality of their litterfall, which is related to the
and extractable nitrogen and lower nitrogen-mineral- lignin : nitrogen ratio discussed above, rather than whether
ization rates than natural second-growth western the species is a conifer or a broad-leaved hardwood.
hemlock—amabilis fir stands growing side by side on
the same ecosystem site type (Prescott et al., 1993). Chemical Composition of the Litter. The chemi-
This was attributed to differences in litter quantity cal composition of the litter influences the acidity
and quality and species effects on tree litter and log (pH) of the forest floor and the nutritional value of the
decomposition rates (Keenan, 1993; Keenan et al., litter to the decomposer organisms. The great varia-
1995a,b), in differences in understory species (Messier tion in litter inorganic chemistry is illustrated in Table
and Kimmins, 1990), and various other factors 5—9, and there is also an important variation in or-
(Prescott and Weetman, 1994). Much of the differ- ganic constituents of litter, such as sugars, amino acids,
ence in soil chemistry and site fertility between differ- and phenols. These may affect the microflora directly,
ent forest ecosystems thus is a consequence of the type or indirectly by regulating the nutritional availability
of species growing there, which, in turn, reflects in of other litter chemicals. The relative abundance of
part the soil condition created by the species of the the different chemicals is also important. The
plant community of the previous seral stage (see carbon : nitrogen ratio has traditionally been consid-
Chapter 16). ered a good indicator of decomposability; litter with a
The effect of tree species on soils has long been a very high carbon : nitrogen ratio generally decomposes
controversial topic. The debate over whether some very slowly. This is partly because microbes need nitro-
species are soil “improvers” and others are soil “de- gen to be able to utilize carbon, partly because of the
graders” originated in concerns about the accumula- high acidity that generally accompanies high
tion of “raw humus” in spruce plantations on former carbon : nitrogen ratios, and partly because the nitrogen
hardwood forest sites in Europe (Bonnevie-Svendsen in high carbon : nitrogen material is often complexed
and Gjems, 1956; Handley, 1954; McNeill, 1955; with decomposition-inhibiting chemicals such as lignin
Siren, 1955). However, there have always been those and tannins. However, a high carbon:nitrogen ratio
who have challenged the “conventional wisdom” that does not necessarily correlate with slow decomposition.
species that produce a thick, acidic forest floor are nec- Low carbon : nitrogen ratio material may decompose
essarily soil degraders that lower site productivity (e.g., very slowly because all the easily decomposed carbon
 The Biogeochemical Cycle: Nutrient Cycling Within an Ecosystem 101

Table 5—9 Variation in the Chemical Composition of Freshly Fallen Leaf Litter

Composition, %
Species N P K Ca Mg Ash

Sweet birch (Betula lenta) 0.72 0.17 Os 1:65 0.28 =

Yellow poplar (Liriodendron tulipifera) 0.51 0.11 0.95 2.56 0.45 10.67
Balsam fir (Abies balsamea) Wes 0.09 0.12 Vale 0.16 3.08
Shagbark hickory (Carya ovata) 0.91 0.16 1.18 2.09 _ 9.59
American basswood (Tilia americanum) 1.01 0.12 0.52 3.84 0.77 1OwIG
Cucumber tree (Magnolia accuminata) 0.58 0.28 0.76 aera 0.29 =
Sugar maple (Acer saccharum) 0.67 0.11 0.75 1.81 0.24 =
Norway spruce (Picea abies) 1.02 0.09 0.39 1.96 0.23 =
Jack pine (Pinus banksiana) 0.58 0.04 0.16 0.61 _ 4.15
Northern white cedar (Thuja occidentalis) 0.60 0.04 0.25 2.16 0.15 —
Eastern hemlock (7suga canadensis) 1.05 0.07 0.27 0.68 0.14 —

Source: Lutz and Chandler, 1946. Copyright John Wiley and Sons. Used with permission.

compounds have been used up, and high carbon : ni- the climate. In warmer, drier climates, decomposition
trogen ratio material with little tannin and lignin (e.¢., rate Is very sensitive to carbon : nitrogen and lignin: ni-
straw) may decompose rapidly because the carbon that trogen ratios, whereas in cool-humid and cold environ-
it contains is accessible to decomposer organisms. De- ments, decomposition of all litter types is slow and is
composition can sometimes be more limited by lack of less affected by the lignin: nitrogen ratio (Meentmeyer,
available carbon than by lack of available nitrogen. 1978). When lignin content is low or when one is con-
Of all the chemical characteristics of litter, the ini- sidering litter with a wide range of lignin contents, car-
tial concentrations of nitrogen and the lignin:nitrogen bon : nitrogen is a better predictor of decomposition
ratio have proved to be the best predictors of litter de- than lignin: nitrogen; nitrogen may be a better predic-
composition rate and nitrogen mineralization. Litter tor of the early stages of decomposition, and lignin of
with one or both of high lignin and low nitrogen com- the later stages (Taylor et al., 1989, 1991).
pose slowly, and the ratio of lignin to nitrogen is an Considering the well-established relationship be-
even better predictor of decomposition than either tween litter nitrogen and both decomposition and im-
measure alone (Fogel and Cromak, 1977; Harmon et mobilization, one would expect that fertilization
al., 1990; Meentmeyer, 1978; Melillo et al., 1982). Lit- would increase rates of decomposition. However,
ter with carbon : nitrogen ratios greater than approxi- most studies have found little or no effect of increas-
mately 30 immobilizes nitrogen. This raises litter ing nitrogen availability (e.g., Prescott, 1995;
nitrogen content, lowers carbon : nitrogen ratio, and Theodorou and Bowen, 1990), although where fertil-
increases microbial activity, which in turn releases car- ization is accompanied by improved moisture status
bon as the litter decomposes, further reducing carbon: (Baker et al., 1990), decomposition may increase.
nitrogen. As carbon: nitrogen approaches approxi- It should be clear that the effect of litterfall quality
mately 30, nitrogen immobilization ceases because the on decomposition rates is complex. Having single-
decomposers now have sufficient access to nitrogen, species litter versus mixed-species litter may or may
and, as decomposition continues, nitrogen is released not affect rates, depending on species and ecosystem
as fast as carbon is lost. type (e.g., Blair et al., 1990, versus Klemmedson,
The relationship between carbon : nitrogen and de- 1992), and the degree to which litter has lost its “read-
composition is modified by the lignin content of the lit- ily leachable fraction” will modify the relationship dis-
ter, expressed as the lignin:nitrogen ratio; the higher cussed above (Harmon et al., 1990). Other references
the lignin content and the ratio, the slower the decom- that should be consulted on this topic include Mac-
position rate. This relationship is further modified by Claugherty et al. (1985).
102 CHAPTER 5 Biogeochemistry

Acidity (pH) of the Litter and of the Forest Floor corrhizal roots reported from New Zealand (Gadg-
That It Produces. Many soil organisms are intol- il and Gadgil, 1975) was not found in northern tem-
erant of very acidic forest floor conditions. Fungi are perate spruce forests (Harmer and Alexander, 1985;
less adversely affected and dominate the soil flora in Staaf, 1988).
the decomposition of acid forest floors. Mites and
springtails, which dominate the soil fauna under Fertility Status of the Soil. A plant growing on a
acidic conditions, decompose litter more slowly than fertile soil tends to have higher concentrations of nu-
bacteria and earthworms that dominate the soil flora trients in both foliage and litter and lower levels of
and fauna, respectively, in less acid soils. Acid sub- chemicals such as lignin and tannin that inhibit de-
strates are normally characterized by slow litter de- composition than does the same plant growing on an
composition, although this is not always the case. Red infertile soil. Fertile soil generally supports plant
alder (A/nus rubra) litter decomposes rapidly despite species that have higher concentrations of nutrients
the low pH found in the forest floor beneath many and lower concentrations of decomposition inhibitors
red alder stands. and physically more decomposable litter than species
that generally grow on infertile sites. Thus, by con-
Soil Microclimate. In common with other organ- trolling plant species and litterfall chemistry, soil fer-
isms, soil fauna and flora have certain climatic toler- tility exerts a major influence on the activity of
ances and are not active in soils that are too hot, too decomposer organisms; in many cases, the species ef-
cold, too wet, or too dry. As a result, decomposition fect will dominate.
is sensitive to climate (Meentmeyer, 1978). The Decomposition of large roots, branches, and stems
fastest decomposition occurs in the tropics, where involves different organisms than decomposition of
the temperature and moisture conditions are almost foliage, twigs, and fine roots, but the same principles
constantly favorable for soil organisms, except hold true. Decomposition of woody material is slower
where low litter quality restricts decomposition. than finer litter but can be reasonably fast once the
The slowest decomposition occurs where the soil is material is comminuted (broken into small pieces) by
very hot, very cold, very wet, or very dry for much of the action of wood-boring animals (largely insects op-
the year. erating in conjunction with fungi), whose fecal pellets
Decomposition is affected by disturbances, such become the substrate for fungal and bacterial activity.
as windthrow or timber harvesting, that alter Fungal activity alone will decompose woody material,
microclimate. In a study of leaf and forest floor but the process is much slower than when fungi are
decomposition in an oak forest in Wisconsin, preceded by fauna.
clearcutting reduced decomposition in litter bags The end products of litter decomposition are
and the surface organic layer but increased it in CO;, water inorganic ions, and a variety of stable
lower organic layers, with no overall effect on for- “recalcitrant” organic substances that are added
est floor decomposition (Binkley, 1984; Whitford et to soil humus or mineral SOM. The nature and
al., 1981; Yin et al., 1989). The depth in the forest abundance of these organic materials plays an impor-
floor at which the switch from negative to positive tant role in the development of the forest floor and
effects on decomposition occurs increases as you the soil.
move toward the equator. The effects of canopy re- Decomposition rates determine the quantity of
moval on overall forest floor decomposition rate carbon stored in the soil and play a significant role in
may be minor except at high or low latitudes (Yin et the global carbon cycle. Conversion of forest to agri-
al., 1989), and research suggests that the biological culture has been claimed to increase the loss of soil
and microbial changes in the soil accompanying carbon and contribute to increases in atmospheric
canopy removal and the loss of mycorrhizal roots CO, levels (Bolin et al., 1979; Clark, 1982;
have a greater influence on decomposition rates Schlesinger, 1977). Although this undoubtedly occurs
than the change in microclimate (e.g., Gadgil and in some cases, reports such as those by Desjardins et
Gadgil, 1978). However, there may be a latitudinal al. (1994) and Johnson (1992) suggested that the issue
control in the relative importance of microclimatic cannot be generalized. It will depend on the type of
versus microbial changes. The suppression of radia- forest, the type of soil, and the type and quality of the
ta pine litter decomposition by the presence of my- agriculture.
 The Biogeochemical Cycle: Nutrient Cycling Within an Ecosystem 103

Role of Understory Vegetation take, and Figure 5—9 compares the importance of un-
derstory vegetation in potassium cycling in two differ-
Discussions of nutrient cycling in forest ecosystems ent forest types. It is apparent from these data that
generally refer exclusively or mainly to the overstory estimates of nutrient turnover based on studies of the
tree vegetation. To understand how forests cycle nu- overstory alone will significantly underestimate the
trients, it is necessary to consider the contribution of magnitude of the forest biogeochemical cycle in most
all the vegetation, not only that of the dominant trees. types of forest.
This requires that our discussion be broadened to in- A particularly interesting example of the signifi-
clude understory trees, shrubs, herbs, ferns, mosses, cance of understory vegetation in nutrient cycling is
and epiphytes (plants that grow nonparasitically on the relationship between black spruce and the carpet
other plants). of mosses on the forest floor in central Quebec
The biomass of understory and epiphytic vegeta- (Weetman, 1967; Weetman and Timmer, 1967). In
tion can vary from nothing to as much as 15 t ha“, ac- these stands, the moss was found to contribute 33 to
cording to the tree species, the stand structure, the site 50% of the total aboveground biomass production,
type, and the climate (Rodin and Bazilevich, 1967). and annual uptake of nitrogen, phosphorus, potassi-
Particularly high values occur in “moss forests” um, calcium, and magnesium by the moss was esti-
(forests typified by a continuous blanket of moss on mated to be between 23 and 53% of the annual
the forest floor but little other minor vegetation). uptake by the trees. Studies of precipitation chem-
Dense shrub and herb growth often looks more im- istry, moss decomposition, the rooting pattern of the
pressive but generally amounts to far less weight per black spruce, and the availability of nutrients in the
hectare than does a thick carpet of moss. forest floor led to the conclusion that the moss layer
Minor vegetation biomass rarely constitutes more was the major source of nitrogen for the trees. Nitro-
than a small fraction of the total forest biomass, but its gen from precipitation and throughfall is absorbed by
contribution to nutrient cycling and total stand pro- the moss layer, which holds it for 1 to 3 years before
duction can be significant. It generally has higher con- the moss starts to decompose. Nitrogen that is re-
centrations of chemicals and higher rates of biomass leased is absorbed by mycorrhizal tree roots that grow
turnover (i.e., less storage of net production) than does in the layer of decomposing moss above the forest
the overstory. As a result, the role of minor vegetation floor. Much of the nitrogen in the tree litterfall is
in total forest productivity and biogeochemistry is thought to remain as unavailable organic nitrogen,
much greater than would be expected on the basis of the nitrogen cycle being driven by the 9 kg ha™ that
biomass. For example, the understory of a 120-year-old the tree canopy and moss layer filter out of the pre-
pine forest in the Soviet Union was found to produce cipitation. The moss thus acts as an efficient biologi-
16 times more biomass annually than the trees cal filter that absorbs nutrients from solution in
(P’Yavchenko, 1960). A study in three subalpine forest precipitation and throughfall and subsequently makes
ecosystem types in coastal British Columbia (Figure them available to the trees. Without the moss, much
5-8) revealed that although minor vegetation litterfall of the circulating nutrients would become immobi-
accounted for only 3 to 11% of the aboveground litter- lized in the forest floor. There may also be some ni-
fall biomass, minor vegetation contributed 16 to 38% trogen fixation by the mosses or associated external
of the nitrogen, 14 to 35% of the phosphorus, 5 to 31% microflora. Many species, especially Sphagnum and
of the calcium, 19 to 55% of the magnesium, and an Drepanocladus, have both epiphytic and intracellular
amazing 32 to 90% of the potassium in the total annual nitrogen-fixing algae and bacteria (Granhall and
aboveground litterfall (Yarie, 1980). The understory Lindberg, 1977; Granhall and Selander, 1973; Rosen
also had a significant effect on nutrients in throughfall. and Lindberg, 1980).
This study demonstrates the importance of even a The relatively high concentration of nutrients in
diminutive understory and that the quantity of nutri- the litterfall of understory vegetation (Tables 5-10 to
ents circulating in the minor vegetation is sometimes 5-12) has a generally beneficial effect on overall litter
greater than that circulating in the tree crop. decomposition in the forest floor and thus on site fer-
Tables 5-10 to 5-12 present some comparisons tility. The beneficial effects on stand productivity of
between the contributions of overstory and minor for- maintaining a light understory has long been recog-
est vegetation to leaching, litterfall, and nutrient up- nized by European silviculturists.
104 CHAPTERS Biogeochemistry

Figure 3-8 Contribution of understory and overstory to annual aboveground litterfall in

«

three subalpine forest ecosystems along a minor topographic sequence of site types from dry
Samm

(xeric) to moist (hygric) sites. The forest, which cor asisted of Pacific silver fir, mountain hemlock,
wt

dar in various proportions, was up to 450 yea rs old. (After Yarie, 1980. Used
hoes

ne author's permissio n.)

Table X—10 Chemical Composition and Total Chemical Content of Understory (U) and Overstory (O) Vegetation
Litterfall in Two Forest Types in Eastern U.S.
White Pine Plantation Hardwood Stand
O

ZSlol/

oO

Pal Claes)
Ceas
#i

Cc
Onl
2
a:

ay
7)
o
-_
©
a

x
iS
©

oO
(=
-
—
©
=
°

x
D

rs
©
D
he
TT

ih

(oe
=
=
£
o
a

| oe oe ee De

| On On Gh Qi a
|@e@nnr

pep
-

ener a

Og OIE)

te.

t~ NOY ®D
rc
eaeDatesSe

Cm

owt

ie aeare

One)
oo
or
oy ee

coy
Of
alerietaeNe 22)
Ge

OO
e

oro
=

<6) r= 2s:
ee
Oo @

5h AS
oOo +
Oo
o

-
ene

ee)
c
Gee
ke
coke
OS

©
DD

51/0
ONAN

ice) wt - 1402
Data from Scott, 1955.
 The Biochemical Cycle: Nutrient Cycling Within Plants — 105

Table 5-11 Contributions to Leachate (Lch) and Litterfall (Ltfl) of Overstory and a Fern Understory in an Oak
Wood in England
N,kgha*yr* P,kghatyr? K,kgha‘yr' Ca, kg ha” yr" Mg, kg ha” yr*
a a ee Litterfall
Lch__Ltfl Lch__Ltfl Lch_ Ltfl Lch_ Ltfl Lch_ Litfl Biomass, kg ha-1 yr

Oak overstory 0.9 48.7 0.6 2.8 16:08 15:3 WAS S25 5.6 4.7 5196
Fern understory 0.3 12.9 0.1 0.9 9.4 0.1 4.0 se) ies 1470

Data from Carlisle et al., 1967.

Direct Nutrient Cycling ing on nutrient-deficient soils in the Amazon Basin,

but it is probably also the dominant pathway of nutri-
In many forests, recovery of nutrients by plants from
ent recovery on nutrient-poor sites in temperate and
decomposing litter involves the combination of uptake
boreal forests.
from soil solution by both mycorrhizal and nonmycor-
rhizal roots and direct uptake by mycorrhizal fungi
from decomposing organic matter. In some forests, nu-
trients are recovered almost exclusively by the latter 5.6 ‘The Biochemical Cycle:
method: the direct nutrient cycling pathway (Stark, 1972). Nutrient Cycling Within Plants
Direct nutrient cycling involves the invasion of
fresh litter by the mycelium of a mycorrhizal fungus. Plants do not rely solely on root and foliar uptake to
The hyphae penetrate into the litter and decompose satisfy the current nutrient demands of height, diame-
it, absorbing the nutrients as they are mineralized. ter, and root growth. They are able to provide at least
Some of these nutrients become available to the myc- some of their needs by internal redistribution of nutri-
orrhizal plant root. This decomposition pathway ents that are already inside the plant. Animals behave
avoids the soil solution phase, which in turn prevents similarly. There is a constant exchange of calcium be-
the nutrients from being leached away and also pre- tween blood and bones, but in lactating mammals,
vents them from being immobilized by nonmycor- there is a net withdrawal of calcium from bones to
rhizal microbes. blood if the dietary intake of calcium is inadequate to
This is the most conservative of the biogeochemi- meet the calcium demands of milk production. The
cal pathways. Direct nutrient cycling ensures a high internal redistribution of nutrients in plants, the
efficiency of recovery of nutrients lost from plants in biochemical cycle, constitutes an important mechanism
litter and thereby contributes to an almost closed bio- by which plants conserve nutrients.
geochemical cycle. It is particularly well developed on Plants would lose a much greater quantity of nitro-
nutrient-poor soils where there has been strong selec- gen, phosphorus, and potassium in leaf litterfall if they
tive pressure in favor of nutrient conservation. The were not able to retranslocate a substantial proportion
pathway was first described for tropical forests grow- of these nutrients from old leaves before shedding

Table 5-12 Contributions of Overstory and Understory Vegetation to Annual Nutrient Uptake in a Pine Stand
in Finland
Percentage of
Percentage of Total Uptake aera

N P K Ca Percentage of Biomass Production

Pine overstory 54.6 53.6 54.1 58.4 84 71

Understory 45.4 46.4 45.9 41.6 16 29

Data from Malkonen, 1974.

106 CHAPTERS — Biogeochemistry

Wood 4—— Tree [Branches Max),Foliag e Wood ¢—Tree [—) Branches [Foliage
>gZ
stem 3
stem &
2 Tree
%, ows Tree
ow leaf-fall
Uptake
G
om ox leaf-fall
Litter Litter

Understory Understor:
litterfall litterfall
Understory
Understory
vegetation
vegetation
understory Uptake by
vegetation Output understory
arcs > Trees vegetation
Scale: 100 kg ha™! ea) ;

>> Understory

(a) (b)

Figure 5-9 Comparison of the role of understory vegetation (shaded) in the potassium cycle
in stands of two different species growing adjacent to each other on similar sites in England.
(A) Pedunculate oak, 47 years old. (B) Scots pine, 47 years old. The magnitude of both the overstory
and the understory cycle varies between the two stands. (After Ovington, 1965. Reproduced by per-
mission of Hodder & Stoughton Limited and J.D. Ovington.)

them. In a study in Finland, it was shown that 4-year- dles (Switzer et al., 1968). The translocation out of
old needles of Scots pine lost 17% of their weight, old Scots pine needles in Finland was found to equal
69% of their nitrogen, 81% of their phosphorus, and 23 and 33% of the nitrogen and potassium, respec-
80% of their potassium before they became litterfall. tively, required for the production of new needles
The nutrients were translocated out of the needles and (Malkonen, 1974). Table 5-14 examines the impor-
stored initially in the bark and new wood of branches tance of internal transfer in comparison with other
adjacent to the old needles (Malkonen, 1974). A study sources of nutrients in satisfying the annual nutrient
in the southern United States reported that loblolly requirements of loblolly pine. Internal cycling is ob-
pine needles lost 44, 38, and 58% of their nitrogen viously very important for some elements in some
phosphorus, and potassium, respectively, and 14% of species.
their weight just before abscission (Wells and Metz, Redistribution of nutrients is important to
1963). Table 5-13 summarizes the efficiency of nitro- plants in many ways. By conserving nutrients within
gen and phosphorus retranslocation in a variety of tree their tissues, plants can sustain growth on sites on
species. More recent studies have shown that re- which the availability of nutrients for uptake from
translocation occurs from virtually all ages of tissues, the soil is very low. They can also sustain growth at
not just at the time ofleaf senescence. Retranslocation a time of year when nutrient availability in the soil
is therefore probably much more important in plant is very low (e.g., in the spring on soils that are very
nutrition than suggested by the earlier studies (Millard cold and wet). Nutrients continue to be absorbed
and Proe, 1993; Nambiar and Fife, 1991; Proe and and stored by plants as they become available in the
Millard, 1994a,b). soil later in the season, even though they may not be
‘Trees are able to satisfy an appreciable propor- needed right away for active growth. Redistribution
tion of their annual nutrient requirements by this also has some practical significance. The conserva-
mechanism. In one study, 20-year-old loblolly pines tion of nitrogen within the tree crown may be one
were found to satisfy 45% of their annual nitrogen reason that the stimulation of tree growth by a sin-
requirement by translocation from yellowing nee- gle addition of nitrogen fertilizer persists for sever-
 The Biochemical Cycle: Nutrient Cycling Within Plants 107

Table 5-13 Percentage of the Presenescence Nitrogen and Phosphorus Translocated out of Foliage
before Litterfall

Nitrogen Phosphorus
(%)
ee
ee
(%)
2 oe
Reference
ee ee ee

Douglas-fir
hygric 61 73 Parkinson, 1984
mesic 69 35
xeric 61 {
Douglas-fir
42 yr 42 46 Turner, 1975
49 yr 36 63
Loblolly pine 45 66 Switzer and Nelson, 1972
Scots pine 69 81 Malkonen, 1975
Evergreens
bog 45-71 45-75 Small, 1972
nonbog 51-73 66-98
Deciduous
bog 23-68 46-80
nonbog oe 4-71
Scots pine: poor site 77 — Stachurski and Zimka, 1975
Hornbeam
poor site 64 _
good site Sif
Oak
poor site 63 —
good site 45 =e

Alder: good site 5) _

Beech 72 70 Staaf, 1982

Chestnut oak
better 80 65 Ostman and Weaver, 1982
poorer 76 61
Eucalyptus — 60 Attiwill, 1980

Eastern cottonwood 74 66 Baker and Blackmon, 1977

Mixed hardwoods Sil 61 Ryan and Borman, 1982

Pin cherry OM 59
Loblolly pine 52 47 Wells and Metz, 1963
upper crown 99 52
mid crown ey 46
lower crown 47 45
Pine 43 = Rapp et al., 1979

Subalpine conifers 54 59 Turner and Singer, 1976

al years after treatment. If this is so, then fertiliza- The efficiency of nutrient recovery and redistri-
tion may have better results in forests where inter- bution at the time of leaf abscission can vary according
nal redistribution is well developed than in those in to the availability of nutrients in the soil. It has been
which it is providing only a minor contribution to shown that the rate of nutrient withdrawal from aging
tree nutrition. radiata pine needles was much greater on nutrient-
108 CHAPTERS — Biogeochemistry

Table 5-14 Relative Contributions of Different Nutrient Sources to the Annual Nutrient Requirements of a
20-year-Old Loblolly Pine Plantation
Percentage Contribution to the Requirement of

Cycle Nutrient Source N P K Ca? Mg

Geochemical Precipitation 16 6 12 Si 16
Mineral soil 0 2 0 0 6

Biogeochemical Forest floor (litter decomposition) 40 23 16 47 38

Canopy washing and leaching 5 9 50 22 16

Biochemical Internal transfer 39 60 20 0 24

Data from Switzer and Nelson, 1972.

“Original data adjusted.

poor soils than on more fertile soils (Florence and immobilization of much of the available nitrogen in
Chuong, 1974). Recovery of nitrogen and phospho- the forest floor. The trees responded to this sudden
rus from senescing foliage in a Jamaican rain forest reduction in nitrogen availability by increasing the re-
was reported to be 14 and 19%, respectively, for a fer- distribution of nitrogen from old foliage to younger
tile, low-elevation site and 50 and 65% for an infer- foliage and by shedding the oldest foliage, presum-
tile, high-elevation site (Tanner, 1980). A confirm- ably because it no longer had sufficient nitrogen to
ation of the relationship between internal cycling operate efficiently; the old foliage was sacrificed in
and external nutrient availability was provided by the face of serious nitrogen shortages to maintain the
Turner (1977), who altered the availability of nitro- production and photosynthetic activity of young fo-
gen in the forest floor of aDouglas-fir stand in coastal liage. ‘Table 5-15 summarizes the effects of the in-
Washington by applying sucrose and sawdust to duced nitrogen deficiency (sucrose and sawdust) and
the ground beneath the trees. The provision of a also of induced nitrogen enrichment (fertilization) on
readily available energy source greatly increased mi- internal cycling and foliage biomass. It can be seen
crobial activity, which in turn resulted in short-term that uptake of nitrogen provided only 15% of the ni-

Table 5—15 Effect of Induced Increases and Decreases in Soil Nitrogen Availability on the Biomass, Nitrogen
Content, and Internal Cycling of Douglas-Fir Foliage
N® Fertilization Untreated Control Carbohydrate Treatment

B> N B N B N

Total foliage 10,340 137 9,400 94 8,480 80

Current foliage 2,550 SL 2,200 22 1,790 19
Foliage litterfall 1,290 1 er 1,880 14.1 2,390 14.3
Uptake of N from soil 63 17 4
Recycling from old foliage -13 14 22
Ratio: Uptake/current
tissue content (%) 126 55 15
Uptake & retranslocation 50 31 26

Source: Turner, 1977. Copyright American Society of Foresters. Used with permission.
aN, nitrogen, kg ha; 880 kg ha~ of urea.
>B, biomass, kg ha”.
 The Biochemical Cycle: Nutrient Cycling Within Plants 109

trogen needed for new foliage production in the car- (a maximum of 31 years) at elevations above 1500 m.
bohydrate treatment, compared with 55% in the un- Retention of a large biomass of old foliage provides a
treated control. The fertilized trees took up 26% large volume from which nutrients may be withdrawn
more than was needed for the new foliage (luxury to support new growth when the air is warm but the
consumption). However, in contrast to these results, soil is still cold or waterlogged—conditions that limit
many studies have failed to demonstrate a strong rela- nutrient uptake. Having a large number of small annu-
tionship between percentage of retranslocation and al foliage increments also means that only a very small
soil nutrient availability (e.g., del Arco et al., 1991; fraction of the foliage is dropped each year. This re-
Helmisaari, 1992; Proe and Millard, 1994a,b). duces the need for a large recovery of nutrients at the
Chapin and Moilanen (1991) concluded that re- time of litterfall, and it has been suggested (D. W.
translocation efficiency is more related to rates of Cole, personal communication, 1972) that the degree
photosynthesis than the nutritional status of plants of internal cycling is inversely proportional to the
(i.e., to demand rather than to supply). number of years of foliage retention. Deciduous
The efficiency of internal cycling is determined conifers, such as larch, and deciduous angiosperms
not only by the fertility of the soil but also by any fac- that grow on infertile soils recycle a very high propor-
tor that influences plant growth and plant uptake de- tion of foliar nitrogen, phosphorus, and potassium.
mand. Moisture thus has a major effect on internal Pines that retain only 2 to 4 years of needles are some-
cycling. Parkinson (1984) found that internal cycling what less efficient but still recover a large proportion
of nitrogen did not vary much along a gradient of in- of these nutrients before leaf fall. Species such as
creasing nitrogen availability in the soil because an ac- spruces and true firs, which carry many years of fo-
companying increase in soil moisture enabled the tree liage, have a much lower need to develop highly effi-
to utilize all the available soil nitrogen even on the cient internal cycling. Escudero et al. (1992)
richest site. The confounding of moisture and nutri- supported the idea that variations in leaf retention
ents may explain some of the inconsistencies in the lit- may be as important to a plant as retranslocation in
erature on internal cycling. Nambiar and Fife (1991) determining the efficiency with which nutrients are
reported that the key factors in determining retranslo- used. Retranslocation may also be related to the dura-
cation are growth and nutrient uptake demand. They tion of leaf abscission; species with gradual loss of fo-
claimed that nutrient retranslocation efficiency is in- liage may have less efficient retranslocation than
creased by high soil fertility and rapid growth; re- species with sudden and predictable leaf fall (del Arco
translocation is driven more by shoot growth than soil et al., 1991).
nutrient supply. In a study of the response of re- Much of the discussion of retranslocation has fo-
translocation in radiata pine needles to fertilization cused on senescing leaves. We now know that re-
and irrigation, Crane and Banks (1992) found that re- translocation can occur at almost any time, according
translocation was correlated with growth only when to the plant internal nutrient requirements (Proe and
stands in the moisture-limited study area were irri- Millard, 1994a,b), and that it occurs in roots (e.g.,
gated. Similar conclusions were reached by Chapin Millard and Proe, 1993) and in stems. The percentage
and Moilanen (1991). recovery of phosphorus as Eucalyptus sapwood be-
It has been suggested that the physiological char- comes heartwood can exceed 90% (Attiwill and
acter of evergreenness (as opposed to deciduousness) Leeper, 1987; Ferreira, 1984). Van den Driessche
may be related to the storage and conservation of nu- (1984) reviewed a variety of nutrient storage sites that
trients (Beadle, 1966; Loveless, 1961; Monk, 1966; are involved in retranslocation.
Muller-Stoll, 1947; Reich et al., 1995; Richards, Considerable disagreement persists concerning
1968). The observation that the degree of evergreen- the relationship of internal cycling to soil nutrient
ness of a given species varies in different habitats may availability, and it is clear that a plant’s retranslocation
in some cases be related to biogeochemical cycling strategy is much more complex and is affected by far
and nutrient use efficiency (see discussions in Sheriff more determinants than originally envisaged. How-
et al., 1995). As noted in Chapter 4, Pacific silver fir in ever, internal cycling obviously plays a vital role in
southwestern British Columbia retains 6 to 8 years of plant nutrition and in determining the efficiency with
needles at low elevation (300 m), 13 to 19 years of nee- which plants can utilize the solar energy available in a
dles at medium elevation (1000 m), and up to 25 years particular ecosystem.
110 CHAPTERS — Biogeochemistry

Until relatively recently, biogeochemical studies cycle incorporate many uncertainties and must be ac-
have tended to focus on one of the three major cycles: cepted with appropriate caution. Nitrogen exists in
geochemical, biogeochemical, or biochemical. This the atmosphere as dinitrogen gas (N2), ammonia gas
was reasonable because the scarcity of data on all as- (NH;), and nitrate (NO;), or ammonium (NH,’),
pects of nutrient cycling required that a lot of detailed ions, and can enter biogeochemical cycles in any of
data gathering be undertaken. However, there now ex- these four chemical forms. Ammonia gas is absorbed
ists an extensive literature on forest biogeochemistry, directly by plant foliage, and such direct uptake from
and increasing attention is being paid to synthesis: to a the atmosphere may account for up to 10% of the
consideration of the overall biogeochemical character plant’s nitrogen needs (Hutchinson et al., 1972). Ni-
of forest ecosystems and how this varies between dif- trate and ammonium ions enter ecosystems as dry fall-
ferent ecosystems and between different developmen- out or in precipitation, and this can add appreciable
tal stages of a given ecosystem. Attiwill and Leeper quantities of nitrogen annually to some ecosystems
(1987) reviewed nutrient cycling and its role in the (Table 5-2). However, it is biological fixation of N>
ecology of Australian forests. Bowen and Nambiar that accounts for the greatest nitrogen input in most
(1984) showed how knowledge of nutrient cycling and ecosystems (Wollum and Davey, 1975), and estimates
tree nutrition is used in the sustainable management of up to 200 kg ha"! yr! have been made for legumi-
of plantations. nous crops (Stewart, 1966) and 320 kg ha! yr‘ for
stands of red alder (Newton et al., 1968). Terrestrial
biological fixation has been estimated at 98 million
5.7 Nitrogen Cycling: The Key to metric tons per year on land with an additional 28,000
Productivity in Many Forests tons in the oceans (Delwiche, 1981). This is probably
an underestimation.
As a gas, nitrogen is abundant: it makes up approxi- Very few types of organism have evolved the abil-
mately 79% of the world’s atmosphere (i.e., 790,000 ity to convert biologically unavailable N, into biologi-
ppm). This vast atmospheric reservoir of nitrogen is cally available reduced or oxidized nitrogen; of those
unavailable to most organisms, and net production in that can, all are microorganisms (Table 5-16). Many
much of the world’s vegetation is limited by lack of ni- species of bacteria and algae are known to fix nitrogen.
trogen. This is a situation not unlike a person dying of Some of these are free living, such as soil bacteria, the
thirst in the middle of the ocean—“water, water every- purple bacteria, several other photosynthetic bacteria,
where nor any drop to drink.”’ The nitrogen molecule and several species of blue-green algae. Others oper-
N,j is an inert, unreactive gas that requires a high input ate symbiotically (a relationship between two different
of energy to break the nitrogen-nitrogen bond before types of organisms that is favorable to both of them)
it will combine chemically with other atoms. Industri- with higher plants. The bacteria invade the living tis-
al fixation of nitrogen in the manufacture of fertilizers sues of the plant and stimulate cell division and
requires a temperature of 450°C and pressure of be- growth. This results in the formation of a nodule in
tween 250 and 1,000 atm. It is remarkable, therefore, which the bacteria live and fix atmospheric nitrogen.
to find that certain organisms have evolved the ability Bacteria of the genus Rhizobium stimulate root nod-
to reduce atmospheric nitrogen and combine it into ules on leguminous plants, and similar nitrogen-fixing
organic molecules under normal atmospheric pressure root nodules are formed on the roots of alder (con-
and growing season temperatures. This biological at- taining species of Frankia) and several other tree
tribute is likely to be heavily exploited in the future as species. Nitrogen-fixing nodules also form on the
the economic and fossil fuel carbon release costs of en- leaves of some tropical trees, and nonnodule nitrogen-
ergy and thus the costs of making nitrogen fertilizer fixing bacteria are known to live on (or in) the leaves
continue to rise. of other trees, including conifers (Jones, 1970, 1976;
The main pathways of nitrogen movement in geo- Jones et al., 1974). With the help of the enzyme nitro-
chemical and biogeochemical cycles are shown in genase, these organisms are able to reduce N, to am-
Figure 5—10, together with estimates of the quantities monia, which is converted to ammonium ions and
involved. Quantitative estimates of the world nitrogen incorporated into the metabolic activities of the plant.
Nitrogenase contains iron and molybdenum, which,
together with cobalt, are essential micronutrients for
From the poem “The Rime ofthe Ancient Mariner” by Samuel Coleridge. nitrogen fixation (Epstein, 1972). Reviews of the role
 (a)
Atmospheric nitrogen
N,, NH3, and some nitrate and
Volcanism ammonium ions (NO;, NH;)

Combustion of fossil fuels

and industrial emissions
A _ Denilndustrial Uptake of atmospheric Biological fixation Atmospheric fixation
/ fixation and combustion NH, 3 by plants N,> NH, 3 >NO, 3
| 2 (lightning
g .) and
56 X 10° Industrial
12 <
photochemical fixation)
i] fixation and combustion
I 56 X 10!" trification
I NO, > N,
| Organisms living in
} the roots or on the
leaves of plants
\ P Free living
i organisms
\\ Industrial fixation
and soil adsorption
of atmospheric NH,
Nitrate nitrogen
Nitrogen in
in soil or water
organic
sediments
a
> Nitrate bacteria
“ Nitrogen in Denitrificat NO; > NO;
: : enitrification
\ \Y Nitrogen in plants ——ennen \K
‘S grazing trophic Bina
Ne webs ee Nitrification
acteria
SS NO, > NO,
~S Denitrifying
[s Nitrogen in bacteria
~ detritus trophic NO, > NH,
webs Nitrate bacteria
NO; > NO;
Ammonification

Organic N > NH, Ammonia or ammonium nitrogen

in soil or water
a+
NH, Nig

(b)
Atmospheric N
39 x 108

Biological
’ fixation |
Animals 98 x 107
Industrial : ;
te cara Z © 5 Biological
fixation and combustion Seiret Renton
: ; 56 x 10!” iologica 3 Z
4 eae denitrification 28 X 10
Poco 42 x 10"

Dead organic
matter
16.8 X 107 | Inorganic
‘| nitrogen |
Organic and inorganic xe] 100X 10°} %
Inorganic nitrogen in river water
nitrogen
15.4 X 10°
Weathering 28 X 10"

Figure 5-10 The nitrogen cycle. (A) Distribution and transfers of nitrogen in the biosphere. Double lines indicate the biogeo-
chemical cycle, single lines indicate the geochemical cycle. Solid lines indicate major transfers, dashed lines indicate secondary trans-
fers. Nitrogen cycles in aquatic and terrestrial ecosystems are not shown separately. (B) Estimates of quantities and rates of transfer of
nitrogen in the biosphere. Units for inventories (numbers in the boxes) are 10'? g (millions of metric tons). Transfer rates are ex-
pressed as g yr!. (Redrawn after Delwiche, 1981. Copyright John Wiley & Sons, Inc., New York. Redrawn by permission.)
18 CHAPTER 5 Biogeochemistry

Table 5-16 Biological Nitrogen-Fixing Systems

A. Symbiotic associations
1. Root nodules
a. Legumes with root-nodule bacteria (Rhizobium): peas, beans, clover, lupin, soybean, etc.
Casuarina,
b. Nonlegumes with root-nodule organisms (bacteria and actinomycetes): A/nus, Arctostaphylos, Artemesia,
Pur-
Ceanothus, Cercocarpus, Coriaria, Discaria, Dryas, Elaeagnus, Hippophae, Opuntia, Myrica, Arafucaria, Ginko,
shia, and Sheperdia
c. Nonlegumes with root-nodule blue-green algae; tropical trees of the order Cycadales
2. Leaf nodules
a. Nonlegume angiosperms with leaf-nodule bacteria; tropical trees; Psychotrica, Ardisia, etc. :
b. Nonlegume angiosperms with leaf-gland blue-green algae; herbaceous species in moist tropical habitats: Gunnera sp.
B. Less intimate associations
1. Bacteria on the leaf surfaces of tropical rain forest trees
2. Bacteria on the leaf surfaces of temperate zone trees
3. Blue-green algae associated with ferns, liverworts, and fungi
C. Free-living organisms
1. Blue-green algae (Cyanophyta), e.g., Nostoc, Anabaena, Gleotrichia, Trichodesmium
2. Fungi, yeasts, actinomycetes
3. Bacteria
a. Aerobic, e.g., Azotobacter, Beijerinckia
b. Facultative, e.g., Aerobacter
c. Anaerobic
i. Nonphotosynthetic, e.g., Clostridium
ii. Photosynthetic, e.g., Rhodospinillium

After Bond, 1970; Epstein 1972; Youngberg and Wollum, 1970.

of nitrogen fixation in the production of forest bio- mental pine systems (no symbiotic fixation). This rate
mass can be found in Fortin et al. (1980) and was approximately 20% of the rate estimated for
Chatarpaul and Carlisle (1983). The use of plants ca- symbiotic nitrogen fixation by black alder. If these ex-
pable of nitrogen fixation will undoubtedly increase perimental estimates can be extrapolated to field
on nitrogen-deficient sites (e.g., Binkley, 1983). conditions, then our understanding of the nitrogen
Although symbiotic nitrogen fixation is the main economy of forests that lack symbiotic nitrogen fixa-
input of nitrogen to ecosystems in which there is a sig- tion will have to be revised.
nificant number of symbiotic nitrogen-fixing plants, it Biological fixation of nitrogen is largely associated
is now recognized that nonsymbiotic fixation can be with early successional species, such as alder and some
important in many forest ecosystems. Nitrogen fixa- herbs and shrubs, which are generally light demand-
tion by free-living microorganisms in decaying wood ing. Thus, symbiotic nitrogen fixation is largely a phe-
has been identified in all forests in which this process nomenon of early successional stages and ecosystem
has been investigated (Hendrickson, 1991; Jurgenson disturbance. In contrast, nonsymbiotic fixation associ-
et al., 1987, 1992). Although the rates are generally ated with decaying wood, forest floors, and nonsymbi-
low (less than 3 kg ha™ yr! and mostly less than 1 kg otic plants will generally supply nitrogen at lower rates
ha yr), the input is important in the decomposition over the entire successional sequence. Arboreal
of wood, and, combined with nonsymbiotic nitrogen lichens that grow on stems and branches can be im-
fixation in the mineral soil, in the litter layer, and even portant in nitrogen fixation in old forests, and also
in the stems of living trees, this addition of nitrogen play a role in the nitrogen economy of younger forests
may play an important role in maintaining ecosystem if the stand density is not too high. Because lichens
nitrogen budgets over long periods. The significance grow slowly, they do not accumulate much biomass on
of long-term nitrogen budgets of reducing the amount stems and branches unless these remain well illumi-
of decaying wood by forest harvesting is a topic of re- nated for many years. They are much less important in
search interest (e.g., Jurgensen et al., 1992). Nonsym- dense young forests.
biotic nitrogen fixation may also be significant over Another important source of nitrogen for terres-
shorter periods. Bormann et al. (1993) estimated fixa- trial ecosystems is industrially fixed fertilizer nitrogen.
tion rates of approximately 50 kg ha” yr"! in experi- This was estimated to be 40 million tons in the mid-
 Biogeochemical Efficiency ofForest Ecosystems 113

1970s and 50 to 60 million tons in 1980 (Delwiche, is by far the more important mechanism. Some fungi
1981; United Nations, 1981), and the rate of fertilizer and autotrophs are also capable of denitrification. The
use has increased greatly in the past 25 years. How- most complete denitrification occurs under anaerobic
ever, continued increase in industrial fixation may be conditions, which are found in waterlogged soils,
limited by the future supply and cost of energy, so ex- poorly aerated soils, or soils rich in decomposing or-
trapolations of historical trends should be treated with ganic matter. Under more aerobic conditions, the re-
caution. Most fertilizer nitrogen is used in agriculture. duction is less complete and may proceed only from
Economic considerations continue to limit fertilizer NO; to NO,, which is then reoxidized by nitrate
use in most forests, despite the considerable potential bacteria. Nitrogen that is reduced as far as N,O, No,
of nitrogen fertilization to increase forest growth. or NH; can be lost from the biogeochemical cycle.
Once nitrogen has entered ecosystems, it under- Under appropriate conditions, denitrification can re-
goes a complex series of alterations. It is taken up by sult in losses of up to 50% of nitrogen that has been
plants as NO; or NH," and converted to organic ni- added to a soil in the form of fertilizer in a period as
trogen before proceeding along either grazing or de- short as 2 to 4 days; it is therefore of considerable sig-
tritus trophic chains to end up as soil nitrogen. nificance in the nitrogen balance of the ecosystem.
Organic nitrogen may be stored in soils for many cen- The circulation of nitrogen thus is seen to be
turies or even millennia, but sooner or later much of it under a greater degree of biological control than is the
is converted to NH; or NH,* by a variety of het- circulation of carbon. The atmospheric pool of nitro-
erotrophic microorganisms (fungi, bacteria, and actin- gen is so large (estimated to be 3.9 X 10°! g or
omycetes) in the process of ammonification. Nitrogen 3,900,000 billion tons [Delwiche, 1981]) relative to
in this form may be taken up by plants, held by the the estimated annual withdrawals (approximately 210
soil, or converted to NO; by a variety of either million tons) that its size can be considered to be inde-
chemoautotrophic or heterotrophic bacteria in a two- pendent of ecosystem processes. For this reason, in-
stage process called nitrification. Nitrite bacteria (e.g., terest in the nitrogen cycle focuses on biological
of the genus Nitrosomonas) convert NH; or NH,* to processes within the biogeochemical cycle that control
nitrite ions (NO, ), which are the substrate for nitrate the circulation and retention of nitrogen and the bio-
bacteria (e.g., of the genus Nitrobacter), which convert logically controlled acquisition from and loss to the
it to NO;. The energy released during this oxidation geochemical cycle. Abiotic inputs from the geochemi-
is used to reduce and incorporate CO), into organic cal cycle are often of secondary importance, although
molecules that the microorganisms require for with the increase in air pollution and the phenomenon
growth. Nitrification only occurs slowly under condi- of “acid rain,” atmospheric inputs of nitrogen have be-
tions of high acidity and low temperature and may be come very important in some areas (see the discussion
inhibited by the presence of certain types of vegeta- in the section “Acid Rain and Its Effects on Forests
tion (Rice and Pancholy, 1972). It is therefore general- and Lakes”). Also, nitrogen dissolved in slope seepage
ly a less important pathway for nitrogen in coniferous waters can be an important input into productive
than in hardwood forests, in northern than in south- lower-slope ecosystems.
ern forests, and in high-elevation than in lower-eleva- Earlier reviews of terrestrial nitrogen cycles can be
tion forests. NO; released by nitrification is taken up found in Bolin and Cook (1983), Clark and Rosswall
by plants, chemically altered by other soil microbes, or (1981), Delwiche (1981), and Sd6derlund and Svensson
held by the soil on the anion exchange capacity. Be- (1976). An example of how this type of information
cause the anion exchange capacity of many forest soils can be applied in forest management is given in Heal
is low, much of the nitrate that is not altered or ab- etrala (1932):
sorbed by living organisms enters the geochemical
cycle; it moves with soil water into other terrestrial or
aquatic ecosystems or back to the atmosphere as a gas. 5.8 Biogeochemical Efficiency
Inorganic nitrogen in soil or water may be subject of Forest Ecosystems
to the process of denitrification, in which NO; is re-
duced to NO,, nitric oxide (NO), nitrous oxide The combined action of the biogeochemical and bio-
(N,O), N>, or NH;. Denitrification to N; can occur chemical cycles in undisturbed forest ecosystems re-
either abiotically or by the action of bacteria, which sults in active accumulation and retention of nutrients
use oxidized nitrogen as a source of oxygen; the latter from the geochemical cycle. Many of the nutrients lost
 114 CHAPTER 5 Biogeochemistry

from forest ecosystems in streamwater come from soil on areas that vary in the inherent fertility of the soil.
below the main rooting zone or enter the stream di- Part of this seeming independence of a mature forest
rectly as throughfall and litterfall from overhanging from the nutritional status of the underlying mineral
streamside vegetation. The organic forest floor that substrate arises from the tendency for forests to build
develops from decomposing litter promotes the chem- up a forest floor, accumulate a capital of nutrients, and
ical retention of nutrients, and the combined action of then operate largely from the nutrients within the bio-
mycorrhizal roots and fungi provides an efficient bio- geochemical cycle. In managing seemingly productive
logical mechanism for nutrient uptake and retention. mature forests on inherently infertile sites, it is vital to
Characteristically, there is a concentration of fine recognize the underlying biogeochemical character of
roots at or near the surface of the soil that are efficient the ecosystem so that the biological mechanisms that
at absorbing nutrients in throughfall and nutrients re- sustain the site productivity are not impaired.
leased from decomposing litter. It is not surprising, The efficiency of the forest in retaining nutrients
therefore, that forest streamwater typically has very can be demonstrated by examining the chemistry of
low levels of dissolved chemicals, a condition referred water as it enters, at various stages as it passes through,
to as oligotrophic (as opposed to a eutrophic, or nutrient- and as it leaves a forest ecosystem. The result is a water
rich, condition). Plants that grow under conditions of chemistry profile (Figure 5-11). Commencing with low
low nutrient availability have evolved mechanisms for levels of chemicals dissolved in rain and snow, water
nutrient conservation, such as long foliage retention, becomes enriched as it leaches the tree crowns. Fur-
leaching-resistant cuticles, chemical defenses against ther increases occur as the water leaches the fresh litter
losses to herbivores, infrequent reproduction, and effi- on the forest floor. Chemical concentrations are then
cient internal recycling. A summary of these strategies reduced as the water percolates down through the for-
can be found in Chapin (1980) and in references in est floor and mineral soil. The water that leaves the
Smith and Hinkley (1995a,b). ecosystem as streamwater or groundwater may be in
The biogeochemical efficiency of forests is one of much the same chemical condition as when it entered
the main reasons that forests are able to grow on soils as precipitation, with the exception of elements such as
of extremely low fertility or on soils that are simply a silicon, which are scarce in precipitation but can be
thin layer of forest floor over unweathered bedrock. It abundant in streamwater as the result of rock weather-
is frequently observed that mature forests of remark- ing. The chemical composition of streamwater is cer-
ably similar composition and productivity can develop tainly affected by the chemistry of the soil parent

Nitrate, ppm Potassium, ppm Calcium, ppm

>
l 3 l 2 3 2 4 6
a 1 =i

Precipitation

Throughfall

Surface runoff

Forest floor leachate

Mineral soil leachate

Ground water
‘.-*,- Mineral soil
KD 5
Stream water

Te: “ raat : ‘ ; ,
Figure ‘-11_ Water chemistry profiles for a mixed conifer forest. Concentrations of nitrate,
potassium, and calcium ions in water are shown at seven stages of its passage through a 90-year-old
west coast conifer stand in British Columbia. The shape of the profiles reflects the relative efficiency
of the ecosystem at retaining these three ions within the biogeochemical cycle (high for NO, and
K*, moderate for Ca**). (Data from Kimmins and Feller, 1976.)
 Nutrient Cycling in Tropical Forest Ecosystems — 115

material and the rocks over which it passes in reaching animals. The forester must know the nutritional re-
the stream. In areas rich in calcium, magnesium, or quirements of the species that are being managed for
other elements, the streamwater will be enriched in and the mechanisms by which they are satisfied. Ex-
the corresponding chemical. Stream chemistry will amples of the types of information required are shown
also reflect regional precipitation, dry areas generally in Figure 5—12 for a conifer stand in Finland. We shall
having higher concentrations of dissolved chemicals in return to this topic later in the book to explain how
streamwater than wet areas. However, in many of the this information can be used as an aid to forest man-
world’s forests, forest streamwater is characterized by agement (see FORECAST in Chapter 21).
very low levels of dissolved nutrients because of both
the efficiency with which forests retain nutrients and
within-stream nutrient cycling processes. 5.9 Nutrient Cycling in
The biogeochemical efficiency of forests poses a ‘Tropical Forest Ecosystems
potential solution to one of today’s chronic environ-
mental problems—what to do with polluted water, Much of the early research on forest biogeochemistry
and, in particular, what to do with human sewage and was conducted in temperate and northern latitude
other “waste” biosolids. The water pollution that oc- forests, and until relatively recently, there has been a
curs when urban and industrial sewage is put into general scarcity of information on nutrient cycling at
rivers or lakes can be avoided by applying the enriched tropical latitudes. This situation has changed. Some of
water effluent to the forest floor. Although this has not the earliest work on tropical forest biogeochemistry
yet been done on a scale suitable for the world’s larger was done in West Africa (Greenland and Kowal, 1960;
cities, several of which have populations between 15 Nye, 1961). Extensive studies were undertaken in the
and 20 million, there are now many examples from 1970s in the tropical rain forest at El Verde, Puerto
small- and medium-sized cities that have demonstrat- Rico (Odum and Pigeon, 1970), in eastern Panama
ed the potential of this approach (Sopper, 1975). The (Golley et al., 1975), in Costa Rica (Gessel et al., 1978;
forest ecosystem extracts most of the chemicals from Johnson et al., 1977, 1979), in Brazil (Herrera et al.,
the effluent, neutralizes pathogenic organisms, and 1978), and in other Central and South American
delivers relatively pure water to the streams, while the countries (see the references in Herrera et al., 1978).
trees benefit by the improved water and nutrient avail- General reviews of tropical ecology can be found in
ability. This method of cleaning up polluted water is Farnworth and Golley (1973), Golley and Medina
not without its biogeochemical problems. For in- (1975), UNESCO (1978), Whitmore (1984), and
stance, litter decomposition can be accelerated to the Goldammer (1992). Nutrient cycling in the tropics is
point of eliminating the forest floor if the sewage solu- reviewed by Cole and Johnson (1980), Jordan (1985),
tion is applied in too large a quantity. If the concentra- Vitousek and Sanford (1987), Proctor (1989, 1995),
tion of heavy metals in the effluent or biosolids is too and Miller and Adams (1992). Nitrogen cycling in
high, then they can accumulate to levels that can influ- Southeast Asian wet tropical ecosystems is discussed in
ence soil biological processes. However, if done with Wetselaar et al. (1981).
an understanding of these potential problems, applica- Great concern has been expressed by many people
tion of sewage effluent and biosolids to forests offers about the exploitation of tropical rain forests for tim-
the possibility of solving a serious environmental ber and the clearing of forest for shifting cultivation or
problem while improving forest growth (e.g., Cole et permanent agriculture (Fearnside, 1995; Johnson and
al., 1986; Dutch and Walstenholme, 1994; Hall, 1991; Gabarle, 1993; Myers, 1982; Repetto and Gillis, 1988;
Harrison et al., 1994). Urban solid waste can also be Schneider, 1989; Skole and Tucker, 1993). The public
applied to forest areas with positive effects on forest has pictured tropical rain forest as a biologically diverse
growth (Smith et al., 1979). and very productive ecosystem characterized by a large
To make management decisions that are biogeo- tree biomass, nutrient-poor soils, rapid circulation of
chemically rational, the forester must have a good nutrients, and the majority of the nutrients on the site
working knowledge of the biogeochemistry of a forest. contained in the vegetation. On the basis of this con-
He or she must understand geochemical inputs and cept of tropical forests, it has been suggested that re-
outputs to and from the forest; the quantities, distri- moval of most of the vegetation will remove most of
bution, and cycling of nutrients within the forest; and the site’s nutrient capital. Like most ecological general-
the internal economy of nutrients inside plants and izations, this concept is valid only for certain types of
116 CHAPTERS Biogeochemistry

NUTRIENT CYCLING, kg ha” yr™

NUTRIENTS IN TREE BIOMASS, NUTRIENT IN UNDERSTORY

7 Mnput
kg ha j ee BIOMASS, kg ha
: ~~ Dust and precipitation, Total annual net 5 5490
T Biomass 95,149 / A Ss bfomuaetphornenan T ,Biomass
o| N 59
fo) N 186 a $4.5 cmyr' aS 8410 kg ha“! T Pp 7
atte. Se NEKO At ok 29
Ue Ca 4 \ SE jah Ket 22
Uptake >. ‘ _Intemal recycling Biomass of net production
ees n 2030 kg ha!
Retained in het tree \ \\
Foliage ey ? ; oe,

ees internal recycle

Se, the crown
p |
Branches 5 6380 AY 4.5 4sOF se
~’ Biomas of net sa S N
Output of — tine “> Ns oi - &

ie | quozne|f 1 \P29) “ieee

harvested Stemflow 0.9 7.4
4 d

| N KCa Il
uFThroughfall N P KCa
1)4.8 '\5.87-4
eee
| Understory 1655 N P K Ca Biomass
vegetation
Biomass
Stembark
Va
wi
ake and return )

ye,
15200,

Total soil reserves to a depth “Available”

of 30 cm, kg ha”! in the soil
Humus Mineral Input
soil Uptake by Output Soil weathering and lateral
1282 O the vegetation 62
10 Soil leaching water flow (no data)
25 18 40 19
0.2.2!
10 Return from 7
80 the vegetation N P K Ca

Figure 5-12 Summary of biogeochemistry of a pine stand in Finland (data from Malkonen,
1974) showing the distribution of four macronutrients in the biomass and soil and the dynam-
ics of nutrients in the biochemical, biogeochemical, and geochemical cycles. The dynamics of
nutrients within the understory plants is not shown. (Data used by permission of E. Malkonen.)

tropical forest and is certainly wrong for many forests tition. The area is then abandoned. Inputs of litterfall
at tropical latitudes (Brady, 1978; Lal and Sanchez, from adjacent forest and reinvasion of herbs and trees
1992; Whitmore, 1989). Many of the ideas about trop- (some of which are nitrogen fixers) gradually return
ical forests that appear in the media and many ecology the area to its original fertility, and if the interval be-
textbooks either are based on slender evidence (Proc- tween successive cut-and-burns is not too short, then
tor, 1995) or are simply inconsistent with our knowl- this type of land use can continue indefinitely without
edge of these ecosystems (Whitmore, 1989). degrading the productivity of the site (see further dis-
The indigenous agricultural traditions in the trop- cussion in the section “Effects of Forest Management
ics involve cutting down a small patch of forest, burn- on Forest Biogeochemistry”). However, because of
ing the slash after it has dried, and planting crops in population growth, the interval between successive
the ashes that enrich the soil with nutrients and raise cut-burn-harvest events has been rapidly decreasing,
the pH ofthe very acid forest soil. This produces one and in some areas, the soil has become so impover-
or two good food crops, but the yield declines rapidly ished that the original forest species can no longer
because of declines in soil fertility and pH and in- grow (Richards, 1973, 1977). There is disagreement as
creases in crop diseases, insect pests, and weed compe- to whether the yield decline is more closely related to
 Nutrient Cycling in Tropical Forest Ecosystems 117

nutrient depletion or to the buildup of pests and nutrients before they can be leached into the soil and
pathogens of the crop plants (Janzen, 1973), and the conducting them directly to the roots (Went and
idea that nutrient leaching is always severe in the trop- Stark, 1968). Radioisotopes have been used to demon-
ics has been challenged (Harcomb, 1977a, b). The role strate the high efficiency with which the mycorrhizal
of the forest fallow period may also extend beyond soil hyphae that cover the surface of the soil and root mat
fertility and weed and pest considerations. Mainte- are able to retain nutrients should they get into solu-
nance of soil structure by soil mesofauna that depend tion (Stark and Jordan, 1978). Almost none of the nu-
on forest litterfall may be essential if soil erosion dur- trients applied as an aqueous solution to the upper
ing heavy monsoon rainstorms is to be avoided. As is surface of this root mat were still in solution in water
usual in forest ecology, this issue is multiply deter- collected below the root mat.
mined and requires a careful assessment of ecosystem Third, nutrients are conserved within the trees by
function. Regardless of the outcome of this debate, efficient internal cycling before leaf shedding (Montes
there is still widespread concern about the effects of and Medina, 1977; Small, 1972) and by the reduction
short-rotation shifting agriculture on the soil and of herbivory through the accumulation of toxic sec-
about the intensive harvesting of tropical forests for ondary metabolic chemicals in leaves and roots. Annu-
timber. The long-term consequences of conversion al loss of leaf area to herbivore consumption is only 1
from forests of indigenous trees to short-rotation ex- to 2%, and the small size and low abundance of
otic species are not well known, but many ecologists arthropods in such forests reflects the high levels of
have expressed urgent concern about this trend (e.g., toxic alkaloids and polyphenols that are found in the
Cornforth, 1970; Richards, 1973). plants (Golley, 1977; Janzen, 1974).
Much of the argument about the effects of manip- Fourth, Plants adapt to very acid soils, which are
ulating tropical forest has lacked a firm basis in scien- very low in calcium and phosphorus, high in H* and
tific fact. Broad generalizations have frequently been Al** ions, and frequently waterlogged. Adaptations to
made, such as that all tropical soils are lateritic, subject these harsh, oligotrophic conditions include sclero-
to irreversible hardening on exposure. to sunlight, phyllous leaves (thick, leathery, and evergreen); toler-
and/or inherently nutrient poor. It is now recognized ance of very low nutrient levels; efficient internal
that the diversity of soils, vegetation, and biogeo- recycling of nitrogen, phosphorus, and potassium; and
chemical characteristics is as great in the tropics as at an evergreen habit. Calcium, which is one of the most
other latitudes and that much more sophistication is limiting nutrients in the Amazon region, is not re-
required in our approach to evaluating the biogeo- translocated but is efficiently recovered from decom-
chemical characteristics of tropical forests and the posing litter.
human impact thereon (Lal and Sanchez, 1992; Whit- Fifth, the morphology of the leaves results in their
more, 1989). arrangement on the forest floor after leaf fall so as to
Studies in the Amazon Basin in South America minimize their time of contact with throughfall, which
have provided some good insights into the adaptation could leach nutrients. The mineral soil tends to have
of tropical vegetation to very nutrient-poor soils. many large pores, which promote rapid drainage and
Forests that grow on white sand soils (which are virtu- minimize leaching (Nortcliff and Thornes, 1977).
ally devoid of nutrients) have evolved a number of Sixth, the multilayered structure of the rain forest
adaptations that cause them to act like “a giant ion- canopy, together with the presence of many epiphyl-
exchange column that extracts nutrients from water lous organisms (living nonparasitically on the leaves)
passing through the ecosystem” (Herrera et al., 1978). such as mosses, algae, lichens, and bacteria, promote
First, the forest develops a dense root mat on the uptake of nutrients from precipitation and throughfall.
soil surface. The rootlets do not have a well-developed The canopy scavenges nutrients from solution and
geotropic response, and they soon grow over and cover combines them into organic forms that become avail-
any litter that falls onto the root mat. Up to 60% of the able to the tree directly or via the litterfall pathway. As
tree biomass in some Amazonian forests is roots. This a result, concentrations of nutrients may be lower in
root mat has a very high nutrient retention capacity. throughfall than in precipitation. Many of the epiphyl-
Second, direct nutrient cycling occurs from the lit- lous organisms are capable of nitrogen fixation.
ter to the roots via mycorrhizae. The fungal hyphae Finally, the high concentrations of secondary plant
rapidly invade freshly fallen litter, absorbing inorganic chemicals in litterfall of plants that grow on tropical
118 CHAPTERS — Biogeochemistry

white sands impede decomposition, other than by my- devegetation is generally greater than at higher lati-
corrhizal fungi. The secondary chemicals and the high tudes, and most native humid tropical vegetation ex-
acidity of the litter restrict the activity of bacteria. hibits adaptations for nutrient conservation. This
This ensures that virtually all the nutrients released should warn us to be careful in our manipulation of
through decomposition return to the plants via myc- the vegetation of these areas and not to replace native
orrhizal fungi by the direct nutrient cycling pathway. vegetation with exotic species until we have demon-
It can also result in the buildup of a considerable forest strated the ability of our managed ecosystems to con-
floor, a situation that is not normally associated with serve the soil nutrient capital.
tropical forests. The decomposition pathways that Not all the tropics are uniformly humid. As one
occur in such forests produce dark-colored organic moves away from the equator, one first encounters sea-
acids as an end product. Under the prevailing acidic sonally dry forests (much of the Amazon basin experi-
conditions, these organic acids are mobile and leach ences a pronounced dry season) and then increasingly
into streams, staining the water brown. This gives rise arid woodlands. These have different soils and vegeta-
to the name black water rivers (e.g., the Rio Negro, tion and markedly different biogeochemical character-
which combines with the clearwater Solimoes River at istics. Consequently, experience of nutrient cycling in
Manaus to form the Amazon). Such rivers have very the humid tropics cannot be transferred to higher
low concentrations of dissolved inorganic nutrients, tropical latitudes without considerable modification,
reflecting both the shortage of nutrients in the adja- just as ecological wisdom gained at temperate latitudes
cent terrestrial ecosystems and the efficient retention is often an inadequate background for the wise man-
of nutrients in the terrestrial nutrient cycle. agement and conservation of the humid tropics.
Removal or destruction of the vegetation in such A comparison between oligotrophic and eutrophic
nutrient-deficient ecosystems will result in rapid loss forests in the tropics and at temperate latitudes (Jor-
of the nutrient capital and can convert tropical forest dan and Herrera, 1981) showed that the productivity
on this particular soil type to “wet deserts”—areas of and biomass of mature, undisturbed oligotrophic
white sand virtually devoid of nutrients and vegetation. forests can be similar to that of mature, undisturbed
Occasional groups of plants may survive on “islands” eutrophic forests, because adaptations for nutrient
of nutrient-containing organic matter. The efficiency conservation in oligotrophic forests greatly reduce the
of nutrient conservation and accumulation in such problems of nutrient scarcity in the substrate. The
forests depends on the living plants. Once these are problems come when forests are removed. In olig-
gone, so are the mechanisms of nutrient conservation. otrophic ecosystems, the nutrient capital is rapidly dis-
The scenario just described does not occur sipated, the soil having little capacity to retain the
throughout the humid tropics. It occurs where the nutrients. The rate of nutrient loss will be greatest in
soils are very old and are the result of several cycles of the tropics because of the more rapid decomposition
erosion and deposition that have left little more than and the greater risk of soil leaching losses. In eutroph-
silica sand and silt. Some of the moist tropical forest in ic ecosystems, plant-related nutrient conservation
some parts of the Amazon basin is growing on very mechanisms are relatively much less important, be-
nutrient-poor soils, and these forests exhibit the adap- cause the fertile soil is capable of retaining most of the
tations described above. Where the soils are younger, nutrients released after deforestation and has substan-
are fine textured, and/or receive additions of nutrients tial nutrient reserves. This is true for both tropical and
from other geochemical pathways, the ecosystem nu- temperate forests, so the difference between temper-
trient capital is much higher and there is far less risk of ate and tropical latitudes will be greatest on the olig-
nutrient exhaustion and ecosystem degradation. In otrophic sites.
areas with soils derived from recent volcanic materials, Herein lies the explanation for much of the dis-
in areas where nutritionally exhausted surface soils are agreement concerning the importance of nutrients in
removed by erosion to expose nutritionally richer forest growth. Both tropical and temperate forests can
lower layers, and/or where fertile alluvial materials are grow on both nutrient-rich and nutrient-poor soils.
deposited periodically, tropical soils can be as rich and Growth on the former depends mainly on the inher-
fertile as temperate soils. However, because of the al- ent fertility of the soil, and this is not usually reduced
most constant humid and warm conditions, the poten- catastrophically by harvesting. Growth on the latter,
tial for rapid decomposition and nutrient loss after however, depends on the plants themselves, and once
 Nutrient Cycling in Tropical Forest Ecosystems 1

plants and their biogeochemical effects are removed, and shorter. With less and less time to recover, the soil
the physical environment may not be capable of sus- fertility began to decline and the forest began to de-
taining plant production. grade. Degraded areas were invaded by clump bam-
In a review of nutrient cycling in tropical forests, boo (Gigantochla spp.), an early seral species of
Cole and Johnson (1980) concluded that because of disturbed sites. The growth form of the clump bam-
our limited data base and the high diversity of tropical boo is similar to that of a rhizomatous grass—an un-
forests, few generalizations can be made. However, it derground stem (rhizome), a massive fine-root system
does seem that, in general, leaf turnover (litterfall) in extending as much as 1 m into the soil, and multiple
lowland tropical forests is more rapid than in other shoots (called culms). These can achieve heights of 21
types of forest, decomposition is generally rapid, and m and diameters of 16 cm at 1.3 m above the ground,
mineral weathering and soil leaching rates are often but when the bamboo is cut down (4 to 5 years of age),
high. They concluded that for many tropical soils, nu- the resprouting culms are initially less than 50 cm tall
trient loss in harvested boles does not pose a serious (the “grass” stage).
threat to soil fertility but that burning could cause sig- Adapting to this degraded forest system, the Ja-
nificant losses of nitrogen. The presence of abundant vanese mountain farmers developed a 6-year cycle that
HCO, ions in tropical soil solutions creates a high sustains a food supply, a supply of firewood and build-
leaching potential, but leaching is often only moderate ing material, and soil fertility, with little or no external
in undisturbed forest because of a relatively low flux of inputs. Mature, 4- to 5-year-old bamboo clumps are
water through the soil Johnson et al., 1977) and the cut down. The litter layer of slowly decomposing
presence of a high anion exchange capacity, which leaves and stem sheaths is raked and burned, and the
limits the movement of phosphorus and sulfur (Borne- ash is saved for fertilizer. The litter layer in the bam-
misza and Llanos, 1967; Mekaru and Uehara, 1972). boo stands reduces evaporative losses, conserving soil
NO; ions may be more important than HCO;° ions moisture and protecting the soil from damage by
in the leaching of some tropical soils (Nye and Green- heavy rain and surface runoff, so it is not raked until
land, 1960). the bamboo is harvested. The harvested area is fenced
Harvesting tropical forests causes a short-term in- to keep out foraging animals and is hoed to a depth of
crease in available nutrients, but this is followed with- 25 cm. This kills the surface roots of the bamboo, pro-
in 3 years by a decline, presumably because of leaching viding an organic “fertilizer” to the soil and reducing
and erosion (Cornforth, 1970; Cunningham, 1963; the culm resprouting to small “grass” shoots that are
Nye and Greenland, 1964). Erosion losses can be se- easily weeded out. Food crops are planted and fertil-
vere on steep slopes, accounting for a 74% decrease in ized with the ash, animal manure (partly from branch-
the original site nutrient reserves. The extent of the es taken from the area and fed to livestock), and
postharvest nutrient loss will depend on the speed compost produced from plant waste from the field.
with which the forest reinvades the site. Because rein- Use of synthetic fertilizer has been introduced only
vasion is normally rapid and because many of the in- recently. Some of the food crops are legumes provid-
vading species are nitrogen fixers, forestry in the ing an input of symbiotically fixed nitrogen, and scat-
tropics is potentially a much less damaging land use tered leguminous trees are left in the field for nitrogen
than agriculture. For a discussion of the biogeochemi- fixation and ultimately to be used for timber.
cal effects of shifting cultivation in Central America, After a year of mixed-species food cropping, the
see Ewel et al. (1981). Evans (1992) and Miller and fertility of the surface soil declines and the farmers
Adams (1992) discuss management of tropical planta- switch to a somewhat less nutrient-demanding food
tions and native tropical forest, respectively. species: cassava. The cassava crop further depletes the
A fascinating example of the importance of under- soil fertility over the second year, and by now the new
standing nutrient cycling in the tropics is a study of bamboo culms are reaching several meters in height,
traditional agroforestry systems used in west Java. In shading the cassava. At this point, the crops are har-
this highly volcanic area, the tropical forest on the vested and the field is abandoned to bamboo and in-
steep mountain slopes was traditionally subject to vading rain forest trees. After 4 years of fallow, the
shifting cultivation, with a return interval of a century fertility and organic matter content of the surface soil
or more. As the population increased, the cycle of has recovered, a new litter layer has formed, and the
cut-burn—food crops—forest fallow became shorter cycle is repeated.
120 CHAPTERS Biogeochemistry

The sustainability of this bamboo talun-kebun sys- and outputs involved in agricultural and forest fertil-
tem reflects the traditional saying of the local farmers: ization and harvesting. Some are deliberate but harm-
“Without bamboo, the land dies” (Christanty, 1989; ful, such as the use of herbicides and insecticides in
Christanty et al., 1996a,b; Mailly et al, 1996). The ill-conceived eradication programs. There are also
bamboo, with its deep and prolific root system and many undesirable biochemical changes that are un-
slowly decomposing litter, rebuilds the surface SOM planned or unexpected consequences of deliberate
and returns leached cations from deeper in the soil to biogeochemical activity; acid drainage often results
the surface. Biological nitrogen fixation helps to re- from open-pit mining, lakes downwind of industrial
store site nitrogen supply. The mass of fine bamboo complexes are acidified, and lakes are over-enriched
roots killed by hoeing enriches the upper mineral soil by excessive use of agricultural fertilizers. We shall ex-
with organic matter, improving the soil’s ability to re- amine two examples.
sist the leaching of cations.
The sustainability of this agroforestry system is
now threatened because it is a subsistence system in a Alteration of the Role of Vegetation:
society that is evolving to a cash economy. Bamboo is A Devegetated Watershed at Hubbard
being replaced by fruit trees or other cash crops. With- Brook, New Hampshire
out the particular biogeochemical functions provided
by the bamboo, it is unlikely that this system can be We have seen that vegetation plays an important role
sustainable. Clearly, there is an urgent need to utilize in ecosystem biogeochemistry. Nutrients are accumu-
our knowledge of nutrient cycling in the tropical envi- lated and conserved by the vegetation, and the capacity
ronment to design a system that is both sustainable and of mature vegetation and soil to recover nutrients gen-
provides the economic benefits desired by the farmers. erally seems to match approximately the rate at which
they are released by decomposition. This results in
characteristically low levels of nutrients leaving the
ecosystem biogeochemical cycle dissolved in
5.10 Effects of Humans on streamwater. However, if either the rate of nutrient up-
Biogeochemistry and the take by vegetation or the rate of nutrient release by de-
Cycling of ‘Toxic Substances composition, or both, is altered, then a change in the
biogeochemical balance may occur. A good example of
In our discussion of geochemical and biogeochemical this is a study that examined the changes in the biogeo-
cycles, we saw that animals play a critical role in forest chemical balance of a mature northern hardwood for-
biogeochemistry. Herbivores accelerate the transfer of est when nutrient uptake by the vegetation was
chemicals from plants to the soil, the atmosphere, or eliminated through devegetation by tree felling and the
the hydrological cycle, and by killing vegetation they prevention of vegetation by applying herbicides. Nu-
can short-circuit important nutrient retention mecha- trient retention in the undisturbed forest resulted in
nisms. Animal migration can import or export nutri- low levels of dissolved chemicals in the stream draining
ents to or from an ecosystem, and the activity of soil the area despite the cycling of substantial quantities of
animals has an important effect on litter decomposi- nutrients in the forest (Table 5-17).
tion. Humans play a similar biogeochemical role, re- In the late autumn/early winter of 1965, all the
distributing nutrients by the harvesting (or mining), vegetation on a small, forested, first-order watershed
transportation, consumption, and waste disposal of was cut down, and all minor vegetation and regrowth
agricultural, forest, fish, and mineral products. of the hardwood stumps were eliminated the following
Any alteration of biogeochemical pathways in- spring and in subsequent summers by herbicide treat-
volves work. Our biogeochemical role is far greater ment. The result was dramatic (Figure 5-13). Previ-
than that of other animals primarily because of the ously constant low concentrations of cations in a very
enormous energy resources at our command. Some of small stream draining the treated area were increased
our biogeochemical alterations of the environment are in magnitude and variability, and the concentrations of
deliberate and generally beneficial (“good” being de- anions were either increased or decreased. Levels of
fined in terms of what society thinks it wants or nitrate increased from less than 0.5 ppm to as much as
needs), for example, the increased geological inputs 80 ppm (eight times greater than the permissible U.S.
 Effects ofHumans on Biogeochemistry and the Cycling of Toxic Substances Wa

Table 5-17 Nutrient Capital and Turnover, kg ha’, in an Undisturbed 55-Year-Old Forested Watershed at Hubbard
Brook, N.H.*

Nutrient

Component N P K Ca Mg S

Nutrient capital
Aboveground plant biomass 351 35 155 383 36 42
Belowground plant biomass 181 53 63 101 13 17
Forest floor 1256 78 66 372 38 124
Annual inputs
Precipitation 6.5 Trace 0.9 2.2 0.6 Ws
Aerosol/gaseous 14.2 — - _ - 6.1
Weathering 0 ? Hol 21 3:5 0.8
Annual outputs, streamwater
Dissolved 3.9 Trace 1.9 13.7 Sul 17.6
Particulate 0.1 Trace 0.5 0.2 0.2 <0.1
Annual turnover
Vegetation uptake 79.6 8.9 64.3 62.2 9.3 24.5
Aboveground litterfall 54.2 4.0 18.3 40.7 5.9 5.8
Root litterfall 6.2 alle eal 32 0.5 0.6
Throughfall and stemflow 9.3 0.7 30:1 6.7 2.0 21.0
Root exudates 0.9 0.2 8.0 3:5 0.2 1.9
Net litter mineralization 69.6 t 20.1 42.4 6.1 Lay//
Annual accretion
Aboveground plant biomass 4.8 0.9 4.3 5.4 0.4 0.8
Belowground plant biomass 4.2 1.4 d25 Pel 0.3 0.4
Forest floor Holl 0.5 0.3 1.4 0.2 0.8

Source: Likens et al., 1977. Copyright Springer-Verlag New York, Inc. Used with permission.
*A comparison of outputs with the nutrient capital and annual circulation shows that the ecosystem is efficient at retaining nutrients. However,
when the watershed was deforested, this ability was drastically reduced.

federal water pollution standards) and averaged 38 and cause of increased soil temperatures. These changes
53 ppm, respectively, during the 2 years after cutting. led to the production of large quantities of nitrate
The increase in dissolved nutrients, together with the ions, which, in the absence of plant uptake, were
increased light and summer water temperatures that leached out of the soil into the stream. For reasons to
accompanied the removal of the forest canopy, led to be discussed in Chapter 11, the presence of large
the prolific growth of algae in the stream. quantities of soluble anions in soil solution results in
The dramatic outcome of this devegetation exper- the leaching of cations that would otherwise remain in
iment is thought to have been the result of alterations the soil. The great increase in levels of cations in
in the pathway of nitrogen transformations in the lit- streamwater that resulted from devegetation in the
ter decomposition phase of the biogeochemical cycle Hubbard Brook study can therefore be explained in
(Likens et al., 1969). Mature forest vegetation appar- terms of the massive production of NO;.
ently inhibits the nitrification pathway, most soluble The findings of the initial Hubbard Brook study
nitrogen in the forest floor being in the form of am- cannot be applied uncritically to managed watersheds,
monium ions or amino acids. These are efficiently re- because the experimental treatment that was applied
tained by soil chemical mechanisms or plant uptake, was not representative of conventional forest manage-
whereas any nitrate that is produced is absorbed by the ment. Subsequent investigations of commercial har-
vegetation. Removal of all plants resulted in several vesting of yellow birch in the Hubbard Brook region
major changes. The inhibition of nitrification was ter- yielded less dramatic and less persistent but qualita-
minated, nutrient uptake by plants was terminated, tively similar results (Hornbeck et al., 1975). Other
and the rate of litter decomposition was increased be- studies showed that each of several different patterns
 122. CHAPTERS — Biogeochemistry

Period of Herbicide Nitrate (NO,) of harvesting had different effects on water chemistry
80.0 vegetation application (Martin and Pierce, 1979). Clearly, under some condi-
60.0
40.0
cutting
tions, some forest management practices do alter bio-
20.0 Treated watershed geochemical mechanisms in a way that leads to
significant changes in the geochemical balance of a
———— Control watershed :
{i
forest. Readers who are interested in the mobility of
bl
\
~"\
nitrate ions in disturbed forest ecosystems and in the
' ¢

\
Sulfate (SO,2-)
rua
regulation of the chemical composition of streams in
4 Y
mel
the northeastern United States should consult Vi-
tousek (1977, 1981), Vitousek and Melillo (1979), and
Vitousek et al. (1979, 1982).

Acid Rain and Its Effects on

Forests and Lakes
The phenomenon of acid rain has been alluded to sev-
L!
concentrations,
Streamwater
mg Calcium (Ca?*)
eral times. Described as Canada’s premier pollution
problem in the late 1970s and the 1980s, acid rain is
no newcomer to many industrialized areas of Europe,
and the problem gained public attention in parts of the
United States. The early evidence of the acidification
of the atmosphere included the absence of lichens on
trees and gravestones in industrial England and Eu-
rope and the crumbling of stone statues and cornices
of buildings that had remained almost unchanged for
J JASONDJFMAMJ JASONDJFMAMJ JASONDJ FMAM centuries before atmospheric acidification. We now
1965 1966 1967 1968 know of a more insidious aspect of acid rain: its effects
(a)
on rivers, lakes, and forests.
The origins of acid rain lie in the accelerated re-
lease of sulfur dioxide and oxides of nitrogen to the air
by the combustion of fossil fuels and other industrial
materials. This greatly increases the atmospheric
loadings of these gases, which in due course combine
with water and are converted to sulfuric and_ nitric
acid. These strong acids are highly ionized, and conse-
quently the concentration of hydrogen ions in the at-
mosphere is greatly increased. Acidity is a measure of
the concentration of hydrogen ions, so these processes
lead to an increase in atmospheric acidity. This acidity is
deposited in ecosystems by rain, snow, fog, or dry fallout
Figure \-13 (A) Effect of the devegetation of a northern
(dust); the last may contribute up to 50% of the deposi-
hardwood watershed in the mountains of Vermont, north-
tion of acidity. Acid rain can also result from releases of
eastern United States, on the chemistry of streamwater. All
vegetation was cut and left in place at the end of 1965, and an
chlorine and hydrocarbons into the atmosphere.
herbicide was applied the following spring to prevent regrowth
Many ecosystems (especially forests) are chemi-
(after Likens et al., 1970). (B) Winter aerial photograph of cally buffered: their chemical systems are able to resist
two experimentally logged watersheds; the farthest one is change in the concentrations of various ions in solu-
where the above data were obtained. (A reproduced by per- tion, and consequently input of hydrogen ions may
mission of the Ecological Society of America and G.E. Likens; not lead to much change in their pH (the measure of
B courtesy of J.W. Hornbeck.) acidity). Acid rain may have relatively little effect on
 Effects of Humans on Biogeochemistry and The Cycling of Toxic Substances 123

such systems, at least in the short run, but the effects has also been shown to affect some soil animals by re-
on poorly buffered ecosystems can be dramatic. Many ducing their food supply (bacteria) and shelter (moss-
oligotrophic lakes and streams in areas underlain by es) (Stachurska-Hagen, 1980). In the long run, this
carbonate-poor granitic rocks are very poorly could have an adverse effect on soil processes, soil fer-
buffered, and inputs of acid rain result in biologically tility, and forest growth. In particular, the accumulated
significant reductions in pH (increases in acidity). effects of acid precipitation may exacerbate leaching
Such aquatic ecosystems are normally near neutral in losses at the time of forest harvest (Tamm, 1976a).
reaction, and the fish and other aquatic organisms na- The effects of acid rain on forest soils may in some
tive to such lakes and streams are sensitive to acidity, cases be masked by the normal changes in soil pH that
especially in their reproductive stages. On average, accompany changes in vegetation over time. Krug and
oligotrophic lakes in areas with acid precipitation have Frink (1983) caution against ascribing declining soil
experienced a pH drop of 2 units during the past 40 pH values to the effects of acid rain unless the effects
years. Because the pH scale is logarithmic, this means of past land use practices, secondary succession
that the acidity in these lakes has increased 100 times. (Chapter 16), and stand development on soil pH have
The concern over acid precipitation was stimulated by been accounted for. A very useful review of the effects
the realization that fish have been eliminated from of acid rain on forest soils and the types of soils on
many oligotrophic mountain lakes in Scandinavia, and which serious effects might be expected is given by
it is now recognized that damage to lakes in eastern Johnson et al. (1982).
Canada and the northeastern United States from the Investigations of dieback of spruce in the north-
effects of acid rain has occurred on a large scale. eastern United States and eastern Canada (see ‘lom-
Paralleling the concern over the effects of acid rain linson, 1981), as well as studies in Germany (Ulrich et
on aquatic ecosystems is concern over its effects on al., 1980), have suggested that on soils that have a low
forests. A considerable amount of research into these capital of calcium, the deposition of acid lowers pH
effects has been conducted in Scandinavia (Braekke, and results in the mobilization of aluminum, which
1976; Jonsson and Sunberg, 1972), but the early re- then becomes toxic to fine roots. The loss of fine roots
sults were contradictory, and by the end of the 1970s renders the trees more susceptible to drought and re-
there was no strong evidence of regional reductions in duces their ability to absorb nutrients, which can lead
tree diameter growth in Norway that could be un- to reductions in growth or even dieback. These effects
equivocally attributed to acid rain (Strand, 1980). will occur most strongly on soils derived from calci-
There is no doubt that acidic precipitation has a direct um-poor parent materials (granitic rocks) and will not
adverse effect on vegetation, including damage to the be so prevalent on base-rich soils. In addition to soil
cuticle, interference with guard cells, disturbance of effects, there are thought to be important direct ad-
metabolism and poisoning of cells, interference with verse effects of SO, and other air pollutants on plant
reproduction, accelerated foliar leaching, alteration of foliage (Knabe, 1976; Materna, 1979). These may in-
mycorrhizal and nitrogen-fixing associations, alter- volve direct effects on leaf physiology and photosyn-
ation of host—parasite relations, and increases in sus- thesis (with accompanying fine-root death) or may
ceptibility to other stresses (Iamm and Cowling, involve accelerated leaching of nutrients from foliage.
1976). However, much of the acidity in the rain is as- We obviously do not yet know the full extent of the ef-
sociated with nitrogen and sulfur, both of which can fects of acid precipitation on forests or the exact
be in limiting supplies in forest ecosystems. Adverse mechanisms of damage, and it would be premature to
effects of the acid rain may therefore be balanced by draw firm conclusions at this time. However, as in
beneficial, nutritional effects, at least initially. most ecological phenomena, it is probably safe to as-
The organic forest floor is a very well buffered sume that a variety of mechanisms are involved.
chemical system, which is already acidic, and it seems The literature on acid rain is extensive. Interested
that it may take many decades of acid rain to induce a readers are referred to Dochinger and Seliga (1976),
major reduction in soil pH. However, there is evidence Tollan (1978), Drablos and Tollan (1980), and an ex-
that acid rain may be accelerating the leaching of cal- tensive series of articles about the effects of acid rain
cium and other cations out of forest floors and de- on forests and fish published by the SNSF project in
creasing the base saturation (see Chapter 10). Acid rain Norway (SNSF Project, Box 61, 1432 Aas-NLH,
124 CHAPTER 5 Biogeochemistry

Norway). Other useful references are Swedish Min- the remaining forest area that is used to supply wood
istry of Agriculture (1982), Livingston (1982), Bangay products. The increasing demand for forest products
and Riordin (1983), Morrison, (1984), Smith et al. can be supplied from a smaller land base by increasing
(1985), Innes (1993), and Freedman (1995). the intensity of production and harvest on the remain-
ing “working forest” (Binkley, 1997; Sedjo and Botkin,
1997). A trend away from conventional stem harvest-
5.11 Effects of Forest Management on ing with long rotations and toward shorter rotations
Forest Biogeochemistry and whole-tree harvesting (removal of all aboveground
tree biomass from the site) developed in the 1970s and
Of all the environmental factors that determine ener- 1980s. Some countries (e.g., Sweden) went the final
gy flow and storage in forest ecosystems, the supply step on some sites and even harvested tree stumps and
and circulation of nutrient elements is one of the most root systems in what has been called complete-tree har-
easily manipulated by the forest manager. Because of vesting (Young, 1968). Stumps and large roots have
this and because of the relationship between nutrition also been harvested in some parts of the southern pine
and productivity, it is important for the forester to un- region of the United States, together with all above-
derstand how forest management can affect forest bio- ground biomass (Koch, 1978, 1980). Experiments
geochemistry. The effects may be either undesirable were conducted on harvesting foliage for processing
(nutrient loss or immobilization) or desirable (nutrient into cattle feed or adhesives. Although some might ap-
additions by fertilization or improvements in rates of plaud the apparent thrift of such developments, it is
cycling and availability). The discussion focuses on necessary to consider the full ecological effects of this
two aspects of forest management: clearcut harvesting type of forest harvesting, including the effects on nu-
and postharvest slashburning. trient withdrawals.
Concern about harvest-induced site nutrient deple-
tion was initiated in 1876 by Ebermayer in a study of
Effects of Clearcutting the effects of litter raking on site productivity. This
Losses in Harvested Materials. ‘Irees contain nu- early concern was reinforced by Weidemann (1935,
trients; consequently, the removal of tree parts during 1951) and by Rennie (1955, 1957), who questioned
a harvest results in a loss of nutrients from the site. whether it was reasonable to expect sustained intensive
The extent of the loss depends on the utilization in- production to continue indefinitely in conifer stands
tensity. The so-called “wasteful” harvesting that char- established on infertile soils. The development of con-
acterized the early days of logging in western North cern over the nutritional aspects of intensive plantation
America utilized only large, sound logs, leaving much management in the 1950s was reflected in the work of
or even most of the forest biomass on the site. This re- Ovington (1957, 1959), who documented the biomass
sulted in only small withdrawals of nutrients because it and nutrient content of an age sequence of pine stands
harvested mainly heartwood, which generally contains developing on sandy soils in eastern England and un-
low concentrations of nutrients. These “bad old days” dertook nutrient-cycling studies in a wide variety of
have largely been replaced by a “less wasteful” type of forest types in Great Britain (Ovington, 1961, 1962).
harvesting that takes a much higher proportion of the However, it was not until the 1970s that the question
tree stems. This is accompanied by an inevitable in- of nutrient withdrawals in harvested materials received
crease in nutrient withdrawals. widespread attention by forest scientists.
The predicted growth in the human population by During the past two decades, the effect of switch-
50% or more during the next century and the associat- ing from conventional to whole-tree harvesting on the
ed growth in demand for wood and other forest prod- extent of nutrient withdrawals has been documented
ucts will not be matched by a similar growth in forest for a wide variety of forest types (Table 5-18). How-
area. On the contrary, many countries will see a reduc- ever, the consequences of these withdrawals for future
tion in their area of forests during this period, and in- tree growth are not easy to predict with confidence.
creasing competition between timber production They vary for different nutrients, with different
forests and management of forests for other land uses species, with stand density, and with stand age because
will see a continuing reduction in the proportion of these parameters determine the relative proportions of
 Effects ofForest Management on Forest Biogeochemistry 125

Table 5-18 Percentage Increase in Removal of Certain Plant Nutrients in Harvested Materials Accompanying the
Switch from Conventional to Whole-Tree Harvesting?

Increase, %

Aboveground
Forest Type Biomass N P K Ca Reference

Hemlock-cedar, <500 yr old 43 165 Alt at 95 Kimmins and Krumlik, 1976

Lodgepole pine, 125 yr old 15 ey6} 54 14 15 Kimmins and Krumlik, 1976
Spruce-fir, <350 yr old 25 116 163 32 50 Kimmins and Krumlik, 1976
Hemlock-fir, <550 yr old 20 86 67 48 48 Kimmins and Krumlik, 1976
Black spruce, 65 yr old 99 288 367 236 179 Weetman and Webber, 1972
Cottonwood
7 yr old 25 27, — - = Carter and White, 1971
9 yr old _ 116 100 74 68 White, 1974
Jack pine, 65 yr old
Whole-tree harvest 26 120 250 80 46 Morrison and Foster, 1979
Complete-tree harvest Oe 149 325 110 78 Morrison and Foster, 1979
Cryptomeria japonica, 25 yr old 50 360 315 270 196 Tsutsumi, 1971
Loblolly pine
4 yr old = 289 483 231 267 Haines and Sanderford, 1976
16 yr old 29 35) 107 85 — Jorgensen et al., 1976
Douglas-fir, 15-20 yr old Si 290 298 265 299 Webber, 1977
Mixed hardwoods, 45-50 yr 39 2 77 78 83 Boyle and Ek, 1972
Boreal forest average 65 Ws _ _— 180 Marion, 1979
Temperate coniferous average 28 100 _- _ 150 Marion, 1979
Temperate broadleaf average 60 PS) ~ _ 110 Marion, 1979

*Additional data of this type can be easily calculated from Freedman (1980) and Kimmins et al. (1985).

stem biomass to crown biomass and therefore the dif- and 173%, respectively. Obviously, fiber farming (full
ference in nutrient withdrawals between conventional utilization with short rotations; Young, 1972) will have
and whole-tree harvesting. The effect of stand age is a much greater effect on site nutrient capital than did
demonstrated in Table 5-19. In younger stands, the the “wasteful” logging of the past, which left most of
ratio of sapwood to heartwood is higher than in older the nutrients on the site where they could eventually
stands, as is the ratio of crown biomass to stem bio- be used by the next crop.
mass. Sapwood generally has higher nutrient concen- It is much easier to evaluate the degree to which
trations than heartwood; crown materials have higher harvest-induced nutrient depletion occurs as the in-
nutrient concentrations than stems. Shortening rota- tensity of utilization increases than to determine how
tions reduces the age at harvest and therefore com- important such increased losses are in terms of re-
pounds the effects of increasing utilization standards duced future wood production (Kimmins, 1977). The
by increasing the proportion of stems that is sapwood latter involves a variety of questions, many of which
and the proportion of crown material in the harvested may be difficult to answer:
biomass. For example, Boyle (1975) estimated that the
change from a single 30-year rotation to three 10-year 1. What proportion of the site nutrient capital is re-
rotations in aspen stands (all with whole-tree harvest- moved in harvested material?
ing) would increase the nutrient removal of nitrogen, 2. What is the magnitude of other harvest-induced
phosphorus, potassium, and calcium by 345, 239, 234, losses, such as soil leaching?
126 CHAPTERS — Biogeochemistry
n from Conventional
Table 5-19 Effect of Age of Stand on Percentage Increase in Loss of Nutrients after Conversio
to Whole-Tree Harvesting
Percentage Increase in Loss of
Age
(Year) N P K Ca Reference
Species ee
eS
Norway spruce 18 195 233 161 206 Tamm, 1969
50° 114 115 26 40 Tamm, 1969
85° 91 104 42 29 Tamm, 1969

Scots pine 18 188 Zale 171 129 Wright and Will, 1958
28 130 149 97 83 Wright and Will, 1958
oS 172 150 102 69 Ovington and Madgwick, 1959
39 164 200 140 88 Tamm, 1969
44 124 133 108 84 Tamm, 1969
64 103 114 94 4 Wright and Will, 1958
15 77 67 56 59 Tamm, 1969

Corsican pine 18 194 229 239 183 Wright and Will, 1958
28 122 155 133 150 Wright and Will, 1958
48 89 107 95 97 Wright and Will, 1958
Averages
Pines 50 - 156 104 100 Rennie, 1955
100 — 87 59 52 Rennie, 1955
Other conifers 50 - 170 127 138 Rennie, 1955
100 _ 87 56 59 Rennie, 1955
Hardwoods 50 _ 122 92 67 Rennie, 1955
100 — 69 47 37 Rennie, 1955

“Average of two stands.

3. How frequently will harvest-induced losses occur; sive biomass harvesting. A simple graphical analysis
what is the rotation length? (Figure 5-14) can show in general terms what may
4. How rapidly does the remaining site nutrient capi- occur, but it is too simplistic to give much confidence
tal cycle? That is, how “available” are the remain- in the predictions. The solution is to develop a com-
ing nutrients to the next crop? puter simulation model that can handle all that we
know about the biogeochemistry of intensive harvest-
5. How rapidly are the losses replaced by natural
ing (see Chapter 21).
processes, and what are the processes? How are
The effects of whole-tree harvesting are not re-
they affected by harvesting and stand treatments?
stricted to nutritional effects. The suppression of
6. What are the nutrient requirements of the next competing vegetation and the shelter afforded to tree
crop? How do these requirements vary during the seedlings by logging slash improves their early
life of the crop? growth, as does the improved soil moisture conserva-
7. How important is availability of nutrients in regu- tion provided by the mulch of slash. The longer-term
lating production of the crop species? effects are probably mainly nutritional (Proe and
8. How easily (economically and ecologically) are the Dutch, 1994).
harvest-induced losses replaced by fertilization or Before we leave this topic, it should be emphasized
some other means? that most investigations have concluded that medium
to long (80 to 120 years) rotating harvesting of tem-
In most cases, we do not know the answers to these perate forests in which only stems are removed poses
and other questions, but even if we did, the diversity of little threat of site nutrient depletion. When estab-
information involved makes it difficult to arrive at a lished on degraded soils (e.g., abandoned agricultural
conclusion concerning the long-term effects of inten- land), soil fertility and site productivity may increase
 Effects ofForest Management on Forest Biogeochemistry 127

concerns about the nutritional consequences of har-

. ee
. tee,
PP ee eee eee Peter ere ee ee er ress
Long
vesting are increasingly restricted to the intensively
— Free,
rotation managed forests that in the future may supply much of
-— fee es
250
EOOSCON AR
emcees “ the world’s demand for wood products.

Nutrient loss at harvest rotation Increased Availability of Nutrients: The Assart

Effect. Clearcutting may accelerate the mineral-
ization of forest floor materials present at the time
of harvesting, and it adds a variable quantity of log-
: ease ee eeescescocecccves
ging slash to the forest floor. The green foliage, fine
te eee Stem twigs, and dead fine roots added as slash contain
harvestin,
os A
PCOS AG Cooney
substantial quantities of readily decomposable tis-
sues that are relatively rich in nutrients, and there is
i pS

Ie eer id | eam Whole-tre

harvesting commonly a “flush” of available nutrients from this
material that starts within a year or two of logging:
the assart effect. The increase in nutrient availability
(b) also reflects reduced nutrient uptake by plants, and
Magnitude
available
the
of
capital
nutrient
site an assart effect can occur even if mineralization rates
ory See Serer rere Corer rrr ere (Pere rriry Serre erry are not increased. The reduction in competition for
Rapid
BE
4 —-4+s
| CG) nutrient nutrients between tree roots and microbes and the
ae
Pee ecees improved soil moisture status in the absence of
water loss by transpiration will generally increase
nutrient microbial activity and the soil fauna that feed on
replaceme
bacteria, so accelerated decomposition is common
after clearcutting. During the initial postharvest pe-
Time
riod, most of the nitrogen released by decomposi-
(c)
tion may be taken up and immobilized by microbes,
Figure 5-14 Simple graphical analysis of the relationship and this may delay the assart flush for a few years. As
between the available nutrient capital of a site and the ro- the readily available carbon in the forest floor and
tation length (A), intensity of utilization (B), and rates of slash is depleted by the microbes, their populations
nutrient replacement (i.e., inputs; C). The graphs are hypo- decline and the nitrogen that they sequestered is re-
thetical because no long-term data of this type are available. leased. In such cases, the assart flush indicates the
The dotted lines indicate the expected long-term trends. Pre- depletion of easily available carbon and decline in
dictions of this type are produced by some of the computer microbial activity. The flush continues until the
simulation models discussed in Chapter 21. readily decomposed organic matter has been
processed into stable microbial biomass, mineral-
ized, or converted into relatively stable humus.
This fertilizing, or assart,* effect of clearcutting,
over successive rotations (Evans 1990; Seely et al., which continues until the nutrients have been taken
1999). It is short rotations combined with intensive up by plants, immobilized by microbes or by soil
biomass utilization that may create problems of re- chemical reactions, leached away, or lost as gases, has
duced soil fertility. This topic received considerable been known for centuries. For millennia, peasant
attention in the 1980s and 1990s, as exemplified by farmers around the world have cleared forest and
several symposia devoted to the question of long-term planted crops that thrived on the accompanying flush
sustained forest productivity under intensive forest of nutrients. As the availability of nutrients and the
management (Ballard et al., 1983; Dyck et al., 1994;
Perry et al., 1989). With concern about the sustain-
ability of non-timber forest values leading to longer ‘An old English word meaning to grub up the trees and shrubs of aforest or
rotations and increasing recognition of the dangers of woodland to make the land arable, this term has been used to refer to the

whole-tree harvesting on nutritionally marginal sites, post-clearcutting flush ofnutrients (Tamm, 1964).
128 CHAPTERS Biogeochemistry

productivity of the crops declined, the land was aban- ability of nutrients. In the presence of nutrient-de-
doned and the forest was allowed to reinvade and re- manding plants, most of the nutrient release is taken
store the diminished fertility. up and retained. However, if the logging has left the
A good example of the assart effect was provided area free of vegetation and if for some reason reinva-
by Weetman (1967). He cut a small area of a 90-year- sion is delayed and the soil is unable to retain them,
old black spruce stand in Quebec and established on then the soluble nutrients released by mineralization
half of it a plantation of black spruce seedlings collect- of organic matter may be leached away or lost as gas to
ed from the vicinity. On the other half, he allowed nat- the atmosphere (nitrogen). A high cation exchange ca-
ural regeneration to stock the area, a process that took pacity will generally ensure the retention of cations,
about 8 years. Eighteen years later, the spruce planted except where there is surface runoff or channeling of
immediately after cutting, which had the advantage of soil water through soil macropores or channels. How-
the assart effect, were approximately 5 m tall. The ever, anions such as nitrate are poorly retained by
trees in the other area, which were more than half the many soils and are readily leached.
age, were only approximately 50 cm, or one tenth, as The most widely published example of the effect
tall. Forestry practice often calls for artificial regener- of vegetation removal on nutrient leaching is the ex-
ation only after a period in which natural regeneration periment conducted in a climax eastern deciduous
is given an opportunity to restock the area. In many hardwood forest in New Hampshire (Hubbard
cases, such a delay will mean that the next crop does Brook). As we have already seen (in the section “Acid
not get the benefit of the flush of nutrients. Rain and Its Effects on Forests and Lakes”), the result
The assart effect can be responsible for fooling was a dramatic and substantial alteration in water
foresters about the quality of a site after clearcutting. chemistry and nutrient budget for the watershed
Because nutrient availability is unusually high for a (Table 5-17). Figure 5-13 shows the changes in the
few years after clearcutting, nutrient-demanding concentrations of various chemicals, and ‘Table 5—20
species may initially grow well even on poor sites. presents annual stream nutrient budget data. During a
However, once the assart effect is over, tree growth 3-year period of deforestation, the total net loss of dis-
on such sites may decline drastically. In fact, once the
flush of nutrients is over, nutrient availability and
growth may actually decline to below the prelogging Table 5-20 Net Gains and Losses, kg ha", of Dissolved
levels. While the increased microbial population is Substances for the Devegetated Watershed and a
mineralizing the easily decomposable needles, twigs, Control Watershed at Hubbard Brook, N.H., for the
and fine roots, it is also invading the larger woody Water Years 1966 to 1969
materials, a process that is greatly aided when wood-
Net Gain or Loss?
boring insects are present because they introduce
fungi into the logs. Before the fungi can utilize the Deforested —CC Ctl
energy contained in the high-carbon : nitrogen-ratio Element Watershed? Watershed
woody material, nitrogen and other nutrients are
Nitrate N —114.1 +1.6
needed. These are believed to be withdrawn from the
Ca -77.7 ~9.0
surrounding area by fungi and translocated into and
immobilized in the large woody material, resulting in Silicate Si -30.6 -15.9

a deficiency of available nutrients elsewhere in the K -30.3 -1.5

soil. Tree growth under such immobilization condi- Al —21.1 -3.0
tions contrasts strongly with tree growth during the Mg -15.6 “2:6
assart period. There is some fixation of N, by mi-
Sulfate S —2.8 -30.6
crobes in decomposing logs, but the relative impor-
Cl —1.7 Eee
tance of fixation versus translocation in increasing the
nitrogen content of decomposing logs has yet to be NH,* +1.6 +2.2
established.
Source: Bormann and Likens, 1979. Copyright Springer Verlag New
York, Inc. Used with permission.
Leaching Losses Accompanying Clearcutting. *Meteorological input minus streamwater output.
The assart effect results from increases in the avail- ’Deforested November to December 1965.
 Effects of Forest Management on Forest Biogeochemistry 129

solved substances was increased approximately eight- logging assart nutrient flush. Fireweed accumulates
fold over the predeforestation value. Increases for in- biomass and nutrient capital rapidly, conserving some
dividual elements were as follows: potassium, 20-fold; of the nutrients that otherwise might be leached away
calcium, 8.6-fold; nitrogen, 160-fold (Bormann and and holding them on the site until trees shade out the
Likens, 1979). Because many ecosystems are nitrogen herb community and utilize the nutrients contained
limited, the last figure is particularly interesting. Sub- therein. Prompt reinvasion of any logged area will act
sequent studies comparing commercial clearcutting to reduce nutrient losses and conserve the site nutri-
with the devegetation experiment revealed less dra- ent capital, although it should be noted that in many
matic and persistent effects on stream nitrogen chem- cases the nutrient content of the pioneer vegetation
istry but similar short-term effects on the nitrogen may be much less than the quantity of nutrients re-
budget (Table 5-21). leased after logging because the biomass of this vege-
The major difference between the Hubbard tation is often small.
Brook devegetation experiment and commercial The results of the Hubbard Brook studies were
clearcutting is that concentrations of nutrients in interpreted by some as evidence that clearcutting
streamwater in clearcut areas rapidly decline to levels causes serious disruption of ecosystem function. This
that approach or even fall below those of the precut- stimulated studies in many other forest areas to as-
ting period. The explanation for this decline is to be certain whether the same thing is true in other forest
found in the rapid reestablishment of the plant com- regions. For example, in a series of watershed studies
munity after clearcutting. Logged areas in the north- in Oregon (Fredriksen et al., 1975), it was found
ern hardwood forest are rapidly invaded by various that small increases in nitrate concentrations in
pioneer shrubs (e.g., Rubus sp.) and deciduous hard- streamwater do occur after clearcutting, typically
woods (e.g., pin cherry [Prunus pensylvanica] and birch with a delay of approximately | year. This delay oc-
[Betula sp.]). These species accumulate biomass very curs because with the low summer rainfall that char-
rapidly, but even biomass accumulation is exceeded by acterizes much of the West Coast, products of
the accumulation of nutrients. Maximum nitrogen decomposition are not flushed out of the soil until
content is achieved within 6 years of clearcutting, midwinter, and because it usually takes at least one
whereas peak potassium levels are reached within 4 summer before appreciable mineralization can occur.
years (Figure 5-15). Many logged areas in western The extent of the increase varied in different parts of
North America are promptly invaded by fireweed Oregon, but peak levels were generally only a few
(Epilobium angustifolium). Vhis is a nutrient-demand- percent of the peak levels observed in New England.
ing species whose presence is indicative of the post- The increases declined rapidly as the watersheds
revegetated, the duration of the increase being relat-
ed to the speed of the revegetation.
The lower peak levels of stream nitrate observed in
Table 5-21 Loss of Calcium and Nitrogen, kg ha", Oregon have been attributed to a greater quantity of
from a Devegetated Watershed Compared with Losses high carbon : nitrogen ratio, slowly mineralizing ma-
from a Comparable Commercially Clearcut Watershed terial in the forest floor in Oregon than in New Eng-
in New Hampshire land, and to a delay in nitrification after disturbance
Losses from Losses from
because nitrifying organisms are less active in the
Devegetated Clearcut
lower-quality forest floor materials of Oregon (Vi-
Watershed Watershed tousek et al., 1982). There may also be a difference in
the within-stream nutrient immobilization between
N Ca N Ca Oregon streams and New England streams. Primary
production in many West Coast streams is limited by
Export in streamwater lack of nutrients (Haydu and Thut, 1971), and streams
1st yr 95 75 38 41
2nd yr 140 90 Sy7/ 48
are known to be active sinks for nitrate (Cummins,
1980; Perrin, 1981; Triska and Cromak, 1980). The
Export in harvested products 0 0 144 221
Total 2-yr export PIS{3} (615) 239 310
difference could also be related to a greater impor-
tance of denitrification in the Oregon forests (cf. Mar-
Data from Likens et al., 1970; Pierce et al., 1972. tin, 1985).
 High

7
/ \
i Pol ‘

a \ ae esessesee Berula alleghaniensis

3 !f ‘ Le ———— Prunus pensylvanica
$ \ 7 —— Rubus sp.
3 H } o” : —2—* Acer saccharum
3 : \ “i ‘% Ssasses Faous erandifolia
a i 4 gt Populus tremuloides
= \
1
\
~

Low
10 20 30 40 50 60 70 80 90 100
Years since disturbance
(a)

200
Total
70

60

Biomass
Prunus,
of 20
tha ha!
Nitrogen
kg
Prunus,
in

40
10
20

0 0
0 2 4 6 NY 10 12 14 0 2 4 6 8 lO i234
Stand age, yrs Stand age, yrs
(b) (c)
eeeeeeeee Stemwood and bark
—e—e Branches
t+ Leaves
———— ROO

Figure )-1) Role of pioneer plant communities in retention of nutrients after clearcut-
ting. (A) Typical sequence of dominant plant species in a secondary succession in a northern de-
ciduous hardwood forest in New Hampshire. Note the very rapid development of the pioneer
vegetation. (B) Accumulation of biomass in components of the pin cherry (Prunus pensylvanica)
community. (C) Nitrogen content of the pin cherry reaches nearly maximum values after only 6
years. Peak potassium values were observed after only 4 years. In this case, the pioneer vegetation
played an important role in conserving the site nutrient capital. In other situations, the pioneer
vegetation only develops a small biomass after clearcutting and may be proportionately less
im-
portant in nutrient conservation. (After Marks, 1974. Reproduced by permission of the Ecological
Society of America and P.L. Marks.)

130
 Effects of Forest Management on Forest Biogeochemistry 131

A study of clearcutting effects in British Colum- est. The study was undertaken after he found that a
bia revealed a somewhat greater response of stream conventional nutrient budget study failed to explain
chemistry than in Oregon but much less than in New the majority of an observed loss of approximately
England (Feller and Kimmins, 1984). Losses of dis- 1,000 kg ha” of nitrogen from the forest floor during
solved nitrogen in streamwater for 2 years after a 10-year postharvest period (810 to 1232 kg ha’ of
clearcutting were increased approximately 4 times nitrogen, depending on how it was calculated). He re-
the first year and 14 times the second year compared ported potential denitrification rates as high as 189 kg
with the losses before clearcutting. However, the ab- ha’ yr’ of nitrogen 5 years after clearcutting, at the
solute losses were small—the increase was from a peak of the assart flush of nitrogen that he observed.
very low level. This is higher than rates reported in most other stud-
Evidently, it is necessary to investigate the biogeo- ies, but high rates have been reported from elsewhere
chemical effects of clearcutting on a site- and region- (Davidson et al., 1990; Robertson and Tiedji, 1984;
specific basis. Also, the considerable difference in Struwe and Kjoller, 1990).
stream chemistry that sometimes occurs in adjacent
and seemingly similar streams means that conclusions
concerning the effect of clearcutting on ecosystem
function based on streamwater chemistry can be valid Effects of Slashburning
only if, among other things, they include a pretreat- Clearcutting inevitably removes some nutrients from
ment comparison of the streams. Broad generaliza- the site in harvested materials, virtually eliminates
tions that do not explicitly account for spatial and the biochemical cycle and much of the biogeochemi-
temporal variations in ecological conditions are un- cal cycle, and can make a major modification in the
warranted in the light of our present knowledge. Mac- geochemical cycle. The extent and duration of these
Gregor (1994) provides a review of the effects of forest effects are highly variable. Whole-tree harvesting
management on water quality. can remove a significant proportion of site nutrients,
and a delay in revegetation may permit leaching
Gaseous Losses of Nitrogen as a Result of Deni- losses to occur on a significant scale. Conversely, nu-
trification. Denitrification is probably the least trient withdrawals in logs can be modest, and with
well understood and most poorly quantified aspect of prompt revegetation, all three cycles return rapidly
the nitrogen cycle in forests (Tiedje, 1978). This is to something approaching the prelogging condition.
partly because it is so difficult to study and partly be- Most of the site nitrogen is often retained by immo-
cause of the assumption that denitrification is in- bilization, lack of nitrate production, or plant uptake.
significant in forest ecosystems (Ineson et al., 1991). Slashburning has a different effect. Although the
This assumption was based, in part, on the belief that question of wildfire effects is discussed in Chapter
rates of nitrification, which produce the NO, that is 12, a discussion of biogeochemical impacts of the
the substrate for microbial denitrification, are low in prescribed use of fire in forest management is pre-
the acidic, well-drained forest floors that characterize sented here.
many forests, especially coniferous forests (e.g., Vi- Combustion of organic matter converts most of the
tousek et al., 1979, 1982). More recently, it has been chemicals therein to oxides. Many of these remain as
recognized that denitrification can be a major but solids because of high melting and gasification temper-
largely unrecognized mechanism of nitrogen loss atures. They remain after the fire as ash, although a cer-
from forests after disturbance. In a study of Sitka tain proportion of this is carried aloft by fire-induced
spruce forest in Scotland, Ineson et al. (1991) found convection currents and distributed to downwind areas.
an annual gaseous loss of nitrogen by denitrification Where these air currents are strong, as in a hot fire cov-
of 3.2 kg ha"! of nitrogen, 80% of which was N,O. ering many tens of hectares, a large proportion of the
This loss was increased after clearcutting, and an esti- ash may be removed. For example, it is estimated that
mated 9 to 40 kg ha’ of nitrogen was lost annually approximately 0.5 to 8 kg of fly ash is produced for
during the first 2 years after felling, declining to the every metric ton of fuel burned. Where there is a re-
pre-felling rates after 4 years. Martin (1985) estimat- duction of 50 t ha™!, this would amount to between 25.0
ed potential denitrification losses after clearcutting of and 400 kg ha"! of mineral material (data in Cramer,
an old-growth, west coast hemlock—amabilis fir for- 1974a). Estimates of particulate emissions as high as 29
132. CHAPTERS Biogeochemistry

kg of fly ash per metric ton of fuel have been reported at Table 5-22 Percentage Loss of Biomass and Nutrients
the head of a wildfire (Cooper, 1973), and presumably During Slashburning of Experimental Plots with Two
Different Levels of Slash Loading
such high values may pertain under certain conditions
ina slash fire. This fly ash may have the following com- Loss with Standard Loss with Heavy
position: 0.0089% nitrogen, 11.7% calcium, 1.61% Slash Loading,* % Slash Loading,” %
Mg, and 10.01% potassium (Grier, 1972).
Biomass 93 96
Oxides that are gaseous at normal temperatures or
the temperatures in the fire are lost as gases in the N 91 92
smoke plume, and most of the nitrogen and sulfur in P 32 56
the burned material will suffer this fate. Losses of the K 16
somewhat more refractory materials may also occur. It Ca 2
is customary to use temperatures lower than 480°C in
Mg 21 19
preparing plant samples for analysis of cations because
above this temperature, potassium can be lost in sig- Na 0 25

nificant amounts. Some phosphorus may also be lost Data from Grier, 1972.
at higher temperatures. @The standard loading was 75 t ha’.
Calculation of the quantity of nutrients lost during >The heavy loading was 320 t ha”.
a prescribed burn is difficult, and there are few pub-
lished data. When heather (Ca/luna vulgaris) was
burned in the laboratory by Allen (1964) at 800 to
825°C, the percentage losses of nutrient content were sium, 51; calcium, 100; and magnesium, 37) and by
nitrogen, 76.2; phosphorus, 3.5; potassium, 4.9; calci- Flinn et al. (1979; nitrogen, 220; phosphorus, 8; potas-
um, 2.4; magnesium, 2.1; and sulfur, 56.3. At 550 to sium, 21; calcium, 123; and magnesium, 13).
650°C, the losses were 68 and 50% for nitrogen and After the slashfire, the area is left covered with a
phosphorus, respectively, and less than 1.4% for the layer of ash and charred material of varying depth.
others. Losses at higher temperatures were not re- The fate of the nutrients contained within this layer is
ported. Knight (1966) burned forest floor material highly variable. If the soils have become hydrophobic’
from a Douglas-fir stand and reported nitrogen losses or the surface soil pores are plugged with ash, then
of 410 kg ha! at 700°C and 167 kg har! at 300°C. much of the ash may be washed directly into streams.
Much of this was probably as molecular nitrogen and is At the other extreme, the ash may be washed into the
unlikely to be returned to the ecosystem by rain. Grier upper layers of the soil, where all the cations are held
(1972) burned Douglas-fir slash on areas with average by the soil cation-exchange capacity. Where water
and four times average slash loadings and calculated the flow through the soil is predominantly in macrochan-
percentage loss of nutrients to gasses and fly ash. The nels, much of the nutrient material leached out of the
results are shown in Table 5-22. In his study, fuel load- ash may bypass soil exchange sites and rapidly reach
ings (and therefore energy release) had an effect on the groundwater and streams. Alternatively, if the burn
loss of phosphorus, potassium, and sodium but little ef- results in the production of anions such as bicarbon-
fect on nitrogen, calcium, and magnesium. ate or nitrate, then significant quantities of cations
Feller and Kimmins (1984) reported on the loss of can be leached right through the soil (see the discus-
nutrients from the slash and forest floor during a sion of soil-leaching mechanisms, Chemical Proper-
slashburn in western British Columbia. On the basis ties of Soil and E. Soil Leaching, in Chapter 11). Any
of both field and laboratory studies, the quantities (kg of these situations can occur, and it is difficult to pre-
ha“) of nutrients lost during the burn were calculated dict the postfire movement of nutrients on slash-
to be nitrogen, 982; phosphorus, 16; potassium, 37; burned areas without a careful preburn site
calcium, 154; and magnesium, 29. These amounts assessment.
represented 41, 8, 24, 25, and 15%, respectively, of the Studies of nutrient leaching in a burned area of
total amounts of these nutrients in the forest floor and pine forest in South Carolina revealed that the fire in-
slash (Table 5-23). This compares with similar esti-
mates (kg ha') made for slashburns in Australia by
Harwood and Jackson (1976; phosphorus, 10; potas- *See Chapter 12.
 Effects ofForest Management on Forest Biogeochemistry 133

Table 5-23 Nutrient Loss, kg ha™', from Two Treated Areas and a Control Watershed" at the Haney Research
Forest, Southwestern British Columbia, During the First Two Years after Treatment (1973 to 1974, 1974 to 1975)
Compared with Nutrient Reserves in the Mineral Soil and Forest Floor in the Areas and Nutrient Inputs
in Precipitation

N P K Ca Mg

A B Cc A B Cc A B Cc A B Cc A B Cc

Exports
Streamwater al 3 1 0 0 0 11 9 3 44 OOmEEOO: y) 10 g
Logs 234 308 — 34 50 - (6Smecor - 260 467 _ 27 38 =
Atmospheric? = 982 — - 16 = - 37 - — 154 - - 29 _
Total 245 = 1298 1 34 66 0 179 =.283 3 304 676 55 36 77 )
Reserves
Forest floor® 1632 2180 1490 WA Wey sles fm hia es 724 742 526 102 OS 29
Mineral soil® 4566 4647 3924 18 16 7 87 148 85 489 332 149 38 25 16
Total 6198 6827 5414 140 190 162 158 258 180 Wiss OV ifs 140 120 145
Average annual
precipitation
input® 4 4 4 0 0 0 1 1 1 i it i 1 1 1

Source: Feller and Kimmins, 1984. Copyright American Geophysical Union. Used with permission.
“A = watershed A, a clearcut area; B = watershed B, clearcut and slashburned: C = control watershed C.
’Atmospheric exports after slashburning were obtained from Feller et al. (1983).
“Total quantities of nutrients present before clearcutting and burning.
‘Quantities of total nitrogen; extractable phosphorus; and exchangeable potassium, magnesium, and calcium present in the soil to rooting depth
(70 cm in watersheds A and C, 65 cm in watershed B).
“Average for the seven water years 1971-1972 to 1977-1978.

creased the solubility of cations as follows: Ca’*, 20- were retained in the upper 30 cm of the soil: 62% cal-
fold; Mg’*, 10-fold; Na*, 2.3-fold; and K’*, 2.3-fold. cium, 41% magnesium, and 81% potassium
Natural leaching reduced the yield of ions from the In a study of the biogeochemical effects of shifting
burned litter during a 1-month period by 80 to 83% cultivation in the tropics, Ewel et al. (1981) cut and
for divalent cations and 45 to 63% for monovalent burned an area of second-growth tropical forest in
cations. The burn did not affect the level of nitrate Costa Rica. Harvest of the larger trees removed 18%
and phosphate ions, but the solubility of nitrogen and of the sulfur and, with the exception of nitrogen,
phosphorus increased some time after the burn, prob- more than 10% of the total initial inventory of other
ably as a result of microbial activity (Lewis, 1974). nutrients in the initial biomass and upper 3 cm of soil.
Grier (1972) reported that during a 2-year period, Less than 5% of the initial nitrogen was removed in
0.86 g m”® of calcium, 4.24 ¢ m~” of magnesium, and logs. Burning the slash volatilized 31% of the bio-
49.1 g m™~ of potassium were leached from an ash mass, 22% of the remaining nitrogen, and 49% of the
layer after burning Douglas-fir slash in western Wash- remaining sulfur. Postburn losses of nitrogen were
ington. These losses compared with the following re- equal to 16% of the initial inventory (probably leach-
lease from unburned decomposing slash: 0.26 g m~ of ing of NO;>). Loss of phosphorus, potassium, calci-
calcium, 1.4 g m~ of magnesium, and 3.67 g m~ of um, and magnesium were 51, 33, 45, and 40%,
potassium. In a study of a ponderosa pine site burned respectively; these occurred after the onset of rains.
in a wildfire in central Washington, losses of 16.3 g The impact of the postburn decomposition of SOM
m~” calcium, 6.3 g m~ magnesium, and 7.9 g im” was found to be as important as the loss during the
potassium were recorded (Grier, 1975). The differ- burn. Decomposition released as much carbon in 154
ences between the two sites were explained in terms of days as was lost in the fire. Obviously, shifting cultiva-
differences in ash composition. Of the cations leached tion involving fire can have a significant effect on site
from the Douglas-fir ash, the following percentages biogeochemistry, especially when reinvasion of vege-
134. CHAPTERS — Biogeochemistry

tation is inhibited, thereby facilitating the loss of ash succession. Figures 5-16 and 5-17 present a general-
minerals and SOM. ized model of variations in tree nutrition and site bio-
After a burn, the availability of many nutrients is geochemistry over a tree crop rotation in an even-aged
initially increased, even if the total site capital of nutri- forest: the variation over a “stand cycle” of a midseral
ents has been reduced. As a result, the vegetation that stage of a management-induced secondary succession.
reinvades a slashburned site is initially well nourished, The geochemical cycle plays a major role in deter-
and the new flora will often contain species typical of mining ecosystem nutrient budgets over long time pe-
much more fertile sites. Fireweed is a species that is fa- riods—several tree crop rotations, or several centuries.
vored by high nutrient availability. It is a fire follower The balance of relatively small annual nutrient inputs
largely because of the increased nutrient availability. and outputs to and from the ecosystem determines the
The postfire flush of nutrients is not sustained, how- quantity of nutrients available for cycling in the bio-
ever; as the available nutrients are leached away, im- geochemical cycle. The important long-term role is
mobilized by microbes, or taken up by plants, accompanied by an equally important short-term role:
indicators of high nutrient availability such as fireweed (1) accelerated nutrient outputs as a result of natural
generally decline. This is widely known to bee keep- disturbance, forest harvesting, or other management
ers, who will keep their hives in harvested areas only activities and (2) accelerated inputs as a result of in-
until the assart flush declines and the loss of flowering creased soil weathering and biological nitrogen fixa-
plants leads to a decline in honey production. tion by early seral plant species after disturbance.
On a fertile site, slashburning will generally have Thus, the geochemical balance plays a major role in
little adverse effect on the site nutrient capital and on defining long-term potential sustainability of produc-
postfire nutrient cycling. The soil is well endowed with tion and the ability of the ecosystem to recover from
nutrients, there is often a good rate of natural replace- disturbance-induced losses. It contributes to ecosys-
ment, and most of the SOM is mixed down into the tem resilience.
mineral soil by animals, where it is protected against The biogeochemical cycle determines the avail-
consumption by fire. Conversely, loss of nutrients from ability of nutrients for uptake by plants. This plays a
nutrient-poor sites, as a result of slashburning can re- major role in determining leaf area, leaf efficiency, and
sult in a significant reduction in biogeochemical cy- carbon allocation. The magnitude and rate of the bio-
cling and a concomitant reduction in subsequent NPP. geochemical cycle thus are key determinants of pro-
The initial site nutrient capital is smaller, rates of nat- duction ecology, and this cycle plays the major role in
ural replacement tend to be low, and much of the site’s determining the sustainability of primary production
available nutrient capital is in the surface organic accu- from decade to decade within the tree crop rotation,
mulation, which may be burned off during the fire. or over a time period of approximately a century.
Burning or mechanical site preparation on such Maintaining a suitable rate of biogeochemical cycling
sites may require longer rotations than currently used thus is fundamental to sustaining growth within a
if site nutrient status is not to be degraded (e.g., Feller managed forest. The slowing down in the rate of cy-
and Hamilton, 1994). cling in some late-seral or climax forests is a major
For many of the nutrients lost by harvesting or nat- reason that the primary production of these forests de-
ural disturbance, replacement will be by precipitation clines over time. Thus, the biogeochemical cycle de-
inputs and mineral weathering processes. The latter is termines the extent to which the potential ecosystem
not well understood and has not received as much atten- production defined by the long-term geochemical bal-
tion as the former. For an introduction to the topic, see ance is actually achieved.
Zabowski et al. (1994) and van Miegroet et al. (1994). The internal cycle within plants acts to even out
short-term variations in the supply of nutrients for up-
take by plants—variations caused by short-term varia-
5.12 Variations in Contributions of the tions in the biogeochemical cycle. These can be
Three Main Cycles Over the “Stand caused by the seasonal cycle of temperature and mois-
Cycle” and Succession ture or by variations in climatic conditions from year
to year. This cycle is responsible for sustaining re-
The relative contribution of the three major cycles to markably uniform tree growth despite these short-
ecosystem nutrient budgets varies over the develop- term variations in nutrient availability. It supports the
ment of a stand (a seral stage) and over the course of tremendous burst of growth in the spring, the nutrient
 Long-Term Declines in Forest Productivity—Are They Fertility Related? 135

---- Soil nutrient availability

Clear- Tree nutrient uptake demand
cutting Contribution to tree nutrition
ete from internal cycling
Assart ' Period when symptoms of
period |nutrient deficiency are common
A ; A ~

Trees rely increasingly

on internal cycling and
less on uptake from the soil

yr-!
ha!
kg ane
ite
--

-_<-- --—.

-
hte
-- =" ——e——
ee

Tine =

Above
ground L

Below
f
ground |
Carbon
%
allocation

Figure 5-16 Generalized model of temporal patterns of variation in soil nutrient availability,
tree nutrient uptake demand, and contributions of internal cycling (A) and consequences for
below- and aboveground carbon allocation (B) over an even-aged tree crop rotation with
clearcutting. The three curves in A are not to scale and are intended to show temporal patterns
rather than absolute values, which will vary from site to site and species to species. The horizontal
time scale might be 10 to 15 years from clearcutting to the end of the assart period and 80 to 100
years for the entire scale.

demands of which generally exceed the rate at which sequences for carbon allocation of the patterns of avail-
nutrients can be taken up at that time. ability and uptake are shown. Figure 5-17 summarizes
All three cycles contribute to the sustainability of a wider range of biogeochemical variables over the
ecosystem processes but over different time scales. stand cycle. The temporal pattern for these variables in
Their contributions vary at different stages of succes- an uneven-aged forest managed with partial harvesting
sion and at different times within a seral stage or a stand systems would be much different, with much less varia-
cycle. Figure 5-16 shows the temporal patterns in soil tion in these processes over time.
nutrient availability over an even-aged tree crop rota-
tion and the patterns of tree nutrient uptake demand
and contributions from internal cycling. From this dia- 5.13 Long-Term Declines in Forest
gram, it is easy to understand why visual symptoms of Productivity—Are They Fertility Related?
nutritional stress generally occur in 15- to 25-year-old
stands when nutrient availability is declining at the end The decline in growth of radiata pine growing on nu-
of the assart period but uptake demand is increasing. trient-poor, summer-dry sandy soil in South Australia
Such symptoms are much less common in older stands, over three successive rotations stimulated interest in
where uptake demand is lower and the contributions the sustainability of plantation forestry (Farrell et al.,
from internal cycling are greater. The anticipated con- 1981; Keeves, 1996; Squire 1983; Squire et al., 1979,
 a Stand-replacing A a -
y* ; disturbance 7 : : y :
de LEN hed
a msn ayd
as SON

|
.

I,

Old growth phase

oh Dic TS RAF, LVS. FOV

Stand initiation phase

Ce Th
=: Pia

wail il
aucilll
(il

Stem exclusion phase

A. Forest floor depth and nutrient availability

77 ~, Assatt
/ \ period
Forest floor depth

Nutrient uptake
{

D. Relative role of understory

E. Nitrogen fixation, assuming the presence of symbiotic N fixing species in the early post-harvest period
| | | |
| we Saks J Symbiotic | |

Non-symbiotic
| ’
7 \
| | |
| | / ‘

aS
77 \

|
F. Snags and coarse woody debris (CWD) in naturally disturbed and managed forests

| Snags |CWD
—— naturally disturbed seem naturally disturbed
y ---- managed ‘ts managed

!
| | ———
' sgn
Ly aly ep gia ee a eii

Figure \-17 Generalized model of patterns of change in biogeochemical parameters over the stand cycle in an even-
aged forest harvested by clearcutting. The diagram shows only temporal patterns. The different variables are not drawn to
scale.
 Summary 137

1985). This reflects similar concerns in Europe that growth decline with age. Increases in allocation to fine
repeated removal of litter by peasant farmers was the roots and declining nutrient availability in older stands
cause of significant declines in Scots pine growth on are probably important, but insufficient data are avail-
poor sandy soils in northern Germany in the early able to confirm a general pattern. On the basis of their
1800s (Ebermayer, 1976; Wiedemann, 1935; Assman, review, they presented the conceptual model of carbon
1970). Research in the United Kingdom by Ovington flux accompanying forest stand development that is
(1957, 1959) and Rennie (1955, 1957) raised similar shown in Figure 5-18.
questions about the nutritional sustainability of
forestry on nutrient-poor sites. The results of short-
term (e.g., 5 years) experiments on the effects of site SUMMARY
mechanical and pyral site preparation on spruce plan-
tation growth in Sweden were reported to give mis- Because of the biochemical nature of ecological energet-
ics, organic production can occur in ecosystems only
leading results because of the confounding effects of
when the appropriate chemical elements are available at
the assart flush (Lundmark 1983, 1986). Forty-year appropriate concentrations, in the appropriate relative
growth results were different from 5-, 10-, and even quantities, and in the appropriate total amounts.
20-year results. What seemed to be good for tree The dynamics of nutrient chemicals in terrestrial
growth in the short term resulted in growth declines ecosystems can be identified with one or more of three
in the long term. cycles: the geochemical cycle, which involves inputs into
The issue of harvesting effects on long-term site and losses of nutrients out of a particular ecosystem; the
productivity has been examined repeatedly (e.g., Dyck biogeochemical cycle, which involves the uptake by, stor-
et al., 1994; Dyck and Mees, 1989, 1990, 1991; Evans, age in, and loss of nutrients from plants within an ecosys-
1990; Gessell et al., 1990; Johnson and Todd, 1998; tem, including movement of nutrients through grazing
and detritus trophic webs; and the biochemical cycle,
Kimmins, 1990, 1996; Messier et al., 1985) and has
which involves an internal redistribution of nutrients
been the subject of forest ecosystem modeling within organisms that permits the organism to satisfy
(Chapter 21). Because of the long time scale of this some of its nutritional requirements for new growth from
issue, ecosystem models will become increasingly use- within its own nutrient capital. For each cycle, a variety
ful in evaluating sustainability of ecosystem function. of pathways vary in relative importance among different
However, such models are only as reliable as the con- species and different environments.
ceptual models on which they are based. Unless these Mechanisms have evolved to conserve and store nu-
models represent all the key determinants of ecosys- trients within an ecosystem. Plants that establish on un-
tem function and reflect the complexity of the issue, colonized mineral substrate will gradually remove
their predictions will be unreliable. What is needed is available nutrients from the mineral layers and transfer
simulation models based on appropriately complex them to living plant biomass and a surface accumulation
of decomposing organic matter. The acids that they pro-
conceptual models (Chapter 21). This chapter has
duce release nutrients from unweathered soil minerals,
presented a fairly complete conceptual model of the and many early seral species are able to fix atmospheric
biogeochemical aspects of the issue. nitrogen symbiotically. The litter layer that they create
Ryan et al. (1997) presented a useful evaluation of promotes asymbiotic nitrogen fixation. Atmospheric in-
the question of decline in forest growth as a forest puts are also accumulated within the ecosystem. Trees are
stand ages. Their conceptual model compares the con- particularly well adapted to accumulate nutrients from
tributions of declining soil fertility, declining tree the geochemical cycle into a tight biogeochemical cycle,
vigor as trees grow older, increasing internal moisture and after a period of time, a forest may be able to live in
stress (hydraulic resistance) in trees as they get taller, virtual nutritional independence of the underlying min-
increasing allocation of NPP to fine roots and repro- eral layers. This important phenomenon permits reason-
ably productive forest growth on some exceedingly
duction, increasing tree mortality, and decreasing pho- nutrient-poor mineral substrates if the ecosystem re-
tosynthetic capacity of foliage as trees get older. They mains undisturbed for long periods and is a major reason
concluded that reduced leaf area (as a result of crown that forests are such a successful form of vegetation.
abrasion and tree death and possibly related to mois- In managing ecosystems, the biogeochemical mecha-
ture stress in tall trees and to declining nutrient avail- nisms responsible for production must be identified and
ability) and reduced foliage efficiency (possibly as a conserved. Over the past 2,000 years, especially in the
result of moisture stress) are major components of the past 50 years, humans have been dislocating nutrient cy-
138 CHAPTERS — Biogeochemistry

cles on an ever-increasing scale. In many areas, the accu- STUDY QUESTIONS

mulated nutrient reserves that have taken centuries or
1. What is the difference between a macronutrient and
even millennia to build up have been dissipated. Natural
a micronutrient?
processes are fully capable of rebuilding these reserves in
time but generally too slowly for human purposes. The 2. What are the three major types of nutrient cycle, and
energy required to rebuild these concentrations of nutri- how is each related to forest productivity?
ents rapidly once they are dissipated would be enormous. 3. How do terrestrial plants acquire nutrients?
Continued life as we know it will require that we con- 4. What are Ingestad ratios, and why are they important
serve available nutrients by maintaining biogeochemical in forestry?
cycles intact, and nutrient management will become as
important in forestry as it is in agriculture. 5. Is nutrition important for animals? For microbes?
6. What are the major inputs and outputs to and from
the geochemical cycle?
TAKE-HOME MESSAGE 7. What are the major pathways of the biogeochemical
cycle?
Life is about energy. Energy in biological systems is
8. Why is internal cycling important? What determines
mainly in the form of the chemical energy of organic
the percentage of retranslocation in a plant?
molecules. The energy that drives life processes cannot
be transferred from inorganic chemical energy or solar 9, Why is it interesting to know how nutrients are dis-
radiant energy and accumulated in biomass without an tributed in different parts of a plant?
adequate quantity and balance of the essential nutrients. 10. What controls litter decomposition rates?
Sustaining the productivity and environmental func- 11. Are understory plants important in the nutrient cycle
tions of forest ecosystems requires an understanding of of a forest ecosystem?
plant, animal, and microbial nutrition; of nutrient cycling
processes; and of the adaptations that permit organisms 12. Why are forests, left undisturbed for a long time, able
to survive, grow, and reproduce in environments in to grow reasonably well on nutrient-poor soils?
which one or more nutrients are at low concentrations or 13. How important is biological nitrogen fixation?
in limiting supply. Where does it occur, what are the mechanisms, and
Forest management in the past has often ignored how should this knowledge influence how you man-
the biogeochemistry of forest ecosystems and the nutri- age a forest ecosystem?
tional basis of sustained production and ecosystem 14. What are the key processes that regulate the avail-
function. Site nutrient management and site organic ability of nitrogen in forest floors?
matter management must become a major component
. What are the major ways in which forest manage-
of sustainable forest management systems. Because of
ment can affect forest biochemistry?
the complexity of the issues addressed in this chapter,
they are best incorporated into ecosystem management 16. What is the assart effect, and why is it important?
by means of ecosystem-level computer models that ex- 17. How can we judge whether the biogeochemical im-
plicitly represent nutrient cycling and its relationship to pacts of forest harvesting threatens long-term ecosys-
ecosystem function, ecosystem disturbance, and ecolog- tem productivity?
ical succession.
 Ecological Diversity and
the Ecological Stage
Physical Determinants of the Ecological Play

T. major differences in vegetation, animal life, and human soci-

eties and culture around the world are related to differences in the
physical environment. Food, clothing, living conditions, and history
of human social and cultural development are more similar between
countries at similar latitudes and locations relative to oceans and
large freshwater bodies than between countries at different latitudes
or continental/island locations. ‘They are more similar between
peoples of boreal forest cultures, between tropical forest human
communities, and between hot desert/semidesert human societies
than between any two of these global environments.
Physical environments set the potential for biological develop-
ment. Variations in solar energy, seasonality, temperature regimes,
wind, water, soil, and fire play a major role in determining the
species composition, structure, productivity, and rates and patterns
of change of vegetation and its resistance to and resilience after
disturbance. The variation in vegetation and autotrophic activity
that is a direct result of this physical or ecological diversity creates
variation in heterotrophic animal and microbial community com-
position, abundance, productivity, and geographical distribution.
Ecological diversity sets the potential—the ecological stage—
for biological diversity and is the physical framework within
which biodiversity and ecosystem function should be studied,
managed, and conserved. Part III addresses this physical diversity.
Starting with a brief review of systems of classification of ecologi-
cal diversity (Chapter 6), each of six physical components of the
environment—light, temperature, wind, water, soil and fire
(Chapters 7 to 12)—is examined. Consideration of physical diver-
sity ends with an assessment of how biotic communities are dis-
tributed along gradients of physical factors (Chapter 13). io
 a ie .ea aon
: UH: Vitis ney lestRolGar
je Binen
> v4 ne
By, i ane b
» o} a. C) {} 7
= i Ty eat Fen’
; =
AVON iwwesteCh: HAauD
~ - 4 al 1 a

Sets ts, 1) Fotiead oeDig tong ae

: ‘tutta 1A (rhe teye
loipintiToy
og of) Gre atear
ho pete say besa tt
hi "0 (eure
hal Sree uw ae
brauuil SMALL ened —
mule sh hey Coe ER nerd Wa 7

|
a EatA da lthi) 6 ii Sea

2) Wei(il gh oO), «Gclelee

na:

s
YaTati
Vet Le tod haus (p a.
~ bx Aree Evrae milircseeeeae .
—— ily <codrooret (hb erreded stad
wail ¥ ~ OWeo nino oe
ci penely 47) mn
2 yale 208} Ha) plow Wan ,
“
; ,
; IT uriti: elie teoh cama
3 leo | Epeedb
19N ome aeeee
S i (Legere 4 1. ‘JTS a} oy
"tives? PepiPpa
~ iqae Torratel nail oe
U4] SAnebeaie ae niga 7
' cmb bersehert
hay (or) wii ixeqpilintd
1
; remem ns SAIL
a GT a
et Oh ives sje) Die
- ieee pa
wi tot! 9 itriow erage
/ hy ee dS aepega:ay
cpu heteeine
iW ars fi
a roe paaee 1 tis a

Pein es al ¥ 4a tne
 Ecosystem Classification
The Ecological Foundation for
Sustainable Forest Management

6.1 Introduction statements varies depending on the breadth (i.e., ho-

mogeneity) of the class and the characteristics used to
The desire to create order where there is disorder, to define the class. The experience of swimming safely in
explain the unknown, and to predict events or condi- a lake that contains trout or in the sea in the presence
tions in the future is a characteristic of most, if not all, of mackerel is not a reliable basis on which to predict
humans. We have an innate desire to explain, under- the safety of swimming in piranha- or shark-infested
stand, and predict our environment and ourselves. Un- waters. The class of objects called fish is very broad,
fortunately, this is a difficult goal to achieve, given the and the definition does not specify whether members
diversity and complexity of ecosystems. We do not of the class might be inimical to human health or
have the time, the energy, or the mental capacity to whether humans constitute part of the group’s diet. To
know everything about all things. Our response to this make such a prediction with confidence would require
problem has typically been to arrange objects or facts splitting fish into two subgroups: human-eating and
about objects into groups of like individuals. This en- not human-eating. The usefulness of a particular clas-
ables us to extrapolate our knowledge about one or two sification and the criteria that should be used in defin-
individuals to all the members of such a group. For ex- ing classes will depend on the heterogeneity of the
ample, by stipulating that a particular animal is a dog, classes and the purposes for which the classification is
it is in most cases unnecessary to say that it has four to be used.
legs, a tail, and a fur-covered skin. One can also infer Classification is, as far as we know, a uniquely
that the animal is most probably larger than a rat and human attribute. It is undertaken with the objective of
smaller than a horse, that it generally does not eat hu- reducing the heterogeneity of a group of objects to in-
mans, and that it probably has a strong predilection for crease our understanding and ability to predict the
chasing cats. Similarly, by saying that an object in a for- characteristics of any one of the objects. In its very
est is a tree, it is immediately apparent that it is an im- essence, it is purposive: it is done for a reason, not as an
mobile organism that requires light, water, nutrients, end in itself. Purpose is implicit in all classifications,
and a secure anchorage and that it is a very uncomfort- and different purposes lead to different classifications.
able experience to have a mature specimen fall on you. There can be no single “best” classification that will
Classification, the act and the result of arranging serve all purposes under all circumstances, although
facts or things into groups or classes of like individu- there will nearly always be a best method for a specific
als, permits confident statements to be made about all application.
the members of a class on the basis of the knowledge The ideal forest classification for forest manage-
gained from the study of only a few members of that ment would be one that satisfies the needs of the for-
class. The degree of confidence that one has in such est resource manager and other forest users with

141
142 CHAPTER 6 _— Ecosystem Classification

minimum cost, time, and commitment of resources. area into such small units that the classification is
Such a classification should be formulated using virtually unusable.
those ecosystem characteristics that will give to the 2. Such a classification can be very expensive to for-
classification the predictive powers necessary to mulate and apply in mapping. Classification is pur-
achieve the stated objectives of forest management. posive, and detail is justified only when the
For example, before the development of an unhar- resulting benefits exceed the costs. Consequently,
vested valley for timber production, we might wish to complex, expensive classifications are generally re-
have a land classification that would predict the loca- garded with less favor than are simpler, inexpen-
tion of unstable soil and shallow, rocky soil to mini- sive ones, all other things being equal.
mize the cost of road construction, to make road
3. Many of the characteristics of ecosystems are
maintenance as easy as possible, and to reduce nega- closely correlated (i.e., when characteristic A in-
tive environmental impacts of the road. However, if creases, so does characteristic B), so that it is not
we wanted to be able to make reliable long-range always necessary to classify every characteristic.
plans concerning site-specific harvesting methods For example, on a given aspect at a given latitude
and postharvesting site treatment or if we wanted to
and longitude, temperature varies in a more or less
predict the recreational potential or wildlife habitat
predictable manner with elevation, so it may not
suitability of the forest, then different classifications
be necessary to include both of these parameters in
would be required. It is unlikely that a single classifi-
a classification.
cation could be optimum for all possible applications,
but some classifications can serve a broader range of
purposes than others. The “best” classification is generally a compromise
Forest classification is undertaken to improve our between the need for simplicity and the need for suffi-
ability to make reliable statements about the charac- cient detail to make the classification effective. In this
teristics, the resource and biological potential, and/or chapter, we review briefly the various approaches that
the environmental constraints on the management of have been made to forest and forestland classification.
any particular forest stand or land area. It is an attempt The important topics of field data gathering, analysis
to improve our ability to predict the characteristics of of vegetation data by traditional and “numerical” tech-
forest ecosystems by reducing the number of un- niques, and vegetation/ecosystem mapping cannot be
known determinants (cf. Figure 2-2). The large num- covered here for lack of space. Throughout the chap-
ber of such determinants, together with the wide ter, we assume that the primary objective is classifica-
variety of forest values and uses of classification in for- tion and mapping of the forest resource, rather than
est management, suggests the need for a variety of dif- ordination of the vegetation and gradient analysis (dis-
ferent classifications. This has the disadvantage, cussed in Chapter 13). The issue as to whether vegeta-
however, that it can result in an undesirable duplica- tion organizes itself into discrete communities that are
tion of effort and expense, and although useful forest repeated in similar environments throughout the
classifications are not necessarily immensely complex, landscape or is a continuum made up of species popu-
single-parameter classifications often have limited lations distributed independently along environmental
utility and may not satisfy the complex requirements gradients is discussed in Chapter 13. Readers who are
of multiple-use forest management. interested in fuller details of classification, field sam-
As an alternative, one might decide that the most pling, data analysis, and mapping are referred to
useful approach to classification of forests is one that Kuchler (1967), Shimwell (1971), Whittaker (1973a),
provides all possible ecological information desired by and Mueller-Dombois and Ellenberg (1974). A useful
forest resource users and managers: a-classification review of site classification in the United States is pro-
that explicitly incorporates all major parameters of the vided by Carmean (1975).
ecosystem. Although this may be true in some situa- There have been many different approaches to the
tions, it is not necessarily the best approach, for the classification of forests. Some have focused on those
following reasons: aspects of the physical environment that determine
the characteristics of the vegetation. Others are based
1. Homogeneous classes defined by all major pa- on the vegetation itself. Some approaches attempt to
rameters of the ecosystem can result in dividing an incorporate all the major ecosystem determinants in
 Climatic Classification 143

combined vegetation-environment or ecosystem tematic approaches incorporate climate as a major

schemes (called ecosystematic approaches). The reader component of the classification. All climatic classifica-
is reminded that there is no single best or correct clas- tions involve various combinations of degree of aridity
sification. “Best” is defined by application and needs (dryness) and warmth. Aridity is not just a matter of
that change with time, and a classification that was precipitation; it is determined by the ecological effec-
good 25 years ago under the prevailing circumstances tiveness of precipitation, as discussed in Chapter 10.
is not necessarily adequate to meet the needs of Perhaps the most durable of the climatic classifica-
present-day forest management and conservation. For tions used in vegetation ecology is that proposed by
example, forest classifications in the past were based K6ppen (1923). Building on the ideas of an earlier bio-
mostly on merchantable timber volumes and the ages climatologist, de Candolle (1874), Képpen made care-
of commercial species relative to timber management ful observations of the climatic conditions required for
information needs. These classifications have proved the growth of various groups of plants, and he related
inadequate for the identification of the extent, loca- variations in vegetation to temperature and precipita-
tion, and characteristics of “old growth” forest, for tion characteristics of the climate. He proposed a
wildlife management, and for management of recre- world classification consisting of five major climatic
ation and other nonconsumptive uses of the forest. types: tropical rainy, dry, humid mesothermal (warm
Classification is not a static, once-and-for-all activ- temperate), humid microthermal (cold boreal forest),
ity, and present-day classifications will undoubtedly and polar (tundra). These types were defined by arbi-
evolve to meet the future needs of forestry. However, trary temperature limits. For example, the warmest
it should be noted that where the initial classification month of the humid microthermal climate has an av-
involves a comprehensive, ecosystematic approach, it erage temperature above 10°C, and the coldest month
should be possible to adapt the basic classification to has an average temperature less than —3°C. The former
changing needs, eliminating the need for costly reclas- is claimed to be correlated with the poleward limit of
sification. If the initial work involves the establish- tree growth and the latter with the equatorward limit
ment of a basic taxonomy of ecosystems, then the of permafrost. Neither of these correlations is precise
classification will be much more adaptable than when (Money, 1972). Each of Képpen’s types is subdivided
it is based on a small subset of ecosystem parameters. by the addition of subscript letters, resulting in more
The classification approaches discussed here are homogeneous climatic subunits (lable 6-1). The
climatic, physiographic, vegetative, and ecosystematic. world distribution of K6ppen’s climatic types is shown
in Figure 6-1.
An alternative system based on the effectiveness of
6.2 Climatic Classification precipitation was developed by Thornthwaite in 1931.
This is now largely of historical interest, and it was re-
Climate is generally accepted as one of the major de- placed in 1948 by a second classification based on po-
terminants of vegetation. Major plant formations (the tential evapotranspiration (PE; discussed in Chapter
major vegetation types of the world, described in 10), defined by temperature and the water budget. Veg-
Chapter 13) are closely correlated with major climatic etation boundaries in the eastern United States coin-
zones, and local variations in climate are reflected in cide reasonably closely with PE values (Figure 10-4),
changes in species composition within a biome (a con- but unlike his 1931 and K6ppen’s classifications,
tinental-level ecosystem characterized by a major cli- Thornthwaite’s 1948 classification does not use vegeta-
matic type, vegetation type, and associated animal and tion boundaries to define climatic areas. Thornthwaite’s
microbial communities and major topographic/geo- system is not satisfactory in tropical and subtropical
logical features; see Chapter 13). Climate 1s important areas (Barry and Chorley, 1968), and the climate/vege-
in forestry because it influences the suitability and tation/soil relationships are useful only at a continental
productivity of a tree species on a particular site and level. Further details may be found in Thornthwaite
because it affects every aspect of forest management (1948) and Thornthwaite and Hare (1955).
from regeneration to harvesting (Thornthwaite and More recent developments in climatic classifica-
Hare, 1955). Many of the earliest attempts at broad tion have stressed the energy-balance approach
vegetation classification by plant geographers were (Budyko, 1974). PE is defined more by available ener-
based on climatic parameters, and present-day ecosys- gy (net radiation) than by temperature per se. Net ra-
Table 6-1 Képpen’s System of Climatic Classification as Modified by Trewartha
Major Climatic Types Climatic Subtypes

Symbol Climate Definition Symbol Climate Definition

A Tropical rainy Hot climate with no cool f No dry season Driest month has >60 mm
season. Temperature of precipitation.
the coldest month m Short dry season Rainfall sufficient to
>18°C support rain forest.
Ww Dry winter, wet summer Driest month has <60 mm
precipitation.
B Dry Evaporation exceeds Wh Hot desert Mean annual temperature
precipitation >18°C.
W = arid, desert Sh Tropical and subtropical A short rainy season.
S = semiarid, steppe semiarid
Wk Middle latitude interior Large annual temperature
desert range. Persistently dry.
Sk Middle latitude semiarid Meager rainfall, mostly in
summer.
Cc Humid Temperature of coldest Subtropical, dry summers Wettest winter month has 3x
mesothermal month between rain of driest summer month.
(moist 18 and 0°C Driest month has <30 mm.
temperate) Hot summers Warmest month averages
EPVAlG;.
Warm summers Warmest month averages
<22°C.
Humid subtropical hot Warmest month >22°C.
summers
No dry season Driest month has >30 mm
rain.
Dry winters Wettest summer month has 10x
rain of driest winter month.
Marine climate, cool-warm Rain at all seasons. Warmest
summers month <22°C.
No dry season
Marine climate, short cool Warmest month <22°C.
summers Less than 4 months >10°C.
Rain at all seasons.
D Humid Temperature of Humid continental Warmest month >22°C. Rain
microthermal coldest warm in all seasons with summer
(rainy/snowy, month <0°C, summers maximum. Winter snow
cold) warmest month >10°C cover. Variable weather;
alternating polar and
tropical air.
Humid continental, cool Warmest month <22°C,
summers as for “a,” but large snow
cover.
Subarctic Warmest month <22°C.
<4 months >10°C. Cold
winter. Light precipitation
Subarctic, very cold Coldest month <-38°C.
winters Very light precipitation.
E Polar Temperature of the Tundra Mean temperature of the
warmest month <10°C warmest month >0°C. Light
precipitation, mostly in
summer.
Ice cap. No month with mean
Perpetual frost temperature >0°C.

Based on Money, 1972.

“f, no dry season; w, winter dry season; s, summer dry season; W, desert climate; S, steppe climate; h, mean annual temperature
>18°C; k, mean an-
nual temperature >18°C; a, mean temperature warmest month >22°C; b, mean temperature warmest month 22-1 0°C,
4 months have mean tempera-
ture >10°C; c, mean temperature warmest month 22-10°C, <4 months have mean temperature >10°C; d, mean temperature coldest month
<-38°C.
144
 Climatic Classification 3)

LASOGG:
se

A Tropical rainy climate

pare B Dry climate
EZ C Moist temperate climate

D Moist cold climate

ee E Polar climate
| ET —J

Af TROPICAL RAIN FOREST CLIMATE BS STEPPE CLIMATE Cs SUMMER-DRY WITHOUT DRY SEASON,
Am TROPICAL MONSOON CLIMATE h Denotes a Hot B Climate Cw WINTER-DRY Cfe COOL SUMMERS
Aw TROPICAL SAVANNA CLIMATE k Denotes a Cool B Climate WITHOUT DRY SEASON, Dfa, Dwa HOT SUMMERS
BW DESERT CLIMATE k’ Denotes a Cold B Climate HOT SUMMERS Dfb, Dwb WARM SUMMERS
n Denotes a B Climate with Frequent Fog Cfb WITHOUT DRY SEASON, Dfc, Dwe COOL SUMMERS
WARM SUMMERS Dfd, Dwd VERY COLD WINTERS
w __ Denotes a Dry Season in Winter ET TUNDRA CLIMATE
f Indicates the Absence of a Dry Season EF FROST CLIMATE
H Denotes Polar Climates E due to High Altitudes

Figure 6-1 Képpen’s system of classification of world climates. Only the major classes of cli-
mate are shown.

diation available annually for evaporation from a wet ecosystem classifications to be described below, and
surface (R,,) is compared with the heat required to climatic data are incorporated into most modern clas-
evaporate the mean annual precipitation (L,). The sification activities.
ratio R,,/L, (the radiative index of dryness) has values less Most climatic classifications result in the prepara-
than 1.0 for humid areas and greater than 1.0 for dry tion of climatic maps. An alternative approach is the
areas; (e.g., desert, >3.0; semidesert, 2 to 3; steppe, | to presentation of individual climatic diagrams for various
2; forest, 0.33 to 1.0; and tundra, <0.33). The relation- geographical locations. These permit the presentation
ships among net radiation, the radiation index of dry- of much more detailed information on moisture and
ness, vegetation zones, and the hydrological balance temperature conditions and are more useful in de-
are shown in Figure 6-2. tailed ecological or resource management work than
The climatic classes in the classifications described are the more general climatic schemes described
above are generally too broad to render them useful above. Climatic diagrams for the whole world have
for site-specific forest management decisions. They been published by Walter et al. (1975), and the result-
do, however, indicate the general vegetation potential ing climatic diagram maps have been synthesized into
and are therefore useful in regional planning. They world climatic maps. Some examples of ‘Thornth-
provide the logical framework for the more detailed waite’s climatic diagrams are presented in Figure 6-3.
146 CHAPTER6 — Ecosystem Classification

Forest
100 100 Runoff, cm yr 7
Tropical
Wet
Savanna Semidesert

yr!
em?
keal
R,,,

Radiative index of dryness (R,,/L,) Radiative index of dryness (R,,/L,)

(a) (b)

Figure 6-2 Relationships of net radiation (R,,) and the radiative index of dryness to vegeta-
tion (A) and river runoff (B). (After Budyko, 1974. Copyright Academic Press, Inc. Used with per-
mission of the publisher and the author.)

—
o—o Potential evapotranspiration
6.3 Landform, or Physiographic,
— — Precipitation
»--@ Actual evapotranspiration
Classification
fxg Water surplus
= Water deficiency The basic resource in forestry is the soil and its associat-
(ZZZ Soil moisture utiliz:ation |
SS) Soil moisture recharge +
ed landforms. Soil and landform play a major role in
determining forest structure, composition, and produc-
Bx
2.7 cm
tivity, and they form the basis of physiographic classifica-
7, oe tions.
The main advantage of the physiographic ap-
proach is that it lends itself to remote sensing, which
enables large areas to be mapped relatively rapidly and
inexpensively. There are several other advantages. Be-
cause it is based on relatively permanent site features

aS {ee = eee
(macroclimate, landforms, and soil), it is a relatively
permanent classification, resulting in relatively un-
Manhattan, Kansas Manaus, Brazil
changing maps. Because the classes are defined by pa-
rameters that determine vegetation composition,
structure, and productivity, the classification has a
sound ecological basis. There are, however, a number
of disadvantages. One of these is that it is difficult to
use the many continuously variable parameters of cli-
mate, landform, and soil to produce units of classifica-
tion that are appropriate to vegetation classification
without reference to the vegetation itself. Thus, phys-
iographic approaches tend to evolve with time into
biophysical or ecosystematic approaches by the explic-
it inclusion of vegetation parameters. For an example
JFMAMJ JASONDJ JFMAMJJASONDJ
of a physiographic system, see Burger (1972) and Hills
Chapingo, Mexico Seattle, Washington
(1976). A brief comparison of physiographic ap-
Figure 6-3 Climatic diagrams developed by Thornthwaite proaches in Canada, Australia, and the United States
and Hare (1955) for locations in Kansas, Washington is given in Wertz and Arnold (1975).
(U.S.), Mexico, and Brazil. (Used with permission of the Food Many physiographic classifications have used both
and Agriculture Organization of the United Nations and the au- soils and landforms to define classes. Some have used
thors.)
 Vegetation Classification Ler

just soils, such as those in Germany (Wittich, 1962) scientist often uses the vegetation to help in the map-
and in the United States (reviewed in Jones, 1969). ping and interpretation of soils, and many people have
Where soils are classified according to ecological believed that it is therefore more logical to classify and
rather than purely pedogenic (soil development) crite- map the vegetation (or the basic biotic community) it-
ria, the correlation among soils, vegetation, and over- self. The biota constitutes the best measurement and
all ecosystem characteristics may be fairly close. integration of the total physical and biotic environ-
However, where soil classes are fairly broad and are ment (Daubenmire, 1976), and it generally reflects the
basically pedogenic, correlations with vegetation may overall ecological characteristics of a site more faith-
be poor and of relatively little use as the basis for for- fully than is possible with other approaches, given the
est classifications. Also, soil data alone cannot provide current level of ecological understanding and ability to
all the information needed by the resource manager measure the environment. The widespread use of veg-
because of the complex interactions among soil, cli- etation in classification and mapping of soil and cli-
mate, topography, water, vegetation, and fauna. mate supports this point of view.
Soil classifications are extremely useful in all as- Vegetation classification attempts to identify dis-
pects of forest management even if they fall short of crete, repeatable classes of relatively homogeneous
the ideal classification. They help identify areas of un- vegetation communities or associations about which
stable or erodible soils, areas of high or low road- reliable statements can be made. Once a classification
building suitability, areas that will be difficult to has been established, field mapping of vegetation may
regenerate, and areas that are suitable for particular be undertaken. Classification assumes either that nat-
crop species. They are also valuable in recreation, ural vegetation groupings (communities) do occur or
wildlife, and watershed management (Crawford, that it is reasonable to separate a continuum of varia-
1975). They can give a good indication of productivity tion in vegetation composition and/or structure into a
if appropriately classified (see Thomas and Burroughs, series of arbitrary classes.
1975). Knowledge of the soil, the forester’s basic re- Vegetation classification can be approached in
source, is the first step toward understanding the over- many different ways: structure and life form, domi-
all functioning of ecosystems, and soil classification is nance, floristic composition, or plant productivity.
an essential component of any ecosystematic or inte- Each of these approaches can involve different meth-
grated classification. ods of investigation and different criteria for the defi-
One of the interesting approaches to soil classifica- nition of classes. The choice of criteria to be used
tion for ecology and forest management is the edaphic depends on three points (Whittaker, 1973b): (1)
grid (Remezov and Pogrebnyak, 1969), which depicts accessibility—criteria that are easily observed and mea-
the occurrence and growth of plants under various sured in the field; (2) stgnificance—ability of criteria to
combinations of soil moisture and nutrients. The distinguish one community from another (i.e., the de-
edaphic grid and its application in forestry are de- gree of correlation between the criteria and character-
scribed in a later section of this chapter. istics of the community and its environment); and (3)
effectiveness—suitability for expressing environmental
differences or producing units of classification at the
6.4 Vegetation Classification required level of detail.
Different approaches to classification have been
Climatic and physiographic classifications are an indi- triumphed vigorously by ecologists who work in dif-
rect approach to the classification of forest ecosys- ferent countries and biomes. These diverse views will
tems. Their classes are defined by factors that probably persist, and it is unlikely that a single, com-
determine vegetation. To the extent that an indirect mon method will ever be adopted because differences
approach can perceive, measure, interpret, and incor- in environmental and biotic conditions and in classifi-
porate all significant physical factors, it should be sim- cation needs will favor different approaches at differ-
ilar to ecosystem classification in its ability to predict ent times and places. In the following section, the
the ecosystem. However, rarely do such classifications various approaches are described briefly. Some of the
incorporate all significant physical factors, and there is more important methods are presented in more detail
frequently a problem of knowing which parameters to in a subsequent section. For a detailed treatment of
measure and how to interpret the significance of the this topic, see Whittaker (1973a) and Mueller-Dom-
resulting data for vegetation. Consequently, the soil bois and Ellenberg (1974).
148 CHAPTER6 — Ecosystem Classification

Classification of Structure and Life Form: Table 6-2 Terrestrial Formation Types of the World
Physiognomic Classification Tropical rain forest

The earliest attempts at vegetation classification were Tropical seasonal forest

made by the plant geographers Humboldt and Grise- Temperate rain forest
bach during the first half of the 19th century. They Temperate deciduous forest
characterized classes of vegetation by the growth form Temperate evergreen forest
of the dominant plants and the type of environment Taiga (subarctic-subalpine needle-leaved forest)
in which they grow. These classes were called
Se
10)
OD.
Bal
ee
Cae Elfinwoods (subalpine zone forest on tropical moun-
formations: major kinds of plant communities on a tains)
given continent characterized by their physiognomy
8. Tropical broad-leaved woodlands
and a range of environments to which that physiog-
nomy is an adaptation. Definition of the type of envi- 9. Thornwoods

ronment is important. Tropical grassland (savanna), 10. Temperate woodlands

temperate grassland (steppe), alpine meadow, and 11. Temperate shrublands
salt-marsh all are dominated by plants of the same life 12. Savannas (tropical grasslands)
form, but they all are regarded as different formations
13. Temperate grasslands
because of differences in the physical environment
14. Alpine shrublands
(Beard, 1973). As is discussed later, formations are
part of biomes and therefore have similar geographi- 15. Alpine grasslands
cal locations. 16. Tundra (treeless arctic plains)
The physiognomic approach was developed by 17. Warm semidesert scrubs
the Danish ecologists Warming (1909) and Raun- 18. Cool semideserts
kiaer (1934). Warming defined an association as a
19. Arctic—alpine semideserts
community of definite floristic composition within
20. True deserts (subtropical)
a formation and noted that a formation is an expres-
sion of certain defined conditions of life; it does not 21. Arctic—alpine deserts
define the floristic character of the vegetation.
After Whittaker, 1975a.
Raunkiaer introduced the /ife-form classification
“The relationship of these formation types to world-scale gradients
(discussed in Chapter 8) and the concept of biological of temperature and moisture are shown in Figure 16-4.
spectrum (percentages of the species in a community
that belong to various life forms), but his approach
has remained less influential in physiognomic clas-
sification than the central concept of the growth former is used to produce a descriptive formula, and
form of the vegetation dominants. A listing of the the latter is used to prepare stylized diagrams of the
major growth forms of terrestrial plants is presented vegetation structure.
in ‘Table 15-1; major formation types are given in A somewhat analogous approach is that of profile
‘Table 6-2, and their distribution in relation to mean diagrams, which were developed for use in the descrip-
annual temperature and precipitation is shown in tion and classification of tropical forests (Davis and
Figure 6-4. Richards, 1933, 1934), where either floristic complex-
Physiognomic classification can be based on a di- ity or incomplete floristic information makes a floris-
rect description of vegetation structure. This method tic approach difficult. This method involves the
was introduced by Kuchler (1947) and_ subsequently preparation of a descriptive diagram of the vertical
modified and improved by Dansereau (1951, 1957). structure of the vegetation, in which the relative posi-
They described vegetation using six parameters: tions of the various layers of the vegetation are accu-
growth form, size, “function” (coniferous, deciduous, rately depicted (Figure 6-5). Such diagrams can be
etc.), leaf shape and size, leaf texture, and ground cov- used to portray the variation in formation types along
erage. Each category is divided into three to six sub- major environmental gradients, and they are helpful in
categories, which are associated with two sets of explaining the complexity of the vegetation mosaic
symbols, one alphabetic and the other pictorial. The that occurs in mountainous topography. They can also
 Vegetation Classification 149

os
Tropical

400

th Warm
temperate

Tropical
rainforest

E300
=]
S |
S
‘6
E2 250
: Temperate
Cool rainforest
Z
temperate
:
Ss Tropical
seasonal
o
= forest
9

Temperate
150 F Cold forest
temperate 4,5
a
2 La
7.
ou) Taiga
Arctic-
alpine

50

Le Semi-deser
20 Desert

5 10 15 20 25 30
Mean annual temperature °C

Figure 6-4 Relationship between world formation types and climate (temperature and hu-
midity). The numbers refer to the formation types of Whittaker (1975) as given in Table 6-2. The
shaded area enclosed by the dashed line is a range of environments in which either grassland or
woody plants dominate. (After Beard, 1973; based on Holdridge, 1947. Copyright Dr. W. Junk Pub-
lishers, Netherlands. Used with permission of the publisher and the author.)

be used in the description and classification ofindivid- communities on different continents and is particu-
ual plant communities. larly useful in the initial description and classification
As the earliest approach to vegetation description of diverse forest or other vegetation types about
and classification, the physiognomic approach, re- which there is little or no immediately available taxo-
mains the principal basis for treating vegetation on a nomic information. However, the approach is less use-
broad scale in relation to climate, and it is an impor- ful than some of the alternatives as a basis for
tant basis for vegetation research in certain parts of site-specific forest management, especially in areas
the world. It is one of the best ways of comparing with relatively simple floras and readily available
150 CHAPTER6 = Ecosystem Classification

meters 120 feet

35
30 + 100

25 + 80 ‘
X Vr
call A feet meters

15 ; Ke
50+ 15
* 40 a 6 : ié nes ead 40
y Vt ood (Adry,
:
7, stale10

Uh sawal thy Fy’LAI Y¢39PD

My. \ 30
: Nae

:
20 Ne : is he n>
OVI] gaye {LO
jt MU Yagil 724 p
Evergreen seasonal forest _|Semi-evergreen seasonal forest Deciduous seasonal forest Thorn Cactus scrub
woodland

Increasing severity of drought

Figure 6-5 Profile diagram showing the sequence of formation types that has been de-
scribed along a moisture gradient in Trinidad. (After Beard, 1973. Copyright Dr. W. Junk Pub-
lishers, Netherlands. Used with permission of the publisher and the author.)

taxonomic information. Description of other physiog- dardized type of community unit. Dominance types
nomic methods of classification can be found in are poorly suited to the construction of a formal hier-
Mueller-Dombois and Ellenberg (1974). archical classification of vegetation in such areas
(Whittaker, 1973b). Conversely, in areas with a simple
flora, dominance types may adequately describe the
Classification Based on Dominance ‘Type major features of the vegetation and may be detailed
Plant formations are often divided into subordinate enough to define at least some of the smaller vegeta-
vegetation units on the basis of their dominant species tion units. The information requirement for this type
(dominant in terms of biomass, density, height, cover- of classification is generally much lower than that of
age, etc.). In both British and American plant ecology, other floristic approaches, which may make the
classification of communities has often been based on method attractive despite its several shortcomings.
the dominant plant species. Individual, easily noticed
plant species provide the simplest floristic tool by
which to create some order in the great variability of Classifications Based on
plant communities, and dominance has traditionally Floristic Composition
been used in describing types of forest: beech forest, Floristic classifications constitute the best developed
pine forest, or spruce forest. The national system of approach to vegetation classification. Most European
forest classification in the United States (forest cover and Soviet schools of plant ecology utilize this ap-
types) is based on the dominant tree species in the proach, and it has been widely applied in Canada and
overstory (Society of American Foresters, 1980; de- elsewhere. The approach can be broken down into
scribed in the section “Overstory Composition”), and three major subdivisions: emphasis on ground or sub-
in the past this type ofclassification has been preferred ordinate vegetation, emphasis on overstory vegeta-
by many applied ecologists, foresters, and range and tion, and emphasis on the entire plant community.
wildlife managers in many countries.
Because some species attain dominance over a Ground Vegetation. Emphasis on ground vegeta-
wide geographical area whereas others occupy a re- tion characterizes the classification scheme developed
stricted range ofhabitats, vegetation classes defined by in the Scandinavian and Baltic region. The approach is
dominance vary greatly in their heterogeneity. In areas especially associated with the work of Cajander (1926)
with a diverse flora, the selection of dominant species in Finland: the forest site-type classification. The Baltic
by which to define the class may be arbitrary, and a region is characterized by few tree species, each of
dominance type therefore cannot be a particular, stan- which can occur over a wide range of sites with differ-
 Vegetation Classification i>

ent undergrowth associates, the distribution of which mixture of tree species. Medium and rich site types
is relatively independent of the overstory. The under- thus can be divided into a number of forest types,
story vegetation provides much more information which are identified by both understory and over-
about and expresses more effectively the habitat story species composition. For example, the
conditions than does the species-poor overstory. Suc- Vaccinium myrtillus site type is divided into several
cessful application of the system requires a good forest types on the basis of the association of this
knowledge of succession and of the undergrowth- species with either pine or spruce in the climax
environment species relationships, because the over- condition and either birch or aspen in the seral
story composition and structure of stands representing condition.
a given habitat or site type can vary considerably. The
site type is an abstraction based on the climatic climax
The Finnish system has been widely applied in
condition (Chapter 17), with a number of variations
according to the seral stage. Finnish forest management. Forest composition and
species productivity have been studied, and volume
The Finnish system is applied in three stages:
versus age curves have been produced for the various
site types (Figure 6—6). Site-type classifications have
1. Site-type classes. Finnish forests are divided into five also been applied in Sweden, eastern Canada (Burger,
such classes on the basis of broad features of the 1972), parts of the United States, and the Soviet
site (moisture and fertility status): dry and poor Union (Frey, 1973), and the units of this classification
(heathlands), fresh-mesic (mossy forests), fresh- have proved to be useful for making decisions about
and-rich (broad-leaved forests), wet-and-rich (in- forest regeneration and forest management strategies.
undated forests), and wet-and-poor (bog forests) The method has been most successful under boreal
site types. Site-type classes represent a broad divi- conditions, where the diversity of the flora, the cli-
sion of the vegetation into plant formations and of mate, and the edaphic conditions is low (e.g., Rowe,
ecosystems into biomes. Each site-type class is di- 1956; see the review in Burger, 1972). Similar applica-
vided into a number of site types. tions of the Finnish system have been made elsewhere,
2. Site types. These are generally identified on the basis as for example the height versus age and volume ver-
of the entire understory vegetation, although where sus age curves that were prepared for Douglas-fir site
the flora is simple, the definition may be based types in Washington, Oregon, and southwestern
largely on one or two dominant species that give the British Columbia (Spilsbury and Smith, 1947). The
name to the site type. In the boreal conditions of method did not work well in this region because ofthe
Finland, Cajander stressed dominance and competi- higher diversity of the trees and minor vegetation and
tion within the understory vegetation, but a full the greater diversity in the physical environment. The
characterization of site type involves specification of same conclusion was reached in Ontario and Quebec
dominant, constant characteristic and differential in eastern Canada: the Finnish system worked well in
species (defined below) of stable climax stands. Be- the central and northern areas of these provinces,
cause the characteristic climax ground vegetation where boreal conditions occur, but the system was re-
establishes itself within decades after disturbance, jected in favor of a total community classification in
this approach permits accurate assignment of seral the floristically richer southern areas. The work on
stands to site-type classes with relative accuracy and the Finnish system in the Pacific Northwest is now
ease. Using dominant ground vegetation, Cajanda largely of historical interest.
separated the heathland site-type class into five site
types: Cladonia type, Myrtillus—Cladonia type, Overstory Composition. The focus of attention in
Calluna type, Empetrum—Myrtillus type, and forestry has frequently been on the tree crop: its com-
Vaccinium type . Other site-type classes are similarly position and productivity. Both Canada and the
subdivided. Most site types are further subdivided United States have national classifications/inventories
into forest types. of forest vegetation based on the species composition
3. Forest types. Except at the extremes of moisture and of the overstory, and these have been widely used by
nutrient conditions, a given site type can carry dif- both the professional and the scientific communities
ferent tree species or have different proportions of a as a standardized description of forests. The U.S.
 a
Oxalis-Myrtillus type

500 Herb types

| a
Myrtillus type
e
e
e
e @
e e

400 ® ° @
e
e oO

e- e o
©. oO "t
Vaccinium
= e oo0 = O (blueberry) type
ls oo
a e a a)
E e eo 0 5 Oo
oO Oe, fa) Oo
2 300
a= e Oo Oo Oo
5 o0 QO
co A =
000 f/f, a Calluna (heather) type
a 00 £
- ooo
e A *
e re 1x a 44
r(a) A
200 ee Oo
a//e yg F
e o.6 60 ee Cladonia (lichen) type
@ 4 DK A
| a oO yay A Aa
ogo fa
4, a®
e 4/7
e@ OA nm A
e@ oO 4 a 4 A
OG av A
100 a Ti Aa ee’ peek
O * =
n A
A yay
of ar ax
Daa
DA gk aa 4 4 A
rN

LY rac —a
|) - ieee RE! Meee eee Pe
0 50 100 150
Stand age

Figure 6-6 Volume-over-age curves for five forest site types in Finland, from a moist, fertile
herb type (Oxalis-Myrtillus) to a poor, dry, lichen type (Cladonia). The spread of the data about
the lines varies between types. (After Frey, 1973; based on Cajander and Ilvessalo, 1921, and Cajan-
der, 1949. Copyright Dr. W. Junk Publishers, Netherlands. Used with permission of the publisher
and T. E. Frey.)
 Vegetation Classification 153

classification’ (Society of American Foresters, 1980) some are more sensitive than others in expressing cer-
divides North America into eastern and western tain relationships, and this approach, which was devel-
halves, for each of which the major forest cover types are oped in central Europe and which is generally referred
listed and described. A forest cover type is a category to as the Braun—Blanquet method, seeks diagnostic
of forest defined by the trees presently occupying the species that are effective indicators of the relationships
area, no implication being conveyed as to whether it is of interest. These diagnostic species are used to orga-
temporary or permanent. A forest type is defined as “a nize communities into a hierarchical classification of
group of stands of similar character as regards compo- which the association is the basic unit. Used in this con-
sition and development due to given ecological fac- text, an association is “a plant community of definite
tors... .” Emphasis is given to “composition” rather floristic composition and uniform physiognomy which
than development, and the classification is based on occurs in uniform habitat conditions.” This definition
existing tree cover. Table 6-3 shows the forest cover was adopted in 1910 at the Brussels International
types for North America. No map of the distribution Botanical Congress and has become the generally ac-
of these types has been provided. cepted use of the term “association” throughout Eu-
The Canadian national forest classification (Rowe, rope. It should be noted that the uniform habitat
1972) identifies forest regions, defined as stable, climat- criterion is not always met, emphasis being placed on
ically controlled plant formations characterized by the floristic composition and physiognomy.
presence of certain tree species: the climax dominants. The Braun—Blanquet approach accepts a view of
The regions, of which there are eight (plus two non- the plant community that is intermediate between the
forest regions: grassland and tundra), are subdivided superorganism concept and the individualistic or con-
into forest sections, which are defined by the consistent tinuum concept (discussed in Chapter 13). It recog-
presence of certain “associations” and which show a nizes the reality of continuous species distributions
character that is different from other parts of the re- but emphasizes interactions between species that lead
gion. The term “association” in this Canadian system to relative discontinuities between communities. A
is used to describe a recurring community of one or more or less discontinuous continuum of species dis-
more tree species. Actually, the Canadian system was tributions results from the combination of competi-
developed as an inventory of the nation’s forests, the tion (and the resulting development of ecological
components of which were subsequently assigned to niches; see Chapter 15), coevolution (development of
classes using criteria that were defined after the inven- obligate or facultative interspecific relationships), and
tory was completed. Thus, it differs somewhat from a that the dominant species in the community often play
classification in the pure sense. a major role in determining the physical characteris-
tics of the microenvironments in which the subdomi-
Composition of the Entire Plant Community. nant species grow. There is continuous variation in
Two classification systems that are based predomi- community composition along an environmental gra-
nantly on the entire plant community are described. dient, but the degree of change per unit length of the
gradient is very low in some sections of the gradient
The Central European, or Braun-Blanquet, Ap-
(associations) and very high in others (transition zones
proach. Perhaps the most widely applied approach to
between adjacent associations).
floristic classification involves the study of entire plant The Braun—Blanquet approach stresses the dis-
communities. These are conceived as types of vegeta- tinction between the abstract idea of the association (an
tion recognized by their structure and floristic compo- idealized vegetation class of floristic classification; for
sition, which are considered to be a better expression a history of the various definitions of the term “associ-
of between-community differences and community ation,” see Westhoff and van der Maarel, 1973) and
environment relationships than any other community the real plant community that is growing in a real phys-
characteristic. Among the species in the community, ical environment and that is assigned to a particular
abstract class or association. This is similar to the dis-
'This was also listed under dominance types of classification. Originally a
tinction between an individual eastern hemlock tree in
dominance classification, it is now a hybrid between overstory dominance a stand and the abstract idea of the eastern hemlock
and overstory composition. species to which this individual plant is assigned.
Table 6-3 Forest Cover Types of North America
Eastern Forest Cover Types Western Forest Cover Types
pn iadehs cheieheinaetdet ian es SY eS fFee ee eee cr SEE ee

Boreal forest region Southern forest region Northern interior (boreal)

Boreal conifers* Southern yellow pines White spruce (201)
Jack pine? (1)° Sand pine (69) White spruce—aspen (251)
Balsam fir (5) Longleaf pine (70) White spruce-paper birch (202)
Black spruce (12) Longleaf pine-slash pine (83) Paper birch (252)
Black spruce-tamarack (13) Shortleaf pine (75) Balsam poplar (203)
White spruce (107) Virginia pine (79) Black spruce (204)
Tamarack (38) Loblolly pine (81) Black spruce—-white spruce (253)
Boreal hardwoods Loblolly pine-shortleaf pine (80) Black spruce-paper birch (254)
Aspen (16) Slash pine (84)
Pin cherry (17) South Florida slash pine (111) High elevations
Paper birch (18) Pond pine (98) Mountain hemlock (205)
Oak-pine types Engelmann spruce-subalpine fir (206)
Northern forest region Longleaf pine-scrub oak (71) Red fir (207)
Spruce-fir types Shortleaf pine-oak (76) Whitebark pine (208)
Red spruce (32) Virginia pine-oak (78) Bristlecone pine (209)
Red spruce—balsam fir (33) Loblolly pine-hardwood (82) California mixed subalpine (256)
Red spruce-Fraser fir (34) Slash pine-hardwood (85)
Red spruce-yellow birch (30) Bottomland types Middle elevations, interior
Red spruce-sugar maple—beech (31) Cottonwood (63) Interior Douglas-fir (210)
Paper birch-red spruce—balsam fir (35) Willow oak-water oak—diamondleaf oak (88) White fir (211)
Northern white-cedar (37) Live oak (89) Western larch (212)
Pine and hemlock types Swamp chestnut oak-cherrybark oak (91) Grand fir (213)
Red pine (15) Sweetgum-willow oak (92) Western white pine (215)
Eastern white pine (21) Sugarberry—American elm—green ash (93) Blue spruce (216)
White pine-hemlock (22) Sycamore-—sweetgum—American elm (94) Aspen (217)
Eastern hemlock (23) Black willow (95) Lodgepole pine (218)
White pine-northern red oak-red maple (20) Overcup oak—-water hickory (96) Limber pine (219)
White pine-chestnut oak (51) Baldcypress (101) Rocky Mountain juniper (220)
Hemlock-yellow birch (24) Baldcypress-tupelo (102)
Northern hardwoods Water tupelo-swamp tupelo (103) North Pacific
Sugar maple (27) Sweetbay-—swamp tupelo-redbay (104) Red alder (221)
Sugar maple-beech-yellow birch (25) Other southern types Black cottonwood-willow (222)
Sugar maple-basswood (26) Ashe juniper-redberry (Pinchot) juniper (66) Sitka spruce (223)
Black cherry—maple (28) Mohrs (“shin”) oak (67) Western hemlock (224)
Beech-sugar maple (60) Mesquite (68) Western hemlock-Sitka spruce (225)
Red maple (108) Southern scrub oak (72) Coastal true fir-hemlock (226)
Other northern types Southern redcedar (73) Western redcedar—western hemlock (227)
Northern pin oak (14) Cabbage palmetto (74) Western redcedar (228)
Gray birch-red maple (19) Sweetgum-yellow-poplar (87) Pacific Douglas-fir (229)
Black ash-American elm-red maple (39) — Atlantic white-cedar (97) Douglas-fir-western hemlock (230)
Hawthorn (109) Pondcypress (100) Port Orford—cedar (231)
Redwood (232)
Central forest region Tropical forest region Oregon white oak (233)
Upland oaks (Florida only) Douglas-firtanoak—Pacific madrone (234)
Post oak-blackjack oak (40) Tropical hardwoods (105)
Bur oak (42) Mangrove (106) Low elevations, interior
Bear oak (43) Cottonwood-willow (235)
Chestnut oak (44) Bur oak (236)
White oak—black oak-northern red oak (52) Interior ponderosa pine (237)
White oak (53) Western juniper (238)
Black oak (110) Pinyon-juniper (239)
Northern red oak (55) Arizona cypress (240)
Other central types Western live oak (241)
Black locust (50) Mesquite (242)
Yellow-poplar (57)

154
 Vegetation Classification p53

Table 6-3 (Continued)

Eastern Forest Cover Types Western Forest Cover Types

Yellow-poplar—eastern hemlock (58) South Pacific except for high mountains

Yellow-poplar—white oak-northern red oak (59)
River birch-sycamore (61) Sierra Nevada mixed conifer (243)
Silver maple-American elm (62) Pacific ponderosa pine—Douglas-fir (244)
Sassafras—persimmon (64) Pacific ponderosa pine (245)
Pin oak-sweetgum (65) California black oak (246)
Pitch pine (45) Jeffrey pine (247)
Eastern redcedar (46) Knobcone pine (248)
Canyon live oak (249)
Blue oak—Digger pine (250)
California coast live oak (255)

Source: SAF, 1980. Copyright Society of American Foresters. Used with permission.
“Type group.
’Type name.
“Type number.

The hierarchical levels of the classification are de- characteristic association between a typical group of
fined by three types of diagnostic species: character species of high fidelity.
species, differential species, and constant companion Subassociations, the level of the hierarchy below the
species. Character species are those that are largely re- association, are defined by differential species. These are
stricted to the unit of vegetation under consideration, species that do not have high enough fidelity or con-
and each level in the hierarchy has a list of character stancy to be character species for the association but
species. Io qualify as a character species for the associ- that permit subdivision of the association into two or
ation level, a species must have a narrow habitat distri- more subunits.
bution and be almost, if not entirely, restricted to that A problem with the definition of fidelity classes
particular association (i.e., the range of the species arises from variations in the adaptations of different
should more or less coincide with the range of the as- ecotypes (Chapter 16). A species that may have a fi-
sociation). Its degree of restriction to one association delity class V in one part ofits range may have a much
is referred to as its fidelity. Table 6-4 gives definitions lower fidelity in other parts.
of both presence and fidelity classes and defines the A full treatment of this and other aspects of the
various classes of diagnostic species. Figure 6—7 illus- Braun—Blanquet approach is beyond the scope of this
trates the concept of fidelity class. book, so fuller treatments in Shimwell (1971), Whit-
Character species are not necessarily dominants. taker (1973a), and Mueller-Dombois and Ellenberg
In fact, they can be species with low abundance and (1974) are recommended to the interested reader.
cover. Fidelity to the association (not occurring else-
where—fidelity classes III-V are required for a char- The Habitat-Type Approach of Daubenmire. A
acter species) and high presence value (present in a somewhat different approach to vegetation classifica-
very high proportion of the plant communities repre- tion that uses the entire plant community is used in
senting the association in question—presence class V the northwestern United States (Daubenmire, 1952;
is often required, but class IV may be accepted if the Daubenmire and Daubenmire, 1968). The vegetation
fidelity is high) are the key attributes. Of the two at- is classified primarily by differences in the overstory,
tributes, fidelity is more important than presence, and the resulting units, or wnions, being subdivided accord-
the presence or absence of a particular character ing to the dominant shrub or herb species. The com-
species is less important in the definition of the bination of overstory and understory unions defines
vegetation unit than the characteristic combination of “associations,” which are named after the dominant
character species. The association is recognized by a overstory species and the dominant or characteristic
156 CHAPTER6 — Ecosystem Classification

Tuble 6-4 Definition of Constancy and Fidelity and the union is said to be more responsive to local variations
Criteria for Differentiating Values of Species in soil moisture, chemistry, and microclimate.
The climax associations are used to define habitat
A. Presence or Constancy Classes
Percent of Plots in
types: physical environments or parts of the land-
Presence Class Which Species Occurs scape that will support particular climax plant asso-
ciations in the absence of disturbance. Habitat types
r <1 are believed to be a more practical classification tool
| 1-20 than associations because once the habitat type is de-
I] 21-40 fined, so is the seral sequence of plant communities
ll 41-60
and the suitability and productivity of the site for
different tree species. As a result, habitat types can
IV 61-80
be recognized regardless of whether the community
V 81-100 has been disturbed.
B. Fidelity Classes (see Figure 6-7) The habitat-type system of classification does
Fidelity Class V: Exclusive species. Species exclusively not incorporate soils data in the development of the
or almost exclusively restricted to a particular vegetation classification, and a given habitat type can include a
unit. variety of soil types. A description of soils is in-
Fidelity Class IV: Selective species. Species with a cluded in a description of the habitat types, and soils
strong preference for a specific vegetation unit but also information may be used to help characterize lower
found infrequently in other units.
levels in the classification such as phases. However,
Fidelity Class Ill: Preferential species. Species often oc-
the system is most commonly used at or above the
curring in other vegetation units but with their optimum
or maximum expression in one unit.
habitat type level and, consequently has traditionally
been a vegetation rather than an ecosystem type of
Fidelity Class Il: Companion species. Species without a
definite preference for certain vegetation units but that classification.
are frequently present on a particular unit. The explanation for the relative exclusion of soils
Fidelity Class |: Strangers or accidental species. Species data from the classification is that an adequate soils
that have a definite preference for other vegetation units classification had not yet been developed for the
but that are occasionally present; may be a relict from northwestern United States at the time the system
previous seral stage. was first developed. Although he noted that soil is a
C. Differentiating Values critically important ecological factor, Daubenmire
1. Character species: must have presence class >IV and believed that vegetation responds to differences in
fidelity class Ill. moisture, fertility, temperature, and aeration rather
2. Constant dominant: presence class V; mean species than to parameters such as color, texture, structure,
significance =3.0. depth, sequence of horizons, and other soil features
3. Constant: presence class V; mean species signifi- that are easily observed by the human eye. The latter
cance <3.0. were the basis for soils classifications available to
4. Important companion: presence class =Ill: fidelity him at the time he was developing his classification
class Il. scheme, which led Daubenmire to develop a vegeta-
Shimwell, 1971; Mueller-Dombois and Ellenberg, 1974.
tion-only classification: a classification of the biotic
potential of the land as expressed through the vege-
tation. Daubenmire believed that soils data should
be included as ecologically useful soils information
understory species (shrub or herb). Only climax asso- became available, and although the system remained
ciations are considered in the nomenclature, but seral largely a vegetation system, recent applications
variants of these are described. The authors of this of the method have involved an increasing compo-
system accept the polyclimax view of succession nent of soils data, and it is likely that this trend
(Chapter 17) and recognize climatic, edaphic, topo- will continue.
graphic, topo-edaphic, and zootic climaxes. The over- Habitat types are the basis for two different classifi-
story union at climax is said to reflect regional climate cation hierarchies: a floristic hierarchy and a landscape
rather than soil variations, whereas the understory hierarchy. In the former, habitat types are grouped into
 Vegetation Classification 157

Fidelity class for Distribution of a plant species

association B

Vv Exclusive
species

Nn
2
iS)
a
3 IV Selective
4 ;
s species
Ss

iS 3o

0 S$
oO
S)
Cg
Ss
iS)
: S|
Il Preferential 5
species DB
Y 6
1S()
a.
72)
o
fe
fe
eS
oO
[a4

I Companion
species

Accidental-species
for Association B;
character species
for Association C.

Association A Association B Association C

Environmental gradient

Figure 6-7 Diagram of fidelity classes. Relative species significance values are given because fi-
delity does not imply high absolute significance value.

series: groups of habitat types that have the same domi- graphic areas having characteristic climate and domi-
nant climax species. Series are successively grouped nant vegetation of a particular physiognomy. The
into subformations and formations on the basis of highest category in this hierarchy is the ecoregion, de-
physiognomic similarities of the dominant vegetation fined on the basis of major differences in understory
at climax. Figure 6-8 shows a topographic sequence of and overstory, which are thought to reflect major cli-
series in northwestern Montana and the relationship matic variations (Pfister, 1976; Pfister and Arno,
between habitat types and potential timber yield. 1980). Each region is characterized by a typical altitu-
In the landscape hierarchy, habitat types are used dinal sequence of 5 to 15 habitat types.
to define vegetation zones: geographical areas with a The habitat-type approach is being used through
uniform climate that supports the same climatic cli- much of the northwestern United States, where it has
max association (i.e., has the same habitat type) in gained acceptance as the ecological basis for forest
zonal ecosystems. Vegetation zones are grouped into management. A map and description of the ecore-
vegetation provinces on the basis of floristic similarities, gions of the United States is available (Bailey, 1976,
and these are grouped into vegetation regions: geo- 1978).
158 CHAPTER6 = Ecosystem Classification

Zz =
Ea &“anfianonghE tigaba
= Gi

a om *

2 . J
|| ste a
2 3 Timberline
3 a BS. ——————
s on LL

o ere
=

| = 3 |

‘aaa : lane
z a eri
2 5 eee a | Tsuga as
| S = reall Ui | Y mertensiana m &
: iF s 2 | 3 5 3S aa climax <5
es Pee i ea hl: Se Hs
Te ae Bqe eves
fp size
tg tS
iz icejg
ts |e bh12 |ian, ve
A Se aHt Als
$ pV TsuGA HETEROPHYLLA par
ae2) aef oan Nees. | r=2 | &Ls | THUJA PLICATA SERIES
4
! S , P, . } me Lak ee ABIES GRANDIS SERIES
E Kea
: sal A! | Ay
i &® | 2
| a | 4
ae
: PSEUDOTSUGA MENZIESII SERIES

l
PINUS PONDEROSA SERIES

grassland

Figure 6-8 Daubenmire’s forest habitat type classification. (A) Altitudinal distribution of forest
habitat series in northwestern Montana. The vertical lines show the elevational range of the tree
species. Dotted line, seral status; solid line, climax status. (B) Estimated production potential (ft
. . . . . ~ . . . 3

acre! year‘) of habitat types in western Montana. (After Pfister et al., 1977. Courtesy of the USDA
Forest Service and the authors.)

6.5 Ecosystem Classification popular for such situations over the past three decades
is ecosystem classification, in which the classes are ex-
In cases in which there is a diverse biota that is in equi- plicitly defined in terms of climate, soils, landforms,
librium with its physical environment, there is detailed and vegetation. Faunal and microbiological parame-
knowledge of the relationships between the compo- ters belong equally in such classifications but have re-
nents of this biota and between the physical environ- ceived less attention than they deserve, for reasons
ment and the biota, and the seral characteristics of such as lack of knowledge, lack of interest, and/or lack
particular ecosystems are known, classification on the of funds and human resources.
basis of the biotic community alone has proved to be a Ecosystem approaches to classification can be di-
very satisfactory approach. Conversely, where the vided into two main types: (1) those that focus on veg-
biota is less diverse, where individual species tend to etation-soil units (ecosystem types) using climatic
have broader ranges and to be less specific in their en- and/or climatic—vegetation relationships as a broad
vironmental tolerances (lower fidelity values), where framework for the classification (e.g., the biogeoclimatic
much of the vegetation has been recently disturbed, classification of British Columbia; also called an
and when this situation is coupled with an incomplete ecosystematic classification) and (2) those that either
knowledge of biotic-environment relationships, a focus on the physical environment, incorporating veg-
purely vegetation classification may prove to be unsat- etation only in the final stages of the classification, or
isfactory for many aspects of forest management. An that classify and map different parameters of the envi-
alternative approach that has become increasingly ronment separately and subsequently integrate them
 Ecosystem Classification 159

YIELD CAPABILITY CLASSES

Vv Ill
|
I
(very low) (moderate)
|
(very high)
|
|
|
ABLA, CACA {13 —6
ABLA, XETE
ABLA, LIBO
ABLA 4 MEFE
ABLA 4 GATR
ABLA 4 OPHO
ABLA CLUN
TSHE g CLUN
THPL CLUN
Abbreviated tree
species name 158
(Thuja plicata) ABGRg&g LIBO (6-3

PICEA Abbreviated understory

VACA
| :
species name (Clintonia uniflora)
PICEA @ CLUN 24 — 13
PSME 4 VACA

90%
-<—of range—>|
60 eal Hite
-SYAL 3 — mean mean no. if no. of
PIPO FEID iKey AMG) stockability site stands
factor trees

40 60 80 100 120 140 160 180

Yield capability (ft* acre~! yr~!)
(b)

Figure 6-8 (Continued)

by overlaying the final maps (e.g., the biophysical classt- cation developed by Krajina (1965, 1969, 1972) in
fication approach developed in Australia and Canada). British Columbia. It divides the province initially into
major climatic regions using Koppen’s classification of
climate (Table 16-1). Four climatic formations (E, D,
Biogeoclimatic Classification’ BSK, and C) and seven biogeoclimatic regions (subdivi-
Introduction to the System. Biogeoclimatic classi- sions of formations) are identified, primarily on the
fication is a hierarchical system of ecosystem classifi- basis of broad climatic features but also on the basis of
broad vegetational and soils characteristics that are in-
duced by climate. These regions are in turn subdivid-
This method of classification is given more detailed treatment than the
ed into biogeoclimatic zones, which are defined as
others because it is currently in operational use in British Columbia as the
geographical areas characterized by a mosaic of vege-
ecological basis for intensive forest management by government and indus-
tation types and soils, the character of which broadly
try foresters. The method is being applied in Alberta and Ontario. Biogeo-
climatic classifications have been proposed for the Hawaiian Islands reflects the regional climate; the spatial boundaries of
(Krajina, 1966) and for Japan (Kojima, 1979). a zone are defined by the dominant climatic climax
160 CHAPTER6 — Ecosystem Classification

vegetation on mesic sites that have zonal soils. The pacity, and no upslope drainage area (no slope seepage
term “mesic” refers to the site hygrotope (soil moisture water to augment precipitation inputs). In contrast,
status, also referred to as soil moisture regime, or hygric sites receive an abundant input of water in the
SMR). Zonal soils are those that have developed pri- form of soil seepage from upslope and often have good
marily under the control of regional climate and the soil moisture storage capacity. Plants on such sites
climatically determined vegetation. Understanding have access to far more moisture than would be sug-
the biogeoclimatic classification method requires an gested by local precipitation data, but the soils are still
understanding of the classification of hygrotopes. well aerated and are rarely saturated in the upper 0.5
to 1.0 m. Hydric sites are those that, because of soil
Classification of Hygrotopes or SMR. The term texture, soil organic matter, poor drainage, and/or
mesic in the biogeoclimatic classification system refers abundant inputs of slope seepage or spring water, have
to sites on which the moisture conditions experienced soils that are saturated almost to the surface for much
by plants are primarily under the control of the local of the year; often, they are semiaquatic. Soil moisture
climate. Xeric sites are those that are drier than would regime classes are illustrated in Table 6-5.
be expected from local precipitation data. This can be On xeric, hygric, and hydric sites, there is strong
because of rapid drainage of water (steep slopes; thin, edaphic and/or topographic modification of the cli-
coarse soils lacking incorporated organic matter; or matically determined availability of moisture to plants.
very deep coarse soils), low soil moisture storage ca- On mesic sites, there is relatively little such modifica-

Table 6—5 Soil Moisture Regime (SMR) Classes and Their Definition*
Code Class Description Primary Water Source

0 Very xeric Water removed extremely rapidly in relation to supply; Precipitation

soil is moist for a negligible time after precipitation
1 Xeric Water removed very rapidly in relation to supply; soil is Precipitation
moist for brief periods following precipitation
2 Subxeric Water removed rapidly in relation to supply; soil is moist Precipitation
for short periods following precipitation
3 Submesic Water removed readily in relation to supply; water available Precipitation
for moderately short periods following precipitation
4 Mesic Water removed somewhat slowly in relation to supply; soil may Precipitation in
remain moist for a significant, but sometimes short period of moderate- to fine-
the year. Available soil moisture reflects climatic inputs textured soils and
limited seepage in
coarse-textured soils
5 Subhygric Water removed slowly enough to keep soil wet for a significant Precipitation and
part of growing season; some temporary seepage and possibly seepage
mottling below 20 cm
6 Hygric Water removed slowly enough to keep soil wet for most of Seepage
growing season; permanent seepage and mottling; gleyed
colours common
i Subhydric Water removed slowly enough to keep water table at or near Seepage or permanent
surface for most-of year; gleyed mineral or organic soils; water table
permanent seepage <30 cm below surface
8 Hydric Water removed so slowly that water table is at or above soil Permanent water table
surface all year; gleyed mineral or organic soils

* More detailed descriptions and keys are given in the DEIF manual (Luttmerding et al. 1990) and in MOF
field guides to site identification and in-
terpretation.
From B.C. Ministry of Environment, Lands and Parks/B.C. Ministry of Forests 1998. . Field Manual for Describing Terrestri al Ec :
Management Handbook Number 25, Victoria, B.C.; www.for.gov.bc.ca/research/becweb/ akaick beak
 Ecosystem Classification 161

tion. Mesic sites can occur on broad flat ridges with to another and from one zone to the next. The ab-
fairly deep (1 to 2 m), medium-textured soil or on solute moisture status of each hygrotope in each sub-
midslopes where downslope drainage is balanced by zone is currently being quantified.
seepage inputs from above (but insufficient to make
the site hygric). Mesic sites can occur on rocky ridges Biogeoclimatic Zones and Subzones. Biogeo-
or on coarse soils if the forest floor has accumulated to climatic zones are identified and named after the cli-
sufficient depth to be able to hold much of the precip- matic climax vegetation on the mesic site type with
itation input for long enough for plants to have the zonal soils. The climax vegetation on hygric and xeric
opportunity to use it. Thus, mesic sites can occur al- sites within a zone reflects these soil moisture condi-
most anywhere in the landscape where the moisture tions and will be similar to the climax vegetation on
conditions experienced by the plants are determined the mesic sites of the next wetter and drier zones, re-
primarily by the local climate. They certainly are not spectively. For example, ponderosa pine is climax on
restricted to midslope positions, although mesic con- mesic sites in the Ponderosa pine (PP) zone in the
ditions are often best developed on midslopes. southern interior valleys of British Columbia but is
Xeric sites can also be found in various topograph- found only on xeric sites in the Interior Douglas-fir
ic positions. Normally an upper slope or ridge-top (IDF) zone, which is elevationally above the PP zone
type, xeric conditions can occur on lower slopes or and is therefore wetter and cooler. Douglas-fir is re-
valley bottoms, where deep, coarse-textured geologi- stricted to hygric sites in the PP zone, is the climatic
cal deposits elevate the rooting zone far above the climax dominant on mesic sites in the IDF zone, and is
slope seepage zone or the phreatic zone (above the restricted at climax to the xeric sites in the next wetter
groundwater table). Hygric conditions are nearly al- zone: the Interior Cedar Hemlock (ICH) (Figure
ways found on lower concave seepage slopes with 6-9A). Similarly, a single plant species or plant associ-
medium soil depth (the seepage zone is within reach of ation may be found in more than one subzone of a
the roots). Hydric conditions are also a lower-slope, zone and may even occur in more than one zone, but
valley-bottom condition but can sometimes be found its position on the topographic sequence will vary ac-
in poorly drained concave depressions on ridge tops. cording to the zone and subzone (Figure 6-9B). The
In exceptionally cool and humid climates, subhydric position of an association on the topographic sequence
conditions (bogs) may develop almost anywhere on will also be influenced by the soil parent material. On
the landscape. nutrient-rich materials, an association will frequently
Use of the hygrotope classification is central to the occur farther upslope (drier) than on nutrient-poor
biogeoclimatic classification. Each biogeoclimatic materials. Thus, each zone and subzone is a mosaic of
zone and subzone is a mosaic of different climax asso- vegetation types, characterized by particular vegeta-
ciations (or their seral equivalents) located at various tion types on particular sections of the topographic se-
positions on the landscape. The structure and compo- quence. The province of British Columbia is a mosaic
sition of these associations reflects both the regional of zones, a predictable sequence of which will be en-
climate and the local soil conditions, but the relative countered while traveling west to east or south to
importance of climate and soil as controlling factors north through the province or while ascending a
varies. On xeric, hygric, and hydric sites, soil condi- mountain anywhere in the province.
tions dominate, and the vegetation does not accu- Most of the biogeoclimatic zones are subdivided
rately reflect the regional climate. Only on the mesic into two or more subzones according to variations in
site do the vegetation and soil truly indicate the cli- overstory, understory, and soils that reflect variations
mate of the area. Consequently, the geographical ex- in climate. These variations are not sufficient to alter
tent of a zone or subzone is defined by the the climax dominant plant species on the mesic site,
composition and structure of the vegetation and the but they do result in significant variations in soil and
nature of the soil on the mesic site. vegetation that are of both ecological and manage-
The terms xeric, mesic, hygric, and hydric are mea- ment significance. The biogeoclimatic subzone is the
sures of relative moisture availability to plants (relative level of the classification that dictates overall resource-
to local precipitation). This means that the absolute use strategies. It is significantly more homogeneous
moisture tensions experienced by plants on the xeric than the zone, but the subzone is still not detailed
site, for example, will vary greatly from one subzone enough for site-specific decisions.
 Dominant plant species in three
biogeoclimatic zones, of B.C.
ponderosa pine Interior Interior
Douglas-fir Cedar-Hemlock
Drier

Ponderosa
Bunchgrass :
pine

= OQ z. Q Ponderosa Western hemlock

pine Western redcedar

Gradient
moisture
Soil

Western redcedar,
grand fir, or Western redcedar
Western hemlock
Douglas-fir

Weer 19 — 36 36 — 56 56 — 170
Average annual precipitation, cm Cooler
wetter
Drier, hotter Climatic gradient Wetter, cooler

eee: Bei ING

a4
3 S5
r= Lodgepole
= pine
Figure 6-9 Changes in the posi- E
tion of tree and minor vegetation ¥ eal
species on the landscape as climate E 2 ( gS
changes. (A) Changes in site location Zz is z >
of climax tree species in different bio. 6 ~ % gradient
Climatic
Increasing
elevation
geoclimatic zones in southern interior
British Columbia. In the hot dry cli-
mate, Douglas-fir is the dominant cli-
max on the hygric site type. As the cli-
mate becomes cooler and more 2
humid, this species is found as the cli- = 2 g=
max dominant on the mesic site, being a 3
displaced on the hygric site by western
redcedar, grand fir, and/or western E
hemlock. As the climate becomes ever &
wetter, hemlock becomes the climax £ 2
dominant on the mesic site, and Dou- 2 Es ay
glas-fir as a climax species is restricted ~ Z
Warmer
to xeric sites. (B) Similar change in
drier
topographic location of three tree Hygrotope
species and one shrub species along a
climatic gradient in coastal British Hygric Mesic
Columbia. (b)

162
 Ecosystem Classification 163
Table 6-6 Soil Nutrient Regime (SNR) Classes and Their Definition
Oligotrophic Submesotrophic Mesotrophic Permesotrophic Eutrophic Hypareutrophic
A Very poor B Poor C Medium D Rich E Very rich F Saline
a a a a ee ek a ee ee
Available excess salt
nutrients very low low average plentiful abundant accum.
Humus form Mor Moder Mull
A horizon Ae horizon A horizon Ah horizon
present absent present
Organic low (light medium (inter. high (dark
matter coloured) in colour) coloured)
content
C:N Ratio high moderate low
Soil depth extremely very shallow
shallow to deep
Soil texture coarse medium to
textured fine textured
% Coarse moderate
fragments high to low
Parent
material
mineralogy base-low base-medium base-high

Soil pH extremely- moderately slightly acid-

mod. acid acid-neutral mildly alk.
Water pH
(wetlands) <4-5 4.5-5.5 5.5-6.5 6.5-7.4 7.4+
Seepage temporary = permanent

From B.C. Ministry of Environment, Lands and Parks/B.C. Ministry of Forests 1998. Field Manual for Describing Terrestrial Ecosystems. Land
Management Handbook Number 25, Victoria, B.C.; www.for.gov.bc.ca/research/becweb/.

Topographic Sequences of Site Types Within a shown in Figure 6-10 are different in zones with ex-
Subzone. Within each subzone, variable topogra- tremes of temperature or low rainfall.
phy results in gradients of soil moisture (SMR) and The magnitude of the vegetation discontinuities
fertility (classes of soil fertility are called trophotopes, or that occur along a topographically induced gradient
soil nutrient regimes (SNR); see Table 6-6). These of moisture and fertility varies from subzone to sub-
environmental gradients are associated with charac- zone. This causes a variation between zones in the
teristic patterns of vegetation and soils that are repeat- ease with which the vegetation can be subdivided
ed throughout the subzone. The variation in into recognizable, discrete classes. Subdivision tends
vegetation along the physical gradients is typically dis- to be easy in zones with steep topographic, soils, and
continuous, which permits the identification of indi- climatic gradients and in which climatic climax veg-
vidual plant communities that have characteristic etation is common. In these areas, the zones of
floristic composition and structure. Figure 6-10 shows transition between adjacent classes are narrow. Sub-
a characteristic topographic sequence within one of division into discrete classes is more difficult in areas
the subzones of coastal British Columbia. The broad that lack steep environmental gradients or in which
structural features of this sequence can be observed in fire or animals (including humans) have created
other subzones in many parts of the province, but of widespread seral conditions. In such areas, the tran-
course the floristic details vary from subzone to sub- sition zones may be as broad as the areas of individ-
zone. Several features of the vegetation structure ual vegetation classes.
TYPE OF OVERSTORY
VEGETATION
Densely stocked with
small trees. Small, thin
No trees —______ crowns permit penetration
of substantial amounts of
light, favoring shrub growth.
Mainly conifers.
Very N
SOIL MOISTURE Xeric —
REGIME Peon Ss
Xeric Densely stocked with
aS medium-sized trees with
ww well developed crowns and
little light penetration.
> Mainly conifers. SS
Non-forested
ecosystem with lichens =".
\ x
and mosses orgie ‘ Variably stocked with
Dry-land Mesic trees of varying size and
shrubs and Ns age. Large gaps in canopy
mosses ~ provide sufficent light for
ae xx shrub and herb growth.
T YPE OF \ \ Conifers and some
UNDERSTORY \ \,___ hardwood species.
VEGETATION "
Hygric
= sy
ic Variably stocked
with deciduous and
Mba coniferous species.
\ 7 \ Abundance of the
\ Se former favors
shrub and herb growth.
\ SS
\ ~~ Sub-hydric

Soil legend

: Bedrock

Basal till Herbs, ferns,

(compacted) mosses, and some
Ablation till or moist-land shrubs
gin colluvial veneer Bs

Y Glacio-fluvial deposits
L4

AN Temporary seepage

Wet-land shrubs, herbs,

Permanent seepage ferns, and mosses

Organic surface soil horizon

Figure 6-10 Variation in vegetation structure, soils, and soil moisture along a minor topo-
graphic sequence in British Columbia.

164
 Ecosystem Classification 165

Methods Used in Biogeoclimatic Classification. species found in the subzone), a grid showing the dis-
For describing and identifying zones and subzones tribution of the major plant associations, and a grid
and subdividing the subzones, the vegetation is ana- recommending (1) the major tree species that should
lyzed using a slightly modified Braun—Blanquet be favored in regeneration and (2) postlogging slash-
method, and the soils and landforms are analyzed for a burning treatment. Other interpretations of the grid
variety of parameters using standard soil and landform have been developed.
description and classification methods. Particular at- Examples of the use of biogeoclimatic classifica-
tention is paid to the nature and condition of the for- tion in forest management can be found in Klinka et
est floor (Klinka et al., 1981). The combination of the al. (1980a,b). More details of the system can be found
resulting environmental (soils and landform) and veg- in Krajina (1969), Kojima (1981), and Pojar et al.
etation data forms the basis for the definition of site (1987). (The B.C. Ministry of Forests web site pro-
types: landscape units that are uniform in climate, veg- vides details of this system; see www.for.gov.bc.ca/
etation, and soil conditions. All site types within a sub- research/becweb.)
zone (1.e., site types that have the same climate) that
have the same climax vegetation potential, irrespective Overall Structure of the Biogeoclimatic Classifi-
of variations in soil conditions, are grouped as a site se- cation System. The biogeoclimatic classification
ries. Where site series in different subzones or zones approach subdivides a region into climatic classes ac-
(i.e., in different climatic areas) have the same climax cording to Képpen and subdivides these into biogeo-
vegetation potential because soil differences compen- climatic zones according to the climatic climax
sate for climate differences, they are grouped as a dominant vegetation on mesic sites with zonal soils.
site association. Site types, series, and associations are Zones are subdivided into subzones according to
generally defined and named in terms of the climax floristic and structural differences in the plant com-
condition (which may be determined by climatic, munity and differences in soils. Within subzones, pat-
topographic, or edaphic factors), but in climatic re- terns of site types are identified along topographic
gions where the disturbance of wind or fire is very fre- sequences. These topographic sequences of site types,
quent and/or where the rate of succession is slow (e.g., which are associated with gradients of soil moisture
the boreal region), some of the ecosystem types recog- and fertility, are summarized in the form of edaphic
nized may be seral. erids, which then form the ecological basis for the de-
The ecosystem types found along minor topo- velopment of guidelines for a variety of silvicultural
graphic sequences within a subzone reflect variations decisions at the stand level. The vegetation compo-
in soil moisture and fertility (see Chapter 13). Other nent of the ecosystem is analyzed according to the
environmental determinants (e.g., temperature, soil Braun—Blanquet system, and standard descriptions of
aeration) also help to determine the location of a plant soil and landform are prepared. Because the system is
association on the sequence, but the former two fac- hierarchical and follows the basic principles of system-
tors are thought to be particularly important, and they atic taxonomy, this system is said to be an eco-
play a major role in determining the productivity of systematic classification: it is a natural, taxonomic
trees. The relationship of these two factors to the dis- classification of ecosystems.
tribution of plant associations and the productivity of The hierarchical biogeoclimatic classification is
tree species can be broadly summarized in a two- summarized in (Figure 6-12).
dimensional edaphic grid (Remezov and Pogrebnyak,
1969) with hygrotope soil moisture regime along one
Biophysical Classification
axis and trophotope (soil nutrient regime) along the
second axis. These grids have proved to be a useful aid Biophysical classification is an approach that has been
in the interpretation and application of the biogeocli- developed and used widely in Canada for broad classi-
matic classification for forest management at the local fication of large areas. As a logical development from
site or stand level. the earlier physiographic approach, it has itself
Figure 6-11 shows a set of three grids for one sub- evolved and is now generally referred to as ecological
zone in British Columbia: a grid of productivity for a classification (Thie and Ironside, 1976). However, the
tree species (one of these is prepared for each tree approach is really a series of inventories of climate,
166 CHAPTER6 — Ecosystem Classification

Soil nutrient regime

Very Very Very Very Very
Poor Medium Rich _ rich poor Poor Medium Rich _ rich poor Poor Medium Rich
poor
Excessively Lodgepole pine
Avoid slash burning
drained
Very rapidly
drained
Rapidly (or ee arenepi
drained Douglas-fir-
= hemlock
W ell draine
ained Moss

Moderately (sword fern)-

Hemlock
well drained
Imperfectly
regime
moisture
Soil
drained
Foam flower- |_ Kor Douglas-firy a
Poorly
drained sword fern-
cedar
Very poorly
drained

(b) hi

VA spot burning only [] high intensity burn (i) light intensity burn LY medium intensity burn

Figure 6-11 Edaphic grids, using the dry subzone of the coastal Western hemlock biogeo-
climatic zone in British Columbia as an example. (A) Productivity and shade tolerance of Dou-
glas-fir for various combinations of soil moisture and fertility. The size of the diagrammatic trees is
proportional to productivity. Open trees indicate that regeneration is light-demanding. Black trees
indicate that regeneration is shade-tolerant. Dotted lines are isoclines of tree productivity. (B) Dis-
tribution of major plant associations in the subzone with respect to soil moisture and fertility. (C)
Major species recommended for regeneration, and recommended slashburning policy. (After Kra-
jina, 1969, and Klinka, 1977. Used with permission of Department of Botany, The University of
British Columbia, Ministry of Forests, and the authors.)

soils, landforms, and vegetation, which are then syn- has largely been done from aerial photographs with
thesized into a series of environmental categories, limited ground checking because of access problems
rather than a true taxonomic ecological classification. and the vast extent of the area being mapped.
The biophysical approach generally involves a The biophysical approach has been proved to be
team of specialists (a geologist, a pedologist, and a enormously successful for many of the purposes for
plant ecologist). When this team works together with which it has been used, but it has not proved to be the
close cooperation and integration, the results should best classification method for forest management. It
be broadly comparable with the biogeoclimatic ap- has tended to be more of a regional than a site-specif-
proach. However, in practice, the geology, soil, and ic system, although there are many examples of its use
vegetation inventories are often undertaken indepen- for specific local resource development and land man-
dently, and this can result in a number of problems agement purposes.
during the final synthesis because ofvariations in scale
of mapping and size of mapping unit. The method has
been developed and applied particularly in large-scale SUMMARY
reconnaissance inventories of the environment for Forest management is being practiced in a climate of
such purposes as hydroelectric schemes and pipeline public opinion that is increasingly intolerant of “mis-
projects in Canada and the United States. Mapping takes.” The need for resource managers to achieve their
 Take-Home Message 167

VEGETATION
foundation will generally result in a house that is unsatis-
ZONAL
CLASSIFICATION CLASSIFICATION factory in one or more ways. Similarly, a sound ecological
foundation in the form of a site classification does not
Plant: Biogeoclimatic: guarantee that forests will be managed sustainably, but it
class formation
order region
is unlikely that sustainability will be achieved without it.
alliance zone Various approaches to classification have been devel-
association subzone oped, reflecting varying physical and biological condi-
sub association variant tions, variable access to and basic knowledge of the
environment, and variable applications for which the
classifications have been developed. There is no single
best approach. Each particular situation defines a best ap-
proach for the prevailing set of circumstances. However,
there has been a historical trend in the development of
SITE forest classifications from broad schemes that use very
CLASSIFICATION general parameters of the physical environment to more
detailed schemes that incorporate all the major compo-
nents of the ecosystem. It has generally been found that
Site:
association prediction of ecosystems is more successful when based
series on an ecosystem classification than on a classification that
type incorporates only one or a small number of ecosystem
parameters. Within the ecosystem approach, classifica-
tion either may work with entire ecosystems (e.g., the
biogeoclimatic approach) or may classify different com-
Figure 6-12 Diagrammatic summary of the biogeocli- ponents of the ecosystem independently, combining sub-
matic classification of British Columbia (based on Pojar et groups of these components to define landscape units
al., 1987). (e.g., the biophysical approach). The merits of the two
approaches must be judged by their utility for resource
management. Undoubtedly, both approaches will find
favor under appropriate circumstances, but it is probable
public statements of intent to practice “good” forest
that the former will achieve greater success than the lat-
ecosystem management has never been greater. Thus,
ter as we move from a focus on timber management to
foresters need to be successful in achieving the objectives of
the management of forest ecosystems and forest land-
their management.
scapes for multiple values.
Success in almost any endeavor in life involves an
ability to make reliable predictions about the object of
our intentions. The better our ability to predict some-
thing, the greater the probability of success in any en- TAKE-HOME MESSAGE
deavor in which it is involved. Thus, to be successful,
foresters must be able to make accurate predictions. A key ingredient in the evolution of forestry from the ad-
Making accurate predictions in an ecologically vari- ministrative to the ecologically based stage and from
able forested landscape is difficult unless we can explicit- there to social forestry is the development and application
ly identify and account for variations in ecological of ecological site classification as the ecological founda-
conditions that determine ecosystem productivity, re- tion for management. Only when we explicitly recognize
silience, and response to management activities. Success in the spatial variations in the forest ecosystem will we be in
predicting the response offorest landscapes to natural and man- a position to predict accurately the temporal changes in
agement-induced disturbance is most effectively achieved ecosystems, with or without management interventions.
through ecological site classification. Such a classification also Different intensities of management, different re-
facilitates the development of reliable generalizations source values to be sustained, and different types of forest
based on experience and research. will suggest different types of classification, and relatively
It is difficult to imagine successful, sustainable man- simple, single-factor systems will sometimes suffice.
agement of forests for multiple values unless this is based However, public demand for the management of a wider
soundly on ecological site classification. A good, strong range of ecosystem values suggests that the trend toward
foundation does not guarantee that a house will have good ecosystematic classification will continue. With increas-
architecture and pleasing proportions and will perform ing interest around the world in the biogeoclimatic clas-
its intended functions well. However, an inadequate sification of British Columbia, it seems that this approach
168 CHAPTER 6 — Ecosystem Classification

to forest classification will become more widely applied . What are the key features of the Finnish system?
in the future. How does it differ from the U.S. national vegetation-
based forest classification?
STUDY QUESTIONS . How does habitat-type classification differ from bio-
geoclimatic classification? What are habitat series,
1. Why do we classify things? and what are site series?
. Why is ecosystem classification important for sus-
tainable forest management? . What is an edaphic grid?
. What is a biogeoclimatic zone? How do we recognize . From what you learned in this chapter and in
different subzones in a zone? Chapter 5, define SMR and SNR.
4. What is a character species? BE: What is the relevance of climatic classification for
. What is a climatic diagram? ecosystem classification?
6. Describe dominance-type forest _ classification. Ls What is the role of an ecologically based approach to
Where is it most widely used? soil classification for ecosystem classification?
 Ecological Role
of Solar Radiation

7.1 Introduction The term “light” is used as a synonym for solar radia-
tion in this chapter because many of the ecological ef-
Solar radiation is the major source of energy for life, fects of solar radiation are the result of wavelengths in
and consequently photoautotrophs are the dominant the visible part of the spectrum.
producers in the trophic web of most ecosystems.
The ecological role of solar radiation is broader than
merely the provision of energy, however. Its fate as it 7.2 Physical Nature of Solar Radiation
passes through the atmosphere and at the earth’s sur- and Its Variations in Time and Space
face plays a major role in determining world tempera-
tures, climates, and weather patterns. Variation in the Before examining the ecological role of and the adap-
wavelength of radiation in the visible part of the solar tations of organisms to solar radiation, we review
spectrum (/ight) gives rise to the visual sense of color briefly its physical nature; how it varies geographically,
in animals, and this has led to a wide variety of adapta- daily, and seasonally; and how it determines the ener-
tions in both plants and animals. These organisms gy balance and temperature of the world.
have evolved the use of color to provide protection Because of its high temperature, the sun is continu-
against sunburn, for display purposes (to repel ene- ously emitting vast quantities of energy in the form of
mies or attract mates and beneficial organisms), and to electromagnetic waves of various lengths. These form
provide camouflage for security against predators. a continuous spectrum from very-short-wavelength
The continuing alternation of day and night provides gamma rays (3 X 10° p, a micron being one thou-
an environmental clock that determines patterns of sandth of a millimeter) to medium-wavelength infrared
physiology and behavior, and the seasonal variation in rays (up to 5 p1). Light is those wavelengths between ap-
the relative length of day and night provides an envi- proximately 0.39 and 0.76 p (i.e., 390 to 760 nm). Most
ronmental calendar that schedules the life histories of solar radiant energy is emitted at wavelengths between
most of the organisms on Earth. The intensity of light 0.4 and 2.0 p (visible light and infrared). Nearly 50% is
regulates rates of activity and behavior patterns in in the visible part of the spectrum; infrared accounts
some animals, influences the pigmentation of both for most of the rest (Figure 7-1).
plants and animals, and can affect the morphology of As solar radiation passes down through the atmos-
plants. Light also acts as one means by which both an- phere, it experiences changes in both quantity and spec-
imals and plants orient themselves in space. tral composition. Clouds and water vapor reflect, scatter,
Solar radiation thus is the ultimate source of most or absorb radiation of all visible wavelengths more or
life and a major determinant of the physiology, mor- less equally, with the result that clouds, overcast skies,
phology, behavior, and life history of most organisms. and very humid skies look white. Atmospheric dust also

169
170 CHAPTER 7 Ecological Role of Solar Radiation

Visible light
0.4 0.45 0.5 0.57 0.61 0.7

|
Yellow Orange |

Radiation reaching the

Solar radiation at outer atmosphere
sea level (clear day)

Intensity Solar radiation at

Radiation
of
sea level (cloud cover)

Solar radiation at
ground level beneath
vegetation cover
Skylight

Vv

0.3 0.4 0.5 0.6 O:7 30.8 09 AeO Zao) 10.0

Wavelength of radiation, microns

Figure 7-1 Change in the quantity (shown on the graph as Intensity of Radiation, also called
Spectral Irradiance) and quality of solar radiation as it passes through the atmosphere on ei-
ther clear or cloudy days and penetrates through a canopy of vegetation. The spectral compo-
sition of skylight (radiation from a clear blue sky) is also shown. This graph expresses radiation in-
tensity as the energy per micron of wavelength x (wavelength’), which causes the line to increase
from UV to IR wavelengths. In fact, the greatest energy levels are found in the visible wavelengths
(approximately 0.5 1). If the values in this graph are divided by (wavelength’), then the true energy
distribution can be seen—strongly peaked in the visible, with low values in UV and IR. (Modified
after Gates, 1965. Reproduced by permission ofthe Ecological Society ofAmerica and D.M. Gates.)

absorbs, reflects, and scatters solar radiation, but the also look red because of the preferential scattering of the
longer visible wavelengths are scattered more than the longer wavelengths.
other visible wavelengths, so very dusty atmospheres Ultraviolet and infrared wavelengths are largely
(heavily polluted air over urban and industrial regions lost from solar radiation as it passes down through the
and very smoky air after forest fires or slashburns) have a atmosphere because of a number of absorption
brownish or reddish tint. Molecules of atmospheric processes. Ultraviolet is largely absorbed by the layer
gases, however, scatter the shorter wavelengths more of ozone (O;) that is present in the upper atmosphere, !
than the other visible wavelengths, with the result that whereas infrared is absorbed by carbon dioxide gas
clear, clean skies look blue. Under these atmospheric
conditions, the sun becomes red when it is close to the
‘Reduction in the ozone layer caused by air pollution is causing an increase
horizon because of the scattering of the shorter wave- in the quantity of ultraviolet wavelengths reaching the earth’s surface.
lengths, whereas the sky on each side of the sun remains This has negative implications for human health (sunburn and skin can-
blue. However, if the atmosphere contains a lot of dust, cer), as well as for aquatic and other terrestrial organisms that are unable
then the sky on each side ofthe sun near the horizon will to protect themselves from the damaging effects of these wavelengths.
 Physical Nature of Solar Radiation and Its Variations in Time and Space ya

and water molecules. Because of this filtering action 12 hours long and there is often a lot of cloud. The
by the atmosphere, the solar spectrum is significantly greatest amount of solar radiation is experienced in
depleted in most wavelengths other than the visible summer in the middle latitudes, where clear, clean,
ones by the time it reaches the earth’s surface. It is low-humidity skies are combined with moderately
therefore not surprising that so much of the ecological long days. It comes as a surprise to some that the total
role of solar radiation (energy source, sight, color, etc.) growing season solar radiation energy at high latitudes
is related to these visible wavelengths. The other can compare very favorably with lower latitudes be-
wavelengths are also very important, of course, and we cause of the long days and clear skies, conditions that
should not overlook the important ecological role of result in the highly productive summer agriculture
ultraviolet and infrared radiation. that can be conducted at high latitudes.
When solar radiation reaches the earth’s outer at- The discussion of radiation energies so far has as-
mosphere, it has a total energy of approximately 8.4J sumed a flat surface at right angles to the sunlight.
cm™~ min” at a surface at right angles to the sun’s rays. However, both slope and aspect can either increase or
This quantity, which is called the solar constant, varies decrease the radiation that reaches the ground, de-
slightly with the seasons (approximately 3%) because pending on the location and the time of year. Aspect
the earth has an elliptical orbit around the sun. As the and slope also have an important effect in determining
radiation passes through the atmosphere, it is reduced the distribution of solar intensities throughout the
in intensity by reflection, absorption, and scattering, day. Easterly aspects will receive maximum intensities
and by the time “sunlight” reaches sea level at midday in the morning, whereas westerly aspects experience
in summer at latitude 40°, it is normally reduced in in- peak illumination in the afternoon. Midday is the time
tensity by between 12 and 50%. During periods of of highest light intensity for northerly and southerly
heavy cloud cover, it is reduced much more. aspects. The effects of easterly and westerly aspects is
Reduction in the intensity of radiation is greatest amplified by increasing slopes.
in areas where the air is very humid, where there is The effect of aspect on radiation intensities is
frequent cloud cover, where and when the sun is low modified somewhat by atmospheric conditions. On a
in the sky, and where the air is very turbid from dust or clear, sunny day, approximately 5% of the radiation
air pollution. Cities tend to have 10 to 25 times as that reaches the ground is skylight (scattered sunlight),
much dust and air pollution, 15 to 20% less total solar whereas under average conditions, the value is com-
radiation, and 5 to 10% more clouds than do compa- monly approximately 17% (Collier et al., 1973). On
rable rural areas (Landsburg, 1958). The importance completely overcast days, it is 100%. In clear, sunny
of atmospheric turbidity is demonstrated by the obser- weather, a northerly aspect in the northern hemi-
vation that during one winter, the city of Chicago re- sphere may receive much less solar radiation than a
ceived only 55% as much radiant solar energy as the southerly aspect, whereas on a completely overcast
less industrialized city of Madison, Wis., to the west day, both aspects will receive virtually the same quan-
(Trewartha, 1954). Smoke from wildfires and slash- tity. Thus, the effect of aspect is most apparent in clear
burns in western Canada in the summer and fall of weather and cloudless regions and least apparent
1950 was implicated in a 54% reduction in normal lev- under overcast conditions and in cloudy regions.
els of solar radiation in the city of Washington, D.C., Similarly, the light intensity that reaches the
more than 3000 miles to the east, and a noticeable ef- ground on a clear, sunny day in a small gap in a tropical
fect on solar radiation was recorded in Europe (Lull, forest (e.g., 0.1 to 0.5 ha) will be the same on the south
Look): and north edges of the gap, whereas under similar
The total amount of radiation reaching the earth’s weather conditions, a seedling at the northern edge of a
surface does not depend only on the attenuation of ra- similar gap in a northern temperate or boreal forest will
diation by the atmosphere. It also depends on the receive much more light than a seedling growing at the
number of hours of daylight. For example, maximum southern edge. Under cloudy conditions, the light in-
solar radiation in the summer in Alaska is only approx- tensity at the north and south edges will be similar.
imately 2 Jcm” min‘! but because the day is 24 hours Some of the solar radiation that does not reach the
long, daily totals of 1255 to 2092 Jcm™ day” are accu- earth’s surface is reflected back to space and plays no
mulated. This is sometimes greater than the daily ac- further role in the earth’s ecology. However, much of
cumulations in the tropics, where maximum solar it is absorbed by the atmosphere, and the resulting at-
radiation may be 6.7 Jcm’ min", but the day is only mospheric warming contributes to the determination
172 CHAPTER 7 Ecological Role of Solar Radiation

of world climates. Uneven heating of different parts of flected or absorbed and then reradiated back to Earth
the atmosphere results in convection currents and at- by the atmosphere. This explains why cloudy days in
mospheric stirring that produce cumulus clouds, help winter can be warmer than clear days. The clouds pre-
regulate atmospheric CO, levels (see Chapter 5), dis- vent much of the reflected radiation from escaping
perse air pollution, and contribute to the determina- from the atmosphere by returning it back to the surface.
tion of regional weather patterns. The solar radiation that is not reflected from the
Once solar radiation has reached the surface of the surface of the earth is absorbed by soil, rock, water, or
earth, it is either absorbed or reflected. The percent- living organisms. The absorbed energy may raise the
age of the incident radiation that is reflected is called temperature of the absorbing material, evaporate
the albedo, and it can vary from as much as 95% (deep, water from it, or, in the case of green plants, be incor-
fresh snow) to as little as 3% (deep oceans and forests) porated into high-energy chemical molecules in the
(Table 7-1). Albedo measures the overall reflectance process of photosynthesis. If the absorption of energy
of a surface, but the various wavelengths of sunlight results in an increase in the temperature of an object,
are not normally reflected equally, which gives rise to then some of this energy will be reradiated. In fact,
the variable color of surfaces. Snow reflects visible most of the visible portion of the solar spectrum that
wavelengths uniformly, giving a white color, and the passes through the atmosphere is absorbed by terres-
green color of forests results from their preferential trial or aquatic surfaces, raises their temperature, and
reflectance of green wavelengths. Invisible wave- is subsequently reradiated. However, this reradiated
lengths are also reflected differentially. The variation energy is at much longer wavelengths than the incom-
in ultraviolet reflection by flowers is “visible” to and ing radiation. Peak terrestrial radiation is at 1.0 pm
important for pollinating insects, whereas both the re- (infrared) compared with peak solar radiation at 0.5
flection and the emission of infrared wavelengths by pm (blue-green); the atmosphere is almost opaque to
plants varies with their physiological condition. This these longer wavelengths (cf. Figure 7-1), which are
is the basis of methods for detecting the effects of dis- consequently absorbed. ‘This warms the air, which
eases and insects on forest and agricultural crops by then reradiates some of the energy back to Earth.
aerial photography (remote sensing) (e.g., Puritch, As a result of these processes, the atmosphere acts
1981; Sabins, 1978). The very low albedo values of in much the same manner as the glass of a greenhouse,
forests show that they are among the most efficient which is transparent to visible solar energy but opaque
radiation-absorbing surfaces on Earth. to reradiated infrared energy, which is consequently
Of the solar radiation that is reflected at the earth’s trapped. A greenhouse acts as a solar energy accumu-
surface, some is lost to space and the rest is either re- lator, maintaining itself at a much higher temperature
than its surroundings. By creating a “greenhouse ef-
fect,” the atmosphere (mainly atmospheric H,O and
Table 7-1 Albedo Values for Various Surfaces CO;) accumulates solar energy and thereby plays a
Surface Albedo, % Reflectance critical role in maintaining the temperature of our
ecosystems within a range that is suitable for existing
Fresh snow cover 75-95 forms of life: a range that maintains that critical medi-
Old snow cover 40-70 um for life, water, in a liquid rather than a solid or
Sand dunes, ocean surf 30-60 vapor form. The growing recognition of the impor-
Sandy soil 15-40
tance of atmospheric O; and CO, concentrations for
the maintenance of life on Earth has led to increasing
Meadows and fields 12-30
concern over the possible long-term effects of the
Fresh grass 26 changes that we are causing in atmospheric chemistry
Dry grass 15-25 (discussed in Chapters 5 and 8). Figure 7-2 summa-
Dark cultivated soil 7-10 rizes the energy balance of the earth’s surface.
Woodland 5-20 Solar radiation varies not only geographically but
also with time. The rotation of the earth on its axis
Forest 3-10
gives rise to alternating light and dark periods, and the
Geiger, 1965. Copyright Harvard University Press. Used with permis- intensity of radiation varies during the day as the dis-
sion. tance that the radiation travels through the atmos-
 ‘ SHORT WAVE RADIATION
Incoming solar
radiation

LONG WAVE
Outgoing reflected RADIATION
Outer
space
radiation

Emission
Absorption by CO, and H,O

by O;
Stratosphere

Reflected
by clouds

13
Net absorption by clouds}
CO, + H,O0

Absorption SS ee
by H,O, a :
dust and
haze Reflected
Troposphere from
clouds
Reflected
and scattered
by dust and
haze

Direct
sunlight

We
from soil
vegetation
and
water

Biosphere

Figure 7-2 Radiation balance at the earth’s surface. The importance of long-wave radiation is
apparent. The numbers show the relative magnitude of different radiant transfers, with incoming
short-wave radiation equal to 100. (After Schneider and Dennett, 197). Redrawn by permission of the
Royal Swedish Academy of Sciences and S.H. Schneider.)
174. CHAPTER7 — Ecological Role of Solar Radiation

phere changes. The long atmospheric path length to have infrared (IR) sensors that enable this type of
early and late in the day results in lower radiation in- snake to detect and strike at living enemies or prey in
tensities than at midday, when the path length through the dark, and IR sensors are thought to be important
the atmosphere is shortest. The duration of the daily for some insects. Certain wood-boring beetles that use
light period (photoperiod) also varies because the plane freshly fire-killed timber as a substrate for their larvae
of rotation of the earth around the sun is different are thought to be able to detect forest fires many miles
from the plane of rotation of the earth around its axis. away using these sensors, and other phytophagous in-
When the earth faces the sun squarely with the equa- sects may possibly use IR sensitivity to detect physio-
tor closer to the sun than any other point on earth (as logically weak plants (cf. human use of IR imagery in
it does at the spring and autumnal equinoxes: March remote sensing). Insect parasites and predators may
21 and September 23, respectively, for the northern use IR detectors to locate concealed living prey.
hemisphere’s spring and fall), the day length and night Ultraviolet (UV) wavelengths are also detected by
length are equal at 12 hours all over the earth. Three some animals. Many insects use the variation in
months after the March equinox, when the earth has UV reflectance for plant recognition, and nectar-
progressed one fourth of its way around the sun, it gathering insects use patterns of UV reflected from
reaches the point at which the Tropic of Cancer is the flowers as a guide in their search for nectar.
closest point of the earth to the sun (June 21: the Much of UV radiation from the sun is absorbed by
northern hemisphere’s summer solstice). Day length the ozone layer in the upper atmosphere, and in the
in the Arctic becomes 24 hours, whereas in the absence of air pollution that damages the ozone layer,
Antarctic there is perpetual night. Equatorial areas UV comprises only approximately 2% of the solar ra-
continue to receive approximately 12 hours of day- diant energy at the earth’s surface. The cuticle of
light, whereas in the temperate regions, the nights be- plants is largely opaque to these wavelengths, and they
come short or long depending on the hemisphere. As are absorbed by the cell sap, so it is not surprising that
the earth progresses further around its solar orbit, it UV plays no known essential role in plant physiology.
reaches the September equinox, when day length re- Cells not protected by a cuticle, by layers of other
turns to 12 hours over the whole earth; 3 months later, cells, or by UV-absorbing pigments can suffer injury
on December 22, the northern hemisphere’s winter from the high photochemical energy levels of these
solstice is reached. Arctic nights become 24 hours wavelengths. Algae, fungi, and bacteria all are sensitive
long, whereas perpetual day occurs in the Antarctic; to UV, a phenomenon that is given practical applica-
the northern temperate areas experience winter day tion in the use of UV radiation to sterilize surfaces and
lengths, whereas southern temperate regions have to kill microorganisms.
long summer days. UV radiation induces a reversible inhibition of
growth in plants, either by destroying the auxins
(growth-regulating hormones) that control cell divi-
i Ecological Effects of Variations in sion and enlargement or by affecting the ability of a
the Spectral Quality of Solar Radiation plant to respond normally to auxins. The very short
a eS growth form and stunting that are characteristic of
Some of the evolutionary adaptations to solar radia- many alpine plants have sometimes been ascribed to
tion reflect the alteration of the solar spectrum by the higher exposure to UV at high elevation, but such a
earth’s atmosphere (Figure 7-1). The wavelengths relationship has not been definitely established.
least affected are those between 0.4 and 0.7 pw (visible Whereas the quantity of UV in direct sunlight in-
light). These are the wavelengths that are absorbed by creases with increasing altitude, the amount of UV in
the pigments of plants and that are perceived by the skylight decreases with altitude. High-elevation areas
eyes of most animals. Some plant functions can be are sometimes cloudier than low-elevation areas,
completed satisfactorily in light depleted in some of which results in greater atmospheric UV absorption;
these wavelengths, but most plants grow best when ex- consequently, the increase in total UV radiation with
posed to the full visible spectrum (Daubenmire, 1974). increasing elevation is not always very great. Also
Some animals have the ability to detect and may even alpine plants have become adapted to the somewhat
be able to “see” (i.e., form an image on a retina) wave- higher levels of UV. They have thicker cuticles and
lengths other than visible light. Pit vipers are known more UV-absorbing purple anthocyanin pigment, and
 Ecological Effects of Variations in the Intensity of Solar Radiation i795

therefore the living cells may not actually experience than for conifer forests with opaque needles, which act
more UV than do the cells of a lower-elevation plant more like neutral density filters, reducing all wave-
(Caldwell, 1968). lengths equally.
The production of protective pigments by plants is In addition to changes in the balance of visible
a common phenomenon. Just as human beings devel- wavelengths, plants with translucent leaves cause a
op a suntan to protect skin cells from damage (sun- change in the ratio of red to far-red wavelengths: the
burn) by photochemically active short-wavelength zeta value. Changes in zeta value are known to have a
solar radiation, so do plants protect their cells by the significant effect on plant physiology (see Section 7.5),
production of anthocyanin pigments. It is a common and the slow growth of plants in the understory of
observation that the fleshy, pale green stems and forests is sometimes as much the result of changes in
leaves of many plants grown indoors over the winter the red to far-red ratio caused by the overstory canopy
rapidly develop purple, red, or brown pigmentation if as it is a response to the reduction in light intensity.
placed in direct sunshine in the spring. Endler (1993) reviewed the color of light in forests
UV radiation produces an important effect when it and its ecological implications.
falls on fatty substances such as sterols. It changes One of the best known and most easily perceived
some of them into vitamin D, an important substance ecological aspects of the spectral quality of light is the
for animal nutrition but one that most vertebrate ani- adaptation of animal species to the selective pressures
mals cannot synthesize for themselves and that occurs of predation by the development of protective col-
at very low levels in plants. UV radiation that falls on oration. The case of the peppered moth (discussed in
the skin of animals (including humans) converts Chapter 16) exemplifies the almost universal phenom-
sterols into vitamin D and acts as the major source of enon of camouflage. In some cases, such as the pep-
this vitamin for many organisms. Children who live on pered moth, the coloration does not change during
inadequate diets in urban areas with heavy air pollu- the life of the adult. In others, it may change as the
tion (which virtually eliminates UV wavelengths from spectral character of the organism’s environment
sunlight) or in other environments in which they are changes. The varying hare (“snowshoe rabbit”) and
not exposed to sunlight tend to develop vitamin D de- arctic ptarmigan have white fur and feathers, respec-
ficiency symptoms (e.g., rickets, the malformation of tively, in winter to blend with the snow and brown fur
bones). These can be alleviated by exposure to sun- and feathers in the summer to blend with their snow-
light containing UV or simply by artificial irradiation free surroundings. Species of flat fish will change color
using UV lamps. The importance of UV synthesis of and pattern within hours or even minutes of being
vitamin D to animal populations is not known, but placed on a different-colored ocean floor, and animals
variations in the abundance of certain animals in the such as the chameleon lizard or the squid are able to
north-central United States have been shown to paral- change color in minutes or less to blend with their
lel variations in the intensity of solar UV radiation background. Not all coloration is for concealment.
(Shelford, 1951a, 1951b). Migratory birds that are Many species of fish, insects, and birds are brilliantly
prevented from traveling south from northern regions colored, although often this is true only for the male.
have been reported to develop vitamin D deficiency Female birds generally have coloration that conceals
during the sunlight-deficient winter (Clarke, 1967). them when they are sitting on their nest. Brilliant col-
Light passing through a canopy of vegetation ex- oration may serve the various functions of sexual dis-
periences changes in spectral composition, as well as plays, aggressive displays, distraction of predators
in intensity, because of differences in the reflectance, from a concealed mate on a nest, warnings of chemical
absorbance, and transmittance of the different wave- defense mechanisms, and mimicry.
lengths (Figure 7-1). The pigments responsible for
photosynthesis absorb radiation most efficiently in the
violet-blue and orange-red wavelengths, permitting 7.4 Ecological Effects of Variations
both reflectance and transmittance of green to exceed in the Intensity of Solar Radiation
those of other visible wavelengths. This is why most
plants are green. The spectral alteration of solar radia- Nowhere on Earth is the intensity of solar radiation so
tion by vegetation varies somewhat between species. It great that it has prevented the evolution or mainte-
is greater for hardwood forests with translucent leaves nance of life. High temperatures and the resulting
176 CHAPTER 7 Ecological Role of Solar Radiation

moisture stress that accompany high intensities of

sunlight do have a marked effect on life processes and Saturation point
Decline at very high intensitie:
may greatly restrict or even prevent the sustained exis-
tence of life. However, where these indirect effects are
not overpowering, ecological adaptation has been able
to solve most of the problems of high levels of light.
Lack of light has a more potent effect, and envi- photosynthesis
Net
ronments that completely lack visible radiation are de-
Compensation point
void of most or all forms of permanently resident
organisms. Chemoautotrophs and the organisms de- Light intensity ——»
pendent thereon can survive in dark places, as can de-
trivores if organic matter finds its way into such Figure 7-3 Generalized relationship between net photo-
environments. This accounts for life in the soil, in the synthesis and increasing light intensity. The CP is the light
depths of lakes and oceans, and in underground caves intensity at which photosynthesis equals respiration. The SP is
and rivers. Alternatively, organisms can exist in dark the light intensity above which no further increase in net pho-
environments if they make periodic forays into the tosynthesis occurs.
light world. This is an adaptation found in bats, which
are equipped with guidance systems (hearing or smell) crease, a point is reached at which respiratory losses
that are independent of light so that they can navigate are exactly balanced by photosynthetic gains. This
in their dark environments, as well as on nocturnal light intensity is called the compensation point (CP).
hunting trips. Plants are organisms that are able to live Above the CP, the rate of photosynthesis continues to
halfin a dark environment (the soil) and half in a light increase rapidly with increasing light intensity, but this
environment. This is made possible by differences in relationship is not sustained. With continued increases
morphology and function in the root and the aerial in light, the rate of increase in photosynthesis dimin-
systems. Another adaptation to life in dark environ- ishes until the saturation point (SP) is reached, beyond
ments is the production of light by the organisms which further increases in light intensity result in little
themselves. This phenomenon, known as biolumines- or no further increases in net CO, fixation. At very
cence, is perhaps best known in oceanic plankton and high light intensities, net fixation may drop because of
deep-sea organisms and in glowworms and fireflies. damage to the photosynthetic apparatus or for other
reasons. When expressed graphically, this relationship
is called the photosynthetic light saturation curve.
Effects on Plants CPs and SPs vary considerably among different
Photosynthesis. Photosynthesis is a light-depen- species, among different individuals of the same
dent process in which the rate of photosynthetic fixa- species, among different parts of a single individual,
tion of both CO, and solar energy is largely dependent and under different environmental conditions (Figure
upon light intensity.” However, this relationship is not 7-4). Leaves with very thin cuticles and algae that have
a simple linear one. It does not hold true at all light in- no cuticle to reflect and absorb light have net photo-
tensities, and it can be modified by a wide variety of synthesis at very low light intensities (i.e., very low
other factors. CPs). Plants with a high ratio of photosynthetic bio-
Consider the pattern of change in the rate of net mass to living supporting biomass (e.g., algae, mosses)
photosynthesis in a plant as the intensity of light in- will have lower CPs than plants with a low ratio (e.g.,
creases from zero (Figure 7-3). Photosynthesis in- trees) because they (the algae and mosses) have less res-
creases rapidly, but initially there is no net CO, piratory loss of CO, for which to compensate. Plants
fixation (and therefore no increase in biomass) because with low CPs often have lower SPs than plants with
the rate of CO, loss in respiration is greater than the high CPs. It takes less light to provide all the solar en-
rate of CO, fixation. As light intensity continues to in- ergy that the photochemical system can use in an algal
cell than in a tree leaf. Within a tree crown, leaves that
‘For simplicity, the term light intensity is used throughout the following grow in full sunlight (sun leaves) have higher CPs and
discussion to refer to the intensity ofphotosynthetically-active radiation SPs than do leaves that grow in deep shade (shade
(PAR). The correct term for this intesity is photosynthetic photon flux leaves) because of differences in leaf morphology
density (PPFD). (Figures 7-4A and Figure 7-5). A single, fully illumi-
 Single leaf

Sun leaves

Whole plants

cm?/hr
Photosynthesis,
CO,/50
mg

cm?/hr
Photosynthesis,
Net
CO,/50
mg 0 0.5 1.0 ES)
Light intensity, Light intensity,
(Thousands of foot-candles) (Thousands of foot-candles)
(a) (b)
Atmospheric CO, concentration

Shade tolerant:
Eastern red oak

Shade intolerant:
Loblolly pine

Net
CO>/min
J
Photosynthesis,

cm2/hr
Photosynthesis,
Net
CO3/100
mg
0 2 4 6 8 10 0) 0.5 1.0 eS 2.0
Light intensity, Light intensity,
(Thousands of foot-candles) (Thousands of foot-candles)

(c) (d)

a European
= larch
fe)
oO
oD
q 100 -
7, Norway io
seee spruce 80 ON,
fe en eS \ SS
sh ee:
S
S 60 Pacific \\ aN ue Grand fir
Oo _|R ° e e e e ee =
silver fir
= ae Stone Ht i
. pine
\ \ 7 Subalpine\
wean Ne
Balsam fir —— ae 5

| 2
Las Maximum
of
%
Photosynthesis
20 30 40 50 5 Oil 5 te 20M 125) e303 405 845
Soil Water (% dry weight) Water Stress (bars) in Twigs

(e) (f)

Figure 7-4 Variations in the rate of photosynthesis with a variety of variables. (A) With foliage type (sun and shade leaves) in
European beech (after Boysen-Jensen, 1932). (B) With degree of mutual shading in white mustard (Sinapis alba) (after Boysen-fensen,
1932).(C) With difference in shade tolerance (after Kramer and Decker, 1944).(D) With concentration of CO) in the atmosphere sur-
rounding the foliage of wheat seedlings (after Hoover et al., 1933).(E) With soil moisture status in three species of conifer (after
Havranek and Benecke, 1978).(F) With variation in internal plant moisture stress for four species of Abies (after Puritch, 1973).
 178 CHAPTER7 — Ecological Role of Solar Radiation

50
Fertilized = 8
3Qa. Jo
= 4042.0 9
Sg
=F
eo)
E°
_
oO a

Control oo 30-15 8
ZE 05
S$ 20+102
| Q.
> e
gfo) sFm
E 10F05 0
S
a Ss
iz
CO,/dm?
Photosynthesis,
Net
leaf/hr
mg 1 2 3 4 2 0.2 0.4 0.6 0.8 1.0 1.2
Light intensity, % N in foliage
(Thousands of foot-candles)

(g) (h)

Clear day Cloudy day

50 S 12
S
og ea em o) = 25
uAO [=catelO
EI : 2 20+ * 20
8 o
&B 30K
wh
3+
as)
8 ee 1 Temp 20
oD 15h
vo
3 15
S
pe
on
= en
B 208 6 15 = = 10- 8 10 10 =
IOP Sree ine eae SG We=
& s-e 5 5 E
r= Q re uecc =~ VPD a
sy % oO.
6.0, a : nner ee ey ? =
a
el \ 5 >
N
~

. 40 =e “NN
4.0 | WA
-
a
=) | i ca
YN ‘\ Noble fir aay
if

ina
9) ie _-
2.0 4 \, \Grand fir
ad
=
} Vicon \

0.0
if
/
\\%
SD UNS pater ane das Grand fir ; ec Noble fir \ oe
> \\
g
o an mae
x ae
<a ee
Oo. 20 Ne
D =.

ie cee carvnciar sca Aconlited ©

12 12 12 12 12 12
Midnight Noon Midnight Noon Midnight

Figure 7-4 (Continued)

(G) With nitrogen fertilization of Douglas-fir (after Brix, 1971).(H) With foliar nitrogen concentrations in Norway
spruce (after
Keller, 1971).(1) With time of day in two tree species (after Hodges, 1967). (Reproduced with permission.)

nated leaf will exhibit lower CP and SP values than the lock, maple, redwood) generally have lower CPs and
leaves in a crowded group of plants, where there is a lot SPs than do species adapted to high-light-intensity en-
of mutual shading. Many of the leaves in such a group vironments (e.g., pines, Eucalyptus), which may not
of plants will be below their CP or SP until high light reach SP even at full sunlight because of the morphol-
intensities are reached (Figure 7-4B). Species that are ogy and arrangement of their foliage (Figure 7-4C).
adapted to growing at reduced light intensities (which Because of the effects of within-canopy shading,
are referred to as shade tolerant, e.g., dogwood, hem- foliage efficiency (the kilograms of net production per
 Ecological Effects of Variations in the Intensity of Solar Radiation 179
Ej=)
E
2 100
E 90
°
e& 80
2 70
2= C0)
2 50
al

2 40
- O47 Do SB He Hee of aay LOM 2" 1334
Needle age-class, year

100 Nitrogen Nitrogen

90
E= 80
— 70
a
E 60
I

2er50
Ss
= 40
5
5 Phosphorous 100 Phosphorous
= Co
‘90
F= 80
S
S 1
60
50 Photosynthetic
leaf
rate/g
%
nutrient,
maximum
of

40
Cig eh 6 Se OM 0811 19813 914 Oe Sah Ab Sy, oh GP ON KO al Wey ale
Needle age-class, year Needle age-class, year
(b) (c)

Figure 7-5 Variation in photosynthetic rates (A), nitrogen and phosphorus content (B), and
nitrogen and phosphorus use efficiency (C) in black spruce needles of increasing age (data
from Hom and Oechel, 1983). Data are expressed as a percentage of maximum values.

unit of foliage mass, leaf area, or foliage nitrogen con- in the air surrounding plants, a fact well known to hor-
tent) declines as the foliage mass or area increases. ticulturalists who raise CO, levels in greenhouses to
However, the rate of decline depends on the photo- increase growth rates of various plants (Figure 7-4D).
synthesis—light intensity adaptations of the species in CPs are higher under conditions of moisture stress.
question (Figure 7-4). Photosynthesis is impaired by water stress, and it drops
The intensities of light at which compensation and off rapidly as the plant approaches its wilting point (see
saturation points occur depends on several other fac- Chapter 10), so much higher light intensities are re-
tors. Photosynthesis is a highly complex chemical quired for survival in dry environments or where there
process that is dependent on a variety of chemical raw is severe competition for water. Water stress may kill a
materials, including CO, and H,O. SP values can be plant by literally starving it to death by preventing net
greatly increased by raising the concentration of CO; photosynthesis (Figure 7—-4E, F). The increased
180 CHAPTER 7 Ecological Role of Solar Radiation

growth and yield after fertilization reflect the im- 40% of maximum values in 13-year-old needles (e.g.,
proved ability of plants to utilize sunlight energy as the Hom and Oechel, 1983). Nitrogen concentrations
supply of nutrients is increased (Figure 7-4G, H). were high in the first eight age classes before declining
The relationship between light intensity and net to 70% of maximum at age 13; phosphorus declined to
photosynthesis is complex and under the control of 55% of the maximum value that was in current foliage.
many factors. It is not surprising, therefore, that net As a result, nutrient use efficiency (or “nitrogen pro-
photosynthesis in natural stands of plants does not al- ductivity” [Agren, 1983]) was highest in 1-year-old
ways follow the daily variation in light intensity. In needles (Figure 7—6) and declined with age. This was
clear weather, there may be a morning peak in net interpreted to reflect both a decline in photosynthetic
photosynthesis followed by a midday dip and a second efficiency and the use of older needles as a nutrient
peak in the afternoon (Figure 7-41). It has been sug- storage site. The physiological significance and eco-
gested that this midday dip may result from one or logical factors that determine variation in leaf life
more of the following factors: spans are discussed in Gower et al. (1993) and Reich et
al. (1995).
1. Overheating of leaves The effects of temperature on photosynthesis and
2. Excessive respiration respiration are shown in Figure 7—7. Gross photosyn-
3. Water deficits thesis responds rapidly to initial increases in tempera-
ture, but the rate levels off as lethal temperatures are
4. Accumulation of products of photosynthesis in the
approached. Respiration exhibits the opposite pat-
leaves
tern, increasing slowly in the lower part of the tem-
5. Photooxidation of enzymes and pigments perature range and rapidly at the upper part of the
6. Closure of stomata range. The sum of these two relationships results in
7. Depletion of CO) in the air surrounding the crown maximum net photosynthesis at relatively low to in-
that accompanies high intensities ofsolar radiation termediate temperatures.
in the middle of the day The relationship of photosynthesis to light inten-
sity thus is seen to be complex, and recent research in
Conifers may or may not show this bimodal pat- this field has used computer modeling as a means of
tern of photosynthesis. Scots pine, noble fir, and grand handling this complexity (see Chapter 21).
fir were reported to have a bimodal pattern on a clear
day with a high vapor pressure deficit (VPD) but a
unimodal pattern on an overcast day with a low VPD
(Hodges, 1967). Gross photosynthesis_ee |,
Yet another source of variation in photosynthetic
rates is leaf age. Foliar nitrogen concentrations gener-
Sia /
7
ally decline with increasing leaf age, and photosyn-
Mh
thetic rates are a function of foliage nitrogen content.
Nosak Ea
As a result, deciduous species whose foliage is mostly “ Respiration
young and generally has relatively high nitrogen con- photosynthesis xy:
xX
centrations generally have higher average rates of CO, aN
fixation than evergreens (Reich et al., 1995). Northern
conifers may retain their needles for up to 30 years,
and if this relationship holds true over this entire life photosynthesis
and
respiration
of
Rate
span, then it is not surprising that northern forests from
Death
cold heat
from
Death
grow slowly (e.g., Chapin and van Cleve, 1981). There AOS 0 10 20 30 40 50
does not seem to be any difference in the decline of
Temperature, °C
photosynthetic capacity with age between conifers and
broad-leaved species apart from the variation with leaf Figure 7-6 Effect of temperature on photosynthesis and
respiration of Pinus cembra. The broken lines for photosyn-
age (Reich et al., 1992). Photosynthetic rates in black
thesis at the higher temperatures are estimates. (Based on Tran-
spruce needles in Alaska were reported to be highest quillini, 1954; after Kramer, 1957. Reproduced with permission of
in l-year-old conifer needles, declining with age to W. Tranquillini and P#. Kramer.)
 Ecological Effects of Variations in the Intensity of Solar Radiation 181

100 100

“a 80 80
‘a
=
= 90 60
a=
6 40 40
I

20 20

20 40 60 80 100 2X2 4X4 6X6 8x8 10 20 30

% crown closure Spacing, ft

meni
Figure 7-7 Effect of forest cover on the intensity of solar radiation reaching the forest floor. (A) Variation with crown closure,
conifer forest, Sierra Nevada mountains, Calif. (after U.S. Corps of Engineers, 1956).(B) Variation with tree spacing; the jack pine plan-
tation at 6 X 6 and 8 X 8 ft had not yet closed canopy (data from Reifsnyder and Lull, 1965).(C) Variation with height aboveground in a
beech stand in sunny and overcast conditions (after Reifsnyder and Lull, 1965).(D) Effect of time of day (angle of sun in the sky) on light
intensity reaching the top of the canopy and the ground in a tropical forest as percent of midday full sunlight (after Longman and Jenik,
1974). (Reproduced with permussion.)

Most of the light that reaches the forest floor in a seedlings and understory plants with light that may be
closed canopy forest during sunny weather is in sun- as much as 50% as intense as full sunlight. In a study
flecks: patches of direct sunlight shining through of a tropical rain forest, it was found that 20 to 25%
small gaps in the canopy (Vézina and Boulter, 1966). of the forest floor at noon was illuminated by sun-
The average light intensity found beneath a well- flecks that accounted for 70 to 80% of the total solar
stocked forest stand is lower than the CPs of most energy reaching the ground (Evans, 1956). The com-
plant species (compare Tables 7-2 and 7-3); under parison of Tables 7-2 and 7-3 should be made with
such conditions, net photosynthesis can occur only this in mind.
when a sunfleck falls on the plant. Consequently, data Although the CPs shown in Table 7-2 were ob-
on the average light intensity beneath a canopy are tained under experimental conditions and will be
often not particularly useful. Whereas much of the subject to some modification for field conditions,
forest floor may receive intensities of less than 1% they give a reasonably good indication of the re/ative
full sunlight, the sunflecks that move over the forest minimum light requirements and shade tolerances of
floor as the day progresses temporarily provide the various species. The highest CP values are for
182 CHAPTER 7 Ecological Role of Solar Radiation

Table 7-2 CPs of Seedlings of Several Tree Species Table 7-3 Light Intensities Beneath Various Types of
Expressed as a Percentage of Full Sunlight Forest Expressed as a Percentage of Full Sunlight
CP, % Percentage of Full Sunlight
Species
Forest Winter With
Ponderosa pine 30.6
Type Leafless Period Foliage
Scots pine 26.1
Northern white cedar 18.6 Red beech 26-66 2-40

Tamarack 17.6 Oak 43-69 3-35

Douglas-fir 13.6 Ash 39-80 8-60

Lodgepole pine 13.6 Birch = 20-30

Oak 13.6 Tropical rain forest - 0.2-2.0

Engelmann spruce 10.6 Jack and red pine 7-15

Eastern white pine 10.4 Eastern white pine 27

Norway spruce 8.7 Western white pine 6-15

Eastern hemlock 8.4 Scots pine 11-13

American beech 18 Norway spruce 2-3

Sugar maple 3.4 Silver fir 2-20

From Reifsnyder and Lull, 1965; Geiger, 1965.

Burns, 1923; Bates and Roeser, 1928.

species such as pines that are adapted to high light already equipped with secondary foliage take root, or
intensities; the lowest CPs are for species (e.g., hem- from layering that may occur when small trees are
lock, beech, maple) whose seedlings often grow in tipped over, or pendulous branches touch the ground
deep shade. (Weber et al. 2003). This topic is discussed again later.
The presence of seedlings on the forest floor is not It has been found that, in general, at least 20% of
necessarily evidence that the total light energy re- full sunlight is required for tree survival over a period
ceived is above their CP. Seedlings of species with of many years (Spurr and Barnes, 1973) and almost all
large seeds can exist for several years below their CPs species make their maximum biomass growth under
by using energy stored in the seed. conditions of full sunlight. As the ratio of photosyn-
The light requirements and CP of a species may thetic biomass to respiring, nonphotosynthetic biomass
change as it ages, so intensities that are adequate for a decreases, the respiratory load on the photosynthetic
young seedling become inadequate as it grows older. apparatus increases, leading to increases in CP values.
Norway spruce seedlings were found to survive for 1 The rate of height growth of seedlings generally
to 2 years at light intensities of less than 5% full sun- increases as light intensity increases from low values.
light, whereas intensities of 15 to 24% were necessary The light intensity at which maximum height growth
if the plants were to grow satisfactorily at 10 to 15 is reached varies considerably, but the seedlings of
years of age. Optimal seedling survival and growth oc- most species attain maximum height growth under
curred at 35 to 50% of full sunlight (Rousell, 1948). partial shade. Light-demanding species, such as pines,
Seedlings of the European ash (Fraxinus excelsior) will do attain maximum height growth in full sunlight, but
grow in heavily shaded hedgerows and limestone light intensity can be reduced by as much as 50% with
crevices, but the species becomes light demanding as it little loss in height growth (although significant loss in
gets older. In contrast, germinants of western redcedar diameter may occur). Seedlings of shade-tolerant
(Thuja plicata), which have needle-like primary foliage species such as sugar maple and sycamore may attain
with very small leaf area, are apparently relatively maximum height growth at intensities as low as 20%
shade intolerant. Once they develop scale-like sec- of full sunlight, and there is a significant growth re-
ondary foliage, the seedlings become very shade toler- duction in full sunlight. Seedlings of other species
ant. Regeneration in shade is largely limited to reach maximum height growth at intermediate light
veglings that develop when broken branches that are intensities (Table 7-4).
 Ecological Effects of Variations in the Intensity of Solar Radiation 183
Table 7-4 Effect of Light Intensity on Height Growth, Biomass Accumulation, and Shoot/Root Ratio in Seedlings of
Several Tree Species

Biomass, g

Species Light Conditions Height Growth, cm Shoot Root Shoot/Root Ratio

Jack pine (4 yr old) Full sun 112 OZ 30.0 W433

Partial shade 99 3251 U2 1.9
Full shade 40 Pt 1140) 2.7
Eastern larch (4 yr old) Full sun 158 31.8 29.8 ileal
Partial shade 170 27.9 20.5 ics)
Full shade 64 4.0 2s) 1.6
Loblolly pine (2 yr old) Full sun 42 25.2 20.1 0.8
Full shade 35 6.1 Ue? Vez
Overcup oak (2 yr old) Full sun 59 Pail ci 44.1 0.48
Full shade 66 20.1 38.7 0.52
River birch (10 wk old) Full sun 4 — — 3.6
Partial shade 52 _ — 8.3
Full shade 26 _ — 13.4

Kozlowski, 1962; Logan, 1966a; McDermott, 1954.

The light intensity that reaches the forest floor Clearcutting or even heavy thinning of beech stands
varies greatly according to the tree species, the density on moist, fertile sites before adequate beech regenera-
of the canopy, and the local solar radiation levels. The tion has been obtained can result in the establishment
light conditions experienced by a forest plant will vary of troublesome weed species, which make reestablish-
according to average light conditions in the stand, the ment of beech difficult and expensive.
presence of sunflecks, and the position of the plant in
relation to the rest of the canopy. A crown dominant Morphology. Most people are familiar with the
will receive full sunlight, whereas codominant, sub- fact that a plant grown in the shade looks different
dominant, suppressed, and understory plants will gen- from another plant of the same or similar genotype
erally receive progressively less light. Table 7-3 gives grown in full sunlight. Phenotypic differences be-
some average light intensities beneath various forest tween individuals of a species often result from vari-
types. The wide variation is the result of differences in ation in the light conditions experienced by the
age and density of stands. Figure 7-5 shows how the different individuals. Such differences may also be
radiation that reaches the forest floor varies with partly due to such factors as temperature, wind, and
crown density, spacing (number of stems per hectare), moisture, which tend to vary as light intensity varies.
height above the ground (note the variation according Individual factors never act alone in ecosystems,
to cloudiness), and time of day. and, as stressed earlier, causal determinism probably
Silviculturists use a knowledge of the relative CPs never occurs.
of crop and weed plant species and of the relationship Some plants will grow well only at high light inten-
between crown density and forest floor light intensi- sities (heliophytes), whereas others grow well only in par-
ties to influence the composition of vegetation on the tial shade (sciophytes). ‘This is a genetic adaptation to
forest floor. Many of the weed species in European increase the fitness of these species in particular light
beech forests have higher CPs than do beech environments, but heliophyte and sciophyte genotypes
seedlings. Careful opening of the beech overstory frequently lack phenotypic flexibility and are not good
canopy by selective removal of individual stems per- at acclimating to different light environments. The lack
mits the establishment of a crop of beech seedlings, of a thick cuticle, supporting and conducting tissues,
which, once established, is resistant to weed invasion. and roots makes mosses efficient at photosynthesis in
The remaining trees can then be safely removed. deep shade, where they are protected from excessive
184 CHAPTER 7 Ecological Role of Solar Radiation

moisture loss, but makes them poorly adapted for sur- British Columbia, may exhibit etiolation in competition
vival in full sunlight. The thick cuticles and mutual with evergreen conifers. An extreme example is a 15-m
shading of pine needles enable them to operate under tall maple with a 3-cm stem diameter 1.3 m above the
conditions of very high light intensity and the tempera- ground (breast height) that Iobserved growing in a 30-
ture and moisture stress that often accompany high year-old Douglas-fir stand.
light intensities but prevent them from acclimating to a Dramatic cases of etiolation are not normally ob-
shaded environment. Other plants, such as many of the served in forest trees, but light-controlled variation
hardwood tree species and understory plants, have great does occur in the allocation of new biomass to various
powers of acclimation, and a single genotype can grow parts of the plant. Under high light intensities,
in widely differing light environments by developing seedling root biomass may increase as fast or even
very different morphological characteristics. faster than stem biomass, but under reduced light in-
The importance of light in controlling plant mor- tensities, most of the net biomass production is invest-
phology can be demonstrated by keeping a light- ed in stem height growth, thus increasing the shoot/root
demanding, herbaceous, perennial plant such as a and height/diameter ratios. The response of trees and
potato or a dahlia in darkness as it begins its spring tree seedlings to reduced light intensity is generally
growth. The distribution of new biomass between not so much one of increasing height growth, al-
roots and stems and within stems between height though this may occur (Table 7-4). It is rather one of
growth and diameter growth is regulated by growth reducing root and stem diameter growth to sustain
hormones that are produced at the growing tips. height growth. Trees that grow in overcrowded plan-
These hormones, which are sensitive to light and tations or densely stocked natural stands develop tall,
which are partially or completely destroyed under thin, and weak stems with very small crowns and very
high light intensity, stimulate rapid cell division and small root systems, which makes them susceptible to
cell elongation. In bright light, most of the hormone damage by wind or snow. ‘Tree seedlings that grow in
produced is destroyed, so growth in length is moder- densely stocked or excessively shaded nursery beds
ate and the remaining growth materials contribute to will also have high shoot/root ratios.
root and stem diameter growth. In the dark, the hor- Ifaseedling is to survive light competition from its
mones are not destroyed; cell division and elongation neighbors, then it must expose its foliage to adequate
are promoted; and development of leaves, supporting light intensities. Producing a strong, thick stem and a
tissues, chlorophyll, and root systems is suppressed. large root system will do little to help the long-term
The result is rapid growth of a long, thin, weak, fleshy survival of a plant if this strategy maintains the plant in
stem that lacks chlorophyll and normally developed the shade of its competitors. Competition for light was
leaves as the plant attempts to grow into an area with probably one of the most powerful selection factors
higher light intensities. Plants that exhibit such mor- during the development of the first land plants, and
phological changes in response to inadequate light in- the development of erect stems to hold the photosyn-
tensity are said to be etiolated. thetic apparatus up to the light was one of the earliest
Etiolation is seen best in herbaceous light-demand- terrestrial plant adaptations (Wilson, 1970). Table 7-4
ing species. However, the ericaceous shrub Gautheria shows how the stem height, stem biomass, root bio-
shallon (salal), which grows in clearcuts in coastal British mass, and shoot/root ratio of seedlings of several
Columbia, is often 50 to 80 cm tall with robust stems in species vary with light conditions. A considerable vari-
clearcuts or other high light environments, whereas in ation in response is apparent. Light-demanding
deep shade in closed-canopy young forests, it can grow species such as jack pine show a much greater reduc-
as a vine-like plant supported by the branches of the tion in height and biomass growth and a greater in-
trees to height of up to 4 or 5 m. Etiolation generally crease in shoot/root ratio as light intensity decreases
does not develop in shade-tolerant plants. Evolution than do shade-tolerant species.
has “taught” them that sacrificing everything in favor of Leaves show a great range of morphological varia-
rapid stem elongation is not the best strategy, and many tion in response to varying light intensities. Leaves
shade-tolerant plants have adapted to restrain rapid grown in deep shade (shade leaves) are normally larger,
elongation under shaded conditions (Grime, 1966). thinner, and less lobed and have fewer layers of palisade
However, many trees that are traditionally thought of as cells than leaves grown in full sunlight (sun leaves)
moderately shade tolerant, such as the big leaf maple of (Figure 7-8). Palisade cells are the closely packed mes-
coastal northwestern United States and southwestern ophyll cells that occur in one or more layers beneath
 Ecological Effects of Variations in the Intensity of Solar Radiation 185

Epidermis

<M LLT j“ilSTE, _*

wo
Palisade mesophyll

Spongy mesophyll

Resin duct
por Pos

Epidermis
Palisade mesophyll
wee

~ In shade

See
Spongy mesophyll

Resin duct

(b)
Figure 7-8 Variation in leaf morphology in response to light intensity. (A) Cross-sections of
leaves of western hemlock developed in full sunlight and in the shade of a dense Douglas-fir canopy
(after Tucker and Emmingham, 1977). (B) Cross-sections of leaves of sugar maple from the southern
edge (full sunlight) and the center of the crown (shade) of an isolated tree (after Weaver and Clements,
1938). (Copyrights Society ofAmerican Foresters [A] and McGraw-Hill Book Co. [B]. Reproduced with per-
mission of the Society ofAmerican Foresters, McGraw-Hill, and WH. Emmingham.)

the leaf epidermis. In most leaves, they are found only very high. For example, the internal leaf surface of a
on the adaxial (upper) side of the leaf, but they can also 21-year-old Catalpa tree was reported to be 5100 m’,
occur on the abaxial (lower) side. Palisade cells usually whereas the outer leaf surface was only 390 m? (Tur-
contain the majority of the chloroplasts that are re- rell, 1934). The ratio of internal to external leaf sur-
sponsible for photosynthesis. In a study of a variety of face is greater in xeromorphic sun leaves (17.2 to 31.3)
herbaceous plant species, it was found that the distri- than in intermediate leaves (11.6 to 19.2) and shade
bution of chloroplasts between the palisade paren- leaves (6.8 to 9.9) (Turrell, 1936). Thus, the area of in-
chyma cells and the spongy parenchyma cells was 69 ternal leaf surface available for gas exchange varies ac-
to 86% in the palisade and 14 to 31% in the spongy cording to the ecological conditions. The larger
mesophyll (Schurhoff, 1924). Leaf size is often related internal surface area in xeromorphic leaves presum-
to light levels and can sometimes be used as a rough ably compensates for the smaller overall size of sun
measure of the light intensity, although many other leaves. For further details of leaf anatomy, see Esau
factors can affect leaf size as well. (1963) or another standard work on plant anatomy.
As light passes through the palisade layer(s), some Shade leaves have a thinner cuticle and fewer cu-
of it is absorbed, and at low intensities, insufficient light ticular hairs, thus reducing reflection and absorption
penetrates through the upper palisade layer to produce and increasing the transmission of light to the palisade
net photosynthesis in lower palisade layers or the cells. The epidermis of shade leaves may transmit as
spongy mesophyll. Under very high light intensities, much as 98% ofthe light energy, whereas the value for
the quantity of light that penetrates the first two layers sun leaves may be as low as 15% (Daubenmire, 1974).
of palisade cells may be greater than the CP of the indi- Shade leaves also have better developed mesophyll
vidual cells, and there may be a third palisade layer. layers through which CO, is absorbed and H,O is
The importance of the palisade layer is reflected in evaporated. As a result of these morphological charac-
the fact that the palisade tissue exposes 1.6 to 3.5 times teristics, shade leaves have a much higher photosyn-
as much surface within the leaf for gaseous exchange thetic efficiency than do sun leaves (as high as 20%;
as the spongy mesophyll. Where there are many pal- Daubenmire, 1974). This does not mean that trees in
isade layers, the internal surface of the leaf may be deep shade will grow faster than trees in full sunlight.
186 CHAPTER 7 Ecological Role of Solar Radiation

The total amount of CO, fixation is normally lower in Shortleaf pine stand, 42 yr old Loblolly pine stand, 31 yr old

shade leaves than in sun leaves despite this high effi- )

o
ciency because of the much lower levels of light to a
ion

which shade leaves are exposed.

2

Ss
a
—
)
Shade Tolerance. For many centuries, foresters Q
oO
5
have classified plants as shade tolerant or light demand- 4
5

ing (shade intolerant). The emphasis of this classifica-

tion has always been on light conditions, but we now =)
=)

know that the reaction of plants to shade conditions in Sb

aS
natural situations is only partly a matter of photosyn- a]
thesis/light intensity relationships. Just as plant height 3
growth is controlled by a large number of genes and :
oa
ro)
environmental variables, so tolerance of shade is a oo
multiply determined characteristic. 0 1 2 3 4 0 1 2 3
Years after trenching
____ Untrenched
Ability to Compete for Soil Moisture and Nutri- control plot
ents Under Shaded Conditions. Where there is Trenched plot
aboveground competition for light, there is usually (a) (b)
belowground competition for moisture and nutrients.
Figure 7-9 Effect of removal of root competition by
Shade tolerance may reflect the ability to compete for
trenching on the height growth of tree seedlings and the
soil resources as much as the ability to tolerate low
total number of plants on the trenched plots in two differ-
light intensities. ent pine stands. (After Korstian and Coile, 1938. Copyright Duke
Much of the early work on CPs showed that under University, School of Forestry and Environmental Studies. Repro-
experimental conditions, individual seedlings can exist duced with permission.)
at lower light intensities than those that occur in wood-
lands and forests that lack any understory vegetation.
Several researchers in the first half of the 20th century
reported that when root competition was eliminated by new growth that reduce a plant’s ability to compete for
digging a trench around an area of shaded forest floor soil resources or render them susceptible to wind and
that lacked any seedlings or minor vegetation, the area snow damage.
was rapidly invaded by overstory and understory The high shoot/root ratios and small root systems
species (Figure 7-9). The trenching increased soil of shade-intolerant plants grown at low light intensities
moisture, leading to improvements in the germination,
put them at a disadvantage in the competition for mois-
growth, and survival of tree seedlings (Korstian and
ture and nutrients with overstory plants whose canopies
Coile, 1938). Table 7-1 presents a comparison between
are fully illuminated. Wind and snow damage to trees
trenched and untrenched plots 8 years after trenching.
with high shoot/root or height/diameter ratios will re-
More recent trenching experiments have demonstrated
duce their ability to compete for light with undamaged
increases in mineralization of organic matter and a
trees. These morphological effects of light will reduce
great improvement in nutrient availability (Gadgil and
the success of the plants in shaded environments, from
Gadgil, 1971, 1975; Vitousek et al., 1979). However,
which they are therefore excluded. Successful plants in
trenching does not always benefit seedlings. Root such environments are those that have been molded by
grafting or the sharing of mycorrhizal fungi between evolution to have adaptations that ensure adequate sup-
overstory trees and seedlings may be a significant fac- plies of light, moisture, and nutrients under the very
tor in the survival of seedlings in the understory, and low light conditions. The rootless mosses that domi-
severing of these connections by trenching may reduce nate the ground vegetation in very shaded conditions
seedling growth in some cases (Simard et al., 2001).
obtain most of their nutrients from throughfall, mini-
mize moisture loss by their prostrate growth form, tol-
Effect of Shading on Allocation of New Growth. erate drying out, and have very low CPs because almost
Low light intensity may cause shifts in allocation of all of their biomass is photosynthetic; there is almost no
 Ecological Effects of Variations in the Intensity of Solar Radiation 187

Table 7-5 Height Growth of Tree Seedlings and Abundance of Minor Vegetation 8 Years After Removal of Root
Competition by Trenching

Control Plot Trenched Plot

Average Average
Height of Height of
Number of Seedlings, Number of Seedlings,
Seedlings cm Seedlings cm

Response of tree seedlings

Eastern hemlock 3 if 11 96
Eastern white pine 6 26 36
Response of herbaceous plants
White violet (Viola sp.) 39 764
Blackberry (Rubus sp.) 18 357
Five finger (Potentilla sp.) 12 279
Percent total vegetative cover 8.1 80

Data from Toumey and Kienholz, 1931.

respiration “tax” imposed by nonphotosynthetic organs and permitting survival in the shade. However, as the size
such as roots and stems. Some tree species that can of the tree increases, this ability declines and the carbon
grow in deep shade have evolved to sacrifice height allocation pattern approaches that of mature trees, which
growth in favor of maintaining branch elongation and appears to vary relatively little between “shade tolerant”
root growth. It is not uncommon in dense old-growth and “shade intolerant” species. If the tree is still growing
subalpine forests in coastal Pacific Northwest United in the shade as it gets larger, this mature carbon alloca-
States and British Columbia to see 1-m-tall Pacific sil- tion pattern is poorly adapted for surval at low light in-
ver (amabilis) fir (Abies amabilis) saplings of up to 100 tensities. Thus, the carbon allocation aspect of shade
years in age with a branch spread of as much as 4 m in tolerance appears to be a juvenile characteristic.
diameter and a height increment of less than 1 cm.
Ability to Alter Leaf Morphology and Orientation
There is a greater chance of finding light by going out-
in Response to Light Intensity. Plants that are able
ward rather than upward, a strategy that maintains an
to grow in low light intensities produce shade leaves
understory of stunted saplings, many of which can re-
that are designed to optimize photosynthesis under
sume normal rates of height growth if a gap is created in these conditions: large leaves with thin cuticles and
the overstory. This is a very different growth strategy few lobes, held horizontally in a pattern that mini-
from the big leaf maple example given above. Evidence mizes mutual shading. The leaves (and the plants in
suggests that there is not a single set of growth strate- general) have a higher ratio of photosynthetic biomass
gies that is adopted by all “shade tolerant” or “shade to nonphotosynthetic biomass than do plants that
tolerant” species. Different species appear to have been grow in high light intensities, and there is a lower res-
selected for different combinations of adaptations to piration rate. These adaptations combine to reduce
light availability and competition. This requires that we CPs, improving the chances of survival in low-light
understand the light adaptations of each individual environments (Table 7-6).
species rather than simply classifying them as a member
of a particular light adaptation group. Effect of Seed Size on Shade Tolerance. Shade-tol-
Plasticity in carbon allocation to different tissues erant trees often have large seeds with abundant energy
varies as a function of tree age and size (Messler and reserves, which enables the seedlings to develop a signif-
Nikinmaa, 2000; Claveau et al., 2002). Seedlings and icant although diminutive root and shoot system before
small saplings of many shade tolerant species are able to they have achieved any net photosynthesis. In forests
allocate more NPP to foliage at low than at high light in- where the surface few centimeters of forest floor dry out
tensities, thereby maintaining high ratios of photosyn- in mid-summer, only seedlings that have developed
thetic biomass to respiring, non-photosynthetic biomass roots to below this dry layer will survive their first year.
188 CHAPTER 7 Ecological Role of Solar Radiation

Table 7-6 Relative Growth and Respiration Rates in The difficulty in defining shade tolerance simply by
Seedlings of Shade-Tolerant and Light-Demanding the correlation between regeneration of a species and
Tree Species light intensity has led to the search for an effective index
Growth rate? _—Respiration rate” of shade tolerance. Carter and Klinka (1992) demon-
Species mg g” hr* mg g” hr" strated variation in shade tolerance of three conifer
species both between and within species. In particular,
Shade tolerant they reported that for the species that they examined,
Eastern hemlock 2.9 22 shade tolerance, expressed as the ability to sustain a par-
Pacific yew — ee, ticular relative growth rate, declined along a gradient of
soil moisture from dry to wet. Klinka et al. (1992) exam-
American beech - 2.6
ined the usefulness of various measures of morphological
Red maple a5 231
response to light levels as an index of shade tolerance and
Sugar maple 0.6 1.8 found that variation in specific leaf area (the ratio of leaf
Chinese chestnut 1.8 1.2 area to leaf mass: cm’ g*') and in the allocation of growth
Northern red oak 0.6 1.8 to height and stem diameter were useful measures of the
Shade intolerant ability of a species to survive and grow in the shade.
Tulip tree _ 3.9
CO, Concentrations. ‘he shade tolerance of some
Black walnut _- PRE
diminutive forest plants such as mosses, liverworts, and
Black cherry _- 4.8
small herbs is related to very high concentrations of CO,
Aspen ~ 4.2 in the 5- to 10-cm layer above the forest floor. These
Sweet birch 4.5 3.6 high concentrations are produced by the soil microbial
Sumac 2.8 6.6 and faunal decomposers of litterfall. Such plants also
American elm 3.9 _
have a low respiratory load because very little of their
biomass is nonphotosynthetic, resulting in very low CPs.
Data from Grime, 1966. The interesting example of shade tolerance in west-
*Growth rate of 1-week-old seedlings in full sunlight: milligram ern redcedar an evergreen conifer of the northwestern
weight gain per gram of dry seedling per hour.
United States and western Canada, has already been in-
>Respiration rate of leaf discs over 12 hours: milligram weight loss
per gram of dry leaf per hour at 25°C.
troduced. Long considered a very shade-tolerant tree
species, redcedar can be the dominant species in old-
growth forests in cool humid and temperate rainforest
areas. This species produces abundant small seeds,
This may limit regeneration to large-seeded species
something that is generally associated in this geographi-
with low CPs. Large seed size is not limited to shade tol-
cal area with disturbance-related, shade-intolerant, early
erance, of course. It may also be an adaptation to dry,
seral tree species. It has abundant regeneration on dis-
high-light-intensity environments (e.g., many pines, co-
turbed sites such as road edges and clearcuts, but it also
conuts). However, for two species with similar CP and
SP values, the species with the larger seed will often be
regenerates in the shade of old forests. Recent studies
(Weber et al., 2003) have shown that the germinants and
more successful at regenerating in the shade where there
is a thick forest floor that dries out in the summer.
some young seedlings of redcedar, which have very
small, needle-like leaves with a very small leaf area, are
Effect of Disease Organisms on Shade Tolerance. relatively shade intolerant, at least under an evergreen
The higher humidity that occurs in shaded environ- overstory. Redcedar is an indeterminate species that is
ments favors the growth of various fungal pathogens able to grow in height at any time of year when temper-
of tree seedlings (Vaartaja, 1952), and shade tolerance atures are favorable, and it has a very short photosyn-
may be related to resistance to these pathogens. thetic induction period that allows it to take advantage
It should be apparent from this discussion that of sunflecks (Karakatsoulis, 2003); these features permit
whether a seedling can establish and grow in a partic- it to grow in the shade of deciduous broadleaved trees
ular shaded environment is not simply a question of and shrubs. After a period that varies from a few months
light intensity and CPs. As with most ecological phe- up to several years, these germinants/young seedlings
nomena, several factors are involved, and shade toler- develop very shade-tolerant, scale-like secondary fo-
ance is a good example of multiple determinism. liage, the time for such development being related to the
 Ecological Effects of Temporal Variations in Solar Radiation 189

success in taking up sufficient nutrients. This in turn is Some heterotrophic organisms have evolved to be-
related to the availability of the appropriate mycorrhizal come active mainly at times of low light intensity be-
fungi. Once equipped with secondary foliage, they are cause of the concealment conferred by darkness, and
able to grow slowly even in fairly deep coniferous shade. some species are entirely nocturnal in their activities.
Thus, the shade tolerance of seedling regeneration is re- The nocturnal habit undoubtedly developed first in
lated to leaf morphology, nutrition, and the presence of herbivores as an adaptation to avoid predation.
mycorrhizal associates, as well as to light levels. So how Respite from their enemies was probably short-lived,
does this species regenerate in the shade of old-growth however, as the eyesight of their predators evolved to
conifer forests? On fertile sites and where there are large enable them to hunt their prey using visual clues at
canopy gaps, the light and nutrients required for germi- very low light intensities. The eyes of vertebrate ani-
nant establishment and secondary foliage development mals are generally able to detect the movement of
are present. However, where they are not, regeneration small objects at light intensities as low as one ten-
is generally by veg/ings. Branches broken in winter billionth of full sunlight (Clarke, 1967). Natural selec-
storms rapidly develop roots in contact with moist forest tion obviously had the genetic material ready at hand
floors. Alternatively, redcedar reproduces by /ayering with which to produce nighttime vision.
that also results in veglings. Attached branches of stand- Light plays an important role in the orientation of
ing trees or branches of trees that have been tipped over both plants and animals. In plants, the response is called
by wind or snow may take root where they contact moist phototropism; the orientation and movement of an ani-
soil or forest floor and grow into independent stems. All mal in relation to a light stimulus is termed phototaxis.
veglings have shade-tolerant secondary foliage from the The orientation may be either toward (positive) or away
outset and so are able to establish and grow in levels of from (negative) the light source or at some angle to it,
shade that exclude seedling reproduction. Redcedar has and the direction of the response may vary under differ-
relatively high height/diameter and shoot/root ratios ent conditions. Spruce budworm larvae have a positive
when growing in deep shade—a characteristic more of phototaxis that leads them out to the end of branches
shade-intolerant than of shade-tolerant tree species. during bright sunshine and negative phototaxis under
This results in the tipping over of many understory overcast conditions when the source of light is diffuse.
trees, which then form veglings. Many animals have a complex orientation behavior in
Clearly, shade tolerance is a complex plant attribute. which the angle of the sun in the sky is used for orienta-
There are many combinations of different plant growth tion, allowance being made for the change in the angle
strategies that can enable a plant to reproduce, survive, of the sun as the day progresses.
and grow under conditions of significant competition Information about its discovery of a new source of
for light. For a comprehensive review of the effects of pollen or nectar is relayed by a bee to other bees in the
light and other factors on photosynthesis and carbon al- hive by means of a waggle dance, in which the distance
location, see Gower et al. (1995), Luxmore et al. (1995), of the new resource is indicated by the duration of the
Stenberg et al. (1995), and Teskey et al. (1995). dance, the magnitude of the resource is indicated by
the enthusiasm and persistence of the dancing bee,
Effects on Animals and the direction relative to the angle of the sun in the
Variations in the intensity of solar radiation are generally sky is indicated by the orientation of the dance on the
of less direct consequence to animals than to plants, but honeycomb (von Frisch, 1967). Both bees and birds
light intensity nevertheless plays an important role in the have been shown to have eyes that are sensitive to the
lives of animals. Most animals depend to a considerable angle of polarization of sky light, an adaptation that
extent on their eyesight for finding food, for detection of permits unerring navigation over featureless expanses
enemies, and for navigation, although in many cases the of forest, ocean, or desert even on a cloudy day.
senses of hearing, smell, and taste may be as important or
even more important. The intensity of light plays an im- 7.5 Ecological Effects of Temporal
portant role in the level of activity of many animals that Variations in Solar Radiation
enter a resting period during darkness (called “sleep” in
some animals) and become active in the light. The speed The survival of an organism is determined by both the
of movement and general activity of many animals, such factors of the physical environment and the other or-
as insects and crabs, increase as light intensity in-creases, ganisms with which it interacts. Success at exploiting fa-
a phenomenon known as photokinesis. vorable opportunities and at avoiding unfavorable
190 CHAPTER 7 Ecological Role of Solar Radiation

circumstances in an environment that varies both daily activity is presumably important to animals that may
and seasonally depends to a great extent on the ability overwinter underground or in polar regions, where
of an organism to do things at the correct time. Preda- the normal environmental stimulus of varying light
tors must be in the right place at the right time if they does not occur throughout the year. Many plants grow
are to eat. Insect-pollinated flowers must flower when normally only when exposed to alternating day and
the adults of their insect pollinators are around, and the night conditions, and their metabolism becomes ab-
food needs of nectar-feeding insects must coincide with normal when they are exposed to continuous day for
the flowering period of their host flowers. Long-dis- prolonged periods.
tance migrations of animals away from environments Seasonal variations in the physical environment
that are about to become unfavorable must often start and the consequent seasonal variations in the biotic
at a time when food and temperature conditions are still environment are largely phenomena of temperate and
favorable if the animals are to reach the security of polar regions. The small seasonal variations in tem-
more equitable environments before bad weather ar- perature and moisture conditions that occur at the
rives. Plants that live in areas with low winter tempera- equator are thought to be of relatively minor conse-
tures must initiate physiological changes (which may quence, but as one moves away from the equator to-
take several weeks for completion) in time to become ward either of the poles, one soon encounters areas in
resistant to unfavorably low temperatures when they which ecologically important seasonal dry periods
occur. In short, organisms need to have a sense of time. occur. In the middle latitudes, marked seasonal fluctu-
The environment exhibits two major types of pre- ations of both moisture and temperature are charac-
dictable variation that require some means of time- teristic, whereas farther poleward, the season of
keeping by the organism: the day—night variation and favorable moisture and temperature conditions is
the seasonal variation. There is also a monthly perio- greatly shortened; a period of unfavorably low tem-
dicity in the activity of some animals in response to the peratures dominates the year.
phases of the moon (Clarke, 1967); most of the exam- Organisms that live in these seasonal regions
ples are marine organisms, for which monthly tidal need a reliable “calendar of environmental events.”
variations are important. Such a calendar cannot be based on light intensity or
The alternation of day and night and the annual temperature alone, because although these vary with
“march of the seasons” give rise to variations in the the season, they are subject to the unpredictable ef-
availability of food (or nutrients) and shelter, in the ac- fects of unseasonal weather. Snow in midsummer and
tivity of natural enemies, in temperature, and in the sunbathing temperatures in midwinter are not un-
availability of water. An organism that is unable to or- known at temperate latitudes, at least for a day or
ganize its daily and annual activities to take advantage two. The calendar must therefore be based on a reli-
of good conditions and to complete all of its essential ably predictable environmental factor that varies in
activities and functions in time to prepare for a period phase with the average onset and termination of un-
of unfavorable conditions is very unlikely to survive. favorable environmental conditions. The annual
Consequently, most organisms are adapted to a diurnal variation in the relative lengths of day and night pro-
(day/night) periodicity, whereas those that live in a vides just such an environmental timekeeper, al-
seasonal environment also have an annual periodicity. though organisms normally use other environmental
Organisms that live in a relatively nonseasonal (equa- clues such as temperature in conjunction with
torial) or unpredictable (desert) environment may lack day/night length.
a strongly seasonal periodicity. The sensitivity of organisms to the length of day-
In some organisms, the variation in day/night ac- light is called photoperiodism. Actually, for plants, it is
tivity is largely a matter of photokinesis; the organisms the length of the uninterrupted dark period that is
are simply more active in the light. In others, the important, as can be demonstrated by interrupting
rhythm is more deeply engrained. Certain species of the middle of a long dark period (short day) with a
small mammals brought into the laboratory and kept brief period of light (Figure 7-10). This produces the
in complete darkness have been found to continue same result as a short “night” (i.e., long day) treat-
their diurnal rhythm of activity for up to 7 months ment, because plant response to light is based on a
(Clarke, 1967). The maintenance of such patterns of light-absorbing pigment known as phytochrome. This
 Ecological Effects of Temporal Variations in Solar Radiation 19]

pigment can exist in two different forms: a biologi-

cally active form, which absorbs radiation in the far-
red part of the spectrum (PFR), and a less active
form, which absorbs in the red wavelengths (PR).
During the day, the red portion of the visible spec-
trum converts most of the phytochrome into the bio-
logically active form, thereby promoting activity.
During the dark period, there is slow conversion to
the inactive form; the longer the dark period, the
greater the proportion of the phytochrome in the in-
active form and the smaller the response of the plants
to light. This can be summarized as follows:

influence of red wavelengths

PR ——_- PFR
inactive form : cti
( ) influence of far red wavelengths pct fom)

4)
ad
Slow conversion during
night to inactive form ("yaa
+ all

short exposure to light

in middle of night

Rapid conversion back to active form 12 hours 12 hours + | hour 20 hours

of light in the
middle of the dark
period
When most of the phytochrome is in the inactive
form, active growth ceases and the plants enter dor- Figure 7-10 Photoperiodism in trees. The effect of inter-
mancy (Downs, 1962). With long-day and short- rupting the dark period on the photoperiodic response in
night conditions, most of the phytochrome is in the plants. Douglas-fir grown with a 12-hour photoperiod enters
active form and the plant responds actively to light. dormancy, but if the 12-hour dark period is interrupted in the
Interruption of a long dark period by a brief period middle by a 1-hour light period, then growth continues and
of light converts much of the inactive phytochrome achieves almost as much total growth as plants under a 20-hour
back to active form and therefore simulates a short- photoperiod. (After Downs, 1962. Copyright Ronald Press. Re-
drawn by permission offohn Wiley & Sons, Inc., and R.F. Downs.)
night condition.
Photoperiodism can be found in all types of organ-
isms and, perhaps surprising, at all latitudes. It is best
developed in areas with highly variable but seasonally erally recognized), or it may simply be the result of an
predictable climates (e.g., alpine or polar regions). Its incomplete understanding of photoperiodism.
occurrence in some tropical organisms may reflect the An enormous variety of biological phenomena
phenomenon of continental drift. Present-day land have been shown to exhibit some degree of photoperi-
masses were not always in their present positions. odism (Vince-Prue, 1975). Its major role is in regulat-
They have moved slowly over the surface of the earth ing the onset of reproduction in plants and animals
as the result of horizontal and vertical movements of and in preparation for dormancy before the onset of a
the earth’s crust (tectonic movements), and present- climatically unfavorable season. Photoperiodic re-
day equatorial areas have occupied positions at higher sponses are classified into one of the following three
latitudes in the past. Equatorial photoperiodism may categories according to the reproductive or other re-
also reflect our failure to detect some current season- sponse of the individual:
ality of ecological importance in equatorial regions
(e.g., there is a much greater seasonality in precipita- 1. Short-day organisms: Animals that reproduce, plants
tion and humidity in many tropical forests than is gen- that flower or flower most rapidly, and organisms
192 CHAPTER 7 Ecological Role of Solar Radiation

that enter dormancy or complete some other as- diate latitudes. At high latitudes, there are too few
pect of their life history with fewer than some crit- days that have both 2- to 15-hour photoperiods and
ical number of hours of daylight. The term favorable temperatures to permit the completion of
“long-night organisms” would actually be more short-day—induced life stages. Short-day organisms
appropriate because the length of the uninter- are therefore most common at the middle and lower
rupted dark period is the critical stimulus. latitudes. Conversely, long-day photoperiods never
2. Long-day organisms: Organisms that respond as occur near the equator and last for only a very short
short-day organisms, but to more than some criti- time (the middle of summer) at middle latitudes. They
cal number of hours of daylight. are most common at high latitudes (northern forest
and polar regions), and this is where long-day organ-
3. Day-length indifferent (day-neutral) organisms: Or-
isms are found.
ganisms that complete their life history events ir-
A few organisms (day-neutral) seem to ignore the
respective of the day length.
seasons and to grow and flower (plants) or reproduce
For short-day organisms, there are two occasions (animals) whenever environmental conditions are
each year when the day-length stimulus and the tem- favorable, irrespective of the length of the photo-
perature conditions satisfy the requirements for repro- period. Such organisms are found only in the less ex-
duction: the spring and the fall. Some organisms will treme seasonal environments. In the more severe cli-
breed at both times (Allee et al., 1967), but most short- mates, most organisms are obliged to “tow the
day organisms require a dual stimulus and reproduce seasonal line.” Determinate tree species in temperate
only once a year. Long-short-day organisms respond and northern latitudes set a terminal bud and cease
to short-day conditions only after a period of long days height growth in response to moisture, temperature,
(i.e., in the fall), and short-long-day organisms re- and photoperiodic stimuli and are constrained by this
spond only to a period of short days when the days are environmental calendar (e.g. pines, spruce, Douglas-
getting longer. Triggering of the photoperiodic stimu- fir and deciduous hardwoods). In contrast, inde-
lus may involve temperature and other environmental terminate species are able to grow in height whenever
factors. The critical photoperiod for short-day organ- the environmental conditions are suitable for photo-
isms is usually fairly broad, from 2 to 15 hours of light, synthesis and cell division/elongation.
whereas the critical photoperiod for long-day organ- The terms determinate and indeterminate actually
isms is from 16 to 24 hours. There may also be some apply to buds, shoots and other organs arising from or
intermediate-day-length organisms that respond to containing apical meristems rather than trees per se.
periods of 12 to 16 hours of light (Figure 7-11). They can apply at all latitudes, but there is a tendency
Short-day photoperiods occur much of the year towards more determinate growth of vegetative shoots
near the equator and in the spring and fall at interme- at higher latitudes. All trees are indeterminate in their

: Short-day Day-neutral organisms

High organisms Intermediate-day
2 yn organisms
5
\
a \
7
o
_
\
\
Long-day organisms

5 \
3
\
S 1
=
3 \
\
Low

18 24
Hours of daylight

Figure 7-11 Photoperiodism. Variation in the reproductive response of the three major photope-
riodic types of organism as day length varies. Other photoperiodic responses show a similar range of
types. (After Vince-Prue, 1975. Amended and redrawn by permission ofMcGraw-Hill Book Company [UK]
Limited and D. Vince-Prue.)
 Ecological Effects of Temporal Variations in Solar Radiation 193

first year of growth, but many then become determi- vorable spring food and temperature conditions re-
nate. Some continue to be indeterminate for several quires that mating take place sufficiently long before
years before producing determinate buds. Yet others, the birth event to allow for the gestation (pregnancy)
such as western redcedar (Thuja plicata) and yellow period. Animals such as deer, sheep, and goats, which
cedar (Chamaecyparis nootkatensis), are always indetermi- have a fairly long gestation period, must mate in the
nate, lacking vegetative buds altogether. Some species fall or early winter, and sexual activity is stimulated by
(e.g., larch, apple) have indeterminate long shoots and short days following long days. Conversely, animals
determinate short shoots (Dr. Robert Guy, Department with a very short gestation period can safely wait until
of Forest Sciences, University of B.C., personal com- late winter or early spring to mate, and they may
munication, 2003). This example shows once again the have a short-day-following-a-period-of-low-tempera-
complexity of things biological and ecological, and the ture photoperiodic response or even a long-day re-
difficulty in finding useful generalizations. There is no sponse. Initiation of preparations for reproduction
substitute for understanding the details of the different must start in advance of sexual activity, so courtship,
adaptations of different species—biological diversity selection and establishment of nesting sites or territo-
must be understood and respected. ries, and preparation of a shelter or nest must be initi-
ated by photoperiodic stimuli that permit sufficient
time for these activities to be completed before the
Role of Photoperiodism in Animals onset of breeding. An illustration of this is that birds
The phenomenon of photoperiodism was first demon- that feed on early spring insects have a shorter day
strated experimentally in plants (Garner and Allard, stimulus for nest building than do summer migrant
1920), but it has subsequently been shown to be the visitors or midsummer insect feeders such as swallows.
factor responsible for the timing of many of the criti- The migration of birds and other animals between
cal events in the lives of animals in seasonally variable polar and lower latitudes to avoid periods of winter
environments. It plays a critical role in regulating the cold has been known for a long time, but only in the
onset of reproduction, the start of migrations, and the past half century was it realized that photoperiod was
seasonal change in color of fur and feathers. the environmental trigger that initiated migrations.
A characteristic of most photoperiodic responses is The really long migrations from polar areas require
that they initiate activities that lead to future events that animals leave before temperature and food condi-
whose timing is critical. Such responses are therefore tions become unfavorable, and migrations are initiated
less useful in highly unpredictable environments than at approximately the same time each year in good or
in regular, predictable environments, although they bad weather. Only the evolutionary experience that to
may still play an important role in the former. For ex- delay can sometimes be fatal, coupled with the de-
ample, the timing of spring rains in temperate deserts pendable environmental calendar set by day length,
and the brief flush of plant and animal activity that fol- can ensure a consistent departure date irrespective of
lows cannot be accurately predicted, and therefore a the current conditions. The rigidity of this adaptation
rigid photoperiodic control of spring reproduction in will sometimes result in birds’ arriving at northern lo-
desert areas would not always synchronize reproduc- cations before winter has given way to spring. As with
tion with the period of abundant food. It has been most adaptations, photoperiodicity represents a com-
found in certain species of desert birds that spring promise between the conflicting pressures of selection
photoperiods initiate the development of the sexual induced by biotic factors and selection induced by
organs but do not actually stimulate reproduction. A physical environmental factors that results in maxi-
state of sexual readiness is prepared and maintained mum fitness of the species.
during the period when rains are most likely to occur, Seasonal variation in the color of the fur and the
but reproduction itself occurs in response to some feathers of terrestrial animals is a common adaptation
stimulus associated with the rains themselves (Mar- to seasonal changes in the color of the physical envi-
shall and Disney, 1957). ronment. Animals that are well camouflaged in the
Most animals in seasonal environments bear their summer would be obvious against a background of
young in the spring so that they have time to grow toa winter snow, and evolution has produced a genetically
size and maturity that will permit them to survive the controlled color change in many of the animals that
following winter. Synchronization of births with fa- spend the winter above the snow in seasonally snowy
194 CHAPTER7 Ecological Role ofSolar Radiation ef}

environments. Animals of the tundra that overwinter plants are to grow, mature, and reproduce before they
beneath the snow (e.g., the lemming) do not change are killed by fall frosts. Spring short-day perennials are
color. The timing of color changes is controlled by those that are likely to experience strong competition
photoperiod, as in the case of the varying hare (or for light, moisture, or nutrients from other species later
snowshoe rabbit, Lepus americanus), which begins to in the season. Many understory species in deciduous
change from producing brown fur to producing white woodlands initiate growth and flower before the over-
fur when the day length drops below 18 hours. The story species come into leaf and light conditions be-
reverse change occurs as day length increases beyond come unfavorable. Flower buds are normally set the
18 hours. White fur can be induced in the middle of previous summer or fall to avoid any delays in the
summer by exposing the animals to light periods of spring. Other short-day perennials flower in the fall.
less than 18 hours, and brown fur is produced in mid- Long-day plants are those that flower only when
winter by animals given an artificially extended pho- the days are long and the nights are short, a condition
toperiod (Allee et al., 1967). that is met only in the higher latitudes. Such plants
Many animals solve the problems presented by break dormancy and return to dormancy at much
winter low temperatures by going into a dormant con- longer day lengths than short-day plants, reflecting
dition (see Chapter 8), and photoperiod is generally the environmental hazards of being an “early bird” or
involved in initiating the physiological changes re- a “hanger-on.” Many economically important
quired for dormancy. conifers are long-day plants, becoming dormant if
grown at photoperiods of 16 hours or less, and many
achieve their full growth potential only under 24-
Role of Photoperiodism in Plants hour photoperiods. Some species, such as the coastal
Photoperiodism in plants plays a major role in the redwood and Monterey pine from California, are rel-
control of the cessation of growth and the onset of atively day neutral (Table 7-7), reflecting the lack of
dormancy in the late summer or fall, and in many any selective advantage of having a photoperiodic re-
plants, it regulates flowering and fruiting in the spring sponse in the equitable environments in which they
and summer. It also plays a role in the breaking ofdor- have evolved. Figure 7-12 shows the variation in
mancy and resumption of growth in the spring in growth response to varying photoperiods between the
some perennial plants. It is thought that control of relatively short-day Japanese larch, which enters dor-
flowering and fruiting in woody plants is generally not mancy at photoperiods of less than 14 hours, and
under direct photoperiodic control, but our under- white spruce, which enters dormancy at photoperiods
standing of many aspects of photoperiodism in trees is of less than 16 to 20 hours.
still incomplete as most of the research on this topic Because of the importance of photoperiodism to
has been conducted on herbaceous plants. Much of the survival and fitness of organisms, it is not surprising
the work carried out on woody plants has used that there is considerable variation in photoperiodic
seedlings or cuttings, and photoperiodic responses of response in a species across its geographical range, and
mature trees, particularly the control of reproduction, photoperiodic ecotypes have been recognized in many
are not yet well known. Flowering in many tree species. The date of occurrence of unfavorable temper-
species seems to be day neutral, and the trees will ature and moisture conditions will vary between the
break spring dormancy equally in total darkness or coast and inland, between a frosty valley bottom and
continuous light, indicating that termination of dor- valley slopes, between high and low elevations, be-
mancy is not a light-mediated phenomenon (Pauley, tween north and south portions of the species range,
1958). Temperature frequently acts as an important and between north and south aspects. Consequently,
complementary factor in photoperiodic responses. the photoperiodic “calendar” selected for by organisms
Short-day plants are those that flower naturally only that live in these different environments may vary.
under short days and long nights and are found only at In a study of ecotypic variation in photoperiodic
middle or low latitudes. Short days do occur at higher responses, seed was collected from black cottonwood
latitudes, but they are more frequently accompanied by (Populus trichocarpa) growing over a wide geographical
frost or snow than at lower latitudes. Spring short-day area and grown near Boston, Mass., at approximately
plants include certain annuals and perennials. The seed 42° north latitude. The date at which the seedlings en-
of the annuals must germinate early in the spring if the tered dormancy was recorded and compared with the
 Ecological Effects of Temporal Variations in Solar Radiation 195
Table 7-7 Effect of the Length of Photoperiod on Growth in Height and Gain in Weight of Various Conifers

Percentage of Maximum Growth in Percentage of Maximum Growth in

Height at Various Photo Weight at Various Photo :
periods, hr periods, hr Duration of
Y Experiment
Species 8 12 14 16 20 24 8 12 14 16 20 24 (months)

Coastal redwood — 77 94 100 90 77 - 68 87 100 93 79 WZ

Monterey pine 55 59 64 64 — 100 55 67 68 60 _ 100 iG
Japanese larch _ 8 44 62 69 100 — 17 71 66 80 100 4
Sitka spruce - 13 ee es —_ 100 = 14 20 71 = 100 U7
Douglas-fir _ 4 11 51 100 88 _ 4 23 83 ThE 100 =
Jack pine 12 20 40 56 - 100 4 14 39 88 — 100 8.5
Eastern white pine 3 13 13 23 = 100 1 3 14 23 - 100 15
White spruce _ 13 16 28 50 100 —_ 14 Ai 28 50 100 4
Norway spruce 1 1 1 8 - 100 1 1 1 9 _ 100 11

Expressed as percentages of maxima achieved with optimum photoperiods.

Based on data from Downs, 1962.

latitude of origin (Figure 7-13). A clear relationship dormancy is delayed, the plants may not become dor-
between latitude of origin and date of cessation in mant before the first killing frosts, and plants that are
height growth was observed, with some southern eco- moved north typically suffer a higher incidence of
types (which had been moved north) continuing frost damage. Flowering may also be inhibited because
growth until late October. In fact, some did not stop of the long photoperiods. However, long-day plants
growing until their shoots were killed by the first fall that are moved south often experience reduced growth
frosts. The date of cessation of growth also showed a or even dwarfing because the photoperiod is so short
good relationship with the length of the growing sea- that it may permit only a short period of release from
son in the region of origin, indicating that the length dormancy or none at all. In the black cottonwood
of the growing season is a genetic characteristic of a study just mentioned, movement of northern ecotypes
species that is related to and modified by variations in southward reduced annual height growth from 130 cm
the photoperiod. Both relationships depicted in to between 15 and 20 cm, height growth being termi-
Figure 7-13 exhibited considerable variation. Seed nated in late June, whereas ecotypes from southern lo-
from latitudes 44 to 46°N varied in date of cessation of calities grew as much as 200 cm (Pauley, 1958).
height growth from mid-July to late October, and eco- Flowering in the northern ecotypes that were moved
types that stopped growing in mid-September were south was inhibited because the photoperiod never be-
from regions with growing seasons of between 140 came long enough to initiate reproduction. Ecotypes
and approximately 220 days. This variation reflects are genetic subdivisions of a species that are adapted to
variations in local conditions as a result of aspect, ele- different physical environments. Photoperiodic adap-
vation, and so on, and the interaction of photoperi- tations are one of the more important aspects of eco-
odic responses with factors such as temperature. It is typic variation (Chapter 16).
also an expression of the natural variation in charac- It should be apparent from these examples that
teristics that can be found in most populations. photoperiodic adaptations will help to limit the nat-
When short-day plants are moved northward, they ural migration and extension of a species’ range.
often grow larger than they would in their native lo- They can also pose a major limitation on the artificial
cality because of the greater number of hours of light extension of a species’ range, such as the planting of
per day and because the short photoperiod required to “exotic” trees (species or ecotypes from a different
stimulate dormancy does not occur until very late in country or region) by foresters. The success in using
the summer or fall. However, because the onset of North American species at more northerly latitudes
196 CHAPTER 7 Ecological Role of Solar Radiation

7.6 Importance of the Light

| WY\\
ys! Factor in Forestry
A\ NN
(ry
There can be little doubt that the light factor plays an
important role in the management of forest ecosys-
AYPAN)
PTA tems. The following are a few examples of the practi-
S Uy
; aN a
Gx PZ. cal importance of the topics discussed in this chapter.
aX? The forest nursery manager should be aware of the
1Sire influence of light on plant morphology. Maximum
Nae
= 50g 7's
seedling height growth for some species may be ob-
: KS uN Z =e

Dy KE DEDD og SSS NUS iy

tained by partial shading or by dense stocking of
Re TEUD) ————] be - a3
seedbeds, but the resulting seedlings will generally
have thinner and weaker stems than seedlings grown
in full sunlight, and they will have higher shoot/root
10 hrs 12 hrs 14 hrs 16 hrs 24 hrs ratios. For plantation establishment in a moist, shel-
(a) tered site, these characteristics may be desirable, but
they are unlikely to help the seedling to survive if
planted on an open, exposed site. Seedlings grown in
the nursery or the forest at low light intensities will
have shade leaves. If suddenly exposed to full sunlight
and the accompanying altered temperature and mois-
ture conditions by being planted on an exposed site,
these plants will experience some degree of light
shock. This may either kill them or result in a reduc-
tion or cessation of growth for | year or more until ap-
propriately adapted sun leaves are produced; the
poorly adapted foliage is frequently shed by the plant.
Post-planting height growth delay (planting shock)
may also reflect the seedling’s development of its root
system to produce shoot/root and height/diameter ra-
12 hrs 14 hrs tios that are appropriate for the site conditions. Nurs-
(b) ery light conditions should be manipulated to produce
seedlings with shoot/root ratios, leaf morphology, leaf
Figure 7-12 Growth of conifer seedlings over a 4-month pigmentation, and height and diameter characteristics
period at various photoperiods. (A) Japanese larch. (B) White
that will give them the maximum chance for survival
spruce. (After Downs, 1962. Copyright Ronald Press. Redrawn by
permission ofJohn Wiley & Sons, Inc., and R.F. Downs.)
and early growth when planted out.
Failure to recognize the role of photoperiodism in
synchronizing the activities of plants and animals with
variations in the physical environment can result in
in Britain, maritime Europe, and Scandinavia reflects planting failures, susceptibility to frost, and poor
the increased growth of ecotypes when moved north; growth. Planted seedlings must have received the cor-
the relatively moderate climate of these areas (be- rect light and temperature conditions in the nursery
cause of the Gulf Stream) removes the normal im- and in storage if their physiology is to be synchro-
pediment on south-to-north movements of plants. nized correctly with their new environment. The
The reverse process of moving high-latitude eco- generally poor adaptation to the physical environ-
types native of Europe into more southerly areas in ment of ecotypes from very different elevations at the
the United States has not met with notable success same latitude or from different latitudes at the same
(Pauley, 1958). elevation renders them susceptible to climate-
For a more detailed treatment of photoperiodism induced stress, competition, and pathogens. Having
in plants, see Salisbury and Ross (1969). said this, it must be noted that establishment of plan-
 Importance of the Light Factor in Forestry 197

60

50

Latitude
origin,
of
N 40
Latitude at which the
trees were grown

30
June July Aug. Sept. Oct. Nov.

150

100
Length
of
growing
in
place
days
origin
season

July August September October

Date of height-growth cessation when grown together at latitude 42° N

Figure 7-13 Experimental demonstration of photoperiodic ecotypes in black cottonwood.

Variation in the date when height growth stopped in black cottonwood ecotypes collected from a
wide latitudinal range and grown together in Boston, Mass. (approximate latitude 42°). The variation
of the data about the mean reflects differences in elevation and climate in different localities at any
one latitude. The date of cessation in height growth, a genetically controlled photoperiodic re-
sponse, reflects the average length of the growing season in the location of origin. (Pauley, 1958.
Copyright Ronald Press. Redrawn by permission offohn Wiley & Sons, Inc., and S.S. Pauley.)

tations with seedlings of nonnative photoperiodic there have been so many cases of plantation failures
ecotypes will not always have undesirable results. A resulting from the use of nonnative ecotypes that the
change in photoperiod may suppress reproduction of forester must give careful consideration to the pho-
forest trees so that all of the available nutrients and toperiodic adaptation of the plants that are to replace
energy are put into vegetative growth. However, the harvested crop.
198 CHAPTER 7 Ecological Role ofSolar Radiation

Damage to plants as a result of a sudden increase in

light intensity often involves the foliage (e.g., damage
to conifers when released from competing vegetation),
but light can also affect stems because of the heating
effects of solar radiation. Sun scald, the killing of bark
by lethally high temperature after sudden exposure to
sunlight, commonly occurs when thin-barked species
that grow in dense stands are suddenly exposed to di-
rect sunlight by the removal of adjacent trees. Sun
scald is mainly a winter phenomenon because it is as-
sociated with rapidly fluctuating temperatures rather yr~!)
ha”!
(Mg
ANPP
than with high temperature alone. Another effect of
sudden stem exposure is the stimulation of new or
epicormic branching. This is common in_ previously
shaded hardwood stems that experience a great in-
crease in light intensity, but it is also seen in some 0 D) 10 15 20
conifers. Epicormic branching can have important im- Foliage biomass (Mg ha!)
plications for timber quality.
Failure to recognize light intensity-plant mor- Figure 7-15 Relationship between net aboveground pro-
phology relationships in the forest stand can result in duction and foliage mass in Douglas-fir growing on low-,
poor resistance to wind and snow damage or poor medium-, and high-productivity sites in coastal British
Columbia. (Redrawn after Kurz, 1989.)
growth form with heavy branching. As noted earlier,
self-shading within the canopy reduces foliage pro-
duction efficiency (Figure 7-14). As a result, maxi-
mum production may not occur at maximum levels of mum growth may occur to intermediate levels of fo-
foliage biomass or leaf area. In shade-tolerant species, liage. The level of foliage that is optimum is a function
total production may increase up to high levels of fo- not only of light intensity but also of site moisture and
liage mass, but in less shade-intolerant species, maxi- nutrient status (Figure 7-15).
The light regime created by stand thinning will
have a major influence on the species diversity of nat-
ural regeneration. In some ecosystem types and some
ecological zones, the high light conditions created by
clearcutting may be necessary to obtain natural regen-
wn—Oo eration of light-demanding species and to ensure that
all the local species, from shade-tolerant to shade-
intolerant, have the opportunity to regenerate. Ma-
nipulation of light condition in the stand to give trees
of desired form and timber quality and to obtain de-
Red pine sired species composition and stand structure is a
major objective of silviculture.
oe Wa Douglas-fir
ee spruce

SUMMARY
!-yr7!)
productivity
Nitrogen
mass-kgN~
(kg
The fundamental role of solar radiation is as the major
Leaf mass (Mg ha!) source of energy for life, but this is not its only ecological
effect. The interaction of solar radiation with water and
Figure 7-14 Variation in foliage production efficiency the atmosphere maintains global temperatures within a
(expressed as “nitrogen productivity”) as foliage mass in- range that has proved to be suitable for the evolution of
creases in red pine (shade-intolerant), Douglas-fir (inter- life. Without the temperature-regulating aspects of solar
mediate shade-tolerant), and Norway spruce (shade-toler- radiation, life as we know it could not have evolved and,
ant). (Redrawn after Agren, 1983.) given our present technology, could not continue.
 Take-Home Message 199

Although the energy aspects of solar radiation are ex- requirements of the crop is important and will play a
tremely important, they are preeminent only for the pho- major role in how a particular site is managed. Radiation
toautotrophs. For most heterotrophic organisms, other periodicity can be varied significantly only by moving
aspects of the sun’s energy emissions are more important. plants to different geographic locations. Thinning or
Vision, the ability to see one’s surroundings, is of critical partial harvesting in broad-leaved forests may alter the
importance to most animals, some of which have become spectral quality of light, and this can affect understory
adapted to “see” wavelengths other than the visible or to plants, but this qualitative change is generally of less im-
see at light intensities that are so low that for most other portance than variations in light intensity.
organisms it is dark. Color, the visual response to differ-
ent wavelengths, adds an ecologically important dimen-
sion to vision, which has led to adaptations for
TAKE-HOME MESSAGE
concealment, mimicry, and a variety of visual signals that Public concern about forests and how they are managed
assist organisms in organizing their behavior in a way has generated strong pressure against clearcutting, tree
that increases their genetic fitness. planting, and even-aged forests and in favor of partial
Variations in light intensity as the result of either lo- harvesting, natural regeneration, and uneven-aged
cation or competition have resulted in a wide variety of forests. At the same time, there is demand for mainte-
adaptations, especially in green plants. For such organ- nance of “biodiversity.” Evaluation of whether the silvi-
isms, light is potentially both beneficial and harmful, and culture demanded by some groups would conserve the
adaptations include protective mechanisms as well as diversity of plant species and the animals and microbes
mechanisms to enhance the exposure to and interception that depend thereon more or less than the silvicultural
of light. These involve both morphological and physio- methods used in the past depends, to a considerable ex-
logical modifications that enable organisms to survive tent, on how the forest practices being advocated would
and compete successfully in various light environments. affect light environments and how native plants, animals,
Adaptations to variations in the quantity of solar radia- and microbes would respond to these light environments.
tion are not restricted to plants; animals also exhibit a va- If foresters are to make appropriate decisions with respect
riety of adaptations to light intensity. to this public pressure, then they must understand these
Most organisms require a sense of time—an environ- responses and be able to predict their consequences for
mental clock and/or a calendar by which their activities the future species composition, productivity, and struc-
can be scheduled to take advantage of favorable condi- ture of the forest ecosystems that they are managing.
tions and to avoid adverse conditions. The daily and sea- With increased interest in natural regeneration for
sonal variation in light intensity and the seasonal both environmental and economic reasons, foresters
variation in photoperiod apparently provide the best must become much more familiar with the light re-
available environmental timepiece; in most terrestrial or- sponses of noncrop species as well as the trees that they
ganisms, light seems to have been selected by evolution wish to regenerate. This was common knowledge among
as the most important timekeeper. However, few organ- silviculturists when natural regeneration was the major
isms exhibit a slavish adherence to photoperiod. It is method of forest renewal, but this experienced-based un-
more common for light to be one of several factors that, derstanding was largely lost as planting in clearcuts be-
in combination, schedule the biological events in the lives came the common system. Among other things, foresters
of living things. will have to become much more familiar with the aute-
Because organisms have adapted to particular solar- cology and ecophysiology of the plant communities that
radiation conditions, care must be taken when moving or- they are managing if natural regeneration is to be accom-
ganisms from one geographical location to another. Rais- plished successfully.
ing a plant ina light environment that is different from that Conservation of genetic diversity involves, among
of its native location can sometimes induce growth pat- other things, conserving the photoperiodic adaptations
terns that are desirable for human uses of the plant, but op- of species. In the face of predicted human-induced cli-
timum growth and survival are generally obtained when a mate change, maintaining this aspect of genetic diversity
plant genotype is kept in its native light environment. is particularly important. Concern about increases in UV
Of the three major aspects of solar radiation that can radiation associated with pollution-caused loss of the
be varied on any particular site, the forest manager can protective ozone layer raises new and at present unan-
control only the intensity factor. This involves the choice swered questions about the response of forest ecosystems
of harvest method (clearcutting versus selective or small to variations in the spectral quality of sunlight.
patch cutting) and the manipulation of light competition Clearly, foresters of the future will have to pay more
between crop trees and other vegetation early in the life attention to the light environments that they are manag-
of the crop and between individual crop trees later on in ing if they are going to achieve the public’s expectations
the life of the stand. Consideration of the light intensity for the management of public forests.
200 CHAPTER 7 Ecological Role of Solar Radiation

STUDY QUESTIONS . Can we predict the regeneration and growth of trees

in a forest from a knowledge of their photosynthetic
1. What is light? light saturation curves?
. In what ways is the spectral quality of light important
. Why is the light regime so important in nursery prac-
in ecosystem function and forest management?
tice and silviculture?
. What is shade tolerance?
10. Why do temperate and northern deciduous trees
. How does light intensity affect plants? generally have a higher photosynthetic efficiency
. What is the role of photoperiodism in organizing the than temperate and northern evergreens?
lives of plants and animals?
EL: How might an understanding of photosynthetic light
6. Why is it important that foresters exercise strict con- saturation curves and other light-related plant adap-
trols on the planting of different photoperiodic eco- tations of different species be used by a forester to de-
types? sign the density, structure, and species composition of
. Can trees grow at light intensities below their CP? forest stands?
 ‘Temperature as an
Ecological Factor

8.1 Introduction The temperature of a body generally goes up as

heat energy is added and drops as it is lost, but this is
Temperature is probably the environmental factor not always the case. Water gives out a great deal of
most familiar to the average person, a fact that sug- heat energy without changing its temperature as it
gests its ecological importance. Our cultural evolu- freezes, and ice absorbs a lot of heat energy as it melts,
tion has made us dependent on an artificially again with no change in temperature. This exception
regulated temperature environment, and many of to the general rule is due to the /atent (or hidden) heat
the cultural variations between different races, such of crystallization. The latent heat of vaporization results
as style of clothing, type of housing, and daily and in a similar phenomenon as water changes from the
seasonal activity patterns, reflect the overriding in- liquid to the vapor state. Because of the large amount
fluence of the temperature factor. A significant of latent heat associated with freezing and vaporiza-
portion of the human economy is expended in over- tion and because of its high thermal capacity, water re-
coming the constraints that the temperature factor tains its liquid properties under widely varying levels
places on human activity and population develop- of heat energy. It has been estimated that the melting
ment. Similarly, much of the energy budget of most of the winter ice on Lake Mendota in Wisconsin in-
terrestrial and some aquatic animals is allotted to volves the absorption of heat energy equivalent to
maintaining within acceptable limits the tempera- 195,000 tons of anthracite coal, with no change in
tures that they experience. temperature (Allee et al., 1967). A similar quantity of
Temperature is a measure of the intensity or con- heat must be given off before the water can refreeze,
centration of heat energy in an object. It is deter- resulting in a substantial delay in transformation of the
mined by both the amount of heat energy in and the liquid back to the solid form. The remarkable temper-
heat capacity of the object. Thus, a piece of dry wood ature properties of water make it a suitable medium
(low thermal capacity) will have a much higher tem- for life processes and provide a partial explanation as
perature than a piece of iron (high thermal capacity) to why all living organisms conduct their chemical
of the same size when both objects contain the same processes in solution.
amount of heat energy; when both the iron and the In this chapter, we consider how the temperature
wood have the same temperature, the iron will con- of the environment varies from place to place and
tain much more heat energy than the wood. A small from time to time and the significance of these varia-
addition of heat energy to the piece of wood will raise tions for plants and animals. We then discuss the role
its temperature significantly, whereas the same of temperature in limiting the geographical distribu-
amount of heat energy added to the iron will alter its tion of species and the importance of the temperature
temperature very little. factor in forestry.

201
202 CHAPTER 8 Temperature as an Ecological Factor

8.2 Geographical and Temporal human-induced climate warming (see Section 8.8.
Human Effects on the World Radiation Budget and
Variations in [emperature Global Temperatures and Chapter 20). There is now
Temperature exhibits a number of well-defined cycles an extensive literature on atmospheric temperatures,
of variation that are directly attributable to the rotation which can most easily be accessed through the Inter-
of the earth around its axis and around the sun. These net or documents of the Intergovernmental Panel on
Climate Change (Watson et al., 2001). This topic is
rotations lead to a daily and seasonal variation in the
amount of radiant energy that reaches a particular part revisited at the end of this chapter.
of the earth and consequently in its temperature. In the The three most important variables that determine
tropics, the diurnal variation in temperature may be the geographical variation of temperature are latitude,
only a few degrees, whereas in continental regions, it altitude, and the proximity of large bodies of water
can be as much as 50°C in either winter (e.g., —49 to such as oceans or large lakes. Both latitude and altitude
6.5°C in Montana) or summer (e.g., 2 to 50°C just influence the radiation budget, whereas convection
above the ground in some deserts). These large ranges and conduction processes move heat from large water
sometimes involve extremely rapid changes in temper- bodies to the air in winter and from land to the air in
ature, such as a 27°C change in 2 minutes recorded in summer. Temperature extremes are greatly modified
South Dakota. Islands in the Pacific Ocean may have in the vicinity (especially downwind) of large bodies of
an annual variation in daily mean temperatures of as water. Aspect and topographic position are also impor-
little as 11.8°C (19.6 to 31.4°C), whereas the compara- tant determinants of local temperature conditions.
ble figure for parts of Siberia is 107°C (—70 to 37°C).
The lowest recorded temperature is —88°C in Antarc-
tica, whereas temperatures as high as 60°C for air and
Radiation Budget
84°C for surface soil have been recorded in desert areas When viewed from space, the earth has an effective
(Clarke, 1967; McWhirter and McWhirter, 1975; radioactive temperature of approximately —18°C,
Spurr and Barnes, 1973). whereas the average surface temperature is 15°C. The
Temperature exhibits long-term as well as short- difference of 33°C is the result of a positive balance
term and geographical variations. On a geological between incoming and outgoing short- and long-
time scale of hundreds of thousands of years, tempera- wavelength radiation (McBean and Thomas, 1991).
tures have varied as ice ages have come and gone, ac- This positive balance is attributable largely to atmos-
companied by shifts in the location of the major pheric water vapor and CQ), which are translucent to
climatic zones. Variations in mean annual temperature visible wavelengths but opaque to long wavelengths
of as much as 10°C have occurred since the last glacial (Figure 7-2). The “greenhouse effect” of the atmos-
period (the Wisconsin glaciation), and although this phere maintains global temperatures at a level suitable
may not result in great changes in a hot climate, it is of for life as we know it and at which much of the H,O
great significance in a cold climate. A 1°C reduction in on Earth is in liquid form. This global importance of
the mean annual temperature (1°C drop in every radiation budgets is mirrored by the importance of ra-
monthly average) of Akureyri, Iceland, would reduce diation budgets at all levels in forests, from regional
the growing season from 158 to 144 days, and a 2.4°C climates to microclimates, and the temperature of in-
drop would reduce it to 118 days (Bryson, 1974). A dividual objects.
“little ice age” occurred from the 16th to the 19th cen- The temperature of a terrestrial object is often
turies, followed by a general warming period of nearly largely determined by its radiation budget: the balance
100 years. There was a rise of up to 1.7°C in the mean between the quantity of radiant energy that impinges
annual temperature of the eastern United States, a on the body from solar or terrestrial sources and the
change in climate equivalent to a 300-m change in el- quantity of energy that leaves the body as a result of
evation or approximately 170-km change in latitude reflection or reradiation. Heat exchange by conduc-
(Spurr and Barnes, 1973). This warming trend, which tion and convection also contributes to the determina-
was reflected in the retreat of glaciers (see Chapter tion of temperature, but radiant-energy exchanges are
17), was thought to be ending (e.g., Wahl, 1968; Wahl frequently more important in terrestrial situations. In
and Lawson, 1970), but predictions of cooling have contrast, the temperature of objects in aquatic situa-
now been displaced by widespread concerns about tions is mainly controlled by conduction.
 Geographical and Temporal Variations in Temperature 203

Because of the close relationship between the agreement with Figure 7-2, which shows the impor-
radiation budget and temperature, geographical and tance of long-wavelength radiation in the global radia-
temporal variation in temperature is similar to geo- tion budget.
graphical and temporal variation in net radiation bud- The radiation budget of an object is greatly influ-
gets. Days are warmer than nights. Cloudy areas are enced by its “view factor”: the proportion of its radiat-
cooler in the day and warmer at night than clear areas ing environment that is accounted for by the sky. A
(Figure 8-1). Northern aspects are cooler than south- rabbit in the middle of a flat area of desert has a “view”
ern aspects (in the northern hemisphere). of all of the sky and therefore has a high view factor; a
The radiation budget of an organism consists of seedling growing beneath a small opening in the forest
energy gains through the absorption of short- (solar) canopy has a very small view factor. Almost its entire
and long- (atmospheric and terrestrial) wavelength ra- radiating environment is occupied by overstory trees.
diation and energy losses from the emission of long- On a clear day, a location with a large view factor will
wavelength radiation. At temperatures above absolute have a strongly positive net radiation budget because
zero (—273°C), all objects emit radiant energy, the of the high intensity of short-wavelength solar radia-
wavelength of which varies with the temperature of tion and high temperatures may occur. On a clear
the object: the higher the temperature, the shorter the night, such a site will have a strongly negative net bud-
wavelength. Table 8-1 presents a radiation budget for get and low temperatures may occur. The “cold,” clear
a meadow community in central Europe in summer sky, on the one hand, will emit or reflect very little
and winter, showing the importance of long-wave- long-wavelength radiation to balance the outgoing ra-
length radiation in comparison with the more readily diation. A situation with a small view factor, on the
perceived short-wavelength radiation. This is in other hand, will have only weakly positive and nega-

Cloudy sky Clear sky

re

>) : eu i SS
YMMIIIM Us, Wl
Strongly positive radiation budget; rapidly increasing
Weakly positive radiation budget; slowly increasing
temperatures; little high-temperature hazard. temperatures; great high-temperature hazard.

(a)

Cloudy sky

Weakly negative radiation budget; slowly dropping Strongly negative radiation budget; rapidly decreasing
temperatures; little ground frost hazard. temperature; great ground frost hazard.

(b)

Figure 5-1 Effect of cloudiness on daytime (A) and nighttime (B) radiation budgets.
204 CHAPTER8 Temperature as an Ecological Factor

Tuble 8-1 Summer and Winter Radiation Budgets* in a Central European Meadow, Langleys/Day
June December

Short Long All Short Long All

Radiation Source Wavelength Wavelength Wavelengths Wavelength Wavelength Wavelengths

Downward from space +248 - +248 +12 - -12

direct solar radiation
Downward from atmosphere +218 +728 +946 +39 +582 +621
skylight and atmosphere
reradiation
Upward from Earth’s -93 -807 -900 -18 -621 -639
surface reradiation
and reflection
Net budget +373 -79 +294 +33 -39 -6

Data from Reifsnyder and Lull, 1965.

4A positive net radiation budget indicates increasing temperatures, whereas a negative budget indicates falling temperatures.

tive day and night net budgets, the latter because in- sulting wide diurnal temperature fluctuations and high
puts of long-wavelength radiation from the surround- light intensity.
ing vegetation at night partially compensate for Radiation budgets at ground level are greatly af-
radiation losses (Figure 8-1). The view factor is less fected by vegetation. We saw in Chapter 7 that there is
important in cloudy than in clear weather because of a great reduction in short-wavelength solar radiation
the reduced intensity of daytime short-wavelength ra- as light passes down from the upper surface of the veg-
diation and because of long-wavelength reradiation etation to the forest floor and that this results in lower
from the cloud at night. In a forest situation, the view summer air and soil temperatures below the forest
factor of a seedling is determined by micro relief canopy than in adjacent clearings. The forest vegeta-
(stumps, logs), aspect, and the size of canopy opening tion also acts as a blanket, eliminating the view factor
relative to the height of the surrounding trees. The and covering the area with a radiation-absorbing and
importance of tree height, clearing size, and view fac- -emitting layer. Net nighttime radiation losses from
tor in determining radiation balances is reduced in the soil, understory vegetation, and ground-dwelling
areas with predominantly cloudy climates. fauna beneath a forest canopy are much less than in
Radiation budgets are greatly modified by albedo. open areas, and there is a reduced risk of ground frost
Energy that is reflected rather than absorbed does not (radiant cooling of the soil and adjacent air layers to
increase the heat content of an object and therefore below 0°C). Winter nocturnal soil and air tempera-
does not affect its temperature. Forest canopies, which tures are often higher inside the forest than in adjacent
have a low albedo, warm up rapidly, whereas a white clearings, but daytime temperatures are generally re-
sandy beach or snow surface reflects most of the solar duced because of the weaker daytime net radiation
energy that is received, and such surfaces warm up budget. Shrubs, herbs, and even a moss layer can exert
slowly. Snow covered with dust melts faster than clean a similar insulating effect on the ground.
snow because of the lowered albedo and increased ab- Forest cover reduces the range of temperatures
sorption. Dark-colored soil absorbs much more radi- (annual, monthly, or diurnal), and the mean monthly
ant energy than light-colored soil and therefore range may be reduced by as much as 7°C in July and
reaches much higher temperatures. However, the air 3°C in January. Table 8-2 shows the influences of for-
temperature above the lighter-colored soil may be est cover on air temperatures in various U.S. forests.
greater than above dark-colored soil because of the Maximum temperatures are generally reduced by for-
greater reflection from the former (Figure 8-2). A est cover, whereas minimum temperatures are gener-
layer of snow increases the daytime temperature of the ally increased. The reduction in monthly maxima is
air just above it (because of light reflection), and plant more pronounced in July (up to 4.6°C) than in January
parts just above the snow may be damaged by the re- (up to 1.7°C), whereas the increase in monthly
 Geographical and Temporal Variations in Temperature 205

minima can be as much as 3 to 4°C in both summer

and winter. These data apply mainly to closed forest
stands. Shrub and open forest communities can pro-
duce the opposite effect, with increased monthly tem-
perature ranges (Kittredge, 1948).
The temperature effects of a net radiation budget
are greatly influenced by the thermal capacity of the
°C
Soil
temperature
objects that receive the radiation. Water has a very
high thermal capacity; therefore, the moisture content
of an object affects its temperature response. A wet
soil warms up and cools down more slowly than dry
soil, and the high water content of living organisms
Noon Midnight Noon
Time of day helps to regulate their temperature fluctuations.
Water bodies such as large lakes and oceans have an
(a)
enormous thermal capacity, and they experience only
minor and very gradual temperature changes.

Conduction and Convection of Heat

The radiation budget is not the sole determinant of
the heat energy content of an object; other processes
°C
Soil
temperature are also involved. Plants, animals, and soil lose or gain
heat from the atmosphere by conduction and/or con-
F ie vection. This can maintain low temperatures even
JV ASS Vine ee Aus SON ID)
when there is a positive net radiation budget, and vice
Time of year (months)
versa. The importance of conduction and convection
(b) to the temperature of an organism depends to some
extent on the thermal properties of its surface. Where
conductivity is low, such as tree bark, seal blubber, or
the fur of mammals at high latitudes, conduction and
convection are of reduced importance. In warmer cli-
mates, where efficient loss of excess body heat is nec-
essary, these processes are more important.
Distance
below
&
above
surface
soil In addition to their contribution to the thermal
balance of individual organisms, conduction and con-
Temperature (°C)
vection play a major role in world ecology by transfer-
(c) ring energy between land and water surfaces. Air in
Figure 8-2 Soil temperature profiles and their diurnal and contact with water exchanges heat by conduction, and
annual variations and variations according to color, mois- the resulting atmospheric temperature gradients lead
ture content, and exposure to light. (A) Diurnal variation in to convection currents that distribute the effects of the
temperature at various depths in a subalpine soil in British Co- heat exchange over large areas. The enormous store of
lumbia. (After Ballard, 1972. Reproduced with permission of the summer heat in a large lake or an ocean warms the air
Regents of the University of Colorado.) (B) Variation in mean passing over it during the winter, raising temperatures
monthly temperature at various depths in the soil at Bozeman, in downwind land areas. In the spring and summer the
Montana. (After Fitton and Brooks, 1931. Reproduced with per- relatively low temperature of large water bodies re-
mission of the American Meteorological Society.) (C) Variation in sults in the opposite effect. The Japanese current in
vertical gradients of soil temperature with color, shading, and the north Pacific and the Gulf Stream in the north At-
moisture content. (After Daubenmire, 1974. Reproduced with
lantic are good examples. They deliver warm water
permission offohn Wiley & Sons, Inc.) Key: | = grey, dry, shaded
from tropical to higher latitudes on the west coasts of
soil; 2 = grey, wet, exposed soil; 3 = white, dry, exposed soil; 4 =
grey, dry, exposed soil; 5 = black, dry, exposed soil. North America and Europe (Scandinavia), which
206 CHAPTER 8 Temperature as an Ecological Factor

Table 8-2 Effects of Forest Cover on Air Temperatures Near to the Ground
Departures from Air Temperature in the Open, °C

Monthly Monthly
Mean Maximum Mean Minimum

Forest Type in the U.S. January July January July

oe

Beech, N.Y. — -4.6 - -2.5

White pine, Idaho _ -3.3 — +3.9

Hemlock, N.Y. - -2.7 — +2.2

White pine, N.H. _ -2.7 _ +1.8

Douglas-fir, 9,000 ft. Colo. -1.7 -1.2 +1.2 +1.4

Jeffrey pine, 6,000 ft. Cal. +0.4 +0.8 +2.0 +3.7

Jack pine, Neb. +0.1 -0.7 +0.6 -1.6
Aspen-Engelmann spruce, 9,000 ft. Colo. -1.0 -0.6 -0.1 -0.6
Ponderosa pine, 7,250 ft. Ariz. -0.6 0 +3.3 -2.8
Chaparral, 6,000 ft. Calif. +0.8 +2.8 -0.9 -1.2

Kittredge, 1948. Copyright D.D. Kittredge. Used with permission.

warms the air moving from the ocean to the land in ty of roots and soil organisms to continue throughout
these areas. This produces a remarkably equitable cli- the autumn and into winter, long after the above-
mate in these coastal areas, permitting the northward ground parts of plants have become dormant.
extension along the coast of many lower-latitude Many northern and high-elevation soils freeze in
species. The Alaskan coast and northern coastal Nor- the winter. In very severe climates, this occurs irre-
way are warmer in January than areas in the middle of spective of snow cover, and the soil may be perpetual-
large continents several thousand miles to the south. ly frozen throughout most of its depth (permafrost). In
In contrast, the east coast of Canada experiences cur- somewhat less frigid climates, soils may remain more
rents from cold northern waters, and these make the or less unfrozen for most or all of the winter if a deep
climate of Labrador and Newfoundland much colder layer of snow accumulates before low temperatures
in the summer than comparable latitudes elsewhere. occur. The snow acts as an efficient insulation against
Conduction plays an important role in determin- low winter temperatures, and animals such as lem-
ing the distribution of temperature with depth in the mings can continue their activities uninterrupted be-
soil. Dry organic soils have an extremely low thermal neath the arctic snow. Snow also reduces radiation
conductivity (heat energy moves through them slow- losses from the soil. Where there is permafrost, plant
ly), whereas the conductivity of mineral soils is low to roots are confined to the superficial “active layer” of
moderate. Wet soils conduct heat better than do dry soil, which melts in the summer. This restricts root
soils, but even in wet mineral soils, heat conduction is growth and the availability of nutrients and is an im-
not fast enough to give a uniform distribution of tem- portant determinant of the low productivity of north-
perature through the soil. Peak soil surface tempera- ern forests. Deeper thawing of permafrost soils as a
tures may be reached at noon or early in the result of warmer summers, removal of vegetation, or
afternoon, but peak temperatures at successively deep- disturbances of surface organic layers improves the
er points in the soil occur at successively later times. availability of nutrients and is associated with im-
Similarly, minimum diurnal temperatures in the soil proved plant and animal productivity (Lutz, 1956;
lag behind minimum surface temperatures (Figure Schultz, 1969).
8-2A). There is also a lag in the annual variation in Where frozen mineral soil is exposed to sunlight in
soil temperatures, with maxima and minima at depth the spring and early summer, its temperature rises
in the soil occurring several months after those at the rapidly. Where it is covered by a layer of organic mate-
soil surface (Figure 8—2B). This may permit the activi- rial, mineral soil warms up very slowly and may remain
 Geographical and Temporal Variations in Temperature 207

frozen for much or even all of the growing season. inversion may develop. During an inversion, air tem-
Much of the northern permafrost regions owes its eco- perature increases with increasing distance above the
logical condition to the layer of insulating organic ground up to a certain height and then resumes the
matter laid down by the vegetation, and massive soil normal lapse rate. Temperature inversions that occur
thawing, slumping, and erosion frequently follows the as the result of nocturnal radiant cooling generally
removal of the organic surface layer and exposure of break up as soon as the ground is warmed by sunlight
frozen soil. The concern over use of tracked vehicles in the following day, but in some circumstances, they can
oil exploration and construction of oil pipelines in the be more persistent.
Arctic is related to the thermal fragility of these north- Inversions can also occur as the result of topogra-
ern ecosystems. Regeneration problems in clearcuts in phy. Radiant cooling of high ground flanking a valley
some boreal forests in Alberta, Canada, have been at- gives rise to a layer of cold, dense air in contact with
tributed in part to low soil temperatures caused by the surface. This air flows slowly down the valley
prolific growth of grass (Hogg and Lieffers, 1991), and slopes, displacing warmer air in the lower part of the
low soil temperature is a dominant factor in the ecolo- valley and creating an inversion (Figure 8-3). When
gy of northern forests (Bonan, 1992). the cold air that drains into the valley is below 0°C,
frost occurs on the valley floor, whereas much warmer
temperatures will be experienced in the “thermal belt”
Role of Topography higher up the slopes. This is of great importance to
High-elevation areas have lower average temperatures fruit growers, and orchards are often located in the
than do low-elevation areas, because air temperatures thermal belt. Forest insect problems are sometimes as-
normally decrease at a rate of approximately 0.4°C per sociated with this belt because the milder winters ex-
100 m of elevation (the /apse rate) as one proceeds up a perienced in the thermal belt may increase the winter
mountain. The air at low elevation is denser and survival of insect defoliators ((e.g., the lodgepole pine
therefore absorbs more radiant energy than the air at needle miner in the Canadian Rocky Mountains
higher elevations; it also receives heat by conduction (Stark, 1959)). The warmer winter air temperature of
and convection from the ground. Under conditions of the thermal belt may also give rise to winter moisture
low atmospheric humidity, the lower, warmer air ex- stress in plants (ved be/t damage: see Chapter 9).
pands and becomes lighter than the colder air above. The cold air drainage that produces valley temper-
Convection currents develop in which plumes of ature inversions can also have important small-scale
warm air rise high into the atmosphere, forming cu- effects. Small topographic depressions tend to fill up
mulus clouds. When the lower air is very humid, the with cold air, creating frost pockets, and small openings
atmosphere is more stable, the convection columns do in the forest may fill with cold air and be frostier than
not develop as readily, and there is little atmospheric a large opening in which the cold air can drain away or
stirring. Should rapid nocturnal radiant cooling occur be swept away by winds. Frost pockets sometimes cre-
under these stable conditions, the air close to the ate local variations in plant and animal life in environ-
ground may become colder than the air above it and a ments that lack apparent variation in any physical
condition of atmospheric stability called a temperature attribute other than their topographic position.

SASK
SOK SS
SSSA eM
Atmosphere
SESS 7
tgs Cold air
¢

Landform |

Figure 8-3 Vertical temperature profiles in a valley on a cool, clear day (A) and a cold, clear
night (B). The daytime profile near the valley floor is inverted during the cold night.
208 CHAPTER 8 Temperature as an Ecological Factor

Soil Temperatures Climate and Microclimate

The temperature of soils has already been mentioned In Chapter 7, we saw that the light conditions expe-
several times, but its importance deserves some addi- rienced by two seedlings growing 10 cm apart on the
tional comment. Because of the high thermal capacity forest floor can vary greatly. Similarly, the tempera-
of water, wet soils require more heat to raise their ture conditions experienced by an insect that feeds
temperature than do dry soils; a volume of water re- on the uppermost branches of an overstory tree are
quires five times as much heat as the same volume of different from those experienced by an insect that
dry mineral soil to produce the same temperature feeds on minor vegetation. In discussing any cli-
change. Poorly drained organic and clay soils thaw matic factor, it is necessary to distinguish regional,
and warm up much more slowly than do loamy soils or macroclimate, from microclimate—the climatic
(moderately well drained), which are in turn slower conditions actually experienced by an individual or-
than freely drained sandy soils. Frost damage to roots ganism. The microclimate of a person dressed in a
in the fall is most common on light, freely drained sweater and overcoat is very different from the win-
soils that rapidly lose their warmth. The heavier, wet- ter macroclimate of the person’s hometown; the
ter soils are kept warm far into the fall by the large temperature experience of a diminutive spring-flow-
quantity of summer heat that they contain. ering herb is very different from that of a nearby
The color of soil has an important effect on its tree 20 m in height. Surface soil and adjacent air
temperature, as demonstrated in Figure 8-2C. Light- temperatures in late winter or early spring in a spot
colored soils absorb radiant energy less efficiently than sheltered from the wind may be high enough to pro-
do dark-colored soils and therefore stay cooler. Soil mote plant growth when the air 2 m above the
blackened by charcoal is an excellent absorber of radi- ground is still too low to break dormancy. An herb in
ation, with an albedo approaching zero. Dry, charred a shallow frost pocket may experience frost damage,
organic matter has a very low thermal conductivity, whereas a taller shrub next to it does not: the frost-
and this, together with its high absorptivity, can give sensitive organs of the shrub are located above the
rise to very high temperatures in a thin layer at the frost layer. In one study, it was found that a small
surface. In a study of soil temperatures, charred or- frost pocket had a frost-free period of only 77 to 104
ganic matter, grey mineral soil, and yellow mineral soil days, compared with 161 days for a nearby location
exposed to full sunlight on a day with an air tempera- with good cold-air drainage (Spurr, 1957). Con-
ture of 38°C were found to have surface temperatures versely, a plant in a small, steep-sided depression will
of 73, 64, and 62°C, respectively (Isaac, 1938). have a smaller view factor than one in a flat, open lo-
The importance of thermal conductivity in deter- cation, and it will be less susceptible to radiation
mining surface temperatures was shown in a study of ground frosts. Clearly, the regional macroclimate
white pine forest floors in the northeastern United merely sets the overall climatic framework within
States. Pine needle litter reached a temperature of which microclimate is determined by many modify-
68°C on a day when the air temperature was 24°C, ing factors.
whereas bare mineral soil reached 36°C and a moss car- In discussing the ecological effects of tempera-
pet reached 39°C. Soil color also affects soil tempera- ture, a further cautionary note is necessary. Temper-
ture by influencing reradiation. Light-colored soils are atures are frequently quoted as annual or monthly
less efficient at radiant heat loss than dark soils, and the means. By itself, the former is an almost worthless
surface of an organic soil can drop to below freezing piece of information, because a mild coastal climate
(because of efficient reradiation and poor conduction can have the same mean annual temperature as a
to replace the lost heat from deeper layers) under con- continental climate with bitterly cold winters
ditions in which a light-colored mineral soil would and very hot summers. Mean monthly tempera-
remain above freezing. The resulting damage to super- tures give a more realistic picture, but the com-
ficial feeding roots in organic soils can be lessened by bined graphs of mean monthly maximum and
covering the surface of the soil with light-colored sand, mean monthly minimum temperatures with an indi-
which reduces reradiation and increases conduction cation of extreme daily values gives a much better
(Haines, 1922). This technique is sometimes used in basis for comparing the temperature regimes of two
forest nurseries to prevent frost problems. environments.
 General Ecological Concepts Concerning Temperature 209

8.3 General Ecological Concepts In defining maximum and minimum temperatures,

Concerning ‘Temperature the length of exposure, time of year, and stage in the
organism’s life history must be defined. The eggs and
There is a great temptation to describe climates as larvae of a species of fruit fly are killed by 7 weeks of
severe, extreme, favorable, or unfavorable. These adjec- exposure to 7°C, but death occurs after 3 weeks at
tives reflect human response to climatic conditions 1°C. Flower farmers sometimes immerse their bulbs
and are not necessarily accurate as descriptors of the briefly in nearly boiling water to kill surface fungi and
response of other, locally adapted organisms. That we pests without harming the bulbs; prolonged immer-
find -30°C very cold does not mean that this is an ex- sion at this temperature would “cook” them. Maxi-
treme or unfavorable temperature for a tree in mid- mum and minimum temperature information is
winter in northern Canada or Russia. The response of obviously of limited value unless accompanied by in-
an organism to a particular temperature is conditioned formation on exposure (Figure 8-4).
by the adaptation of the organism to that temperature. A further complication arises because different
A species that has survived in a particular environment parts of the same organism may have different cardinal
is adapted to the conditions of that environment, and temperatures. Roots have lower minimum tempera-
those genotypes that cannot tolerate the conditions tures than shoots, so the relative temperatures of soil
are eliminated. The unfavorableness of a temperature and air may have a pronounced effect on the welfare of
therefore depends on whether it is “unusual” in terms some plants. The extremities of polar animals have
of the organism’s recent evolutionary history. Very low temperatures in winter that would be lethal to the
or very high temperatures that have been a regular main body of the animal. Cardinal temperatures vary
feature of the species’ experience over many genera- at different times of year and at different stages of an
tions are not necessarily unfavorable. Similarly, if tem- organism’s life history.
perature extremes that have a short-term negative To clarify the situation and render comparisons
affect on some physiological or behavioral attributes meaningful, the following terms should be used (Table
of a particular species are even more deleterious for 8-3). The effective temperature range is that range of
competing organisms, pathogens, parasites, and temperatures over which the existing organisms can
predators, then the overall long-term effects on that conduct all their normal life functions and persist in-
species may be beneficial. Temperatures that are mod- definitely. It is limited by the maximum and minimum
erate but that are outside the species experience may effective temperatures. The survival temperature range
be unfavorable and can accurately be described as “se- is the maximum temperature range over which the
vere.” Fish from arctic seas will die if transferred to population can survive. It is limited by the maximum
tropical waters, and the same fate would befall tropical and minimum survival temperatures, which will vary
fish put into the sea off the north coast of Greenland.
Mild frost at the time of flowering may be disastrous
30 wa Northern catalpa
for a plant population that can tolerate extremely low \
winter temperatures.
Other adjectives that are commonly used to de-
scribe temperature are optimum, maximum, and
minimum (referred to as cardinal temperatures’).
These terms can be used correctly only when qualified Eastern white
American elm
time,
Exposure
min
adequately. Optimum for what? A temperature that is pines

optimum for survival may not be optimum for repro- ae

Se

duction. Growth is not causally determined by tem- 57 60 65 6!

perature, and the optimum temperature for growth Maximum survival temperature, °C
will probably depend on conditions of moisture, nutri-
tion, light, competition, and so on. Figure 8-4 Variation in the maximum survival temperature
of three tree species with different lengths of exposure. The
‘Cardinal temperatures are the set oftemperatures that define the tempera- graph shows the exposure time at different temperatures that is
ture adaptations ofan organism, including the temperature requirements and required to kill cortical parenchyma cells. (After Lorenz, 1939.
tolerances for different life stages, life processes, and parts of the organism. Reproduced with permission of the University of Minnesota.)
210 CHAPTER 8 Temperature as an Ecological Factor

Table 8-3 Temperature Relationships of Living rate at which it is replaced by photosynthesis or inges-
Organisms tion/assimilation (leading to starvation) or by mobiliza-
tion from reserves (leading to heat exhaustion). In plants
High
==} ator survival temperature -———-——— 46.5
“C
Death after a few minutes
and cold-blooded animals, metabolic rates commonly
——— fat various exposure times 44.6 Heat coma double or triple for every 10°C increase in body temper-
ature. Such organisms are said to have a Q10 value of 2
40.1 Excessive activity

or 3, respectively. Different metabolic functions have

——— Maximum effective temperature .
different Q10 values, and the effective temperature
range may be that range over which all functions are
Optimum temperature range a tee a .) synchronized. As this range is exceeded, the different
for a particular biological 23.0
function §Normal activity Q10 values put different aspects of the metabolism out —
~ of synchrony and the organism eventually dies. This
_-7~ 10.8 Feeble movement
Effective
temperature
range
——- Minimum effective
~
temperature “i
-
6.7 Movement begins
:
condition occurs in homeotherms with a much smaller
change of internal temperatures than in poikilotherms.
6,0 Clold coma
Temperatures above 50°C induce changes in the
---) Minimum survival temperature ——————
~— fat various exposure times
~ 5.0
- 8.0
40min
20min fDeath
molecular structure of most proteins. Boiling an egg at
Low
-12.0 Smin 100°C causes the proteins to coagulate, but sustained
temperatures of only 30 to 50°C will cause sufficient
alterations in most proteins to interfere with their
After Clarke, 1967. normal function (Daubenmire, 1974; Spurr and
Barnes, 1973). A few organisms can survive at temper-
with time of year, stage of life cycle, duration of expo- atures of approximately 90°C, such as algae in hot
sure, and rate of temperature change at the extreme springs and certain desert animals, but there are few
temperatures. The rate of temperature change is records of poikilothermous animals living permanent-
sometimes more important than the actual tempera- ly at temperatures above 45°C (Clarke, 1967). Poikilo-
ture. A plant that is gradually cooled to below freezing therms can tolerate much greater increases in body
and then gradually thawed may suffer no harm, temperatures than homeotherms, for which the mar-
whereas it would be killed if frozen and thawed rapid- gin of safety between normal and lethal body temper-
ly. Acclimatization enables organisms to change their ature may be only a few degrees (2°C for some birds; 3
cardinal temperatures, but it will occur only when the to 5°C for human beings). However, for most poikilo-
rate of temperature change is not too great. therms, optimum temperatures for many functions are
Organisms can be classified into two major groups: only just below maximum. survival temperatures
those that regulate their internal temperature within (Figure 8-5), so temperature regulation at the opti-
narrow limits so that for much of the time it is differ-
ent from that of the environment (homeotherms, en-
dotherms, or warm-blooded animals) and those that
Fast
allow their internal temperature to equilibrate more
or less with that of their environment (poikilotherms or
ecotherms—plants and cold-blooded animals). Some bs}
-

animals (notably hibernating birds and mammals) are rs)

homeotherms most of the time but sometimes allow uo)
vo
vo

substantial changes in their body temperatures. Such N

a.

animals have been referred to as heterotherms.

Slow
Cold coma Heat coma
Increasing temperature»
Excessively High Temperatures
Figure 8) Generalized relationship between animal ac-
‘Temperatures may be too high for a variety of reasons. tivity and temperature. The range in body temperature be-
Metabolic and respiratory rates increase with tempera- tween cold coma and heat coma is much greater for poikilo-
ture, and for every organism there is a temperature at therms than for homeotherms. (After Allee et al., 1967.
which the loss of energy in respiration is greater than the Copyright W.B. Saunders Company. Reproduced by permission.)
 Effects of Temperature on Plants 211

mum range is just as important as for homeotherms. permanent water body before the stream dries up and
Maximum survival temperature (with long exposure) they are killed. The rate of movement of poikilo-
for plants is commonly approximately 55°C, but respi- therms is greatly reduced at low temperatures, render-
ration generally exceeds photosynthesis at approxi- ing them more susceptible to predation. Low
mately 50°C (Spurr and Barnes, 1973), although some temperatures can cause the precipitation of proteins or
desert plants have been shown to grow when their tis- lead to the formation of intracellular ice crystals.
sues are at 56.5°C with an air temperature of 58°C These may rupture the cells, destroy their internal or-
(MacDougal and Working, 1921). ganization, or dehydrate them by withdrawing liquid
Perhaps one of the most important effects of high water; frozen tissues often exhibit symptoms of dehy-
temperature is the associated loss of moisture. Water dration (Weiser, 1970).
is lost either as a cooling adaptation or simply because Although freezing damage occurs in many species,
of increased evaporation at higher temperatures, and a freezing is not always harmful. Ice crystals are known
significant proportion of deaths of nonaquatic organ- to form in the tissues of Norway spruce without dam-
isms from high temperatures are probably cases of aging them, and a species of algae has been unharmed
dehydration. Changes in behavior and metabolic de- after being frozen at -182°C for 1 hour (Clarke,
rangement as a result of loss of water and attempts to 1967). Alternate ice crystal formation and melting can
locate water and cooler environments all can have im- occur many times in a single day in pine needles with-
portant effects on the survival of animals. out apparent harm (Christersson and Sandstedt,
High temperatures may increase the susceptibility 1978). Certain species of algae that inhabit snow or
of aquatic organisms to diseases and reduce the con- marine environments may complete their life cycle
centration of dissolved oxygen to critical levels. Exces- entirely at temperatures below 0°C. Freezing of ani-
sive fluctuation of water temperature can affect the mal tissues frequently ruptures blood vessels and leads
solubility of nitrogen in the blood of aquatic animals, to the degeneration of cells, but the Alaska black fish
leading to death from the “bends”: bubbles of nitro- can recover normal activity after 40 minutes at -40°C.
gen forming in and blocking the bloodstream. Large Obviously, generalities about the biological conse-
increases in stream temperature as a result of clearcut- quences of temperature are meaningless unless ade-
ting can sometimes cause these problems, but more quately qualified.
subtle effects can also occur. Clearcutting-induced
warming of Carnation Creek on the west coast of
Canada increased the numbers and rate of develop-
ment of young chum salmon, but it also resulted in an
8.4 Effects of Temperature on Plants
earlier migration of the young fish to the ocean, where
water temperature and food availability were not yet
Growth Response
optimum. A reduction in numbers of mature salmon Both temperature and physiological response to it
returning to spawn several years later was attributed in vary considerably in different parts of a plant. Roots
part to the effect of stream warming on the timing of normally assume the temperature of the soil around
this migration (Hartman and Scrivener, 1990). them, which in areas away from the equator is lower
than shoot temperature in the summer and may be
higher in the winter. Shaded stems approximate the air
Excessively Low ‘Iemperatures temperature below the crowns, although the insulat-
There is also a variety of reasons that temperatures ing properties of bark cause the temperature of the
can be too low. Variations in Q10 values can be re- stem to lag behind that of the air. A stem illuminated
sponsible for metabolic derangement at low tempera- by full sunlight may experience appreciable increases
tures, or metabolism may simply be slowed to the in temperature, particularly if it has a dark color, al-
point at which other factors become lethal. If unusual- though movement of cool water up the stem from the
ly low winter temperatures delay the development of roots may keep stems cooler than the air and prevent
salmon eggs and alevins (young fish) in the gravel beds excessive temperatures in exposed stems. This can be
of a temporary stream (one that dries up in the sum- particularly important at the root collar of a seedling,
mer), then the young fish may not be sufficiently well which may find itselfincontact with a midsummer soil
developed to emerge from the gravel and migrate to a surface temperature of more than 60°C. Leaves may
212 CHAPTER 8 Temperature as an Ecological Factor

have temperatures higher, lower, or the same as the air. nal temperatures commensurate with their environ-
Thin leaves with high rates of transpiration may be as ment. High-altitude, high-latitude plants have lower
much as 15°C cooler than the air in summer, even in full values than low-altitude, low-latitude plants (Figure
sunlight, because of evaporative cooling. Thick leaves 8-6). The variation in cardinal temperatures as a plant
may be as much as 30°C warmer than the surrounding ages has already been mentioned.
air in winter. In some cases, the respiratory activity of Plant metabolic activity and growth depend on the
the plant may increase tissue temperatures slightly, but availability of water in liquid form, and they cannot
this is normally unimportant (Daubenmire, 1974). continue when the tissues are frozen. Plants can con-
The cardinal temperatures for different plant tinue to photosynthesize at leaf temperatures a few de-
functions vary enormously. Gross photosynthesis grees below freezing because cells do not freeze at
reaches a peak at a much lower temperature than res- these temperatures, and photosynthesis has been mea-
piration, which increases almost up to the maximum sured in some conifers at even lower air temperatures
survival temperature (see Figure 7-5). Net photosyn- (Parker, 1953). However, the needle temperature in
thesis reaches a maximum at intermediate tempera- these studies was probably substantially higher than
tures. This may help to explain why plants that are the air temperature. Leaves can act as miniature green-
transferred to cooler climates than those to which they houses, maintaining temperatures many degrees above
are accustomed (higher latitude or altitude) frequently that of the surrounding air (see Figure 9-5). Low-
grow faster, and plants that are moved to warmer cli- temperature photosynthesis permits alpine and arctic
mates (lower latitude or altitude) are frequently less plants to start growth in the spring beneath the snow
productive than in their native environment. The and winter photosynthesis by evergreen plants in
midday decline in net photosynthesis and the rapid sunny weather. The extent of this winter activity is lim-
carbohydrate accumulation in some crop plants as ited, however, because of the reduced ability of plants
temperatures decline at the end of the growing season to absorb nutrients and moisture and to translocate
may also reflect differences in respiratory and photo- them from the roots to the shoots at low temperatures.
synthetic cardinal temperatures. All plants experience variations in temperature as-
Cardinal temperatures vary between different sociated with diurnal variations in the net radiation
plant organs. As noted above, roots have lower mini- budget. Plants that live away from the equator also ex-
mum effective temperatures than shoots and can con- perience seasonal temperature variations. Plants are
tinue to grow in the fall when the shoot has become generally sensitive to these variations and will grow
dormant. Spring root growth can begin when the soil normally only when exposed to the particular diurnal
is still cold but increasing air temperatures have stim- and seasonal temperature changes to which they are
ulated shoot growth. Predictably, plants exhibit cardi- adapted, a phenomenon called thermoperiodism.

e Citrus limonum
Quercus pubescens
¢ Laurocerasus sp.
i)oO he
Betula
© Taxus baccata pendula
Oleae =e Fagus silvatica
europea e Larix decidua
e
Larix decidua
_ nN > Quercus ¢ == Picea excelsa e

ilex Abies Stee, DELULA

alba é —— pendula
pinus cembra ¢
°C
CO,
photosynthetic
uptake,
optimum
Mean
for
temperature Picea excelsa
10} >
0 500 1000 1500 2000 750 1000 1250 1500
Elevation of seed source, m Elevation of seed source, m

Figure 8-6 Variation in the temperature optimum for photosynthesis in trees. (A) An eleva-
tional sequence of tree species from central Europe. (Data from Pisek et al., 1969. Reproduced by per-
mission of VEB Gustav Fischer Verlag.) (B) An elevational sequence of balsam fir ecotypes from New
England. (After Freyer et al., 1972. Reproduced by permission ofthe Canadian Journal ofBotany.)
 Effects of Temperature on Plants = 213

Diurnal thermoperiodic responses generally in- warm nights (23°C/23°C) in Washington and Col-
volve a difference in temperature between day and orado ecotypes, and in both warm day/cool night
night. Low night and moderate day temperatures are (23°C/14°C) and cool day/warm night (17°C/22°C)
important to flowering and fruiting and affect the combinations in a Californian ecotype (Callaham,
quality of crop plants and fruit (Treshow, 1970). Maxi- 1962). The height-growth response of red maple
mum height growth in Loblolly pine and Douglas-fir seedlings in the United States to day and night tem-
seedlings occurred when day temperatures were 12 peratures also varied in different parts of its range, op-
and 10°C warmer, respectively, than night tempera- timum growth occurring at temperature combinations
tures (Hellmers and Sundahl, 1959; Kramer, 1957). that resembled those normally prevailing in the area
For some species, such as Engelmann spruce and pon- from which seed was collected (Perry, 1962).
derosa pine, the night temperatures are more critical Seasonal thermoperiodic responses play an impor-
to seedling growth than day temperatures. Conversely, tant complementary role to a plant’s photoperiodic re-
variation in day temperatures produced a greater vari- sponses. Whereas photoperiod plays the major role in
ation in redwood seedling height growth than did the initiation of winter dormancy, temperature condi-
variation in night temperatures (Figure 8-7). Maxi- tions predominate in the termination of dormancy in
mum height growth of red fir seedlings was achieved the spring and in control of the timing of flowering.
at both 17 and 23°C daytime temperatures but only Release from dormancy in many plants requires chill-
when the former was accompanied by a 4°C and the ing at temperatures of 0 to 10°C or below for periods
latter by a 10°C night temperature (Hellmers, 1966a). of 260 to 1000 hours, depending on the species. Peach
Cool days required cold nights, and warm days re- trees require 400 hours, blackberry bushes need 800
quired cool nights. The diurnal thermoperiodic re- hours, and apple trees require even longer. In some
sponses of ponderosa pine have been found to vary climates, this chilling requirement may have been sat-
somewhat over its range, with maximum height isfied by January or February, but new growth does
growth occurring with cool days and warm nights not normally commence until temperatures become
(17°C/22°C) in an Arizona ecotype, warm days and favorable (Wareing, 1969). However, if plants from

é 90,
s Ee
3oO ¥
nN

2 S) 2
S 1
“e
5
=
sn
~ 60 2s a
2S oa aS)
fo)
att | av)
N L @

aS
ee
ep
ay x
2
Euc0)
a
am[o)
v s °C
Night
temperature °C
Night
temperature
s 3
i) 2
= S
eh
0) @ 0
ap Pie} 19 15 iil an a) 27 23 LOLS seedlings,
14-week-old
of
Growth
Height
cm.
Night temperature °C Day temperature °C Day temperature °C

(a) (b) (c)

Figure 8-7 Height growth in California redwood (A), Colorado Engelmann spruce (B), and
Arizona ponderosa pine (C) seedlings exposed to various combinations of day and night tem-
peratures (after Callaham, 1962 Hellmers, 1966; and Hellmers et al., 1970). The difference be-
tween day and night temperatures clearly has an important influence on height growth of these
species. (Reproduced with permission.)
214 CHAPTERS Temperature as an Ecological Factor

lower latitudes are moved poleward, then their modest high latitudes and flowering may be initiated when as
chilling requirements will be met before the risk of few as eight leaves have been produced. Vernalization
late spring frosts is over, and they may suffer frost in- is more complex than merely a temperature treatment,
jury. Conversely, high-latitude plants that are moved and the desired results cannot be obtained without the
toward the equator may not get sufficient winter chill- correct oxygen and light conditions (Hillman, 1969).
ing, and bud break and leaf and flower development all The response of a plant to temperature is not
may be inhibited. Plants from winter-cold environ- fixed. It depends on the temperature experiences in
ments may exhibit a similar response if kept in a warm the plant’s recent past because of the ability of plants
greenhouse over winter. Not all species require chill- to acclimatize. If a plant is exposed to slowly diminish-
ing. Growth of birch and European beech seedlings ing temperatures, then it will eventually be able to sur-
can be stimulated simply by transferring from short to vive at very much lower temperatures than if it is
long photoperiods. suddenly exposed to a low temperature. This ability of
One of the possible effects of global warming metabolic processes to acclimatize is exemplified by
caused by the release of greenhouse gases is a reduc- the cardinal temperatures of photosynthesis, which
tion of winter chilling. This could pose problems for change in response to changes in environmental tem-
many northern tree species near the southern edge of peratures. For example, a Californian shrub was found
their range (e.g., Leverenz and Lev, 1987). Changes in to attain its peak photosynthetic rate at 12 to 15°C
winter temperature regimes could cause increased after being exposed for 12 days to a 15°C day/2°C
frost damage if dormancy is broken before the risk of night temperature regime (Figure 8-8). After being
major frosts is past (Cannell, 1989; Cannell et al., switched to a 30°C environment for 23 hours with
1986, 1989) or may interfere with cone and flower de- continuous light, the photosynthetic rate increased
velopment, fertilization, seed development, and ger- and the optimum temperature range occurred at ap-
mination (Zasada et al., 1991). van der Kamp and proximately 21 to 28°C. Photosynthetic temperature
Worrall (1990) described an example of frost damage acclimatization in a Californian perennial herb
that can occur when winter temperature regimes do showed the same pattern, although the shift in opti-
not match the species thermoperiodic adaptations. mum temperature range was much smaller for an
The impacts of climate change on woody plant growth alpine ecotype than for a coastal ecotype. The phe-
were reviewed in Lavender (1987). nomenon of acclimatization poses yet another prob-
Seeds of many species require chilling before they lem for the definition of cardinal temperatures, which
will germinate. Temperatures of 0 to 5°C are most ef- are valid only for the temperature conditions under
fective, and adequate aeration and moisture are re- which they are measured. The degree to which plants
quired. Germination of some hardwood tree seeds is can alter their cardinal temperatures in response to
greatly enhanced by stratifying them in damp sand at slow changes in environmental temperatures will vary
cool temperatures for a period before they are sown. greatly between species.
The chilling requirements of seed of some species can Because the rate of metabolic processes increases
be eliminated by removing the seed coat, but in many with temperature, there is a reasonably good correla-
species the embryo itself has a chilling requirement. tion between the growth of experimental plants and
Initiation of flower buds and flowering in many the heat sum expressed as daily degree hours [the daily
species requires an exposure to low temperatures, a sum of: temperature (in degrees Celsius) * number of
fact that limits the spread of certain species to warmer daylight hours at that temperature). The heat sum is a
climates. The stimulating effect of low temperature on measure of the amount of heat to which the plants are
flowering is used to advantage by farmers in the exposed, and different species exhibit optimum
process of vernalization: the induction of flowering by growth at different daily heat sums (Figure 8-9).
cold treatment. The short growing season of northern Rate of growth is strongly influenced by tempera-
areas poses problems for the production of grain crops ture, and because plants must complete a certain min-
such as rye, which is day-neutral but will not flower imum amount of growth before they can flower, the
until it has produced 22 leaves. This requires most of poleward and upper altitudinal distribution of plants
the growing season, but if the seed is moistened and corresponds to some extent to the geographical distri-
held at 1°C for several weeks before planting, then the bution of annual heat sums. Many attempts have been
plant becomes sensitive to the long days that occur at made to correlate these two parameters, with variable
 Effects of Temperature on Plants — 215

PERENNIAL HERB
SHRUB Polygonum bistortoides
Encelia californica
Valley ecotype Coastal ecotype Subalpine ecotype

PN
a My
\,

rateCO,/gm
d.w./hr.
leaf
mg
Photosynthetic

Response of cold-
acclimated plants
—-—: Response after heat
treatment

of
Percentage
maximum
photosynthesis
net

Temperature °C
(a) (b) (c)

Figure 8-8 Effect of varying temperature on the photosynthetic rate of two California plant
species. Encelia (Encelia californica) and snake weed (Polygonum bistortoides), conditioned to low and
high temperatures. (A) The photosynthetic response to various temperatures was tested on the shrub
species after exposure to 12 days of low temperatures (15°C daytime, 2°C nighttime). The plants
were then exposed to 23 hours at 30°C, and the photosynthetic-temperature response was redeter-
mined. The experiment was repeated for coastal (B) and subalpine (C) ecotypes of the perennial
herb, with 10 days of cold treatment and 25 hours at 30°C. Differences in photosynthetic response
are apparent between the two species and between the two ecotypes of the herb. (After Mooney and
Shropshire, 1967. Reproduced from Oecologia Plantarum with permission.)

success. Heat sums have been calculated in many ways lated earlier in a hot summer than in a cold summer,
(Wang, 1960), but the commonly used measure of an- contributing to variation in date of flowering.
nual heat sums is degree days: the annual or growing The use of heat sums to predict plant distribution
season sum of the daily difference between mean daily has been criticized because it ignores the complications
air temperature and some threshold temperature. For of thermoperiodism, the variation in cardinal tempera-
example, the heat sum required for flowering of grain tures as the plant matures, differences between species
crops in Ohio is 600 to 1050 degree days, using 6°C as and between ecotypes, the variation in the day length
a threshold (Clarke, 1967). This heat sum is accumu- during which high temperatures occur, and because
216 CHAPTER 8 Temperature as an Ecological Factor

logical symptoms in plants, but low growing-season

and high winter temperatures are generally much
more damaging, particularly when accompanied by
rapid temperature changes. If dormant woody plants
are cooled slowly, then they can survive temperatures
far below those attained even in polar regions (Weiser,
1970), reflecting the ability of plants to change their
cardinal temperatures to some extent by acclimatiza-
tion. In addition to the general effects of temperature
extremes discussed in the preceding section, plants ex-
hibit the following responses.
seedlings
hemlock
of
growth
Height
pine
after
of
weight
Total
6seedlings
g
mo.,
200 400 600 800 Low-Temperature Injury. Frost damage to plant
Heat sum, daily deg-hr tissues as the result of seasonally low temperatures is
Figure 8-9 Seedling growth of three tree species at vari-
uncommon. The delicate leaves, flowers, and stems of
ous different daily heat sums. (After Hellmers, 1962. Copy- deciduous and annual plants are generally shed before
right Ronald Press. Redrawn with permission of John Wiley & low temperatures occur, and the leaves of dormant
Sons, Inc., and H. Hellmers.) evergreens are very resistant to low-temperature dam-
age. In contrast, unseasonably low temperatures can
be very damaging, even when they are not actually
growth and distribution are also influenced by other very low. Unseasonal frost can kill flowers, which are
environmental factors. Generally, the heat relation- the most frost-sensitive parts of a plant. Freezing of
ships of plants are too complex to permit simple corre- leaves during the growing season can lead to disorga-
lations with geographical distribution, but despite nization, wilting, water loss, and death, and freezing of
these criticisms, northern tree lines do show a good stems can result in the death of patches of the camb-
correlation with temperature. North of the line, the ium, which develop into lesions and permit the entry
warmest summer month generally has a mean temper- of pathogens. Efficient emission of radiation and low
ature of less than 10°C, whereas south of the tree line, conductivity lead to rapid surface cooling of woody
the value is higher (Mikola, 1962). Vegetative growth stems on clear nights with low air temperatures. The
usually requires lower temperatures than reproductive outer layers of the stem contract more rapidly than the
growth, and the tree line may reflect a failure to repro- inner layers, which creates tensions that can cause the
duce rather than a failure to grow. Rehfeldt et al. stem to crack. These frost cracks are particularly com-
(1999), working with lodgepole pine (Pinus contorta) in mon in regions subject to sudden drops in air temper-
western North America, showed genetic variation in ature, such as the west coast of North America. The
growth response to temperature and cold hardiness in normally mild winter air masses on the coast are peri-
genotypes taken from different parts of the geographi- odically replaced by polar air masses spilling over the
cal range of this widely distributed species. Coast Mountains, producing dramatic drops in tem-
Heat sums are important in determining spring perature (Figure 8-10). As the temperature in the
bud burst in many plants. Once their winter chilling stem again becomes uniform, the cracks close and heal
requirement has been set, the timing of bud break is over, but they remain permanent points of weakness.
largely under the control of the accumulated heat sum The differential stem shrinkage that causes frost
(e.g., Worrall, 1983). cracks can also cause bark to separate from the stem.
An indirect form of low-temperature damage that
is of both ecological and practical interest results from
‘Temperature-Related Injuries rapid alternate freezing and thawing of the soil. Rapid
Two aspects of temperature are responsible for most radiation cooling results in the freezing of soils from
of the temperature-related injuries to plants: rapid the surface downward. Water is drawn up to the
change in temperature and the occurrence of unsea- frozen layer, where it freezes and forms a gradually
sonal temperatures. Unusually hot summers and thickening layer of vertically oriented ice crystals
unusually cold winters can certainly produce patho- (needle ice). The frozen surface soil together with small
 Effects of Temperature on Plants = 217

10 viscosity of water between 25 and 0°C, which makes

v 5 Normal water uptake more difficult at temperatures approach-
s temperature
5 0) range ing freezing. Plants that grow on soils that are cold or
ss frozen in winter often exhibit the same morphological
aes}
adaptations as plants that grow on summer-dry sites.
5
eae LO The water imbalance caused by high air temperatures
15 and low soil temperatures is referred to as physiological
drought. When severe, it can cause browning of the fo-
November 1955 liage (e.g., the red belt damage in the mountains of
western North America) and even the death of the en-
Figure 8-10 Effect of the invasion of a polar air mass into tire plant.
a coastal region on the diurnal temperature range. Data are By killing buds or new terminal twigs, early frost in
for Centralia, Washington. Plain bar indicates above-freezing the fall or spring can alter the shape of a plant. Stems
temperatures; striped bar indicates below-freezing, tempera- become crooked or have multiple leaders and may be-
tures. (After Daubenmire, 1974. Redrawn with permission offohn
come bushy. In extreme cases, the plant may be killed
Wiley & Sons, Inc., and R.E. Daubenmire.)
back to ground level each year and never gain the size
and shape that it would have in a less frosty environ-
plants can be lifted as much as a decimeter by this nee- ment. Plants that are damaged by frost or subjected to
dle ice and then lowered again as the ice melts. Roots a prolonged period below their optimum temperature
that are pulled up from lower unfrozen soil layers can- range are frequently more susceptible to damage by
not return to their original position, and over several insects and diseases.
freeze-thaw cycles, small plants such as tree seedlings
may be lifted (frost-heaved) right out of the soil. Par- High-Temperature Injury. The most common
tially frost-heaved plants are rendered more suscepti- deleterious effects of high temperatures are the exces-
ble to drought and wind damage, and damage to roots sive loss of moisture and the stimulation of excessive
can permit the entry of root pathogens. Frost heaving respiration rates. Sun scald, mentioned in Chapter 7,
occurs only on soils that have a high moisture content results from the rapid heating and cooling of stems
and that cool rapidly: it is generally restricted to ex- during the winter dormant period, but this is more a
posed mineral soils of medium to fine texture. function of rate of temperature change than of high
Alternate freezing and thawing of soil results in so7/ temperatures per se. High temperatures are responsi-
churning. This can have beneficial effects on the phys- ble for stem girdle in seedlings. Because of the low
ical properties of soil but may render fine-textured albedo and low conductivity of many soils, surface
soils prone to erosion by loosening the surface; in arc- temperatures frequently become very high, and young
tic and alpine environments, it may inhibit growth of plant stems that are not yet protected by thick layers
most or even all of the vegetation. In severe cases, of bark may be damaged where they contact the soil
rocks and stones are brought to the surface and surface. A band of cambium a few millimeters wide is
arranged in various patterns (stripes or polygons). On killed around the stem, and this results in the death of
steep slopes, it can cause soil creep (solifluction), in the plant either because of the interruption of internal
which wavelike lines of soil are gradually moved translocation or because of the entry of pathogens.
downslope by the alternate freezing and thawing.
Plants are tipped over, roots are torn from the ground,
and over a geological time period large quantities of Adaptations of Plants to
soil are moved downslope. Unfavorable Temperatures
Another indirect injury associated with low tem- In common with other organisms, plants have adapted
peratures is water stress. Warm air temperatures in to the temperature factor both morphologically and
winter (especially when accompanied by wind) or an physiologically.
early, warm spring in areas where the soil is still frozen
can remove water from plants at a time when it cannot Morphological Adaptations. Morphological adap-
be replaced. Even if the water is not frozen, winter tations to high temperatures include leaf shape, size,
water stress can occur because of the doubling of the and orientation. Long, thin leaves are oriented to
218 CHAPTER 8 Temperature as an Ecological Factor

maximize mutual shading (e.g., pines) or minimize closed by resin and will open only to release the seed
sunlight interception during midday heat (e.g., after exposure to high temperatures. Such tempera-
Eucalyptus). Cuticles are thick, scaly, or hairy to in- tures occur during a forest fire or when the cones are
crease albedo and reduce transmittance, and protec- lying on hot ground in midsummer after removal of
tive pigments further reduce the heat balance of the tree cover by wind, insects, disease, or logging.
foliage. Rapid evaporation of water from leaves cools The occurrence of serotiny seems to be related to the
them, but this is a luxury on which many plants in hot incidence of forest fires (Ledig and Fryer, 1971).
environments cannot rely. In fact, an important adap- Adaptation to low temperature includes the forma-
tation to high-temperature environments is a leaf tion of a morphologically (and physiologically) differ-
morphology that reduces evaporation. Adaptation to ent stage, such as a seed or spore, that either resists
physiological drought associated with low tempera- low temperature or finds its way to a more sheltered
tures also involves leaf morphologies that regulate location (e.g., the soil), or the development of differ-
water loss. ent overwintering forms (/ife forms) of the permanent
Many trees have morphological adaptations to en- plant. Morphological adaptation to low temperature is
able them or their progeny to survive brief exposure to so common that it is the basis for a plant classification
high temperatures in forest fires. The thick bark of scheme developed by the Danish botanist Raunkiaer
redwoods, Douglas-fir, and many species of pine con- (1934). The classification is based on the vertical loca-
fers a high degree of resistance to surface fires. Some tion—and hence the degree of protection—of the dor-
trees have the power to resprout if crowns are burned mant buds during the period of dormancy. Some of
off, such as the epicormic sprouting of redwoods and the major categories of the classification are shown in
the production of new foliage by seedlings of long leaf Figure 8-11.
pine (Pinus palustris) after a fire. Many species of pine The relative abundance of the different life forms
have serotinous cones in which the scales are held (the /ife form spectrum) varies in environments that dif-

(a) (b) (c) (d) (e)

Phanerophytes Chamaephytes Hemicryptophytes Cryptophytes Therophytes
and
Geophytes
Overwintering
buds

IRE
EASY.
a
=—wa— > Gy 0,
'\Y, KA WY
Vly
%

Ys ORY

Increasing degree of morphological adaptation

Figure 8-11 The major life forms of plants according to Raunkiaer (1934). Lighter-colored
parts are ephemeral; darker parts are perennial. (A) Trees and tall shrubs with dormant buds and
shoot tips that project high into the air with little or no protection other than bud scales and sticky
substances. (B) Low shrubs with dormant buds and shoot tips borne above but close to the ground.
(C) Herbaceous plants with dormant buds and shoot tips at the soil surface or beneath an organic
layer. (D) Tuberous and bulbous herbs with dormant buds buried at various depths in the soil. (E)
Annuals with the embryonic bud protected by a seed coat; located on or in the soil surface during
the dormant period.
 Effects of Temperature on Plants 219

Table 8-4 Variation in the Frequency of Five Life Forms in the Plant Communities of Various Environments
Approximate Percentage Occurrence of Each Life Form?

Environment Phanerophyte Chamaephyte Hemicryptophyte Cryptophyte Therophyte

Wet tropical 92 7 { 0
Subtropical 62 13 3 12
Temperate 17 & 46 28 9
Hot desert 19 14 19 6 42
Arctic 0.5 22 60 ie 2

After Dansereau, 1957; Richards, 1952.

*Mean of two or three locations.

fer in their temperature characteristics (Table 8-4). Physiological preparations for dormancy are pri-
Phanerophytes dominate tropical environments, where marily under the control of shortening photoperiods,
there is little or no need to protect the dormant bud and city plants that are maintained artificially on a
from extreme temperatures. Temperate, moist cli- long photoperiod because of illumination by street
mates are dominated by hemicryptophytes, whereas tun- lights may have a long delay in the onset of dormancy.
dra environments are dominated by chamaephytes and However, photoperiod does not have exclusive control
hemicryptophytes. Cryptophytes are most common in of dormancy. Low temperatures, nutrient supply, soil
temperate climates, and therophytes dominate the floras moisture, and other factors can modify or even substi-
of desert regions, reflecting the severe temperature tute for the photoperiod stimulus in some plants.
stress of this type of environment (Raunkiaer, 1934). Dormancy can occur in several degrees of intensi-
The survival of chamaephytes in very cold environ- ty. Before the onset of true dormancy, there is a
ments may reflect the presence of insulating layers of predormancy, which is marked by a cessation in growth
snow and the increase in air temperature close to the and an increased resistance to low temperature (first
ground. Although these life forms are adaptations to stage of acclimation; Figure 8-12). This condition can
temperature, they are almost certainly adaptations to be broken and growth resumed if there is a return to
other factors as well, such as wind and water stress. warmer temperatures and longer photoperiods before
the development of full dormancy (second stage of ac-
Physiological Adaptations. Physiological adapta- climation), which occurs as photoperiods continue to
tions are largely related to the water aspect of the tem- shorten and temperatures drop. Full dormancy, which
perature factor. Winter freezing damage is avoided in may be triggered by exposure to the first frost, is
seeds and spores by the elimination of virtually all free marked by deciduous leaf fall and a dramatic increase
water, and dry seeds have been successfully germinat- in cold resistance. A period of chilling is required be-
ed after 3 weeks of exposure to a temperature of fore full dormancy can be broken and growth re-
~190°C (Clarke, 1967). Alternatively, where water is sumed. A temperature of 5°C is optimum, and freezing
retained in a plant, freezing can be avoided at moder- is not required. In fact, temperatures below 0°C and
ately low temperatures by increasing the concentra- above 10°C will not satisfy the requirement. The chill-
tion of dissolved and colloidal substances. This ing must continue for a specific period, the length of
enables metabolic processes to continue at reduced which depends on the species, but for most northern
temperatures, but at very low temperatures, freezing hemisphere plants, the requirement is satisfied by the
becomes inevitable and plants become dormant (a end of December. Trees that are transferred to the
physiological condition in which plants become very greenhouse at this time will break dormancy and re-
resistant to damage by low temperatures). The dam- sume growth, although expansion of new leaves and
aging effect of both high and low temperatures rarely active growth normally requires a period of exposure
occurs in the absence of physiological activity, sug- to some critical higher temperature. Elm trees near
gesting that such damage is associated with a derange- Chicago require 310 hours at approximately 25°C be-
ment of active metabolic processes. fore leaf break; temperatures above 30°C and below
220 CHAPTER 8 Temperature as an Ecological Factor

Start of first stage

First
of acclimation
autumn frost

____ Injury occurred

in the field
= #5)
vil
Start of second
Northern ecotoype,
stage of acclimation
North Dakota
0)

1p)
___ Coastal ecotype,
Washington
°C
survival
Lowest
temperature
—100 °C
survival
Lowest
temperature Midwestern ecotype,
Minnesota

Sept. Oct. Nov. Sept. Oct. Nov.

(a) (b)

Figure 8-12 The development of cold resistance in the living bark of a hardy shrub, red-
osier dogwood (Cornus stolonifera), in Minnesota. (A) Typical pattern. (B) Variation in the pat-
tern among ecotypes from three different climates when grown together in Minnesota. Only the
coastal ecotype suffered frost damage because the other two ecotypes had already passed through
their first stage of acclimation by mid-October. (After Weiser, 1970. Copyright American Association
for the Advancement ofScience, 1970. Redrawn with permission of AAAS and C.F. Weiser.)

10°C are ineffective. There are many minor deviations tures increase, activity rates increase up to a maxi-
from this general pattern, but they are beyond the mum, above which activity declines and the body en-
scope of this book (see Perry, 1971; Weiser, 1970). ters a heat coma. The temperature of optimum
For an earlier review of cold resistance in plants, see activity and efficiency is closer to the maximum than
Parker (1963). to the minimum survival temperature (Figure 8-7). As
The date of onset and termination of dormancy with plants, the cardinal temperatures of animals vary
and the degree of acclimatization varies from species between species, with age and physiological condition,
to species and between different ecotypes of a single with the duration of exposure, and according to the
species, as was the case for photoperiodicity. recent temperature experiences of the animal. Most
The seeds and resting stages of some plants will poikilotherms are active between 6 and 35°C, but
emerge from dormancy and resume activity only after there are examples of activity well outside this range.
exposure to a high temperature. This may be an adap- The relationship between animal morphology and
tation to keep them dormant in a deeply shaded loca- environmental temperatures has attracted attention
tion in which they would die for lack of light. When for a long time, and various general relationships, or
light conditions become favorable for survival, the ac- “rules,” have been proposed. The Allen rule states
companying high temperatures stimulate growth. It that, in general, the extremities of homeotherms are
may also be an adaptation to fire (see Chapter 12). smaller in cold than in hot climates, such as the
marked reduction in size of the ears of mammals as
one moves from hot to cold environments. Another
rule states that the bodies of homeotherms are larger,
8.5 Animal Adaptations to Temperature are stockier, and have a lower surface area/weight ratio
than species of the same genus at lower latitudes
General Relationships (Bergman’s rule). Yet another rule says that races of
At temperatures below their effective range, animals birds and mammals become increasingly darker in
enter a cold coma and if not aroused within a certain skin, fur, and pelage as one goes from high to low lati-
period of time will eventually die. As body tempera- tudes (Golger’s rule). These rules are based on observed
 Animal Adaptations to Temperature 274

patterns of variation, but they have been criticized be- when the air temperature in their burrow, shelter, or
cause of the existence of exceptions and because of the nest, is favorable. Arctic homeotherms such as seals
implication that temperature acts as a causal determi- and whales provide an exception to this rule because
nant in producing these variations. It has been argued their young are born well equipped to regulate their
that the variation described in Bergman’s rule could be internal temperatures. The degree of homeothermy
as much an adaptation to predator-prey relationships may vary through the life of an animal; birds are es-
as it is for heat balance, which could be achieved more sentially poikilothermous until they develop feathers,
efficiently by improved insulation. It is probable that when they become homeothermous (Allee et al.,
temperature is only one of several determinants in- 1967). Homeotherms have to pay a high energy bill to
volved in the variation. maintain a steep gradient between their bodies and
their environment, and some animals permit their
Poikilotherms and Homeotherms internal temperatures to drop at certain times to
improve their energy budget. Winter-hibernating
It has already been mentioned that animals exhibit mammals permit their internal temperatures to vary
two major responses to temperature: poikilothermy seasonally as an adaptation to low temperatures and
and homeothermy. The more limited case of het- low availability of food, and hummingbirds permit
erothermy has also been mentioned. their temperature to drop at night as an adaptation to
Poikilotherms in aquatic environments generally conserve energy, a necessity considering their high
show little deviation from the temperature of the metabolic rate. Sometimes the distinction between
water because of the efficiency of conduction in ex- homeotherms and poikilotherms becomes a little
changing heat between the organism and water. Their fuzzy, as in the case of honeybees, which are able to
temperatures are stable because there is little diurnal regulate the temperature of their hive between fairly
temperature variation in most aquatic environments. narrow limits by varying their behavior pattern.
Marine tidal rock pools are an exception, becoming
warm or even hot during daytime low tides and drop-
Adaptations of Animals to
ping as much as 25°C with the first large wave of the
incoming tide. Terrestrial poikilotherms behave dif-
‘Temperature Conditions
ferently; they normally experience a wide diurnal tem- Adaptations by animals to temperature extremes may
perature fluctuation, and their body temperature can involve attempts to regulate their heat balance and,
differ substantially from air temperature. An animal therefore, their temperature by varying conduction,
with a low skin albedo on a clear but cold day may at- convection, radiation, and evaporative cooling. Alter-
tain a body temperature well above that of the air, just natively, the tolerance of the animals may be changed
as a conifer needle may attain temperatures that per- by alteration of their cardinal temperatures.
mit active photosynthesis when the air temperature is Methods of regulating heat balance vary depend-
far below freezing. The inefficiency of thermal energy ing on the relative importance of the different path-
loss by the processes of conduction and convection in ways of heat loss. In insects such as cockroaches, 80%
dry air permits the existence of steep gradients of tem- of the total heat energy loss is by convection and 20%
perature between a terrestrial poikilotherm and its is by radiation. Very little (approximately 4%) is lost
medium. This enables terrestrial poikilotherms to by evaporation of water because, in common with
control their temperature somewhat. By regulating many other organisms, they have evolved efficient
their radiation budget (see the section Behavioral methods of water conservation. The python and the
Adaptations), they can control their heat load and, tortoise, which inhabit hot, dry environments, lose
consequently, their temperature between unexpected- only 0.1 to 0.3% of their body weight per day in evap-
ly narrow limits. orated water. In contrast, woodlice, which live in
Homeotherms are able to maintain remarkably humid microenvironments and will die if left for sev-
constant body temperatures over wide limits of envi- eral days in a dry environment, lose 14%. Salamanders
ronmental temperature. This ability varies with age lose 95% of their body weight of water per day if
and is poorly developed in the very young and very old placed in a dry environment and are completely intol-
of many species. The young of such species must be erant of dry microclimates. Evaporative cooling could
born in sheltered environments and at a time of year potentially be important to woodlice and salamanders,
222 CHAPTER 8 Temperature as an Ecological Factor

but because they confine themselves to humid mi- sometimes be satisfied by midwinter, but activity is not
croenvironments, water loss is not a significant mech- resumed until temperatures increase in the spring. In-
anism of heat loss (Gates, 1962). sects may also have a period of physiological rest in re-
Birds lose most of their excess heat by radiation sponse to stressfully high temperatures. This is known
because their feathers greatly reduce conductive and as summer (estival) diapause and is also initiated by
convective losses. Birds that have access to ample temperature and/or photoperiodic stimuli. Many aesti-
water supplies may use evaporative cooling by panting vating animals do not require any specific stimulus (as
and what is called gudlar flutter (rapid flapping of the they do in winter diapause) for the resumption of nor-
floor of the mouth). This is absent in birds from hot, mal activity, which occurs as soon as the temperature
dry regions. In general, birds tend to conserve water. returns to a favorable level. Dormancy in insects can be
For sheep, 48% of their heat loss is due to radia- found in the egg, larval, pupal, and adult life stages.
tion, 38% is by convection, and 14% is by conduction. While in diapause, insects are highly resistant to
By comparison, 47% of the heat loss from a naked cold, and they can safely overwinter in environments
human is by radiation and 53% is by convection; in a subject to very low temperatures. Many other poikilo-
fully clothed human, the figures are 55 and 45%, re- therms are not resistant to freezing and seek out shel-
spectively (Gates, 1962). tered habitats for their winter dormancy. Many
Adaptation to low temperatures is necessary be- amphibians and aquatic animals bury themselves in
cause most animals are injured or killed if their tissues the mud on a lake bottom or seek the deeper waters
freeze or if their internal temperature drops below their that do not freeze. Snakes retreat into deep rock
lower effective or survival temperature for a critical pe- crevices or caves, frequently with large numbers occu-
riod. Mortality may be a consequence of physiological pying the same shelter. Some poikilotherms bury
or morphological damage or may result from increased themselves deep in the soil. All these environments are
susceptibility to disease or natural enemies. Adaptation characterized by stable temperatures above freezing.
to high temperatures is necessary to avoid death after Many homeotherms that live in environments with
heat coma, derangement of physiological processes, ex- low winter temperatures are able to continue normal
cessive loss of moisture, or physiological exhaustion as a physiological activities by means of behavioral and
result of hyperactivity. Many animal adaptations paral- morphological adaptations. They may confine their
lel those of plants because the biochemical systems of activities to below the snow, as do the lemmings of the
plants and animals have many similarities, reflecting Arctic, or develop better insulation as polar bears or
their common biogeochemical origin. arctic foxes do. However, where these adaptations are
Animal adaptations can be classified as physiologi- inadequate, increased metabolism may be used as a
cal, morphological, or behavioral, although many means of creating metabolic heat to compensate for
adaptations fall into more than one of these categories. increased losses to the environment. Figure 8-13
shows how the metabolic rate increases with decreas-
Physiological Adaptations. The adverse effects of ing temperature in homeotherms from three different
temperature extremes on physiological processes and temperature environments. This physiological adapta-
on the availability of food and water are avoided in tion is most pronounced in the tropical homeotherm
many animals by undergoing a period of reduced phys- that has no morphological adaptation to deal with low
iological activity. This results in an alteration in their temperatures. The Eskimo dog is so well insulated
cardinal temperatures. Many poikilotherms undergo a that it uses this physiological adaptation only at very
period of winter rest that involves a variety of physio- low temperatures.
logical as well as behavioral and morphological There is a high metabolic price to be paid for sus-
changes. Such a rest period in insects is called-a diapause. tained activity during a period when food and water
As with dormancy in plants, diapause is initiated by may be scarce, and some animals that are normally
decreasing photoperiods and temperature, and termi- homeothermic during the active season avoid this
nation of diapause has a chilling requirement. Dia- metabolic cost by permitting their body temperature
pause can be affected by photoperiod, moisture, the to drop during a period of winter dormancy or at
level of nutrition before diapause, and other factors. other times. These animals are the heterotherms, and
The requirements for termination of diapause may their winter dormancy is called hibernation. It is a
 Animal Adaptations to Temperature 223

A
or close to freezing. Hedgehogs can hibernate in air as
7 cold as —5°C, maintaining a body temperature of 2 to
300 Tropical raccoon,“ zi
vf 5°C, but the optimum air temperature for their hiber-
nation is 4°C. In contrast, arctic marmots can hiber-
nate at air temperatures as low as —48°C with
peripheral body temperatures as low as —5°C Termina-
Metabolic
rate tion of hibernation normally occurs as temperatures
Eskimo dog increase. Hedgehogs wake up after 4 to 5 hours of ex-
posure to 12°C, and arctic marmots wake up at tem-
40 S020 10 0 10 20 30 —40 peratures above 14°C; the Mohave ground squirrel,
Temperature, °C however, can remain dormant even at 27°C (Folk,
1974), this high value corresponding to the higher-
Figure 8-13 Increase in the metabolic rate of homeotherms temperature regime of its environment.
from warm, cold, and very cold environments in response to Insect diapause and heterotherm hibernation are
decreasing environmental temperatures. Metabolic control of accompanied by reduced metabolism and energy uti-
internal temperatures is least important for the well-insulated dog lization. They not only are adaptive characteristics in
and most important for the tropical raccoon, which is not mor-
terms of surviving low temperatures, but they also ef-
phologically adapted to maintain internal temperatures in a cold
fect a considerable savings in an animal’s energy bud-
environment. (After Irving, 1964. Copyright American Physiological
Society, Washington, D.C. Reproduced with permission.)
get. Small homeotherms, such as mice, bats, and
hummingbirds, which have very high metabolic rates
during the day (or night for nocturnal animals), would
highly variable phenomenon that may be initiated by suffer substantial losses of energy if they sustained this
shortening photoperiods, decreasing temperatures, or rate during their period of rest. This is avoided by be-
shortage of food and water. It may be permissive, in coming heterothermic, permitting the body tempera-
which case hibernation is an optional rather than an ture to drop and entering a condition of torpor during
obligatory response to unexpected stress. It may be the nonactive hours. Some birds undergo a similar
seasonal, in which body temperatures drop in a regular torpor during the coldest parts of the year.
seasonal rhythm under internal control even in a con- A second major type of physiological adaptation in-
stant temperature environment. No stress is required, volves avoidance of freezing by lowering the freezing
and the adaptation is initiated before the period of point of cellular fluids. Very few animals can survive
stress. Alternatively, hibernation may be obligate, as actual freezing, but by lowering the freezing point of
when it is triggered by stress such as low temperature the body, many poikilotherms can survive very low
or reduced food supply. Some carnivores, such as temperatures. Reducing the water content and increas-
bears, badgers, and raccoons, do not go into a true hi- ing the fat content of the body, increasing the concen-
bernation. They merely experience a period of winter trations of organic and inorganic materials dissolved in
lethargy when their body temperature drops 4 to 5°C cellular fluids, and the formation of colloids can lower
and their metabolic rate falls by 40 to 50%. The heart the freezing point of insects by as much as 50°C (Allee
of these animals ceases to function at a body tempera- et al., 1967). An alternative to this antifreeze strategy is
ture of 17 to 21°C, whereas the heart of some true hi- the elimination of all free water, such as in the eggs,
bernators continue to beat at temperatures of 0°C or spores, or resting stages of some poikilotherms. Once
even slightly below. in a waterless condition, temperatures approaching ab-
Heterotherms seek out sheltered habitats for their solute zero (-237°C) can be tolerated.
hibernation, and hibernation is less common in envi- A third type of physiological adaptation involves
ronments in which animals cannot find shelter from different cardinal temperatures for different parts of an
very low temperatures. Some of the hibernators allow animal. With an air temperature of -16°C, the foot of a
their body temperature to drop to within a few de- seagull may approach 0°C while the upper leg is at
grees of the surrounding air temperature, but most are 38°C. An Eskimo dog may have a central body temper-
aroused as their bodies approach 0°C, so they can hi- ature of more than 37°C while the air temperature is
bernate only in microenvironments that remain above —30°C, its snout is at 5°C, and its foot pad is at 0°C
224 CHAPTER 8 Temperature as an Ecological Factor

Eskimo dog Gull

Air temperature Air temperature
SMOG.

Figure 8-14 Temperature variation between the body and the extremities in two
homeotherms living in cold environments. The temperatures at various points from the body to
the extremities are shown. (After Irving, 1964. Copyright American Physiological Society, Washington,
D.C. Reproduced with permission.)

(Figure 8-14). Heat loss from the body is greatly re- vection to the cold venous blood, which thus is warmed
duced by tolerating reduced temperatures in those parts and carries heat back to the heart. Many animals also
of the body from which heat could be lost most easily. have the ability to constrict the flow of blood to the
skin or extremities. Vascular constrictor muscles divert
Morphological Adaptations. Morphological adap- much of the warm arterial blood directly into veins,
tations are largely concerned with regulating radiation, thus reducing heat loss. Heat loss from extremities in
convection, and conduction. The development of fur, contact with cold surfaces is also reduced by having a
feathers, or thick layers of fat (blubber) beneath the horny, scaly, or fatty pad with low thermal conductivity,
skin acts to lower heat loss from the surface by reducing very low freezing point, and little or no blood or living
surface temperatures, whereas the layer of low-conduc- cells. This is a common adaptation in animals, although
tivity air trapped beneath fur and feathers greatly re- other adaptations can be found. Wolverines have pads
duces conduction of body heat to the surface. Variation on their feet, which have high thermal conductivity and
in the thickness of fur and in total insulation reflects the which aid in the dissipation of excess heat in summer.
winter temperature environment of the animal and the Hair grows to cover these pads in the winter, and the
degree of reliance on subcutaneous fat deposits and feet can be in contact with snow at —55°C with little
metabolic heat to regulate internal temperatures. The temperature loss.
reduction in the size and projection of extremities be- Morphological adaptations to promote heat loss in
yond the body and fur in high-latitude animals (Allen’s the summer are numerous. Shedding of heavy winter
rule) lowers the surface area over which heat loss can fur and reduction of fat layers under the skin promote
occur; in addition to insulation, many animals that live summer heat loss. Male deer and other male cervid
in cold (or rapid heat loss) environments have a heat- ungulates develop antlers in the summer that serve as
exchange mechanism that reduces the flow of heat heat-exchange organs. The developing antler is cov-
from the center of the body to the skin or cold extrem- ered by a furry skin (velvet) that is rich in arteries and
ities. Arteries that carry warm blood from the central veins, promoting convection and radiant losses of
body area are in close contact with or are surrounded heat. As temperatures drop in the late summer, the
by an intertwining network of veins that carry cold blood supply to the velvet is cut off and the velvet dries
blood back from the extremities. Much of the heat in up and is removed when the animal rubs its antlers
the arterial blood is transferred by conduction and con- against vegetation. After the mating season in the fall,
 Animal Adaptations to Temperature = 225

the antlers are knocked off, only to regrow the follow- distance or local movements undertaken to regulate
ing spring. The antlers are used in sexually oriented body temperature are referred to as thermal migrations
aggressive activities in the fall, but it is thought that (hibernation involves thermal migration to local envi-
heat regulation is one of their primary functions. That ronments that have a favorable temperature regime
the females do not grow antlers may explain in part for this physiological adaptation). Where vegetation is
their preference for the shade. used to help regulate the internal thermal regime of
Animals in hot environments tend to have longer animals, it is referred to as thermal cover. Maintenance
extremities and thinner fur than those in cold envi- of appropriate thermal cover, for shelter both from
ronments. Many desert animals have long legs to winter cold and from summer heat, is an important
raise their bodies off the hot surface and above the component of wildlife habitat management. Where
surface few millimeters of hot air. The ears of ele- thermal migrations are inadequate to regulate daytime
phants provide this bulky animal with a large surface body temperatures in very hot environments, organ-
area for heat exchange. The ears are continually isms may become nocturnal or restrict their activity to
flapped like fans to keep the air in contact with the dawn and dusk.
body in motion, thus promoting convection heat Many organisms that live in environments subject
loss. This source of heat loss is of vital importance; if to temperature extremes avoid the associated prob-
an elephant is immobilized in the hot tropical envi- lems by undertaking long-distance migration to more
ronment and unable to maintain the flow of air over equitable climates. Most birds that breed in the Arctic
its body, then it will soon enter heat coma and will fly south for the winter to avoid low temperatures and
eventually die. the accompanying lack of food and water. In large,
Some animals have morphological skin adapta- well-insulated birds such as geese, the migration away
tions that promote heat loss by evaporative cooling. from the poles may not involve long distances, but
Humans have specialized skin glands that emit smaller birds such as swallows and hummingbirds,
water (sweat) when increased heat loss is required. which are dependent on airborne insects and flowers
The high latent heat of vaporization provides an ef- for food, may follow the summer from one hemi-
fective cooling mechanism as this surface water sphere to the other. Annual migrations of as much as
evaporates. 25,000 miles are made by shorebirds utilizing the Arc-
tic and South America; the arctic tern holds the long-
distance commuter record with a yearly round trip of
Behavioral Adaptations. Most organisms augment 35,000 miles (Wetty, 1963).
their morphological and physiological adaptations Most of the long migrations are made by birds, but
with behavioral adaptations that promote either heat it is not an exclusively avian adaptation. ‘The monarch
conservation or heat loss. It has been noted that poik- butterfly makes an annual migration from its winter
ilotherms do not regulate their temperatures internal- location at Pacific Grove, California, to Alaska and
ly and will tend to equilibrate with the temperature of back. The barren ground caribou of northern Canada
their environment. However, studies have revealed migrate south from their summer range in the arctic
that many poikilotherms are able to maintain their tundra to their winter range in the northern spruce
body temperatures within fairly close limits despite forests, where they find some shelter from the arctic
fluctuating environmental temperatures. Lizards, in- weather, and food in the form of lichens and mosses
sects, and snakes shelter overnight in environments (Kelsall, 1968). Many animals also make altitudinal
that remain warmer than in the open. In the morning, migrations, moving down the mountains into valleys
they move to the warmest location, if possible in full in the winter to avoid deep snow, low temperatures,
sunlight. Once they have reached “operational” tem- and lack of food. All these migrations are regular sea-
perature, they go about their daily activities, distribut- sonal events in response to temperature or tempera-
ing their time between sunny and shaded locations ture-related variations in the availability of food
and orienting their bodies to the sun to maintain al- energy. Other long-distance migrations that are not
most constant optimum body temperature (Bogert, directly related to a temperature factor occur. Many
1959). If they get too hot, then they move to aquatic, species of anadromous fish migrate from their spawn-
deeply shaded, or windy locations, where they can ing and rearing grounds in freshwater lakes and rivers
reestablish an optimum body temperature. Such short to distance oceans, where they feed and mature, re-
226 CHAPTER 8 Temperature as an Ecological Factor

turning to the same stream to spawn several years sult of fluctuating temperatures. Unseasonable low
later. Such migrations are probably the result of a temperatures in early fall or late spring frosts may also
complex of factors, many of which are not yet com- be more important than low winter temperatures, be-
pletely understood. cause they can determine the limit of species by re-
A combined behavioral—physiological adaptation stricting flowering and/or by killing seedlings.
to high temperature that is found in many animals is Low-temperature limitation results in an approxi-
panting combined with salivation. This induces evapo- mately east-west poleward limit to a species distribu-
ration of water from the mouth and lungs, resulting in tion, which bends north along the coast (effect of the
cooling. In some animals, these air movements are re- ocean) and south in the interior (cold, continental
stricted to the mouth and throat (e.g., gullar fluttering winter). The latitudinal distribution of migratory ani-
in desert birds) to reduce water loss or to avoid unde- mals varies during the year, moving poleward in the
sirable changes in blood chemistry that can result from summer and toward the equator in the winter.
excessive ventilation of the lungs (Ricklefs, 1973).
Insufficient Summer Warmth for Growth and Re-
production. Where summers are short and/or cool,
8.6 Role of Temperature in Limiting insufficient gross photosynthesis may occur to balance
the Distribution of Organisms the respiration and litterfall of plants, eliminating net
growth. Long-term occupation of an area is obviously
Ifapopulation of animals is placed in an experimental impossible under these conditions. Even greater quan-
environment that varies in temperature, then they will tities of summer warmth (greater heat sum) are neces-
tend to gather in that part of the environment that sary to produce viable seeds. The upper altitudinal and
most closely matches their temperature adaptations. poleward limits of a species range are often set more
Similarly, if plant seed is sown throughout the experi- by the temperature requirement for reproduction than
mental environment, then the resulting plants will for growth alone. Animals similarly require a mini-
grow best in the temperature regime that most closely mum period above some critical initial temperature
matches that of their native environment. Tempera- if young are to develop successfully, with poikilo-
ture exerts a powerful influence in determining the therms being more sensitive than homeotherms.
geographical distribution of organisms. Although nat- However, the warmer microclimate near the ground,
ural selection and acclimatization can change the tem- the sheltered microclimate of a nest or a den, and
perature relations of a population, there is a limit to homeothermy make animals generally less sensitive to
the genetic and phenotypic adaptability of any popula- low heat sums than are plants.
tion. Competition from species that are adapted to If lack of summer warmth were the factor control-
different temperature regimes also acts as a powerful ling distributions, then it would also produce an over-
deterrent to major changes in the temperature adapta- all east-west poleward limit, but unlike the distrib-
tions of a species that is therefore largely restricted, ution determined by low winter temperature, it would
under undisturbed conditions, to those areas that pro-
bend south on the coast and north in the interior. The
vide optimum temperature conditions for that species
reduced vigor and slower development of organisms
(Rehfeldt et al., 1999). Temperature conditions can
near these limits render the organisms increasingly
limit the distribution of populations in several ways.
susceptible to competition, diseases, and exploitation
by natural enemies.
Polar and Upper Altitudinal Limits
Low Winter Temperatures. Low winter tempera-
tures can derange metabolism, freeze tissues, reduce Lower Latitudinal or Altitudinal Limits
the availability of food and water, or cause mechanical High Summer Temperatures. High summer tem-
injury. The direct effects of low winter temperatures peratures result in the derangement of metabolic
on organisms can determine the upper altitudinal and processes, heat mortality, excessive loss of water, high
poleward extension of species, but they are frequently respiration rate, and a reduction in the availability of
less important than indirect effects such as soil churn- food and water. Plants that have a favorable photosyn-
ing, frost heaving, and solifluction, which are the re- thesis/respiration balance may be restricted from
 Role of Temperature in Limiting the Distribution of Organisms 227

moving to lower elevations or latitudes because the in- results in a complex overlapping pattern of the ranges
crease in nocturnal respiration is greater than the in- of different species. It also results in spatial variations
crease in daytime photosynthesis. These factors result in the overall character of the vegetation that are re-
in an east-west equatorward and lower altitudinal peated all over the world wherever similar tempera-
boundary to species’ distributions that bends south on ture conditions occur. Traveling from the equator to
the coast and north in the interior of continents. the poles over the continents, one passes along a tem-
perature gradient that results in bands of major vege-
Lack of a Winter Chilling Period. Inadequate du- tation types running roughly east-west across the
ration or intensity of winter chilling will limit the continents. A similar temperature gradient and pat-
lower latitudinal or altitudinal range of species that re- tern of vegetation can also be found with increased el-
quire chilling to break dormancy or stimulate flower- evation near the equator. The evergreen tropical rain
ing. Insects and plants of the temperate region will not forest found at the equator at sea level is replaced by
spread into subtropical regions partly because of the semideciduous and deciduous hardwood forest toward
lack of chilling. It is widely known that apples cannot the poles or at higher elevation (Figure 8-15). This in
be grown in fruit-growing areas at lower latitudes for turn gives way to dry forest, grassland, and, in some
this reason. areas, desert. Farther poleward or higher up this is re-
Different species of plants and animals vary as to placed by deciduous forest, which gives way eventu-
which temperature factor limits distribution. Some are ally to evergreen conifer forest. At high latitudes and
limited poleward by low winter temperatures. Others high tropical altitudes, conifer forest is displaced by
are limited by inadequate summer heat. This variation tundra vegetation, which continues until bare rock or

High i ame IR AP a Sa eS
TS ar re =
Rock and ice |
|
eel a er a a eS oR a a ee 7 |
Tundra or alpine vegetation | :

Saat aa eae RR — i | |
Conifer forest _ ern
| |

Nai ce Ldtice PP tae aoe Od eee nL | | |

25 Deciduous forest
“]| |
|
aan |
jaa

madgucil acu Senor... teae 7 | | | |

sat irep os Grassland or desert | | | | |
mountain | : | | l

eC paar in ele ae : |
Dry forest | | | | |

Bee neh tebe ort weous a

|
| | | | |
Low
Equator Tropics Temperate Boreal Polar
Trapical region region region
rain forest LIFE ZONES
0) Latitude 90

Figure 8-15 Latitudinal and altitudinal patterns of vegetation that reflect gradients of tem-
perature and moisture. The exact vegetation patterns shown in the diagram rarely occur along a
straight line from the equator to the poles, and they do not always occur as shown from sea level to
mountaintop, but the general sequence can be observed in many parts of the world.
228 CHAPTER 8 Temperature as an Ecological Factor

permanent ice is reached. In South America, the ent. The plant community of a frosty valley floor is
alpine tundra is called paramo. The dry desert or grass- often markedly different from that of the thermal zone
land vegetation may be absent from this sequence if on the valley slope, and as one travels poleward from
the interaction of temperature, moisture, fire, and the equator, the point at which winter frost first occurs
herbivory permits continuous tree growth. is marked by the sudden loss of many frost-sensitive
The characteristic bands of vegetation and associ- species. Perhaps the best-defined interface between
ated animal and microbial communities that occur life zones is that between conifer forest and either
along altitudinal and latitudinal temperature gradients alpine or arctic tree lines. The environmental deter-
are called biomes, and the regions that they occupy are mination of tree lines is extremely complex, as we shall
called /ife zones—areas in which mature or climax veg- see in Chapter 13, but temperature plays an important
etation (see Chapter 17) is dominated by plants of a role through its influence on snow depth, soil churn-
given life form. Tropical evergreen hardwood forest, ing, solifluction, permafrost, flowering and seed pro-
grassland, and conifer forest are three examples of duction, seed germination, and seedling survival.
biomes. The overall east-west alignment of biomes is The role of temperature in the distribution of
greatly modified by mountains, lakes, oceans, and re- plants and animals has attracted a lot of attention and
gional atmospheric circulation patterns, all of which deserves a fuller treatment than is possible here. The
modify temperature gradients. The distribution of interested reader is referred to the extensive treatment
biomes is further modified by local variations in mois- of temperature and life in Precht et al. (1973). The ef-
ture, fire, soils, and animals and by human effects on fects of historical changes in temperature regimes on
the distribution of plants and animals. Biomes are the growth and distribution of conifers are reviewed in
shifted toward the equator as one goes up in elevation Graumlish and Brubaker (1995), and the effects of
and reach higher elevations on southern aspects than possible future climatic change are discussed in
on northern aspects. Gucinski et al. (1995). The ecophysiological processes
The consistent patterns of biomes around the and effects involved in these distribution changes are
world gave rise to Hopkins’ bioclimatic rule, which says considered by Woodward (1995). We will encounter
that biological events in the spring are delayed 3 to 4 the temperature factor again in subsequent chapters
days for each higher degree of latitude and for each because of its important interaction with several other
100 to 130 m increase in altitude. The situation is re- parameters of the physical environment. Extensive
versed at the end of the summer. Hopkins (1920) also modeling has been done to assess the possible effects
claimed that there was a seasonal retardation of 4 days of global temperature changes on the location of bi-
for each 5° longitude from west to east in the United omes (Watson et al., 2001). The present consensus is
States. It is generally believed that although some that there are unlikely to be shifts in the geographical
species may conform to the bioclimatic rule, the com- location of entire biomes because of the diversity of
plex determination of most ecological events means temperature adaptations of the component species.
that the rule will not apply to all circumstances. However, it is thought that continued temperature
Because of variations in temperature adaptations, change will alter the species composition, function,
species vary in their geographical ranges. Species ap- and character of biomes, especially at the edges of
pear and disappear irregularly along the gradient from their distributions.
the equator to the poles or from low to high elevation,
and this often results in indistinct boundaries between
different biomes or community types within a biome. 8.7 Importance of the Temperature
This is not always the case, and groups of species Factor in Forestry
sometimes replace each other completely over rela-
tively short distances (see Chapter 13). Where this oc- Approximately 10% of the world’s surface and 33% of
curs, there will be a sharp boundary between life the land surface is covered with forest, woodland, and
zones, and many species may have the same limit to shrubland (Whittaker, 1975a). Closed forest accounts
their distribution. Sometimes this may coincide with a for approximately 20% of the land surface (Council on
physical discontinuity in the environment, such as a Environmental Quality, 1981). Of the forest area, ap-
change in soil, but it is frequently related to a biologi- proximately half is tropical forest growing in environ-
cally critical point in a continuous temperature gradi- ments where temperatures are relatively unvarying and
 Importance of the Temperature Factor in Forestry 229

generally not extreme. The other half is characterized root tips in deeper, colder soil layers. These deeper
by temperatures that periodically limit biological ac- layers may also have lower levels of available nutrients.
tivity and may pose problems for the forest manager at Another temperature-related problem results from
some time of the year. Much of the world’s timber sup- variation in the temperature adaptations of different
ply is grown and harvested in environments where ei- ecotypes. Winter hardening of seedlings is regulated
ther macroclimatic or microclimatic temperatures can by the combination of exposure to lower temperatures
influence forest land management in some manner. and decreasing photoperiod. Ecotypes from different
Extreme temperatures and rapid changes in tem- regions have a different “environmental calendar,” and
perature can damage trees in many ways. Bud killing if moved to a new climatic region, they may suffer frost
can result in multiple leaders. Stem girdling or patch damage because of inadequate hardening. Seedlings
killing of the cambium can either kill trees or create that are grown in nurseries at lower latitudes or alti-
lesions that lead to infection by decay organisms. tudes may suffer frost damage when planted at higher
Rapid drops in temperature or very low temperatures latitudes and altitudes if their temperature experiences
can result in frost cracking of stems, which reduces in the nursery have not prepared them for the temper-
timber values and can lead to invasion by decay organ- ature conditions at the planting site.
isms. ‘This is frequently more severe on the most pro- Species vary greatly in their temperature adapta-
ductive sites, where very fast-grown trees seem to be tions. The planting of a single species over a wide
more susceptible than their slow-grown counterparts range of sites, aspects, altitudes, and latitudes that vary
on drier, less-productive sites. greatly in temperature has naturally resulted in vari-
Periods of warm weather during winter, when soils able regeneration success. A species that may regener-
are either dry or frozen, can lead to damage to tree ate satisfactorily in the cool microclimate of an open
crowns because of moisture stress. Hot weather in the forest on a southerly slope may fail entirely in the hot
spring at the time of tree planting can result in heavy microclimate of a large clearcut on the same site. Sim-
mortality of planted seedlings even if the soil is moist ilarly, radiation frost damage to seedlings in a clearcut
because the seedlings have not had time to produce area with very clear, dry summer air is often absent in
enough new roots to compensate for the damage done adjacent areas managed under the shelterwood system
to their roots during transfer from the nursery and or by uneven-age management. A species that regen-
planting. Fall (autumn) planting may avoid this prob- erates and grows well in a midslope position may fail
lem. Soils remain warm in the fall much longer than or grow poorly on a frosty valley bottom or at higher
the atmosphere, which permits continued root growth elevations.
so that the trees are equipped to deal with spring ‘Temperature effects can also be indirect. Douglas-
moisture stress when it occurs. However, soils are fir in coastal British Columbia suffers serious leader
often drier in the fall than in the spring, and “Indian branch and stem damage in areas of heavy wet snow-
summer” weather in the fall can induce lethal mois- fall and freezing rain, whereas the same ecotype may
ture stress in planted seedlings. grow well 100 m lower in elevation, where the tem-
Except for the surface few centimeters, soil tem- perature is only slightly warmer but where most of the
peratures in northern areas may remain low far into precipitation falls as rain.
the growing season. This limits root growth and Microclimate, especially the thermal regime of mi-
makes regeneration after clearcutting very difficult. croenvironments, is often important for forestry. Mi-
The seedlings develop shallow root systems, which crosites that are only a few meters apart may differ
makes them susceptible to spring and fall frost heaving greatly in temperature regimes according to aspect,
if the forest floor is removed, and summer drying of slope, view factor, shading, soil moisture, and soil
the surface of the soil if the forest floor is left undis- color. Frost pockets may preclude reforestation of mi-
turbed. If seedlings are planted in the spring so that crosites that are adjacent to areas that are only a cou-
their roots are in cold soil layers, then they will grow ple of degrees warmer in which regeneration is
more slowly than naturally regenerated seedlings that successful.
develop their roots in warmer, surface layers. This The factors responsible for the natural distribution
problem can be exacerbated by “container-grown” of many tree species are still not completely under-
seedlings if all the root tips are concentrated at the stood, but temperature plays an important role. Ade-
bottom of the root “plug,” because this places all the quate summer warmth for growth and reproduction,
230 CHAPTER 8 Temperature as an Ecological Factor

winter temperatures that satisfy chilling requirements, sequently more of the absorbed energy is available to
freedom from damage caused by temperature ex- raise the temperature of objects (M), which in turn
tremes or rapid rates of temperature change, soil tem- warms the air, which then carries the heat away (sensi-
peratures, and the interaction between temperature ble heat flow, H). Approximately 30% of the net radia-
and moisture all are important. In addition, tempera- tion in a forest with adequate water supplies is
ture plays a major role in determining nutrient avail- dissipated as sensible heat. In contrast, the sensible
ability. Decomposition is much slower in cold than in heat flow from a dry, bare soil may be as much as 70%
warm environments, and fertilization experiments of the net radiation. Table 8-5 compares the parame-
have demonstrated that the temperature-related cli- ters of the energy-balance equation for a clearcut and
matic limitation of tree growth in the north is as much an 18-year-old Douglas-fir plantation in coastal
the result of temperature regulation of nutrient cy- British Columbia.
cling as of more direct effects of temperature on plant Alteration of the radiation balance after clearcut-
physiology. Some of these temperature limitations on ting influences rates of snow melt in the spring, which
where trees can be grown can be modified by natural can have a profound effect on the hydrological cycle,
disturbance or forest management, but some cannot, the availability of moisture in the summer, and the
and this places constraints on how forest ecosystems stream environment. Radiation frosts and cold air
are managed. drainage can create frost pockets after clearcutting or
Of all forest management activities, it is clearcut- patch harvesting, which may lead to the failure of
ting that has the greatest effect on temperature be- plantations or significant damage to the trees.
cause of its alteration of the radiation-energy balance The reduction in forest cover increases the view
(Figure 8-16). Forests have a much lower albedo than factor of seedlings in a clearcut. The farther from the
clearcuts, so less total energy is absorbed by clearcut stand edge, the greater the exposure of the seedling to
than by forested areas (R,, is lower). However, most of the sky (Figure 8-17). This in turn results in higher
the absorbed radiation reaches the ground, and much summer daytime temperatures and lower summer
more heat is conducted down into the soil in a clearcut nighttime temperatures than in the adjacent forest
(G is higher) than in the uncut forest. Because of the (Figure 8-18), the degree of difference depending on
active transpiration by an uncut forest, much of the net the size of the harvest area. Table 8-6 summarizes the
radiation is lost as latent heat of evaporation (LE). LE effect of harvest patch size on a number of meteoro-
has a low value in a clearcut, at least initially, and con- logical parameters.
Slope, aspect, altitude, and latitude all affect the
change in temperature conditions that accompany
R,=G+H+LE+M forest harvesting. Similarly, the radiative and ther-
Incident F's ": mal properties of the soil surface are important. Sur-
radiation f:
Reflection faces such as a dark-colored organic forest floor,
and reradiation which has low thermal capacity, low thermal con-
Atmosphere
ductivity, and low albedo, develop very high daytime
surface temperatures. Light-colored mineral sur-
faces that have high albedo, high thermal capacity,
and high thermal conductivity do not develop high
Forest/soil surface temperatures. The former type of surface
surface can reach such high temperatures that seedling mor-
tality occurs as a result of the death of stem tissues.
The importance of this is seen in the variation in
mortality of seedlings that, because of the angle at
Figure 8-16 The radiant-energy balance of a forest. R,, = which they are planted, either shade their own stems
net radiation absorbed by the ecosystem; G = conduction of at ground level or do not. A study 60 years ago re-
heat away through the soil; H = convective removal of heat by ported that seedlings inclined to the south on
the atmosphere; LE = heat lost through evaporation of water in southerly aspects (in the northern hemisphere) had a
transpiration; M = increase in the heat contained in the ecosys- higher survival rate than seedlings inclined to the
tem (i.e., its temperature). north (Table 8-7).
 Importance ofthe Temperature Factor in Forestry 231

Table 8-5 Energy Balance Comparison Between an 18-Year-Old Douglas-fir Plantation and a Clearcut at Solar
Noon on a Clear Day at the UBC Research Forest, Haney, B.C.
Energy Flux, W m2

Clearcut
Douglas-Fir (No Vegetation
Parameter Stand Regrowth)

R,, Net radiation flex density 520 550

G Soil heat flux density 32 40
LE Latent heat flux density 301 70
H Sensible heat flux density 183 440
M Rate of heat storage in vegetation (on an area basis) 4 0

After Black, 1977; McNaughton and Black, 1973.

Surface conditions also influence the occurrence of ter and in late winter/early spring can dramatically re-
radiation frosts in harvest areas. Organic layers are ef- duce overwinter survival. The Mountain Pine Beetle,
ficient radiators of energy but poor conductors of heat which currently has killed lodgepole pine in central
from the underlying soil. Undisturbed forest floor, es- British Columbia over an area of more than 6 million
pecially if dry, will promote radiation frosts. In com- ha, is held in check by low early winter temperatures.
parison, the good thermal conduction properties of Several years of warm winters or late occurrence of
moist mineral soil will often prevent radiation frosts in low temperatures has created the potential for the epi-
clearcuts where the mineral soil has been exposed. demic. The abundance of mature and overmature
Low winter temperatures are important in the pine, caused by a combination of past extensive fires
population regulation of forest insect pests (Carroll et and/or beetle outbreaks and more recent forest fire
al., 2002; Stark, 1959). Very low temperatures in the control, has turned the climatically determined poten-
middle of winter may have little effect on overwinter- tial in one of the largest epidemics on record (Carroll
ing insects, but low temperatures in late fall/early win- et al., 2002).

S oo

Clearcut area
Tree height 4
18.9 m
factor
View
S 3)
Minimum
°C
air
temperature,

0.4 T Sai) 0 +1.0 2.0

0 | 2 3 4 5 6 Distance from stand edge (tree heights)
Distance from stand edge (tree heights)
Figure 8-18 Average minimum air temperature 6.4 cm
Figure 8-17 View factor (proportion of sky that is visible) above the soil surface for 18 days in summer in a clearcut
at ground level in a large rectangular opening in a lodge- and an adjacent lodgepole pine stand. Radiation frost oc-
pole pine forest as a function of distance from the stand curred in the clearcut but not in the uncut stand. (After
edge. (Afier Cochran, 1969a. Reproduced with permission of PH. Cochran, 1969a. Reproduced with permission ofPH. Cochran.)
Cochran.)
Zoe CHAPTER 8 Temperature as an Ecological Factor

Table 8-6 Effect of Forest Clearing Size (Expressed in Terms of Diameter and Diameter/Tree Height Ratio) on
Some Microclimatic Parameters
Size of Forest Clearing*

Diameter, m: 0 12 22 38 47 87
Microclimatic
Parameter Diameter/tree ht:0 0.46 0.85 1.47 1.85 3.36
pee oe Se ee ee
Mean angle of shielding (h), degrees 90 Ue 59 48 40 26

Outgoing radiation as percentage of open land value 0) 11 31 52 66 87

Precipitation as percentage of open land value 87 - 105 — 102

Midday July temp: excess of clearcut over forest, °C 0 ON 2.0 5.2 5.4 41

Geiger, 1965. Copyright Harvard University Press. Used with permission.

@Only the largest clearing would be considered to be a clearcut.

Tuble 8-7 Summer Mortality of Engelmann and Norway Spruce Transplants in Their First Year in the Field as a
Result of Stem Girdling, as Affected by Aspect and Direction of Inclination of the Trees
Species Aspect Inclination? Mortality, %

Engelmann spruce Level South 3.7

Level Vertical 11.4
Level North 30.6
Norway spruce Level South 8.5
Level North 33.5
Engelmann spruce South Vertical 7.0
Level Vertical 11.4
North Vertical 2.4

Data from Korstian and Fetherolf, 1921.

“Where the trees were inclined so that they shaded their own stems (south), mortality was much lower.

Low winter temperatures are very important for on the global radiation budget and global tempera-
northern (boreal) forest harvesting because many bo- tures. In Chapter 5, the question of human activities
real soils are too wet in the summer to permit equip- on atmospheric CO, levels was introduced. The possi-
ment access. Most logging is done in an 80- to ble effects of elevated atmospheric concentrations of
100-day period when the soils, rivers, and lakes are CO, and other gases on world temperature are consid-
sufficiently frozen for the construction of ice roads ered here briefly. The role of forests in carbon storage
and bridges and for logging equipment to drive off the and regulation of atmospheric CO, will be considered
roads. If global warming leads to a significant shorten- again in Chapter 20.
ing of the frozen soil period, then northern forestry There has been an enormous international effort
will become very difficult. over the past two decades to document and understand
past global climate change, especially global tempera-
tures (ERBY, 2002; Hansen et al., 2001, 2002; Watson
8.8 Human Effects on the et al., 2001). The 2002 meteorological year was the
World Radiation Budget and second warmest since the late 1800s. The mean global
Global Temperatures surface temperature was 0).51°C warmer than the 1951
to 1980 average. The warmest year since the late 1800s
We cannot close this section of the chapter without was 1998; the third warmest was 2001. The greatest
speculating briefly on the effects of human activities warming in 2002 was in Siberia and the Arctic. How-
 Human Effects on the World Radiation Budget and Global Temperatures 233

ever, the monthly fluctuations in temperature for any

given region are generally larger than the magnitude
of global warming, making it difficult to interpret the
local ecological consequences of the global trend.
Many areas experience cooler-than-normal tempera-
tures while the global average is rising. Figure 8-19
shows the long-term (160,000 years) temperature
record and its correlation with atmospheric CO); also
shown are the 1,000-year record, the details of the past
140 years, and predicted trends for the next century Concentratio
CO,
(ppm)
based on computer models. The broad spread of pre-
dictions reflects the considerable uncertainty in mak-
ing global temperature forecasts. Nevertheless, all
predictions suggest a biologically significant increase
well within the life span of most tree species.
CO, and certain other gases (greenhouse gasses) in (°C)
from
Departures
todays
temperatures
160 140 120 100 80 60 40 £20 0
the atmosphere act much like the glass in a green- Age (kyr BP)
house: they permit light to enter but trap long-wave- (a)
length (thermal infrared) radiation. This contributes
to atmospheric warming (McBean and Thomas, 1991;
Ramanthan et al., 1989), although the atmospheric
regulation of temperature is much more complicated
than suggested by the simple greenhouse analogy. Part
of the uncertainty relates to the role of clouds in
the earth’s radiation budget. High clouds (which are
mainly ice crystals) differ in their opacity to different
solar radiation wavelengths, complicating our under-
standing of their role in the greenhouse effect. High
cirrus clouds are thought to warm the earth because
they reflect little solar radiation while trapping the
earth’s radiation, whereas low clouds cool the planet
by effectively reflecting solar radiation, but there is average 1000
(°C)
1961-1990
from
Departures
temperature 1200 1400 1600 1800 2000
still considerable debate about the relative heat trap- Year
ping by high and low clouds. Nevertheless, the (b)
“greenhouse” concept is well established; overall, the
+0.6
atmosphere does function much like a greenhouse,
and human-caused increases in atmospheric concen-
trations of greenhouse gasses do appear to be raising
atmospheric temperatures.
Human activities have increased the atmospheric 13012)
CO, concentrations by approximately 26% from the

Figure 8-19 Variation in global atmospheric tempera-

(°C)
anomaly
Temperature
tures over: (A) the past 160,000 years (data on temperature and —5 year mean
atmospheric CO}, redrawn from Barnola et al. 1987); (B) the e Annual mean
past 1000 years showing departures from the 1961-1990 mean —0.4
1880 1900 =—:1920 1940 1960 1980 2000
(redrawn from Watson et al. 2001); (C) deviations from the
Year
1951-1980 mean over the past 120 years (redrawn from
(c)
Hansen et al. 2002).
234 CHAPTER 8 Temperature as an Ecological Factor

preindustrial level of 280 ppm to 353 ppm in 1990 heterotrophs and the availability of nutrients to au-
(Watson et al., 1990) and 368 ppm in 2000 and may totrophs. Slower decomposition and a reduction in nutri-
rise to between 540 and 970 ppm by 2100, under the ent uptake by plants when temperatures drop results in a
present range of assumptions (Watson et al., 2001). significant difference in the biogeochemistry of forests
that grow in warm and cold environments and between
Emissions of other greenhouse gases, such as methane
summer and winter in the latter. Temperature influences
(CH,) and nitrous oxide (NO), are also increasing. It the geographical distribution of different organisms and
is projected that if these releases were to continue to plays a major role in determining the life form and
increase at the present rate, then their combined effect species composition of plant communities.
on global temperature would be equivalent to dou- The temperatures experienced by animals are modi-
bling the atmospheric CO, by the 2030s. During the fied by physiological, morphological, and behavioral
1980s, the increase in long-wavelength radiation trap- adaptations. Both homeotherms and poikilotherms gen-
ping by the atmosphere was attributed as follows: 56% erally are able to restrict their temperature experiences to
by CO), 11% by CHy, 6% by N,O, and 24% by chlo- a narrower range of temperatures than that occurring in
rofluorocarbons (Hare, 1988; McBean and Thomas, their environment. Plants have somewhat fewer options,
1991). Approximately 50% of the CO, released by fos- owing to their lack of mobility and fewer possible physio-
logical and morphological adaptations, and they general-
sil fuel burning has remained in the atmosphere; it is
ly cope with problems by becoming inactive during
believed that the rest has been absorbed by the oceans
periods of unfavorable temperatures. Their physiological
and terrestrial ecosystems. adaptations involve sensitivity to a variety of “clues” that
Prediction of future changes in temperature re- increase the chances of being prepared to meet unfavor-
quires prediction of future changes in release of CO, able conditions. These clues include photoperiod and
and other greenhouse gases and their behavior in the temperature experiences. Plants also exhibit a consider-
atmosphere and their absorption by aquatic and ter- able ability to alter their cardinal temperature by ac-
restrial ecosystems. There is great uncertainty about climatization in response to changed environmental
future trends in fuel consumption, which accounts for temperatures. This ability permits them to continue
approximately 77% of CQO) release, about the capacity growth under fluctuating environmental conditions and
of forests to sequester and store atmospheric carbon, in a variety of environments. However, every species has
and about the rate of uptake by lakes and oceans. This a range of tolerances beyond which its biological func-
tions will be impaired.
leads to variation in predictions concerning future ef-
Recognition of the temperature factor is important in
fects on world temperatures. The climate is also sensi- forest management, whether the main concern is with
tive to the reflection of short-wavelength radiation by timber, wildlife, fisheries, range, or recreation. The
aerosols from volcanic eruptions and air pollution. forester can modify the microclimate significantly within
Estimates of changes in the average surface air any macroclimatic region by appropriate manipulation of
temperature accompanying a doubling of CO, levels, the forest but must recognize the overall limitations on
or its equivalent, made in 1990, range from 1.5 to the options imposed by the macroclimate.
4.5°C, with a “best estimate” of 2.5°C (Houghton et
al., 1990). 2001 projections of the expected increase in
the globally averaged surface temperatures in the 1990 TAKE-HOME MESSAGE
to 2100 period based on very complex computer mod-
As a physical factor that plays a key role in determining
els of the earth’s atmosphere and climate are between
the biological potential of an ecosystem, temperature
1.4 and 5.8°C, which is equivalent to 2 to 10 times must be considered in planning the management of for-
larger than the observed warming over the last centu- est ecosystems. Whether the objectives of management
ry, a rate of warming that seems to be without prece- are primarily concerned with the environmental values of
dent over the past 10,000 years (Watson et al., 2001). wildlife and biodiversity or with the social values of em-
ployment, harvest of wood products, and creation of eco-
nomic value, the constraints posed by temperature must
SUMMARY
be understood and respected.
‘Temperature is one of the more important environmen- If predictions about climate change caused by green-
tal variables. It plays a critical role in the availability and house gas emissions prove to be correct, then tempera-
ecological effectiveness of that vital component of all life ture regimes will change significantly. Climates will
systems: water. It influences the availability of food to become more variable, and the seasonal patterns of tem-
 Study Questions 235

perature will become less reliable. This has important STUDY QUESTIONS
implications in seasonally extreme environments where
both photoperiodic and thermoperiodic adaptations en- 1. If low soil temperature is a problem, then when
able species to cope with otherwise unfavorable tempera- would you plant seedlings on a wet site and on a
tures. Although global warming may render some dry site? Why?
ecosystems more productive, the desynchronization of 2. What are cardinal temperatures? What do we mean
temperature, photoperiod, and temperature adaptations by optimum temperature?
threatens negative consequences for many species. How- 3. What role does winter chilling play in the annual life
ever, many species have considerable powers of acclima- cycle ofa tree?
tion, and within-species genetic diversity may equip
4. Which species will have the greater heat sum re-
many species to adapt to temperature changes. The com-
quirement for bud burst—one that grows in a conti-
plexity of biological responses to possible global warming
nental climate with a long cold winter, or one that
is such that we cannot make predictions with much con-
grows in a coastal climate with a mild winter but with
fidence. With the increasing risk of human-induced cli-
occasional, unpredictable cold spells, especially in
mate change, it becomes even more important to
late winter?
conserve genetic and species diversity and to manage for-
est landscapes in a way that confers on them the greatest 5. What are greenhouse gases, and why do they matter?
possible resilience to changes in climatic regimes. 6. What is view factor, and how is it important?
‘Temperature relationships are usually important in 7. What are life form spectra? How do they relate to
managing wildlife habitat and in selecting forest harvest- biomes?
ing and silvicultural systems and almost always important
8. How does the temperature factor determine latitudi-
in managing riparian (streamside) ecosystems. They af-
nal limits of plant species?
fect forest health and forest-insect and forest-pathogen
relationships. In short, the temperature factor should al- 9. What are the major adaptations of animals to the
ways be considered in our quest to understand forest temperature factor?
ecosystems and in our design of their management. 10. How is temperature important in forestry?
 Wind
Ecological Effects of Atmospheric Movements

9.1 Introduction wind plays an essential role in the reproduction and

dissemination of vegetation. Animals are also trans-
The quantity of incident solar energy and the propor- ported by wind currents. Seagulls will follow a ship for
tion that is absorbed by either the atmosphere or the miles without wing beats by riding on the air turbu-
surface of the earth varies from place to place. This re- lence in the wake of the ship. Insects, spiders, and
sults in regional variations in atmospheric and surface other small animals are dispersed by wind, and it is be-
temperatures, which in turn result in differences in lieved that surprisingly large animals are occasionally
atmospheric pressure. Such differences lead to the transported long distances by unusually high and tur-
movement ofair (i.e., wind) from areas of high to areas bulent winds. This is believed to be important in the
of low pressure. Wind thus is the result of the conver- development of both flora and fauna on new islands
sion of solar radiation to thermal energy and thermal formed by volcanism.
energy to the kinetic energy of wind systems. Wind is important to humans in many ways. It is
Winds have a wide variety of ecological effects. an important means of dispersing atmospheric pollu-
They transport water vapor from lakes and oceans to tants, and many cities would have far more serious air
the land and thus ensure the supply of rainfall. The pollution problems if it were not for the cleansing ac-
friction of one turbulent air mass on another generates tion of wind. In contrast, high winds can destroy prop-
static electricity, which results in lightning, which in erty, uproot trees, sink ships, erode soil, and destroy
turn may start wildfires and contribute nitrogen to farm and forest crops.
ecosystems by producing oxides of nitrogen in the at- This chapter examines the spatial and temporal vari-
mosphere. Wind removes dust and organic particles ations of wind and the relationship of wind to topogra-
from one ecosystem and deposits them in another, phy. Effects of wind on plants, animals, and entire
sometimes to the extent that a deep layer of /oess soil ecosystems are discussed, as well as the important recip-
develops. The removal of fine soil particles by wind rocal effect of plants on wind. The chapter concludes
can convert medium-textured soils to sand or rocky with a discussion of the importance of wind for forestry.
substrates, with accompanying changes in soil fertility
and moisture. This is an important mechanism in the
process of desertification. Wind influences evaporation 9.2 ‘Temporal and Spatial
and transpiration and can produce desiccation damage Variations in Wind
and even the death of plants. Wind transports heat en-
ergy. By carrying warm air to cold regions and hot air Wind results from the differential heating of the at-
to cold regions, it serves to moderate temperatures. By mosphere, which in turn results largely from differ-
transporting pollen grains, spores, and seeds of plants, ential heating of the earth’s surface. As the earth

236
 Temporal and Spatial Variations in Wind — 237

rotates, areas are sequentially warmed and cooled, shores of Lake Erie was shown to result from air pol-
giving rise to global variations in air pressure, which lutants that originated from industrial areas around
drive the global air circulation systems and thereby the lake. Ozone was formed from these pollutants by
determine to a large extent the world’s variation phytochemical action at some height above the lake
in climate. and brought down to low levels by the daytime circu-
Low-latitude regions between the tropics and the lation (Figure 9-1A). The ozone was then carried
equator are constantly warm, whereas the polar re- with the onshore breezes to the land, where it dam-
gions are always cold. In between lies a broad zone in aged the crops (Mukammal, 1965). In the second
which there is an exchange between equatorial and case, blister rust disease was found to be infecting
polar air. This exchange serves to prevent equatorial pine trees in an area approximately 15 to 20 km in-
areas from getting ever hotter and polar areas ever land from Lake Superior. To the surprise of investiga-
colder. The transfer of excesses and deficits of heat is tors, the disease was not found in pine stands between
accomplished by a systematic global circulation of air the infected area and an area 8 km closer to the lake in
masses. In simple terms, air rises in equatorial re- which currant bushes (the alternative host for the dis-
gions and travels poleward, sinking as it goes, return- ease and the source of the spores infecting the pine
ing to the equator from approximately 25 to 30 trees) were growing. The explanation for this unex-
degrees of latitude as surface winds that blow from pected distribution of the rust lies in the fact that the
the northeast in the northern hemisphere and the spores of the rust that was growing on the currant
southeast in the southern hemisphere; these winds bushes were released only at night, when they were
are known as trade winds. Polar air sinks and flows to- carried out over the lake by the nocturnal offshore
ward the equator, only to rise again between 60 and breeze. The nocturnal counterflow (Figure 9-1B)
50 degrees of latitude in a counterflow back to the then carried the spores to higher altitudes and back
pole. Between these two circulation systems is a mid- over the land, where the subsequent downflow de-
latitudinal circulation that gives rise to the westerly posited them on the trees in the infected area. This
winds of midlatitudes. This is a highly simplified rep- explanation of the unusual pattern of infection was
resentation of world air circulation. In reality, the supported by tracing the air movements with smoke
pattern shows much more diversity as the result of and balloons (van Arsdel, 1965).
the continental land masses and the change of the Valleys, especially those in mountainous regions,
seasons. produce their own wind systems as a result of differen-
Land surfaces warm up faster during the day than tial heating of different aspects or parts of the valley.
do water surfaces. The air that is in contact with As with the water-land circulation system, such winds
land is heated, causing it to expand and rise, whereas are best developed during anticyclonic summer weath-
cooler air moves landward from over the water. This er. There are two types of valley wind. One type re-
creates local onshore breezes that are independent sults from the daytime warming and nighttime cooling
of regional air circulation systems. The reverse oc- of the valley, which results in daytime upslope winds
curs at night. The land cools faster than the water (anabatic) and nighttime downslope winds (catabatic)
surface, and the direction of local breezes becomes (Figure 9-2). The second type involves a plain-to-
offshore (Figure 9-1). Such land—water air circula- mountain flow of air in the daytime because air in val-
tion systems develop only during anticyclonic leys becomes heated and rises more than air over
weather, when the atmosphere is relatively stable. plains; this results in up-valley air movements (valley
They are prevented by cyclonic winds and also by winds). At night, the mountain ridges cool by radiation
cloudy weather, which prevents the development loss faster than the plain. The lower layers of the air
of thermal differences between the land and the over these ridges cool and slide down into the valley,
water. The onshore winds can extend inland as far as where they combine to form a down-valley air move-
30 km and affect airflow up to a height of | to 2 km ment (mountain winds) (Figure 9-3). Glaciers and
(Oke, 1978). snowfields can result in catabatic air movements even
The ecological significance of such winds is exem- in the daytime, although such winds are often over-
plified by two interesting situations that occurred come by the daytime valley winds.
around Lakes Erie and Superior. In the first case, Air may move either as /aminar flow or as turbulent
ozone damage to tobacco plants growing near the flow. In the former, the different air layers ride over
238 CHAPTER9 Wind

i= aot eeeeEEEERIREEEEEEEEeee Ye ‘ ‘s heating

Counter flow

#5
——s
) On-shore sea breezes generated f ,

os by inland convection Life

SS Lake Ee —=— = Agricultural crops ee heating of land

damaged by air pollution
i to

SK
Rapid radiant
cooling

/ Spores carried Off-shore breezes generated by C@™TY!N8 aad: ; y

+ upwards over convection over lake, and Poe A
sinking air over land carried /
lake iS down in: /
a Spores given sinking / /,
. ep hae 4 air } : / /
off at night 27
I SY AY PA
4, x Sy
/ Vv
gf
Lyfe v; u
cage
aq ‘i
Q [>
ay Syty oa CES eee
2,Sy HS
$m wer

SS — Currant bushes Uninfected “Infected

—— Lake Superior ==
SSS infected with pine trees pine trees

\Convection of the warme rust ~ ~~ X55\

& air over the lake an S Oo AN

(b)

Figure 9-1 Air circulation systems between land and water bodies are found along ocean or
lake coastlines during anticyclonic weather. (A) The daytime water-to-land system for Lake Erie
showing the transport of ozone, as described in the text. (B) Nighttime land-to-water system for
Lake Superior, showing the transport of blister rust spores, as described in the text.

each other with little mixing. Such flow is rare in nat- convectional currents that arise from differential heat-
ural situations. More commonly, airflow is turbulent, ing of the earth’s surface.
with large-scale mixing of air from different layers and Mountain topography produces large eddies in the
repeated changes in the direction of the movement of lee of ridges, especially during high winds. Airflows
the body of air. Turbulence results from friction be- tend to be more laminar upwind and downwind of
tween the air and the surface or because of obstruc- such features; the slower the wind, the more laminar
tions to laminar flow. It can also occur as a result of the the flow in the lee area. The narrowing of valleys ac-
 Effects of Wind on Vegetation 239

PL se Catabatic winds

Figure 9-2 Anabatic and catabatic winds shown in a cross-section of a mountain valley.
(Modified after USDA, 1964.)

celerates winds and can lead to eddying. Rough vege- and attain a great drying power, which can result in
tation surfaces and the heating of the land surface can rapid evaporative loss of snow packs in winter and in
also give rise to turbulence (Figure 9-4). moisture removal from the vegetation and soil in sum-
Chinook, or foehn, winds are a particular type of mer or winter.
wind experienced on the lee side of mountain ranges
(also the Santa Anna winds of California). Moist ocean
winds that move inland over mountains lose more and 9.3 Effects of Wind on Vegetation
more moisture as they are forced up and over succes-
sive mountain ranges. On descending the lee side of Wind affects plants in a variety of ways: by disseminat-
the last few ridges, the winds warm up as they are ing pollen, spores, or seeds; by influencing plant phys-
compressed by increasing atmospheric pressure (ap- iology (Whitehead, 1968); and by influencing plant
proximately 1°C increase for every 100 m dropped) morphology.

Cold air draining

down the valley
Daytime off the glacier
valley wind L\ against the valley

p Valley.
wind 7
LE

Ze

Nighttime
mountain
wind

(b)

Figure 9-3 Valley and mountain winds during the day (A) and night (B) in a mountain valley.
(Modified after USDA, 1964.)
 AAAS
eS 0a 8 | |
MM-MMMMHHHHHMéwt$tMMM
)

mens PEGS
MMMM bbb.
ver vege-
ulence with strong

narrow valley produces lateral turbulen ce. (D) Turbulence occurs when laminar flows from two valleys meet at a va
-curs when air moves over er surfaces
s of varying temperature. (Modified after USDA, 1964.)
 Effects of Wind on Vegetation 241

Dissemination of Reproductive Propagules they flap in the wind, then air is pumped in and out of
Most plants depend on wind at least in part for the dis- the leaf air spaces through the stomata, and this accel-
erates the exchange of CO, and O). Wind prevents or
semination of reproductive structures. Dissemination of
reduces nocturnal accumulation of respiratory CO,
pollen by wind (anemophily) is common among the more
within and below the vegetation canopy, thereby re-
primitive taxonomic groups (e.g., the conifers), and most
ducing the rate of early-morning photosynthesis.
plants of cool and cold climates are wind-pollinated.
Loss of water from plant surfaces is markedly af-
Such plants require the production of much larger quan-
fected by wind. Removal or reduction in the thickness
tities of pollen than plants that are animal-pollinated.
of the boundary layer of humid air around leaves by
The extent of pollen production is reflected in the fre-
wind accelerates the diffusion of water vapor out
quently observed coverage of soil and water by a yellow
through the stomata. This can increase evaporation
film of pollen in the spring in areas of coniferous forest.
even when the vapor pressure deficit is zero because of
Evidently, the advantages of widespread distribution of
the important effect of convection on evaporation
pollen by wind justify the apparently profligate expendi-
(Daubenmire, 1974). However, water loss does not in-
ture of energy and nutrients on pollen production.
crease linearly with increasing wind speed; it increases
Wind dissemination of reproductive propagules
with the square root of wind velocity. Initially, there is a
(anemochory) is even more widespread than anemophi-
large increase in loss of water as wind speed increases
ly. It usually involves seeds and spores but can also be
from zero, but there is little difference in the rate of loss
involved in asexual reproduction, as in the case of the
between medium and strong winds. Above wind speeds
distribution of living twigs of poplar, which are physi-
of 6 km h", evaporation is mainly determined by the
ologically abscised (cladoptosis), complete with func-
vapor pressure deficit of the air. Wind also accelerates
tional green leaves (Galloway and Worrall, 1979), and
water loss by the pumping action, as already mentioned.
will take root if disseminated by wind and/or water to
The effect of wind on the moisture balance of
a suitable rooting medium.
plants is demonstrated by red be/t damage to conifers
Some plants produce seeds that are so small that
during the winter in the mountainous areas of west-
they can be carried considerable distances by wind.
ern North America (MacHattie, 1963) and other
Members of the orchid and heather families produce
topographically similar areas in the world. The Chi-
seeds that often weigh no more than 0.002 mg. Such
nook, or foehn, winds that occur in such areas have a
seeds, together with spores of algae and fungi, can be remarkable drying power. As they flow down over
carried to great heights in the atmosphere and trans- the mountain forests, they accelerate transpiration
ported over long distances by even moderate winds. loss, which desiccates the foliage if the soils are
Larger disseminules are often provided with long frozen or dry. The desiccation damage (browning of
hairs, which increase the frictional surface (i.e., act as a
foliage) occurs only in a horizontal midelevation belt
“parachute”) and enable the wind to transport them because cold air in the valley bottoms prevents the
over considerable distances. Seeds of plants such as warm, dry air from affecting valley bottom stands,
dandelions (Compositae), willow (Salicaceae), clema- whereas at high elevations, the winds do not have as
tis, and fireweed (Epilobium sp.) can be carried for much drying power. Winds during dry, cold, sunny
hundreds of kilometers. anticyclonic winter weather can produce similar
Even larger seeds or one-seed fruits (e.g., elm, maple, desiccation damage in other areas. Tree seedlings
ash, birch, members of the Pinaceae family) are provided that grow in forest nurseries can suffer extensive
with wings to ensure dispersal away from the parent damage during such weather unless precautions such
plant. Distance carried depends on the height of the tree, as spraying (to encase the seedlings in ice) or wind-
the strength of the wind, and the size and shape of the breaks are used.
disseminule and its wing. Douglas-fir seed in Oregon is Wind-induced desiccation damage is not restricted
regularly carried 300 m from the parent tree (Isaac, 1930) to cold winter weather, and even plants that grow on
and much farther during periods of strong winds. moist soil can be exposed to moisture stress if exposed
to drying winds. Lethal leaf water deficits can develop
Effects on Physiological Processes in many species at winds of 50 to 70 km h'! when soil
Wind has a marked effect on the exchange of gases be- moisture is not limiting (Whitehead, 1968). The result
tween plants and the atmosphere. If leaves are bent as can be the killing of needles and even the death of buds.
242 CHAPTER9 = Wind

Bending of plants by wind influences the physiol-

ogy of growth. The distribution of available growth
substances between root growth, stem diameter
growth, and stem height growth is influenced by stem
movements. Experiments in which trees were pre-
vented from swaying by guy lines resulted in the de-
velopment of smaller root systems and thinner, °C
Needle
temperature,
weaker stems in comparison with trees that swayed 1300
naturally in the wind. Trees guyed halfway up devel- Time, hr

oped normal stems above the point of guying, but the

Figure 9-S Temperature variation in a Scots pine needle
stationary stem below the point of guying failed to in- in April, with and without shelter from the wind. The shad-
crease in strength and became progressively weaker ing indicates the presence of a windshield around the needle.
relative to the unguyed stem. A significant difference The air temperature was | to 2°C, and the needle was illumi-
was observed after only 2 years of guying. Trees that nated by full sunlight. (After Christersson and Sandstedt, 1978.
have grown in a dense stand where they are protected Used by permission of the Canadian Journal of Forest Research and
from wind have slim and cylindrical stems that are the authors.)
easily broken if the stand is thinned or suddenly ex-
posed to wind. Trees in such dense stands usually
Effects on Morphology
have poorly developed root systems, so they are also Wind plays an important role in determining the phe-
subject to windthrow (Larsen, 1963). However, the notypic appearance of plants. We have already seen
reader should remember that wind does not act as a that wind affects stem and root morphology by affect-
causal determinant in this matter. The influence of ing the distribution of growth. When exposed to fre-
light competition on shoot/root ratios and factors quent drying winds, many plant species fail to obtain
that influence the distribution of photosynthate be- the size that is characteristic of the species in a shel-
tween growth in height and growth in diameter are tered location. Such plants become dwarfed because
also important. their cells have insufficient water to expand to full size
By increasing the evaporation of water from and because the moisture stress may inhibit cell divi-
leaves and by removing a layer of warm air from sion. Leaves that are exposed to wind become thicker
around the plant, wind acts to lower plant tempera- and smaller and have a lower rate of water loss per unit
tures. This is beneficial to plants in hot environ- area than do leaves that grow in nonwindy situations.
ments because it can prevent high-temperature The cooling effect of wind also contributes to dwarfing
damage, but in cold environments, the cooling effect by reducing growth rates, and the death of buds and fo-
can be detrimental. Metabolic rates, which are al- liage by desiccation also acts to prevent an increase in
ready low because of suboptimal temperatures, may stature of the plants. For many plants, a 27 to 36 km
be reduced to the point at which plants are unable to h continuous wind results in some reduction in leaf
complete necessary physiological processes. It has size and internode length. Continuous wind of more
been reported that wind in arctic and alpine envi- than 70 km h™ results in great growth reduction and
ronments can reduce leaf temperatures by 7°C (Wil- ultimately in death (Whitehead, 1968). Plant dwarfing
son, 1959), and a study of Scots pine needle by wind is a common phenomenon in alpine environ-
temperatures in central Sweden showed that even a ments, but once again it must be stressed that wind is
gentle wind can have an important cooling effect. not a casual determinant of alpine plant morphology.
On an April day with variable cloudiness, very light Later in the book, we shall examine the multiplicity of
wind, and an air temperature of —1 to 0°C, leaf tem- factors that determine plant growth at alpine tree lines.
perature was equal to air temperature in cloudy peri- The effect of wind on plant morphology some-
ods but rose to 7 to 8°C above air temperature times results from materials carried by the wind rather
during sunny periods. On a sunny day with an air than from the wind itself. In windy areas near the
temperature of 1 to 2°C and gentle wind, needle ocean, where the wind carries salt or sand, all the fo-
temperature varied from 3 to 7°C but rose to 12 to liage and buds that are exposed to the full effects of the
15°C above air temperature when the needle was wind can be killed. This can be the result of desicca-
sheltered from the wind (Figure 9--5). tion, the toxic effects of salt spray, or the physical abra-
 Effects of Wind on Vegetation 243

Figure 9-7 Wind training of Pinus albicaulis on Mt.

Lassen, California. Only buds in the lee of a rock and the
plant stem survive. This results in the tree’s developing a
procumbent stem that is, in fact, a branch, all leaders having
been killed.

ter of rocks or slight depressions. In extreme cases in

which the terminal bud is killed repeatedly, only lee-
ward branches survive, with the result that the plants
develop a horizontal stem growing in the downwind
direction of the prevailing wind. Any plant part that
grows upward out of the sheltered area is killed
(Figure 9-7).
Wind training can also occur simply as the result
of prolonged unidirectional wind pressure on develop-
ing branches. ‘Iwigs that start growing in an upwind
Figure 9-6 ¥lag-shaped crown of Douglas-fir at tree line direction are bent around to a downwind direction
caused by prevailing winds killing the buds on the wind- and eventually become permanently oriented in that
ward side of the tree. position (Lawrence, 1939).
The input of sea salts along the coastal fringe is
often responsible for a narrow coastal strip of vegeta-
sion caused by wind-driven particles: the foliage and tion that differs in both form and species composition
buds are literally sandblasted. from that inland where the spray input is less intense.
Windblown sand can remove all the bark from Cordes (1972) found that the narrow strip of Sitka
trees that are planted in sand dune areas, and wind- spruce that grows along the sandy beaches of western
blown ice crystals can kill all the foliage and buds of Vancouver Island, B.C., was correlated with a high
trees in a narrow zone just above the maximum depth input of salt spray from the nearby ocean. Western
of winter snow. Buds and foliage that grow in the lee hemlock and western redcedar, which are less tolerant
of the stem are less affected, with the result that of the salt spray, are largely excluded from this strip.
crowns become flag-shaped (Figure 9-6). Modification of stem shape is a common response
Such wind training of vegetation is responsible for of trees to wind. The bending forces in coniferous
the typical wedge shape of tree crown development tree stems during prolonged periods of unidirec-
along windswept coastlines and for the development tional wind result in the development of enlarged an-
of krummbolz tree growth. This term refers to the nual rings of compression wood on the leeward side and
shrub-like growth form of trees at altitudinal tree the reduction or even elimination of some annual
lines, where dwarfing and wind training result in rings on the windward side of the tree. The stem de-
dense, low, multistemmed growth of trees in the shel- velops an elliptical cross-section with the long axis in
244 CHAPTER9 Wind

the direction of the prevailing wind. Angiosperm tree established. “Tatter flags” are then put out wherever
species have a somewhat different response to such plantations are contemplated in very windy areas to
wind pressure. They develop tension wood on the ascertain the likelihood of success of reforestation
windward side, which serves in the same manner as efforts and the elevation beyond which planting
compression wood to resist the bending pressure of would be unsuccessful (Lines and Howell, 1963)
the wind and maintain the stem in an upright posi- (Figure 9-8).
tion. An extreme example of wind modification of Crown tattering also occurs when the branches
stem shape is a case of a Monterey cypress stem of adjacent trees batter each other during a wind-
growing on the coast near Carmel, Calif. At a height storm. Large quantities of small branches that bear
of 7 m from the ground, the stem was 188 cm in di- green foliage may be broken off. Sometimes this re-
ameter in the direction of the prevailing wind but sults in the forest floor of coniferous forests being
only 23 cm in diameter at right angles to the wind. covered by an almost continuous carpet of green fo-
The central growth ring was situated only 8 cm from liage after a winter wind storm, with significant con-
the windward side, there being 180 cm on the lee- sequences for tree growth, decomposition, and
ward side. A total of 304 annual rings were counted nutrient cycling.
to the leeward of the central growth ring, whereas Exceedingly violent winds can result in stembreak.
only 50 growth rings were counted on the windward The stem is snapped off at some height above the
side (Oosting, 1956). ground, while the lower part of the stem remains up-
Stem tensions that result from the swaying of trees right, still anchored by its roots. Except in trees with
produce another form of morphological adaptation: fungal decay in the stem, such breakage in undisturbed
buttress formation. Ring growth is accelerated in the stands is not common except during periods of very
angles between the lower stem and the larger roots, and high wind velocity (typhoons, tornadoes, or hurri-
this results in the formation of large stem but-tresses, canes) or excessively turbulent windstorms. It is most
especially at positions on the trunk that give the tree common where trees that previously were sheltered in
most support against the prevailing winds. Such but- dense stands are exposed to even moderate winds by
tresses are particularly well developed on very large clearcut harvesting of the adjacent stand or by road
trees in tropical forests that are subject to high winds, building activities through a previously undisturbed
but they can also be seen to a lesser extent on large trees stand. Stem breakage tends to be more common in
in windy locations at temperate latitudes. Stem but- dense stands, where the stems are tall and slender,
tresses result in a serious loss of timber because but- than in open-grown stands, where the stems are thick-
tressed stems are difficult to process in a sawmill. er and more tapered. It also occurs more frequently
The effects of wind on plant morphology dis- where the trees have well-developed root systems and
cussed above are generally the result of moderate are well anchored. In most other situations, such dam-
wind speeds sustained over long periods. More dra- aging winds result in windthrow rather than wind-
matic effects can occur when the vegetation is sub- break: the roots are pulled out of the ground and the
ject to high winds or very turbulent winds over intact trees are toppled over (Figure 9-9).
shorter periods. The whipping of branches in the Windthrow is more common in shallow- than in
wind can result in wind tatter. Just as the free end of a deep-rooted species; more common in shallow and/or
flag tatters in high winds, so the foliage at the end of wet soils than in deep and/or dry soils; and more com-
a branch becomes tattered during periods of high mon where soil, stand, or pathological conditions have
and turbulent air movement. This phenomenon has resulted in small, shallow, weak or asymmetrical root
been put to practical use in afforestation of systems. Thus, trees that grow on thin soil over
windswept areas in Great Britain. Small nylon flags bedrock, saturated soil, compacted soil, or soil with
are sometimes placed along a gradient of increasing shallow pans and trees in which deformation to roots
wind (e.g., with increasing elevation on exposed has occurred during nursery growth and/or planting
coastal mountains), and the rate at which the free tend to have a higher-than-normal susceptibility to
edge tatters over | year is measured. This rate is then windthrow. Windthrow is also a common result of
compared with the survival and growth of trees damage to root systems by pathogenic organisms (root
planted along the same transect, and a threshold tat- rot fungi) or some mechanical agency. Turbulent wind
ter rate beyond which tree growth is unsatisfactory is tends to produce windthrow more than laminar flow
 Effects of Wind on Vegetation 245

22.9

Or ~~)

7.6

height
annual
Mean
pine
lodgepole
of
growth
(cm)

Sal 6.3 9.4 2S 15.6

Rate of flag tatter (cm day!)

880 m
732m
634 m
442 m
408 m
(cm?
day!)
Tatter
rate 381m
VipAG Vine Ac S tOnINGDivie EIMICAS Mie Act oe OnNa Die aE IV
Date (month)

(b)

Figure 9-8 ‘The use of “tatter flags” to gauge the effect of wind on planted tree seedlings. (A)
Relationship between the rate of tatter of the free edge of a cloth flag and the mean annual height
growth of planted lodgepole pine trees in Scotland and northern England (after Lines and Howell,
1963) (B) Rate of tatter of standard flags at various elevations in the western Cairngorm Mountains of
Scotland (after Pears, 1972). Areas above 400 m elevation are questionable sites for initial plantation
establishment. The dotted line indicates suggested critical level of tatter for adequate seedling
growth. (Lines and Howell, 1963. Used by permission ofthe Forestry Commission, U.K., and the authors.)

because the rocking motion produced by turbulence aged trees being associated with shallow soil, logging
reduces the friction between the roots and the soil damage, dwarf mistletoe, and stem deformities. Wind
more effectively than does unidirectional wind pres- damage accounted for 60% of all tree damage and
sure from laminar wind flow. A study of wind damage death, compared with 30% for insects and 6% for
in mixed white fir—red fir stands adjacent to clearcuts snow (Gordon, 1973).
in California implicated bole rot (30%), fire scars Windthrow played an important role in the devel-
(19%), and root rot (18%) as the main causes of wind opment of the classical methods of European and
damage to mature trees, the remaining 33% of dam- tropical silviculture (Godwin, 1968). Bad experiences
246 CHAPTER9 Wind

Figure 9-9 Wind damage to an Engelmann spruce-lodgepole pine-Douglas-fir stand in cen-

tral British Columbia. The majority of the trees were blown over, but trees with particularly strong
roots and/or weak stems experienced stembreak. (Photo courtesy ofMinistry of Forests, British Columbia.)

with clearcutting in windy regions led to the develop- ing increase in spring soil moisture may enable trees
ment of a wide variety of silvicultural systems and pat- to invade. Adjacent areas not associated with winter
terns of forest harvesting, each one tailored to a snowdrifts may be too dry for tree establishment and
particular combination of topography, soils, tree survival during the spring and summer. Once trees
species, and wind. have established, perhaps in a moister year, they facili-
tate further tree invasion by increasing storage of win-
ter moisture.
Influence Through the Distribution of
Precipitation and Snow
Wind affects the availability of moisture to plants by 9.4 Effects of Wind on Animals
transporting moist air to a locality and by removing it
again. Atmospheric circulation systems determine the Wind affects animals in a variety of ways. Although
precipitation of a region, and wind greatly affects the wind effects have generally not received as much at-
loss of that moisture by evapotranspiration. Wind re- tention from animal ecologists as have the effects of
distributes winter snow, and trees can trap snow to other physical factors, such as temperature and mois-
form snowdrifts. This can have important conse- ture, the effects of wind are nevertheless of consider-
quences for vegetation. Soil beneath snowdrifts in able importance.
grasslands receives more water in the spring than soil Wind influences the behavior of animals. Many
in areas that lack snow accumulations, and the result- species of birds and insects are restricted to sheltered
 Effects of Wind on Animals — 247

microenvironments during high winds, and prolonged introduction of fish into landlocked ponds or lake sys-
periods of wind may seriously interfere with feeding tems not connected with other freshwater areas in
activities with potentially adverse consequences for which the species occur.
the animals’ energy balance. Because of this, small an- There are many instances of insects being trans-
imals generally live in microhabitats that experience ported over long distances by wind. Species such as the
very much lower winds than occur in the general envi- spruce budworm (Choristoneura fumiferana) are dis-
ronment. Many leaf-eating insects will drop to the persed over many hundreds of kilometers by convec-
ground when the vegetation on which they are feed- tive windstorms as a result of the behavior of the adult
ing is violently shaken by wind and will remain there moths. A sudden decrease in light intensity caused by a
until the wind drops. Butterflies and moths flatten heavy cloud or a sudden drop in air pressure ahead of a
themselves against the ground to avoid being swept convective storm stimulates the moths to rise above
away. There are even birds in the windy desert of high the trees for a brief period of flight. Ifa cold front with
Tibet that are reported to build a rampart of pebbles its associated convective updrafts passes over the area
on the windward side of their nests (Allee et al., 1967). at this time, then the moths are carried aloft and trans-
Homeotherms are affected by the “chill factor” of ported for long distances and then deposited as the
wind, which increases both the convective loss of heat storm disperses. Wind dispersal can initiate outbreaks
and cooling by evaporation of water. Greater insulation of destructive pests in areas far removed from infesta-
or restriction of activities to sheltered locations is neces- tion centers and can hasten the decline of the infesta-
sary during periods of high winds and low temperatures. tion in the source area by removing large numbers of
Poikilotherms are also affected by wind chill. Their in- the female moths (Henson, 1962).
ternal body temperatures are largely determined by be- The location of areas plagued by locusts in East
havior, and therefore wind-induced changes in their Africa has been shown to be influenced by winds,
thermal load may require a change in behavior to main- which tend to concentrate the swarms into fairly well-
tain body temperatures within a particular range. defined areas. Locust problems are particularly severe
Rapidly moving air has a substantial kinetic energy in the intertropical convergence zone (Figure 9-10),
and can transport surprisingly large objects. Even where winds seem to keep the swarms within a fairly
gentle winds play an important role in transporting well-defined east-to-west band. Movement of the
smaller animals, but windstorms of great violence can swarms to the north or south of this band is correlated
transport sizable organisms. The fauna of the Greater
Antilles Islands in the West Indies is evolutionarily re-
lated to the fauna of Central America, yet no convinc-
ing evidence exists that there was ever a land bridge \ Persiah Gulf
connecting the island to the mainland. It has been sug- Iran
Pakistan
gested that the similarity of animals as large as rodents
and snakes between the two areas is the result of wind
transport of the species between the two areas (Dar-
lington, 1938). (This does not discount the important
contribution of water transport). The exceptional lift-
ing and carrying power of wind is illustrated by the
“rain of fishes” that occurred in Marksville, Louisiana,
in 1947 (Bajkov, 1949). On the morning of October
—— Intertropical front at the surface ¢ Reports of locust swarms
23, fish ranging between 5 and 23 cm in length fell
from the sky in a density of up to 1 m~. The species -—-— Limit of upper northerlies ~~ww~ Major air movements

involved were the same as those in nearby ponds, and Figure 9-10 The intertropical convergence zone and the
several small tornadoes had been noticed in the area reported distribution of locust swarms during the period
that day. The incident was explained in terms of local July 12-31, 1950, in eastern Africa, the Middle East, and
high-velocity winds that had sucked up the surface of India. The convergence of major air circulation patterns serves
the water of the lakes together with the aquatic organ- to concentrate locust swarms within a well-defined east-west
isms contained therein. Fish transport by local violent band. (After Rainey, 1973. Used by permission of the Royal Meteo-
air currents may sometimes have been involved in the rological Society and the author.)
248 CHAPTER9 Wind

with north or south shifts in the intertropical front. be subject periodically to the deposition of as much as
Similarly, infestations of the African army worm several meters of fine-textured volcanic ash. Close to
(Spodoptera exempta), which attacks cereals and grasses the eruption, where the ash is hot, the trees are killed,
in eastern and southern Africa, can be predicted using but farther away, where the ash is cool, the trees are
synoptic meteorological data (Rainey, 1973). able to survive the deposition of more than 1 m of ash
by the development of aerial roots.
Wind swaying of very large trees in steep mountain
9.5 Effects of Wind on Ecosystems topography is sometimes responsible for landslides.
The soil is loosened by the heaving of the root systems
Wind can alter ecosystems in a variety of ways. Wind as the trees sway, and the slope hydrology may be al-
transport of pathogens and insects can lead to the de- tered by the pumping action of the roots. The net effect
struction of all or some of the vegetation. The loss of is to increase the soil shear stress and reduce its shear
vegetative cover alters the habitat for other animals in strength to the point at which slope failure occurs and a
terms of both shelter and food. If the loss is extensive, landslide is triggered. This effect of wind is thought to
then it may expose the soil to sunlight, wind, and rain, contribute to the occurrence of many of the landslides
with a variety of consequent effects. in old-growth forests in the coastal mountains of
Wind erosion of soil can significantly alter an British Columbia during wet and windy winter weather.
ecosystem. The surface layers of the soil are generally Air transportation of pollution can have a dramatic
the most important in terms of plant production, and effect on ecosystem structure and function. Figure
their loss through wind erosion can lower soil fertility. 9-11 shows an area downwind of a smelter at Sudbury,
This is particularly serious in dryland arable farming Ontario, where all the vegetation has been eliminated
where soil cultivation techniques render the surface by sulfur fumes. Ponderosa pines in the mountains
soil particularly prone to removal by wind. The dust to the east of the city of Los Angeles are adversely af-
storms that are an unfortunate characteristic of hot, fected by smog.
dry agricultural areas represent both a nuisance to Wind also plays an important role in soil develop-
human comfort and a degradation of ecosystem pro- ment and ecological succession (Chapter 17). When
ductivity in the eroded areas. large trees blow over, they create canopy gaps that have
Deposition of wind-transported soils can have a important implications for forest structural diversity and
variety of effects. Small annual additions of fine min- succession, but they also cause considerable upheaval
eral and organic particles act as a fertilizer and gener- and mixing of soil. The microtopography (hummocks
ally improve the productivity of a site, but where the and mounds) created by windthrow can create drier,
deposition rate is too great, the vegetation may be warmer microsites for seedling establishment on cold or
killed. Most plants cannot tolerate the sudden addi- wet sites and can reduce competition between tree
tion of many decimeters of soil parent material to the seedlings and established herbs and shrubs. In cool, wet
soil surface. Small plants are totally buried, and larger climates, slow decomposition may result in the accumu-
plants may die because their deeply buried roots are lation of deep, acidic forest floors dominated by decay-
unable to obtain the necessary oxygen. ing wood. Soil acidification and the accumulation of
Certain plant species have become adapted to and organic acids that are an end product of organic matter
can survive burial by wind deposits. Marram grass decomposition under these conditions can result in low-
(Amophila arenaria), which grows on coastal sand dunes, ered soil fertility and impeded soil drainage and aeration.
is able to survive burial to a depth of approximately | m. In the long-term absence of soil disturbance by landslide
It responds by sending up a rhizome to within a few or windthrow, the soils may degrade to the point at
centimeters of the new surface and growing new aerial which they no longer support closed forest. Windthrow-
shoots from the elevated rhizome. The buried portions related soil disturbance in such forests plays an impor-
of the plant then die. When a sand dune is excavated, tant role in maintaining soil conditions that are suitable
the remains of marram grass can be found at many for the productive growth of closed forest (see discussion
depths within the dune, tracing the history of growth of in Chapter 17, and Bormann, 1989; Dunn et al., 1983;
the dune by deposition during storms. Another exam- Kramer, 2001; Peterson et al., 1990; Prescott and Weet-
ple of adaptation to burial is exhibited by the ohia tree man, 1994; Schaetzl et al., 1989; Ugolini, 1966).
(Metrosideros collina), which grows in areas of recent vol- In high latitude coastal forests, such as portions of
canic activity on the islands of Hawaii. Such trees may western North America and of Europe, periodic ex-
 Effects of Vegetation on Wind 249

Figure 9-11 Landscape denuded of vegetation by sulfur fumes in Sudbury, Ontario. (Photo
courtesy of Dr. T: Hutchinson.)

treme wind events cause widespread forest disturbance Stand-replacing wind disturbance prevents infection
(Prescott and Weetman 1994; Borman et al. 1995; by the dwarf mistletoe (Arceuthobium tsugense) and
Kramer, 2001). These events alter soils, succession, leads to stands that are susceptible to subsequent
hydrology, nutrient cycling, stream chemistry, and windthrow. Long periods without windthrow lead to
stand characteristics. Studies in southeastern Alaska mistletoe-infected forests that respond differently to
found that stands that had experienced more intense extreme wind events—stem break leads to gap forma-
soil mixing caused by windthrow had less strongly hu- tion and greatly reduces the probability of the whole
mified organic matter and more particulate, partly de- stand blowing down. Thus, wind interacts with this
composed OM than less disturbed stands. Streamflow plant parasite to determine the effects of wind on the
following rain storms peaked 4 to 12 hours later and forest (Kimmins, 2003, Weber et al. 2003).
contained more base cations in the more disturbed
than in the less disturbed watersheds. The disturbance
increased water infiltration and water storage in the 9.6 Effects of Vegetation on Wind
soil profile, and thereby smoothed the stream response
to rain events. Clearly wind effects in these forests play Vegetation exerts a marked effect on winds close to the
a major role in determining their character. ground. The roughness of the vegetation provides a
Frequent wind disturbance (every 80 to 200 years) friction surface that significantly reduces the overall
is associated with high levels of forest productivity in wind velocity but increases wind turbulence. Forests are
coastal western hemlock forests in British Columbia. particularly effective at modifying winds because al-
250 CHAPTER9 = Wind

Tuble 9-1 Roughness Length (Z)) Values for Various The effects of vegetation on wind have been recog-
Surfaces nized and used to advantage for centuries. Shelterbelts
Type of Surface Roughness Length, m
of trees are planted by farmers to reduce wind velocity
over fields and to provide livestock with shelter from
Water (calm, open sea) O—=10:0 10° the wind (e.g., Read, 1964). In many parts of the
Snow 0.5-10.0 x 10~ world, agriculture would be much less productive
Sand desert 0.3 x 10°
without such shelter, and shelterbelts would probably
be of great benefit in many areas that currently do not
Soils 0.1-1.0 x 10°
have them. Trees are also planted to reduce heat loss
Grass (0.02-1.0 m high) 0.03-1.0 x 10°! from buildings and to reduce wind erosion of soil.
Agricultural crops 0.04—0.20 They are used to capture blowing sand and have been
Orchards 0.05-1.0 used extensively to stabilize areas of shifting sand
Forests 1.0-6.0 dunes. They are also important in causing the deposi-
tion and accumulation of blowing snow. The last is of
Data from Oke, 1978. great importance to agriculture in areas where the soil
moisture required for spring and summer crops comes
though they are porous, allowing some wind movement more from melting snowpack than from growing sea-
through them, the very large surface area of leaves, son precipitation.
branches, and bark provides a very large frictional sur- The effect of shelterbelts on wind depends very
face, which effectively reduces wind speeds. ‘Table 9-1 much on the permeability of the belt. Impermeable
lists values for Zp, the roughness length (an index of the belts result in maximum downwind reduction of veloc-
frictional resistance of surfaces to air movement), for a ity, but speeds increase rapidly farther downwind and
variety of surfaces. there is a lot of turbulence. Permeable belts are associ-
The profile of wind in a forest depends on the tree ated with less reduction in velocity but also with less
species, the density of the stand, and the stand structure. turbulence and a more persistent downwind effect
Even-aged stands without any understory may have low (Figure 9-12). The choice of windbreak will depend on
wind speeds in the crown but appreciable winds between the local wind conditions, the type of crop, and the
the crown and the ground, whereas multilayered forest type and degree of wind modification desired by the
communities have reduced wind speeds at all heights land manager. Deciduous hardwood belts are not used
below the upper crown. The reduction of wind speed is to provide livestock with shelter from winter winds be-
much greater for high winds than for low winds. A 6-km cause of inadequate reduction of wind speed in the lee
h'' wind in the open may be reduced by only 2 km h” at of the belt, and impermeable conifer belts will not be
the same height in the forest; a 50-km h™ wind in the used to shelter grain crops from summer winds because
open can be reduced to as little as 7 km h™ in the forest of the problems of downwind turbulence and the short
(Anonymous, 1964). Table 9-2 lists examples of the re- distance over which wind-speed reduction is achieved.
duction of wind by various types of forest stand. The flow of wind over unbroken forest in level ter-
rain is approximately laminar (except where small-
scale turbulence occurs within the tree canopy), but it
Table 9-2 Effect of Forests on Wind Velocities
may become turbulent as the wind passes into har-
Wind Speedin Reduction vested areas (Figure 9-13A). Turbulence often devel-
Open Forest, in Forest, ops along the windward edge of a clearcut, and this
km h"' % can result in windthrow or other damage. However,
turbulence is not usually the major cause of tree dam-
Western white pine, Idaho 2.4 82
age in clearcuts. That place is reserved for the acceler-
Jack pine, Neb. 138A 72
ation of wind as it leaves the clearcut.
Aspen, Colo. ne 68 As wind passes over the leading edge of a logged
Douglas-fir, Colo. 3.9 67 area, the wind drops from canopy height to move over
Ponderosa pine, Ariz. 9.0 49 the ground. ‘To leave the clearcut again, the wind must
regain its original height by rising up over the uncut
Data from Kittredge, 1948. stand at the downwind end of the opening. This involves
 Significance of Wind for Forestry 251

Wind direction away from clearcuts. Trees in a fully stocked stand

120 provide each other with mutual shelter, and only the
uppermost part of the crowns of the largest trees expe-
80 “! i ere
ae
eh rience the full force of the wind. Additional stability is
provided by the intertwining and grafting of root sys-
tems; each tree gains some stability from the root sys-
a ye Impermeable
40
Se -— Moderately dens tems of the adjacent trees. Trees at the edge of a
the
in
that
(%)
open t-- —-—-— Permeable clearcut are exposed to the full strength of the wind,
Wind
relative
velocity
to secs Very permeable
and as the roots of the adjacent cut stumps decay and
10 20 30 lose strength, these edge trees lose some of their root-
Distance from shelterbelt measured in ing stability. Windthrow is most common where
units of shelterbelt height newly exposed trees experience wind acceleration and
(a) turbulence, although sometimes they can be blown
Impermeable belt over even in moderate wind speeds without turbu-
lence. For a discussion of windthrow and forestry, see
Palmer (1968).

SOS SSS

Permeable belt 9.7 Significance of Wind for Forestry

Wind has played an important role in the develop-
ment of forestry. Windthrow was one of the factors
that wrecked early attempts to provide secure future
a Maximum _ Maximum Slight
velocity velocity 2] velocity supplies of timber in Europe, and it was one of the
reduction reduction reduction main stimulants behind the development of the vari-
(b) ous patterns of forest harvesting used in mountainous
Figure 9-12 Effects of shelterbelts on wind velocities. (A)
European forests.
Extent of wind reduction by shelterbelts of different permeabil- The effect of wind on the water balance of plants
ity. (B) Turbulence and generalized wind velocity profiles over can determine the success or failure of planting. The
permeable and impermeable shelterbelts (hypothetical). (After moisture stress caused by wind can be fatal to a
Nageli, 1945. Used by permission ofL’Institut Féderal de Recherches young seedling even on a moist site because of the re-
Forestiers, Zurich.) sistance to water uptake that is caused by the lack of
fine roots, root hairs, and/or mycorrhizal associa-
tions. Reducing the size of clearcuts and planting in
some acceleration of the wind. The degree of accelera- the lee of stumps, microtopographic features, or log-
tion depends on the shape of the clearcut. Where the cut ging debris, all of which reduce the exposure of
is wedge-shaped with the narrow end pointing upwind, seedlings to wind, can help to reduce desiccation-
there may be little or no acceleration; the wind enters induced seedling mortality. Wind also affects natural
the clearcut along a narrow front and leaves on a broad regeneration by determining the distribution of
front. Where the broad end points upwind, a large vol- seeds and can damage young plantation trees by loos-
ume of wind enters the area along the broad front but ening their root systems in the soil by frequent vio-
has only a narrow area through which it must pass on its lent rocking (Edwards et al., 1963).
exit. This results in an acceleration of the air mass in the Windthrow continues to be a problem in many
same way that a broad, slow-moving river speeds up as it parts of the world. Planted trees sometimes develop
passes through a narrow gorge. The increased velocity root forms that differ from those of “natural”
increases the kinetic energy of the wind, and some of seedlings. This may result from root damage during
this energy is transferred to the trees as the wind leaves the nursery phase or at the time of planting. The ab-
the area. This can result in windthrow (Figure 9-13B). normal root systems tend to reduce the wind stability
Windthrow is far more common in stands that are of the trees, even though the tops of the trees may
adjacent to clearcut areas than in stands that are well grow normally. Consequently, the problem is not
 252 CHAPTER9 Wind

WV
PAA sei we
MWA NANA MsCu 1 eA Resumption
/ } / i
: AMINE
vy {|
4 of
Laminar NAA AA, VAN 5d gage
flow over > rit )’-- Slight N se
ANA laminar
forest y KON
NG Pe ; AN): flow
got os wind editers:.; - 4 «as clearout . = = 2". e yu at original
clearcut.” = fo si dais Ute Eme 4 {N speed
Psi 15

ent
/

JAN
cel Tr
i\

y'

ANN . © ~ Acceleration: of.

Lamina ie * §wind'as clearcut
flow over =a in ‘narrows

ape
forest A is Te 3 ay tae

K ae .
ts
hy Ge ad crt

Windthrow in area
of high-speed, turbulent
wind

(b)

Figure 9-13 The influence of clearcut shape and orientation on the incidence of windthrow.
(A) Cone-shaped opening with narrow end toward prevailing high winds. This minimizes the risk of
wind damage. (B) Cone-shaped opening with broad end toward prevailing high winds. This pro-
duces the greatest risk of wind damage. The wind damage continues downwind until speed and tur-
bulence decrease or until wind-firm trees are encountered.

noticed until the trees blow over. Site preparation by seedlings in forest nurseries resulted in root deforma-
plowing can result in an abnormal orientation of the tion and/or abnormal development of the root system
root system, with concomitant reductions in stability. of some species. Modifications to the design of
Several of the earlier “container” methods of growing seedling containers have reduced root deformation
 Take-Home Message 253

problems for most of these species but not for all of leaves. Although a certain amount of wind can have a
them. Bare root stock' may remain preferable to con- beneficial effect, moderate wind speeds generally exert
tainer stock’ for the sensitive species. an adverse influence on tree growth. High winds can
The problems of windthrow should play an over- produce mechanical damage, and sustained, directional
riding role in determining the shape, size, and layout winds can reduce the quality of stemwood for human
of clearcuts in areas where wind is an important factor. uses thereof by causing the formation of tension or
Small patches of trees that are left to protect streams compression wood, ring shake, stem buttresses, and
marked stem taper. Mechanical damage can become
and lakes or to provide wildlife habitat often blow particularly severe if the wind carries ice or mineral
down, and foresters frequently resist leaving such particles.
patches because of the risk of windthrow. Many animals are affected by wind. Small animals
Uneven-aged forests with multiple canopy layers can may be unable to conduct their normal activities when
be more resistant to windthrow than dense, even-aged winds are strong because of the risk of damage or dis-
forests. The larger trees in the former tend to be wind- placement. High winds may aggravate the loss of water
firm because they have been exposed to wind stress for and heat, which may lower body temperatures below the
much of their life, and uneven-age forestry with partial effective or even the survival range. Conversely, the cool-
harvesting can sometimes reduce wind problems com- ing effect of wind may permit an animal to survive in a
very hot environment.
pared with even-age systems. However, if even-age
‘Transportation of materials and objects undoubtedly is
stands are thinned regularly, then their wind stability can one of the most significant ecological roles of wind. It re-
be increased, permitting shelterwood harvesting even in moves human atmospheric pollution from its source area,
windy areas. Shelterwood systems often fail because of transferring any resulting problems downwind (e.g., acid
windthrow when they are attempted in dense, even-age, rain). It transports small animals, some of which are to-
previously unmanaged forests, which is one reason that tally dependent on air currents for their dispersal to new
such forests are often harvested by clearcutting. areas. It carries seeds and spores over long distances and
Wind damage to tree crowns and wind-induced therefore constitutes a major force in determining the pat-
water stress can reduce seedling survival and tree terns of vegetation development in an area. The wind
growth. Wind pressure on trees can result in asym- transportation of pollen is essential in the life history of
metrical stems and lop-sided tree canopies, both of many plant species. Without wind, many ecosystems
would be much less productive because wind provides a
which can reduce the quality of the wood that can be
continuing input of nutrient-containing dust particles and
produced from such trees, and thus reduce their eco- precipitation.
nomic value. Excessive bending of stems by wind pres- Wind has always been a major factor in the manage-
sure can cause internal stem fractures (ring shake), ment of forests. Periodic destruction of forests by wind,
which lower the economic value of the trees when the effect of wind on seed distribution (and therefore on
they are harvested. natural regeneration), the effect of wind on the survival
of planted seedlings, and wind damage to stands that are
adjacent to clearcuts were important factors in the devel-
SUMMARY opment of a variety of silvicultural systems and forest
The atmosphere is the medium that surrounds terrestrial management strategies in Europe over the past three
animals and the aboveground parts of terrestrial plants. centuries. The ecological effects of wind remain
Movements in that medium, known as wind, are likely to an important consideration in the management of both
have significant effects on these organisms. plant and animal populations, and it will always be an
Wind affects the growth and morphology of plants unwise forester who fails to give the wind factor due con-
by influencing their water balance, their gaseous ex- sideration.
changes, the distribution of net growth between differ-
ent parts of the plant, and the survival of buds and
TAKE-HOME MESSAGE
‘Bare root stock: seedlings that are grown in open soil beds in the nursery, Wind plays an important ecological role in forests. It dis-
dug up when large enough, and planted in the forest with bare roots. tributes pollen and seed, pests and pathogens, water,
*Container stock: seedlings grown in the nursery in the soil that is held in snow, and heat. It creates disturbance in the ecosystem,
some type of container, and planted out when large enough either in the influencing soils and succession. Forests modify winds,
container or removed from the container with the “plug” ofsoil around and there is a change in forest—-wind relationships when
which the roots have grown. the forests are managed.
254 CHAPTER9 Wind

Historically, wind played a major role in the devel- STUDY QUESTIONS

opment of European and Scandinavian silvicultural
systems. Wind is also affecting the choices that 1. Why does wind occur?
foresters in the United States and Canada have with 2. How could you predict whether wind will be a limita-
respect to harvesting and silvicultural systems as they tion on how you manage a forest?
convert unmanaged forests to managed forests. The 3. What role does wind play in ecological succession?
ecological role of wind and the interactions between
trees and wind will have to be understood better if the 4. How does wind affect insect and disease problems in
limitation on choice of silvicultural system imposed by forests?
wind is to be reduced. 5. How could you reduce the damage done to forests by
wind?
 Water
The Material That Makes Life Possible

10.1 Introduction temperature fluctuations, a fact to which a comparison

of coastal and continental climates will attest. A high
Second only to energy, liquid water is the most impor- heat of vaporization (heat absorbed during the change
tant prerequisite for life as we know it. It is commonly from liquid to vapor phase; 2.24 MJ are required to
believed that organisms evolved after the synthesis and convert | kg of liquid water to water vapor) inhibits
subsequent combination of simple organic molecules evaporation and acts to maintain water in the liquid
in shallow bodies of liquid water, and all life processes form. It also means that water is an extremely effective
involve chemical reactions in aqueous solutions. coolant. Large quantities of heat are absorbed by the
Water vapor in the air protects organisms from the evaporation of modest quantities of water. Life in ex-
harmful effects of ultraviolet radiation, and by trap- posed or hot environments would be impossible for
ping long-wavelength radiation it contributes to the many plants and animals without the cooling that is
maintenance of atmospheric temperatures in the provided by evaporation. Of equal ecological impor-
range over which water can exist in the liquid form: tance is the relatively high heat offusion (heat given off
the greenhouse effect. The oceans and large lakes help during the change from liquid to solid phase; 0.34 MJ
to regulate atmospheric concentrations of CO, which must be removed from 1| kg of liquid water at 0°C to
influences both atmospheric temperature and global freeze it), which acts to resist the freezing of water and
organic production. Water thus is the very essence of reduces the rate at which the temperature of aquatic
life, and it is almost impossible to discuss the ecology ecosystems drops in the winter. Finally, water reaches
of any organism without reference to its adaptations to maximum density at 4°C, not 0°C, a fact that prevents
the moisture conditions of its environment. lakes and rivers from freezing solid. Ice has lower den-
The remarkable fact of life is to a great extent a re- sity than water and therefore floats; the heavier 4°C
flection of the remarkable properties of water. It is one water sinks to the bottom where it is protected from
of the few inorganic substances that is liquid at sum- further cooling.
mer air temperatures, and without a liquid solvent, the In addition to these temperature attributes, water
chemical processes of life could not occur. It has a is an excellent solvent for ionic substances and has a
higher specific heat (the amount of heat required to high density and viscosity. The high density provides
raise the temperature of | g of a substance 1°C) than support to aquatic organisms, but it means that water
any other common substance, which gives it the abi- is crushingly heavy at depth and the high viscosity im-
lity to absorb or give out large quantities of energy pedes movement. Water also has a poor ability to dis-
with relatively minor changes in temperature. The solve oxygen, a fact that may limit the activity of
abundance of liquid water at the surface of the earth aquatic organisms to shallow areas or the surface of
and water vapor in the atmosphere greatly reduces deep water bodies.

henN wal
256 CHAPTER10 Water

All these physical characteristics of water can be 10.2 The Water Cycle
attributed to the nature of the water molecule. Al-
though it is electrically neutral, with two H* atoms Like nutrient cycles in general, the water cycle is dri-
and one O? atom, the two hydrogens are arranged ven by inputs of solar energy. Vast quantities of radiant
asymmetrically, so one side of the molecule is some- energy are absorbed in the process of evaporating
what positive and the other side is somewhat nega- water from the warm areas of the world’s oceans. The
tive. This type of molecule is called a dipole. The energy is transferred to the atmosphere as the water
negative side of one molecule attracts the positive vapor condenses, thereby driving our climate and cre-
side of another, resulting in hydrogen bonds that hold ating our weather. The warm, moist air creates clouds
the molecules together. This hydrogen bonding ac- as it rises, and the winds formed by the resulting
counts for the high values of such things as heat of processes of atmospheric stirring move the clouds
vaporization and specific heat. Heat energy is ab- over the land, where some of the moisture falls as pre-
sorbed in overcoming hydrogen bonds rather than cipitation. Some of this is re-evaporated directly back
raising the temperature. It also accounts for the sol- to the atmosphere, and some is subsequently tran-
vent properties of water. The attraction of the differ- spired by plants. The rest enters water courses and re-
ent sides of the water molecule to oppositely charged turns to lakes and eventually to oceans, from which it
ions of a solute is so great that it overcomes the is once again evaporated. We shall discuss each stage
attraction that holds the two ions of the solute to- of the water cycle (Figure 10-1) in turn.
gether. The ions separate and become associated
with the water molecule; the solute “goes into solu-
tion.” For example, when soluble ammonium nitrate Precipitation and Other Inputs of
fertilizer is placed in water, the force holding the Moisture into Ecosystems
NH,* and the NO, ions together is less than the at-
traction of these ions to different sides of the water Precipitation occurs when air becomes saturated with
molecule. Consequently, the fertilizer molecule dis- water vapor, and condensation occurs to form clouds.
sociates, and the two ions are held by hydrogen Under certain conditions, the cloud droplets coalesce
bonds to the water molecules. and grow in size until they are too large to remain sus-
Water is of enormous significance to all plants pended in the air, and they fall as rain. The quantity of
and animals and no less so to humans. It acts as the water vapor that a volume of air can contain depends
medium in which all life processes (i.e., chemical on its temperature: the higher the temperature, the
processes) occur. It acts as a transport system permit- greater the quantity. As air temperature drops, the
ting nutrient uptake from the soil, moving metabo- amount of vapor that it can hold also drops, until at a
lites around, supplying oxygen to oxygen-demanding certain temperature (the dew point) it becomes satu-
tissues, removing waste products and regulating the rated and condensation occurs. The conditions for
temperature of organisms, and supporting and pro- precipitation are created when air masses that contain
tecting aquatic organisms. For humans, water has the water vapor cool to below their dew point. Such cool-
added significance of being a source of power (hy- ing can occur in several ways, and consequently there
droelectricity) and a means of disposing of undesir- are several types of precipitation.
able waste heat and materials. Lack of water may well
become more important than lack of energy as a 1. Convectional rainfall. This results when differential
major factor limiting our population growth: water solar heating of the earth’s surface and lower air
to grow food, water to flush away our wastes, and layers results in air masses rising by convection. As
clean water to drink. the air rises, it expands because of lower atmos-
In this chapter, we examine the water cycle and the pheric pressure. As it expands, it cools, and con-
availability of water to plants and animals. An exami- vectional clouds are formed. The resulting rain is
nation of adaptations to the water factor is followed by usually local in distribution and of short duration,
a consideration of how water affects the distribution of but it can be of high intensity. This type of precip-
plants and animals. The chapter concludes with an in- itation is most common in the summer, when there
troduction to the human effects on the water cycle and is moist, warm air near the ground and sufficient
the significance of water to forestry. solar energy to cause convectional stirring.
 Evaporation
from snow
Fog drip
and rime

Snowfield

Transpiration
from plant foliage
: Evaporation
Wind transport jtrom soil surfaces
over land aad

Atmospheric { Interception
humidity loss from plant
surfaces SX
Surface
\ runoff /::/

vy a Wy, -
‘Throughfall A heh
ie Pe La . Infiltration
Stem
2 eeZ ot Sb v oN J;
COND pe EOE Ve
Runoff back 1a Cas? Bead Re | if aE Ph " Slope
to water body ty” 7. + JV Percolation fA seepage
Ye — spon monee Waly er. f "4
E vaporatio
ati ye , Ww
tl Sas £ table = a H

“UY - /- Ae eres, Pantie ee) is Ee eep seepage into fractured

a bi Loos ¥Rees
et Sea
:
aT
PoSoe
aes ce Neeles EE SS
<2 bedrock
a soe PRE SON
pep seer Ba rn
: lee
Sa IEG 5 cer Ne
wes (SS
—

oun
A
==
==
Bedrock

Figure 10-1 Diagrammatic summary of the major components of the water cycle. This cycle
is driven by inputs of solar energy that generate evaporation and atmospheric stirring.

Li)wa~I
258 CHAPTER10 Water

2. Orographic precipitation. This occurs when warm (growing tips), resulting in forked stems; break the
moist air is forced upward over hills or mountain stems; or result in the uprooting of the whole tree.
ranges by the general movement of the air mass. Sometimes this is highly destructive. At other times, it
Mountains, particularly those that have mild cli- merely results in the death of diseased or suppressed
mates and are immediately downwind of large trees, leaving more growing space for those remain-
bodies of water, typically have clouds around their ing. Snow toppling of etiolated understory trees may
tops for much of the year even when the sky else- be a component of shade tolerance in some types of
where is cloud-free. Some mountains are associat- forest. Other ecological aspects of snow are discussed
ed with “standing clouds,” which never move later.
despite strong winds. They represent the cooling Under certain low-temperature conditions, rain
of air as it passes over the mountain and the subse- forms ice particles (hai/) rather than snow. Hail can reach
quent rewarming as the air drops in elevation be- several centimeters in diameter and can damage vegeta-
hind the mountain. tion by stripping foliage and branches. Animals that are
3. Frontal or cyclonic precipitation. This precipitation caught in such heavy hail can be killed or wounded.
occurs when warm moist air is forced to move up- Dew is the condensation of water vapor on snow,
ward over a layer of cold air. This generally pro- cold soil, or cold plant surfaces, and it can be an eco-
duces prolonged rain over large areas, frequently logically important addition of water in some types of
of only moderate or low intensity. Local high- ecosystems. It usually occurs on clear nights when
intensity frontal precipitation may occur at the in- such surfaces cool by radiation loss. By forming a film
terface when large masses of warm, moist air meet of moisture on foliage, overnight transpiration is re-
large masses of cold air. duced, conserving soil moisture for use the following
day. When the dew is heavy, water may run down
Moisture may be transferred from air to land in a stems or drip to the ground, where it becomes avail-
variety of ways. The most obvious of these are rain able for daytime use. This can be an important source
and snow, the ecological effects of which are deter- of soil moisture for dryland species such as bunch
mined not only by the total quantity per year but also grasses, and small, shallow-rooted desert annuals may
by their seasonal distribution and their intensity: the depend as much on dew as on rain. The maximum
rate of precipitation per unit of time (usually per hour amount of dew that can be deposited per night is ap-
or per day). Precipitation with a droplet size of less proximately 1 mm, although it is usually much less
than 0.5 mm and an intensity of approximately 0.25 than this (Slatyer, 1967b). It is generally thought that
mm hr is called drizzle; heavy rain has droplets 2.5 to dew is not of great significance to the overall water
6.4 mm in diameter, intensities of 15 to 100 mm hr“, balance of mature plants in nondesert areas, but dew
and a velocity of approximately 20 mm sec”, nearly may be important at the time of germination and in
twice that of drizzle (Hewlett and Nutter, 1969). Rain- the survival of young trees planted on dry sites. Dew
fall intensity has important ecological implications. has been shown to increase the survival of pine
Heavy rain can strip foliage from plants; beat down seedlings rooted in dry soil, and this may be important
herbaceous vegetation; alter the structure of the sur- in determining tree regeneration in dry areas (Stone,
face layers of the soil; erode soil and silt up streams; 1957, 1958; Stone and Fowels, 1955). Dew can also be
and kill, displace, or interfere with the feeding activi- a significant input of moisture to alpine snowfields.
ties of small animals. Drizzle makes a given amount of Another, less well-recognized input of moisture is
rain more available to plants, as it is delivered over a the result of the formation of rime. This is a layer of
longer period. It is also more effective in leaching ice that develops on surfaces by one of two processes:
chemicals from plant foliage than is a brief period of the impaction of fog particles on freezing surfaces or
heavy rain delivering the same quantity of water. Con- the growth of ice crystals on freezing surfaces that are
versely, heavy rain can be more effective than light in contact with warmer, moist air. When the ice melts,
rain at leaching chemicals through soils. the water can be absorbed by plants either directly or
The distribution, intensity, and characteristics of after the water has dripped down to the soil. The ac-
snow are also important. A heavy fall of wet snow can cumulation of rime can be an important moisture
trap and strand wildlife or inhibit their migration. It input in some environments. For example, the amount
can strip branches from trees; break off their leaders of water that reaches the ground beneath stunted trees
 The Water Cycle 259

at high elevations in the New England mountains was increasing height of the vegetation: the taller the veg-
found to be approximately four times that in the open etation, the greater the volume of air from which fog
as the result of the formation of rime on the vegeta- particles could be scavenged.
tion (Schlesinger and Reiners, 1974). Rime con- The atmosphere is not the only source of water for
tributed 7 to 10 cm additional precipitation over the an ecosystem. In fine-textured soils, water can reach
winter in a lodgepole pine forest in Washington vegetation by capillary rise from a water table well
(Berndt and Fowler, 1969). Rime can accumulate to a below the depth of rooting. Deep-rooted plants that
thickness of several centimeters on minor vegetation grow in dry areas that are underlain by moister miner-
in clearcuts in clear, cold, fall or spring weather. al layers may increase the supply of moisture in the
Fog drip is an input of water that results from fog surface soil by hydraulic lift. At night, when the sto-
and mist particles (0.01 to 0.1 mm in diameter) com- mata are closed and the temperature of the surface soil
ing in contact with vegetation surfaces. Fog drip is drops by radiant cooling, water moves in the roots
particularly important in forests at high elevations or from deeper, moister layers to roots in the dry surface
in humid coastal or foggy areas. For example, fog drip soil, where it is lost to and condenses in the dry soil.
contributed water equivalent to 150 mm of precipi- This moisture is then available to shallow-rooted
tation during a 40-day summer rainless period at a lo- plants during the subsequent day (Caldwell, 1990).
cation near San Francisco (Oberlander, 1956). Hydraulic lift is not restricted to arid areas. It can also
Twenty-five percent of the water that reached the soil occur in mesic environments during periods of water
beneath a ridge-top conifer forest approximately 4 km deficits (Dawson, 1993). Lower-slope ecosystems re-
inland from the Oregon coast was found to be from ceive seepage water flowing downslope through the
fog drip (Isaac, 1946). An 18-week study in a Sitka soil over an impermeable layer. Springs can augment
spruce/western hemlock stand in Oregon recorded 29 the water supply of lower-slope areas even where slope
cm of fog drip while precipitation was recorded as 64 seepage does not occur. Snow melting in midsummer
cm. It has been suggested that water input below for- in subalpine forests may contribute significant inputs
est cover in foggy weather generally exceeds that in of moisture to downslope forests when there is little
the open by 30 to 50%: peak values are generally growing-season precipitation.
found in ridge-top locations. Table 10-1 shows some From this brief review, it should be clear that al-
fog drip data obtained under different vegetation types though rain and snow are the major sources of water
on Mount Wilson, California. Fog drip increased with input for many ecosystems, there are several other
ecologically important sources.

Table 10-1 Fog Drip Under Vegetation at 1,783 m

Elevation on Mt. Wilson, California Do Forests Influence Water
Inputs into Ecosystems?
Precipitation
Beneath the Fog In the earlier part of the 20th century, there was a
Vegetation Vegetation, Drip, great deal of controversy about the relationship be-
Type cm cm tween forest cover and precipitation. It was claimed by
some that in comparison with other forms of vegeta-
Open (no vegetation) 58 0
tion, forests increase precipitation and also recharge
Shrubs: the air with moisture that promotes precipitation in
Ceanothus sp., downwind areas. Forest removal was thought to re-
2.4 m tall 58 0
duce precipitation and hence stream flow.
Trees Analysis of this question for temperate forests (e.g.,
Canyon live oaks, 121 63
Kittredge, 1948) has revealed that although the re-
13.7 m tall
moval of forest can reduce total stream flow in some
Pseudotsuga macrocarpa, 122 64
12.2 m tall
situations, this is mainly through reductions in fog
drip; direct effects of forests on precipitation are prob-
Pinus ponderosa, 154 96
ably minor. It has been suggested that the maximum
24.4 m tall
effect that a forest could have on precipitation in tem-
Data for October to May 1916-1917 from Kittredge, 1948. perate regions is a 5% increase but that this maximum
260 CHAPTER 10 = Water

is attained only in particular circumstances (Golding, in needle angles, and so on, the water capacity of which
1970). However, there is growing concern that defor- varies on different types of vegetation. Much of this
estation in the Amazon basin could significantly affect water will be released if the vegetation is shaken by
atmospheric humidity and precipitation over large wind, which serves to reduce interception loss. Heavy
areas downwind, and this topic will undoubtedly be- rain also serves to shake much of the stored water from
come a focus of future research as the deforestation of the crown. Evaporation of the stored water requires en-
the humid tropics accelerates. Much of the almost daily ergy and is promoted by air that has a high vapor pres-
rainfall in the Amazon basin is promptly re-evaporated sure deficit (VPD), which is the difference between the
and transpired back to the atmosphere, maintaining actual vapor pressure of the air and the saturation vapor
high atmospheric humidity and generating the clouds pressure at that temperature. It is a measure of the dry-
that provide the next day’s rain. Without forest vegeta- ing power of the air. Evaporation loss thus is high dur-
tion, most of the water would enter rivers, resulting in ing summer, when there is a high VPD, abundant
the progressive drying of the air. Even with the forest, energy for evaporation, and a long time between
some trees die or are damaged each year during the dry storms. Winters in winter-wet climates have little in-
season, and there is concern that large-scale conversion terception loss because of little energy for evaporation,
of forest to agricultural land could so reduce atmos- low or zero VPDs, and short periods of time between
pheric humidity and precipitation that many of the tree storms, which gives little chance for evaporative loss to
species in the remaining forest would die. occur. However, some evaporation does occur even in
The role of forests in maintaining the humidity of winter, and there can be surprisingly rapid evaporation
the air and thereby contributing to downwind precipita- between winter showers (Rutter, 1963).
tion is believed to be relatively unimportant in areas The absolute amount of interception loss is rela-
such as the west coast of Canada or Scandinavia, where tively independent of storm size because it is deter-
precipitation patterns are mainly a function of mountain mined by interception storage capacity, which is more
ranges interacting with regional air mass movements. or less constant. By the same argument, it becomes a
The forest in these areas does exert an important decreasing percentage of total precipitation as storm
influence on microclimatic humidity, however. In con- size increases. Interception storage for tree and shrub
trast to coastal British Columbia, the return of mois- cover has been reported to range between 0.25 and 7.6
ture to the atmosphere by forest evapotranspiration as mm of rain and up to 2.5 cm (water equivalent) of
the air mass crosses the undulating interior plateaus of snow (references in Satterlund, 1972). Interception
British Columbia certainly contributes to the precipi- loss of rain by trees can be 100% for light summer
tation that occurs as these air masses rise up over the showers in dry climates and 0% for heavy or continu-
mountain ranges to the east. Reynolds and Thompson ous rain in winter in humid climates. It is much re-
(1988) provided a discussion of the links among duced in deciduous trees in the nonleafy period.
forests, climate, and hydrology. Table 10-2 presents some figures for interception
loss in various forest types in the United States. The
loss of water through interception is dependent on
Interception of Precipitation by Vegetation crown structure, which in turn varies with the tree age
Not all the precipitation that reaches the surface of the and the density of stems in the stand (Table 10-3). In-
vegetation arrives at the soil surface, as anyone who terception can also occur for snow that builds up on
has taken shelter beneath a tree during a rainstorm can tree branches. In humid climates, there is little evapo-
tell you. Similarly, much of the dew or rime accumu- rative loss for such snow, but in continental climates,
lated overnight is evaporated the following morning there may be an appreciable loss to evaporation; the
and never reaches the soil. The loss back to the atmos- snow loses water vapor directly to the air without
phere of precipitation that has been intercepted by melting. This is discussed in more detail later.
vegetation is called interception loss. Interception losses make it difficult to arrive at ac-
The magnitude of interception loss depends on the curate figures for fog interception, and vice versa. Most
interception storage capacity of the vegetation, on the total fog drip estimates are the net contribution of fog drip
energy available to evaporate water held by the vegeta- after interception loss has occurred. Interception loss
tion, and on the movement and humidity of the air. also reduces the quantities of dew and rime water that
Water is held as a film on plant surfaces, in bark cracks, reach the ground. Summer interception loss is not a
 The Water Cycle 261

Table 10-2 Annual Percentage Interception Loss in the portion of the incident precipitation that drips
Various Different Mature Forest Types in the United from or falls through the vegetation canopy (some-
States
times divided into canopy drip and throughfall that
Interception, does not touch any vegetation), and (2) stemflow—the
Forest Type Location % portion that reaches the soil by flowing down the stem.
The importance of stemflow varies greatly. It is
Mature hardwoods S. Appalachian Mts. 12 generally insignificant in climax humid West Coast
Oak-pine New Jersey 13 forests, where the pendulous branches serve to divert
Quaking aspen Colorado 16 almost all precipitation to throughfall (Figure 10—2A).
Ponderosa pine Idaho 22 Immature forests, forests of hardwoods such as cherry
Lodgepole pine Colorado 32
(Prunus sp.) or alder (A/nus), and forests of pine tend
to have more stemflow because their upturned
Douglas-fir Washington 34
branches act as a funnel. Stemflow also varies with
Ponderosa pine Arizona 40 stand density. A 6-month study (May—December) in
Maple—beech New York 43 an 18-year-old Douglas-fir spacing plantation near
Eastern hemlock Connecticut 48 Vancouver, B.C. (Kimmins, unpublished data), re-
vealed that interception loss varied from 15 to 26% of
Data from Kittredge, 1948. incident precipitation as the number of trees per
hectare went from approximately 10,000 to approxi-
mately 730 and that the proportions of the water that
complete loss to a site because the stored water in the reached the forest floor by stemflow declined from
canopy may cause a small decrease in transpiration, 44% in the most dense stand to 3% in the least dense
thereby conserving soil moisture. However, the water stand. This difference was attributed to differences in
thus conserved will be less than the loss by interception. crown morphology (Figure 10-2B). The increase in
interception loss with decreasing stand density re-
Redistribution of Water by Vegetation sulted from the greater interception storage capacity
Precipitation in unvegetated areas reaches the ground of the large, long crowns of the more widely spaced
fairly uniformly over distances measured in tens of me- trees. The small crown storage and upturned branches
ters (heavily localized convectional storms excepted). of the trees in the dense stand delivered a higher pro-
Beneath all but the most diminutive vegetation, the portion of the incident precipitation to the soil via
situation is different. Water that is intercepted by tree stemflow than did the less dense stand.
crowns is redistributed into two major subtypes and Stemflow is also affected by bark roughness.
reaches the floor very nonuniformly: (1) throughfall— Smooth-bark species have little stem water storage ca-

Table 10-3 ¥ffects of Stand Age and Tree Spacing on Interception Loss
Interception Interception,
Age, yr Loss, cm %

Eastern white pine, 10 30.5 AS

S. Appalachian Mts.* 35 38.1 19
60 58.8) 26

Spacing Stems/ha
Douglas-fir, 18 yr old, 10,000 WaT 15
Coastal British Columbia? 3,000 19.5 23
1c 20.4 24
730 22.0 26

49The data for white pine are for annual interception losses (Kittredge, 1948).
’The data for Douglas-fir apply to an 8-month period, May~December
(Kimmins, unpublished data).
 assobsaat phe a We Dt
QP.lolol dha cohtt aectalaiaeeamane
REY, DRT hall a ean
SN i dyogemulPi Le ha
en
Nla a Ade i ag tee ee ee gee
Pee ne no eee bis) a PERL e 4
ea OF: age Pest oe
pan nae ee men
id i pe : TS aNtse
ae - a PP RA ees
ind > ae: ) ae 3 aI WEL
oe ‘a hie . a \ cot
aks ies hare a ee 2 :

i: - i: ae Con, atone

YYyY YY y Y YY y Se
en
et
te
tt
ee
ee
ee
on

i 2 3
Western Hemlock Pine Bitter cherry
or Red alder

Relative A
amount of C C G Cc S
water reaching LOE ins hoDee I ON, I C C I
the ground

I = incident precipitation C = canopy edge drip S = stemflow

(a)

Throughfall Stemflow
54% 46%
Throughfall Stemflow
97%
3%

weg? iy tie uel Tay ¥ eg hy

age °C; Sa ne " aw “bs eS

sgh 7 Rat Wa We
me
ES .
Pati ek a Z AP. fattn
wee a! shew A ng
¥ = Uy re A 5 “5 ney

1 X 1 m spacing 3.7 X 3.7 m spacing

Figure 10-2 Effect of variation in crown morphology and stand density on the relative im-
portance of stemflow and throughfall. (A) Tree species with pendulous branches tend to have lit-
tle stemflow and redistribute much of the precipitation into canopy-edge drip. Species with erect or
acute-angled branches have much more stemflow and less canopy drip. (B) The closer the spacing in
a stand, the higher the proportion of acute-angled branches and the greater the importance of stem-
flow: data from an 18-year-old Douglas-fir plantation.
262
 The Water Cycle 263

pacity, and stemflow will commence on smooth-barked 1972). In contrast, snow does not remain on tree
species such as beech after only a little more than 1 mm crowns for very long in snowy environments that are
of rain has fallen (Leonard, 1961; Voigt, 1960). Rough- subject to periodic above-freezing temperatures, such
barked species such as spruce, cedar, or mature Douglas- as subalpine coastal forests in western North America.
fir have a large stem storage capacity, and appreciable When snow falls from tree crowns, much of it is
stemflow may not reach the ground until more than 2 redistributed into canopy-edge snowfall. This con-
cm of rain has fallen (Helvey and Patric, 1965). tributes to the characteristic pattern of peak snow ac-
Where stemflow is appreciable, it serves to con- cumulation under canopy edges and minimum
centrate a lot of the incident precipitation close to the accumulation close to the stem, but redistribution by
base of the trees. In a study of the distribution of ra- crowns is not the sole cause of this pattern of accumu-
dioactive fallout in the soil beneath beech trees in lation. Redistribution by wind, water dripping from
Ohio, it was found that soil at the base of a tree where snow melting in the crowns, and melting of snow by
there was abundant stemflow contained five times as radiation from tree trunks also help to determine the
much radioactivity as soil where there was no stem- observed patterns of snow depth in forests. Snow
flow. The amount of water that arrived per unit area tends to accumulate in small openings in the forest
over a small area close to the tree bases was found to more than in closed forest or large openings because
be seven times that of the incident precipitation wind eddies in small openings favor the deposition of
(Gersper and Holowaychuk, 1970). The water deliv- snow. In a study in the eastern foothills of the Rocky
ered to the base of trees as stemflow penetrates the soil Mountains in Alberta, greatest snow accumulations in
rapidly by following large roots. Therefore, forests a lodgepole pine forest were found in openings equiv-
with a high proportion of stemflow may have a differ- alent to two to three times the tree height, but the
ent hydrological character from those with a low pro- longest duration of snowpack was found in openings
portion of stemflow. The rapid entry and deep equivalent to the height of the surrounding stand. In
penetration of stemflow into the soil reduces losses by smaller openings than this, the long-wavelength radia-
evaporation and therefore increases the availability of tion from the trees accelerated snow melt, whereas in
water to plants. This can be important to vegetation in larger openings, the greater exposure to sunlight ac-
dry areas, and many species in such areas have a mor- counted for a faster rate of melt (Golding and Swan-
phology that promotes stemflow. son, 1978).
It has been suggested (Horn, 1971) that the geom-
etry of belowground biomass reflects the effect of Infiltration into the Soil
aboveground biomass geometry on water distribution.
Water that reaches the ground can either flow lateral-
Tree species with crowns and stems that promote
ly over the surface or penetrate the soil in a process
stemflow tend to have deeply penetrating taproots,
called infiltration. Once within the soil, the movement
whereas those with little stemflow and a lot of canopy- of water is known as percolation. The term infiltration
edge drip tend to have more superficially located root
can apply either to the organic forest floor or to the
systems: heart roots and flat roots. This will optimize
underlying mineral soil, but because the rate of water
the recovery of moisture and nutrients by the plants. movement into the forest floor almost always exceeds
rates of precipitation and because the condition of the
forest floor is subject to modification and is therefore
Interception and Redistribution of Snow less permanent than the mineral soil as a site feature,
Snow is also intercepted and stored on the crowns of the term is applied most frequently to the mineral soil.
plants, sometimes to a greater extent than rain. Except Entry of water into the forest floor is normally
in windy weather, it remains on trees longer than rain rapid because of the many large pores and the organic
and is therefore potentially susceptible to greater evap- nature of the forest floor, which gives it a high mois-
oration losses. Snow was observed to remain on tree ture-holding capacity. However, forest floors that have
crowns for up to 50% of the snow season at a study site become very hot and dry during the summer may ex-
in the Sierra Nevada Mountains in California and hibit Aydrophobicity (literally, water hating), which
more than 60% in some upstate New York locations; makes them very difficult to wet. As a result, the rate
there may be continuous retention of snow on tree of entry and movement of water during the first fall
canopies in northern Canada (references in Satterlund, rains may be reduced. Hydrophobicity can also be in-
264 CHAPTER 10 = Water

duced by fire (see Chapter 12). It is not a permanent soil material (e.g., a suspension of clay particles or or-
condition, and once the forest floor has been rewetted, ganic matter), then the surface pores may become
hydrophobicity does not re-form upon mild drying. clogged and infiltration may be reduced.
Once wet, forest floors can hold between one and In all but very coarse types of soil, the crumb struc-
five times their own weight of water (Kittredge, 1948); ture of the surface mineral soil is critically important.
the more decomposed the organic matter and the If this is destroyed, infiltration is greatly reduced.
more rotting wood in the forest floor, the more water Compaction by grazing animals or harvesting equip-
it can hold. Only the water in excess of the fie/d ca- ment, loss of structure by plowing, or damage to the
pacity of the forest floor will infiltrate into the mineral surface structure of exposed mineral soils by the im-
soil, and it may take many weeks and many centime- pact of rain or by fire all can reduce infiltration. Heavy
ters of rain before the mineral soil beneath the forest raindrops that fall from branches can rapidly damage
floor becomes wet in the fall after a dry summer. exposed mineral soils. The kinetic energy of these
Infiltration into moist forest floors is normally rapid, drops is transferred to soil crumbs upon impact, and
and it has been widely thought that little or no surface this process may separate the crumbs into their com-
flow or runoff occurs. This is not always true, because ponent particles. It is important to maintain a forest
short-distance runoff can occur on sloping forest floors floor in which the kinetic energy of the falling rain can
as the result of the arrangement of leaf litter (leaves or be dissipated without affecting the soil structure. The
needles arranged like the tiles of a roof) or because of a ecological role of forest fallow in tropical shifting cul-
compact surface growth of liverworts or mosses that tivation (slash-and-burn, or swidden, agriculture) may
makes surface flow easier than infiltration. Considerable have as much to do with maintaining soil structure that
runoff has been measured on sloping sites in wet west- can resist the physical impact of monsoon rains as it
ern hemlock forests in British Columbia (Feller, 1974). does with sustaining soil fertility and controlling dis-
Such flows occur only over short distances (a few cen- eases, pests, and weeds. The forest fallow provides lit-
timeters to a few meters), because before moving very terfall that sustains populations of soil animals, which
far, the water forms into pools behind logs or collects in are vital for the maintenance of soil structure and
microdepressions and then infiltrates the forest floor. thereby help to prevent surface runoff and soil erosion.
Alternatively, the water may encounter a vertical plant
stem, down which it flows rapidly.
Infiltration into mineral soils is determined by Water in the Soil
both soil texture and soil structure, parameters that This topic is examined in more detail in Chapter 11,
define the size and arrangement of pores in the soil. but some concepts are discussed here. Water in the soil
Infiltration involves movement under the action of is classified as gravitational, available, and unavailable.
gravity into larger pores or old root channels (tubular The relative proportions of these three types of water
spaces ramifying through the soil, formed by roots vary according to the relative abundance of different
that have died and decayed). Where large pores are pore sizes, which in turn depends on soil structure and
frequent (sands, gravels, or well-structured silts or texture. Coarse soils (sands and gravels) have many
clays), infiltration rate will be high. Where they are large pores, and most of the water in the soil at field
few, much of the water must infiltrate the smaller capacity is available. Fine soils (clays and silts) have
pores by the process of capillary action (you will recall many capillary-sized pores, and much of the water at
that this is the process that causes water to rise up in- field capacity (an imprecise term for such soils) is un-
side a narrow (e.g., | mm diameter) glass tube when its available. The movement of water through soils under
surface is touched to the water surface). The smaller saturated conditions follows D’Arcy’s law, which says
the pores, the more rapid the initial entry of water into that water flow is proportional to the product of hy-
the surface pores but the slower the movement draulic conductivity (a function of pore size and distri-
through the soil once the surface pores are filled. bution), the hydraulic gradient, and the cross-sectional
Movement of water into and through the soil is af- area of soil through which the water is flowing.
fected by the amount of organic matter and swelling Water moves through the soil in response to hy-
clays in the soil. These swell on wetting and act to clog draulic gradients, capillary forces, and gravity. This
the pores, reducing rates of water movement and thus combination generally results in downward movement
the infiltration rate. If the infiltrating water is carrying during wet periods, when there is a high input of water
 The Water Cycle 265

to the soil surface, but in summer, when there is a net water) and an upward flow of water from lower in the
loss by evaporation from the surface, upward move- soil to maintain water in the surface layer, where the
ments may occur. Water that moves down through the energy is available for evaporation. Evaporation from
soil may encounter less pervious or even impervious bare soil in the summer is initially rapid, but water
layers. These may be lenses (strata, or pans) of finer- evaporates first from the larger pores, and when the
textured materials in stratified transported soils, or rate slows as these are emptied, water is withdrawn
layers in which pedogenic processes have deposited from smaller pores. Upward capillary movement is
materials. Iron, calcium, organic matter, clay, and slow, and under conditions of rapid evaporation, all
other materials are often deposited or accumulated in water may be removed from the surface soil. When
lower horizons, and this tends to obstruct percolation. this happens, the continuity of the water columns
Where such layers are impervious, downward move- from the surface to the lower soil layers is broken,
ment is prevented and a perched water table is formed. continued upward capillary movement essentially
Such perched water tables may be extensive and result ceases, and evaporation is limited to the slow upward
in swampy conditions overlying dry, freely drained ge- movement of water vapor. This sequence of events oc-
ological deposits. More frequently, they are intermit- curs more in coarse-textured than in fine-textured
tent and result in a mosaic of wetter and drier soil soils. Thus, summer evaporation from bare coarse-
conditions. and medium-textured soil is initially rapid, but once
In the absence of impermeable pans, water the surface layer dries out, it drops to very low levels,
drainage will continue until the groundwater table is and soil a few centimeters down may remain near field
encountered. This is a layer of saturated soil in contact capacity. In fine-textured soil, capillary columns re-
with an underlying impermeable layer of bedrock or main intact for much longer and the lower soil layers
compacted material. Water will remain at this level, may therefore become much drier. In the winter, there
percolate slowly down into cracks in the underlying is usually insufficient energy for much evaporation
material, move laterally under the influence of gravity and the soil surface of most soils remains moist.
(seepage water), or move back up through the soil as it Under a continuous cover of vegetation, insuffi-
dries from above (a process known as capillary rise). cient energy reaches the soil to produce much evapo-
The layer of soil above the water table that remains ration, even in the summer. Where there is an organic
moist because of capillary rise is called the capillary accumulation, evaporation from the mineral soil is fur-
fringe. Seepage water moves downhill until it joins ei- ther reduced because of the lack of continuous pores
ther a water surface or a subterranean body of water. from the mineral soil to the surface of the forest floor;
water that rises in the mineral soil is blocked by the hy-
drological discontinuity of the mineral—organic inter-
Loss of Water to Evaporation face. A well-decomposed forest floor resists drying out
and ‘Transpiration because much of the water is held by colloids, but even
Water is lost from soil by three major pathways: if it does dry, it constitutes an efficient vapor barrier
drainage to groundwater, evaporation back to the at- that reduces loss of water from the mineral soil.
mosphere, and uptake by plants. The equivalent of Evaporation of snow occurs in continental climates
760 mm of precipitation is delivered to the 48 coter- where there are many clear sunny days with dry air in
minous U.S. states each year, and of this, approximate- the winter. The rate is much lower than the evapora-
ly 370 mm is lost back to the atmosphere by tion of water from soil because snow has a higher al-
evaporation from forests and wildlands (Hewlett and bedo (most of the energy is reflected and therefore not
Nutter, 1969). By comparison, Australia receives 420 available for evaporation); much of the energy is used
mm of precipitation and loses 370 mm by evapotrans- to raise the temperature to 0°C, and 0.34 MJ kg" is re-
piration. Canada receives the equivalent of 600 mm quired to melt the snow. A total of approximately 2.8
precipitation, of which 230 mm is lost to evapotrans- MJ is required to evaporate 1 kg of snow at 0°C. In
piration (den Hartog and Ferguson, 1978). coastal climates, evaporative losses from snow will be
less because of the lower levels of solar energy avail-
Evaporation. Evaporation from the soil surface re- able during the cloudy winter. In both climates, the
quires two preconditions: energy in the form of solar small vapor pressure gradient between the snow and
radiation (2.24 MJ are required to evaporate 1 kg of the atmosphere acts to limit evaporation.
266 CHAPTER 10 Water

Measurements at an elevation of approximately usually very small unless the cuticle has suffered abra-
2,100 m in Utah revealed evaporation of approximate- sion or cracking, something that can happen to older
ly 7 cm of snow over a 180-day period, which repre- leaves; consequently, older leaves may lose more water
sented approximately 14% of the snowfall (Kittredge, through the cuticle than younger leaves with intact cu-
1948). Studies in an open meadow, an open pine for- ticles. This vaporization produces a water deficit with-
est, and a closed fir stand near Lake Tahoe, Nevada, in the foliage cells that is transmitted via the water
revealed a November to March loss of snow equiva- columns in the branches, stems, and large roots to the
lent to 22, 12, and 6 cm of water, respectively. The dif- fine roots, where uptake from the soil occurs. The re-
ference was related to differences in the quantity of sultant reduction in water in the rhizosphere sets up
solar energy available for evaporation at the surface of moisture gradients that result in movements of water
the snow, which resulted from differences in shading from the surrounding soil to the roots.
by the vegetation. The amount of water transpired by forests varies
Evaporative loss is particularly rapid during peri- according to the species, the quantity of foliage, the
ods of warm winds in the winter (e.g., the Chinook availability of water in the soil, and the availability of
winds in the Canadian Rocky Mountains).-On the solar energy to evaporate water from the leaves. The
basis of an energy calculation, it was estimated that variation between species results from differences in
there was an evaporative loss of snowpack of 1.2 and stomatal resistance (the resistance to diffusion of water
2.0 mm day"! during Chinooks in 1975 and 1976, re- through the stomata, which will be higher where the
spectively, on the east slope of the Rockies west of stomata are sunk deeply in a thick cuticle and lower in
Calgary, Alberta. The higher value in 1976 was attrib- leaves with a thin cuticle), internal resistance to water
uted to higher wind speeds that year. The highest movement in the plant (a function of internal mor-
daily evaporative loss calculated during these 2 years phology of conducting elements), and resistance to
was estimated to be 10.4 mm (Golding, 1978). uptake from the soil (a function of root morphology).
Table 10-4 shows how transpiration varies according
Transpiration. Loss of water from within the living to species and foliage biomass. Birch has a very high
cells of plant tissues to the atmosphere by vaporization transpiration rate per unit of leaf weight, but because
is called transpiration. Water that is absorbed by roots it has a modest leaf biomass, its annual transpiration
from soil is translocated upward to the foliage in the rate differs little from that of species with low transpi-
xylem of the roots and stem. This uptake and translo- ration rates per unit of weight but high foliage bio-
cation is driven by solar energy falling on the leaves mass. Thus, it can be seen that annual transpiration
and stems, which causes water to evaporate from the per unit area of fully stocked forest varies much less
moist outside surfaces of mesophyll cells into air than the rate per unit of leaf biomass. Within one
spaces within the leaf. The water vapor either diffuses species or type of plant (e.g., conifer versus deciduous
out to the atmosphere through stomata or evaporates hardwood), leaf biomass is related to leaf area, so the
directly through the cuticle of leaves. The latter is more fully stocked a stand or the greater the leaf area

Table 10-4 Variation in Transpiration Losses of Water from Forest Trees

Daily Transpiration Annual Transpiration
Biomass of Foliage, of Leaves, g HO g™ of 40- to 50-Yr-Old Stands,
Species kg ha" green leaves cm

Birch 4,940 9.5 17

Beech 7,900 4.8 14
Larch 13,950 3.2 17
Pine 12,550 1.9 8.6
Spruce 31,000 1.4 16

Douglas-fir 40,000 1S) 19

Data from Rutter, 1968. Copyright Academic Press. Used with permission.
 The Water Cycle 267

Table 10-5 Utilization of Water by Forests in Various Parts of the World

Annual
Evapotranspiration Loss
Annual Precipitation, Growing Season
Forest Type cm cm % Soil Water Deficit

Northern taiga 62:5 28.6 54 Negligible

conifer forest,
Russia
Southern taiga 60.0 32.9 55 Negligible
conifer forest,
Russia

Spruce stand, 135.0 80.0 59 Negligible

Great Britain
Mixed conifer and 165.0 86.1 52 Negligible
deciduous stand,
Switzerland

Mixed conifer and 261.7 54.2 21 Negligible

deciduous stand,
N. Japan
Evergreen rain 195.0 oa) 81 Small
forest, Kenya

Deciduous forest, 45.7 42.4 93 Moderate

European Russia
Coulter pine, 123.0 63.7 52 Severe
California

Ponderosa pine, 126.0 58.0 46 Severe

California
Coulter pine, San B25) 63.7 7A5) Extreme
Dimas, California

After Rutter, 1968.

index, the greater the transpirational loss if there is ter precipitation experiences much smaller evapora-
sufficient solar energy available. tion losses than does summer precipitation. At San
Table 10-5 shows how the annual evapotranspira- Dimas, there is little or no water to transpire in the
tion rate varies in different types of forests around the summer, when the potential for loss is high: the rains
world. It can be seen that the percentage of the annual come in the winter, when potential transpiration is
precipitation that is transpired varies considerably but much lower.
not necessarily in proportion to the growing-season The importance of stomatal opening in the control
soil-moisture deficit (annual growing-season precipi- of transpiration loss warrants a brief discussion of the
tation minus potential growing-season evapotranspi- topic. Stomatal opening and closure are thought to be
ration). Northern areas tend to have negligible deficits largely determined by intercellular CO, concentra-
and lose only a modest percentage of annual precipita- tion; when this is low, stomata open, and when it is
tion by transpiration. Much higher transpiration high, they close (Slatyer, 1967). Both temperature and
losses might be expected in a hot, dry climate, yet two light have a direct effect on stomatal opening, but
studies involving ponderosa and coulter pines at San most of this effect is indirect, via the effect of temper-
Dimas, California, found only 46 and 52% transpira- ature and light on internal CO, concentrations. These
tion loss, respectively, associated with severe soil- drop during the day because of photosynthesis (sto-
moisture deficits. The lack of the expected correlation mata open) and increase at night (stomata close) be-
relates to the annual distribution of precipitation: win- cause of respiration. Stomatal opening is also
268 CHAPTER 10 = Water
te

influenced by water stress, because stomatal opening pas) bpi

depends on the turgor of the guard cells; this can over-

ride the effects of light, temperature, and internal CO;
levels. However, water deficits do not affect closure Moving air
until some critical level of water stress has been
reached. Sudden increases in stress can actually cause a
temporary increase in stomatal aperture because more
rapid water loss may occur from cells that are adjacent
to the guard cells than from the guard cells themselves.
Transpiration rate is also affected by wind. As iS= nnvox)
water vapor emerges from the stomata and the cuticle,
it enters the thin layer of air that is in contact with the
leaf (the boundary layer). This increases the partial
pressure of water vapor in the boundary layer, which in
S)
turn depresses the diffusion of vapor out ofthe leaf and
reduces the rate of transpiration loss. If this layer is
continually removed by wind and replaced by drier air, 1077)
sec!
em?
xTranspiration
(g
rate
then transpiration continues unimpeded. Figure 10-3 0.5 Still air
shows the effect of air movement on transpiration loss
from leaves. As stomatal aperture increases in still air,
transpiration slows because of a buildup of water vapor
around the leaf. In moving air, transpiration continues
to increase with increasing stomatal aperture. Wind
5 10 15 20
speeds of 13 km hr! have been estimated to increase
Stomatal aperature (u)
transpiration by 20%, with a 50% increase in 40-km
hr! winds (Clarke, 1967). The uptake, internal trans- Figure 10-3 Effect of air movement on transpiration loss
port, and loss of water by coniferous trees and stands is from Zebrina leaves. The graph shows transpiration rates at
reviewed in Pallardy et al. (1995). different stomatal apertures in still and moving: air. In still air,
water vapor accumulates in a thin layer of air around the leaf.
Water Balance of Plants This layer inhibits transpiration, so water loss becomes rela-
tively independent of stomatal aperture. In moving air, this
The water balance of plants is a function of three resis- boundary layer is removed and transpiration becomes sensitive
tances: (1) the resistance to water movement in the soil to stomatal aperture. (After Bange, 1953. Used by permission of
to the root and its entry into the root; (2) the resistance the Koninklijke Nederlandse Botanische Vereeniging, The Nether-
of water movement within the plant; and (3) the resis- lands, and North-Holland Physics Publishing.)
tance of the plant to loss of water by transpiration,
mainly from the leaves (Timmis and Worrall, 1974).
The resistance to water movement in the soil and The movement of water through a tree is related
its entry into the plant depend on the hydraulic con- to the cross-sectional area of sapwood and the size of
ductivity of the soil, the length and surface area of the “pipes” in the conducting tissues. A tree with slow
water-permeable roots, and the extent of the mycor- growth will have a higher resistance to internal water
rhizal system that can assist the plant with water up- movement than a faster growing tree because of less
take. Lack of an adequate fine-root and mycorrhizal sapwood area and the smaller diameters of the vessels
system, such as in recently planted tree seedlings, and fibers through which the water has to move.
greatly increases this resistance, as does fine soil tex- Resistance to water loss from the foliage depends
ture and low soil temperatures. Recently planted on the ability of the plant to limit stomatal and cuticu-
seedlings with poor root systems on moist, cold, fine- lar water loss. This depends in part on leaf area and
textured soils can have a negative moisture balance leaf morphology.
that leads to desiccation and can result in seedling Where resistance to internal water movement is
death. Dry, coarse-textured soils can also have a high higher than the resistance to loss from foliage, a plant
resistance to water movement to the roots. will be susceptible to internal water deficits, especially
 Availability of Water to Plants and Animals — 269

if damage to roots has increased the resistance to

10.3 Availability of Water
water uptake. This can occur if tree seedlings that are
raised under adequate moisture in forest nurseries are
to Plants and Animals
planted out with damaged or poorly developed root
The hydrological cycle is a description of the distribu-
systems under conditions of high atmospheric mois-
tion and movement of water between water bodies,
ture demand. Animal or fungal damage to roots or
the atmosphere, and terrestrial ecosystems. It does not
root death due to anaerobic soil conditions, can pro- define the consequences for living organisms of the
duce the same result. distribution and movements of water.
The availability of water to organisms, its ecologi-
Loss of Water to Streams cal effectiveness, is determined by several factors.
Water that is not lost by evaporation and transpiration
and is not held in the soil either enters long-term stor- Timing. ‘The moisture requirements of terrestrial or-
ganisms vary from season to season. Greatest demands
age in groundwater aquifers or moves to a stream. The
are usually during the growing season or metabolically
rate at which water moves from the land to a stream
active period, and requirements may be very small dur-
depends on whether it infiltrates into the soil or runs
ing the rest of the year. Therefore, annual precipitation
off the surface. The former results in a greater delay in
figures are often of limited use in ecology. Many parts
reaching a stream than the latter. Once water is in the
of California, which have extreme soil moisture deficits
soil, the speed of movement to the stream depends in
during the growing season, have the same annual pre-
part on soil texture and structure, which help to deter-
cipitation as some continental locations that have only a
mine soil hydraulic conductivity. The speed also de-
small growing-season soil-moisture deficit. The differ-
pends on the abundance of old root channels—“pipes”
ence results from the California sites receiving almost
in the soil left when roots die and decompose. These
all of their rain in the winter, whereas in the interior of
are Important in many forested watersheds and can
the continent, there is a more even distribution
contribute to a rapid response of stream flow to pre-
throughout the year. Only precipitation that is available
cipitation events even where there is no surface runoff.
to an organism when it is needed is of significance to
Watersheds with thin- and coarse-textured soils
the water balance of that organism.
store little water and deliver precipitation rapidly to
streams, whose flow rates rise and fall rapidly in re- Soil Moisture Storage. Dry sites are very often dry
sponse to alternating rainstorms and dry periods. because of rapid drainage and a small water storage ca-
Areas with deeper and finer-textured soils or with pacity in the soil, rather than because they receive less
abundant organic matter and deep forest floors store precipitation. A coarse gravel soil will be much drier in
more water and generally release it to streams more the summer than an adjacent loam soil. In the former,
slowly, giving less variation in stream flow events, un- water tensions may approach the wilting point within
less water movement through the soil is dominated by a few days of a rainstorm, whereas a subsequent rain-
flow through old root channels. Once the soil water storm may well occur before the moisture that is
storage capacity has been exceeded, such as after stored in the soil pores of the loam soil has been de-
abundant rainfall, stream flow response to further rain pleted to the wilting point.
may be relatively independent of the soil condition
and forest floor depth. Rapidity of Drainage Losses. The longer that
Water that enters the ecosystem as snow is stored moisture remains within the rooting zone, the more
as snowpack until spring snow melt. The rate of melt ecologically effective it will be to plants. Steeply slop-
depends upon shading of the snowpack, air tempera- ing land sheds moisture more rapidly than flat land
ture, rain, wind, aspect, and the influence of tree stems and consequently is drier, unless it receives seepage
on snow melt. Snow melt in a watershed with patches water from upslope drainage. In very deep soils, mois-
of forest of different ages and stand density and signif- ture may drain so far below the rooting zone that it is
icant variability in aspect will occur over a longer time of no consequence to the vegetation. As a result, medi-
period than a watershed that is relatively uniform in um-depth coarse soils over an impermeable layer can
stand conditions and aspect. This will result in very sometimes be moister than very deep, finer-textured
different temporal patterns of stream flow in these dif- soils. In the former, rapid drainage is arrested at a
ferent watersheds. point that is still accessible to the plant.
270 CHAPTER 10 = Water

Physical State of the Water. Frozen water is un- cal adjustments are made. They will be killed if trans-
available to most organisms, and water vapor is of ferred directly from salt to fresh water, or vice versa.
much less direct significance to organisms than liquid Tolerance to a range of salinities may therefore be de-
water. Although ice, snow, and water vapor all have pendent on the rate of change in salinity.
important ecological effects in terrestrial ecosystems,
as a usable source of water, they are generally insignif- Intensity of Precipitation. Water that falls on veg-
icant. The availability of water varies somewhat be- etation only to be lost through interception is not
tween the extremes of freezing point and boiling available to plants, apart from minor quantities that
point. At the upper end of this scale, water is generally might be absorbed by the foliage. Similarly, small
unavailable because it is above the cardinal tempera- quantities of precipitation that reach the soil surface
ture of the organism. At lower temperatures, uptake may be re-evaporated rapidly and serve no useful
by plants may be reduced because of an increase in the function for plants. Precipitation that is distributed
viscosity of water in both the soil and the plant. into many light showers therefore may be less avail-
able to plants than precipitation in fewer but moder-
Chemical Composition of the Water. Water is ate-intensity showers. If all the rain falls in a single
available to organisms only when it is in the appropri- intense storm, then its availability is again reduced be-
ate chemical condition. Humans will not drink heavily cause much of it is lost as surface runoff or subsurface
polluted water and will die on a prolonged diet of sea- storm flow. Maximum availability occurs when the
water. Water with high concentrations of solutes can- rain comes in moderate-intensity storms in which the
not be absorbed by the roots of most plants. In fact, rain penetrates more deeply into the soil, where it re-
roots will tend to lose water and become desiccated if mains available to plants.
immersed in water with high concentrations of
solutes. Plants that grow in bogs frequently experience
high moisture tensions because of the tendency for the Evaporative Power of the Air. Except during a
solute-rich bog water both to resist uptake and to re- rainstorm—and sometimes even then—the atmos-
move moisture from the plants. Only specially adapted phere is almost always dry enough to permit evapora-
plants can survive in such ecosystems. As a result of tion and transpiration to occur. The drier the air (high
these adaptations, some tree species that grow on bogs VPD values) and the greater the energy available for
can also be found growing on dry, gravel, or rock evaporation, the shorter the period of time over which
ridges. Both black spruce and lodgepole pine behave water is available to plants and the less effective a given
this way in western Canada, and western redcedar can amount of precipitation. As already discussed, wind in-
be found growing on both wet lower-slope sites and fluences evapotranspiration greatly, so moving air nor-
dry ridge tops. Presumably, the adaptations that per- mally reduces the residence time of moisture in an
mit water uptake in the physiologically dry bog also ecosystem. Approximately 38% of the 600 mm of an-
equip the trees for survival on the dry ridges. nual precipitation in Canada (a cool-to-cold high-
The effects of water chemistry on an organism’s latitude country) is lost to evapotranspiration (den
water balance are of particular significance to aquatic Hartog and Ferguson, 1978). As noted earlier, approx-
organisms. These are divided into two groups accord- imately 70% is lost in this way in the coterminous
ing to their adaptations: stenohaline organisms are United States, a lower-latitude country with a hotter
those that are adapted to only a narrow range of solute climate, whereas the comparable figure for Australia, a
concentrations, and ewryhaline organisms are tolerant country characterized by an arid climate, is estimated
of awide range of solute concentrations. Plants or an- to be 88%. Obviously, evapotranspiration has a great
imals that live exclusively in the ocean or in fresh effect on how long rainfall remains available in the soil.
water are generally stenohaline, whereas those that
live in the tidal portions of estuaries and are alternate-
ly exposed to salt and fresh water are euryhaline. The 10.4 Adaptations of Plants to
difference between euryhaline and stenohaline is not Excess or Deficit of Water
absolute. Anadramous fish (those that spend part of
their life cycle in the ocean and part in fresh water) Plants require water for almost every part of their life
may have to spend a period in the brackish waters at cycle. Seeds require moisture to germinate and will re-
the mouth of the river while the necessary physiologi- main dormant if they fall on dry soil, even if chilling
 Adaptations of Plants to Excess or Deficit of Water 271

and/or heating requirements have been satisfied. This Table 10-6 Effect of Soil Texture on the Availability
moisture requirement may result from the need to of Water to Plants
leach growth inhibitors from the seed coat (particular-
Moisture Permanent
ly important in desert plants, where inhibitors prevent
Soil Textural Class Equivalent? Wilting %°
germination after a light shower and delay it until the
heavy winter rains), to soften hard seed coatings so Clay 28.4 13.4
that the root radicle can emerge, and/or to permit the Loam Pai sll 10.3
seed to swell and burst its covering. Silt loam 16.1 Ud)
Moisture is required for growth because with inad-
Sandy loam 9.5 2.9
equate moisture, gaseous exchange in the foliage is
Fine sand 3.2 1.0
drastically reduced. Plants regulate water loss from
leaves by developing a waxy surface layer, the cuticle, Data from Veihmeyer, 1956.
which is nearly impervious to water. Unfortunately, *Moisture equivalent is the percentage of water in an initially saturat-
cutin is also impervious to CO, and O,, which must be ed soil remaining after a centrifugal force equal to 1,000 x g has
been applied to it for 10 to 40 min.
exchanged if the plant is to grow. This problem was
®PW% equals the weight of water in the soil at the permanent wilting
solved with the evolution of stomata, which connect point (PWP) as a percentage of the weight of the soil.
the leaf atmosphere with the external atmosphere.
Active uptake of moisture is important to plants
because they rely on the transpiration stream to trans-
two measures of soil moisture that are of significance to
port nutrients from the roots to the leaves. Movement
plants.
of water and nutrients from roots to leaves promotes
The variation in the PW% of different species of
uptake of both from the rhizosphere, which in turn
plants growing on the same soil is generally much less
causes water and its solutes to move from remote soil
than the variation for a single species growing on dif-
pores to the rhizosphere.
ferent soils, and some ecologists have claimed that the
Plants withdraw water from the soil until there is no
PW% is essentially common for all plants—that it is a
longer a gradient of moisture tension between the plant
soil parameter rather than a plant parameter. Contrary
and the soil. Transpiration creates a water stress in the
to this point of view, it has been shown that there is a
plant that results in moisture tension within the roots.
significant variation in PW% between different plants
When the soil is moist, water moves into the roots in re-
and that a variation in PW% as small as 0.5% or less
sponse to this tension. As the soil dries, the gradient be-
(which is within the limits of experimental measure-
tween the plant and the soil diminishes and less water is
ment of PW%) may be of great significance to a plant.
absorbed. When a soil has dried to the point at which
This is because a small variation in PW% may be as-
soil-moisture tension approaches the plant-moisture
sociated with a large variation in soil-moisture poten-
tension, uptake suddenly declines sharply, the plant tial (lable 10-7). In some soils, such a small difference
wilts, the stomata close, and transpiration virtually ceas-
es. The soil-moisture potential at which this occurs is
referred to as the temporary wilting point (temporary wilt-
ing can also occur because transpiration exceeds uptake Table 10-7 Variation Among Plants in the PW%, Soil
Water Potential at the Permanent Wilting Point and
and translocation even when the soil is wet). Plants will
the Final Soil Water Content after Permanent Wilting
recover turgidity when soil-moisture tension drops
again or the transpiration rate drops. However, if the Permanent Soil Water Final Soil
soil water potential decreases further, then plants will Wilting Potential, Water Content,

wilt to the point at which they will not recover turgor. Plant % atm %?

The soil-water potential at which this occurs is called Tomato 11.8 -18 9.8
the permanent wilting point. The amount of water in the
Cotton 10.2 -38 7.0
soil (percentage by weight) at the permanent wilting
point is called the permanent wilting percentage (PW%). Privet 9.7 -47 6.9

This varies greatly according to the texture of the soil Data from Slatyer, 1957.
because texture dictates the tension with which the re- aWater uptake by plants generally continues after the permanent
maining water is held in the capillaries as the soil dries. wilting point is reached, so the final soil water content is lower than
Table 10-6 shows the relationship between texture and the PW%.
272 CHAPTER 10 = Water

in PW% could supply a plant with enough moisture to substrate contains none (Daubenmire, 1974). Restric-
keep it alive for 6 full days (Levitt, 1958), which could tion of roots to the better aerated surface soil, devel-
make the difference between survival and death of the opment of pnewmatophores (special root branches that
plant. Another argument against the theory that grow vertically upward until they emerge into the air
PW% is only a soil parameter is that early investiga- and that are usually equipped with stomata and a well-
tions of PW% ignored plants that do not conserve developed intercellular system of air spaces), the abili-
water: called the water spenders (discussed below). Al- ty to respire anaerobically for limited periods of time,
though water absorption by these types of plants may and an overall low requirement for O, are other mod-
be greatly reduced under conditions of low soil water ifications that permit plants to grow in saturated ter-
availability, they are able to resist permanent wilting to restrial environments. Some species of mesophyte
a lower soil water content than other types of plants. have the ability to develop these hydrophytic adapta-
Plants can be classified according to their adapta- tions when growing in wet soil.
tions to water stress.

1. Those that are adapted to partial or complete sub- Moisture Deficits

mergence in free water are called hydrophytes. Such Although hydrophytes are an important group of
plants cannot grow at soil moisture tensions plants, there has been much more interest in xero-
greater than 500 to 1,000 kPa.’ phytes and the adaptations of plants to moisture
2. Terrestrial plants that are adapted to moderate deficits (e.g., Kozlowski, 1968; Levitt, 1972; Slatyer,
water supplies are called mesophytes; they can grow 1967). Xerophytes can be classified according to the
at soil moisture tensions of up to 2,000 kPa. They manner in which they respond to moisture stress
lack the specialized adaptations to excess or deficits (Levitt, 1972):
of moisture found in hydrophytes and xerophytes,
respectively. 1. Drought escapers: ephemeral plants.
3. Plants that are adapted to arid zones are called 2. Drought Resisters.
xerophytes. They can grow under moisture tensions a. Avoidance of drought stress: water savers,
as great as 4,000 kPa. They are characterized by a water spenders.
variety of xeromorphic and physiologic adapta- b. ‘Tolerance of drought stress: avoidance of dehy-
tions, but attempts to define xerophytes on the dration, tolerance of dehydration.
basis of morphology have been unsuccessful.
Drought Escapers: the Ephemerals. Ephemerals
are plants that escape from the adverse effects of
Excess Moisture drought on internal water balance by eliminating most
The term hydrophyte refers to strictly aquatic plants or all of their water and suspending metabolic func-
and to those that grow in bogs, swamps, and marshes. tions that depend on water. They can do this by enter-
Nonhydrophytic plants are generally excluded from ing the seed or spore stage. Arid regions usually
these saturated semiterrestrial habitats because of the support a wide diversity of small annual plant species
soil O, deficiency that accompanies saturation. The that flower only during rainy seasons, which may
tolerance of hydrophytes to this condition can be ex- occur annually or only every few years. These plants
plained partly by the characteristic presence of inter- often lack the xeromorphic characteristics of other xe-
nal cavities and spongy tissues in many hydrophytes. rophytes, and they cannot tolerate soil drought. Their
These facilitate the movement of O, from leaves and main characteristics are small size, a high root/shoot
stems to roots, thus permitting them to function in an ratio, and an ability to complete their life cycles very
O,-deficient environment. A good example of this is rapidly. Germination, growth, flowering, and seed set
the rice plant (Oryza sativa), in which the submerged all are completed within the few weeks after the rainy
roots may contain up to 18% O,, whereas the muddy period, when the soil has sufficient moisture for the
plant’s needs. Because desert annuals cannot actually
tolerate soil drought, they are not always considered
‘PA = pascal, the metric unit ofpressure. 1 PA = Nm, where 1 N (New- to be true xerophytes. It is worth noting that the seeds
ton, unit offorce) =kg ms”. of most plants are drought resistant, that not all annu-
 Adaptations ofPlants to Excess or Deficit of Water = 273

als can survive in arid conditions, and that most desert even more and may become negligible before wilting
ephemerals do not grow outside arid regions. occurs. The deposition of additional lipids in the cuti-
cle also helps to slow cuticular transpiration.
Drought Resisters
Avoidance of Drought Stress. Avoidance of drought WATER SPENDERS. Accelerated water absorption
stress involves the maintenance of a high water poten- permits the maintenance of high internal water con-
tial when a plant is exposed to an external water stress. tents despite high transpiration rates by xerophytic
There are two major adaptations: water conservation water spenders. It enables them to keep their stomata
(water savers) and water utilization (water spenders). open in the day, and water spenders are associated
Early investigators noted the striking morphological with higher rates of photosynthesis and growth than
characteristics of xerophytes and assumed that drought the water conservers.
resistance resulted from water conservation. This as- Very high root/shoot ratios provide plants with a
sumption was challenged when it was discovered that larger soil water capital to draw on, which helps to
some xerophytes lose water faster than some meso- sustain transpiration, and the larger root systems may
phytes when both plants are grown in a moist habitat. penetrate down to water supplies deep in the soil.
It is now known that both adaptations exist. Water These species may provide some additional moisture
savers may lose as little as 1/4,300 of their weight of for shallow-rooted plants by the process of hydraulic
water per day, whereas water spenders may lose up to lifting (“Precipitation and Other Inputs of Moisture
five times their weight of water per hour (Kilian and into Ecosystems”). Artificial reduction of the
Lemee, 1956). Thus, water spenders may lose water as root/shoot ratio reduces drought resistance in water
much as 500,000 times as rapidly as water savers. spenders. For example, it was found that Cryptomeria
japonica was more resistant than Chamaecyparis obtusa
WATER SAVERS. The best examples of water savers when the roots were allowed to penetrate freely in
are the succulents (e.g., cacti) and some sclerophylls. deep soil, but the resistance order was reversed when
Both are able to restrict transpiration long before wilt- rooting was restricted by growing the trees in shallow
ing occurs. Succulents avoid drought by absorbing containers (Satoo, 1956). Pinus densiflora, which had
large quantities of water whenever it is available and the highest root/shoot ratio, was the most drought re-
storing it in swollen parenchymatous tissues and vac- sistant of several conifers tested. Plants that character-
uoles. Their roots tend to be very superficial so that istically root down to a water table are called
they can exploit even light rainfall that wets only the phraeatophytes, an extreme case of which 1s Prosopis fa-
surface soil. The roots are often drought-deciduous, vata, a desert plant of the Near East, the roots of
dying shortly after a rain and regrowing when mois- which extend 15 m down to the water table (Oppen-
ture is next available. Loss of water is avoided because heimer, 1951). The roots of Jarrah trees (Eucalyptus
the stomata open only at night and for a short time in marginata) that grow in southwestern Australia have
the morning. They remain closed during the heat of also been measured to a depth of 15 m (Kimber, 1974).
the day. The CO, required during the day for photo- A final root adaptation is the ability to send out new
synthesis is obtained from organic acids accumulated roots into unexplored soils during a drought. The sur-
at night (Ting and Duggar, 1968). Water loss is also vival of many species in dry regions depends on this
reduced by having a very small surface area-to-volume adaptation. Water spenders also have the adaptation of
ratio. Leaves are reduced or even eliminated, the plant a reduced permanent wilting point.
being merely a swollen stem. Water loss as a result of Water spenders may be able partially to alleviate
damage to the cuticle by herbivores is avoided by the drought stress by absorbing atmospheric moisture or
development of sharp spines and chemical defenses. nocturnal dew directly through the foliage. This may
Nonsucculent water savers include drought-resis- be of importance to plants in foggy localities or areas
tant conifers and evergreen angiosperms. Their pri- characterized by hot, dry days and cold nights. It is
mary adaptation is having very impermeable cuticles. thought that dew absorption may play a significant
Cuticular transpiration in water savers is a much role in the survival of planted tree seedlings in dry
smaller fraction (1/5 to 1/50 of stomatal transpiration) sites that are subject to dew in the evening. Water
than in mesophytes (1/2 to 1/5 of stomatal transpira- spenders also have the ability to become water savers
tion). As water is lost, cuticular transpiration decreases when the water supply becomes extremely limiting.
274 CHAPTER 10 Water

Tolerance of Drought Stress. Tolerance to Table 10-8 Effect of Moderate Drought Precondi-
drought conditions, which is exhibited by many plants tioning on Drought Tolerance in Two Tree Species
and may involve either tolerance or avoidance of de- Preconditioning Drought
hydration, can be achieved by a variety of mechanisms. Water Stress, Tolerance,
High levels of solutes in cell sap give cells a low os- Species bars bars
motic potential (high osmotic pressure), which keeps
Birch 0.1 34
them turgid even under drought stress, thereby avoid-
ing dehydration. Death from drought may result from 1.0 50
the “starvation” that accompanies restricted exchange 2.0 60
of CO. Tolerance to drought may therefore be Aspen 0.1 48
achieved by maintaining stomata open despite reduced 0.5 56
plant water content, thereby avoiding CQ) starvation.
1.0 58
Under these conditions, transpiration is reduced by
low cell osmotic potential, but CO, exchange contin- 4.0 7S;
ues. Alternatively, starvation can be avoided by having
Data from Jarvis and Jarvis, 1963.
a lower metabolic rate. This reduces the demand for
CO, uptake and makes the available photosynthate
last longer. The low dry matter content of succulents
is associated with low metabolic demands per volume Reduced size of leaves
of leaf and a tolerance of CO, starvation. Similarly, the Thicker leaves
drought-related death from starvation that accompa- Smaller leaf cells
nies a loss of protein may be either avoided or toler- Thicker cell walls
ated. Finally, some plants simply are not damaged by Smaller stomata, crowded together more densely
being dried out. Mosses, lichens, and some ferns shoPresence of leaf hairs (pubescent)
Bae
Pum
(sometimes known as “resurrection” plants) can re- Thicker cuticle with more lipids
main in an air-dry condition for prolonged periods Better development of palisade mesophyll
without injury. Growth resumes when the plants are preaLess pronounced development of spongy mes-
able to reabsorb water. Epiphytic ferns are often par- ophyll
ticularly tolerant of being dried out. For example, a Smaller intercellular spaces
fern Polypodium polypochoides was found to survive un- Smaller xylem cells
injured after losing 97% of its normal water content Higher proportion of heavily lignified tissues
(Stuart, 1968). oP
eS More deeply sunken stomata
Drought tolerance, like tolerance to low tempera- i) Physiological characteristics
tures, is influenced by preconditioning, or “harden- Accumulation of sugars
ing”: the exposure to a sublethal stress that results in Higher cell sap concentration
resistance to an otherwise lethal stress. Table 10-8 Lower osmotic potential
shows how moderate levels of water stress result in an Lower cellular water content
increase in drought tolerance of birch and aspen. Higher rate of transpiration per unit area of leaf
More rapid photosynthesis per unit area of leaf
Xerophily: Morphological and _ Physiological
Lower starch/sugar ratio
Adaptations to Drought. The following are com-
Lower protoplasmic viscosity
mon xeromorphic characteristics (Daubenmire, 1974;
Increased protoplasmic permeability
Levitt, 1972). They may be under genetic control and
Greater resistance to wilting
therefore be invariable. In some species, they vary ac-
Earlier flowering and fruiting
cording to the environmental moisture conditions to
Greater longevity
which the plant is exposed.
oe
Bo
we
eS
SG
ae Increase in the percentage of bound water ab-
1. Morphological characteristics sorbed by colloids per unit weight of dry tissue
a. Reduced shoot size For a review of water relationships in plants, see
b. Increases in the extent of the root system Kramer (1983).
 Adaptations ofAnimals to Excess or Deficit of Water = 275

10.5 Adaptations of Animals to Excess water is lost rapidly from such surfaces, which
or Deficit of Water both impairs the efficiency of gaseous exchange
and threatens the water balance of the animal. Or-
Water makes up a large proportion of the body weight ganisms were unable to invade terrestrial environ-
of animals (70 to 90% of protoplasm), all of which ments until evolution had overcome this problem.
must maintain an appropriate water balance to sur- The solution for insects is a series of sinuous
vive. Most of the adaptations to water found in terres- branching pipes called trachea that leads from the
trial animals are related to restricting the loss of water exterior opening (the spzracle) to the site of gaseous
while permitting exchange of respiratory gases and the exchange. Most of the exchange is by diffusion
elimination of wastes. Aquatic animals also have adap- rather than by pumping, which reduces the force-
tations to maintain gaseous exchange (gills and/or gas- able expulsion of moisture vapor. The spiracles
pervious outer layers), and they can experience an close under excessively dry conditions, restricting
excess or deficit of water if their internal osmotic po- moisture loss still further; the insects rely on O,
tential does not match that of the surrounding water. within the trachea while the spiracles are closed.
The necessary physiological changes required during Moisture loss from animals with lungs is restricted
the migration of anadromous fish have already been by the location of the lungs deep within the body.
mentioned. Interesting though such aquatic adapta- The O, exchange sites (alveoli) are joined to the
tions are, the following section focuses on adaptations exterior by the system of finely divided pipes
in terrestrial animals. called the bronchioles.
Many land animals use the evaporation of water as a Dry excretion. An important role of water in ani-
cooling mechanism. Sweating and panting are impor- mals is to transport waste materials out of the
tant means of maintaining internal temperatures in body. In many environments, use of water for this
homeotherms, but this can result in excessive loss of purpose would constitute profligate waste, and
moisture. Animals in dry environments have evolved a animals in such environments have evolved a
variety of mechanisms to ensure moisture conservation. more conservative approach to excretion. Urine
is concentrated to a much lower osmotic poten-
1. Impervious integument. Only animals with a rela- tial than blood by active recovery of water, and in
tively impermeable body covering (e.g., reptiles, the most water-conserving animals such as in-
birds, mammals, insects) can remain permanently sects, reptiles, and some birds, nitrogenous
in dry terrestrial environments. Animals with wastes are excreted as solid uric acid. Animals
moist skins, such as amphibians, are restricted to that inhabit moist environments may have liquid
sites such as swamps or moist soils and are fre- feces (e.g., cows). Animals of drier environments
quently nocturnal in habit. Desert environments (e.g., rodents, sheep, horses, antelopes) have rela-
are normally inhabited by reptiles and insects, tively dry feces. In humans, the concentration of
which have a dry and extremely impervious exte- urine increases and the moisture content of feces
rior, but certain nonsweating mammals and birds decreases as the body becomes progressively de-
can also survive in such environments. The sur- hydrated.
vival of these dry-skinned animals is totally de- Suspended animation. Animals can avoid drought
pendent on the integrity of their impervious outer by suspending the metabolic activities that re-
layers, and insects such as cockroaches can be quire moisture or increase the loss of moisture.
killed almost as readily by carborundum powder Such periods of suspended animation are called
as by insecticides. If the powder becomes trapped aestivation.
between the overlapping scales of the abdomen, Burrowing and nocturnal behavior. Restriction of
then it abrades the waterproof layer, permitting activity to the dark hours reduces the moisture
unregulated water loss and the subsequent death stress experienced by organisms by exposing
of the animal by desiccation. them to lower temperatures and higher atmos-
2. Internal lungs. Exchange of O, and CO, between pheric humidity. During the day, nocturnal ani-
the blood and the atmosphere requires exposure of mals generally remain underground, where hu-
moist, blood-bearing tissues. In a dry atmosphere, midity is much higher. During prolonged
276 CHAPTER 10 Water

drought, they seek out areas of moist soil and re- Various combinations of these moisture adapta-
main there until the end of the drought. tions permit animals of some type to survive in vir-
Drought-resistant stages. Some animals die during tually any terrestrial environment. Only the very
drought periods, leaving drought-resistant eggs hottest and driest environments are unsuitable for per-
that hatch and resume activity when the availabil- manent occupation by animals, and even these may be
ity of moisture increases again. occupied for at least a part of the year. As a conse-
quence, life in the desert is generally abundant and var-
ig Humidity control. The construction of enclosed ied despite what seems to us to be a harsh environment.
nests and covered runways, as exhibited by termites
and some species of ant, permits animals that are
susceptible to moisture loss to exist in dry environ-
ments, with only short forays into fully exposed 10.6 Effects of Moisture on Plant
habitats. Animals such as isopods or amphibians Distribution and Production
generally restrict their activity to high-humidity en-
vironments such as beneath rocks or in rotting logs. Plant geographers of the 19th century noted a close re-
lationship between major types of vegetation, such as
Migration and nomadism. Many birds and animals
deserts, grasslands, deciduous forests, and coniferous
migrate or become nomadic in response to
forests, and major differences in the climate. As one
drought. Migration of birds and animals from
travels from sea level to mountaintops or from the
cold winter areas is in part a response to the re-
equator to the poles, one experiences predictable
duced availability of water as a result of low tem-
changes in temperature and precipitation, and these
perature. Nomadism in desert areas is largely a
are accompanied by fairly predictable changes in the
response to variations in available water either di-
life form (physiognomy) of the vegetation (Figure
rectly from water sources or indirectly through
8-11). Although the relationship between these pat-
the availability of water-containing food items.
terns of vegetation and climate is complex, the correla-
vr Water from food. Some animals depend entirely on tion between the two is good enough that the early
water contained in their food and rarely consume plant geographers used climate to classify and map
free water. The only moisture available to herbi- vegetation, and early climatologists used vegetation to
vores and carnivores for prolonged periods in dry classify and map climates (see Chapter 6). These classi-
grasslands may be the water contained in plants or fications involve various combinations of temperature
in the blood and flesh of herbivores, respectively. (or net radiation) and moisture, and it is certainly not
For animals that consume high-moisture-content possible to separate out the effects of water alone.
food, relatively modest water conservation adap- Some of the more recent ideas about
tations may ensure an acceptable water balance. vegetation—climate relationships have stressed the en-
However, for animals such as the granary weevil, ergy balance approach, in which the climate is defined
which feeds on dry materials such as stored wheat, in terms of precipitation and the availability of solar
there obviously have to be highly efficient water energy to evaporate water. These ideas are centered
conservation adaptations. The moisture content on the availability of water to plants. It has been found
of the granary weevil is approximately 48%, com- that the ratio of precipitation (centimeters of rain per
pared with 9% for its food, indicating very effi- year) to potential evaporation (centimeters of evapora-
cient water acquisition and conservation. tion per year from an open water surface), or P/E
10. Utilization of metabolic water. Insects such as dry ratio, during the frost-free season correlates reason-
wood termites, death watch beetles, and powder ably well in the coterminous United States with very
post beetles survive in environments in which broad patterns of vegetation (Figure 10-4). P/E ratios
there is virtually no free molecular water. Such greater than 100 (P/E - 100; values greater than 100,
animals obtain the necessary water by oxidizing which are found in the southeast, northeast, and
their food to create water metabolically. Some northwest of the United States) indicate that the water
desert animals use a combination of metabolic supply is greater than the potential loss during the
water and other sources. The camel, for example, frost-free season. Values less than 100 indicate the re-
can provide some of its internal water require- verse. P/E ratios of less than 20 correspond approxi-
ment by oxidation of the fat stored in its hump. mately with the occurrence of deserts, whereas the 60
 Effects ofMoisture on Plant Distribution and Production — 277

Palouse
Columbia forest

Laurentian mixed
conifer/hardwood forest

Fasten
eciduou
forest

Southeastern
mixed forest

California chaparral
0 200 400
Shrub miles
steppe Outer coastal
plain forest
Desert Southern
floodplain
forest
Everglades

Figure 10-4 (A) Comparison of the geographic distribution of broad vegetation types (based
on Ecoregions) in the United States (only major divisions of the ecoregion map are shown)
with (B, next page) variation in the P/E ratio shown as (P/E) x 100 during the frost-free pe-
riod. (After Bailey, 1976 and Livingston and Shreve, 1921. (A) Carnegie Institution of Washington Publi-
cation 284. (B) Forest Service, USDA Miscellaneous Publication 1391.)

line approximately separates the short-grass and the vealed predictable differences between a hot, dry envi-
tall-grass prairies (Clark, 1954). ronment and a cool, moist environment (Table 10-9),
Studies in Oregon forests have established rela- and it is possible to characterize plant communities in
tionships between the dominant tree species and gra- terms of these indices.
dients of both moisture (expressed as predawn plant The production of photosynthate per unit of water
moisture stress measured during the driest part of the transpired, or the biomass produced per unit of water
year) and temperature (a physiological assessment of used, is a measure of how efficiently plants use water.
daily soil and air temperatures throughout the grow- Water use efficiency (WUE) has been investigated in a
ing season) (Figure 10-5). These relationships permit wide variety of plants (e.g., Raper et al., 1992; Sands et
a reasonably accurate prediction of potential vegeta- al., 1984; Sands and Mulligan, 1990; Sheriff et al.,
tion using temperature and moisture indices. The re- 1986). Large variations in WUE have been reported
generation potential of different tree species can be between species, between sites of different productivi-
predicted in a similar manner. Studies of a variety of ty, between different families within a species under
other indices of plant-water relationships have re- different climatic conditions, and according to meth-
278 CHAPTER 10 = Water

40

miles

(b)

Figure 10-4 (continued)

es es x
Dry variant ods used to calculate it. Very different values result if
WUE is based on aboveground production or total
production. This variability reduces the value of the
ouglas-fir
WUE concept, although it is still interesting to know
variant how efficiently a species can produce biomass for a
Modal
variant given supply of moisture. Improved nutrition im-
proves WUE, but in many studies, this may reflect
changes in allocation between above- and below-
index
Temperature/growth
E ground production as much as changes in stomatal
gut Ey ies taal control of moisture loss and efficiency of photosyn-
ieee Boones Gh Pease thesis per unit of transpiration loss.
Plant moisture stress, bars Plant moisture stress, bars
(a) (b)

Figure 10-5
4 £8 Nex
Relationship of moisture and temperature
10.7 Interaction Between Moisture
indices to the distribution of forest types (A) and natural and Nutrients
regeneration of tree species (B) in Oregon. (After Zobel et
al., 1974; and Waring et al., 1972. Used by permission of the au- The relationship between moisture and plant produc-
thors.) tivity is extremely complex, and it is very difficult to
 Interaction Between Moisture and Nutrients 279

Table 10-9 Selected Plant-Water Relationships for Various Tree, Shrub, and Herbaceous Plant Species Along a
Gradient from a Hot-Dry Climate to a Cool-Moist Climate

Maximum Potential Maximum Plant

Transpiration, Moisture Stress,
Climate Plant Species g cm” (April-September) atm

Trees
Hot-dry Arbutus menziesii 30.0 30.0
Quercus kelloggii 30.0 30.0
Pinus jeffreyi 25.0 25.0
Pinus ponderosa 21.4 25.4
Pseudotsuga menziesii 21.4 25.4
Abies magnifica 12.5 19.1
Picea engelmannii Wile eas}
Cool-moist Tsuga mertensiana Wo) 19.1

Minor vegetation
Hot-dry Rhus diversiloba 30.0 25.4
Arctostaphylos viscida 30.0 25.4
Xerophyllum tenax 19.5 19.1
Rubus parviflorus 16.8 16.2
Achlys triphylla 16.8 16.2
Viola glabella Asai 12.8

Cool-moist Valeriana sitchensis 7.8 SZ

Data from Waring et al., 1972.

separate the influence of moisture from the influence of these animals promotes the formation of mor humus
other closely related factors. In particular, the availabili- forms (Chapter 11) and lowers rates of nutrient avail-
ty of moisture is very closely related to plant nutrition. ability. Obviously, water stress is inextricably related to
Soil water stress can result in reduced photosynthesis plant nutrition.
because the closure of stomata limits the uptake of The joint effect of moisture and nutrients is seen in
gaseous nutrients from the atmosphere. In extreme characteristic patterns of variation in tree height (or site
drought, the plants may actually be CO, starved and lose index) and volume growth along topographic sequences
biomass. Lack of an active transport stream limits inter- of plant communities from xeric ridge tops downslope
nal movements in plants of inorganic nutrients, amino to moist, fertile, lower-slope situations (cf. Figures 6-10
acids, and the products of photosynthesis, thus interfer- and 6—-11A). Figure 10-6 shows the variation between
ing with metabolism and growth. Lack of soil moisture height growth class and soil moisture for three tree
restricts the mass movement of nuttients into the rhi- species in coastal British Columbia, and Figure 10-7
zosphere, the uptake of nutrients from the rhizosphere, shows estimates of the variation in growth class as a
and the growth of fine roots into soil that contains unex- function of moisture and nutrients for two tree species.
ploited nutrients. The lack of uptake of nitrogen may Forest ecologists have been interested for many years in
lead to protein starvation and a consequent loss of pro- the relative importance of moisture and nutrients in de-
duction. Lack of soil moisture reduces litter decomposi- termining leaf area, total net primary production
tion and mineralization, reducing the levels of soil (NPP), and allocation of NPP between aboveground
nutrients available to plants. The larger soil animals and belowground biomass in forests. This topic was dis-
such as earthworms, beetles, and centipedes, which are cussed in Chapter 4. We will probably never get a de-
important in fragmenting plant litter and mixing it into finitive answer about which of the two factors is most
the mineral soil, are largely excluded from dry soils or important, because they are so interrelated. However,
soils that have a pronounced summer drought. This in the climatically humid environments where most
slows decomposition and permits fungi to dominate forests grow, nutrient availability seems generally to
over bacteria in the decomposer community. The lack of dominate over moisture availability in the regulation
280 CHAPTER 10 = Water

x———_ Douglas-fir often produce much higher densities of very fine roots
| e— — — — Western hemlock than trees, which makes them very competitive for
o—— Western red cedar A moisture, although the extensive mycorrhizal systems
of tree roots help to compensate for their lower levels
of fine roots. In dry areas, failure to control moisture
competition from grasses and other herbaceous plants
or from shrubs may result in plantation failure or very
slow early growth of trees until they produce sufficient
leaf area to shade out these competitors. Increasing
moisture competition from rotation to rotation was
implicated as one of the factors in the rotation to rota-
class
Growth tion decline in growth of radiata pine in dry areas of
South Australia. Herbicide research on dry sites and in
dry climates has demonstrated the importance of com-
petition for soil moisture between tree seedlings and
other plants (see Nambiar and Sands (1993) and refer-
ences therein).
Competition among trees, herbs, and shrubs is
particularly intense when the trees are small, because
Very Moist Very
at that time, all the plants share the same soil layers
dry wet and compete for the same soil moisture. Many of the
Site moisture status tree species that invade grass or shrub communities in
dry areas have tap roots, and as the trees get larger,
Figure 10-6 Relationship between growth class (based on they are able to reach moisture lower in the soil that is
height) and three species of tree growing in the Coastal unavailable to the smaller plants. Once tree roots have
Western Hemlock Zone of British Columbia. The moisture
reached these layers, tree leaf area can increase rapidly
scale is relative, rather than absolute. It represents the range of
(as long as nutrients or nutrient competition is not
edaphically and topographically induced soil moisture condi-
tions found within a particular climatic zone. All three species limiting), reducing the competition from the minor
are adversely affected by excess soil moisture, but especially vegetation for moisture and nutrients in the surface
Douglas-fir. (Data from Klinka, 1976. Used by permission of the soil layers by shading.
author.) It has long been asserted by both scientists and
practitioners in the field of agroforestry that the pres-
ence of trees can actually improve the moisture status
and allocation of NPP. In climatically dry environ- of food crops. This may result from both shading and
ments, the absolute levels of available moisture will the reduction of wind speed that reduce both evapora-
often limit leaf area and NPP (partly through its ef- tion from the soil and transpiration losses from the
fect on nutrient availability) and is probably the dom- crops. It may also reflect hydraulic lifting and “water
inant factor in allocation. For a further discussion of parasitism” (Caldwell, 1990). Trees with deep roots
this complex topic, see Gower et al. (1992), Linder that can access moisture in lower soil layers have much
(1987), Margolis et al. (1995), and Nambiar and higher moisture levels in their surface roots on sites
Sands (1993) and references therein. with dry surface soil layers than the roots of superfi-
cially rooted herbs and shrubs. During dry summer
periods, these well-hydrated surface tree roots may
10.8 Competition for Water Between lose water vapor to the surrounding dry soil, and this
Trees and Minor Vegetation moisture may be available to the roots of the under-
story vegetation. More research is needed to deter-
Farmers and foresters have had experience-based mine the ecological importance of this phenomenon
knowledge of competition for water between food and the extent to which it may explain the observed
crops and trees and between crop trees and minor veg- benefits of trees for food crops in agroforestry systems
etation, respectively, for a long time. Herbs and shrubs in water-limited areas.
 Forestry and the Water Factor 281

Black Spruce, Sub- Lodgepole Pine, Interior

Boreal Spruce Zone Cedar Hemlock Zone
Dry

Moisture
regime

Wet
Poor Rich Poor Rich

Nutrient regime

Site index at age 100 yrs.

Gr. Class SI
Gr. Class SI

~ Nn
n

DNRWNHNHE
OMANI
MOIDNAWN
SP

Figure 10-7 Relationship between tree-height growth (expressed as site index—the average
height in meters of the 200 tallest trees per hectare at some index age (e.g., 100 years)) and
gradients of soil moisture and fertility for black spruce and lodgepole pine growing in differ-
ent biogeoclimatic zones in British Columbia. (A) Black spruce, Sub-Boreal Spruce Zone. (B)
Lodgepole pine, Interior Cedar-Hemlock Zone. This type of graphical presentation is based on the
edatopic grid matrix (or edaphic grid) of Pogrebniak (1930). The grids show isoclines of tree-growth
class for all combinations of moisture and fertility. (After Krajina, 1969. Used by permission of the De-
partment ofBotany, University ofBritish Columbia, and the author.)

10.9 Forestry and the Water Factor Earlier in this chapter, brief mention was made of
the controversy as to the effects of deforestation on re-
Forests grow in the humid regions of the earth, and as gional precipitation, but this is only one aspect of the
a result of interception loss, redistribution by crowns, interaction between forest management and water.
evapotranspiration, and control of snow melt, they Manipulation of forest vegetation results in many
exert a major influence on the water cycle. There is modifications of the hydrological cycle in the treated
considerable interest in what happens to the cycle area. Removal of tree cover reduces interception and
when trees are temporarily removed and the forest eliminates canopy redistribution. It eliminates transpi-
floor is altered by harvest. ration but may also increase evaporation. It influences
282 CHAPTER 10 Water

water input by reducing or eliminating fog drip, and it shading, which has implications for stream tempera-
alters the patterns of snow accumulation and the rate ture, aquatic primary production, and the behavior of
of snowmelt. Movement of water into and through the aquatic organisms. The change in the timing of snow-
soil may be effected if the harvesting has affected infil- melt and the alteration of evapotranspiration affect the
tration rates and soil hydraulic conductivity. This will timing of spring runoff with significant consequences
depend on the degree to which harvesting has re- for summer stream flow and summer soil moisture.
moved or modified the forest floor and compacted the Moisture stress (excess or deficit) plays a major
mineral soil. Increased overland flow and accompany- role in determining the success of artificial regenera-
ing erosion may occur. Total stream flow is increased tion programs. There has been a general lack of ap-
by the reduction of interception and transpiration, and preciation of the resistance to water uptake in newly
peak storm flow will be increased if surface runoff to planted trees, which suffer high mortality if exposed
streams is increased significantly. (as they often are) to moisture deficits before they are
The effects of forest harvesting on the hydrologi- able to develop a new fine-root and mycorrhizal sys-
cal cycle and on regional and global climate has at- tem. The alterations in the hydrological cycle after
tracted a lot of public attention and concern. logging can play a significant role in determining the
Examples of major alterations of stream flow and ac- success or failure of regeneration programs.
companying local or regional effects on erosion, silta- Tree productivity in the developing stand is closely
tion of rivers and lakes, and climate have occurred in related to soil moisture conditions, both because soil
extensively and permanently deforested areas in tropi- moisture dictates the biomass of foliage (Chapter 4) and
cal and subtropical third-world countries: Brazil, because moisture stress affects the photosynthetic effi-
many African countries, and China, for example. The ciency of that foliage. Any forest management activity
impact is much less and is generally temporary in na- that reduces the moisture status of either dry or moist
ture where the rate and distribution of forest harvest- sites will probably be accompanied by a loss of produc-
ing are controlled and forest regeneration is prompt. tion. On an excessively wet site, a reduction in moisture
Research suggests that because of the great year- will normally be highly beneficial for tree growth.
to-year natural variability in stream flow and regime Water is critically important in forestry. It is needed
(the timing of delivery of stream water: the daily and to grow trees, but at the same time, forests are responsi-
seasonal patterns of stream flow), in many forests, ble for maintaining the quality and regime of streamwa-
trees must be harvested from more than a third of a ter that make it a suitable environment for aquatic life
watershed before harvest-related effects on these hy- and a useful commodity for agricultural, domestic, and
drological variables can be identified with any confi- industrial use. In some parts of the world, the socioeco-
dence. However, the extent of forest cover removal nomic value of stream flow regulation, streamwater
needed to produce a significant change in stream flow quality control, and maintenance of fish habitat may ex-
will depend on the character and timing of precipita- ceed the socioeconomic value of timber in many water-
tion and snowmelt events, the size and character of the sheds. In such areas, timber management must take
watershed, and several other factors. Where a signifi- second place to watershed management. This may re-
cant increase in water yield has been demonstrated quire radical modification of harvesting plans or even a
after logging, this declines as the watershed is refor- total restriction on harvesting.
ested, and there may be a net reduction in water yield
as the forest passes through the dense, pre-self-thin- SUMMARY
ning stage that is characterized by maximum leaf area.
If forests are harvested to create a mosaic of forests of The presence of liquid water on the earth is believed to
many different ages across the landscape, then there have been the vital factor that permitted life as we know
should be little effect on the local hydrology. Reynolds it to evolve. The unique physical and chemical properties
of water make it an ideal medium for biological pro-
and Thompson (1988) provided reviews of various as-
cesses, and access to water or the ability to create it meta-
pects of this issue. bolically is indispensable for life. The character of
Increased stream discharge and peak flows produce ecosystems is closely related to the quality and quantity
a variety of changes in the stream environment, such as of water and how these vary through the year.
bank erosion, flooding, and disturbance to fish habitat. Understanding the water factor in ecosystems must
Removal of streambank vegetation reduces stream be based on knowledge of the water cycle: the inputs of
 Study Questions 283

water to, movement and storage of water within, and loss pated effects of the rate and pattern of harvesting on
of water from ecosystems. stream flow and regime in their harvest planning and
The moisture status of a site is not determined simply consider how these effects will influence water quality,
by precipitation inputs, and the availability of water to fish habitat, stream diversity, and streambank stability.
organisms is not determined simply by the moisture sta- The moisture status of the site and the ability of
tus of the site. The physical character and timing of water planted seedlings to maintain a positive water balance
inputs to the site, the nature of the soil and topography, must be identified, and appropriate species, seedling
the physical and chemical state of the water, the nature of types, planting methods, and time of planting must be
the climate, and the adaptations of the organisms all in- used. The consequences of harvest-induced alterations
fluence the availability and ecological efficiency of water. in microclimatic humidity and seedling moisture stress
Both plants and animals exhibit a broad range of for natural regeneration must be considered before
adaptations to either excesses or deficits of water. These harvest if natural regeneration is being relied on to re-
adaptations permit life to continue in almost any type of forest the harvested area. Moisture competition be-
water environment that can be found on Earth, but for tween young trees and other vegetation must be
most species, there is a limited range of moisture condi- evaluated and controlled to achieve regeneration and
tions that can be tolerated, and the water factor is a major early stand growth objectives. Regulation of snow stor-
determinant of plant distribution. The effect is not nec- age and melt rate must be considered in designing
essarily direct, however; water interacts with a wide vari- stand management in snowy areas and the overall fea-
ety of other ecological factors, especially nutrients. tures of the local and regional water cycle must be un-
Although the total quantity of water in the world far derstood to ensure that forest management does not
exceeds human requirements, most of it is unavailable, and alter it in an unacceptable way.
lack of water of suitable physical and chemical quality will
be a significant barrier to human ambitions in the 21st
century. Because forests grow in the humid parts of the
world, they have a major influence on the quantity and STUDY QUESTIONS
quality of streamwater and the timing of its delivery to 1. What determines the ecological effectiveness of water?
streams. Management and harvesting of forests can have a
major impact on all three of these attributes of the water 2. How does vegetation modify the availability of water
cycle, and watershed management is an important compo- to it?
nent of forest management in most forests. Because of the 3. What determines the speed with which precipitation
importance of water in growing tree crops, the forester inputs are lost from an ecosystem?
must understand the hydrological character of forests, the
4, How can we evaluate the likelihood of moisture defi-
adaptations of the forest organisms to water, and how
ciencies for planted seedlings?
management influences this important ecological factor.
5. How does moisture affect the NPP of an ecosystem?
6. How have desert dwelling plants and animals adapted
TAKE-HOME MESSAGE to the moisture conditions that they experience?
Forest managers should consider the water factor in all 7. What is the role of moisture in the distribution of for-
aspects of their work. They must include the antici- est types?
 Soil
The Least Renewable Physical
Component of the Ecosystem

11.1 Introduction Geologists use the term in a broader sense to refer

to all materials produced by weathering of the surface
The terms plant and animal are widely understood by of the earth. To an engineer, soil is ground material
both the lay and scientific communities. The term sod/, that can be excavated by earth-moving equipment
however, has many connotations. Soil scientists have without blasting. To the layman, soil is simply “dirt”—
traditionally considered soil to be the “unconsolidated the substance that sticks to the shoes, dirties the gar-
mineral material on the surface of the earth that serves dener’s hands, or has to be moved while digging a
as a natural medium for the growth of land plants.” hole. Ecological definitions include “any part of the
More recently, they have added to this definition the earth’s crust in which plants are anchored” (Dauben-
requirement that soil can be either mineral or organic mire, 1974) and “the net result of the action of climate
material, that it must be at least 10 cm thick, and that and organisms, especially vegetation, on the parent
it must be “naturally occurring” material. This results material of the earth’s surface” (Odum, 1971).
in the following definition for soil: ‘To a forester, soil is those upper layers of the uncon-
solidated surface of the landscape that provide forest
the naturally occurring, unconsolidated, mineral or plants with the following major necessities of life: water,
organic material at the earth’s surface that is capa- nutrients, a supply of oxygen to the roots, and a firm an-
ble ofsupporting plant growth. Its properties usually chorage. It may be mineral or organic, deep or shallow,
vary with depth and are determined by climatic fac- extensively weathered or only slightly weathered. This
is a somewhat broader definition than that used by the
tors and organisms, as conditioned by relief and
Soil Science Society of America (1973):
hence water regimes, acting through time on geolog-
ical materials to produce genetic horizons that differ
from the parent material. In the landscape, soil The unconsolidated mineral matter on the surface
merges into nonsoil entities, such as exposed, consoli- of the earth that has been subjected to and influ-
dated rock or permanent bodies of water at arbi- enced by genetic and environmental factors such
trarily defined boundaries (L. M. Lavkulich, De- as parent material, climate (including moisture
partment of Soil Science, University of British and temperature effects), macro- and microor-
Columbia, personal communication, 1980). ganisms, and topography, all acting over time
and producing a product—soil—that differs from
The disturbed surface of alogged site or a farmer’s the material from which it was derived by many
field is soil, but displaced materials such as a gravel physical, chemical, biological, and morphological
dump or a mine spoil tip are not. properties and characteristics.

284
 Physical Properties of Soil 285

The Soil Science Society of America definition is atmosphere. ‘Together, these components form a dy-
useful because it reveals the complex nature of the namic system in which the line between living and
phenomenon that we call soil. It is sometimes ex- dead frequently becomes academic. Together, they
pressed in the simpler form (Jenny, 1961): Soil = provide the physical and chemical conditions neces-
(parent material, climate, biota, topography, time) sary for plant life and, consequently, for most forms of
where {| means “a function of” or “is determined by.” animal and microbial life.
Soil thus is not merely the unconsolidated mineral Both ecologists and natural resource managers
material at the interface between the solid geological should learn as much as they can about soil—a critical-
surface of the earth and the atmosphere. It is not pri- ly important but somewhat neglected ecosystem com-
marily a physical, geological phenomenon. It is the com- ponent. The acquisition of a complete understanding
bined result of atmospheric and biological processes of soil is a lifetime activity, and the present chapter can
acting on geological materials over a period of time. serve only as an introduction to one of the most im-
The term solum is used to refer to that part of the portant aspects of both ecology and resource manage-
unconsolidated surface material that is soil. Between ment. Readers who are interested in soils are referred
the solum and the bedrock is relatively unaltered ma- to such books as Armson (1977), Brady (1984), Hunt
terial that has traditionally been referred to as parent (1972), Jenny (1980), Pritchett (1979), Remezov and
material (the unconsolidated and more or less chemi- Pogrebnyac (1969), Ulrich et al. (1981), or any stan-
cally weathered mineral or organic matter from which dard soil science text.
the solum or soil is developed by pedogenic pro-
cesses). The solum, however, may have developed in
transported materials that are unrelated geologically 11.2 Physical Properties of Soil
and pedologically to the underlying materials, which
cannot therefore legitimately be called “parent” mate- ‘Texture: The Particle Size
rial. The alternative term nonsoil, has been suggested Characteristics of Soil
for surface materials that do not meet the accepted de-
Soils are made up of various combinations of organic
finition of soil.
matter and mineral particles of various sizes organized
Soil is the basic resource of foresters and agricul-
in characteristic patterns. Variation in the relative
turists; trees, cows, and cabbages are merely the crops.
abundance of different-sized particles in the <2 mm
A forest or a farm is a landscape unit with a climate
diameter mineral material results in differences in the
and soil that create the potential for organic produc-
texture of soils. Three major size classes in the
tion that can take the form of the biota of an unman-
<2 mm material are recognized—sand, silt, and clay—
aged ecosystem or a harvestable economic crop.
the size limits of which are shown in ‘JTable 11-1.
Destroy the soil and that potential is lost or dimin-
ished. Careful harvesting of these crops does not “de-
stroy” the forest or farm; it merely creates temporary
Table 11—1 American and International Classifications
changes in ecosystem structure and function until the
of Soil Particle Size
tree or crop community is reestablished. If, however,
the harvest is accompanied by widespread loss or cata- Particle Size, mm
strophic alteration of the soil, then one can indeed say
American system
that the forest or the farm, at least the present form Clay <0.002
thereof, has been “destroyed” for the period of time Silt 0.002-0.05
that it will take for natural or human-accelerated Very fine sand 0.05-0.1
processes to return the ecosystem to a productive con- Fine sand 0.1-0.25
Medium sand 0.25-0.5
dition (see Chapter 17).
Coarse sand 0.5-1.0
Soil is one of the three major components of the Very coarse sand 1.0-2.0
ecosystem. To understand it requires every bit as much
International system
effort as that required to understand the other two (at- Clay <0.002
mosphere and biota). It is a highly complex assem- Silt 0.002-0.02
blage of geological materials, dead organic matter, Fine sand 0.02-0.2
living roots, animals and microbes, soil water, and soil Coarse sand 0.2-2.0
286 CHAPTER 11 Soil

silty clay loam

> S > Ss
y > ¥

% silt

Figure 11-1 ‘Triangular summary of the sand, silt, and clay content of the major textural
classes recognized in North America. The relative frequencies of these three particle-size classes
can be read from the diagram by projecting scale lines from each of the scales. The example shown,
which has a clay texture, contains 60% by weight of clay, 20% by weight of silt, and 20% by weight
of sand. (After Armson, 1977. Used by permission ofthe University ofToronto Press and the author.)

These particle sizes are combined in various propor- than 2 mm. It also reflects the historical fact that farm-
tions to produce the various textural classes: sand, ers recognized the importance of understanding soils
sandy loam, silt loam, clay loam, and clay (note that long before most foresters did. In contrast to agricul-
sand, silt, and clay can refer both to a particular size tural soils, particles larger than 2 mm occupy a consid-
class and to a textural class). Figure 11-1 shows how erable portion or even the majority of the soil volume
these and other textural classes are defined in North in many forest soils, and it is not uncommon to find
America, and Table 11-2 gives a guide to the field trees growing directly on fractured bedrock or on or-
recognition of the six basic classes. ganic layers that lack a substantial mineral particle
Particles larger than 2.0 mm in diameter (gravel (2 component. Consequently, the use of texture to indi-
to 20 mm), cobbles (20 to 200 mm), and boulders (200 cate soil fertility is not always as useful in forestry as it
to 2000 mm)) have traditionally been excluded from is in agriculture. (Field guides for the description
textural classifications. This reflects the preoccupation of forest ecosystems, including soils, as used in Brit-
of soil science with agricultural soils, in which parame- ish Columbia can be found at www.for.gov.be.ca/
ters such as aeration, soil moisture, root penetration, research/becweb. Descriptions of other systems of soil
soil structure, and fertility are generally most closely classification can be accessed on the Internet; e.g., the
related to the relative frequencies of particles smaller U.S. system of soil classification—7th Approximation.)
 Physical Properties of Soil = 287

Table 11-2 Field Determination of Soil Texture: Features That Are Useful in Hand Texturing
Soil Texture Features

Sand Loose and single-grained. Individual grains can be seen and felt. Squeezed in the hand when dry, it will fall
a
apart when pressure is released. Squeezed when moist, it will form a cast that will crumble when touched.
Sandy loam Contains enough silt and clay to make it somewhat coherent. Sand grains are readily seen and felt.
Squeezed when dry, the cast will readily fall apart. Squeezed when moist, the cast will bear careful handling
without breaking.
Loam Mellow, with a somewhat gritty feel, yet fairly smooth and plastic when moist. Squeezed when dry, the cast
will not break if handled carefully. Squeezed when moist, the cast can be handled freely without breaking.
Silt loam When dry, it may appear cloddy, but the lumps are easily broken. When pulverized, it feels soft and floury.
When wet, the soil puddles. Casts formed of either dry or moist soil can be readily handled without break-
ing. When moistened soil is squeezed between thumb and finger, it will not “ribbon” but will form flat “pas-
try flakes.”
Clay loam When dry, it forms hard lumps or clods. When moist, it can be squeezed to form a thin ribbon, which will
break readily, barely sustaining its own weight. When moist, the soil is plastic and will form a cast that will
take much handling.
Clay Forms very hard aggregates when dry. When wet, it is plastic and sticky. Moist clay can be pinched out be-
tween thumb and finger to form a long flexible ribbon. Note that some clays are friable and lack plasticity in
all moisture conditions.

Armson, 1977.

NO
Does the soil ——> _SAND
form a cast?

YES

NO NO
Does the soil — Does the soil —————>>- The soil forms a
form a form a weak cast
strong cast weak cast that does not
that allows that allows allow handling.
ready handling? careful handling?

YES

O
The soil barely Oe Does the soil ae Does the soil <— Does the soil — Does the soil Does the soil —» The soil does
begins to form a short form a thin form a thin form a thin flake rather not ribbon.
ribbon and thick ribbon that ribbon that ribbon longer than ribbon?
ribbon? © breaks readily and holds its own than 7.5 cm?
barely supports weight?
own weight?

YES YES

SANDY SANDY LOAM

CLAY LOAM

NO : A
The soil feels—— Does the soil * Does the soil —*> The soil has Does the soil —*> The soil feels Does the soil —* The soil feels
moderately feel smooth feel smooth? substantial feel very floury with feel grainy grainy with
grainy. and floury? graininess. floury? slight graininess. with a small a considerable
amount of amount of
YES YES YES YES YES YES floury material? —_floury material.
YES {yES

CLAY LOAM SILTY CLAY SILTY CLAY SANDY CLAY SILT SILT LOAM LOAMY SAND _ SILTY SAND
LOAM
 —
288 CHAPTER 11 Soil

Structure: The Architecture of Soils that have colloidal properties. Roots exude organic
substances, which, either alone or in association with
In most soils, groups of individual mineral particles microorganisms, bind particles into aggregates. The _

are bound together (cementation) into aggregates, or formation of root channels opens up the soil, and the
peds: units of soil structure formed by natural process- churning action of roots helps to break up massive ma-
es. The binding agents may be colloidal clay or organ- terials. Animals mix soil, which promotes the incorpo-
ic matter, but other chemicals can also be involved.
ration of organic matter deep into the mineral soil.
The size and shape of the aggregates vary greatly, Their burrowing opens and loosens the soil, and the
from spheroidal aggregates as large as 10 mm in diam- feces of many soil animals become small peds.
eter (called crumbs if they are porous and granules if The structure of soil is analogous to the architec-
not) up to blocks or prism-like units as large as 50 mm ture of a building. It is the spatial arrangement of the
in diameter. Where there is no such aggregation, the component parts so that they function as a system that
soil is described as being structureless or massive. Struc- fulfills certain roles. Soil architecture facilitates the
ture may be a fairly permanent feature, but in some movement of water and gases, the activity of soil or-
types of soil (e.g., certain clays), it can change accord- ganisms, and a myriad of soil processes without which
ing to the prevailing soil moisture condition. soil could not provide life-support functions to the
Structure develops as a result of physical and chem- plant, animal, and microbial communities.
ical processes in the soil, often in association with bio- When a house is knocked down, all the parts are still
logical activity. Wetting and drying cycles in soils that there but it is no longer a house because it has lost its
contain clays that shrink on drying can develop a blocky architecture. Without the appropriate spatial arrange-
or prismatic structure. Alternate freezing and thawing
ment of its parts, it no longer functions as a house. Sim-
can break such large aggregates into smaller ones or
ilarly, a soil that has lost its architecture because of
initiate development of structure in massive materials.
erosion, fire, or compaction no longer functions as a
The presence of colloidal clay or organic matter binds
soil until soil structure has been reestablished.
larger particles (silt and sand) together to give a crum-
by or granular structure, although the binding efficien-
cy of the cementing material may depend on chemical Porosity: The Distribution of Space in Soils
factors. Where there is an abundance of monovalent Porosity is a measure of the amount of pore space in a
sodium (Na’*) in the soil, clay particles tend to disperse soil, or the volume of the soil not occupied by solids. It
and peds will break down or not form. Where divalent is a critically important parameter of soil because it in-
cations such as calcium (Ca**) dominate in the soil, for- fluences the movement of water and gases, which in
mation of stable aggregates is promoted by the process turn determine the activity of roots and soil organisms.
of flocculation. These doubly positively charged ions ef- The porosity of structureless (massive) soils is much
fectively neutralize the negative charges on clay parti- lower than that of well-aggregated soils because the
cles that create mutual repulsion between the particles latter have abundant larger pores between the peds.
and resist aggregation. Flocculation is a less permanent However, total porosity may be less important than
condition than aggregation by cementation but seems the distribution of the pore space. If all the space oc-
to be a prerequisite for it and is important in fine-tex- curs as large pores, as it does in gravel or coarse sandy
tured soils. In very acid or very alkaline soils, where soil, then water will drain freely and the soil will be
there is little free calcium or magnesium or where subject to drought. Conversely, if all the pores occur as
there are high levels of sodium and to a lesser extent minute spaces between clay particles, then the move-
potassium, flocculation is inhibited; such soils tend to ment of gases and water will be extremely slow, and
be structureless or massive. Colloidal materials (clay plants that grow in the soil will experience waterlog-
and organic matter) tend to be washed down to deeper ging and oxygen deficiency when the soil is wet and
layers of the soil under such circumstances. Coarse- will have difficulty in taking up strongly held water as
textured soils (sands and gravels) also have little struc- the clay-rich soil dries. The ideal pore distribution is
ture and are sometimes termed single-grained soils. that which retains sufficient water and yet permits ad-
Organic matter, the presence of roots, and the ac- equate diffusion of oxygen and CO, and movement of
tivities of soil animals are important in the develop- water to satisfy the requirements of desired species of
ment of soil structure. Many types of organic plants, soil animals, and soil microbes. Generally, this
compound can act as a binding agent, especially those is found in a well-structured soil with most of the small
 Physical Properties of Soil 289

pores within and the larger pores between the aggre- Aeration: Gaseous Exchange in Soil
gates. ‘Total pore space in a poorly structured soil may
be as little as 35%, whereas a well-structured soil with Organisms that live belowground exchange gases just
the same texture may have as much as 65% pore space. as their aboveground counterparts do. Most soil or-
ganisms require an external supply of oxygen and re-
lease CO,, which must diffuse out of the soil, be
carried away dissolved in water, or otherwise be ab-
Consistence: The Physical Resilience of Soil sorbed if toxic accumulations are not to occur. Inade-
Soils vary in the degree to which the particles or peds quate gaseous or dissolved oxygen results in the
stick together. Soil consistence refers to the stickiness of reduction of oxidized soil chemicals such as iron and
soils and their ability to resist deformation under sulfur (which frequently have low solubility) by anaer-
stress. A soil that can be rolled into a cylindrical shape obic soil microbes. This leads to the production of
when moist is said to be plastic. This is a feature of materials such as ferrous iron and hydrogen sulfide
clay-rich soils, which is used to assess the textural class (which have high solubility), which are toxic to the
of a soil in the technique of hand texturing (cf. Table roots of many plants. Figure 11-2 illustrates the effect
11-1). If the clay becomes too wet, then it loses its of reduced oxygen concentrations on height and root
plasticity and becomes sticky. When dry, it is hard. growth in seedlings of a number of tree species. The
The consistence of soil is important because it helps to root surface area of black spruce seedlings (2 years old)
determine soil stability, erodibility, and trafficability was found to be more sensitive to soil oxygen levels
under various moisture conditions. than was height growth.
Soil aeration is determined by total soil porosity,
the distribution of the pore space, and the proportion
of the space that is filled with water. Soils with high
Bulk Density: The Mass of porosity can be poorly aerated if constantly wet,
Soil per Unit Volume whereas soils with medium to low porosity can be rea-
The structure, texture, and porosity of soils, together sonably well aerated if not too wet and if the pores are
with their organic matter content, combine to deter- of medium size. Even soils with good drainage, high
mine the bulk density of a soil: the kilograms of soil ma- porosity, and good pore distribution can become
terial per cubic meter of soil volume. Measurement of poorly aerated if the structure of the upper few mil-
bulk density is widely used to estimate porosity. limeters of the soil is damaged by such things as heavy
Expressed in units of Mg m® (mg/m*), the bulk rain or mechanical compaction. A thin layer of soil in
density of clay, clay loam, and silt loam surface soils which structure and porosity have been lost can effec-
normally ranges from 1.00 to as high as 1.60, depending tively block gaseous exchange between the soil and the
on their condition. Sands and sandy loams have values atmosphere. A soil whose surface structure has been
of 1.20 to 1.80. Very compact nonsoils may have bulk lost as a result of mechanical impacts when wet is said
densities of 2.0 or even higher (Brady, 1984). Forest to have been puddled.
floor bulk densities also vary. For example, values of
0.12 to 0.16 g cm™® have been reported for yellow
birch—red spruce stands in the Adirondack Mountains Temperature: Soil Warmth
of New York (McPhee and Stone, 1965). A 500-year-
old subalpine forest in coastal British Columbia was Chapter 8 presented a brief summary of soil temperature
found to have mean values of 0.14 to 0.18 (Kimmins,
relationships, and the reader may wish to refer to the ap-
unpublished data). Much higher values (0.27 for a forest propriate section of that chapter before proceeding.
‘Temperature is an important parameter of soil be-
in New Hampshire; Hoyle, 1973) and much lower val-
ues (0.056 for undecomposed moss peat; Boelter, 1964)
cause it influences the activity of roots and soil organ-
isms, rates of decomposition, and nutrient and water
can also be found. One reason for this variation is the
uptake. Organic materials have a lower thermal con-
variation in biomass of live small and fine roots (<5 mm
ductivity and volumetric heat capacity’ than do inor-
diameter) in forest floors and in the extent to which
they were removed before bulk density was calculated.
Forest floors in areas of heavy winter snowpack tend to ‘Volumetric heat capacity: the quantity of heat required to raise the tem-

have higher bulk density than those in snow-free areas. perature of 1 cm’ of the material by 1°C.
290 CHAPTER 11 Soil

Black spruce
100 10

Thome 2
length ;

50 nN
Scots pine —~.—
Root surface
Jack pine —+Q}-— area
length
Shoot
(cm)
(cm?)
surface
Root
area

maximum
of
elongation
%
Root
a
as

0 10 20 °0 10 20
% concentration of oxygen in rooting medium
(a) (b)

Figure 11-2 Effect of reduced oxygen levels on the growth of shoots and roots of several
conifers. (A) Variation in the rate of root elongation in seedlings of three species growing in solu-
tions containing various concentrations of oxygen. (After Leyton and Rousseau, 1958. Redrawn by per-
mission ofJohn Wiley & Sons, Inc.) (B) Variation in shoot length and root-surface area of black spruce
seedlings under simulated out-planting conditions and varying oxygen tensions. (After Armson, 1977.
Redrawn by permission ofthe University of Toronto Press.)

ganic materials, and consequently the forest floor serves 11.3 Soil Water
to modify greatly the temperature of the underlying
mineral soil. The presence of a forest floor reduces Although water relationships are, strictly speaking, a
daily temperature variations and retards the spring physical characteristic of soils, they are sufficiently im-
warming and autumn cooling of the mineral soil. portant that the subject is treated in a separate section.
Soil temperature regimes are influenced by soil Some aspects of soil water were discussed in Chapter 10.
porosity and texture. Thermal conductivities generally Water can exist in soil as a liquid, a vapor, or a
increase with increasing particle size, so sands have solid. Water vapor can be significant in dry soils with
higher conductivities, warm up faster in the spring, fluctuating temperatures because vapor movement
and cool down more rapidly in the fall than do clays. and condensation can appreciably effect the availabil-
Denser materials have higher conductivities than ity of liquid water. Water vapor diffuses upward from
those that have a lower bulk density. lower layers of the soil and condenses in surface layers
Water content has an important effect on soil tem- that have undergone nocturnal radiative cooling.
perature. Wet soils have a higher thermal conductivity Plants may also play a role in moving water from
than dry soils, but they generally warm up and cool down deeper, moister layers to surface drier layers in a
more slowly because ofthe high volumetric heat capacity process called water lifting (see Chapter 10). Ice or
of water and because much of the incoming solar energy frost can have an important effect on soil water con-
may be used in evaporating water at the surface. tent and water movement in soils in cold areas. Frost
Surface temperature of exposed soil is largely de- churning can affect soil structure and development,
termined by the balance of incoming and outgoing ra- whereas frozen soil restricts the infiltration of water,
diation. By shading the soil, vegetation exerts a thus reducing the total amount of water in the soil
dominant influence on the radiation balance at the soil later in the season. Ice influences soil strength, and
surface. Vegetative cover results in a reduction in max- some forests can be logged only in the winter, when
imum and an increase in minimum soil temperatures, the soil is frozen. These points notwithstanding, the
although the latter may be due more to the formation most important form of soil water is liquid.
of a surface litter layer than to shading. The influence Aspects of soil water that have attracted most at-
of soil color was mentioned briefly in Chapter 8. tention are its movement and storage in the soil. Both
 Soil Water 291

of these are in turn determined by the energy relation- As soon as the rain stops, water in the larger pores will
ships that exist between water and soil particles. Water drain downward toward the water table because the
moves from areas of higher energy to areas of lower low {psi},, in these pores cannot prevent the water
energy and is held within the soil when the energy of moving from soil with higher {psi}g (the surface lay-
the water is less than the energy of the soil holding it. ers) to soil with lower {psi}g (the lower soil layers).
Water in the soil has three types of energy (gener- The water that is lost is called gravitational water.
ally called water potential). Water will remain in those pores for which the effect
of {psi}, is greater than that of {psi}g The soil is then
1. Osmotic potential ({psi}o) is a measure of the con- said to be at field capacity: the condition that occurs in
centration of dissolved chemicals (solutes). These nonvegetated soil 2 to 3 days after a rainfall that satu-
lower the vapor pressure of water, reducing evapo- rated it. It is an imprecise term because water drainage
ration. There will be a net movement of water continues for much longer than 2 to 3 days. This is es-
vapor from water with low solute concentrations pecially true in a fine-textured soil (Figure 11-3) be-
(high {psi}o) to water with high solute concentra- cause as the soil becomes drier, capillary forces begin
tions (low {psi}o). This type of water potential con- to predominate and water drains more slowly. The
tributes relatively little to the movement of soil term is most useful for coarse-textured soils in which
water except in saline and desert soils, but it is of {psi}; is less significant.
considerable importance in plant—water relations. The majority of the water in a saturated, coarse-
textured soil is gravitational water, and it is lost by the
2. Gravitational potential ({psi}c) is the potential ener-
time field capacity is reached (Figure 114A). In fine-
gy that the water has by virtue of its position or el-
textured soil, there may be little gravitational water, so
evation within the soil. Water at the surface of the
at field capacity, this type of soil retains a much higher
soil has a higher ({psi}g) than does water 1 m be-
percentage of its saturated water content than does a
neath it. The potential is zero when the water
coarse-textured soil. However, most of this water is
reaches the surface of the water table. ({psi}q) is the
held in such small capillaries that the {psi}, is greater
energy that is generally (but not always) responsible
for the drainage of water down through the soil.
3. Matric potential ({psi}y) is the energy of water asso-
ciated with the forces of capillarity and adsorption. It
is a negative potential (and is therefore sometimes |cravitationl water
called matric suction), and it generally opposes the Field capacity
action of {psi}g and {psi}. Capillary forces occur in
the smaller pore spaces; they act to lower the ener-
gy level of the water and thus hold it in the pores.
Adhesive forces that result from the molecular Gravitational water
structure of water bind water molecules to the soil
particles with which they are in contact. In coarse- Water
soil,
of
%content
Field capacity
textured soils, the surface area of pore walls avail-
able for adsorption and the capillarity of these
pores are negligible compared with those of fine- 0 x 3 5 10 15 2C
Days since saturation
textured soils, and the importance of {psi}, in de- Soil
termining water movement is negligible or sizable, saturation
respectively. The importance of {psi}, changes as
Figure 11-3 Loss of water from two different soils after sat-
the soil dries out, because as it does so, the remain-
uration. Water content at field capacity (defined as water content
ing water is restricted to smaller and smaller pores, after 2 to 3 days of drainage after saturation) is shown. Water in
in which {psi} is increasingly significant. excess of field capacity is gravitational water. It is, obviously, easier
to distinguish gravitational from capillary movement in the
The amount of water stored in soil is determined coarse- than in the fine-textured soil because of the marked
largely by the hydraulic conductivity of the soil. Con- change in slope of the graph in the former, before and after the 2-
sider a saturated soil in which all pore spaces are filled to 3-day time period. (Hewlett and Nutter, 1969. Redrawn by per-
with water (e.g., at the end of a prolonged heavy rain). mussion ofthe University of Georgia Press and the authors.)
292 CHAPTER 11 = Soil

1300

1000

saturation
at Elevation,
m ponderosa pine-savanna
500 forest
of
categories
soil
water
of
magnitude
Relative
three
the 300
Gravel Sand Loam Clay
Sand Sandy Loam Silt Clay Clay
loam loam loam Soil textural class
Soil textural class

Figure 11-4 Relationship between the availability of water to plants and soil texture. (A) The
distribution of soil water between the three main categories of soil water varies in soils of different tex-
ture. Very fine- and very coarse-textured soils have less available water than does medium-textured soil.
(After Hewlett and Nutter, 1969. Used by permission of the University of Georgia Press and the authors.) (B)
Relationship between major climatic climax vegetation types, soil texture, and elevation in south-central
British Columbia. The mean tree heights of dominant trees at age 100 years are also shown. The vege-
tation type and the tree height growth vary as soil texture changes, which reflects texture-controlled soil
moisture conditions. The interaction between elevation and soil texture results from elevation-depen-
dent precipitation changes and the effects of texture on the availability of precipitation to plants. (After
Brayshaw, 1970. Used by permission ofthe British Columbia Provincial Museum and the author.)

than the plant potential (the ability of plants to with- Water flow in unsaturated soils behaves differently.
draw water from the soil) and the water is unavailable As soil dries out, hydraulic conductivity decreases be-
to plants. The {psi},, at which soil water becomes un- cause the water is flowing in smaller and smaller capil-
available (referred to as the permanent wilting point) is laries (rate of flow varies as the fourth power of the
considered to be approximately —1500 kPa (a soil radius of the capillary, so a reduction of radius by 50%
water tension of approximately +15 atm). A fine sand reduces flow rate 16-fold). Also, as water leaves the
will have approximately the same quantity of avail- larger pores, they fill with air, which provides a barrier
able water at field capacity as a clay, even though it to the flow of water remaining in the isolated small
contains only approximately half as much water. pores that connect the larger ones. As a result, water
Loamy soils can contain nearly twice as much avail- movement is rapid in coarse soils when they are satu-
able water as clays, although they contain significant- rated but declines dramatically and becomes very slow
ly less total water. These generalizations will as the soil dries. Water movement at saturation is slow-
obviously be modified by variations in soil structure. er in fine-textured than in coarse-textured soils, but as
An example of how variations in soil texture interact- the soil dries, the rate decreases more slowly and re-
ing with climate (expressed as elevation) can affect mains appreciable even when the soil becomes fairly
vegetation through effects on soil moisture availabili- dry. This means that although a coarse soil initially
ty is shown in Figure 11-4B. loses water rapidly by gravity and evaporation after the
The movement of water in saturated soil approxi- termination of a rainstorm, losses decline rapidly and
mates Darcy’s law: the remaining water may be held for a long time, main-
taining a moist soil condition. In contrast, although a
Q=kA X A/L fine-textured soil does not lose water as rapidly at first,
losses may continue until the soil becomes very dry. It
where Q is the volume of flow per unit time, & is the hy- also means that movement of water to the zone of de-
draulic conductivity, A is the cross-sectional area of soil pletion around a root is much slower in a coarse-tex-
through which water is flowing, and H/L is the hydraulic tured soil than in a fine-textured soil, rendering such
gradient (hydraulic head per unit of column length). coarse soils “drier” to plants than they really are.
 Chemical Properties of Soil 293

11.4 Chemical Properties of Soil than the original primary mineral. When there is
abundant CO, present in the soil solution, hydration
Soils have characteristic chemical properties that are together with carbonation can alter feldspars to clay
the result of weathering of their mineral components, minerals.
decomposition of organic material, and the activity of
plants and animals. After a discussion of weathering,
CaAl,Si,03 - 2NaAISi;0g+ 4H,CO3+ 2(nH,0) >
we examine the following soil properties: cation and Sodium-calcium feldspar Carbonic Water
anion exchange capacity, pH, and the forms and avail- acid
ability of macronutrients.
2Al, (OH), Si,O,9nH,O+ Ca(HCO, )o ae 2NaHCO,
Clay Calcium Sodium
Weathering bicarbonate bicarbonate

Weathering of mineral materials involves both physi- (Hunt, 1972)

cal and chemical processes. Physical weathering is the
breakdown of rocks into increasingly smaller particles
Oxidation and reduction are both important types of
by mechanical action during transportation by water,
chemical weathering. Oxidation is the loss of electrons
wind, or ice or as the result of temperature fluctua-
from a molecule; reduction is the gain of electrons.
tions, particularly freeze-thaw cycles. It is more im- These processes not only alter the chemical form of the
portant in the preparation of mineral particles for
material but also render the material more susceptible
soil-forming processes than in the actual development to other weathering reactions such as hydrolysis. Re-
of soil, a process that is dominated by chemical weath- duction of some insoluble substances (such as ferric
ering. Chemical weathering results in the release of el- iron to ferrous iron: Fe** > Fe*) makes them more sol-
ements from soil minerals and the alteration of the uble; oxidation reverses the process. Oxidation—
chemical form of the residual materials. reduction weathering is particularly important in soils
Chemical weathering involves a variety of where there is a seasonal fluctuation between saturated
processes. Simple exposure to water may result in and nonsaturated conditions. Such soils, which have a
solution or hydrolysis, a reaction in which a compound mottled, gray-brown coloration as a result of the alter-
reacts with water to form two different molecules. nating oxidation and reduction, are called gleysols; the
For example, water combines with some primary process by which they are formed is called gleyzation.
minerals such as feldspars to yield salts and complex Loss of silica from primary minerals is an impor-
compounds that contain aluminum and silicon. These tant type of chemical weathering under certain cli-
compounds can recrystallize into clay minerals (alu- matic conditions (hot and moist). This type of
minosilicates). weathering (/aterization) is important in the formation
of Jaterites, a type of soil found in some parts of the
KAISi,O, + HO = HAISi,O, + KOH tropics and in some wet and warm temperate areas.
{decomposes Laterization results in the formation of plinthite, a
nonindurated (the soil scientists’ term for not physi-
H,SiO,+ Al,O; -3H,O
Silicic Gibbsite cally hardened) mixture of iron and aluminum oxides,
acid clay, quartz, and other materials that commonly oc-
(Armson, 1977)
curs as red soil mottles. Plinthite is usually arranged
in platy, polygonal, or reticulate patterns through the
The intensity of hydrolysis increases with increas- soil matrix. It changes irreversibly to ironstone hard-
ing temperature as long as moisture is adequate. It is pans or irregular aggregates on exposure to repeated
therefore a more active process in the tropics than in wetting and drying. Removal of vegetation from lat-
temperate regions, and tropical soils of comparable eritic soils (oxisols) in the tropics exposes plinthite to
age tend to be more deeply and completely weathered heating and drying, and this has frequently resulted in
than do their temperate counterparts. Water can also the development of serious land management prob-
combine with minerals in the process of hydration. lems. An example of laterization is the conversion of
This results in the formation of secondary minerals, the dark-colored mineral biotite to iron oxide, clay,
which are softer and more easily weathered physically salts, and silica.
294 CHAPTER 11 — Soil

2KMg,Fe(OH), AISi;0;9+ O+ nH,O => a clay-humus complex, have the ability to absorb and
Biotite hold positively charged ions (cations). This gives soil
that contains clay or organic matter a cation exchange
Fe,0, v H,O +Al,(OH), Si,Oj9 ° nH,O +2KHCO,3+
capacity (CEC).
Iron oxide Clay Potassium
bicarbonate The CEC of soil clay minerals and organic matter
varies widely. Measured in units of milliequivalents’
4Mg(HCO;),+ 2Si0,+5H,0 (Hunt, 1972)
per hundred grams of dry material (mEq per 100 g),
Magnesium Silica | Water
bicarbonate
CEC values vary from as low as 3 to 10 for kaolinite,
15 to 40 for illite, and 30 for vermiculite to 80 to 120
Another category of weathering that is thought to for montmorillonite (Brady, 1984). Values for organic
be particularly important in forest soils is chelation, or matter vary from 50 to 250, but because the bulk den-
the formation of organic matter-element complexes: sity of organic matter is much lower than that of clay,
the reaction between metallic ions and certain organic organic matter may have a smaller total CEC than
compounds that can act as electron donors. Metallic would the same volume of clay. The CEC generally
elements such as iron, aluminum, and copper are fre- declines as the soil becomes more acid, but the rela-
quently held and moved in the soil as organic chelates tionship depends on the material. The CEC of humus
or complexes. It is believed that chelation can lead to is more affected than that of clays.
the decomposition of silicate minerals. Chelation re- All cations can be held on the CEC, but they vary
duces the toxicity of substances such as iron and cop- in how firmly they are held. Generally, the greater
per, increases their mobility in the soil, and may the number of electrical charges, the firmer the ab-
facilitate their uptake by plants. sorption, so Ca’* ions are held more firmly than K*
Weathering is important for a variety of reasons. ions. This does not apply to ions such as Al** and
By physical comminution, smaller particles with in- Mg** if they are hydrated, in which condition they
creased surface area are produced, a prerequisite for are held less firmly than unhydrated K* and NH,’.
the formation of mineral soils. Chemical weathering Also, the density of negative charges on the clay or
releases nutrient elements required by plants and mi- organic matter affects which ion is held. The compo-
crobes and results in the formation of clay minerals. sition of the ionic population associated with the
The most abundant primary minerals are the silicates, CEC 1s also influenced by the chemical composition
the most common of which are the feldspars. As these of the soil solution. Where this has high concentra-
weather, they produce one of several different types of tions of any particular ion, that ion will tend to dis-
clay, including montmorillonite, a type that shrinks place other ions on the exchange sites, irrespective of
and swells as its dries and wets; illite, a nonexpanding their relative absorption or replacing powers. For ex-
clay; and vermiculite and chlorite, clays that expand ample, soil solution that contains large numbers of
but less so than montmorillonite, kaolinite, and hal- hydrogen ions will result in the displacement of
loysite. In all clays except kaolinite and halloysite, much of the population of other types of ions from
there is considerable replacement of aluminum by the CEC. The degree to which basic ions occupy the
iron or magnesium. In some of the clays, elements CEC is called the percent base saturation: the sum of
such as potassium can fit between the layers of the the “exchangeable” cations, other than hydrogen and
crystal lattice and become firmly bound. Clay exerts a aluminum, on the CEC expressed as a percentage of
dominant influence over the chemical properties of the total CEC.
mineral soils and also on their physical properties Surfaces of roots have negative charges that attract
(porosity, structure, texture, drainage, etc.). Perhaps and promote the absorption of cations, and organic
its major contribution to soil chemical properties is substances seereted by the roots create a CEC in the
the phenomenon of cation exchange. rhizosphere (soil zone immediately around roots). Liv-
ing roots can therefore provide a CEC in a soil in ad-
dition to that provided by clay and humus.
Cation and Anion Exchange Capacities
Both clay and colloidal organic matter are negatively ‘Milliequivalent is the amount of an element or compound that will com-
charged and therefore can act as anions. As a result, bine with or replace 1 mg of hydrogen. It equals the atomic weight in mil-
these two materials, either individually or combined as ligrams divided by the valence. MEq per 100 g = cmol per kg.
 Chemical Properties of Soil = 295

Soils also have an anion exchange capacity (AEC), the organisms may cause alterations in their chemical
but this is generally much less than CEC. Some clay environment that lead to the release of H* ions (e.g.,
and organic colloids are amphoteric (they can carry ei- organic nitrogen ammonium ions > nitric acid), and
ther negative or positive charges), and any iron or alu- acidity is formed when respiratory CO, combines with
minum oxide present provides additional positive water to produce carbonic acid. Acids may also be pro-
charges to which anions can be attracted. The AEC is duced during the decomposition of soil organic mat-
pH dependent; generally, it is greater in acidic than in ter; litter of most conifers and of mosses such as
neutral or basic soils. Sphagnum becomes very acidic as it decomposes.
Some clays can trap K* and NH, ions in intermi- Soil alkalinity originates largely from salts, such as
cellar spaces, but this involves more than simple cation calcium carbonate, that hydrolyze to form strong
exchange. bases. These salts originate from the weathering of
rocks such as limestone or dark-colored igneous rocks,
which yield basic compounds as they decay. Alkalinity
Soil Reaction or pH: Acidity and Alkalinity can also originate from the importation of basic salts
Soil reaction refers to the activity of hydrogen ions in in drainage water, which subsequently evaporates.
the soil. It is measured in units of pH, which is the log- Highly alkaline (solonchak) or saline (solonetz) soils
arithm of the reciprocal of the hydrogen ion activity. are frequently formed by this process in arid regions.
There is a 10-fold difference in concentration be-
tween successive units (e.g., pH 3 is 10 times more
acid than pH 4 and 100 times more acid than pH 5). Availability of Nutrients
At 22°C, pure water will contain | g of hydrogen ions The nutrients required for plant growth are present in
per 10 million liters, giving a concentration of the soil in a variety of forms. They may be dissolved in
0.0000001; the logarithm of the reciprocal of this con- the soil solution, from which they can be utilized di-
centration is 7. Consequently, the pH of pure (or neu- rectly. They may be absorbed onto exchange sites,
tral) water is 7: solutions that have a pH of less than 7 from which they may enter soil solution or be directly
are acid, and solutions that have a pH of more than 7 exploited by tree roots or microorganisms that contact
are alkaline. Hydroxyl and hydrogen ions are present the exchange sites. Alternatively, they may be firmly
at equivalent concentrations at pH 7. Hydroxyl ions fixed in clay lattices, immobilized in decomposition-
are less abundant than hydrogen ions below pH 7 and resistant organic matter, or present in insoluble inor-
more abundant above pH 7. ganic compounds. pH is important in determining the
“Soil pH” refers to the pH of the soil solution availability of many elements because of its relation-
when it is more or less equilibrated with the soil solids. ship to solubility and rates of decomposition. Figure
Some of the hydrogen ions on the cation exchange 11-5 shows approximately how the availability of vari-
sites are held so tightly that they are not measured ous nutrients varies with pH. Most of the macronutri-
under most experimental conditions: this is called ents exhibit maximum availability at pH values
exchange or reserve acidity. These “reserve” H” ions are between 6.5 and 7.5 (slightly acid to slightly alkaline),
released as the acidity of the solution is reduced, and although metallic ions are generally less available
this tends to buffer the pH of the soil; it serves to resist above pH 7. Optimum availability of nutrients for
changes in soil pH. Both of these types of acidity vary most plants seems to be between pH 6 and 7 (slightly
with the type of soil, depth in the soil, type of vegeta- acid to neutral). Of course, pH is not the sole determi-
tion, season, and weather. Consequently, generalized nant of availability, and Figure 11-5 is only a broad
statements about soil reaction should be treated with generalization to which there are many exceptions.
appropriate caution. Nitrogen occurs in both organic and inorganic
Soil acidity can develop in several ways. Acid pre- forms in the soil. Except in carbon-rich sedimentary
cipitation is a significant source in some regions, but rocks, sources of soil nitrogen are exclusively from the
generally it is relatively unimportant. Weathering of atmosphere (pollution and fertilization being notable
light-colored igneous rocks, which are rich in silica, exceptions in certain localities). The most frequent in-
can result in the production of acidic materials. Soil organic forms of nitrogen are nitrate (NO; ) and am-
organisms produce acids in several ways. Organic acids monium (NOQO,*) ions, but in poorly aerated soils,
may be created as a by-product of their metabolism, nitrite (NO, ) may be formed and accumulate to toxic
296 CHAPTER 11 Soil

reacts with calcium to form insoluble calcium phos-

phate, although there is generally not a great amount of
free calcium in the soil until the pH rises to pH 7 or
above. Maximum availability occurs at approximately
pH 6.5. As a result of these processes, phosphorus is
found only at very low levels in the soil solution (and is
therefore not subject to leaching) and many forest
plants must depend on mycorrhizal fungi to obtain the
phosphorus that they need. Phosphorus can exist in a
number of organic forms, including chelates of iron and
aluminum phosphate, and these may increase its avail-
availability
Relative ability to plants. It is also held on anion exchange sites.
Potassium occurs in a wide range of soil minerals
and is readily available where there is active weather-
ing of such materials. In the soil, it exists almost en-
tirely in inorganic forms, being rapidly released from
organic matter after death. It is found in solution, on
cation exchange sites, or “fixed” in the lattices of ex-
4 5 6 7 8 9 l panding clays. There is an equilibrium between these
pH different pools of potassium, which serves to maintain
Figure 11-5 Effect of soil pH on the relative availability of fairly constant levels of availability to plants despite
12 different nutrients. (After Truog. Reproduced from Soil Soci- additions to or losses from the soil solution or CEC.
ety of America Proceedings, vol. 11, 1946, pp. 305-308, by per- Levels of available potassium are generally higher in
mission ofthe Soil Science Society of America.) organic than in mineral horizons because “fixing” in
clay lattices in the mineral soil removes it from solu-
tion, but organic soils can be short of potassium be-
concentrations. Nitrate occurs almost entirely in solu- cause of the lack of inputs from mineral weathering.
tion and is therefore readily available to plants. Most As with potassium, calcium and magnesium in soils
ammonium ions are held in a readily exchangeable form originate largely from the weathering of primary min-
on cation exchange sites. Others are fixed between the erals. In alkaline soils, they may accumulate as free car-
lattices of clay minerals from which they are released bonates (detected by the effervescence of CO; upon
only slowly. Plants use both forms of inorganic nitro- treatment with dilute hydrochloric acid), whereas in
gen. Some species seem to favor nitrate, some flourish acid soils in humid climates, much of the calcium and
on ammonium nitrogen, and still others seem to grow magnesium released by weathering may be leached out
best with a balance of the two. Some plants actually ac- of the soil. Calcium is found on cation exchange sites
cumulate nitrate ions in their foliage. Plants also exploit and also in soil solution. It is present in organic matter
nitrogen in organic compounds. Simple amino acids and released on decomposition but more slowly than
can be absorbed by roots, and mycorrhizal fungi can ex- potassium because it is a structural component of plant
ploit several forms of organic nitrogen. Nitrogen is tissues, whereas potassium is a component of the easily
available over a wide pH range but decreasingly so decomposed cellular protoplasm.
below pH 6 and above pH 8, an important reason being Sulfur is present in soil minerals as sulfides, which
the adverse affect of pH extremes on soil microbes. De- are weathered to sulfate (SO,”>) ions. It also reaches
tails of the nitrogen cycle are given in Chapter 5. soil from the atmosphere as the result of direct uptake
The main source of soil phosphorus is weathering of of gaseous SO, by plants and in precipitation. As with
soil minerals, there being very little phosphorus in the nitrogen, the dynamics of sulfur in soil are largely
atmosphere. It is available over a narrower range of pH under the control of microbes.
than nitrogen, availability declining above pH 7.5 and There are small quantities of micronutrient and
below pH 6.5. In acid soils (pH less than 5), phosphorus nonnutrient trace elements in most minerals (ore de-
in the H,PO, form reacts with iron and aluminum to posits excepted, of course). Their availability is low in
form insoluble compounds. Above pH 6, phosphorus alkaline soils. It increases as the pH drops, but in
 Chemical Properties ofSoil = 297

highly weathered and acidic soils, availability may held on the negatively charged clay humus complex will
again be low because of leaching losses and the lack of be displaced. The water will pass down through and out
a reservoir in unweathered materials. of the soil, carrying with it salts of these anions unless
they are taken out of solution by plant and microbial up-
take or unless changes in solution pH (because of the
Leaching
loss of H* ions) cause them to precipitate out of solution.
The chemical composition of soil is strongly affected In soils characterized by high rates of CO, produc-
by its water relationships. The chemical products of tion and a pH value that permits both the formation of
mineral weathering accumulate in the upper layers of carbonic acid and its dissociation into bicarbonate ions
soils in regions in which evaporation exceeds precipi- (HCO;,) and hydrogen (H”) ions, leaching is largely
tation, resulting in alkaline (solonchak) or even saline controlled by the pH-dependent abundance of bicar-
(solonetz) soils. In humid areas, where there is an ex- bonate. This has been shown to be of importance in
cess of precipitation over evapotranspiration, at least the tropics and some temperate forests (Cole, 1976;
seasonally, chemicals released by weathering may be Cole et al., 1976). In soils where there is a great abun-
leached out of the upper layers of the soil. However, dance of nitrate ions (NO;) released by chemoau-
leaching of chemicals from soils is not dependent totrophic nitrifying bacteria, the leaching pathway
solely on water movement through the soil. Whether may be dominated by this ion (Likens et al., 1970). In
it occurs also depends on the cation and anion ex- situations such as polluted environments or near
change capacities of the soil, the uptake of chemicals coastlines with prevailing onshore winds, chloride
by plants, and the balance of ions in solution. (CI) or sulfate (SO,’-) may be important (Kimmins
There are two requirements for the leaching of and Feller, 1976). The addition of nitrate or sulfate
chemicals out of soils. First, the chemicals must be in
ionic form, and second, there must be hydrogen
ion—mobile anion couples in the soil solution. These Carbonic acid Nitric acid Sulfuric acid Hydrochloric acid

requirements are necessary because water draining out pathway pathway

Ammonium
pathway
Sulfur compounds or
pathway
— Atmoshperic sources
of soil is electrically neutral; soil water contains as H,0 + CO, compounds ("me sulfates of chloride ions

many cations as anions. Ions on the exchange sites will HCO nitrifying Sulfur-oxidizing Ck +H.O
oF bacteria bacteria 2
not be leached out of the soil despite large water flows low
pH
high
pH NO, + H,O t a
{

unless there is an adequate supply of ions in the solu- H*t+HCO. s0,> +H,O |

tion: hydrogen ions to displace the other cations and HNO,E + H+ Cl

:
anions with which the displaced cations can associate. J H,SO,
H*+ NO, +
The process of leaching is illustrated in Figure 11-6. ¢H*+S0,>
Soil water entering the forest floor may pick up a variety
of anions, such as bicarbonate anions formed from the
dissociation of carbonic acid (which is first formed by MM ltt
the combination of water and respiratory CO,); nitrate
ions produced by nitrifying bacteria; sulfate ions derived
from atmospheric inputs, from decomposition, or from
the activities of sulfur oxidizing bacteria; and chloride
ions released by decomposition or from direct inputs K*+ HCO, NH,’+NO, Gat + SO Nat + Cl-
from the atmosphere. The relative importance of these
different ions varies according to the climate and soil Composition of water draining out of the soil

type. In some soils, bicarbonate will be the most impor- Figure 11-6 Schematic representation of mechanisms of
tant anion, in some nitrate will dominate, whereas in soil leaching. The anions involved may be predominantly bi-
others there may be a mixture of the various different in- carbonate (HCO;)), nitrate (NO;), sulfate (SO,7), chloride
organic anions. The soil solution in some forest soils is (C1), or anions of organic acids (e.g., fulvic acid) produced dur-
dominated by organic substances such as fulvic acid or ing decomposition of organic matter. (After Cole et al., 1975;
oxalic acid, which provide the majority of the anions that Kimmins and Feller, 1976; Likens et al., 1970. Based on a figure in
are involved in the leaching of cations. As this hydrogen- Cole et al., 1975. Redrawn with permission of Les Presses de L’Uni-
and anion-rich water enters the mineral soil, cations versite Laval and D.W. Cole.)
298 CHAPTER 11 Soil

ions with fertilizers may result in the leaching of horizontal layers: the litter layer (L), the intermediate
cations from the surface layers of soils. layer (F; sometimes called the fermentation layer, but
because decomposition in this layer is generally aero-
bic and fermentation is anaerobic, the term is a mis-
11.5 Soil Organic Matter nomer), and the humus layer (H).
The L layer consists of recently fallen above-
Because of its intimate association with soil chemistry, ground litter that has not markedly changed in shape,
the discussion of soil organic matter could have been form, or color from when it fell: its origin is still obvi-
included in the preceding section. It is accorded a sec- ous. The F layer is where most of the decomposition
tion of its own to stress its preeminent importance in occurs, and it is frequently rich in fungal mycelia and
most forest soils. fine tree roots, which together bind the disintegrating
The input and decomposition of organic matter in litter into a loose mat. Physical comminution and de-
the form of litterfall was discussed in Chapters 4 and 5. composition reduce the litter to small pieces in this
In many forests, the largely unmeasured belowground layer, but much of it is still recognizable as to its ori-
litterfall (annual mortality of small roots and their myc- gin. The material is moister and darker in color than
orrhizal symbionts) may equal or exceed aboveground that in the L layer.
litterfall, so most existing estimates of total annual addi- The H layer is normally more compact, darker,
tions of organic matter to the soil are probably low. and moister and is made up of well-decomposed mate-
Litterfall is important as the source of the majority rial that is generally unrecognizable as to its origin
of the nutrients taken up annually by plants. It forms a (the remains of cones, rotting logs, or large bark slabs
superficial organic layer that plays an important role may be an exception). It may or may not be mixed with
in the protection ofsoil against erosion and in regulat- the underlying mineral soil.
ing soil moisture status. Organic matter favors the ac- Not all of these layers are always present or recog-
tivity of soil animals, and colloidal organic matter nizable, and their depth and detailed characteristics
promotes soil aggregation and a favorable structure vary considerably according to the type of forest floor.
and porosity. Once decomposed into humus, the or- Often, the designation of layer boundaries is difficult,
ganic matter contributes significantly to CEC, where- as they tend to grade into each other. In cool, moist en-
as organic acids released during decomposition vironments, the mortality of fine roots may be the
promote the weathering of soil minerals and the major addition of organic matter to the F and H layers.
leaching of mineral nutrients. The presence of or- Where this is the case, litterfall may be added annually
ganic matter is one of the parameters that differenti- throughout the forest floor, and differentiation of F and
ates soil from “parent material” or “nonsoil.” H horizons may be difficult. In very humid, acid envi-
When litter reaches the surface of the soil, it un- ronments, much of the colloidal humus of the H layer
dergoes physical and chemical degradation in the is leached down to the mineral B horizon, so a typical
process of decomposition (see Chapter 4). In warm H layer may be absent and the forest floor may be pre-
humid environments, this is rapid and virtually com- dominantly a deep F layer. On some sites, this layer
plete to inorganic constituents, so no permanent or- may have such profuse fungal mycelium that the mate-
ganic accumulation develops. In most forests, rial is light gray rather than the normal dark brown.
decomposition is slower than additions, with the result Despite these local variations in the general pat-
that a layer of organic matter called the forest floor tern, forest floors can generally be divided into three
builds up. In very dry environments, decomposition major categories, which represent the two extremes
may be exceedingly slow, with relatively little change and the median condition of what is in reality a con-
in the litter other than oxidation and minor fungal de- tinuum: mor, moder, and mull.
composition. In such environments, frequent ground Mor (also called raw humus) forest floors are gener-
fire may have been the major natural agent of mineral- ally acid and occur in cool, moist climates under tem-
ization in the past. In more humid environments, mi- perate or boreal vegetation. Decomposition of litter is
crobial decomposition is the major agent. generally slow and incomplete, leaving a residue of
Mature forests floors (those that have developed humus, and there is an abrupt transition from the for-
for a sufficient period of time to exhibit the features est floor to the underlying mineral soil. The L layer
that are characteristic for the vegetation, soils, and cli- varies in depth from a few millimeters to many cen-
mate of the area) can be subdivided into three distinct timeters and overlies an F layer rich in fine roots and
 Soil Biology 299

fungal mycelia. This layer, which is of variable thick- perceptibly into underlying inorganic horizons. In
ness, grades into a well-decomposed H layer, also of some mull forest floors, the upper layer of the soil is
highly variable thickness depending on time of devel- composed entirely of earthworm casts. The active
opment, soil, and climate. The humus can vary in tex- mixing of the organic material inhibits the develop-
ture from granular to fibrous to greasy-amorphous. It ment of fungal mycelium; consequently, bacteria dom-
may contain large quantities of fungal mycelium and inate the microflora. Mull forest floors are
fine roots or relatively little. In cool, humid climates, characteristic of mild, moist climates and fertile soils,
humus may be leached down into the mineral hori- conditions that promote rapid decomposition and
zons, endowing them with a dark color and leaving rel- vegetation that produces readily decomposable litter-
atively little H layer in the forest floor. In mor forest fall. In less favorable climates, they can be found only
floors, there is generally little activity by soil animals on very fertile sites. In more favorable climates, they
that are large enough to produce significant mixing of can be found on most sites.
mineral particles into the humus layer. Consequently,
mor humus layers usually contain very small amounts
of mineral particles. The scarcity of large soil animals 11.6 Soil Biology
is beneficial for the development of fungi, which dom-
inate the microflora. Mor forest floors are characteris- Irrespective of the physical characteristics of the land
tic of climates that inhibit decomposition, of plant surface and the degree of physical and chemical
species that produce slowly decomposing litter, and of weathering, the abiotic surface constitutes only a soil-
infertile soils, irrespective of climate (mor forest floors forming potential and not a soil. Only when the life-
can be found from the tropics to the Arctic). less geological substrate is occupied and modified by
Moder (also called duff mull) forest floors are charac- living organisms does the dynamic physical-chemi-
terized by a higher pH, more rapid and complete de- cal—biotic system that we call soil develop. Without bi-
composition of litter, and a more gradual transition from otic modification, the physical components of many
the forest floor to the mineral soil than in the mor type. soils would be completely unproductive. The biota lit-
The L layer, of variable thickness, overlies an F layer that erally create their own soil environment.
has less abundant fungal activity than in mor forest The term soil biota refers to the organisms that spend
floors (because of the increased activity of soil animals) all or part of their life cycle in the soil. They can be di-
but generally an abundance of fine roots. It grades into vided into soil flora and soil fauna. The flora is consid-
an H layer, which is generally thin (only a few centime- ered first because it provides energy needed by the fauna.
ters) and which frequently is partially mixed with and
grades into the underlying mineral soil. This reflects
greater soil animal activity than in mor types. Moder for- Soil Flora
est floors are generally more acid than mull forest floors The term soi/ flora refers to a diverse group of nonani-
(mull humus under red alder is an exception; the high mal soil organisms that include bacteria, actino-
level of nitrogen in the alder litterfall releases nitrate-N mycetes, fungi, and algae (all members of the
as it decomposes. Combined with water, this produces microflora) and those parts (the roots) of higher plants
nitric acid), and there is correspondingly less faunal mix- (macroflora) that occupy the soil. The macroflora are
ing. The humus generally has a loose, crumbly texture given only partial treatment here because they are dis-
(but can be greasy on wet, alder-dominated sites). cussed in a subsequent section.
Mull forest floors are characterized by little or no
L layer for most of the year despite considerable litter
Microflora. The soil microflora can be classified ei-
inputs, although significant autumnal accumulations
ther functionally or taxonomically. The group in-
may occur in deciduous hardwood forests. The F layer
cludes autotrophs and heterotrophs. The former
is virtually absent because decomposition is so rapid,
consists of both photoautotrophs and chemoau-
so for much of the year, the soil surface is an Aj,’ layer.
totrophs; the latter includes symbiotic, parasitic, and
This is characterized by intimate mixing of organic
saprophytic species.
matter and mineral material, and it grades almost im-

Surface mineral horizon enriched with organic matter; contains less than
Bacteria. Soil bacteria can be classified by their
17% organic C by weight. shape or by how they take up stains. For our purposes,
300 CHAPTER 11 — Soil

a classification based on their ecological function is ammonium nitrogen (NH,") to nitrite nitrogen
more useful. (NO,), and Nitrobacter oxidizes it further to nitrate
nitrogen (NO; ). This process of nitrification is partic-
AUTOTROPHIC BACTERIA. Autotrophic bacteria are ularly active in soils at pH values close to neutral, with
all chemoautotrophs that satisfy their energy needs by mild temperatures, moist soils, and a supply of readily
utilizing energy released during transformations of decomposable ammonium substrate. Relatively little
inorganic compounds. The genera Nitrosocystis, nitrification occurs in forest soils with low pH, but
Nitrosogloea, and especially Nitrosomonas can convert there may be a great increase in the activity of nitrify-
ing bacteria when such soils are limed (Figure 11-7) or
after clearcutting (Martin, 1985) or devegetation
(Figure 5-16). Other bacteria use nitrate nitrogen
= pH 5.5
Z 30,000 Nitrosomonas as a source of oxygen, reducing it to gaseous
x 12 (NH, ++ NO, ) pH 5.0 N, or N,O, a process called denitrification. Certain
=
£
a
10,000 Nitrobacter :
(NO, > NO,-) species of Bacillus, Pseudomonas, and Achromobacter are
& 5000 involved in this process.
4
Bve Both sulfur and iron are also utilized by
=
3ij chemoautotrophs. Some species of bacteria can oxi-
i 1000
2 H 3.9
dize sulfur or sulfides to sulfate ions (and hence to
36 5004 pH38 Den.
a sulfuric acid): Thiobacillus, Beggiotoa, and Thiothrix.
6 This ability is put to practical use in forest nurseries
2
100 : where “flowers of sulfur” mixed with organic matter
Control Phosphate Lime Phosphate +
lime is added to forest nursery soils to reduce an undesir-
Fertilizer treatment ably high pH. Sulfur may also be added with insolu-
(a)
ble fertilizers such as rock phosphate. The sulfuric
Without CaCO, at time of With CaCO, at time of acid that results from the bacterial activity converts
perfusion perfusion the insoluble material to soluble monocalcium phos-
Pies 150 Lime + phosphate phate, which is readily available to plants. Oxidized
S83
Z 2 me Lime + phosphate | why Lime sulfur compounds may also experience reduction by
5 5 80 KG Se

ze 60 oe 7 Control anaerobic bacteria. For example, Sporovibrio desul-

oe 40 os Lime Le phuricans is important in reducing sulfates to hydro-
Zo eA rg
—=
$3
o 20 ILA
S Control
gen sulfide in poorly drained organic soils. The
fS ee resulting toxic accumulations of this gas are partly
10 20 30
Duration of perfusion, days
responsible for the poor growth of trees in stagnant
swamps. Apparently, these sulfur-reducing bacteria
(b)
are not very active at pH values below 5.5.
Figure 11-7 Effect of liming on the abundance and activity Iron bacteria such as Crenothrix, Leptothrix, and
of nitrifying bacteria in an acidic forest soil under a beech- Gallionella derive their energy from the oxidation of
maple stand in Ontario. The study was conducted 8 years after soluble ferrous iron to almost insoluble ferric iron,
the lime (in the form of ground dolomitic limestone) and phos- whereas under anaerobic conditions, other bacteria
phorus were applied. (A) Effect of lime, with or without phos- reconvert ferric forms back to the ferrous state. The
phorus fertilizer, on the abundance (expressed as most probable
red staining of rocks by water draining from puddled
numbers) of nitrifying bacteria and the pH of the humus layer of
or gleyed soils is largely the result of the activities of
the forest floor. The limiting pH for the bacteria was apparently
these iron bacteria. Manganese experiences similar
between 3.9 and 5.0. (B) Effect of lime, with or without phos-
phorus, on rate of nitrification of added ammonium sulfate and
transformations, and toxic levels of the soluble, re-
with or without additions of CaCo, at the time of the study to duced forms of these two elements are partly responsi-
further increase the pH (perfusion is a laboratory technique used ble for poor root development in such anaerobic soils.
in studies of nitrification). The delay in the nitrification response Water flowing out of such soils often has what appears
is time needed for bacterial numbers to increase in response to to be a film of oil on the surface. Closer inspection re-
the addition of ammonium nitrogen. (After Corke, 1958. Used by veals that the water is coated with a thin film of micro-
permission of Michigan State University and the author.) scopic crystals, which form as the ferrous iron-rich
 Soil Biology 301

deoxygenated water contacts atmospheric oxygen. One group of heterotrophic bacteria has the abil-
These crystals act as a defraction grating, which pro- ity to fix atmospheric nitrogen (N, sometimes called
duces the colors of the spectrum. dinitrogen). Both aerobic (Azotobacter, Achromobacter,
and Biejerinckia) and anaerobic (Clostridium) nitrogen
HETEROTROPHIC BACTERIA. The abundance of fixers have been identified. Bacteria of the genus
heterotrophic bacteria in soils seems to be related to Clostridium are also noteworthy because of their abil-
two major factors: the availability of a suitable energy ity to decompose cellulose and because they are more
substrate and a tolerable pH. Temperature and mois- tolerant of acid soils than are Azotobacter. For example,
ture are also important regulators of these saprotrophs, the most probable numbers’ of Clostridium in a fertil-
as is the inhibitory effect that other soil flora have on ized maple—beech stand in Ontario 8 years after fer-
them. Heterotrophic bacteria utilize simple sugars and tilization were 19,000 (lime treatment; pH 5.0) to
starches but can also utilize cellulose, hemicellulose, 92,500 (lime + phosphate treatment; pH 5.5) per gram
and proteinacious substances (especially species of of humus, whereas Azotobacter were undetected. In the
Bacillus) by means of enzymes that they secrete. untreated control plot (pH 3.8), a value of 80,000 was
An important role of heterotrophic bacteria is the found. Clostridium was thought to be the major organ-
breakdown of complex nitrogen-containing molecules ism responsible for the fixation of 44 kg ha yr! of ni-
to ammonium nitrogen, which then becomes the sub- trogen in a mull humus forest floor beneath a stand of
strate for the chemoautotrophic nitrogen bacteria. A sugar maple (this was increased to 60 kg hat yr"! by
lot of the nitrogen in readily available forms (NO; and the addition of glucose as an energy substrate), where-
NH’) is rapidly taken up and retained in bacterial bio- as less than 1 kg ha"! yr! was fixed in a mor humus
mass (in a process known as microbial immobilization), forest floor beneath American beech (Knowles, 1965).
so much of the nitrogen released by microbial decom- Fixation of approximately 33 kg ha’ yr! under aspen
position is not available to plants. Bacteria require ni- (Brouzes et al., 1969) and 15 to 18 kg ha! yr! under
trogen and other nutrients to exploit the energy in the Douglas-fir (Fortin et al., 1979) has been reported.
cellulose and other organic molecules of undecom-
posed organic matter. Large quantities of such matter SYMBIOTIC BACTERIA. The symbiotic (mutually
in soils can result in nitrogen-deficiency symptoms in beneficial) relationship between various nitrogen-
plants, as most of the available nitrogen is immobilized fixing bacteria and higher plants was discussed in
by the microbes. Similarly, nitrogen fertilizer is often Chapter 5 (see Table 5-4). The relationship between
less effective on soils that have a high carbon-to-nitro- Rhizobium and many species of legume is well known,
gen (C/N) ratio and a lot of undecomposed woody ma- but many nonleguminous perennial plants also have
terial. As the C/N ratio is lowered and wood materials root nodules that contain nitrogen-fixing bacteria.
are decomposed, much of this microbial nitrogen is re- These associations give higher plants the ability to
leased and plant nitrogen nutrition is improved. colonize and grow in nitrogen-deficient environ-
The overall role of heterotrophic bacteria in the ments and are an important aspect of plant succes-
decomposition of complex organic matter is hard to sion (Chapter 17). Such nodulated plants have a
assess. They are known to prepare such material for stimulatory effect on nonnodulated plants growing
faunal consumption and are important in fertile soils with them.
of near-neutral reaction. They also play an important
role in organic decomposition as the gut flora of soil Actinomycetes. ‘This is a group of organisms that
animals. They are much less important than fungi in sometimes (but wrongly) are grouped with the fungi.
acid forest soils, and their activity is probably inhibited They exhibit characteristics of both fungi and bacte-
by soil fungi in much the same way that pathogenic ria, consisting of nonseptate branching hyphae, which
bacteria are inhibited by secretions from fungi of the are often brightly colored. These hyphae are easily
genus Penicillum. It has been shown that elimination of broken into short rods that resemble bacteria. In con-
mycorrhizal fungi from the forest floor of a radiata trast to fungi, they are sensitive to acidity and are ab-
pine plantation in New Zealand results in a greatly ac- sent at pH 4.7 or lower. An outstanding characteristic
celerated activity of nonmycorrhizal saprophitic or- is their ability to produce antibiotics of high toxicity to
ganisms and more rapid litter decomposition: the
Gadgil effect (Gadgil and Gadgil, 1975). *A commonly used unit of estimated bacterial numbers.
302 CHAPTER 11 Soil

other microbes. They are most abundant in grassland layer, where they form an interconnecting web of
or agricultural soils but are also found in forest soils. white, yellow, or dark-colored strands that frequently
As heterotrophs, they can decompose cellulose, hemi- form small fan-shaped mycellial networks clearly visi-
cellulose, and simple carbohydrates, and some species ble to the naked eye. Microscopic examination reveals
can decompose chitin (found in insect exoskeletons their presence in both the L and the H layers of the
and cell walls of fungi). forest floor and also in the underlying mineral hori-
Of major ecological significance is the involve- zons, but their abundance drops off rapidly with
ment of soil actinomycetes of the genus Frankia in depth. A study in a pine stand in Scotland revealed
symbiotic nitrogen fixation. The actinomycetes invade concentrations of fungal hyphae (expressed as meters
the root hairs of nonlegumes and form distinctive of hyphae per cubic centimeter of soil) of 5.6 m cm™
nodules that can equal or exceed the nitrogen-fixation in the organic H horizon and only 0.4 m cm? in the
rate of Rhizobium-legume nodules. Approximately 160 mineral B horizon (Burgess, 1963).
species from 13 genera of nonlegumes are known to Fungi vary in their ability to utilize organic matter.
develop nodules (Bond, 1977), of which the most sig- Some species utilize cellulose but not lignin, darkening
nificant for forestry is the A/nus—Frankia relationship. the material that they decompose and making it weak
Conifers such as Douglas-fir attain much _ better and crumbly (brown rot). Others attack lignin but
growth on nitrogen-deficient soils when there is a leave the cellulose, which results in a fibrous residue
subordinate layer of red alder. As much as 320 kg ha! with a lighter color (white rot). There is often a charac-
yr peak additions of nitrogen have been estimated teristic sequence of fungi participating in the decom-
for temperate red alder stands (see Chapter 5), and position of decaying wood. Such tissue is first invaded
mean values under five species of alder range from 12 by fungi utilizing sugars and other simple carbon com-
to 300 kg ha yr (Tarrant and Trappe, 1971). An un- pounds. They are followed by cellulose decomposers,
derstory of another actinomycete-nodulated species, and finally lignin decomposers take over. This se-
snow brush (Ceanothus), in Oregon forests has been quence of decomposition may vary in different soils.
shown to contribute up to 20 kg ha! yr‘ (Figure Perhaps the most significant group of fungi are
11-8). For further reading on N-fixation, see those that form symbiotic mycorrhizal associations
Chatarpaul and Carlisle, 1983 and Fortin et al., 1979. with tree roots. The mycorrhizal phenomenon is not
yet completely understood, but good reviews can be
Fungi. ‘The fungi make up the majority of the found in Bjorkman (1970), Hacskaylo (1971), Harley
saprotrophic heterotrophs (decomposer organisms) in (1969), Marx (1991), Read (1991), and Séderstrém
many forest soils, especially in acidic soils with mor (1991). The nutritional significance of mycorrhizae
humus. They are most abundant and obvious in the F was mentioned in Chapter 5. It has been shown that

S =)S

200

ha!)
litter
soil
the
(kg
surface
at )dominated
alder
red
by
ha~
(t
Accumulation
nitrogen
of
in
snowbrush nitrogen
of
Accumulation
in
ecosystems
Time (years)
(b)

Figure 11-8 Accumulation of nitrogen by communities of snowbrush, Ceanothus velutinus

(A), and red alder (B) in Oregon. (A, after Zavitovskil and Newton, 1968a. Redrawn by permission of
the Ecological Society ofAmerica and the authors; B, data from Newton et al., 1968.)
 Soil Biology 303

they increase the efficiency of uptake of nutrients that into the cells, where they form swellings called vesicles
are at very low concentrations in soil solution, and it is or arbuscules. This type is also refered to as vescicular-
believed that they accelerate mineral weathering and arbuscular mycorrhizae (VAM). Ectomycorrhizae in
absorb the nutrients that are released thereby. They particular are thought to be an adaption to nutrient-
facilitate the uptake of phosphorus and organic forms deficient soils, and most of the work on the value of
of nitrogen and promote the fixation of nitrogen by mycorrhizae to trees has been done on this form. Rel-
other organisms (see references in Voigt, 1971). atively little work has been done on endomycorrhizae
Mycorrhizae may prolong the life of roots, in- and VAM, but it seems that they confer the same gen-
crease their diameter and branchiness, and even inhib- eral benefits. They have been shown to increase
it the attack of pathogenic fungi. It has been estimated phosporus uptake and photosynthesis in tropical tree
that mycorrhizae increase the surface area of pine species (Lovelock et al., 1997) and to improve nutri-
roots by hundreds or even thousands of times in com- tion and growth of temperate hardood tree seedlings
parison with unaffected roots. In fact, pine trees do in Italy (Bragaloni et al., 1999).
not exist in nature without mycorrhizae. It has also Mycorrhizal associations are not permanent. Initial
been observed that pines have a stimulatory effect on formation occurs in the first growing season after the
other trees that grow with them in nitrogen-limited primary leaves of the seedling have been produced.
environments that resembles the effects of nitrogen The association lasts for approximately 1 year, after
fertilizer. The process involved is not known, but it which the infected root tip may die or may grow rapid-
may be either that the pine mycorrhizae are mobiliz- ly in length, thereby throwing off the fungal sheath and
ing previously unavailable nitrogen resources or that becoming a normal nonmycorrhizal root. New mycor-
they stimulate nitrogen fixation. Mycorrhizae increase rhizal associations are formed each year. Formation of
the tolerance of roots to extremes of pH caused by mycorrhizal associations is promoted by moderate nu-
high levels of soil sulfur and aluminum. They increase trient deficiency, and seedlings grown in pots with re-
the resistance of roots to other toxins, to high temper- peated fertilization have fewer mycorrhizae than do
atures, and to drought (Marx and Brian, 1976). Myc- unfertilized control trees. When the nutrient deficien-
orrhizae also improve the ability of trees to exploit soil cy is extreme, the development of mycorrhizae is in-
water (Read, 1980). They may assist regeneration by hibited because the trees are unable to provide the
transferring nutrients, carbohydrates, and moisture roots with the supply of carbohydrates that is apparent-
from overstory trees to understory seedlings and link ly necessary for the maintenance of the association.
the root systems of trees that are not in root contact, Inoculation of nursery stock with mycorrhizal-
thereby evening out soil resources that are not uni- forming fungi offers great potential for improvement
formly distributed aross a site (Simard et al., 1997). of artificial forest regeneration (Marx, 1991). Many
With this long list of advantages, it is not surpris- cases of poor growth of seedlings on infertile nursery
ing that the mycorrhizal habit is the most common soils; of failures of introduced species (especially
root condition for a great many plants, and it seems pines) in Australia, New Zealand, and Great Britain;
that mycorrhizae have been selected in evolution and of failures in reforestation of prairies and heaths
rather than uninfected root systems (Harley, 1969). have been corrected by inoculating the problem area
There are two major forms of mycorrhizae: with appropriate fungi. Failure of trees to invade
ectotrophic and endotrophic. Ectotrophic mycorrhizae, grasslands may in part reflect the antibiotic effect of
or simply ectomycorrhizae, consist of a short, thick- grass roots on the fungi that form mycorrhizae with
ened side branch of a fine root around which fungal trees. Several studies have demonstrated satisfactory
mycellium has grown as a sheath. This sheath is con- tree growth when such areas are planted with true
nected to hyphae throughout the surrounding soil, seedlings grown in nursery soils inoculated with spe-
and hyphae from the sheath penetrate the outer layers cific ectomycorrhizal fungi (Lamb and Richards, 1971;
of the root and form a network (the Hartig net) around Moser, 1959; Theodorou and Bowen, 1970).
the cortical cells. Endotrophic forms (endomycor- There has recently been great interest in the fungus
rhizae) do not have the outer fungal sheath, and the Pisolithus tinctorius as a mycorrhizal symbiont. It stimu-
hyphae of the Hartig net penetrate the cells of the lates greater development of dichotomous roots com-
cortex (Figure 11-9). An intermediate form, the pared with natural symbionts and increases the survival
ectendomycorrhizae, has a sheath but also penetrates and rate of biomass accumulation in nursery seedlings.
 ECTOMYCORRHIZAE

ENDOMYCORRHIZAE

Penetration sites

Ss

Cortex
Resting (@)
spore

| Arbuscule
Cortical cells ——

(c) (b)

100 25

80 = 204 a
=
=: 606
z=3 15 &
2‘O2
2 20 =
A 40 = 10 =
5)
5 0)
= Y
vo

20 5

0! 0

| non-mycorthizal YD, mycorrhizal

(d)

Figure 11-9 Mycorrhizae. (A) Ectomycorrhizal shortroots formed by Pisolithus tinctorius on

lodgepole pine. (B) Diagrammatic cross-section of ecto- and endomycorrhizae on tree roots. (C)
Photo of Douglas-fir seedlings in an Oregon forest nursery recently converted from a grass field.
Only the larger seedlings in the center were infected with mycorrhizal fungi. (D) The growth and
survival of mycorrhizal seedlings shown in C was substantially better than the nonmycorrhizal
seedlings. (A, courtesy of VG. Marshall, Canadian Forest Service. B, C, and D adapted from DeYoe and
Cromak, 1983. Used by permission of Oregon State University Extension Service and the authors.)

304
 Soil Biology 305

Table 11-3 shows the effects of this fungus on the tain their mineral nutrients directly from rock surfaces
growth of lodgepole pine, Douglas-fir, and loblolly by this means (lithoponics).
pine seedlings in the nursery. The performance of in- The rhizosphere is that portion of the soil in which
oculated seedlings is reviewed by Marx (1991). the abundance and composition of the microbial popu-
lation is influenced by the presence of the roots. The
Algae. Free-living algae occur at the interface of roots release carbohydrates, vitamins, and amino acids
the soil and the atmosphere. Of the several groups that into the surrounding layer of soil, which results in in-
occur, the blue-green algae are capable of nitrogen fix- creases in bacterial numbers by as much as 20 times
ation, and this is thought to be of considerable impor- (Katznelson, 1965). This increase in microbial meta-
tance during the colonization of exposed mineral soils bolic activity accelerates mineral weathering, and the
by plant communities. rhizosphere microbial community is frequently en-
riched in species that either fix atmospheric nitrogen or
Macroflora (Plant Roots). Roots play an important hydrolyze organic nitrogen to ammonium forms (am-
role in soil development and function. They penetrate monifiers). Formation of soil structure is also facilitated
mineral soil, sometimes to considerable depths. Or- by the increased microbial activity. Some of the organ-
ganic matter is contributed to the lower soil horizons isms in the rhizosphere produce growth-regulating
when roots die, increasing CEC and promoting the compounds (auxins) that modify the form of root
formation of soil structure. The channels left in the growth. These would include root nodule-forming
soil after the dead roots have decayed improve soil bacteria and mycorrhizal-forming fungi (Slankis, 1958).
aeration and facilitate the movement of gravitational One of the important effects of roots on the soil
water—important changes in fine-textured soils. results from the high turnover of fine roots. Root
Roots increase the rate of weathering of primary soil mortality in grasses results in a great enrichment of
minerals and rock, either because of organic com- organic matter in mineral layers in prairie soils, and
pounds that they secrete or because of the activities of the accumulation of forest floors can be as much or
the microorganisms that live in association with roots. more the result of the production and death of fine
These include most mycorrhizal fungi and a variety of roots and mycorrhizal fungi as of aboveground litter-
microorganisms that live in the rhizosphere. Where fall. As already mentioned, this poses problems in the
there is a lack of mineral soil, trees may be able to ob- classification of forest floors into L, K, and H layers

Table 11-3 Growth of Seedlings of Three Species With and Without Incorporation of Pisolithus tinctorius Inoculum
Into the Nursery Soil Mix in Which the Seedlings Were Grown
Weight at 16 Weeks After Seeding, mg
Shoot/Root
Species and Treatment Shoot Root Total Ratio

Douglas-fir?
Inoculated 354 151 505 2.36
Control 239 Wile 347 ull

Lodgepole pine*
Inoculated 129 120 249 1.08
Control 83 81 164 1.03

Weight in g of 7-Month Seedling

Shoot/Root Stem Diameter
Shoot Root Total Ratio Height (cm) (mm)

Loblolly pine®
Inoculated 8.9 2.6 Wilds) 3.42 36 8.2
Control 4.2 1.4 5.6 3.00 26 6.0

4Data from J. A. Dangerfield (unpublished).

>Percentage increase attributable to inoculation.
°Data from Marx and Bryan (1976).
306 CHAPTER 11 Soil

because fine-root mortality may add fresh “litter” to The stability that roots confer is particularly impor-
each of these three layers annually. tant for tall plants such as trees. The resistance to wind-
Roots provide plants with a number of important throw depends partly on rooting depth and volume and
requisites for life: stability, water, and most of the nec- partly on the tolerance of the tree species to soil condi-
essary nutrients. Their importance is reflected in the tions. Spruces, which tolerate severe restriction of root
considerable investment by plants in root biomass: as penetration when they grow on swampy soils, are often
much as 20 to 45% of tree biomass in polar areas and more resistant to windthrow on such sites than are
as little as 10 to 20% in tropical areas. More than 50% pines. Species of the latter genus are adapted to drier
of annual biomass production may be invested in root sites and are generally less tolerant of the extreme root
production, and one reason for the small aboveground modification that occurs on the wet sites. However,
production on infertile sites is believed to be the high root development and its relationship to windfirmness
production of fine roots necessary to obtain the re- are multiply determined, and there is much variation in
quired nutrients. any general relationships that are proposed.
Root systems can take a variety of forms (taproot, Nutrients and moisture are absorbed from the soil
heart root, and flat root [Figure 11—10]) as a result of with an efficiency that depends to a considerable ex-
the influence of a variety of factors (see reviews in tent on root distribution relative to the resources of
Bilan, 1971; Herman, 1977; and Sutton, 1969). These moisture and nutrients. Most nutrient uptake occurs
forms are more under environmental than genetic in fine roots (less than 2 mm in diameter), which tend
control. For instance, spruce trees, long thought to be to be concentrated in the surface soil horizons (cf.
shallow-rooted, can have heart roots and even tap- Figure 5-3), especially in acidic, nutrient-poor soils.
roots extending to a considerable depth when grown Conversely, moisture is taken up throughout the pro-
on suitable soils. Pines, which have traditionally been file wherever the moisture is available to the roots.
thought of as having deeply penetrating taproots, can The following factors are important in determining
have flat roots on soils that restrict rooting. The evo- the vertical distribution of roots in forest soils.
lutionary relationship between certain tree species and
certain types of sites certainly does result in trees’ hav-
ing inherent root growth forms, but these are broadly 1. Physical soil properties. Roots cannot penetrate into
modified by the rooting environment. Some species excessively hard material such as solid bedrock or
have highly plastic root forms: environmental factors hardpan (a compacted or cemented horizontal
can easily modify them. Other species have nonplastic layer in the soil) and will be shallow where such
root forms: the genetically controlled form is apparent obstructions are close to the surface.
irrespective of soil types (e.g., black spruce). 2. Soil moisture and aeration. Roots cannot grow in-
definitely in dry or oxygen-deficient soils. Fine
roots may grow into such zones but will die sooner
or later, so larger roots can never develop. In
humid forests, there is abundant rooting in the F
and even in the L layer of the forest floor; in dry
climates, fine roots are absent from the L and
much of the F layer of forest floors and may even
be absent from the H layer. In such forests, they
are restricted entirely to the moister mineral soil.
Lack of oxygen and/or toxic accumulations of CO,
may exclude fine roots from lower layers of soil
Taproot Heart root Flat root with poor aeration and from structureless soils.
Figure [I-10 ‘The three major types of root system com-
3. Soil temperature. Although roots have lower cardi-
monly found in trees. In some tree species, the form of the nal temperatures than shoots do, reduced temper-
root is under strong genetic control, whereas in others, the root atures do limit root growth. Surface soils warm up
form is more a reflection of the soil conditions under which the much earlier in spring than deeper soils do, so new
tree is growing. Intermediates among these three main types root growth tends to be concentrated in the sur-
are common. (After Armson, 1977. Used by permission ofthe Uni- face layers. Permanently frozen soils completely
versity ofToronto press and the author.) restrict root growth.
 Soil Biology 307

4. Nutrition. Roots that grow into more fertile soil c. Microfauna. Microscopic organisms less than
tend to grow faster than those that grow into less 0.2 mm: smaller mites and nematodes and
fertile soil. Consequently, root growth tends to be protozoans.
more prolific in the surface organic horizon or the 2. By habitat
most fertile mineral horizons. The poorer the soil,
the more superficial the fine roots; the richer the a. Animals that inhabit the surface litter (e.g., snails)
soil, the more uniform the distribution of fine because they are too large to penetrate the soil.
roots with depth. This is true, however, only where b. Animals that live in the pore spaces of struc-
moisture is not severely limiting. tured soil.
c. Burrowing animals such as worms or ants (bur-
5. Competition or interaction with roots of other species.
Roots of established lesser vegetation (e.g., grasses, rowing vertebrates are generally of local rather
shrubs) compete effectively for moisture and nutri-
than general importance).
ents in the surface soil and may effectively exclude 3. By trophic relationships
or restrict the roots of planted trees or invading a. Bacteria and some protozoans are preyed on by
woody plants. Some plants exude from their roots other protozoans, and mites graze on the fungi
chemicals that inhibit the root growth of other that are decomposing the organic matter.
species (alellopathy; see Chapter 15). Root compe- b. Mites are consumed by other arthropods,
tition can result in the failure of regeneration on which together with earthworms are preyed on
grassy and shrubby sites. by burrowing mammals.
6. Soil chemistry. Toxic accumulations of chemicals c. The feces of soil animals are rapidly invaded by
such as iron or aluminum or a chemical in a toxic fungi and bacteria, which are then grazed by
reduced form may inhibit fine-root growth. This various animals, including earthworms. ‘There
may prevent roots from penetrating down into the is a marked succession of fungi and fauna dur-
soil B horizon. The damaging effects on trees of ing the various stages of breakdown of organic
acid rain is thought to be related to the mobiliza- matter.
tion of iron and aluminum, which inhibits fine-
root development, impairing tree nutrition and Macrofauna
resistance to summer drought, unless the trees can
release into the rhizosphere organic molecules that Soil vertebrates. Soil vertebrates influence soil and
complex with the toxic ions and render them vegetation in several ways. Their burrowing activities
harmless or at least less harmful to the roots. influence soil aeration and structure, soil drainage,
and soil development. By active churning of the soil,
the development of typical soil horizons may be pre-
Soil Fauna vented. Many soil vertebrates feed on insects and plant
The soil fauna include all animals that spend at least part seeds at the surface of the soil and thus have a signifi-
of their life in the soil. They are a diverse group, ranging cant effect on the overall composition of the soil fauna
from moderately large mammals that excavate under- and on vegetation development. Tree invasion of
ground burrows to microscopic mites, nematodes, and prairies and subalpine meadows may be inhibited by
even protozoans that live in films of water that coat soil animal consumption of tree seed and damage to roots
particles (Figure 11-11). Reviews of soil animals can be or girdling of the stems of — seedlings.
found in Jackson and Rowe (1966) and Wallwork Plant community species composition and _ rates
(1970). Soil fauna can be classified in a number of ways. of succession in general are affected. Predation of
Douglas-fir seed by the white-footed deer mouse
ee by size (Peromyscus maniculatus) in British Columbia frequent-
a. Macrofauna. Animals whose body size is ly precludes natural regeneration of this tree except in
greater than | cm: vertebrates, mollusks, earth- years ofvery heavy seed crops. This mouse, which can
worms, and larger arthropods. reach densities of 30 per hectare, can consume up to
b. Mesofauna. Body size from 0.2 mm (the small- 300 seeds per night (Sullivan, 1968b). It has been
est size visible with a hand lens) to 1 cm: mites, suggested that the marked periodicity of Douglas-fir
springtails (Collembola), potworms (Enchytraeids), seed crops could be in part an evolutionary response
and larger nematodes. to this very heavy seed predation.
308 CHAPTER 11 = Soil

(e)

Figure 11-11 Some of the more common types of soil animal found in forest floors: a spring-
tail (Colembola; Arthropoda; (A)), a pseudoscorpion (Pseudoscorpionida; Arthropoda; (B)), a milli-
pede (Diplopoda; Arthropoda; (C)), Enchytraeid worm (Annelida; (D)), and Orabatid mite (Acari;
Arthropoda; (E)). (Photos courtesy of Dr. VG. Marshall, Canadian Forestry Service.)

Mollusks. Mollusks are particularly abundant on Earthworms. Earthworms have received more
richer soils, where the calcium needed for their metab- attention than any other group of soil animals. There
olism is abundant. Some feed on living plants, some are are many different genera and species with widely
saprophytic and utilize litter, and some are preda-tory varying habitat requirements, which occur from low-
and feed on earthworms. In some forest types, large elevation agricultural soils to high-elevation subalpine
slugs may be the major herbivore feeding on minor forest soils, from tropical to boreal forests. Earth-
vegetation (e.g., in the very humid forests on the coast worms are responsible for large-scale soil mixing. This
of British Columbia); they can also be major herbivore takes surface organic matter deep into the mineral
pests in agricultural ecosystems (Rollo, 1978). soil, promoting good soil structure, and bringing min-
 Soil Biology 309

eral material from lower horizons to the soil surface. with mor forest floors (Langmaid, 1964) (Figure
By their burrowing activities, earthworms can, over a 11-12). Earthworms do not tolerate flooding and will
period of centuries, bury large rocks and even the come to the surface during heavy rain. Neither do
ruins of old stone buildings. Their burrows aerate the they tolerate drought. During hot, dry weather,
soil and provide channels for water movement. Their worms will move to deeper, moister layers in the soil.
feces are invaded by bacteria and form fairly stable soil
aggregates, so soils with abundant earthworm activity
Millipedes. Millipedes play an important role in
are generally well aerated, well structured, and fertile.
the initial comminution and decomposition of organic
Forest floors influenced by earthworms are mulls
matter in forest soils. They tend to be most abundant
and may consist almost entirely of worm casts and ag-
on moist sites with moder or mull humus types.
gregates formed by microbial activity. The worms
convert the organic matter to a condition that is more
favorable to microbial decomposition, and this may be Termites. “Yermites play a major role in many
the most significant overall effect of earthworms tropical soils. Soil mixing and incorporation of or-
(Satchell, 1967). Mulls can develop within a few years ganic matter occurs as with earthworms. The con-
of the introduction of earthworms to wormless soils struction of termite mounds from the finest soil

Soil Depth
(cm)

oH HORIZON

ee
6
6.8 See. L layer
4
eae ee, F layer
2 HORIZON
P.H
H layer Surface of mineral soil 6.8 L layer
‘Eluviated mineral
Well-mixed
layer: Ae
sp 2 f mineral-humus
layer: Ah
E Illuviated mineral
af4 5.0
layers

i6

a8 Se
Bfh horizon
(Iron and humus
Illuviated
accumulation) +10 : mineral layers
Bfh
12
Bf horizon ak 4.8
(Iron accumulation) + 14
dh Bf
16

3 years of earthworm activity

Figure 11-12 Conversion of a podzolic mor forest floor to a mull forest floor after the inva-
sion of the site by earthworms in a New Brunswick coniferous—deciduous forest ecosystem
with a silt-loam soil. Over a 3-year period, virtually all of the forest floor was incorporated into the
upper mineral soil. Similar effects were observed in sandy, silty, and clayey soils. In all cases, there
was a general increase in the pH of the surface mineral layers, in some cases by as much as 0.6 pH
unit. (After Langmaid, 1964. Used by permission ofthe Agricultural Institute of Canada.)
310 CHAPTER 11 — Soil

particles apparently results over long periods of time are not thought to be involved in the decomposition of
in a fine-textured, gravel-free upper soil layer that is organic matter.
more susceptible to some forms of soil erosion than
the original soil (Nye, 1954).
Microfauna
Smaller Mites and Nematodes. Smaller mites
Mesofauna and nematodes are essentially similar to larger mites
Ants. Ants alter the bulk density of soils in the vicin- and nematodes of the mesofauna. Owing to their great
ity of their burrows, and this change can extend to numbers, mites play a major role in producing the fri-
some depth in the soil. Their social organization en- able crumb structure of some humus forms.
ables them to establish populations rapidly. In areas of
prolific ant activity, considerable quantities of material
are brought to the surface from lower horizons. This 11.7 Soil Fertility
material accumulates at the surface, and after a pro-
Soil fertility is “the status of the soil with respect to its
longed period, it forms the surface mineral layer in
ability to supply the nutrients essential to plant growth”
much the same way that termites affect surface texture
(Soil Science Society of America, 1973). Soil fertility
in the tropics. As predators, ants affect the composi-
can be considered either in terms of the availability of
tion of the soil fauna. Their overall effects on soil have
nutrients to plants or in terms of plant nutrition. Nutri-
not been studied in detail.
ent availability is determined by the sum total of the
Mites (Acari). Mites are one of the most numer-
physical, chemical, and biological characteristics of the
ous types of animals in most forest soils; their numbers soil; consequently, assessment of soil fertility is no easy
can go as high as 10,000 per cubic meter. They are lo- matter. Tree nutrition is an equally complex subject be-
cated mainly in the organic surface layers, where they cause of the interactions between different nutrients;
are important in decomposition. In addition to feed- the variation in nutrient demand with species, age, and
ing directly on organic matter, they graze on bacteria physiological condition; and a general lack of knowl-
and fungi. Some are predaceous. edge as to the nutrient requirements of trees.
Soil fertility is so complex that only a brief sum-
Springtails (Collembola). Springtails are another mary is possible here. A more detailed treatment of
very prolific group of soil animals, especially in the sur- the topic can be found in Armson (1977), Morrison
face organic layers. ‘They have been reported to reach (1974), and Tisdale and Nelson (1966). However, de-
densities approaching 50,000 per square meter in a spite the difficulties in defining soil fertility, it contin-
Douglas-fir plantation (Poole, 1961). They are more ues to be of great interest to growers of plants because
abundant in moder and mor forest floors than in mulls, the nutrient regime is one of the principal environ-
possibly because of their habit of feeding on fungi, mental factors over which some degree of control can
which are much less abundant in mull than in mor or be exercised. Various approaches have been made to
moder humus. They also feed on living and dead plant the assessment of site fertility.
material, feces, bacteria, and algae (Armson, 1977).
Measurement of Levels of Nutrients
Potworms (Enchytraeids). Potworms are small
white worms that can attain amazing numbers The total quantity and the “availability” of nutrients in
(250,000 per square meter [O’Connor, 1957]). Pot- the soil can be measured analytically. Interpretation of
worms are thought to feed primarily on dead organic such measures is usually based on the results of agricul-
matter but also on small feces. They ingest small min- tural studies, but often there is little correlation be-
eral particles and are probably important in mixing or- tween conventional analytical measures of nutrient
ganic matter into the mineral soil. availability and the growth of forest trees. Different
analytical methods yield widely varying estimates of
Nematodes. ‘Vhese are small round worms that availability, and it is frequently unclear which estimate
are best known for their parasitism of plants. Some ne- is the most reliable. Analytical measurement of the
matodes are predaceous on other soil organisms. They total nutrient levels in the soil is more reliable, but al-
 Soil Fertility 311

though such totals represent a long-term reservoir, tance movements are also important for nutrients such
they may contribute little to current fertility because as potassium and nitrate nitrogen, which can reach the
much of the reservoir may be in a form that will remain roots from variable distances in the soil by either diffu-
unavailable to the plants for a long period of time. sion or mass flow (Ballard and Cole, 1974).
These problems are compounded by the difficul- Other problems with the soil analysis method of
ties of defining rooting volume. Estimates of soil nu- determining fertility are that it gives a static picture of
trient reserves are often based on the soil to the depth what is a dynamic soil parameter, and a small capital of
of maximum rooting (root system sorption zone: rooting nutrients being rapidly cycled will result in a more fer-
depth X area of root spread). In Chapter 5, it was tile soil than a large capital of nutrients cycling slowly.
noted that virtually all of the fine feeding roots may be Fertility is also related to the demands of the plant
concentrated in the upper few centimeters of the soil, crop. A site may be fertile in terms of the low demands
even though the depth of maximum rooting may be as of a pine crop, whereas the soil might be infertile in
great as 2 m. The nutrient reserve in the root system terms of the requirements of a more nutrient-
sorption zone therefore may be a gross overestimate demanding crop such as spruce or fir. A soil that is fer-
of the pool of nutrients being exploited by the trees. tile in terms of the modest requirements of many
Alternatively, one can consider only the nutrients species of conifer may be infertile in terms of the de-
within some arbitrary distance of the individual roots mands of many species of deciduous hardwood.
(e.g., 1 cm), a zone defined as the root surface sorption Once the relationships between plant growth and
zone (Voigt et al., 1964). Table 11-4 compares the some selected soil analysis parameters have been
available amounts of three nutrients in these two established, the method is relatively quick and inex-
zones. From this discussion, it should be obvious that pensive. Unfortunately, establishing these relation-
conclusions concerning fertility depend very much on ships can involve considerable effort and expense, and
the analytical procedures used and on the volume of the method is compounded by the problems of soil
soil that is being considered. heterogeneity. The high cost of soil analysis often re-
Another problem is that conventional analysis of a stricts the number of samples so much that this
soil sample taken at a single point in time ignores mass method cannot give an accurate picture of site fertility.
movement or diffusion of elements to tree roots in soil
solution. Nutrients contained in downslope seepage are
a major factor in the fertility of lower slope sites, where
Foliar Analysis
the rooting zone is underlain by an impervious layer An alternative and commonly used approach is to
(Klinka, 1976), but this important type of nutrient input measure the concentration or content of nutrient ele-
is ignored by conventional soil analysis. Shorter-dis- ments in plants, although the relationship between fo-

Table 11-4 Comparison of Exchangeable Nutrient Reserves in the Root System Sorption Zone and Root Surface
Sorption Zone for 5- to 9-Year-Old Pitch Pine Trees
Nutrient Availability, mg

Pp? Ca K

Amount in the root 2p Shs 198 35

surface sorption zone
Amount in the root 148 2430 16,400 3530
system sorption zone
Total in tree 154 415 247

Annual uptake 19 53 28

Data from Voigt et al., 1964.

For phosphorus, two different estimates that were derived from two different analytical methods are given. Obviously, estimates of soil nutrient
reserves vary enormously depending on the method of chemical analysis and the method of calculating reserves from the data.
°H,O soluble.
°NH,F-HCI soluble.
312 CHAPTER 11 Soil

liar chemistry and soil fertility is not simple. If the Possible foliar respon
plant is growing particularly well, then it may have following fertilizatifn 7 y
Unit foliage weight, (e.g., g per 100 leaves)
lower concentrations of certain elements (because of —— Isolines of unit /
foliage weight /
the dilution by all the carbon, oxygen, and hydrogen / v4 as
Z
that is being incorporated in the biomass) than if the al
plant were growing only moderately well. Very low
concentrations of essential nutrients are normally as-
sociated with slow growth, but the relationship is not
necessarily causal. Factors such as adverse climate,
root pathogens, and soil moisture conditions can re-
sult in low foliar concentrations on fertile soils or high
foliar concentrations of some nutrients in trees that
are growing slowly on dry sites.
Foliar analysis is quick and relatively inexpensive foliage,
in
concentration
Element
weight
dry
%
once reliable relationships between foliar chemistry
Element content (e.g., g element per 100 leaves) of foliage
and growth performance have been established, but
there are some drawbacks. Considerable effort and ex- Interpretation of the Change
pense are involved in establishing such relationships,
especially because ratios of elements may be more im- Response in Interpretation
portant than absolute levels, and the technique is hard
to apply once trees get larger because of the difficul- Direction Needle dealled Nutrient Possible
ties of sampling crowns in tall trees. Despite these of Shift Weight Conc. Content Status Diagnosis
problems, it is generally a more successful means of
A = =f Dilution Nonlimiting
assessing fertility than soil analysis.’ For reviews of this
approach, see Morrison (1974), van den Burg (1976), B 4p Unchanged Nonlimiting
and van den Driessche (1974). C ae + Deficiency Limiting
A recent development of foliar analysis as a means D af Luxury Nontoxic
of predicting the nutritional status and potential re- consumption
sponse of trees to fertilization is the use of short-term E ~ ara 5 Excess Toxic
foliar response to fertilizer additions (Figure 11-13). F Excess Antagonistic
The technique involves plotting foliar concentration
(%) against the nutrient content (weight X concentra- Figure 11-13 Graphical analysis of the nutritional status
tion) of foliage (expressed as mg 1007! leaves or some of trees. The change in the relationship between nutrient con-
similar unit) and superimposing on this graph lines of centration, nutrient content, and dry weight of foliage after fer-
tilization. (After Timmer and Stone, 1978. Adapted from Soil Sci-
weight per 100 leaves (or some similar unit). Data on
ence Society of America Journal, vol. 42, pp. 125-130 by
nutrient concentration and content obtained before permission of Soil Science Society ofAmerica.)
and one growing season after a spring addition of fer-
tilizer are plotted, and interpretations of nutritional
status and fertilizer response potential are made ac-
cording to the change in the position of the foliage The question of nutrient ratios has been investi-
data on the graph. For further details of this useful gated extensively by Ingestad (e.g., 1967a,b, 1971,
technique, see Mead and Mansur (1993), Morrow 1979, 1981), who reported that optimum growth of
(1978), Munson and Timmer (1989), Timmer and seedlings oceurs only when the ratios of macronutri-
Stone (1978), and Weetman and Fournier (1982). ents lie within a fairly narrow range. For birch, pine,
and spruce seedlings, optimum nutrition occurred at
"Recent developments in this field include computer programs that will the following relative levels: N = 100, P = 13, K = 65,
compare foliar nutrient analysis data with data banks on foliar chemistry Ca = 6, Mg = 8.5 (Ingestad ratios). Alternatively, one
and growth ofthe species involved and produce a nutritional diagnosis and can express each element as a percentage of total con-
fertilizer recommendations (e.g., the work of Dr. T: M. Ballard, Faculty of centration; for pine, N/P/K = 67:7:26 (Lavrichenko,
Forestry, UBC, Vancouver, B.C.). in Morrison, 1974).
 Soil Fertility 313

Assessment of Fertility by Plant Production was determined for each species, and charts were pre-
The fertility of the soil is ultimately expressed in the pared that enable the user to estimate site index for
growth of plants. Plant productivity is a good empiri- each species even when the site is devoid of vegetation.
cal measure of soil fertility as long as climate, By way of example, the charts for cottonwood (Populus
pathogens, herbivores, plant competition, and rooting deltoides) and a sample of the use of the system are
patterns permit the plants to exploit and reflect the shown in Figure 11-14. The method was tested for
availability of nutrients in the soil. Where nutrients approximately 20 sites for each species and resulted in
are not the major limiting factor, this method will not a correlation coefficient between observed and pre-
dicted values of 0.99 for cottonwood (best) and 0.93
yield a good assessment of the fertility of the soil. A
for sweet-gum (worst).
fertile soil in a climate that restricts plant growth will
Eis (1962) correlated eight topographic features
not yield high plant production. Nevertheless, the
and eight soil and water regime features with the site
method will give a good estimate of the total produc-
index of various west coast tree species at the Univer-
tive potential of the site, and ultimately this is the ob-
sity of British Columbia Research Forest near Van-
jective of soil fertility assessment. A major drawback of
couver. The soil and moisture regime accounted for
the method is that it takes a long time to assess fer-
55% of the variability of site index, whereas landform
tility in this way.
accounted for 40%. Soil and moisture variables were
able to account for 81% of the variability in site index
observed within individual plant associations (defined
Plant—Site Relationships by the composition of the minor vegetation).
Another way of assessing soil fertility is to examine
site—plant relationships. Plant production is examined
Field Fertilizer Trials and Assessment
as moisture and nutrient conditions change, and math-
ematical relationships between biomass parameters by Visual Symptoms
and various soil parameters (depth, texture, moisture, Perhaps the most reliable and widely used method of
etc.) are developed. This method requires a lot of ini- assessing soil fertility is the empirical approach of field
tial work, and it is unclear how general such relation- fertilizer trials to establish nutrient deficiencies: the
ships may be; interaction and compensation of “try it and see” approach. Such trials are time consum-
environmental factors may restrict the use of relation- ing and can be expensive but are generally favored for
ships to the region in which they were developed. An predicting response to fertilizers. Although less costly
example of this type of approach is that of Steinbren- and time consuming, pot trials in the greenhouse have
ner (1976). In this study, site index (average height of proved to be less satisfactory because it is often difficult
dominant trees at some index age [formerly 100 years; to establish relationships between pot and field re-
recently 50 years]) of western hemlock (Zsuga hetero- sponses. Field trials have often made use of visual
phyla) was related to total and effective soil depths, symptoms such as foliar discoloration, needle twisting
gravel content, depth of the A horizon, percentage of and fusing, premature leaf fall, resin exudation, and/or
clay in the A horizon, silt and clay in the A horizon, el- deformation of young shoots to indicate deficiencies.
evation, precipitation, slope position, and several other The predictive efficacy of such symptoms depends on
soil parameters. Separate equations were formulated the degree of correlation between the symptoms and
for residual and glacial soils. The best equation for field response to fertilizers. ‘The increasing use of color
residual soils was found to be: Site index (in m) at 50 charts that provide objective descriptions of color
years = 32.75 + 0.265 (depth of A horizon) — 0.12414 should help to improve the technique. Some of the
(depth of A horizon)’ — 0.24409 log (depth of A hori- major problems with the method are that many symp-
zon) + 101.42 (silt and clay content of A horizon) — toms are nonspecific and are usually evident only when
0.00111 (elevation)? — 0.04795 (slope position). deficiencies are severe, and even then only at certain
In a somewhat more comprehensive approach, times of the year. Also, it has been suggested (Ingestad
Baker and Broadfoot (1977) related the site index of and Lund, 1979) that visual deficiency symptoms may
eight species of hardwood trees in the southeastern be a better indicator of changing nutrient conditions
United States to a complex of 23 nonvegetation para- than of a permanent deficiency. If plants are given suf-
meters. The contribution of each factor to site index ficient time to adapt to low nutrient availability, then
314. CHAPTERI11 Soil

Maximum tree height growth 130’

Physical condition Available moisture Nutrient availability Aeration

(46’) (46') | (26’)

| e
e
ee
ce
or
aa
Soil depth Water table depth Geologic source Soil structure
(16’) (10’) (8’) (3’)
+ ~ + -
Soil texture Presence of pans Past use Swampiness
(il) (9) (5’) (3’)
+ oa -
Bulk density Topographic position Organic matter Mottling
(9") (7') (4’) (3')
- + + -
Soil structure Microsite Topsoil depth Soil color
(5’) (7') (3’) (3’)
~ + Soil age
Past use Soil structure (3’)
(5’) (5') +

Soil texture pH
(4’) (3’)
a

Flooding
(2')

Past use
(2')

Figure 11-14 Site evaluation based on four major soil properties: physical condition, avail-
able moisture, nutrient availability, and aeration. Contribution of various soil-site properties to
the four major soil factors and the estimated contribution of each property to the height growth of
eastern cottonwood (130 ft at age 30 years) on an ideal site. Addition of all estimated contributions
produces a total of 130, A key is provided giving ranges for each property, from best to poor. Using
this key, the site index (see Chapter 6) of any site can be determined. (After Baker and Broadfoot,
1979. Used by permission of the authors.)

deficiency symptoms may disappear. Table 11-5 pre- ecosystems are disturbed and recover, as vegetation
sents the principal visual symptoms that typically ac- changes and as stand development proceeds, and as
company severe deficiencies of various elements. ecological succession occurs (the temporal dynamics
of nutrient cycling are discussed in Chapter 5). As a
‘Temporal Aspects of Soil Fertility: consequence, a snapshot measure of soil fertility may
be poorly related to rotation-length forest ecosystem
The Use of Ecosystem Management
NPP and other processes and characteristics. Soil fer-
Models ‘To Assess Fertility
tility during the post-disturbance assart flush is a poor
Most assessments of soil fertility refer to instanta- predictor of long-term site fertility and plant growth.
neous measures of nutrient availability-or plant re- Humus form has traditionally been interpreted in
sponse to it, or to relatively short-term bioassays of terms of nutrient-supplying power and site fertility.
fertility. This reflects the heavy influence of agricul- “Raw” humus in northern forests has traditionally been
tural soil science where short-term fertility is a key considered a sign of low nutrient availability and low site
issue in crop yields. However, soil fertility and nutri- productivity. However, such raw humus can contain a
ent availability are not static characteristics of a par- significant proportion of the total site capital of the most
ticular forest soil or site—they change as forest limiting nutrients (usually nitrogen in these forests), and
 Soil Development — 315

humus accumulations may be very important for long- ecosystem models that represent the dynamic processes
term site productivity while limiting short-term produc- of tree growth, tree nutrition, and nutrient cycling,
tivity by reducing nitrogen availability (Prescott et al. such as the ecosystem management model FORE-
2000). Whether or not such humus accumulations rep- CAST. These models are discussed in Chapter 21.
resent fertile or infertile soil conditions will be related to
whether the disturbance event that creates a new young
forest with high nutrient uptake demand in its early 11.8 Soil Development
years also activates decomposition processes, thereby re-
leasing some of the accumulated nutrient capital when it Major Factors That Determine
is most needed. Once the trees have absorbed sufficient Soil Development
soil nutrients to satisfy much of their nutrient require-
The nature of a soil depends on several factors: the
ment for new growth by internal nutrient cycling, re-
character of the materials from which it develops, its
duced nutrient availability as the raw humus layer
topographic position in the landscape, the climate, the
rebuilds may not be as inimical to growth as suggested
biota (mainly the vegetation), and how long soil devel-
by a snapshot soil evaluation.
opment has been occurring.
This very dynamic character of forest soil fertility,
in which fertility is the result of a temporal balance be-
tween periods of high nutrient uptake demand and pe- Materials From Which Soils Develop. ‘The char-
riods of soil nutrient supply, suggests that soil fertility in acter of the soil is markedly influenced by the geology
many forests may be assessed more accurately by using and mode of origin of the parent material from which

Table 11-5 Principal Visual Symptoms Accompanying Severe Instances of Deficiency as Observed in Both
Culture- and Field-Grown Trees

Element General Appearance

Nitrogen General chlorosis and stunting of needles, which increases with severity of deficiency; in the most severe
cases, needles are short, stiff, yellow-green to yellow; in some cases, purple tipping followed by necrosis of
needles occurs at end of growing season.
Phosphorus Youngest needles green or yellow-green, older needles distinctly purple-tinged; purple deepens with severity
of deficiency; in very severe cases in seedlings, all needles purple.
Potassium Symptoms vary: usually needles short, chlorotic, with some green near base and, in some severe cases, pur-
pling and necrosis with top dieback; sometimes little or no chlorosis, but purpling, browning, or necrosis of
needles evident.
Calcium General chlorosis followed by necrosis of needles, especially at branch tips; in severe cases, death of terminal
bud and top dieback; resin exudation.

Magnesium Yellow tipping of current needles followed in severe cases by tip necrosis.
Sulfur General chlorosis of foliage followed in severe cases by necrosis.
Iron More or less diffuse chlorosis; in milder cases, confined to new needles; in more severe cases, bright yellow
discoloration with no bud development.

Manganese Needles slightly chlorotic; in severe cases, some necrosis of needles.

Boron Tip dieback late in growing season with associated chlorotic-to-necrotic foliage, intergrading to dieback of
leading shoot with characteristic crooking.

Zinc Extreme stunting of trees with shortening of branches; needles yellow, short, crowded together on twig, some-
times bronze-tipped; older needles shed early, with resultant tuffing of foliage; in severe cases, trees rosetted
with top dieback.

Copper Needles twisted spirally, yellowed or bronzed; “tipburn” or necrosis of needle tips evident; in severe cases,
young shoots twisted or bent.

Molybdenum _ Chlorosis of leaves followed by necrosis of tissue, beginning at tip and eventually covering whole leaf.

Morrison, 1974.
316 CHAPTER 11 Soil

it develops. The geological origin determines the min- Topography. Topography—slope, aspect, elevation,
eral content, the resistance to weathering, and the re- and position in the landscape (ridge top, midslope, or
lease of nutrients during weathering of parent valley bottom)—exerts a strong influence on soil de-
material. Soils derived from light-colored, quartz- velopment. Steeper slopes and ridge tops undergo
rich, decay-resistant plutonic rocks are generally more rapid erosion and consequently often have thin-
coarser textured and less fertile than soils derived from ner soils with a coarser texture than do more gentle maa
doe
es
Ne
oe

dark-colored geological materials such as volcanic slopes. Water drains rapidly from such sites; therefore,
the soils are drier. Colluvial action on steep slopes stirs
—
rocks that are rich in easily weathered minerals. Sand-
stone rocks derived largely from quartz particles will the soil, restricting the development of characteristic
yield less fertile soils than siltstones and shales. Lime- horizons. Valley bottom and lower slope soils are
stones will give rise to richer soils than granite. moister, finer textured, and deeper because they re-
The mode of origin of the parent material influ- ceive additions of fine-textured materials eroded from
ences the texture and coarse fragment content of soil: upslope and are more fertile either because they re-
ceive additional nutrients in downslope seepage water
1. Residual soils are developed by in situ weathering of and/or because the improved moisture status increases
bedrock, are usually finer textured at the surface, the availability of nutrients by favoring soil weathering
and become less weathered and have a higher and litter decomposition. Equatorward aspects (.e.,
coarse fragment content with depth. southern aspects in the northern hemisphere) produce
2. Transported soils tend to be more uniform in hotter, drier soils than poleward aspects. High-eleva-
textural composition with depth but can vary tion soils are generally cooler than low-elevation soils
greatly in texture. Soils derived from wind- and may be wetter in the growing season because of
transported materials (aeolian or Joess soils) and snowmelt water.
from materials deposited by water in lakes
(lacustrine) tend to be uniformly fine textured Climate. Given sufficient time, climate may ulti-
(clays, silts, and fine sands). Materials deposited in mately become the dominant factor controlling soil
flood plains by rivers (a//uvial) are fine to coarse development. This is referred to as the zonal concept of
textured (silts, sands, and gravels). Materials de- soil development: a zonal soil is one whose character-
posited by glacial melt waters (g/acio-fluvial) are istics are determined predominantly by the regional
normally very coarse (gravels and stones). The macroclimate.
coarse fragments in these water-transported mate- In areas where there is substantially more precipi-
rials are generally well rounded. Materials trans- tation than evaporation, at least for a major part of the
ported and deposited by glaciers (glacial till) are year, there is net drainage down through the soil. This
characterized by an unsorted mixture of a wide results in the leaching of certain chemicals from the
range of particle sizes from large stones to clay, the surface to lower layers and the development of soil
particular mixture depending on the geological na- horizons, horizontal layers that differ in physical
ture of the material. The coarse fragments vary and/or chemical properties. The abundant moisture
from well rounded to sharp edged. The texture of facilitates the development of forest vegetation and a
colluvial materials that have been transported characteristic accumulation of organic matter at the
downslope under the influence of gravity will surface of the mineral soil, the forest floor (called the
depend on the nature of the material before LFH, or O, horizon). Decomposition of the organic
downslope movement (e.g., glacio-fluvial, till). matter in the forest floor produces organic acids,
Talus parent materials are composed of rock frag- which are leached down to the surface of the mineral
ments that have been dislodged from cliffs or steep soil (the A horizon), where they accelerate the weath-
rocky slopes and accumulated downslope. They ering of primary minerals. Salts produced by this
are extremely coarse textured and generally unsta- weathering are then leached (a process called
ble, although they may overlie more finely tex- eluviation) farther down into the mineral soil, either in
tured colluvial materials. Colluvial materials can solution or chelated by soluble organic matter. This
move downslope either slowly or rapidly; the for- removes colored substances such as iron, leaving a pale
mer is referred to as soil creep, and the latter is re- or bleached surface horizon (the A, or As, also called
ferred to as mass wasting. the e/uvial horizon). Much of the eluviated material is
 Soil Development 317

deposited in the next lower layer, where it produces a eral cations from these soils, they tend to be acidic (lost
strongly colored horizon: the B, or illuvial, horizon. cations are replaced by hydrogen ions), at least in the
The coloring can also result from the release of iron surface horizons, and they are sometimes less fertile
by weathering of minerals in the B horizon. than the other major group of soils: the pedocals (incom-
Such soils have been broadly referred to as pedalfers pletely leached soils; e.g., chernozems, some brunisols).
(leached soils; e.g., podzols, some brunisols, luvisols) Pedocal soils develop where the combined evapo-
(Figure 11-15). Because there is a net leaching of min- ration and transpiration exceed precipitation for

SOIL-FORMING Rainfull
PROCESSES

Rain water enriched

in H® and organic
acids by coniferous
vegetation.

CHARACTERISTIC
THICKNESS
SOIL HORIZONS
Acidic, slowly
decomposing litter
releases organic acids Dark brown to black-
(e.g., fulvic acid). “Litter layer: L 5 = 1OOlcn
Decomposing layer: F
Accumulation Humus layer: H
of layer of “raw” AS AVS :
humus. Forganie matter,” \\ WCE Se ereNe 0— 15cm
fuimiatyM ; < Fa ass Eluviated horizon: Ae
Iron, aluminum, mee AP eS BS
soluble organic Dark red brown
matter, and mineral to reddish-orange 10 — 60 cm
elements leached Mluviated horizon: B
out of forest floor,
and the upper Yellow to grey-
mineral horizon slightly weathered variable
which loses color. horizon: BC

bore es . RESELL \N Horizon influenced tens

aaah ney RV RROvyg7;5 by ground water
organic matter. VSS Sy
Yy
x fy Unweathered rock
4 OX OSG or compacted variable
Loss of water from Os Uh Ea Meg
the rooting zone
containing
dissolved chemicals
(e.g., NO3K *Ca**)

Figure 11-15 Characteristic major horizons and soil-forming processes of a typical conifer-
ous forest podzol, a pedalfer type of soil. (After Clarke, 1967. Copyright 1967 by John Wiley & Sons,
Inc., New York. Used by permission.)
318 CHAPTER 11 Soil

Precipitation
Transpiration

SOIL-FORMING
=e a
PROCESSES

pn ty \ Wy Uy \\ Wy Ty }
Calcium-rich \ Wi } h SOIL HORIZON THICK
litterfall, dominated ytt
a \{I}
by root mortality of 3 = = WK <== Dark brown to black
grasses and other herbs, °#-.'s577 j cay Torey Fi Se: organic-mineral
enriches the upper eg 6 \ CONT ( ¥*- horizon: Ah 10-40.
mineral soil with Beres Yy Vi} a Ae
organi matter. rae ee OSE

“ ae +) \Uptake’of ne ca : Pilea
Soluble, organic i: “4a ~ nutlients b Fea«Soluble organic PN nue
CoS ses
ts ee 4» » plant Dots i bate and inorgai are spies
and inorganic
materials leached me tec riseon nek ae eer Preis Sane HS eer Minera orizon
downward. Ns P water
y : 2 > :~*- >.>
PS stained brown : by
ce Sec cee -o.* soluble organic
‘ i a Na cas rs j +, .7o matter: Bm 0-20 ¢
Most of water ie Spe : a igPTs
<a cas Fo Sode a ik Witanw &
and dissolved x . os \ \a cose: wa Deposition of soluble. oe ~ veeas a 2 Greyish or
minerals returned ne SN ea cokes - Organic matter : ", tat, yellowish-grey
toward surface wee oN ed =: e Se, . ‘~ . mineral horizon
by upward water Rae ES Ne Le -.-© enriched in calcium
movement and riage iene os‘ ; oe Bete carbonate: Ck variat
plant uptake. jc os ans alien ‘sition ofesicina ees Totes
tng sare iS Sat te and other soluble“. ‘., Compacted layer
Deposition of organic ‘.. Ae aieN as ea inorganic chemicals -, + *: of calcium
matter and calcium Res Sa ; cm ta heres © B 6t carbonate: Cca variat
carbonate at limitof =.» Sid det cook asst
penetration of the ate oan oe ore
majority of soil water... ik PN CAN y+; Unaltered parent
ee Abe, On “sy-. 275) material: C variat
= a SS ett ‘ Soe
Z Unweathered
rock or compacted
YY material variat

Y
Figure 11-16 Characteristic major horizons and soil-forming processes of a typical grass-
land chernozem, a pedocal type of soil. (After Clarke, 1967. Copyright 1967 byJohn Wiley & Sons,
Inc., New York. Used by permission.)

most of the year. The soil is rarely wet enough to per- herbivores, and a high frequency of fire. Most of the
mit appreciable drainage to the groundwater table. soil organic matter that accumulates in these soils is in
Products of weathering are leached downward from the upper mineral horizons as the result of production
the surface to deeper horizons during a rainfall but are and turnover of roots. The annual addition of large
returned to the surface either via uptake by plants or quantities of dead roots to the surface layers of grass-
by upward capillary movement as the surface soil lay- land soils results in the formation of a characteristic
ers dry. The drier, hotter climate of the areas where dark-colored, organic-rich surface horizon. Such soils
such soils develop usually produces savanna forest, are called chernozems (Figure 11-16).
grassland, or desert types of vegetation. There is little
or no surface accumulation of organic matter in these Biota. Although climate has an important direct in-
types because of one or more of the following: rapid fluence on soil development, it has an equally impor-
and complete decomposition, low productivity (and tant indirect influence because of its effect on the
hence low aboveground litterfall), heavy browsing by fauna and flora. Two areas of the same parent material
 Soil Development 319

and the same climate will produce two different types considering the development of residual soils from
of soil if one is maintained under a conifer forest and solid rock. In less extreme environments and/or on
the other under herbaceous vegetation. Differences in materials in which pedogenesis is rapid, characteristic
the patterns and extent of rooting, in nutrient uptake, horizons may develop much faster. Perceptible pod-
in the chemistry and decomposition of the litter, and zolization can appear in sand dune soils in 100 to 200
in the associated soil fauna and microflora lead to dif- years, especially under forest, although 1000 to 1500
ferences in horizon development. Pedocal-type soils years is considered necessary for the development of
can develop into pedalfer-type soils if the vegetation is typical podzols (Lutz and Chandler, 1946). In many
altered (e.g., afforestation of grasslands with conifers, situations, zonal soils may never develop because
and vice versa). The effects of the presence or absence processes of soil additions, soil erosion, and soil mix-
of earthworms on the soil profile beneath a given veg- ing occur more rapidly than climatically determined
etation type is illustrated in Figure 11-12, and Figure soil horizon development or because factors of topog-
11-17 shows the effect of changing from a predomi- raphy, parent material, fire, and biota exert an overrid-
nantly conifer to a predominantly deciduous forest ing influence on soil development. This is leading to a
cover. Such a change may require a long period of decline in the popularity of the zonal concept of soils,
time to occur, or it may be relatively rapid, as was the which nevertheless remains a useful generalization in
case in Figure 11-17A. some regions.
Over the past few millennia, humans have become
an important factor in soil development. Plowing and
fertilization of agricultural soils gives them a character Rates of Development in Different
that is distinct from that of unplowed soils. Erosion Types of Parent Material
resulting from agriculture, forestry, fire, and other as- Residual soils generally develop slowly (although the
pects of human stewardship of the earth has altered or rate is greatly influenced by the climate and vegeta-
removed soils in many parts of the world, stripping tion), and in the absence of major disturbance, they re-
material from slopes and filling valleys and lakes. main approximately the same over prolonged periods
Human effects certainly constitute one of the major (e.g., many centuries). Ice-transported materials also
biotic factors that influence the present-day character change slowly once the ice has retreated, little change
of the world’s soils. occurring over several centuries after an initial period
An important aspect of forest soil development is of more rapid change (discussed in Chapter 17). In
the mixing of horizons that accompanies the uprooting contrast, types of parent material that are the result of
of trees. In windy areas, this may happen sufficiently continuing deposition are subject to continuing
frequently that the normal development of soil hori- change, as in the case of alluvial deposits. Strang
zons is prevented completely. In many forest areas, it is (1973) estimated that silt deposits in a northern Cana-
characteristic for the horizons to be discontinuous or dian valley were being added at a rate of between 0.3
to occur in unexpected vertical sequences as the result and 1.3 cm yr'. Aeolian materials such as loess or sand
of the upheaval of tree roots. Even though windthrow dunes are frequently subject to additions, losses, or
or the uprooting of dead trees may happen relatively movement, whereas colluvial soils are frequently sub-
infrequently in terms of the human time scale, the re- ject to movement and mixing.
sultant churning action is often rapid relative to the Figure 11-18 shows the changes in organic car-
rate of pedogenic processes (pedogenesis, or soil forma- bon, nitrogen, sulfur, and organic phosphorus in sandy
tion) that are responsible for horizon differentiation. soils of aeolian origin in New Zealand varying in age
from 0 to 10,000 years since deposition. After a rapid
Time. Soil-forming processes are generally slow gain of carbon, nitrogen, and organic phosphorus dur-
compared with most biological time scales, although ing the first 500 to 1000 years, the rate of accumula-
the time required for the development of characteris- tion slowed down, especially for phosphorus and
tic soil properties is highly variable. In extreme envi- nitrogen. Initially, the increases were restricted to the
ronments (e.g., very hot, cold, arid, saturated) or surface horizons, but with time, the lower horizons
where one is dealing with structureless or consolidated also became enriched in these nutrients, which are still
materials, many millennia may be required for the accumulating in the soil even after 10,000 years. Cu-
processes of pedogenesis to result in the development mulose (organic) soils can develop over rock surfaces
of characteristic horizons. This is especially true when rapidly (several centuries) in humid environments or
 Soil beneath a 60-year-old Soil Depth
eastern white pine stand (cm)
gL 10

=F Soil beneath a mixed

5 deciduous hardwood stand
18 years after logging

Panetey |
ee at Mineral soil Surface — ——
ec Ah
Bf
5

10
BC =| BC
15

Eastern White Pine Deciduous hardwood communities

L layer
6 |
MMU Lr R

floor
(cm) A I ea
a
a a a a ar = I,
forest
of
Depth
0 . TFN SS We
= Oak
k Bee
e En B, horizon BS _= nee seas EIEN NS
2 10 oe a ‘ Oy Variation in
3 s ie) | _ Teh Maple o< depth of B,
= = “Se ansereteensnsanracasaenes serene ey in
& : | Ash Elm Birch Basswood’ three hardwood
£ 30 | communities
= B, horizon : x
g a6 Conversion from
Gg a0 conifers to hardwoods |
by
= 50
10 20 30 40 50 60 70 80 90 “100 101205 "130 140 160
Time in years from abandonment of an old agricultural
field and the invasion of eastern white pine
(b)

Figure 11-17 Effect of vegetation change at Harvard forest on characteristics of the soil pro-
file. (A) Logging of a 60-year-old eastern white pine stand, growing in New England on an acid pod-
zol with a thick (10 cm) mor forest floor, was followed by invasion of mixed deciduous hardwoods
(birch, oak, maple, ash, and cherry). Eighteen years later, after 12 to 15 years of occupancy by hard-
woods, the forest floor was mull with <1 cm surface organic accumulation, and the upper mineral soil
was enriched with humus as a result of abundant earthworm activity. (After Fisher, 1928. Used by per-
mission of the Ecological Society of America and Harvard Forest.) (B) Changing characteristics of the soil
profile in an age sequence of white pine stands that developed on abandoned fields, and in an age se-
quence of mixed hardwood stands that developed after the pine stands were logged. Organic matter is
transferred from mineral soil to forest floor in the pine sequence, whereas the reverse occurs under
hardwoods. The depth of the B, (“dark brown” horizon) varied in proportion to the rate of decompo-
sition of leaf litter. In ash-elm—birch—basswood stands, litter decomposed rapidly. In oak—beech-pine
stands, it decomposed more slowly. (After Griffith et al., 1930. Used by permission of Harvard Forest.)

320
 Ecological Significance of Soil 321

for plants; a refuge and a place to live for soil animals;

and a substrate for aboveground terrestrial animals.

i)j=)

Organic carbon Importance for Plants

Anchorage. A rough surface such as that provided
by soil is necessary for colonization by most vascular
plants. Seeds that fall on bare rock are usually washed
——
—_—— or blown away before they can become established.
Se
Once established, the plants must be able to hold their
S oe Organic phosphorus _,,,.++*
photosynthetic organs in the appropriate position and
Nitrogen
arrangement relative to the sun’s rays. This generally
requires a firm anchorage for the roots. Where there is
little or no soil, adequate anchorage may be provided
ma 102456108
ha!)X10”
(C
10)
Pex
Nex
Weight
(kg
elements
of
depth
|to by the penetration of roots into cracks and crevices,
and trees that grow on rocky ridges can be very wind-
firm if the rock is fractured. Such roots may penetrate
more than 20 m down into fractured or eroded rock.
0 5,000 10,000
Where there is no mineral soil and the bedrock is un-
Age of soil, year fractured, a surface accumulation of organic material
Percent Change in Soil Chemical may suffice for anchorage, although trees are generally
Content Per Year susceptible to windthrow on such sites. As a result,
areas of thin soil over unfractured bedrock often re-
Soil Age Organic Organic
main unforested for a long time after a disturbance.
(yr) Carbon’ Nitrogen Sulfur Phosphorus
The relationship between soil characteristics and the
0-50 +4.64 +2.76 —0.48 +0.66 stability of plants is complicated by the fact that plants
provide each other with mutual shelter and support.
50-500 +0.36 {O27 +0.02 +0.52
‘Trees in dense forests have interwoven root systems and
500-3000 +0.06 +0.04 +0.03 +0.03
crowns, which may provide stability in situations in
3000-10,000 +0.01 +0.01 +0.01 +<0.01 which isolated root systems on their own are inadequate
Figure 11-18 Weights of organic carbon, nitrogen, sulfur,
to support the aboveground parts. The stability provid-
and organic phosphorus in mineral soil of different ages, ed by root anchorage often proves to be inadequate in
derived from windblown sand in New Zealand. The maxi- the face of strong wind once the continuity of such a
mum rate of change occurred during the first 50 years, very stand is broken. This can happen when trees are re-
slow changes occurring after 500 years. (After Syers et al., 1970. moved during road construction and when previously
Used by permission ofBlackwell Scientific Publications, Ltd., Oxford, sheltered trees are exposed at the edge of a clearcut.
England, and the authors.) Uulization of soil by roots is strongly conditioned
by both physical and chemical conditions. Poor aera-
tion, physical compaction, chemical cementation (iron
more slowly (several millennia) in drier climates as the or calcium pans), high water tables, toxic chemical con-
result of the natural succession from a shallow pond or ditions, and low soil temperatures can restrict root ac-
lake to a fen or bog. Several more examples of tempo- tivity to a shallow surface layer of soil. Consequently,
ral changes in soils are given in Chapter 17. deep mineral soils do not necessarily provide better an-
chorage and wind stability for trees than do shallower
soils. Forests on deep but saturated (either permanent-
11.9 Ecological Significance of Soil ly or temporarily) soils are often more susceptible to
windthrow than are trees on thin rocky soils.
Soils provide a variety of essential resources for plants
and, hence, for animal life. These include anchorage, Supply of Moisture. One of the major ecological
moisture and nutrient supply, and adequate soil aeration roles of soil is to act as a sponge: to absorb and store
322 CHAPTER 11 Soil

water. A major reason for the lack of vegetation in areas is almost permanently saturated, which makes it un-
that lack soil of any kind is the lack of moisture storage. suitable for most plants.
Life is thought to have evolved in water, and only when Soil texture becomes critical at the dry end as well
adaptations such as roots and lungs evolved did terres- as at the wet end (saturation) of the soil moisture spec-
trial life evolve. These adaptations maintain at least a trum. We have already seen that although fine-
part of the organism in a humid environment where textured soils generally hold more water than coarse-
the vital process of chemical exchange can occur. textured soils, they may sometimes be drier in terms of
Soils that have extremely low moisture storage ca- moisture availability to plants. Brayshaw (1970) found
pacity (e.g., coarse gravels, pumice soils) may remain that Douglas-fir (only moderately drought tolerant)
unvegetated or carry only a sparse growth of drought- was restricted to coarser soils than ponderosa pine
resistant plants. Soils that have an excess of water are (very drought tolerant) in the hot, dry, southern inte-
also unsuitable for most forms of plant growth because rior of British Columbia because the coarser-textured
they lack the oxygen necessary for root growth and soil had higher soil moisture availability than the
frequently have soil chemical conditions that are toxic finer-textured soil (Figure 11—4).
to plant roots. Conversely, soils that are well drained
but that remain moist throughout the growing season Supply of Nutrients. We have already seen that
are generally highly productive and support lush vege- plants obtain the nutrients required for new growth in a
tation. These are useful generalizations, but it must be number of ways: by absorption from the atmosphere
remembered that one cannot predict the composition and soil solution, from weathering of soil minerals, from
and productivity of the vegetation of a site merely on decomposing organic matter, and by internal redistribu-
the basis of soil moisture conditions. Soil structure, tion of their existing capital of nutrients. It has also been
porosity, aeration, and the local climate are all impor- noted that mycotrophy is almost universally important.
tant in determining the relationship between plant Trees obtain most of their annual uptake of most
growth and soil moisture conditions, and optimum nutrients from the biogeochemical cycle, and the
plant growth occurs only in soils with the appropriate major contribution of the soil to this cycle is usually
balance between air-filled and water-filled pores. the forest floor (especially for nitrogen and phospho-
Soil organic layers are often particularly favorable rus). The geochemical cycle adds to and sustains the
to good moisture status because they can hold a lot of biogeochemical cycle, the relative contributions of
water but also contain a lot of larger air spaces. In cool, the different geochemical inputs varying from site to
humid regions where forest floors can accumulate to site. The contribution of the mineral soil depends on
depths approaching | m and where they commonly ex- its mineralogy, its physical and chemical characteris-
ceed 20 cm, potentially xeric sites may in fact have a fa- tics, its water relationships, and the distribution of
vorable moisture status. For example, many productive the roots; in some cases, the contribution may be
forests in very humid climates in western Canada are very minor. For example, forests with a well-devel-
growing on thick accumulations of organic matter over oped forest floor, an active biogeochemical cycle, and
solid rock. These stands frequently achieve levels of or- a humid climate may grow productively on rocky
ganic production rivaling that of adjacent stands that landscapes that have little or no mineral soil; inputs
are growing on glacial till soils several meters deep. In from the geochemical cycle via /ithoponics are the
a dry climate, this is less common because of the much major source of mineral nutrients on such sites. Such
thinner forest floors and the general lack of moisture. examples notwithstanding, the mineral soil is nor-
Under dry conditions, organic layers are generally in- mally a major source of nutrients for plants. Also, the
ferior to mineral soils in providing moisture for plants. presence and the physical and chemical condition of
Organic layers can dry rapidly because of the large air the mineral soil have an important influence on the
spaces, and the lost moisture is not replaced efficiently soil fauna, which in turn affect nutrient cycling and
from the underlying mineral soil because of the hy- soil fertility.
draulic discontinuity between the mineral and organic From this brief discussion, it should be clear that it is
layers, because dry organic matter has a low hydraulic difficult to generalize about the ecological significance
conductivity and because it may become hydrophobic of soil. The answer to the question, “Do plants require
(water repellent). Organic soils can also pose problems soil?” requires a consideration of what plants require
of excess moisture. Soils derived from sphagnum moss and how they are able to satisfy these requirements.
may retain so much water that the surface organic layer Under certain climatic and geological conditions, plants
 Importance of Soils for Forest Management — 323

are able to satisfy their minimum requirements for an- rocky areas unless there are natural caves, and animals
chorage, moisture, and nutrients in the virtual absence do not hibernate in areas of frozen soils because of the
of mineral soil. However, before this can occur, a con- difficulty of excavating suitable shelters. Animal ap-
siderable organic accumulation is us-ually necessary, and pendages are adapted for the type of substrate that
the highest levels of plant production are normally at- they contact; animals that walk on moist, low-bearing-
tained on a well-structured, medium-textured, well-aer- capacity soils have larger feet than those accustomed to
ated mineral soil with a rapidly decomposing surface dry, firm ground. Animal nutrition is influenced by
accumulation of organic matter. soils because plant chemistry reflects soil chemistry. In
The organic and mineral soil horizons contribute humid climates or areas with low soil-sodium levels,
differently to the needs of plants in different situations, plants may be short of sodium and other minerals. In
but rarely can plant production be improved by the some areas, micronutrients are in short supply in the
large-scale removal of any of the soil horizons. There soil, and this is also reflected in plant chemistry. Ani-
are some exceptions, of course. Where the soil in the mals that feed on plants from such areas may develop
rooting zone is completely weathered and largely de- deficiency symptoms unless they can make up the defi-
pleted in nutrients (this occurs in some very old resid- ciencies by eating soil or licking mineral deposits. The
ual soils in the tropics), erosion of the nutritionally availability of such sa/t licks may be a major factor in
exhausted surface layers can expose the less weathered determining animal abundance, and animals will travel
deeper soil layers, which have greater reserves of un- considerable distances to get to a lick. Soils in alpine
weathered minerals. This may result in increased plant environments are typically short of sodium, and it is
production. Similarly, excessive accumulations of cer- not uncommon to have the straps of a pack-sack or a
tain types of organic matter can render a site swampy, wooden spade handle gnawed by salt-hungry animals.
cold, and/or depauperate in available nutrients. Re-
moval of some or all of such organic layers can im-
prove moisture, temperature, and nutrient conditions. 11.10 Importance of Soils for
The complexity of soil—plant relationships has led Forest Management
some ecologists to focus their attention on the plant
components of the ecosystem rather than on the Soils are a major determinant of the productive poten-
plant-soil complex. The often overriding influence of tial of forests. They are one of the least renewable
climate, the importance of light and the frequently components of the renewable forest resource. (For a
dominant role of light competition in controlling plant discussion of the renewability of resources, see
community development, and the ability of plants to Chapter 22.) Certainly, there are other ecosystem
modify the moisture and nutrient conditions of the soil components with low renewability, such as the genetic
sometimes create the impression that plants can be- constitution of the biota. If an entire gene pool (e.g., a
have relatively independent of the soil, and this may species) is lost, then it will probably never be re-
lead to reduced concern about the soil resource. How- created, but of all the physical components of the for-
ever, although it is true that plants can create their own est ecosystem, soil is generally the hardest to replace
soil environment (see Chapter 17), there can be little once lost. Biota can normally be reintroduced from
question that soil conditions exert an extremely impor- other areas. Except on a limited basis, this is not usual-
tant influence on the biotic composition and produc- ly possible for soil.
tivity of ecosystems. In Chapter 13, we examine the Much of the really fertile soil on moderate topogra-
distribution of plants along environmental gradients phy in areas with climates that are suitable for high-
that are largely gradients of soil conditions, and we will production forestry have already been sequestered by
see that soil-plant relationships, although complex, are agriculture. Our desire for food has traditionally over-
a dominant feature of most ecosystems. The ecology of ridden our desire for forest products. The forester is
an area cannot be understood adequately unless there frequently left with the less productive soils, soils that
is an appropriate understanding of the soil. for reasons of topography, texture, depth, mineralogy,
moisture status, chemistry, temperature, fertility,
and/or economics are not required for food production.
Importance for Animals Because of these limitations, the forester generally has
Animals are affected by soils both directly and indi- fewer options for manipulation of soil physical and
rectly. Burrowing animals cannot live in very stony or chemical parameters than does the farmer. Certainly,
324 CHAPTER 11 Soil

the forester can fertilize some types of sites, and the use along which the logs are moved. According to the
of fire or mechanical treatment may improve the fertil- topography, the front end of the log may be raised off
ity of some areas after harvesting, but compared with the ground, but often the log is in contact with the
most farmers, foresters must generally produce their ground along most of its length for much of the trip
crops with the soil resource that they have inherited. from stump to landing. The highest degree of soil dis-
Forest harvesting is an engineering activity that is turbance is associated with tractor or skidder yarding, in
rife with possibilities for adverse effects on the soil but which a wheeled or tracked vehicle travels to the log
that can also have beneficial effects. There are a vari- and then drags it to the landing by cable.
ety of potential impacts on soils. Table 11-7 presents some results of a detailed
study of soil disturbance, vegetation cover, and regen-
eration on clearcuts in the mountains of southeastern
Soil Disturbance British Columbia. Both yarding method and season of
Forest harvesting generally produces disturbance to logging were important in determining the degree of
the soil surface. This can vary from a slight disturbance soil disturbance. Roads (main haul roads and skid
of the forest floor to the complete loss of the upper soil trails) constituted by far the most important source of
horizons. The most important factor determining the soil disturbance. Skyline yarding, which may not re-
degree of ground disturbance is the type of harvesting quire the harvested area to be roaded, produced the
equipment used, followed by the time of year. Table smallest degree of soil disturbance. Ground skidding
11-6 shows a comparison of the degree of soil distur- produced the highest degree of disturbance. However,
bance associated with various different extraction there is a wide variation in the extent of disturbance
methods (yarding) in the western United States. associated with this method in different regions.
The least soil disturbance occurs with skyline
yarding, in which the logs are moved from the stump to
the landing (point of loading onto trucks) sus- Effects on Soil Stability
pended in the air by a taut cable. Relatively few data are Unfortunately, damage to soils by clearcutting is not
available on the soil effects of horse yarding, but the ex- limited to surface disturbance. On steep slopes, the sta-
tent of soil disturbance seems to be relatively small. bility of the soil may be reduced to the point at which
High lead yarding, perhaps the most common method mass wasting occurs (slumps, slides, and debris avalanch-
on the west coast of North America, produces signifi- es). A study of soil mass movements in the H. J. Andrews
cantly more soil disturbance than does either of the pre- forest in the Cascade Mountains of Oregon revealed
vious two methods. In high lead yarding, the logs are that 83% of all soil mass movements were associated
dragged to a central landing by a slack cable, and there with slopes of more than 45% (248) and that 72% of all
is considerable soil disturbance on the “yarding roads” events and approximately 50% of all soil material moved
were associated with roads; 17% of the observed events
Table 11-6 Degree of Soil Disturbance Associated were in logged areas between roads, and 11% occurred
With Various Methods of Harvesting Trees in Western in areas not disturbed by humans (Dyrness, 1967).
North America* A study of the impact of clearcutting and road con-
Total Soil Mineral Soil struction on soil erosion by landslides in the western
Method Disturbance, Exposed, Cascade Range in Oregon demonstrated the impor-
of Logging % of Area % of Area tance of the nature of the underlying geological mate-
rials (Swanson and Dyrness, 1975). At elevations above
Horse 12 12 900 to 1000 m in the H. J. Andrews experimental for-
Tractor 21-87 21-69 est, the soil is underlain by lava-flow bedrock. This
Jammer 15 igo: type of material produces a stable soil mantle that has a
High lead 11-90 5-56
low susceptibility to landsliding after clearcutting, and
in the 1950 to 1972 period, only two small road-related
Skyline 6 6
slides occurred in clearcut areas. At lower elevations,
Helicopter 36 5
altered volcanic-clastic rocks have a very unstable soil
*Data from Bockheim et al., 1975; Dyrness, 1965; Garrison and Rum- mantle. Clearcutting in this area has increased slide
mell, 1951; Ruth, 1967; Wooldridge, 1960. erosion by 2.8 times from prelogging levels. Along
 Importance of Soils for Forest Management 325

Table 11-7 Soil Disturbance Associated With Forest Harvesting and Postharvest Site Treatment in Southeastern
British Columbia

Area Disturbed by Roads, % Area Disturbed Between Roads, %

Logging Method Slashburned Not Slashburned Slashburned Not Slashburned

Summer ground skidding 36 42 4

Winter ground skidding 19 25 5
Summer high lead 15 fs) 14 8
Winter high lead _ 16 ~ 0.5
Summer grapple yarding _ 27 - 2
Winter grapple yarding - 22 - 0
Summer jammer yarding — ~ 5
Summer skyline _ _ 8

Data from Smith and Wass, 1976.

road rights-of-way, the increase has been 30 times. and partly to the loss of root strength as stumps and
With roads covering approximately 8% and clearcuts large roots decay. Roots are believed to play a major
92% of the harvested area, roads contribute approxi- role in holding soil mantles in place on steep slopes in
mately half of the management-related slides. Roads areas such as western North America. After clearcut-
and clearcuts combined have increased slide activity by ting, there is a delay of 3 to 10 years in the onset of
approximately five times over a 20-year period. slides, which corresponds to the timing of loss of root
A study of landslides in the Maybeso Creek Valley strength (Bishop and Stevens, 1964; Fujiwara, 1970;
in southeastern Alaska revealed a dramatic effect of Rice and Foggin, 1971; Swanston, 1969, 1974; ‘Tur-
clearcutting on the incidence of soil mass movements. manina, 1965). This is followed by a period of several
Most of the increase occurred on the slopes of steep, decades of soil instability until new root systems are
V-shaped side drainages, but there was also some in- formed, which restabilize the soil mantle (Figure
crease on smooth uniform slopes of the main valley 11-19). Loss of root strength was reported to con-
(Table 11-8). tribute to soil mass movements after high-elevation
The increase in slides is due partly to the disrup- clearcutting in southwestern British Columbia (Utzig
tion of slope hydrology and slope stability by roads and Herring, 1975).

Table 11-8 History of Landslides’ in the Maybeso Creek Valley Near Hollis, Southeastern Alaska, Illustrating the
Effect of Clearcutting on Soil Stability
Length of Number of Number of Total Area Area Per
Time Period Period, yr Slides Slides of Slides, ha Year ha yr“

1848-1948 (approx.) 100 12 0.12 aed 0.11

1948-1952 4 1 0.25 1.6 0.40

Clearcutting started 1953

1952-1959 7 20 28) 2S) 0.36

1959-1961 2 28 14.0 ll 3.85

1961-1962 1 68 68.0 36.8 36.8

Source: Bishop and Stevens, 1964. Used with permission.

@The number and area of slides were monitored by aerial photography.
326 CHAPTER 11 Soil

an

Unstable soil type

Western redcedar

300

200
to

of
kg
strength
Tensile
roots,
*
em
Stable soil type
forest
relative
landslides
of
Frequency
uncut
to
0
0 2 - 6 8 0 10 20 30
Years since clearcutting Years since clearcutting and road construction
(a) (bd)

Figure 11-19 (A) Loss of root strength in a clearcut in coastal British Columbia. (/ter
O'Loughlin, 1974. Used by permission of the National Research Council of Canada and the author. See also
O'Loughlin and Ziemer, 1982; Ziemer, 1981.) (B) The consequent period of soil instability on
steep slopes. The graph shows relative frequency of landslides on stable and unstable soil types in
the western Cascade Range of Oregon. The curve for unstable soil is based on data; that for stable
soil is hypothetical. (After Swanson and Dyrness, 1975.)

summer, can render seedlings susceptible to drought

Loss of Soil Organic Matter and Nutrient injury or death. Soil disturbance that exposes mineral
Content and Changes in Soil Chemistry soil and the reduction in the depth of forest floor that
These topics have already been covered in the discus- generally occurs after logging can have beneficial ef-
sions of effects on biomass and energy (Chapter 4) and fects on soil temperature and plant productivity in such
on biogeochemistry (Chapter 5). environments. Exposure of mineral soil also reduces
diurnal temperature extremes at the soil surface with
beneficial consequences for seedlings. In summer-hot
Alteration in Soil Temperature environments, temperatures at the surface of exposed
Important changes in soil temperature occur as the re- but undisturbed organic layers may increase to levels
sult of clearcutting. The extent of these changes de- that are lethal for many plants (see Chapter 8).
pends to a considerable extent on the degree of min-
eral soil exposure and the reduction in the depth of
the surface accumulation of organic matter. In Alteration of Soil Physical Properties
winter-cold environments, the productivity of the Use of heavy machinery can alter soil structure, poros-
plant community may be influenced markedly by the ity and density, pore-size distribution, aeration, water
late thawing and slow warming of the mineral soil. retention, infiltration capacity, and hydraulic conduc-
Roots are largely confined to the surface organic lay- tivity. The degree of alteration depends on the weight
ers, which warm up somewhat more rapidly than the of equipment and logs being yarded, the number of
deeper layers because of their lower thermal capacity times that heavy objects pass over the soil, and the soil
and greater exposure to sunlight. Where these organic water content. In one study, soil density increased ex-
layers contain nutrient-poor and slowly decomposing ponentially with increasing number of passages of a
material, the plants will experience inadequate nutri- tractor (Foil, 1965); one trip of the tractor on wet soil
tion and grow slowly. Also, the restriction of roots to resulted in as much compaction as four trips on dry soil
the surface of the forest floor, which may dry out in the (Steinbrenner, 1955). Compaction during the harvest-
 Summary — 327

ing of tree stems by tractor in Minnesota was reported Ponderosa pine Douglas-fir
to be approximately half that with full-tree harvesting
(stem plus crown) (Mace, 1970). Loss of the forest
floor exposes the mineral soil to compaction and to the
physical impact of rain, both of which can lead to loss of
Depth
soil
eroded
of surface soil structure and “puddling.” A structureless
surface layer only a few millimeters in thickness can be 50
soil
uneroded
enough to alter the infiltration rate radically and result
in surface runoff and erosion on steep slopes. i)nN

Unless an ecological approach is taken to forest soil

which
on
harvesting and site treatment, forest management is on %
30
depth
growth
of
cm),
(to
a
as
on
remaining
of
Growth
seedlings
soil
tree remaining
of
Depth
0 3 10 15 0 5 10 1 seedlings
grown
were
unlikely to attain its objectives; to a considerable ex- Thickness of surface soil lost by erosion, cm
tent, this is the result of adverse effects on the soil.
One of the essential educational requirements for Figure 11-20 The importance of upper layers of soil for
foresters and other renewable resource managers must site fertility and tree growth is shown by the loss of
surely be a sound working knowledge of soils. seedling growth when grown in pots of soil from profiles
from which progressively thicker surface layers have been
Above all, it must be remembered that in most
removed to simulate progressively more serious erosion.
cases, the productive potential of the site is vested in a
The four lines are for soil samples taken from four different
relatively thin surface layer of soil, and erosion of only areas in Washington. The significance of soil erosion for the
the surface few centimeters may be accompanied by growth of seedlings of these two tree species varies, obviously,
significant reductions in plant growth. Figure 11-20 from one site to another, and there is a difference in the re-
shows the reduction in seedling growth when seedlings sponse of the two species. (After Klock, 1982. Used by permission
are grown on soil from different depths in the profile. of the American Society of Agronomy, and the author.)
This is a measure of the growth of seedlings after pro-
gressively more serious soil erosion. A loss of the upper
2 cm of mineral soil from a site on the east side of the slide, then it may be many decades, centuries, or even mil-
Cascade Mountains in Washington was associated with lennia before new soil has developed to the extent of that
which has been lost. Thus, for most forests managed for
an 80% reduction in seedling growth.
timber production, the soil is the resource; the trees are
Knowledge of soils is important in all phases of merely a crop. In forests managed for multiple values,
forestry. The choice of harvesting method and type of maintenance of soil in an appropriate condition is key to
equipment will be influenced by soil fertility, soil stabil- sustaining the many biologically related and physical val-
ity, soil compactability, and soil erodibility. Postharvest- ues and environmental services. This requires that
ing site treatments, choice of species, and method of foresters have at least as good a knowledge of the soil as of
regeneration will similarly be strongly influenced by the the trees and other organisms, a situation that rarely oc-
soil properties of the site. The behavior of herbicides in curs. A sound appreciation of the nature and ecological
the environment is largely determined by soils, as are role of soils is equally important for ecologists who wish to
the effects and fate of fertilizers. In short, soil consider- understand ecosystems. As is the case for foresters, many
ecologists lack an adequate knowledge of soils, especially
ations should influence every aspect of forestry.
population and zoological ecologists. Because plants are so
intimately related to soils, botanical ecologists are often
better informed on this important aspect of ecosystems.
SUMMARY Soil provides plants with a variety of basic necessities
Cows and cabbages are merely crops; it is the soil that is for life: anchorage and support, moisture, nutrients, and
the essence of a farm. Crops can be replaced. Once agri- a supply of oxygen for the roots. The availability of these
cultural soil is lost, it may be economically and/or ecologi- necessities depends on the characteristics of the soil.
cally impossible to replace it or restore it to its original Soil texture, structure, porosity, and temperature exert a
condition within our lifetime. The same holds true for major influence over soil moisture conditions, aeration, the
forests. The trees and other plants are replaceable, as long potential fertility of the soil, and the ability of root systems
as the gene pool is not lost. Displaced microbes and ani- to occupy and exploit the soil. The physical properties of the
mals will reinvade harvested sites from populations in adja- soil also influence soil stability, erodibility, and trafticability.
cent areas as soon as the redeveloping vegetation provides Soil water is one of the major factors controlling
a suitable habitat, but if the soil is lost through fire or land- plant growth. The amount ofwater stored in the soil, the
328 CHAPTER 11 Soil

availability of water to plants, and the rate at which water from one place to another across the forest landscape and
is lost through drainage and evaporation all are deter- from region to region. Consequently, what is an accept-
mined by physical features of the soil, as well as the cli- able level of soil management or soil disturbance on one
mate and topography. site or in one region may be unacceptable in another.
Soil fertility—the ability of soil to provide plants with Sustaining the functional role of forest soils does not
the nutrients that they require—is a complex phenome- imply no disturbance. In climates where low soil temper-
non. It is determined by soil chemistry, soil organic mat- ature and/or slow decomposition of organic matter can
ter content, soil moisture, soil aeration, soil animal and develop, ecosystem function may decline over time un-
microbial activity, and soil temperature. These are not less both plant community ad soil are disturbed periodi-
independent factors. They all interact with and influence cally. Some level of soil disturbance to create appropriate
each other and other soil parameters. Additional com- seedbeds may be essential to maintain stand-level biodi-
plication arises from the fact that soil fertility is more versity as expressed by species richness and evenness (see
closely related to the balance of different nutrients and to Chapter 15). Conversely, excessive levels or patterns of
their rate of circulation within the ecosystem than to ab- disturbance may lower ecosystem productivity and have
solute levels of nutrients in the soil. The fertility of forest adverse effects on the biotic community and associated —

soils is often more closely related to the input of organic resource values. Much of the negative environmental im-
matter to and rate of decomposition of organic matter in pact of forestry in the past has been related to effects on
the soil than to the chemistry of the mineral layers. soils, and in areas of steep topography, much of the im- —

Soil is the product of biological processes acting in pact on soils has been related to roads. Because of the ur-
concert with physical and chemical processes to alter soil gent need to achieve sustainable forest management, an
parent material. These biological processes involve the improved understanding of and respect for forest soils
activity of soil fauna and flora, and plants require the should rank high on the agenda of forest companies, gov-
presence and activity of at least some soil organisms. The ernment agencies, and public interest groups.
mycorrhizal relationship between fungi and plant roots is
particularly important, but the roles of other soil mi-
crobes in the nitrogen cycle and of soil animals and STUDY QUESTIONS
microbes in organic matter decomposition and the devel-
opment of soil structure and porosity are equally signifi- 1, What is soil?
cant. Soil is partly biological in nature, and due 2. Why is soil important?
consideration must be given to this fact. 3. Can plants grow without soil?
The luxuriance of undisturbed, mature plant communi-
ties has often fooled people into misinterpreting the ability 4+. Why is “soil architecture” important?
of soil to sustain plant productivity in the face of intensive 5. What are the key physical characteristics of soil that a
cropping. Experience has shown that many soils are fertile forester needs to understand and respect?
only when left under undisturbed vegetation for consider- 6. Why is soil chemistry important?
able periods. Frequent disturbance and intensive harvesting
7. Are soil animals important? Why?
of biomass may degrade soil fertility because of the physical
loss of soil through erosion, the loss of soil organic matter, 8. What roles do soil microbes play in ecosystem function?
changes in soil physical properties, the loss of nutrients, or 9. Why do we consider soil fertility to be a critically im-
any combination of these factors. Soil is not an inert, phys- portant aspect of ecosystem sustainability? How can
ical phenomenon. It is a dynamic physical—chemical- We assess it?
biological entity. This must be recognized if ecosystems are 10. Is soil always a nonrenewable resource?
to be managed intensively for plant crop production.
11. What factors determine the type of soil that you find
in a forest?
TAKE-HOME MESSAGE 12. Will the type of soil in an area always be the same? If
We cannot manage forest ecosystems sustainably unless not, then what might cause it to change?
we maintain those soil conditions and processes that help 13. Does a little bit of surface soil erosion matter when
to determine forest ecosystem function and ability to re- there is a deep layer of unconsolidated mineral mate-
cover from disturbance. Many aspects of soil must be con- rial on the site?
sidered: physical properties, chemical characteristics, and j '
14. What are the key processes in the development of
soil biology. All of these are important. They vary greatly forest soils?
 Fire
A Pervasive and Powerful
Environmental Factor

12.1 Introduction Alaska, accounting for 75% of the area burned, have
been attributed to lightning (Hardy and Franks, 1963).
Precipitation is a common event and one with which Fire was perhaps the first major force used by hu-
most of us in the humid parts of the world are all too mans to alter their surroundings (Stewart, 1956). In
familiar. We all are aware of light, wind, and tempera- addition to using fire for heating and cooking, fire was
ture as factors of our environment. We experience soil used to drive game, improve grazing, remove vegeta-
by growing plants in it, by having to move it, or simply tion to facilitate travel, clear land for agriculture, and
by walking on it. Fire, however, is an environmental in both aggressive and defensive encounters with
factor with which most people have had little firsthand other humans. Human use of fire extended over most
experience. We have been conditioned by years of an- of the earth, and there is evidence that fire of anthro-
tifire propaganda to consider fire as inherently bad pogenic origin has influenced even the humid tropical
and largely the result of human activities. There has forest for many thousands of years in the form of cut-
been a widespread failure to recognize fire for what it and-burn shifting agriculture.
really is: a natural ecological factor that is as important Fire is generally an environmental factor of low
as wind or precipitation in determining the structure frequency but of considerable potency. Because it pro-
and function of many of the world’s ecosystems. duces such radical environmental changes so rapidly, a
Except in the very wettest, very hottest, and very frequency of once in several centuries is sufficient to
coldest environments (there is nothing to burn in the maintain an ecosystem condition different from what
last two!), fire has played a role virtually everywhere would eventually develop in the absence of fire. This is
on terrestrial earth. Fire was ecologically important seen in some of the wetter eucalypt forests of Aus-
long before it was used by humans, as evidenced by the tralia, which are maintained by a fire frequency as low
presence of charcoal buried deeply in ancient sedi- as once in 350 years. In the prolonged absence of fire,
mentary deposits. Such fires were the result of vol- the eucalypts would be replaced by forests of southern
canic activity, spontaneous combustion, or lightning, beech (Nothofagus sp.) and tree ferns (Dicksonia)
the last being the most important. It has been esti- (Gilbert, 1959; Gill, 1975).
mated that there are about 40,000 thunder clouds ac- Fire frequency varies enormously from one re-
tive over the whole world every day (Schonland, gion to another and from one ecosystem type to an-
1950). For Canada alone, almost 10,000 lightning fires other within a region. Dry ridge tops may be burned
were recorded during a 10-year period (Bennett, every few decades or centuries, whereas moist valley
1960), and lightning is believed to account for 42% of bottoms may burn only once every few millennia.
all Canadian forest fires (Canadian Forestry Service, Table 12-1 shows a range of fire frequencies for se-
1996-1997). Approximately 25% of forest fires in lected vegetation types in North America.

329
330 CHAPTER 12 _— ‘Fire

Tuble 12-1 Fire Frequencies From Selected Vegetation Types in North America
Record Type Fire Frequency
Vegetation Type and Location and Period (years) Reference

Prairie, Missouri = 1 Kucera and Ehrenreich (1962)

Long-leaf pine and blue stem _ Duvall and Whittaker (1964)
ranges, southeast U.S.

Mixed conifers, California Fire scars on trees 7-9 Wagener (1961)

(1581-present)
Ponderosa pine, Arizona Fire scars on trees 5-12 Weaver (1951)
(1700-present)

Mixed forest, Minnesota Lake sediments 60-70 Swain (1973)

(1,000 years)
Boreal conifers, Northwest Territories Fire report records 110 Johnson and Rowe (1975)
(1966-1972)
Mixed forest, New Brunswick Fire report records 230-1000 Wein and Moore (1977)
(1920-1975)

Source: After Wein, 1978. Copyright Plenum Press. Used with permission.

In this chapter, we review briefly the types of fire ter; (2) surface fires—rapidly burning fires that sweep
that occur and their effects on soils, plants, animals, quickly over an area, consuming litter and the above-
and ecosystem function. General reviews of the eco- ground portions of herbs and shrubs; and (3) crown
logical role of fire in Canada and elsewhere can be fires—fires that burn through the crowns of woody veg-
found in Agee (1993), Ahlgren and Ahlgren (1960), etation, frequently leaving most of the stem and the for-
Broysen and Tainton (1984), Cooper (1961), De Bano est floor relatively untouched. Various technical terms
et al. (1998) Gill et al. (1981), Goldammer (1990), are applied to fire, including severity, intensity, and rate
Johnson (1992), Kayll (1968), Kelsall et al. (1977), of spread. Severity refers to the degree of impact of fire
Kozlowski and Ahlgren (1974), Lutz (1956), Payette on organic matter. Intensity refers to the rate of energy
(1992), Slaughter et al. (1971), Walstad et al. (1990), released by the fire and is usually expressed as frontal
Washburn (1973), Wein and MacLean (1981), Wells fire (Canada) or fireline (United States) intensity
et al. (1979), and Whelan (1995). (kW/m). Rate of spread refers to the speed with which
the leading edge of the fire travels downwind.
The three types of fire can occur in any combina-
12.2 ‘Types and Occurrence of Fires tion. Sometimes a crown fire will be accompanied by
both surface and ground fires, resulting in the total
In the mind of the average layperson, a fire is simply a consumption of all organic matter above the mineral
fire. Unfortunately, things are not so simple. The eco- soil. Even roots deep in the mineral soil may be
logical effects of fire vary enormously according to the burned. Alternatively, crown fires driven by strong
time of year; the quantity, condition, and distribution winds may race through the tree crowns consuming
of the fuel; the prevailing climatic conditions; the foliage and twigs only, leaving stems and the forest
severity and intensity of the fire; the slope, aspect, and floor virtually intact. Such fires leave most of the bio-
elevation; the type of vegetation and soil; and so on. mass and almost all of the minor vegetation intact.
Generalizations about fire and its effects on ecosystems They have high intensity but low severity. Some trees
are fraught with the danger of inaccurate representa- are killed by having their crowns burned, but others,
tion of the real situation. Discussion of fire should al- such as redwoods (Sequoia sp.), are able to regenerate
ways specify the type of fire and type of environment. branches and foliage.
Fires can be divided into three major types: (1) Surface fires generally burn off just the litter layer
ground fires—largely flameless fires that burn slowly and the aboveground parts of herbs and shrubs. These
through thick surface accumulations of organic mat- are often able to resprout from below-ground pere-
 Effects ofFire on Soil 33.1

nating organs, depending on the depth of heat pene- tems in which fire is a frequent and natural component,
tration into the soil and the depth of the lowest pere- the effects on soil may be benign or even desirable.
nating organ of the plant (Flinn and Wein, 1977). ‘The degree to which soil properties are altered by
‘Trees may or may not be killed, depending on their fire depends on the fire intensity and the fire severity
bark thickness. Ground fires tend to be more destruc- (amount of organic matter that is consumed). These in
tive, because they kill and consume all the roots in the turn are influenced by the amount of organic fuel
forest floor, which generally prevents resprouting available and its distribution and moisture content and
from underground organs. Ground fires can kill large the prevailing weather conditions. Because these pa-
trees by this means while the stems and crowns remain rameters are so variable, it is difficult to make reliable
untouched. Consumption of forest floor may elimi- generalizations about the effects of fire. A low-severity
nate most of the dormant seeds on the site, slowing surface fire may have little effect on soils, and a ra-
revegetation of the area. However, viable seeds are pidly moving crown fire might also have little effect if
sometimes found buried in the mineral soil, where the vegetation resprouts rapidly. Conversely, fire may
they may escape destruction by ground fires and con- have a great effect if all the vegetation is killed, be-
tribute to revegetation (Moore and Wein, 1977). The cause the resulting losses of shade, root strength, and
loss of forest floor exacerbates the ecological conse- transpiration can lead to significant changes in soil
quences of the loss of all living vegetation, but the sub- temperature, stability, and moisture. Ground fires
sequent fall of dead but unburned shrub and tree generally have a marked effect by removing the sur-
materials contributes to a rebuilding of this layer. face accumulation of organic matter, whereas high-
Fire occurs mainly in the dry season, which occurs intensity ground-surface—crown fires during the heat
at different times of year in different parts of the of the summer can result in a massive alteration or the
world. It can occur at the beginning, middle, or end total loss of soil.
of the dry season, with markedly different ecological Because fire can create major changes to soil char-
effects. A dry spring may be associated with high fire acteristics, because soil is so important to the forester,
hazard in areas without a snowpack, but in many re- and because fire, both wild and managed, is such a
gions, middle to late summer has the highest risk of common feature of the forest environment, it is very
fire. A spring surface fire will have a very different ef- important that resource managers have a clear under-
fect than a late summer fire, which is often a com- standing ofits effects. These can be divided into phys-
bined crown-surface—-ground fire. After a spring fire, ical, chemical, and biological. For a detailed treatment
plants will resprout and there may be relatively little of the effects of fire on soil, the reader is referred to
change in the vegetation for the following year. In Bell et al. (1974a,b), Cramer (1974b), De Bano et al.
contrast, a late summer or early winter fire will leave (1998), Kozlowski and Ahlgren (1974), and Wells et al.
the ground bare of vegetation throughout the follow- COTO URS
AD A107 191970):
ing winter.
Information on the causes of fire is of some interest
because it permits a comparison of the importance of
Physical Changes
humans as a cause of ignition with that of other causes. Organic Matter. Loss of organic matter is one of
However, such data are often of little value in deter- the most important effects of fire on soils. Fire speeds
mining the contribution of human-caused fires to the up the normal process of mineralization of organic
overall ecological effect of fire because many of these matter, achieving in a few minutes what would have
fires are extinguished promptly. They generally do not taken microbes several years for materials such as dead
burn large areas, whereas many lightning fires do. foliage or fine roots, and decades or even centuries for
large fuels such as stumps and logs. Generally, fire is
restricted to the surface of the organic accumulation
12.3 Effects of Fire on Soil because of the need for oxygen and because deeper
layers are often too wet to burn. However, ground
Soil properties are strongly influenced by living vegeta- fires can smolder slowly into deep layers of even moist
tion and accumulated dead organic matter, both of which organic matter, the heat from the fire drying the mate-
are removed to a variable degree by fire. Consequently, rial ahead of it. Organic matter incorporated into the
fire has the potential to induce major changes in soils. mineral soil is normally unaffected by fire, but in ex-
Such changes are sometimes deleterious, but in ecosys- tremely severe fires, the penetration of heat down into
332 CHAPTER 12 Fire

TS
eee
e
C
n

(a) (b)

Figure 12-1 Loss of forest floor after slashburning. (A) Rocky slope in southwestern British
Columbia, which had a thick forest floor beneath a mature stand of western hemlock, Pacific silver
fir, and western redcedar before clearcut harvesting and a hot fall slashburn. (B) Close-up of a stump
on a bare rock, which had previously been covered with at least 15 cm of organic matter (based on
the space between many of the large roots and the rock surface).

the mineral soil can destroy colloidal organic matter. whereas a fire that removes all the forest floor and ex-
Fire can penetrate deeply into the soil by burning poses the mineral soil to the impact of raindrops can
along dead roots. lead to a loss of structure in the surface layers. This
The loss of surface organic accumulations depends can reduce infiltration rates and increase surface
on fire duration, intensity, and fuel moisture. A fire in runoff, which can lead to erosion. In a severe ground
the spring when the forest floor is moist may burn off fire, colloidal organic matter in the surface mineral
the fresh litter only, leaving the F and H layers intact, layers may be destroyed, accompanied by a loss of
whereas a midsummer fire might remove the L, the F,
and much of the H. Severe fall fires can remove up to
50 cm of organic matter over rock on dry sites (Figure 4.0
12-1). Armson (1977) reported that a light early spring
(April) surface fire in mixed white pine and sugar maple Original biomass of F + H
forest in southern Ontario resulted in no significant re-
duction in surface organic matter. In contrast, natural Se o

fires in the boreal region of Ontario reduced the LFH

layer by approximately 50%. Viro (1974) reported a
25% reduction in the F and H layers from 33 to 25 t i) =
ha’! after burns in Finland, whereas organic matter in
the mineral soil fell by 17% in the upper 10 cm and 7
to 10% lower in the profile. Consumption of F and H
kg
F+m
consumed,
H =)
layers in pine stands in eastern Canada was found to be
related to the percentage of moisture content before
the fire (Figure 12-2). The effects of fire on total soil
organic matter will depend on how much organic mat-
ter there is in the mineral soil. Where there is a lot, the 0 50 100 150
loss of surface organic accumulations may not consti- % moisture content

tute as serious a loss as first seems. Figure 12-2 Effect of F and H layer moisture content on
F and H layer consumption by fire in pine stands in On-
Structure and Porosity. Fires that remove only the tario. (After van Wagner, 1972. Reprinted by permission of the Na-
L horizon will have little effect on soil structure, tional Research Council of Canada and the author.)
 Effects ofFire on Soil 333

structure and a reduction in porosity (Fuller et al., mately 500°C or greater do not induce hydrophobicity
1955). Opposing these changes is the increase in pH because the hydrophobic substances are destroyed.
and divalent cations that accompanies fire, which can ‘Temperatures of 425 to 500°C begin to destroy the hy-
lead to an increase in flocculation and an improvement drophobic property after 10 minutes, whereas 260 to
in the structure of surface layers of finer-textured soils. 315°C for 10 to 15 minutes will produce a persistent
and highly water-repellent layer (Debano et al., 1967).
Moisture. Fire reduces transpiration and intercep- The best known examples of hydrophobicity occur
tion losses in proportion to the reduction in foliage. under chaparral brushlands in southern California.
Where all vegetation is killed, the soil may be corre- Wildfires in moister and cooler northern forests are
spondingly wetter, but this depends on the effects of less likely to cause a significant increase in hydropho-
the fire on soil organic matter and soil structure. It is bicity. This is because mor forest floors under these
true only where the fire leaves the forest floor more or forests are normally hydrophobic when dry, a condi-
less intact. In coarse-textured soils, much of the soil tion that is attributed to the abundance of fungal
moisture storage capacity is provided by organic mat- mycelium and spores in these F and H horizons (De-
ter, and if this is burned off, then the soils may become bano and Rice, 1973). Intense fires can produce a
much drier. If there is a reduction in infiltration be- short-term increase in mineral soil hydrophobicity
cause of the fire, less water will enter the soil and less (Table 12-2), but some fires may actually reduce it by
will be available for storage; again, the soil becomes eliminating the fungus (DeByle. 1973). Water repel-
drier. Fire will also increase soil evaporation losses if lence has been reported in burned Jeffrey pine stands
all the forest floor is removed: losses that are restricted in Nevada (Hussain et al., 1969) and lodgepole pine
in the absence of fire by the hydraulic discontinuity at stands in Oregon (Table 12-3).
the mineral/organic interface. In a hot, dry climate Because of the natural hydrophobicity of dry mor
with a medium-textured soil, such evaporation losses forest floors, exceptions can be found to the general-
may be considerable. . ization that organic matter improves the moisture sta-
Fire can reduce infiltration in several ways: by loss tus of coarse soils. For example, organic matter
of mineral soil structure, by the plugging of macro- contributes little to soil aggregation in a sandy soil,
pores with ash, by the formation of a charred crust, and by coating the mineral particles, it may make the
and by the development of a water-repellent layer. In soil water-repellent and therefore drier. The LFH lay-
an extremely severe and adequately aerated fire, all ers that accumulate on such soils in dry climates may
carbon compounds are oxidized to CO or CO). How- be very hydrophobic during the growing season, pre-
ever, in many fires, there is insufficient heat and/or venting the infiltration of summer rain down to the
oxygen for combustion and many organic compounds rooting zone. Removal or diminution of this layer by
are simply vaporized. Much of the vapor leaves in the fire can improve the moisture status of the site. In
smoke column, but some of it is driven downward into contrast, the large surface area of the particles in a clay
the unburned soil, where it condenses on the cooler, soil diminishes water repellence, whereas the organic
unburned materials. Subsequent penetration of heat matter promotes aggregation, structure, and drainage.
drives the most volatile constituents even deeper into This improves the permeability and moisture status of
the soil, broadening the band of soil affected by the clay soils. Removal of surface organic matter from clay
condensate (Savage, 1974). This leads to the develop- soils can have adverse effects on soil moisture status.
ment of a well-defined layer of material, which, be- It is generally thought that hydrophobicity lowers
cause of the coating of volatile substances, displays the soil moisture levels by impeding infiltration, although
phenomenon of hydrophobicity or water repellence. this effect may be offset by a reduction in evaporation
This hydrophobic layer reduces the rate of infiltration loss. This occurs because the hydrophobic layer im-
into the remaining forest floor and/or underlying pedes nonsaturated water flow upward through the
mineral soil. surface layers of the soil, which acts to conserve soil
The occurrence of hydrophobicity is highly vari- moisture (Figure 12-3).
able. It varies according to the type of fuel, the type of
fire, and the type of soil. Sand, with its relatively small Temperature. Low soil temperature is a major lim-
surface area, is much more affected than silt or clay iting factor in many poleward and high-elevation for-
soils, which have a very large surface area (DeByle, est ecosystems. It is primarily the result of long cold
1973). Fires with surface temperatures of approxi- winters and thick forest floors that prevent summer
334 CHAPTER 12 Fire

Table 12-2 Percentage of Soil Samples Found to be Hydrophobic on Various Aspects in Clearcut and Slashburned
Areas’ in Montana and the Duration of the Condition
% of Samples That Were
Hydrophobic, by Aspect
Soil Depth,
Time of Sample cm North East South? West
e
Dire ger e
ee a e
Preburn 0-5 3.1 0 10.5 (h4

5-10 0 0 0 0

Immediately postburn 0-5 14.3 21.4 56.2 27.3

5-10 0 Ua 25.0 9.1

1 year postburn 0-5 0 6.2 16.7 5.0

5-10 0 0 5.6 0
2 years postburn° 0-5 0 0 25.0 0
5-10 0 0 0 0

Data from DeByle, 1973.

“The area had supported old-growth stands of western larch, Douglas-fir, and Engelmann spruce.
>Hydrophobicity was greater on the southerly aspect both before and after burning.
°On most areas, pretreatment levels were approached after 2 years.

warmth from reaching the underlying mineral soil. In amount of CO, given off from the forest floor materi-
very cold climates, this leads to the lower layers of the als incubated at various temperatures. Viro (1974)
soil’s being permanently frozen. found that raising the temperature at which Finnish
Low soil temperature slows decomposition and forest floor material was incubated from 6°C to
limits rooting depth, both of which reduce plant pro- 12.5°C doubled CO, evolution, and 20°C quadrupled
duction by limiting the supply of nutrients. The effect it. Permafrost close to the surface also leads to poor
of temperature on decomposition is indicated by the soil drainage, which further limits forest growth.
Fire affects soil temperatures in both the long and
Table 12—3 Effect of Fire on Infiltration Rates (Water the short term. The long-term effects generally in-
Drop Penetration Time) in Lodgepole Pine Forests volve an increase in soil temperatures. By darkening
in Oregon the soil surface, fire promotes absorption of solar en-
Soil Depth, Water Drop
ergy, and by reducing the depth of the surface organic
Fire Treatment cm Penetration Time
accumulation, transfer of heat to the mineral soil is
promoted. ‘Table 12-4 shows soil temperatures in for-
Unburned 0-5.0 1346? est, clearcut, and burned areas. Removal of tree shade
5.0-91.4 0 increased soil temperature, but fire increased it by as
Light burn 0-2.5 1 much again. The effect can be seen down to a depth of
20 cm in the mineral soil.
2.5-7.6 b
In addition to the general increase in soil tempera-
7.6-15.2 170
ture, fire sometimes results in a reduction in surface
20.3-30.5 1 temperature. In hot climates with open forests, very
Severe burn 0-6.4 0 high temperatures can occur at the surface of the forest
6.4-15.2 floor. This is a consequence of the high solar energy
input to and the low thermal conductivity and thermal
Source: Debano and Rice, 1973. Copyright Society of American capacity of dry litter. Removal of the LFH exposes
Foresters. Used with permission.
mineral soil, which has both a high thermal conductiv-
*Note the hydrophobicity of the unburned surface layer and the pro-
gressive increase in depth of the hydrophobic layer in the mineral soil
ity and a high thermal capacity. Because of these ther-
with increasing fire severity. mal properties, high mineral soil surface temperatures
>Water drop evaporated without penetration. are uncommon. However, if the fire leaves a layer of
 Effects ofFire on Soil 33)

300

Wettable soil
SS
Vv |
s |
200 5 \ |
e \ |
a
— ———

:
©
| Initial
|
ie) moisture vA|
Hydrophobic & content |
100 soil == |
(mm)
evaporated
Water S
A |

20 40 60 80
Time (days) % moisture content

(a) (b)

Figure 12-3 Effect of hydrophobicity on soil moisture conditions. (A) Cumulative evaporation
from soil columns containing wettable and nonwettable soil. (B) Soil moisture distribution in the
columns after 109 days of evaporation. Hydrophobicity reduced evaporation, which resulted in the
soil’s retaining a higher moisture content. (After Debano et al., 1967. Used by permission of the USDA
Forest Service, and the authors.)

black charcoal, then it will have the opposite effect; it slash in Australia, whereas at 5 to 10 cm depth in the
will result in increased surface temperature because of soil, the temperature was only 100°C (Humphreys and
the lower water-holding capacity and greater ab- Lambert, 1965). In a jack pine prescribed burn in
sorbancy (lower albedo) of the burned surface. Minnesota, surface temperatures exceeded 800°C for
Energy released during a burn creates short-term 1 minute, 500°C for 9 minutes, and 300°C for 17 min-
effects on soil temperature. Surface soil temperatures utes. The organic—mineral interface at a depth of 5 to
as high as 1000°C have been recorded (references in 8 cm from the surface reached 300°C for 14 minutes
Ahlgren, 1974), but because of the remarkable insulat- and was above 50°C for 72 minutes (Ahlgren, 1970).
ing properties of forest floors, heat penetration is gen- An experiment in Australia examined the soil heat-
erally limited. For example, temperatures of 350 to ing in fires ranging from a surface fire of 45 minutes’
900°C were recorded at the soil surface under burning duration to an 8-hour fire that consumed all the trees

Table 12-4 Effect of Burning and Shading on Summer Soil Temperatures

Temperature, °C

Sampling Depth Uncut Forest* Clearcut Area Slashburned Clearcut

Air, 5 cm above forest floor 20.6 24.8 23.6

Forest floor surface 18.0 24.4 31.3

Mineral soil
Surface 9.1 12.8 16.0

10cm 8.7 10.6 WES)

20 cm 8.3 10.2 220

Source: Viro, 1974. Copyright Academic Press. Used with permission.

@The data are for 1400 hours in a 100-year-old spruce stand in south Finland.
336 CHAPTER12_ Fire

and shrubs on the test plot. The short-duration fire elements such as calcium, potassium, and magnesium,
resulted in a maximum temperature of 50°C at a 2.5- which form cations. The ash left by the fire consists
cm depth. The 8-hour fire raised the temperature at a largely of soluble oxides of these alkali earths. These
7.5-cm depth to 223°C (Beadle, 1940). Fire of high oxides are rapidly changed to carbonates, which have
severity generally occurs on only a small proportion of an alkaline reaction and tend to neutralize acidity in the
an area, so heating of the mineral soil beneath burning soil. Consequently, soil pH generally increases after a —
forest floors is not normally very intense. In a study of fire. The extent and duration of the increase will de-
44 fires in Florida, it was found that temperatures pend on the intensity of the fire, the amount of organic
rarely exceeded 52°C for more than 15 minutes at matter consumed, and the buffering capacity of the
depths 3 to 6 mm below the surface (Heyward, 1938). soil. Viro (1974) reported an increase of 2 to 3 pH units
The major temperature-dependent stages of igni- in the forest floor after burning in Finland, returning to
tion are (Ralston and Hatchell, 1971): (1) 100 to original levels after 50 years (Figure 12-4). The under-
200°C—nondestructive distillation of volatile organic lying mineral soil was less affected, but an increase of tas
viel
ial

compounds, (2) 200 to 300°C—destructive distillation 0.4 pH unit persisted for 20 years; even after 50 years,
of up to 85% of organic substances, and (3) >300°C— there was still a difference of 0.2 pH unit.
ignition of carbonaceous residues. A comparison of Smaller changes in pH have been reported in areas
these numbers with those in the preceding paragraph of the eastern and southeastern United States, where
explains why much forest floor remains unburned burns of various frequencies produced pH changes of
even under very hot surtace fires. only approximately 0.5 pH unit. The degree of change
is related to the cation exchange capacity of the soil, as
is the speed with which the pH returns to its original
Chemical Changes levels. The growth of herbs and shrubs that frequently
pH. When fire oxidizes organic compounds, ele- follows a fire helps to slow the return to the original
ments that form anions (e.g., nitrogen, phosphorus, level of acidity by reducing the leaching of cations and
and chloride) are lost in much greater quantities than by active circulation of nutrients. The pH change that

——* data from Finland

data from Ontario

st

pH
floor
forest
the
of
== Preburn pH

Preburn pH

0 10 20 30 40 50
Years since burning

Figure 12-4 Effect of fire on the pH of the LFH layer and rate of recovery of the original
pH. (From data in Armson, 1977, and Viro, 1974. Used by permission of Academic Press, Inc., and The
Finnish Forest Research Institute: the University of Toronto Press; and the authors.)
 Effects of Fire on Soil 337

accompanies fire may be one of the main benefits of fire nitrogen availability may actually be increased be-
burning in tropical shifting agriculture. Highly cause of a pH- and temperature-induced increase in
leached tropical soils derived from poor parent mate- mineralization of the remaining F and H materials.
rials often have such a low pH (in the 3 to 4 range) that Where the fire burns off all the forest floor, nitrogen
food crops cannot be grown satisfactorily unless the availability will be greatly decreased unless this results
pH is raised by ash from burning the vegetation. in postfire invasion of the site by free-living or symbi-
otic nitrogen fixers. Fire-induced changes in site bio-
Site Nutrient Capital and Nutrient Availability. geochemistry are discussed further in the section
Fire induces a variety of chemical changes in the soil. Forest Biogeochemistry.
As organic matter is burned, carbon is released as
gaseous oxides and nitrogen is lost increasingly as
temperatures rise above 300°C. Sulfur and phospho- Biological Changes
rus are also subject to gasification losses, and some Although soil temperature may not be high enough
potasstum may be lost at temperatures greater than during a burn for ignition to occur, it may be high
500°C. Boron is also subject to loss during a fire enough to affect soil animals. The meso- and micro-
(Moore and Norris, 1974). Many other nutrients are fauna of the L and F horizons have limited mobility
removed from the site in the form of fly ash that is car- and are normally killed during a fire, but survivors in
ried up with the smoke. In very intense fires with high the lower H horizon or in islands of unburned LFH
fire-induced winds and a strong convection column, provide a source for recolonization, and numbers re-
most of the ash and the nutrients contained therein cover within a few years. The effects are therefore re-
may be removed from the site. In less intense fires, lated to fire frequency, as has been demonstrated in
most of the elemental content of the burned material studies of prescribed burning of pine stands in South
remains on site. Because of this, the total quantity of Carolina. Annual burning significantly reduced the
chemicals such as calcium and magnesium in the for- abundance of mites and springtails, whereas burning
est floor may be increased significantly by a low- to every 5 years produced no long-term change in num-
moderate-intensity fire through the addition of ash bers (Metz and Farrier, 1971). Where severe fires
from incinerated minor vegetation and tree crowns. eliminate all surface organic accumulations, virtually
This increase generally does not persist indefinitely. all meso- and microfauna can be eliminated for many
Some of the chemicals in the ash are leached down years, although such fires are generally limited to the
into the mineral soil, and some are taken up by the summer, when surface dryness and heat would have al-
vegetation. ready driven these animals down into the moister,
A major effect of fire is to convert unavailable min- cooler, mineral soil, where they would survive the fire.
eral nutrients in undecomposed organic matter to a The literature on fire reveals that effects on mi-
soluble form that is available to plants. This means croflora are highly variable because of the great vari-
that even when fire causes a substantial decrease in the ability in the effects of fire on soil temperature,
total soil mineral nutrient capital as a result of gasifica- moisture, and pH (Ahlgren, 1974). Because of the mo-
tion, fly ash, and/or leaching losses, it may actually im- bility of their reproductive propagules, bacteria and
prove the availability of nutrients. The higher nutrient fungi are usually able to reinvade and recolonize a
availability is seen first in the forest floor (with the ex- burned area promptly; therefore, fire-induced micro-
ception of very soluble elements such as potassium), faunal changes often tend to be less persistent than
but forest floor levels decline with time because of other fire-related environmental changes. Typically,
leaching to lower horizons, uptake by plants, and fires produce an initial reduction in microflora popula-
microbial conversion back to unavailable forms. In- tions, followed in many cases by an increase, which
creases in levels of available nutrients lower in the pro- frequently occurs after the first postfire rainfall
file occur after a delay of several years. (Ahlgren and Ahlgren, 1965). The abundance of these
Of all the macronutrients, nitrogen is the most postburn populations may exceed preburn levels be-
susceptible to loss during a fire, and postfire net pri- cause of increased pH, reduced microbial competition,
mary production may be significantly reduced because and improved availability of nutrients. Sometimes fire
of a reduction in soil nitrogen. However, if the fire can sterilize an area, permitting the replacement of the
merely burns off the L and upper F layers, then post- original microfauna by a different set of species. Loss
338 CHAPTER 12 ‘Fire

of mycorrhizal fungi from a burned area as a result of

such microfaunal replacement may be related to the
regeneration failures and/or poor seedling growth that
sometimes occur after a burn. The persistence of mi-
crofaunal changes will depend on the persistence of
changes in critical soil conditions.
It has sometimes been suggested that the initially
improved nitrogen nutrition of plants after a moderate
burn may reflect atmospheric nitrogen fixation by
free-living bacteria. Grass growing in burned areas is
frequently reported to be more lush, to have a darker
green color, and to have a higher protein content than
grass in unburned areas, suggesting higher availability
of nitrogen. Both wildlife and domestic livestock
will seek out and browse the vegetation of burned Figure 12-5 Douglas-fir stump in south-central British
areas. Direct evidence of such fire-induced fixation is Columbia, showing the great thickness of bark that con-
lacking, but it may occur as the result of the increased fers on this species a high resistance to surface fires.
pH (Armson, 1977). Burning has been reported
to increase the abundance of the nitrogen-fixing
Azotobacter and Clostridium species. Fire can increase of different species has a similar thermal diffusivity ir-
nitrogen mineralization for up to 12 years (references respective of a wide range of moisture content and
in Ahlgren, 1974), although the opposite effect may structure: that bark thickness is the most important
occur in dry soils (e.g., Meiklejohn, 1953). variable. Others claim to have demonstrated wide
species differences in heat penetration of bark (see
Gill, 1975). The protection conferred by bark will also
12.4 Effects of Fire on Plants depend on its flammability and how quickly the
burned bark is replaced. Where replacement is slow, a
Because fire has been such a characteristic feature of severely burned bark may be susceptible to a subse-
most ecosystems, a wide variety of plant adaptations quent fire of much lower intensity. Species that shed
has evolved. Fire may affect any stage in a plant’s de- bark rapidly and accumulate bark and other above-
velopment (vegetative, flowering, fruiting, or dor- ground litter around their base will be more suscepti-
mant), and there is a corresponding variety of adaptive ble to fire damage than will those without such an
traits (Brown and Smith, 2000; Whelan, 1995). accumulation.
Reduced flammability of tissues will reduce the
spread and intensity of a fire, reducing the risk of fire
Adaptations to Fire in the Vegetative Stage damage. Species with high foliar moisture content and
Plants can be classified into several categories on the low resin or oil content will generally be much more
basis of vegetative characteristics that affect their reac- fire resistant than the resin- and oil-rich conifers.
tion to fire. Some plants achieve resistance by protect- Mixed angiosperm—coniferous stands will be more re-
ing themselves from damage, whereas others are able sistant than pure coniferous stands partly because of
to tolerate the damage. the break in continuity of flammable crown fuels.
Fire-resistant bark is one of the more common Protected buds confer on plants the ability to con-
adaptations to surface and ground fires. Trees such tinue to grow and to recover from the loss to branches,
as western larch (Larix occidentalis), Douglas-fir foliage, or even the entire aerial shoot. Some conifers,
(Pseudotsuga menziesii), and ponderosa pine (Pinus pon- such as pitch pine (Pinus rigida), redwoods (Sequoia
derosa) develop a very thick layer of dead bark as they sp.), and ponderosa pine, are able to replace the foliage
mature, which enables them to survive severe ground and branches lost in a crown fire by means of adventi-
and surface fires (Figure 12-5). Some angiosperms, tious or latent auxiliary buds. Similarly, leaf scorch of
such as some Eucalyptus sp., also have fire-resistant eucalypts stimulates the development of epicormic
bark. There is disagreement over the nature of a bark’s shoots, although the ability to recover from crown
protective value. Some researchers claim that the bark damage in this manner varies considerably between
 Effects ofFire on Plants — 339

different species. Younger trees of any species have a Fire has been shown to cause a stimulation of flow-
greater ability in this respect than do older trees. ering in some plants. For example, the reproductive
Many species are able to replace entire aerial shoots success of some members of the genus Xanthorrhoea, a
by developing new shoots from underground buds that fire-adapted Australian shrub, is greatly enhanced by
survive the fire. Many angiosperm trees, shrubs, herbs, fire, although some interfire flowering does occur
and even some conifers have this ability. When the aer- (Specht et al., 1958). This may be a temperature re-
ial shoot of aspen (Populus tremuloides) is killed by fire, sponse or may merely reflect a reduction in competi-
sucker shoots develop from adventitious buds on later- tion for light, moisture, and nutrients. It could also be
al roots growing in the surface layers of the soil. When an effect of smoke. It is known that ethylene, a compo-
the crown and stem of a coast redwood (Sequoia semper- nent of wood smoke, stimulates flowering, a fact that
virens) are consumed by fire, sprouts form around the was once put to good use by pineapple farmers in
base of the stem, which results in the clumping of Puerto Rico, who flooded their fields with wood
stems in the subsequent stand around the old stumps. smoke to initiate flowering. Ethylene gas is sometimes
Plants that have /ignotubers are able to produce new used by itself to stimulate flowering (Komarek, 1971).
shoots more rapidly than those that lack this adapta- Seed dispersal is influenced by fire in some species.
tion because of the food reserves provided by the Lodgepole pine, jack pine (Pinus banksiana), and black
tuber. A lignotuber is a conspicuous swelling of the spruce (Picea marina) are three North American
main axis, mostly or entirely below the ground sur- conifers that have serotinous cones, the scales of
face. The tuber forms at the root/stem junction by the which are prevented from opening by a resinous bond
process of bud multiplication, and it is consequently a (Figure 12-6). The resin of jack pine cones melts at
rich source of sprouts if the aerial shoot is killed. This temperatures greater than 60°C, and the seeds can re-
adaptation enables some eucalypts to reoccupy and main viable within cones exposed to a temperature of
dominate a site very rapidly after a fire (Mount, 1969). 370°C for 60 seconds (Beaufait, 1960). Certain species
It is most obvious at the environmental extremes of of several Australian genera (e.g., Banksia, Casuarina,
the genus, and the adaptation is absent in the most Calothamnus, Hakea, and Xylomelum [references in
productive forest sites of southern Australia. It is Gill, 1975]) also depend on heating for seed dispersal.
therefore thought that lignotubers probably represent For example, the shrub genus Banksia produces flow-
a generalized adaptive trait for recovery from stress- ers in dense, spike-like inflorescences. The resulting
induced damage rather than a specifically fire-induced hard woody fruits (or follicles because they are devel-
adaptation (Gill, 1975). oped from a single carpel) open freely without fire in
Plants that have rhizomes (horizontal, underground some species, whereas in others, only approximately
stems) are also able to sprout rapidly after fire and are 1% of the fruits open unless the plant is exposed to
highly resistant to damage by fire. The success of the
rhizomatous fireweed (Epilobium angustifolium) and
bracken fern (Pteridium aquilinum) in fire environ-
ments illustrates the success of this adaptation.

Adaptations to Fire in
the Reproductive Phase
Evolution has selected a variety of reproductive traits
that increase the fitness of plants exposed to frequent
fires.
Precocious flowering reduces the time from ger-
mination to seed production in perennial plants, and
species such as lodgepole pine (Pinus contorta) can pro-
duce seed in as little as 5 years (Fowells, 1965), which
is normally before the next fire occurs. Frequent fire Figure 12-6 Serotinous cones of lodgepole pine. The cone
eliminates species that flower infrequently or only scales will open only after being heated to a critical tempera-
after a long juvenile period, unless there are other ture, such as in a fire, or when lying on a forest floor exposed to
adaptations. full sunlight (e.g., after clearcutting).
340 CHAPTER 12. Fire

fire. In all these genera, seed release involves the the vegetative stage to fire, coupled with seed adapta-
breaking of a zone of abscission cells. Desiccation in- tions. It is intriguing that rather than being fire-
creases the tension across this abscission zone, and resistant, many plant species in fire environments ac-
in many species, this is sufficient to open the fruit; tually burn more readily and destructively than plants
in others, the additional heat treatment provided by of environments in which fire is uncommon. It has
fire is necessary. In genera such as Eucalyptus and been suggested (Mutch, 1970) that this represents a
Casuarina, the protective woody fruits open only upon successful adaptation that increases the fitness of the
desiccation. This occurs after death by fire but does species. For example, jack pine (a species with preco-
not actually require heat treatment (Gardner, 1957). cious flowering and protected seeds) forms stands that
under natural conditions burn with monotonous regu-
Effects of Fire on Germination larity. Fire eliminates the light-demanding pine, but it
also kills the competing tree species that would replace
The seeds of many species lie dormant in the soil until it in the absence of fire. Continued occupancy of the
the area is burned. Shrubs of dry areas, such as Acacia,
area by jack pine is thus ensured because the preco-
Arctostaphylos, Ceanothus, and Rhus produce large ciously produced seeds protected in their serotinous
quantities of hard-coated seeds that germinate only
cones promptly regenerate the burned area. A similar
when they have been heated. This may be achieved by
example is provided by fireweed and bracken fern,
exposure to full sun as well as to fire, but fire is gener-
which produce highly inflammable litter that accumu-
ally the agent that removes the shade and is therefore
lates because of slow decomposition. This increases
either directly or indirectly responsible for the heating.
the chance of fire, which benefits these rhizomatous
Many plants germinate better on mineral soil than
species while eliminating the woody species that
on a loose surface organic accumulation. This can be
would shade them out.
explained in terms of better moisture and temperature
conditions on the mineral soils and/or the removal of
chemicals that may serve to inhibit germination (al- Other Adaptations
lelopathy; see Chapter 14). A feature of chamise com-
Another adaptation to fire that is found in some fire-
munities (Adenostoma fasciculatum, a shrub found in
adapted species is very rapid early growth. This un-
Californian “hard chaparral”) is the virtual absence of
doubtedly benefits the plants in other ways as well, but
herbs until after a fire that removes the aerial shoots of
rapid elevation of the terminal bud and foliage out of
the shrub. Experiments showed that although some of
reach of surface fires and rapid development of thick
the increase in herbaceous growth could be attributed
bark are undoubtedly of advantage in a fire-dominated
to the fertilizer effect of the ash and to some reduction
ecosystem. An interesting and widely known example
in small-mammal seed predation, fire seemed to be
of this type of adaptation is provided by longleaf pine
the most important factor (McPherson and Muller,
(Pinus palustris). Vhe terminal bud remains close to the
1969; Muller et al., 1968). Other chaparral species are
ground for approximately 5 years after germination,
known to release a variety of terpenes, phenols, alka-
while the seedling develops a large root system. Dur-
loids, and other organic chemicals that inhibit the ger-
mination and growth of competing species. These
ing this period, the bud is protected from frequent
allelochemicals accumulate in the surface soil, from
surface fires that characterize the region where this
which they can be removed by fire, permitting the in- species grows by long fire-resistant needles that form a
vasion of previously excluded species. The rapid and dense circle around the bud (the so-called grass stage
of the tree). Many 1- and 2-year-old seedlings are
prolific invasion of burned-over forest in northern
killed by fire, but for those that survive the first 2
areas by fireweed and its superior growth in such areas
years, the fires merely scorch the ends of the needles.
may well involve a similar release from inhibition, al-
The grass stage ends when the well-rooted seedling
though the improved nutrient availability in burned
begins a period of rapid height growth that carries the
areas is probably the major factor.
fire-sensitive bud well above the reach of surface fires.
Longleaf pine not only is extremely resistant to fire
Evolution of Increased Inflammability damage, it also depends on fire. This species is suscep-
Some fire-adapted species seem to have evolved along tible to a foliage fungus (brown spot disease, Septaria
a totally different line. Rather than evolving resistance acicola), which is eliminated or reduced by fire, and
to fire, they have evolved an increased susceptibility of long-leaf pine stands are deliberately burned in the
 Effects ofFire on Animals 341

grass stage to control the disease. The species is also popular view. Much of the unreferenced discussion
unable to compete with the understory hardwood below is based on the excellent review of the extensive
species that would invade the site in the absence of fire. literature on the effects of fire on birds and mammals
Because of the variable degree of fire adaptation in given by Bendell (1974). A recent review of the effects
different plant species, fire plays a major role in deter- of fire on fauna is given in Smith (2000).
mining the structure and composition of many of the Fire affects animals in two major ways: the direct
world’s plant communities. It has been called the dom- effects during the burn and the indirect effects that re-
inant fact of forest history by Spurr and Barnes (1973), sult from changes in the animal’s environment.
who reviewed the role of fire in determining the com- Evidence for the direct effects of fire on animals is
position of forests in the United States. Fire has been mixed. These effects depend on the mobility of the
implicated in the maintenance of grassland around the animals, the completeness and size of the burn, the
world (although some grasslands are climatically de- rate of spread, and the intensity of the fire. There are
termined), and extensive heath lands in Europe were reports of many mammals, large and small, swimming
created by deforestation followed by frequent fires. across large rivers to escape a 1915 fire that covered
Much of the vast area of pine, Douglas-fir, and euca- 1,600,000 km? in western Siberia. The smoke from
lypts in the world are of fire origin, and extensive areas this fire is reported to have covered an area the size of
of oak in Europe and the eastern United States owe Europe. In contrast, neither birds nor animals under-
their existence to a history of forest fires (Brown, took more than local movements during fires that cov-
1960). Without it, the species would be partially or ered approximately 35,000 ha on the Kenai peninsula
completely replaced by other species, and fire is im- in Alaska in 1969. Small mammals were affected more
portant in controlling populations of parasites and dis- than large mammals and birds. The literature suggests
eases of trees. For example, fire has historically been that most animals are able to avoid adverse effects of
important in regulating the infection of lodgepole fire by moving into burrows (e.g., small mammals), to
pine (Pinus contorta) forests in the interior of British islands of vegetation that are not affected by the fire,
Columbia by the parasitic vascular plant dwarf mistle- or into lakes or rivers. Obviously, fire will have adverse
toe (Arceuthobium sp.). Mistletoe infections of western effects on animals with low mobility, such as the
hemlock (Tsuga heterophylla) in some coastal forests young, the egg and fledgling stage of birds, the old,
have similarly been controlled by fire. It should be the maimed, and the sick. These are animals that are
noted, however, that only intense fires that kill all the normally subject to above-average mortality rates, and
infected trees have this effect. Lower severity fires that in many cases fire-induced mortality may only hasten
leave scattered infected trees alive will not prevent the the inevitable. In very large and intense fires, smoke
next generation of trees from becoming infected and/or lack of oxygen may cause more harm to ani-
(Alexander and Hawksworth, 1976; Unger, 1992). In mals than the direct effects of heat.
many cases, a relatively low frequency of fire is all that The indirect effects of fire on animals are of much
is necessary to control the mistletoe. greater import than the direct effects. Most animals
are highly habitat specific. For reason of food, cover,
microclimate, and competition from other species,
12.5 Effects of Fire on Animals each type of animal tends to be associated with spe-
cific vegetation types in a specific type of landscape. As
Antifire educational programs have engendered the fire changes the environment, there is normally a
popular conception that forest and grassland fires are change in the abundance, distribution, productivity,
always accompanied by a mass evacuation of the local and species of animal occupying an area. When a fire
fur and feather populace in a general state of panic. burns erratically through an area, it produces a mosaic
The image of the singed bear cub or fawn left to a of old and young vegetation types, each with its own
fiery fate by the general exodus of animals has con- characteristic fauna. Such fires serve to increase both
vinced many that fire is the enemy of both human and the diversity and the abundance of fauna. In contrast,
beast and that at the first smell of smoke, animals evac- many species of birds and animals are absent in either
uate. However, the scientific literature does not give continuous, dense old-growth forest or an extensive,
unequivocal support for this view, and that animals can completely burned area.
be more abundant in fire-dominated than in fire-free Many animals use different types of vegetation for
environments does not lend much credibility to this different stages of their life cycle and therefore require
342 CHAPTER 12 Fire

Age of stand (years)

0 10 20 30 40 50 60 70 80 90
| | | | | | | | | (

uff Oth
Asven
Hardwood forest succession 10, HOSOI
V]OOM fares

OXY Oj —- 15
0 — 10
Y havo ( iil
il i |
Ill=

Aspen
per ha.
stems _all ieel if
‘il Tree
height
(m)

30,000 15,000 7,500 2,500

Nesting cover —————+

Hardwood
forest
Ge Brood Breeding
cover cover

Breeding (<><
§ | >< 0 —»___ 27 10 —9+_ 0
densities
(grouse per square km)

Figure 12-7 Variation in the utilization of aspen forest by ruffed grouse and other wildlife
species during the first 90 years postfire. (After Gullion, 1972. Used by permission ofthe author.)

a mosaic of vegetation. For example, Figure 12-7 willow ptarmigan; and some of the waterfowl are ben-
shows the use of hardwood forest of different ages by efited by fire, whereas birds such as the spruce grouse,
ruffed grouse. Mammals that depend on late stages of which depend on dense forests, are adversely affected.
forest development and that may be eliminated or dis- Table 12—5 summarizes some of the literature on
placed by fire are mountain, woodland, and barren- species changes after fire.
ground caribou; marten; red squirrel; grizzly bear; Fire plays an important role in maintaining grass-
wolverine; and fisher. Species of large mammal fa- lands against invasion by trees. It has been suggested
vored by fire include moose, white- and black-tailed that herbivores contribute to this role of fire by feed-
deer, elk, cougar, coyote, black bear, beaver, and hare. ing more on the fire-sensitive, less flammable species.
Birds such as the wild turkey; ring-necked pheasant; By leaving the most flammable species, the animals
bob-white quail; short-tailed, ruffed, and blue grouse; contribute to the future reburning of the area and the

lable 12-5 Change in Species of Breeding Birds and Mammals (Excluding Predators and Large Mammals)
After Burning
Number of Species

Foraging Zone Before Burn After Burn No. of Sp. Gained % No. of Sp. Lost %

Species of birds
Grassland and shrub 48 62 18 38 4 8
Tree trunk 29 26 S 20 4 16
Tree 63 58 6 10 11 Aig
Totals 136 146 29 21 19 14
Species of mammals
Grassland and shrub 42 45 7 17 4 10
Forest 16 14 2 13 4 25
Totals 58 59 9 16 8 14

Source: After Bendell, 1974. Copyright Academic Press. Used with permission.
 Effects ofFire on Ecosystems and Ecosystem Processes 343

maintenance of habitat suitable to them. Examples to Table 12-6 Comparison of Some Population
the contrary can also be found. Characteristics of Fire-Adapted (Moose) and Non-
There is a particularly important relationship be- Fire-Adapted (Bighorn Sheep) Large Mammalian
tween fire and large mammals in northern areas. Herbivores
Where vegetation develops in the continued absence Population Bighorn
of fire, postfire birch and aspen stands are replaced by Characteristic Moose Sheep
dense spruce stands. These shade the soil and lead to
the development of a thick moss layer and forest floor. Birth rate High and variable Low and constant
This insulates the soil and leads to a rise in the level of Dispersal rate High Low
permafrost in the soil, which in turn leads to a decline Abundance Fluctuating Steady
in the chemical quality of the vegetation (see Chapter Major population
15). Fire leads to a warming of the soil, a drop in the limitation Food supply Food supply
level of permafrost, the replacement of conifers by
hardwoods, and a general improvement in plant and Geist, 1971.

animal nutrition. Wildfire from lightning is an impor-

tant natural component of Alaskan ecosystems with-
out which the wildlife ecology of the area would be so dispersal is not an advantage, and the birthrate is
radically altered (Lutz, 1956). In fact, it is believed stable and low with a death rate that is determined
that more ecological damage is done by fighting fire in mainly by food supply. Table 12-6 summarizes some
northern areas than by the fire itself (references in characteristics of the two populations.
Slaughter et al., 1971). Fire can increase the depth of
the active layer (soil unfrozen above the permafrost in
midsummer) from approximately 55 cm to 100 cm, 12.6 Effects of Fire on Ecosystems
with recovery to prefire values in approximately 50 and Ecosystem Processes
years (Viereck, 1973). However, fire in the north is not
always benign, and it can have an adverse effect on By influencing plants, soil, and animals, fire induces
fine-textured, high-ice-content soils of arctic and sub- several changes in ecosystem processes. We discuss
arctic regions. When fire removes the protecting veg- the effects on energy flow and forest biogeochemistry.
etation and organic blanket, the ice melts, leading to
soil subsidence and silt flows. Once this process is
started, extensive areas of stable soil and vegetation
Energy Flow
can become an eroding morass of silt (Heginbottom, Energy flow out of the system is greatly increased dur-
1971; Mackay, 1970; Viereck, 1973). ing a fire, and decomposition after fire may also be ac-
Animals that live in fire-affected environments celerated, although decomposition of charred logs
have evolved various adaptations to this environment. may be slower than uncharred logs. Primary produc-
These adaptations include high dispersal rates and the tion is decreased by the reduction or elimination of
ability to respond to the newfound opportunities cre- plants, and where a fire has reduced the moisture and
ated by fire by having a high and variable birthrate. fertility status of the soil, primary production may be
Species of such environments must be able to tolerate depressed for a long time. Alternatively, primary pro-
rapid changes in their environment. Animals of fire duction may be increased by fire because of the
areas tend to be larger and of the browser and grazer change in species composition and more favorable soil
types, whereas animals of relatively fire-free forests conditions. After a fire, primary production is initially
generally tend to be smaller. An example of adaptation by ephemeral species with little biomass accumula-
to fire habitats is provided by Geist (1971), who com- tion. These species are replaced by herbaceous and
pared population characteristics of a fire follower, the shrubby perennials, in which much of the initial bio-
moose, with those of the bighorn sheep, which lives in mass accumulation may be below ground. Only when
a stable, relatively fire-free environment. Moose are trees recolonize the area does a significant above-
adapted to exploit new habitat by their ability to dis- ground biomass begin to reaccumulate.
perse and quickly increase numbers in a new, fire- Secondary productivity in grazing food chains is
created habitat. For sheep, new habitats appear rarely, generally increased by fire except where severe soil
344 CHAPTER 12 Fire

damage has occurred. Immediately after a fire, there additions of plant ash to the forest floor, it was esti-
may be no secondary production, but as herbaceous mated that 39, 11, 15, 35, and 83% of the nitrogen,
and shrub primary production increases, herbivore calcium, magnesium, potassium, and sodium capital of
production generally increases, usually to above pre- the ecosystem was lost. These most probably are con-
fire values. This continues until plants that have lower servative figures.
value to herbivores invade and dominate the area. Losses of these chemicals in streamwater in-
Much of the effect of fire on energy flow is the re- creased, partly because of increases in streamwater
sult of change in the biogeochemistry of the ecosystem. concentrations but also because of the increased
stream flow (a 15% increase, equivalent to 8.9 cm of
precipitation over the first postfire year (Berndt, 1971;
Forest Biogeochemistry Helvey, 1972). Concentrations of nitrate-nitrogen
In August 1970, much of north-central Washington rose from 0.005 ppm to 0.042 ppm in the burned wa-
State, including the Entiat Experimental Forest, was tershed and to 0.310 ppm when a burned watershed
swept by lightning fires that covered approximately was subsequently fertilized with urea. This is equiva-
47,000 ha. Streamflow and streamwater chemistry had lent to an annual loss (in kg ha“') of 0.008 before the
been measured on several watersheds (473 to 564 ha in fire, 1.92 after the fire, and 3.28 after the fertilization,
area) in this experimental forest for nearly 10 years be- which was at a rate of 78 kg ha™'. Thus, the losses in
fore the fire. The watersheds carried forests of pon- streamwater were insignificant compared with the
derosa pine, Douglas-fir, and lodgepole pine with an losses in the fire itself.
understory including Ceanothus species, bitter brush The small contribution of stream export to nitro-
(Purshia tridentata), pinegrass (Calamagrostis rubescens), gen losses reflects both the loss of most of the nitro-
and numerous forb species. The watersheds, which gen on the site to the atmosphere during the fire and
range in elevation from 550 to 2100 m, are on the east the efficient retention of most of the leached nitrogen
side of the Cascade Mountains. They receive an an- by the upper mineral soil. It may also be the result of
nual precipitation of approximately 58.4 cm. the fact that energy flow in the aquatic food webs of
The Entiat fire produced an average ash mass on many streams and rivers is nutrient limited, especially
the soil surface of 2900 kg ha”, containing the follow- by nitrogen and phosphorus. Much of the inorganic
ing amounts of nutrients (in kg ha“): nitrogen, 23; cal- nitrogen that enters a stream is rapidly taken up by
cium, 314; magnesium, 54; potassium, 70; and sodium, microbes and deposited as organic sediments within
22 (Grier, 1975). Nutrient losses during the fire as the the stream, immobilized by large woody debris, car-
combined result of volatilization and ash convection ried downstream as living organisms or suspended or-
were estimated to be (in kg ha™') nitrogen, 855; cal- ganic matter, or lost to the atmosphere by microbial
cium, 75; magnesium, 33; potassium, 282; and sodium, denitrification. Any or all of these processes can act to
698. The nitrogen loss was considered to be directly reduce streamwater exports to a value well below the
proportional to loss of plant biomass and forest floor. actual loss of nitrogen from land to the stream.
During the first year after the burn, leaching of ash Loss of chemicals from an ecosystem after fire can
resulted in the following measured transfers down into lead to the creation of shrub-dominated heathlands,
the mineral soil (in kg ha™'): nitrogen, trace; calcium, particularly on coarse-textured, low-fertility soils.
149; magnesium, 50; potassium, 92; and sodium, 33. These so-called fire barrens can be seen in Scandinavia
Of these amounts, the following percentages were re- (Viro, 1974), in Canada (Damman, 1971; Strang, 1972),
tained in the upper 19 cm of mineral soil: 90% of cal- in the United States (Forman, 1979), and as extensive
cium, 96% of magnesium, and 91% of potassium. A heathlands in Great Britain (Gimingham, 1972). The
net loss of 29 kg ha! of sodium was observed in this ericaceous shrubs that dominate these heaths are
soil layer. Because the nutrient content of the vegeta- thought to impoverish the soil further by promoting
tion before the fire was not estimated and because podzolization and inhibit invasion of trees by interfer-
much of this would have been added to the soil as ash, ing with the formation of mycorrhizal associations.
it was not possible to measure accurately the percent- Fire in chaparral has been shown to increase the
age of the ecosystem nutrient capital that was lost as a availability of soil nitrogen (Christensen, 1973). Both
result of the fire. However, by ignoring the problem of nitrate and ammonia levels increase greatly after burn-
 Effects ofFire on Ecosystems and Ecosystem Processes 3.45

ing from typically low preburn levels. This may be be- nutrients caused by fire. Few studies have quantified
cause fire removes chemicals that are present in the long-term effects on ecosystem function. Faced with
unburned chaparral, which inhibit nitrogen mineral- this high complexity and frequent lack of long-term
ization by bacteria. Peaks in the nitrate content of un- response data, forest scientists often resort to the de-
burned soil normally occur after the first fall rains after velopment and use of computer simulation models.
a dry summer, but these are thought to be the result of This topic is discussed in Chapter 17, but a good ex-
foliar leaching of nitrate rather than a stimulation of ample of the application of modeling to the effects of
nitrification. The ash from the burned chaparral con- wildfire on a forested ecosystem can be found in
tains appreciable quantities of ammonium nitrogen MacLean and Wein (1980). A useful review of the ef-
because of incomplete combustion (the ash contained fects of fire on nutrient cycling is provided by
38% by weight of organic matter), and the high post- MacLean et al. (1981); see also DeBano et al. (1998).
fire nitrate level undoubtedly results from microbial
nitrification of this ammonium nitrogen.
Of all the losses of nutrients during a fire, perhaps Effects of Fire on Carbon Storage in Forests
the most difficult to measure is the loss in smoke and A severe forest fire can release large amounts of ener-
fly ash. Several studies have reported large losses of gy and CO, back to the atmosphere. Forests that his-
volatile nitrogen during fires (Allen, 1964; Debell and torically had a high fire frequency tend to have little
Ralston, 1970; Feller et al., 1984; Knight, 1966), and surface accumulation of carbon in the forest floor, de-
large losses of phosphorus during the burning of caying branches and logs, and standing dead trees.
herbaceous communities have been reported (Lloyd, Such forests are also generally open with few trees per
1971). Not all of these local losses are permanently re- hectare (e.g., Covington and Moore, 1994). In con-
moved from the ecosystem. A fire in a South Carolina trast, forests in humid climates that experience fire
pine forest resulted in a doubling of the cation content only every few hundred years may accumulate large
of rainfall and dry fallout downwind (Lewis, 1974), quantities of organic matter in live and dead trees, de-
whereas 30% of the elements lost as fly ash during the caying logs, and, depending on climate and site, a
burning of an old field in Ontario were deposited in thick forest floor. According to fire severity, a highly
the adjacent downwind area (Smith and Bowes, 1974). variable portion of this aboveground accumulation
An intense fire that occurred in a Douglas- may be returned to the atmosphere by the fire, and in
fir-Ponderosa pine forest in Idaho produced a large some forests, much of the organic matter that is not
pall of smoke through which rainfall fell over the fol- consumed will decompose over the subsequent
lowing few days. This precipitation was found to have decades. These forests are characterized by large accu-
concentrations of nitrogen, potassium, calcium, mag- mulations of organic matter and carbon between fires
nesium, and sodium that were 21, 38, 19, 27, and 71 and large releases during and after the fire.
times higher, respectively, than rain falling before the It is postulated that there was just such a build-up
fire. Although this seems to be a dramatic increase, the of carbon in boreal forests during the “little ice age”
estimated addition of nutrients gained from the smoke and that, on the basis of the present age class structure
through wet and dry impaction were calculated to of the Canadian boreal forest, much of this accumula-
provide only 1 to 4% of the average net annual in- tion was released in the early and mid-19th century,
crease in forest biomass nutrient content. Conse- when massive fires swept the region. Climate change
quently, this return is probably of limited ecological thus could cause large fluctuations in carbon storage
significance, most of the fire losses being removed over extensive areas of forest by its effects on forest
from the general area of the fire (Clayton, 1976). fires. This has important implications for our under-
The effects of fire on nutrient cycling are complex, standing of the role of forests in regulating atmos-
and our knowledge of these effects is still incomplete. pheric CO) concentrations (Kurz and Apps, 1992).
Consequently, it has proved to be difficult to make The possible relationship between sunspots and
quantitative predictions concerning the long-term ef- forest fires and between forest fires and the 10-year
fects of fire on forest biogeochemistry on the basis of cycle of snowshoe hare in Canada (Sinclair et al.,
conventional research approaches. Most studies have 1993) illustrates yet another way in which forest fire
focused simply on the changes in the distribution of may help to determine the ecological character of
346 CHAPTER 12 ‘Fire

northern forests. The issue of the 10-year cycle is dis- tree mortality. Severe wildfires promote the growth of
cussed again in Chapter 14 (see “Cyclical Population legumes, which increase site nitrogen status through
Fluctuations”). It is mentioned here to illustrate the symbiotic nitrogen fixation, thereby increasing the
important and frequently overlooked linkages be- vigor of jarrah and its resistance to Phytophthora. Se-
tween physical factors and the population dynamics of vere fires also remove Banksia grandis, another tree
plants and animals. species that is highly susceptible to Phytophthora.
Dieback of jarrah forest is particularly severe when
there is an understory of banksia because the fungal
12.7. Effects of Fire Exclusion infection builds up on the banksia and then spreads to
the jarrah. Low-intensity prescribed fires, conversely,
Because fire is a natural factor of the environment, promote the growth of banksia and the subsequent
fire-affected ecosystems have become adapted to a par- dieback of jarrah. Control of wildfires and their re-
ticular range in frequency and intensity or severity of placement by prescribed burning has rendered the jar-
fire and will remain in their natural condition only if rah forest far more susceptible to this root pathogen
this range remains the same (Bradstock et al., 1995). (Christensen et al., 1981; Shea, 1979). Additional ex-
Human activities have often altered both of these. Fire amples of the complex interactions between fire and
has been introduced to ecosystems in which it has his- the biota can be found in Gill et al. (1981).
torically been rare, and the occurrence of fire has been Fire is thought to be the primary natural ecological
greatly reduced in some ecosystems in which it was al- factor governing the distribution and abundance of
most an annual event. In the former case, fire-sensitive dwarf mistletoes in North American forests (Alexan-
communities have been replaced by fire-resistant com- der and Hawksworth, 1976). Its effect is complex, and
munities. In the latter case, conditions in which the fire may either encourage or discourage these para-
relatively benign natural fire has gained the potential sites. Where the burn is complete over large areas, fire
to produce widespread destruction have been created. excludes dwarf mistletoe because this parasite recolo-
In fire-adapted forests with natural variation in fire nizes an area slowly. Conversely, partial burns that
frequency, fuel accumulation is prevented and regen- leave scattered groups of infected trees create ideal
eration is limited, so surface fires tend to be of low in- conditions for rapid spread of the parasite throughout
tensity and do not turn into crown fires. Such the subsequent young stand. Wildfires also exacerbate
low-intensity surface fires generally have a net benefi- this pathological problem by converting nonsuscepti-
cial effect on ecosystem function. When fire is ex- ble climax (see Chapter 17) forests, such as spruce-fir
cluded from this type of forest, surface fuels in the Rocky Mountains, into susceptible seral forests
accumulate, eventually resulting in intense ground such as lodgepole and jack pine forests. Frequent
and surface fires. Dense regeneration of trees provides fires also prevent the natural selection of mistletoe-
a fire ladder by which a crown fire can be created from resistant genotypes.
a surface fire (Cooper, 1961). In fact, the net result of Some northern high-elevation and cool, humid,
50 years of successful fire suppression in fire-adapted temperate forests remain as closed-canopy forests only
forests has been the creation of a greatly increased risk if they are periodically disturbed. In the long-term ab-
of fire of greatly increased destructiveness. This prob- sence of fire or other severe ecosystem disturbance,
lem has attracted a lot of attention, and attempts are these forests may be replaced by open woodland with
being made to reintroduce fire to fire-adapted forests ericaceous shrubs or by sphagnum bog forest
(see Agee, 1974; Biswell, 1960; Covington and Moore, (muskeg). The causes of this phenomenon are dis-
1994; Dodge, 1972; Habeck and Mutch, 1973; Kil- cussed in Chapter 17. Suffice it to note here that the
gore and Briggs, 1972; Oberle, 1969; Sampson and occurrence of closed forest over much of this land-
Adams, 1994; Wickman, 1992; and Wright; 1974). scape reflects the historical role of fire.
An interesting example of the effect of fire on the Fire may either increase or decrease biodiversity,
interaction between organisms is the case of dieback depending on which measure of biodiversity you are
of jarrah (Eucalyptus marginata) in Australia. This large considering, and the fire regime. The effects of dis-
tree is susceptible to an introduced root pathogen, turbance on biodiversity are discussed again in Chap-
Phytophthora cinnamomi, which can cause extensive ters 15 and 17.
 Take-Home Message 347

12.8 Fire and Forest Management tation can vary from beneficial to highly detrimental,
whereas the major effects on animals will depend on
Uncontrolled wildfire and intensive forest manage- whether the species is favored by the changes in the envi-
ment have traditionally been considered incompati- ronment wrought by the fire.
ble. The forester wants to harvest much of the Fire does have direct adverse effects on plants and an-
biomass that is consumed in a wildfire and does not imals, but these are often relatively short-lived. In con-
trast, the indirect effects that result from fire-induced
want the undesirable changes to soil and the hydro- alterations in soils can be much more persistent. Soil
logical cycle that are the frequent aftermath of an in- structure, organic matter content, and nutrient status all
tense summer wildfire. Conversely, fire does have can be adversely effected by hot, persistent fires, with
some uses in forestry. It can be used to manipulate the consequent reductions in plant productivity. On some
depth, chemistry, and decomposition of the forest sites in some forest types, the long-term reductions in
floor and the temperature of the mineral soil. It can forest productivity may be more serious than the killing
be used to sanitize an area of diseases, parasites, and of the present forest cover.
insect pests. It can dispose of unwanted biomass Foresters have traditionally considered wildfire as
(postharvest logging residue or slash), thus reducing an enemy to be excluded at all costs. However, there is
the fuel accumulation problems, improving regenera- growing evidence that fire plays an important and either
a beneficial or a benign role in some ecosystems. The
tion possibilities, and improving access for wildlife.
organisms native to such areas have evolved with fire
Brush problems can be abated temporarily by fire, and may grow better with a natural fire regime than
and established stands can be fireproofed by periodic with no fire. Overprotection from fire can result in un-
prescribed burning. desirable changes in the plant community and render
There can be little doubt that managed or pre- the community susceptible to serious damage should a
scribed fire has an important role to play in forest fire occur. Much more needs to be known about the ef-
management, especially during the conversion of fects of fires of different types on particular ecosystems
overmature virgin forests into second-growth forests. so that we can prevent destructive wildfire but retain
There can also be little doubt that we will continue to benign and beneficial fire where it is a natural and desir-
suppress many wildfires. To be useful to the forester, able environmental factor. Where historical wildfire
regimes have been altered because society is not pre-
prescribed fire must be applied at the time of year
pared to accept the risks to life, property, and resource
when there is a reasonable chance of achieving given availability or the impacts of wildfire on forest ecosys-
objectives. The important difference between man- tems, forest managers should attempt to mimic the de-
aged fires and wildfires must be made clear to the pub- sirable effects of natural wildfire regimes in their
lic so that they will accept what may to a layperson management.
seem to be an extraordinary policy. They must be
helped to understand why a forester may deliberately
burn an area in the spring or fall, when money and TAKE-HOME MESSAGE
time were invested putting out a fire the previous Fire is as natural a component of many of the world’s
summer in the same area. forests as rain or wind. Although it is much less fre-
quent, its overall effects can be just as important. In
some forests, the disturbance caused by fire maintains a
SUMMARY low species diversity and/or low site productivity,
Relatively few terrestrial ecosystems have not, at one whereas in others, periodic fire is the factor that main-
time or another, been affected by fire. For most ecosys- tains diversity and productivity. Clearly, the ecological
tems, the frequency of fire is low, but fire is such a pow- role of fire varies from one type of forest to another,
erful ecological factor that even at a frequency as low as just as different fire severities result in different ecolog-
once every two or three centuries, it may be a major de- ical effects.
terminant of ecosystem character. Because fire threatens human safety, habitation, and
Generalizations about the ecological effects of fire property, societies that have the human resources
are unreliable because of variations in the frequency, and/or technology needed to control wildfires have
severity, intensity, and type of fire; in the character of the changed the frequency, scale, and severity of fires. This
physical environment (climate and soil); and in the has led to fundamental changes in the ecological char-
species and their adaptations to fire. The effects on vege- acteristic of the affected forests and in their resistance
348 CHAPTER 12 Fire

to disturbance, including fire. Forest managers must STUDY QUESTIONS

understand the ecological role played historically by
fire in the forests under their control and recognize 1. What is the major cause of forest fire?
that if the “health,” “integrity,” diversity, and produc- 2. List and briefly describe the major ecological effects
tivity of these ecosystems are to be maintained within of fire.
the historical range of values, then fire effects that are 3. How have plants adapted to fire?
no longer occurring will have to be created or mimic-
ked by their management practices. To do this success- 4. How are animals affected by fire?
fully requires a thorough knowledge of forest 5. How does fire affect ecosystem function?
ecosystems, the ecological effects of fire, and the eco- 6. What is the relationship of forest fire to climate change?
logical consequences of different fire management 7. What role does fire play in maintaining the health of
practices. forests?
 Patterns of Biotic
Communities Along
Environmental Gradients

13.1 Introduction involves the philosophical basis for conflicting views

about the nature of plant communities.
We have examined the classification of spatial diver- The sequence of biotic communities along an en-
sity in forest ecosystems (Chapter 6) and the six vironmental gradient is called a coenocline, or, simply, a
major factors that define variations in the ecological community gradient. Vhe assemblage of physical envi-
stage and the associated spatial biotic diversity. This ronmental factors that change as one moves along a
chapter examines the distribution of species along coenocline is called a complex gradient; the combined
environmental gradients defined by these key physi- community—environment gradient is called an ecocline.
cal factors. One can observe an ecocline as one travels from sea
Communities vary in their composition and struc- level to the top of a tall mountain or from the tropics
ture according to variations in the physical environ- to the tundra. Local ecoclines can be seen along soil
ment, and both local and regional gradients of moisture and fertility gradients associated with local,
physical factors are associated with characteristic pat- relatively small-scale topographic gradients from
terns of ecosystem types and biome types, respectively. ridge top to valley floor.
The variation in physiognomy along such gradients is
accompanied by variations in species composition.
Classification assumes that the objects being classi- 13.2 Possible Patterns of
fied can logically be arranged into classes. In reality, Species Distribution
most of the environmental factors that define the eco-
logical stage change along a continuum of variation. There are several possible ways in which plants could
Sometimes this continuum is remarkably even, ren- distribute themselves along environmental gradients
dering classification difficult and arbitrary, as many (Figure 13-1).
objects are intermediate between adjacent classes as
are legitimate members of the classes. Classification in 1. If the dominant tree species occupy sharply defined
such environments tends to have low predictive power segments of the environmental gradient and if there
and is an inefficient way of explaining and reducing is a strong association between the dominant trees
spatial diversity. Analysis of distributions of species and subordinate vegetation, then plants may form
along environmental gradients in such landscapes may characteristic groupings of species from all layers,
be a more useful approach. The characteristics and and these groupings may exclude each other at sharp
causes of such distributions ranked as one of the more boundaries along the gradient (Figure 13-1A).
controversial topics in plant community ecology for Strong competition between the dominant species
several decades. The topic is still important because it produces strong mutual exclusion, so each dominant

349
 Species of:
- Tree

O—O
TOS g
eSee Saee iL
ae

Species
importance
value

Environmental gradient

Figure 13-1 Seven hypothetical patterns of distribution of

plant species along an environ-
mental gradient such as moisture or elevati
on. (A) Mutually exclusive associations with
overlap. (B) Mutually exclusive associations with little
broad overlap. (C) Mutual exclusion within
but no association between layers. (D) Similar to C, a layer,
but with broad overlap of species distributions
within a layer. (E) Tree layer shows broad overlap
of species distributions, but lower layers exhibit
mutually exclusive associations. (F) The species in
all layers enter and drop out of the community ir-
regularly along the gradient without any identifiable
groupings or patterns of exclusion. (G) Species
distributions exhibit various combinations of the
above, with different patterns of species replace-
ment in different parts of the gradient. Each line
can represent a single species or a group of species
in a particular layer. (Reprinted with permussion
of Macmillan Publishing Company from Communities
and Ecosystems. Second Edition by Robert H.
Whittaker. Copyright © 1975 by Robert H. Whittak
er.)
350
 Possible Patterns of Species Distribution 351

species occupies a particular section of the complex climatic factors (which vary gradually along a
gradient. Each dominant species creates or is associ- major complex gradient), and there are many
ated with a particular habitat in which a characteris- species in the lower layers that are responding pri-
tic group of subordinate species develops. The marily to a complex of local site factors (e.g., soil
groupings represent adaptation of the subordinate characteristics) that vary considerably over short
vegetation to the presence of the dominant species, distances (Figure 13-1E).
to each other, and to the physical environment. This 6. If all the species distribute themselves along the
pattern could also result in other ways. For example, environmental gradient solely in response to their
if the plants in all layers respond similarly to the gra- individual environmental tolerances and indepen-
dients of the major environmental factors and com- dent of the presence of other species, then species
petitively exclude other species in the same layer, in all layers would appear and drop out again in an
then grouping could occur without any necessary in- irregular manner as one moves along the complex
teraction between the species in different layers. gradient. There would be neither identifiable
This pattern is characteristic of steep environmental grouping between layers nor sharp exclusion of
gradients (i.e., a marked change in en-vironmental species within a layer. Rather, there would be a
conditions over relatively short distances). continuum of vegetation change along the environ-
Similar to pattern 1 but with broadly overlapping mental gradient (Figure 13-1F).
ranges because of less intense competition and 7. The distribution of species along the gradient can ex-
less competitive exclusion among the dominant hibit various combinations of the foregoing patterns
species. There are identifiable groupings involving at different locations on the gradient or can have a bi-
all layers, but their ranges overlap broadly (Figure modal or multimodal distribution (Figure 13-1G).
13-1B). This pattern is characteristic of gentle en-
vironmental gradients.
. If the species in any layer compete strongly with Major Schools of Thought Concerning
other species in that layer but exhibit only a weak Species Distributions
response to species in other layers or if the species
European schools of plant synecology have tradition-
in different layers are responding to different envi-
ally maintained that species are associated in charac-
ronmental gradients, then there may be very little
teristic groups that are more or less mutually exclusive
overlap in the ranges of species within a particular
and limited by fairly distinct boundaries (Figure
layer but little or no grouping of species between
13-1A). Such groups, which are characteristic of par-
layers. There is little association between species
ticular types of physical environment, are termed
in different layers and therefore no clearly identifi-
associations: plant communities of definite species com-
able plant associations as one proceeds along the
position and characteristic physiognomy, growing in
gradient. This occurs where the complex gradient
particular, homogeneous habitat conditions. The term
and intralayer competition are the major determi-
“association” has not always been used in this context.
nants of species distributions (Figure 13-1C).
American ecologists in particular have tended to apply
Similar to pattern 3 but with broadly overlapping it more broadly to include all communities dominated
ranges of species with any layer. There is some de- by a particular species. For example, a Douglas-fir “as-
gree of mutual exclusion within a layer but no ob- sociation” would include all forest plant communities
vious associations of species in different layers in a region that are dominated by Douglas-fir.
(Figure 13-1D). The essence of the European association concept is
. If the dominant tree species do not exhibit mutual that (1) the individual species in the association are, to
exclusion but the subordinate species do, then the some extent, adapted to each other; (2) the association
species in the upper layer (i.e., the tree canopy) is made up of species that have similar habitat require-
have broadly overlapping distributions along the ments; and (3) the association has some degree ofinte-
complex gradient, whereas species in the lower gration.
layers show marked grouping. This pattern occurs The association concept involves the idea
where there are relatively few species in the domi- of ecological groups. Although no two species have
nant layer, all of which are responding primarily to identical ecological relationships, many species are so
352. CHAPTER 13 __ Patterns of Biotic Communities Along Environmental Gradients

similar to certain other species in their tolerances and along environmental gradients. A large number of
distribution that they can be considered to have essen- ecologists have subscribed to this viewpoint, and it is a
tially the same overall adaptations to their environment common experience to observe discrete associations
and belong to the same ecological group. The term can while passing through a forest, especially where envi-
be applied to either the total set of species from all veg- ronmental gradients are steep and the area has not
etation layers that share the same tolerances or to been seriously disturbed recently by human or natural
groups of plants restricted to species with the same disturbance factors. Figure 13-2 illustrates the idea of
growth form within a vegetation layer (a synusia; see an ecological group.
Chapter 15, “Community Structure as a Result of A different concept of vegetation was developed
Plant Life Form”). It can refer to the total set of eco- independently by two plant ecologists in the 1920s:
logical attributes of a species or to a restricted set, such Ramensky (1924) in the Soviet Union and Gleason
as the moisture or the nutritional tolerances. The exis- (1926) in the United States. They considered that
tence of ecological groups and the adaptation of the species are distributed along environmental gradients
members of an association to each other results in solely according to their individual adaptations and
some degree of mutually exclusive grouping of species tolerances. The community of plants observed in a

Ca H,0 Mg
Oryzopsis
Thalictrum
2 Urtica Cornus
= Sorbus
t= Sorbus

3 Lycopodium \ Mineanie ene

gs : 7 Pinus ) : on
9 Pinus / /
O. If /
n
/

iY \ A
Ly gee 3.4 10 20 40 100 0 20 0 60 80 100 2 34 10 20 40 100
% available calcium % available water % available magnesium
(a) (b) (c)
Calcium Group Water Group Magnesium Group

Ecological No. Species No. Species No. Species

Group Limit in Group Limit in Group Limit in Group

| <2.3 14 =15 8 <I 14

Il 2.3-4.5 31 16-30 49 1.5-3.0 30
il 4.6-9.0 34 31-45 56 3.1-6.0 19
IV 9.1-18 26 46-60 29 6.1-12 32
V 18-36 10 61-75 5 12-24 31
Vi 36-73 20 =76 9 24-48 10
Vil >72 6 >48 9

Figure 13-2 Distributions of selected plant species along gradients of available soil calcium
(A), moisture (B), and magnesium (C) in north-central British Columbia. Ecological-group
limits for these three parameters in this area and the number of species in each group are listed
below. The horizontal axis and group limits are expressed in terms of percentage of the maximum
value recorded for the parameter. For example, 80 on the calcium scale means a value 80% as large
as the highest calcium value recorded. Only the generic names of the species are shown. (Modified
after Wali and Krajina, 1973. Used with permission of Dr. W. Funk Publishers and M.K. Wali.)
 Possible Patterns of Species Distribution — 353

particular habitat consists of populations of those bution of species along gradients that distinct group-
species that are able to invade, survive, and reproduce ings could not be identified (Figure 13-3).
successfully in that environment. Each species is dis- Although the entire community was studied by
tributed independent of other species according to ge- these ecologists, the focus of attention tended to be on
netic, physiological, and life-cycle characteristics that the tree layer, and the resulting analyses lend support to
determine how it relates both to its physical environ- the continuum theory. This may be, however, as much
ment and to other species. Because of the multifactor- a result of the emphasis on overstory species as of a lack
ial nature of species characteristics, no two species are of associations. Identifiable groupings of plant species
alike in distribution. This lack of similarity results in a tend to be much more evident in subordinate or under-
scattering of the centers of species distributions along story vegetation than in the overstory layer, because
the gradient, generally with broad overlap. Communi- many tree species grow on a range of sites within a
ties that consist of species whose distributions happen given climatic region, whereas many herbs and shrubs
to overlap intergrade continuously except where are found grouped only in restricted locations along an
marked environmental discontinuity or disturbance environmental gradient. Although all plants are affect-
by fire, logging, etc. occur (Whittaker, 1975a). This ed by both climatic and soil factors, trees tend to be
individualistic and continuum hypothesis of plant distri- more influenced by the former, which tends to change
butions views the vegetation of a region as a series of slowly and continuously across a landscape. Minor veg-
plant species populations distributed independent of etation is affected more by the latter, which tend to vary
each other along the physical gradients of the environ- more abruptly and over shorter distances. ‘Thus, analy-
ment, as suggested in Figure 13-1F. sis of tree data will often support the continuum hy-
The evidence used by Whittaker in support of the pothesis, whereas analysis of minor vegetation will
individualistic hypothesis was obtained in studies of the often support the individual association concept.
distribution of tree and subordinate plant species along Analysis of community data from areas with a diverse
altitudinal gradients in various mountainous regions of tree flora and a poorly developed or poorly studied sub-
the United States (Whittaker, 1956, 1960). Other evi- ordinate flora will tend to support the continuum theo-
dence has come from studies of forests in Wisconsin ry more than data from areas with very few overstory
(e.g., Curtis, 1959; Curtis and McIntosh, 1951). Using species and a well-documented subordinate flora.
the approach to vegetation analysis known as gradient A major problem in considering the distribution of
analysis and the technique known as ordination, it was plants along environmental gradients arises from the
concluded that there was so much overlap in the distri- confounding of gradients of soil moisture and fertility

400

400

Stems
hectare
per

ere Dry
Moist
Moisture gradient

Figure 13-3 Distribution of tree species along a topographic moisture gradient. The upper
graph uses data from the Siskiyou Mountains in Oregon (760 to 1,070 m). The lower graph uses
data from the Santa Catalina Mountains in Arizona (1,830 to 2,140 m). The graphs show numbers of
stems greater than 2 cm diameter per ha. (Reprinted with permission of Macmillan Publishing Company
from Communities and Ecosystems, Second Edition by Robert H. Whittaker. Copyright © 1975 by Robert
H. Whittaker.)
354 CHAPTER 13 Patterns of Biotic Communities Along Environmental Gradients

(associated with local topographic features) with cli- grass prairie to eastern deciduous hardwood forest in
matic gradients (associated with major topographic the Midwest of the United States.
features (Figure 13—4)). Species may have discontinu- The western low-elevation tree lines are often the
ous local distributions and form local associations and result of sudden changes in soil texture that induce sud-
at the same time exhibit more of a continuum of den changes in soil moisture availability. For example,
change along regional climatic gradients. valley bottom soils in many interior valleys in southern
British Columbia are deep, fine-textured lacustrine de-
posits laid down in extensive postglacial lakes. With the
13.3. Major Forest Ecotones: present-day climate, such soils are too dry for tree
The Interface Between Forests and growth and support native bunchgrass or semidesert
Communities of Different Physiognomy shrub communities. In contrast, the valley slopes have
medium-textured till soils, often with slope seepage
If it is agreed that a coenocline may sometimes be within the rooting zone. Such soils generally support
made up ofa series of discrete communities, then tran- Douglas-fir-ponderosa pine forests (Figure 13-5).
sitions between the communities can be identified. The broad prairie ecotone is under the primary
These transitional areas are zones of tension between control of climate (Curtis, 1959; Weaver, 1968).
the two adjacent biotic associations in which the mem- There is a gradual increase in precipitation and P/E
ber species of each community compete with those of ratio (see Chapter 10) as one moves east from the open
the other community for resources and occupancy of grassland to the closed forest, and the location of the
the land. These transition zones, which are called ecotone correlates reasonably well with a critical range
ecotones, can be seen most clearly where the adjacent of P/E values. However, insufficient precipitation
communities have a different physiognomy as well as a alone is not an adequate explanation for the distribu-
different species composition, as they do at the inter- tion of the prairie grasslands and the ecotone. Scat-
face between grassland and forest or between the com- tered patches of tree growth occur throughout most of
munity of a clearcut and the adjacent uncut forest. the prairie (Wells, 1965), and the climate of much of
Ecotones are interesting because the plants in the area is not too dry for trees (Eyre, 1968). Fire and
them are often growing under conditions of physical grazing have also played important roles, both singly
and biotic stress. By studying the reasons for natural and in combination.
ecotones, we can reveal much about the complexity of Fire has been an important natural component of
the way in which the structure and function of biotic grasslands, especially autumnal fires started by light-
communities are determined. Perhaps the best known ning or humans. The early colonization of the prairies
examples of ecotones are tree lines: the interface of was apparently affected by these grass fires, because
forest and grassland, of forest and alpine meadow, and settlers would not build their houses beyond the shel-
of forest and arctic tundra. ter of woodlands as a result of the danger posed by
grass fires (Sauer, 1969). Fire kills tree seedlings that
are invading grasslands, but it can actually increase the
Low-Elevation Tree Lines
productivity of the grasses (Kucera and Ehrenreich,
Low-elevation tree lines occur where forest gives way 1962). This may be because the removal of the mulch
to grassland. The ecotone may be very narrow, with a of dead grass permits the soil to warm up more rapid-
sharp transition from grassland to forest, such as the ly in the spring (promoting root growth) while burn-
conifer forest-grassland transition common in the ing releases nutrients and may destroy toxic organic
western mountains of North America. Alternatively, it compounds in the litter (Weaver and Roland, 1952).
may be more gradual, as in the transition from long- However, fire does not always have a beneficial effect

Figure 13-4 Distribution of major tree and selected herb and shrub species along local topographic sequences and along
the major elevational gradient from sea level to 1,500 m near Vancouver, British Columbia. Species can have unimodal or bi-
modal distributions along local topographically induced moisture/nutrient gradients with sharp boundaries to their discontinuous dis-
tributions, whereas their overall elevational distribution may be unimodal (shown for trees by dotted lines). Note that the distribution
of most of the understory species is much narrower than that of most of the overstory species. The trees are strongly influenced by the
climate as well as the soil, whereas the understory vegetation is more strongly influenced by local soil conditions. The wet or dry sub-
zone definitions are a simplification of the subzones defined for this area.
Tree species

Pacific
madrone

Lodgepole
pine

Bigleaf
maple

Western
redcedar

Douglas-fir

Western
hemlock

Pacific
silver fir

Mountain
hemlock

Understory species
Bearberry
Salal
Juniper
Sword fern
Deer fern

Salmonberry

Twisted stalk
Alaskan
blueberry
Black mountain
huckleberry
White
rhododendron

Local
topographic
position

Relative
moisture
status

Average ee
elevation, aa

“ eee 7711111 A
Coastal western hemlock zone See ane
Saige
ee
re Coastal Douglas-fir zone
es
Subzone Dry | Wet i Dry Wet Forested
L
356 CHAPTER 13 Patterns ofBiotic Communities Along Environmental Gradients

fire, lower winter snowpack, and increased grazing tend

to result in a slow expansion of the grassland. Patterns
of tree seedling invasion of grassland that are often hard
to explain on the basis of midsummer studies may be re-
lated to differential snow accumulation in the winter.
Seedlings in low-elevation forest-grassland ecotones
have a higher survival rate in areas that carry large win-
ter snow drifts than in areas of average or below-aver-
age snowpack because the prolonged snowmelt of the
former area keeps the surrounding soil charged with
moisture far longer into the hot, dry summer.
In addition to the factors of precipitation, P/E ratio,
soil texture, fire, and grazing, low-elevation tree lines e
can be a product of cold-air drainage. Billings (1954) re-
Figure 13-5 lLow-elevation tree line where forest gives
way to grassland—the interface between valley bottom
ported that the lower limits of a pine—-juniper woodland
grasslands and ponderosa pine-Douglas-fir forest in on a mountain in Nevada seemed to coincide with a
south-central British Columbia. pronounced temperature inversion. ‘Temperature inver-
sions in valleys can result in grassland or meadow vege-
tation because of frost damage to trees or because cold
on the growth of grasses, and it can increase tree inva- air increases snow duration. Although snow accumula-
sion by reducing competition for soil moisture. Burn- tion generally promotes tree invasion into grasslands, if
ing very dry grassland may adversely affect both the snow-free period is too short, then the trees will be
grasses and trees because the accumulated grass litter adversely affected and a meadow will develop.
serves as a vapor barrier that limits evaporation and The low-elevation forest—-grassland ecotone is of
thus conserves soil moisture (Dix, 1960). Also, by re- considerable interest to forest management. Should
moving the shade provided by the grass, fire can raise such areas be subjected to periodic prescribed burning
summer surface soil temperatures to levels that are to promote grass production for use by wildlife and/or
lethal to tree seeds and young seedlings. domestic livestock, or should they be protected and
Grazing affects the nature and location of the eco- tree production be encouraged? World food shortages
tone by killing small seedlings and modifying competi- and pressure favor the former, whereas from a wood
tion between trees and grass. Occasional heavy products perspective, the land base for tree production
grazing, such as was produced by the original roaming should be maintained against encroachment by other
herds of native ungulates, does not adversely affect the land uses or expanded.
production of native grasses (Alcock, 1964), but sus-
tained heavy grazing can lead to a change in the species
composition of the grass community and/or a replace- High-Elevation Tree Lines
ment of the grasses by shrubs or trees because of re- The ecotone between alpine and montane forest plant
duced competition for soil moisture (e.g., Buffing- communities is perhaps the best known and most
ton and Herbel, 1965). studied example of tree lines. The term timberline is
The location of the low-elevation tree line shows a sometimes used synonymously with tree line, but the
temporal variation because of alternate invasion of trees two terms are in fact different. Timberline refers to
into the grassland and loss of trees from the ecotone. the upper elevational limit of commercial timber,
These events are erratic and depend to a considerable which is normally the upper elevational limit of closed
extent on climatic cycles. Periods of wetter summers, forest. The timberline thus is synonymous with the
greater snowpack, lower fire frequency, and reduced forest limit. Trees in the ecotone area (correctly re-
grazing pressure are generally accompanied by an inva- ferred to as the subalpine zone (Love, 1970)) either do
sion of the grassland by trees. Periodic intense droughts not reach commercial size or, if they do, are scattered
(e.g., the 1934 to 1941 period in the Midwest, when it in “tree islands” that have low commercial value and
was estimated that 50 to 60% of the trees in Nebraska therefore cannot be considered as “timber” (Figure
and Kansas died [Weaver, 1968]), higher frequency of 13-6). Stunted individual trees (krwmmbolz) may be
 Major Forest Ecotones: The Interface Between Forests and Communities of Different Physiognomy 357

found growing many hundreds of meters above the

forest limit, so the tree /ine or tree limit is often very
different from the forest limit or timberline (Figure
13-7). In some areas, such as the high-elevation &
Nothofagus forests of New Zealand, the tree limit and
the forest limit coincide; there is a sudden transition
from commercial forest to alpine vegetation.
Elevational timberlines may be formed by either
conifers or angiosperms. The Pinaceae are the pre-
dominant timberline trees in much of the northern
hemisphere (mainly species of Pinus, Picea, Abies, Tsuga,
and Larix, although Juniperus and Chamaecyparis of the
Cupressaceae are also represented). However, species
of Betula and Alnus form both northern and upper ele-
vational tree lines over much of Scandinavia and Asia
Figure 13-6 ‘Tree islands of Abies lasiocarpa in the sub- and in parts of the Alps and British Columbia. Fagus
alpine ecotone of the coastal mountains of British Colum- sylvatica is an important timberline species in central
bia. Once established as a tree island, the trees grow relatively and southern Europe and the Caucasus, and Acer, Sor-
well. Establishment in the meadow is very difficult. bus, Populus, and Quercus species are also found at the

Forest tree Tree islan Cushion

Closed krummholz krummholz
montane Subalpine ————$ > Alpine
forest vegetation

Foliage just above the winter

snowpack killed by high light
intensity and wind-driven snow
and ice
Wer Height of winter
snowpack
Forest tree Multi-stemmed ~ Elfinwood
krummholz

Increasing harshness of the physical environment

Figure 13-7 Variation in tree growth form in the subalpine zone as one proceeds from
closed montane forest to open alpine communities. (A) Engelmann spruce, Rocky Mountains,
Colorado. (B) Lodgepole pine (given as Pinus murrayana), Sierra Nevada, California. (After Wardle,
1968, and Clausen, 1965. Used by permission of the Ecological Society ofAmerica and Society for Study of
Evolution and the authors.)
358 CHAPTER 13 Patterns of Biotic Communities Along Environmental Gradients

Tuble 13—1 Elevation and Timberline Species at Various Locations Around the World

Abies lasiocarpa 1,850-1,900 50°N Garibaldi Park, B.C.

A. mariessii 1,400-1 ,500 41°N Mt. Hakkoda, Japan

Betula pubescens 1,000 63°N Harjedalen, Sweden

B. utilis 3,500-4,200 29-35°N Western Himalaya

B. verrucosa up to 2,500 41-44°N Caucasus Mts., U.S.S.R.

Eucalyptus coccifera 1,200-1,300 43°S Mt. Field, Tasmania

E. niphophila 1,850-2,000 36°S Snowy Mountains, N.S.W., Australia

Larix decidua up to 2,300 46°N Poschiavo, Switzerland

L. griffithii 3,800 28°N Eastern Himalaya
Nothofagus menziesii 1,200-1,300 42°S West coast South Island, N.2Z.
N. pumilo 1,650 40°S Andina de Nenquen, Argentina
Picea engelmannii 2,150—2,300 50°N Rocky Mts., Alberta
3,350-3,600 39°N Frong Range of Rocky Mts., Colo.
P. likiangensis 4,500 38°N Southeast Sinkiang, China
P. sitchensis 900 60°N Haines, Alaska
Pinus albicaulis 3,300 38°N Sierra Nevada Mts., Calif.
P. cembra 1,900 47°N Glarns, Switzerland
P. hartwegii up to 4,100 19°N Iztaccihuatl, Mexico
P. sylvestris 500 57°N Cairngorm Mts., Scotland
(600-700 potential)
Podocarpus compactus 3,900-4,100 6°S Mt. Wilhelm, New Guinea
Polylepis tomentella up to 4,900 19°S Northern Chile
Rhododendron sp. 3,600-3,800 27°N Eastern Himalaya

Data from Wardle, 1974.

timberline (Wardle, 1974). In the southern hemisphere, imately 30°N and 25°S, with a possible dip near
Eucalyptus and Nothofagus are the important timberline the equator. This change in elevation approxi-
genera. ‘Table 13-1 summarizes the elevation and mates that of isotherms (110-m change per degree
species at timberline in various parts of the world. of latitude) in North America.
The explanation of the location of either the tree . The elevation of timberline is lower in coastal than
line or the timberline is very complex. Numerous en- in interior mountains, which mirrors summer tem-
vironmental factors act to inhibit the upward extension perature isotherms that dip farther south (in the
of closed forest into the subalpine and alpine zones. northern hemisphere) along the coast than in the
The importance of any one factor varies from place to interior of continents. The difference can also be
place, so no single explanation will fit all circum-
due to the Massenerhebung effect: there is a lower
stances. Each potential factor is considered in turn. temperature lapse rate and therefore higher cli-
matic-biotic zones on larger and higher than on
Temperature. Several facts suggest that tree lines smaller and lower land masses (Wardle, 1974).
are largely determined by temperature: The greater temperature range of continental in-
terior areas also serves to raise timberlines because
1. The elevation of timberline increases from polar the lower winter temperatures have much less ef-
regions toward the equator, leveling off at approx- fect than do the greater summer temperatures. For
 Major Forest Ecotones: The Interface Between Forests and Communities of Different Physiognomy 859

a given average annual temperature, interior areas the tree limit (Figure 13-7). Fully illuminated foliage
will be warmer in the summer than coastal areas. of Nothofagus solandri 2 m above the ground in New
3. High-elevation tree lines are higher on south as- Zealand was observed to be 0.5 to 2°C above air tem-
pects than on north aspects in cold, humid climates perature during much of the day, with occasional in-
in the northern hemisphere, where lack of summer creases to 6°C above air values. In contrast, fully
warmth is critical. In hot, dry climates, where high illuminated shoots of the sprawling shrub Podocarpus
summer temperatures are an important limiting nivalis within 3 cm of the ground usually exceeded air
factor, low-elevation tree lines are lower on the temperature by 5 to 11°C and occasionally more
cooler polar aspects. (Wardle, 1974). Thus, trees are able to satisfy their
heat requirements as elevation increases by growing
These observations, together with the broad rela- closer and closer to the ground.
tionship between temperature and biological phenom-
ena in general, support the idea that temperature plays Carbon Balance. Several authors have suggested
a key role in determining tree line elevations. The that the tree limit represents the elevation at which
concept of heat sums and its relationship to northern trees can no longer achieve a positive carbon balance
tree lines was mentioned in Chapter 8. Nineteenth- from year to year. That most timberline species are
century work on this topic suggested that the climatic evergreen has been cited in favor of this idea because
limits of cold timberlines roughly parallel the deciduous species would require higher levels of sum-
isotherms for the mean temperature of the warmest mer energy fixation (more positive CO, balance) to
summer month (Daubenmire, 1954). Above tree lines, compensate for the annual loss of all their foliage and
there is insufficient warmth to permit tree growth. the lack of any winter photosynthesis to offset winter
Success of woody plants at high altitudes depends to a respiration (Bliss, 1966). At a certain elevation, there is
considerable extent on their ability to ripen and insufficient summer photosynthesis because of low
“harden off” new shoots in time for this new growth to temperatures and short growing season to generate
survive low winter temperatures and winter desicca- enough net photosynthesis to support winter respira-
tion, the latter being the most important. Where there tion and growth, let alone reproduction. For example,
is insufficient total summer warmth (heat sum) to per- bristlecone pine (Pinus aristata) at timberline in Cali-
mit the necessary physiological processes to be com- fornia requires 117 hours of photosynthesis at the
pleted in time, trees cannot survive. peak summer rate (equal to half the growing season) to
Other aspects of the temperature explanation of redress the total winter negative respiration balance of
tree lines are lethal maximum summer temperatures 140 mg CO, g dry wt! (Schulz et al., 1967).
and frost heaving. High-elevation plants may experi- Studies of Pinus cembra at timberline in Austria
ence intolerably high temperatures for short periods have shown that this species used only 33% of the car-
of time (Gates and Janke, 1966). For example, surface bon assimilated annually for respiration and increase
temperatures of 84°C have been recorded in subalpine in dry matter, which suggests that carbon balance may
areas in Austria (Aulitzky, 1961), and the surface tem- not be limiting. It is thought that the remaining 67%
perature in a subalpine meadow in British Columbia of assimilated carbon may have been transferred to
was reported to go as high as 49°C, approaching the mycorrhizal fungi (Tranquillini, 1959), which are
range that is lethal for some young tree seedlings (Bal- known to be capable of using a significant portion of
lard, 1972). Frost heaving and its effects on tree the net photosynthate produced by the plant (e.g.,
seedlings were discussed in Chapter 8. Subalpine areas Schweers and Meyer, 1970). This “loss” to mycor-
can experience radiation frosts almost throughout the rhizae represents the energy cost to plants of obtain-
year, and tree invasion of subalpine meadows is often ing the nutrients and moisture needed for survival in
restricted to drier mounds, where frost heaving is less what is a dry and nutrient-poor environment.
intense. Frost heaving is often the factor that excludes Some authors question the importance of the
all vegetation from alpine areas. carbon-balance theory. Nevertheless, there can be lit-
The temperature explanation of tree lines is sup- tle doubt that alpine herbs are at an advantage over
ported by the observation that tree growth-form trees in that they produce less nonphotosynthetic
changes from upright to krummbholz to a shrub or aboveground biomass and do not have to support the
scrub form as one proceeds from closed forest up to respiration of overwintering aboveground biomass.
360 CHAPTER 13 Patterns of Biotic Communities Along Environmental Gradients

They are able to use much more of their photosyn- ‘Typically, tree growth on windy ridges is luxuriant
thate for the production of photosynthetic and repro- up to the height of winter snowpack. Any growth above
ductive structures. the level of snowpack is killed during the winter by me-
chanical damage or desiccation (also by high light in-
Wind. ‘Timberline trees in most extratropical regions tensities just above the snowpack). Where this winter
are so dwarfed and wind-trained (Figure 9-6) that for wind damage is serious, the height of krummholz may
many years ecologists have attributed tree lines to be limited by snow depth. Wind aggravates the prob-
wind. Projecting twigs frequently show the effects of lem of low temperature and desiccation, and wind-in-
fatal winter desiccation (often as a result of wind-in- duced mechanical and physiological damage near the
duced transpiration loss), and at the upper tree limit, tree line is more significant than at lower elevations be-
the krummholz or scrub growth is typically confined to cause of the very slow growth to replace these losses.
the lee of rocks, other vegetation, or sheltered depres- Against the wind theory must be set the observation
sions (Figure 9-7). Timberline can extend more than that high winds are often more common in valleys than
500 m higher on leeward than on windward slopes of on the higher slopes above them and that deformed but
mountains (Schroter, 1926) unless snow accumulation surviving trees are found in very windy locations. Obvi-
on leeward slopes is limiting (Figure 13-8). Pears ously, wind alone cannot explain tree lines.
(1968) showed that tree lines in the Grampian Moun-
tains of Scotland have been considerably depressed Snow Duration. The transition from closed forest
by human activities over the past few millennia and as- to alpine conditions may represent a critical point on
serted that wind is a major factor that has prevented a the snow duration curve at which the snow-free period
return of the forest limit to its original elevation. is too short to permit the establishment and growth of
Reestablishment of conifer plantations on windy moor- trees. Long snow duration lowers the total warmth
land in Great Britain is similarly limited by wind (Lines (heat sum) available to seedlings and small trees in cli-
and Howell, 1963). Wind velocities often rise sharply mates that more than satisfy the requirements of trees
with elevation, in contrast to the more gradual increase above the snowpack. This affects the hardening off
in temperature (Daubenmire, 1954). and net carbon balance of the seedlings and may pre-
vent their establishment and/or survival. The snow
craters that melt around the stems of trees in the
closed forest near timberline permit the establishment
and growth of shade-tolerant seedlings close to the
parent tree, whereas the snowpack between trees may
persist an additional month and preclude seedling
establishment (Figure 14-1). This leads to the charac-
teristic clumping of trees in the closed forest-sub-
alpine interface. In southwestern British Columbia, it
is possible to predict the location of this ecotone by
measuring the length of the snow-free period (or vice
versa) (Brooke et al., 1970).
In northern latitudes, tree growth sometimes oc-
curs well above the general timberline along the south-
facing rim of exposed east-west ridges, despite the low
temperatures and high winds to which such a location
Figure 13-8 Effect of snow duration on mountain vege-
is exposed. This observation can be explained by the
tation in high snowfall areas, facing east along a ridge in
fact that these locations lose their snow cover many
coastal British Columbia. Trees grow on the south side of
the ridge crest, where wind prevents deep snow accumulation weeks earlier than either the north-facing rim, where
and there is early snowmelt. Deeper, more prolonged snow- cornices form if the prevailing winter wind is from the
pack on the north side and lower on the south slope precludes south (Figure 13-8), or the slope below, where the
the establishment of trees. In such environments, the south deep snowpack persists long into the summer.
crest may be sparsely forested, whereas the north aspect has Another line of evidence in favor of the snow dura-
meadow vegetation. tion hypothesis comes from observations of widespread
 Major Forest Ecotones: The Interface Between Forests and Communities of Different Physiognomy 361

invasion of subalpine meadows by conifers in the Cas- usually lower the tree line, whereas grazing eliminates
cade Mountains in Oregon, Washington, and British tree seedlings, so the tree line is gradually lowered as
Columbia (Brink, 1959; Franklin et al., 1971; van the old trees die. The loss of forest cover changes the
Vetchen, 1960). Tree invasion of subalpine and alpine microclimate (mainly wind and temperature), and this
meadows is normally a slow process. One tree estab- may preclude reinvasion. The herbaceous vegetation
lishes and survives. Its presence reduces snow duration, above tree lines encourages burrowing animals such as
leading to further seedling establishment around it and marmots and pocket gophers. The soil churning, the
the formation of a tree clump. Vegetative reproduction winter consumption of vegetation below the snow, and
(layering, in which lower branches that touch the damage to seedling root systems caused by these ani-
ground develop roots and form new individuals that are mals may serve to limit tree invasion of the meadows in
sometimes called veg/ings) and/or sexual reproduction which these herbivores live. Bark stripping of seedlings
then enlarge the groups to tree islands, which may even- by small mammals can also be a potent force restricting
tually coalesce into closed forest, thus raising the tree tree invasion into meadows.
line. This is a slow process that requires many centuries
and is different from the sudden appearance of Other Plants. The area above tree lines is charac-
seedlings over large portions of the subalpine such as terized by vigorous communities of shrubby or herba-
occurred between 1923 and 1944. Studies of these in- ceous growth forms that are well adapted to the alpine
vasions eliminated grazing and fire as causative agents. environment. Tree lines may sometimes represent the
They revealed that at the time of the invasion, the local elevational point at which competition between trees
glaciers were in decline, with a steady recession from and alpine-subalpine plant species is balanced.
1910 to 1953. After this, the glaciers readvanced. The
warmer, drier period that triggered the glacial recession Pathogens. ‘Irees at timberline are under physio-
would have been accompanied by lower snowpacks and logical stress and are therefore more subject to disease
earlier snow melt, facilitating the invasion of snow- and parasite attack than when growing at lower eleva-
determined meadows by trees. tions. Trees in alpine meadows are often adversely af-
fected by snow fungus, which may kill part or all of the
Fire. There can be little doubt that fire has de- foliage buried in the winter snowpack, weakening or
pressed alpine tree lines in many areas. It is the most killing the tree.
important catastrophic event affecting timberline, and No single explanation for elevational tree lines
its effects last longer than those of windthrow, that is satisfactory for all situations has emerged. It is
avalanche, or insects because it usually kills seedlings clear that these ecotones are almost always determined
and consumes forest floor and seeds as well as killing by a complex of environmental factors, the relative
the overstory. Burned areas are usually invaded by lux- importance of which varies from place to place. Some-
uriant communities of flowering plants, and many times it is wind, sometimes it is snow duration, and
subalpine meadows of high recreational value owe sometimes it is summer warmth that is important, but
their existence to lightning fires. Reinvasion by trees is never do these and other factors act alone. Sometimes,
often slow because fire followers such as lodgepole tree lines coincide with abrupt changes of physical fac-
pine and aspen are not normally timberline species tors along an elevation transect. More often, they rep-
(Wardle, 1974). In the Australian Alps, timberline resent the point along a steady gradient of physical
Eucalyptus can survive destructive crown fires by re- conditions at which the trees can no longer tolerate
generating from lignotubers. their environment or can no longer compete with
other life forms. Readers who are interested in this
Animals. Domestic grazing, excessive wild popula- topic are referred to the detailed analysis of tree
tions of introduced grazers (e.g., goats), and outbreaks growth at high altitudes by Tranquillini (1979) and to
of insects (defoliation or bark beetles) all can influence Benecke and Davis (1980).
elevational tree lines by killing seedlings or mature
trees. Grazing tends to be a more insidious factor than Polar Tree Lines
catastrophic insect outbreaks, because the latter occur
only periodically and usually do not kill all the trees and The most northerly forest type in the northern hemi-
their seedlings. Consequently, insect outbreaks do not sphere is the boreal, which gives way on its northern
362 CHAPTER 13 Patterns of Biotic Communities Along Environmental Gradients

boundary to the shrub-herb formation of the tundra. median location of the arctic front. Exactly how these
The species composition at the northern forest air masses influence the vegetation is still a subject of
ecotone varies in different parts of the world. In most speculation (Larsen, 1974).
of North America, black and white spruce are domi- Temperature regimes in arctic and alpine regions are
nant. In Scandinavia, pine, spruce, and birch combine significantly different from each other because areas
at the forest margin, whereas larch dominates at the with similar mean temperatures differ in the amplitude
ecotone in Siberia (Hustich, 1953). of temperature fluctuations. In alpine environments, the
As with elevational tree lines, it is much easier to low mean daily summer temperature is the combination
define the northern limit of continuous forest (25 to of warm daytime temperatures and low night tempera-
30 km) than to define the northern tree limit, because tures. In the Arctic, where the difference between sum-
the latter depends partly on how one defines “tree.” mer day and night temperatures is much smaller as a
The northern limit ofa tree species is highly irregular, result of the short nights, the same mean daily tempera-
diffuse, and difficult to map. ture is associated with much cooler days. This has impli-
It is generally believed that the position of the north- cations for many life forms. For example, the number of
ern forest ecotone is predominantly determined by cli- insect species and their rate of development and the rate
mate, especially temperature. There is a good correlation of organic decomposition all are greater in alpine than in
between potential evaporation (a direct function of tem- arctic areas of similar mean summer temperatures
perature) and the major forest types in Canada (Table (Remmert and Wunderlung, 1970).
13-2). ‘Temperature is more important than moisture, Temperature affects arctic vegetation in several
and the isopleth of 10°C accumulated month-degrees ways. Low temperatures not only reduce metabolic rates
above a 6°C threshold shows a close correlation with the but also produce soil churning and permafrost. Soil
northern forest limit in Canada. In Alaska, there is a cor- churning inhibits the establishment of tree seedlings and
respondence between the forest limit and both day-de- damages the roots of established trees, whereas per-
grees above 10°C and the mean temperature of the mafrost limits root activity, tree stability, and tree nutri-
coldest month (references in Larsen, 1974). tion. Soil under trees is often colder than under the
The climate of the arctic ecotone is influenced by diminutive tundra vegetation to the north, and trees at
airmass climatology. It has been observed that the eco- the forest limit are often associated with permanent ice
tone coincides with the path of summer cyclonic lenses. As a tree grows, it shades the ground beneath its
storms and that the forest border coincides with the crown, thereby preventing the summer melting of per-
mean summer (July—October) position of the climatic mafrost in the shaded area. This leads to the develop-
boundary between arctic air to the north and pacific ment of a lens of ice that grows in thickness each year to
and continental air masses to the south. Similarly, the form a mound with the tree on top. Eventually, the moss
forest-tundra ecotone in Eurasia approximates the layer over the mound becomes broken, exposing the ice
lens, which then melts, collapsing the mound and killing
the tree. The result is a frost scar (a gap in the organic
Table 13—2 Relationship Between Potential layer that exposes the mineral soil) along the edge of
Evaporation (cm) and Major Ecotones in Canada which tree seedlings establish to repeat the cycle (Figure
13-9). For a discussion of permafrost, see Ives (1974).
Typical PE Value,
Ecotone cm As with elevational tree lines, both wind and fire in-
fluence the polar extension of trees. Wind, which helps
Tundra-forest/tundra 30-32 to determine the potential location of the forest eco-
Forest/tundra—open lichen woodland 36-37 tone, acts more through its effects on temperature than
Open lichen woodland-closed
its mechanical effects. Fire is an important factor in de-
boreal forest 42-43 termining the actual location of the ecotone, which
Closed boreal forest-mixed conifer/
often differs from the potential location. Lightning
hardwood forest 47-48 fires burning through the ecotone eliminate trees, the
Mixed conifer/hardwood forest—Great
reinvasion of which is very slow. The action of fire may
Lakes/St. Lawrence mixed forest 50-51 be one reason that the northern forest ecotone is fre-
quently broad, with islands of trees occurring far to the
Data from Larsen, 1974. north of the main body of continuous forest.
 Forestry and Environmental Gradients—The Question of Site 363

Tree and forest floor

‘i are elevated by
growing ice lens.

Disturbance of the
Growth of white forest floor exposes
spruce shades the the ice lens which
ground and leads melts, often tipping
to the development the tree over, creating
of an ice lens. a frost scar that exposes
the mineral soil.

Organic layer ———=

Mineral soil —_*

White spruce seedlings germinate
at edge of old frost scar.

Figure 13-9 Cycle of development and collapse of frost mounds associated with white
spruce in Mt. McKinley National Park, Alaska. (After Viereck, 1965. Used by permission ofthe Arctic
Institute of North America and the author.)

As with the elevational forest ecotone, the polar that can be managed as a spatial unit. For as long as
ecotone varies in location with time. Griggs (1946) re- forestry has existed, foresters have recognized that the
ported on a major northern advance of the forest bor- character of a stand is determined to a large degree by
der around Kodiak, Alaska, over the past two the physical factors that collectively define the site. Site
centuries. Presumably, this reflects the changes in is the physical components and characteristics of the
world temperatures that have occurred over this time forest ecosystem—the local “ecological stage.” Recog-
period. The determination of the northern ecotone is nition of site is one of the reasons that foresters under-
complex, and the interested reader is referred to stand that when managed properly, harvesting trees
sources such as Hansen (1966), Ives and Barry (1974), does not destroy forest ecosystems—as long as the site
Pielou (1991), and Sirois (1992). is left in the appropriate condition and the processes
of reestablishing a population or community of trees
are active and operating at an appropriate rate.
13.4 Forestry and Environmental Site is the soil, geology, topography (aspect, slope,
Gradients—The Question of Site slope position, elevation) climate, wind exposure, cold
air drainage, frostiness, slope stability, hydrology, risk
Field foresters and silviculturists have long been con- of fire, and any other physical and chemical condition
cerned with forests stands—local forest ecosystems or factor that sets the potential for vegetation develop-
characterized by a population or community of trees ment, net primary productivity, and the accompanying
of fairly homogeneous age, structure, and character animal and microbial communities.
364 CHAPTER 13 Patterns of Biotic Communities Along Environmental Gradients

Site may be classified in terms of tree growth po- Finally, the actual location of a species along an envi-
tential (e.g., site index), vegetation community poten- ronmental gradient may reflect the relative competitive
tial (ecological site classification), wildlife habitat abilities of the different species as much as the direct ef-
potential, or some other interpretation of the biotic fects of the physical environment itself. Change in the
relative competitive ability of different species as the
potential or condition of the ecosystem. It should be physical environment changes may be the major factor
and increasingly is the physical framework within determining how different species divide up the environ-
which the stand level objectives of management and mental gradient.
the choice of management practices to achieve these
objectives are chosen. In assessing site, it is always im-
portant that individual factors not be considered in
isolation. It is the combined consequences of the di- TAKE-HOME MESSAGE
rect and indirect (as a result of factor interactions) ef- The spatial complexity of the forest landscape must be
fects of all the physical and chemical site factors that divided into more homogeneous units if forest manage-
collectively determine the biotic potential of a local ment practices that respect the local ecological character
forest ecosystem. of forest stands are to be selected. Where biotic commu-
nities form repeated patterns of groups of species that
are reliably associated with repeated patterns of combi-
SUMMARY nations of environmental factors, it is generally possible
Complexity and interaction between components and to classify the forest landscape into classes of forest
processes are two of the fundamental characteristics of an ecosystem that can be expected to respond similarly to a
ecosystem. The major physical factors described in the particular management regime. This situation tends to
last six chapters do not act in isolation. They are inti- occur most frequently where environmental gradients
mately interconnected, and separation of the effects of are steep and where the plant community is relatively
individual factors is at best difficult and frequently im- species rich. Where physical factors vary continuously
possible. Consequently, physical environmental variation along gentle environmental gradients and where this is
is best considered as a complex gradient that sometimes combined with relatively low species diversity, classifica-
varies continuously and sometimes discontinuously. tion is more difficult and less useful. In such environ-
The response of biotic communities to the complex ments, recognition of how ecosystem function varies
gradient sometimes mirrors the spatial variation in the along the environmental gradient may be a more reliable
physical factors, but often it does not. 0°C on the tem- basis for allocating management objectives and design-
perature scale is biologically important because of its re- ing appropriate practices.
lationship with the liquid/solid phases of water.
Consequently, dramatic biotic changes can occur along a
uniformly changing temperature gradient in the vicinity STUDY QUESTIONS
of 0°C. Similarly, thresholds of tolerance to wind, water
balance, soil texture, fire, and other physical factors can 1. What do we mean by plant physiognomy?
result in abrupt biological changes along a relatively uni- 2. What is a biome, and how is it defined?
form physical gradient.
3. What is a plant association?
Modification of the physical environment by the
4 How are plants distributed along environmental gra-
dominant vegetation adds further complexity. The physi-
cal location of a dominant species along a uniform envi- dients?
ronmental gradient can result in a discontinuous gradient a Which ecological factors act to determine high eleva-
of environmental conditions for nondominant species. tion tree lines?
 Biological Diversity
Population, Community,
and Genetic Determinants
of the Ecological Play

[.organisms existed apart from each other, then evolution

would have resulted in adaptations only to physical factors, and
there would probably be much less diversity of species today than
actually exists. A much more modest array of species—ecological
actors—could have occupied the diversity of physical environ-
ments—ecological stages—that the earth has to offer. That this is
not the case can be attributed to the important role that other or-
ganisms play in the life of any particular plant, animal, or microbe
and the key role that biotic interactions have played in the evolu-
tion of the wide diversity of life forms and species that is our in-
heritance from 4 billion years of evolution.
The ecological play—the change in species composition of the
biotic community (the ecological actors) and the change in the
physical features of the site/ecological stage over time—is to a
large degree the result of processes associated with groups of sim-
ilar organisms—populations—and groups of dissimilar organisms—
communities—the interactions of which are conditioned by the
genetically controlled characteristics and tolerances of the organ-
isms involved.
In the most severe physical environments, organisms may not
have been abundant enough for interactions between them to
have played the dominant role in their evolution. In more moder-
ate environments and because species have evolved adaptations to
permit them to tolerate an amazingly wide range of physical con-
ditions, organisms generally exist at levels of abundance at which
biotic interactions have been a major determinant of their diver-
365
366 PARTIV _ Biological Diversity

sity, abundance, and distribution. Over most of the earth’s surface,

the evolutionary development of organisms and the form and
functioning of ecosystems have been very strongly influenced by
inter- and intraspecific interactions. In such environments,
coevolution (the evolution of interacting species under the influ-
ence of the selective pressures that they impose on each other) has
frequently been a major force in determining the character of the
community and the functioning of the ecosystem.
Managers of renewable resources must be mindful of the wide
variety of interrelationships between organisms of the same and of
different types and of the genetic control over the adaptation of
organisms to their biotic and physical environment. Such rela-
tionships and adaptations are of key importance in determining
the success of forest resource management. This part of the book
addresses intraspecific interactions and processes—population
ecology—and then considers how these are modified by the rich
diversity of interactions between different species—community
ecology. The final chapter in this part deals briefly with the im-
portant topic of genetics in ecology.
These three chapters collectively address the issue of biological
diversity, including the question of the number of actors in the
ecological play: species diversity.
 C H E R

It
Population Ecology
Study of the Abundance and Dynamics
of Species Populations

14.1 Introduction predation is examined. The chapter closes with a dis-

cussion of plant population ecology. Because of this
A stand of beech trees, a herd of deer, a school of fish, plant focus in the final section, most of the examples in
a flock of seagulls: living organisms rarely exist for the earlier sections are taken from animal ecology.
long as isolated individuals. Most exist for most of
their lives as members of a population: a group of in-
teracting individuals of the same species. Certainly, in-
Spatial Arrangement of Organisms
dividual trees can be found growing by themselves and the Origin of Populations
many kilometers from any other trees, and individual The individuals of most plant, animal, and microbial
animals may be driven by the elements far from the species are not distributed across the landscape at ran-
other members of their population. However, such in- dom. They exhibit a definite spatial pattern, or
dividuals do not remain alone indefinitely. They may dispersion. Many types of organism in most types of en-
reproduce (if they are able to do so on their own) and vironment exhibit marked clumping of the individuals,
create a new population, return to the population with the clumps distributed either randomly or non-
from which they came, be joined by other individuals randomly. In contrast, organisms in some types of en-
of the same species, or die. Other individuals of the vironment are regularly spaced apart; their dispersion
same species are often as important a component of approaches a uniform distribution. It is common to
the environment of an individual as are other species find coniferous trees in an even-aged, single-species
and the factors of the physical environment. The ecol- stand fairly evenly spaced as the combined result of
ogy of individual species populations is, therefore, a past disturbance and subsequent competition for light,
major component of the overall science of ecology. whereas competition for moisture may lead to an even
Understanding the population ecology of the plants, spacing of individual cacti or pine trees in desert and
animals, and other organisms in the forest is of major dryland forest, respectively. The same thing can be
importance in the conservation and sustainable man- seen in animal populations. Many bird species divide
agement of forest ecosystems. the environment into uniformly sized territories so
In this chapter, we review how populations come that the birds may have a fairly uniform rather than a
to be and the advantages and disadvantages to an indi- random distribution. However, completely uniform
vidual organism of being a member of a population. distributions are very rare in nature.
We then examine how populations grow over time and Clumping of individuals into a population (they
what determines this growth. Some theories about are said to be contagiously distributed or to exhibit
population regulation are presented, and the role of contagion) can result from a variety of processes.

367
368 CHAPTER 14 — Population Ecology

Reproduction and Dispersal. Lack of medium- drip and radiation of heat to the snow from the stem.
and long-distance dispersal of reproductive propa- This accelerated snowmelt exposes the ground close
gules or offspring may result in the clumping of subse- to the stem several weeks earlier than in the sur-
quent generations as a population around the site of rounding meadow, lengthening the snow-free grow-
the parent organism(s). Longer-distance dispersal will ing season close to the tree. The earlier snowmelt
tend to produce less obvious clumping. Vegetative re- facilitates the growth of shrubs and other trees in the
production will lead to the establishment of new “snow crater” and leads to the formation of a tree
shoots in the vicinity of the reproducing plant and group (in closed forest) or tree island (in subalpine
produce a spreading mat or clump of that species. Be- meadows). In the meadow away from the tree island,
cause of wind and water movements, many seeds and the prolonged snow duration precludes the growth of
spores may come to rest in the same general area, and trees and large shrubs. In this example, the presence
this can lead to clustering of the individuals in the next of one individual modifies the environment in a way
generation. Similarly, wind-dispersed insects all may that allows other individuals of the same or other
land in the same locality and therefore have a nonran- species to survive. Similarly, the growth of mosses
dom distribution. Collection and storage of seed by adapted to acid conditions on sites with nonacidic
animals can result in clumped growth of plants. soils is made possible by the presence of trees with
acidic litter or decaying logs, and the growth of calci-
Variations in the Physical Environment. Dis- um-demanding herbs or mosses on calcium-poor
persal of seeds or spores can result in either a random soils is facilitated by the presence of hardwood tree
or a uniform distribution of plant propagules over a species that have significant levels of calcium in their
large area, but only those that land on a suitable sub- throughfall and litterfall.
strate will become established. For moisture-requiring
species, only those propagules that land on moist soil Aggregation as the Result of Biotic Factors.
will survive, giving rise to a clumping of those species Contagion in animal populations can occur because
on such sites. Seeds of light-demanding plants may the presence of other individuals modifies the biotic
germinate all over the forest floor, but only those lo- hazards of the environment. Species may be restricted
cated under gaps in the canopy will survive. Similarly, to, or clumped in, those environments or microenvi-
spatial variations in soil chemistry and microrelief can ronments in which they can resist diseases or escape
result in a markedly nonrandom distribution of plant the depredations of natural enemies. Heavy grazing
individuals and species. Many animals also have nar- pressure may eliminate forbs (herbs other than ferns
row habitat requirements and will be found aggregat- and grasses; Table 15-1) from grasslands, restricting
ed in areas where they have access to appropriate food, them to clumps of shrubs that protect them from her-
shelter, and environmental conditions. bivory, and animals in metapopulations may be
clumped as a result of predation on local subpopula-
Active Locomotion in Response to a Common tions. Animals often move in groups to reduce the risk
Stimulus. Members of a particular animal species from predation.
will tend to have the same or a similar response to en-
vironmental stimuli, which can cause them to aggre-
gate in the same general area. Deer will congregate in Advantages and Disadvantages of Being a
specific areas of winter habitat during heavy winter Member of a Population
snow or cold, and freshwater fish will school in deep Aggregation into populations confers both advantages
cool river pools or the deeper parts of lakes during hot and disadvantages on the member organisms (Allee,
days in summer. 1931, 1951). We discuss the advantages first.
Modification of the Physical Environment by the
Organisms. An interesting example of clumping is
Advantages
the characteristic tree islands found in subalpine Protection. Plants gain mutual protection from the
meadows near the tree line (Figure 14-1). The snow rigors of the physical environment, such as wind and
close to the stem of an isolated tree melts earlier than temperature extremes. Many mature forests just below
in the surrounding meadow, as a result of canopy timberline are often difficult to replace if the forest is
 Introduction 369

KX.
Pa)

ze
nat,
Recession of snow “S AN Recession of snow
1962 a 1961
May 25 open \} SAS May 15
May 29 >, NE
yin Position of snow-

Ashe
J Sy
pack on various
June 15 g f dates
June 22- -

1.5 12 030 O32 hy

Distance from center of tree island ag
(a)

Figure 14-1 (A) Diagram of a small subalpine tree island in a subalpine meadow in coastal
British Columbia, showing its associated snow crater and the springtime recession of the
snow in 1961 and 1962. The area within 30 cm of the central tree had almost 4 weeks more snow-
free period than the area 3 m away from the tree. (After Brooke et al., 1970). Around the tree island,
one characteristically finds concentric bands of vegetation that identify different snow-free periods:
white rhododendron, blue-leaved huckleberry, red heather, meadow spirea, and sedges, from the
longest to the shortest snow-free period, respectively. (Used with permission of the Department of
Botany of The University of British Columbia and the authors.) (B) Snow craters in closed forests can
sometimes be 2 m deep with vertical walls at a distance of 50 cm from the tree trunk. The
photo shows a snow crater in closed montane forest close to the forest limit in southwestern British
Columbia. It is common to find two or three trees growing close together in these craters.

removed because the combined effects of wind, full in- prevented. Undoubtedly, there are other determinants
sulation (high summer temperatures), and/or frost of the lowered tree line in the Grampians, but the loss
heaving can make reforestation by natural or artificial of shelter by other trees is probably a major cause of
means difficult. The present tree line in the Grampian regeneration failure.
Mountains of Scotland is thought to be approximately Many animals also depend on mutual shelter. Bees
200 m lower than the climatic potential for the area form tightly packed swarms inside hives in the winter
(Pears, 1968) because historic deforestation has al- to conserve heat, whereas in the summer, hive temper-
lowed wind velocity to increase near the ground to the atures are kept down by some of the bees fanning their
point at which seedling establishment and growth are wings at the entrance to the hive (Uvarov, 1931).
370 CHAPTER 14 Population Ecology

Larger animals will huddle together in winter for mu- Genetic Diversity. The aggregation of organ-
tual warmth and shelter from wind, and even animals isms promotes genetic variation and polymorphism in
that are aggressive and solitary in the summer may be- a population. High genetic diversity improves the
come gregarious during cold winter weather (e.g., the ability of a population to survive in a changing, unpre-
winter wren [McLachlin, 1983] and flying squirrels dictable, and variable environment. It facilitates the
[Layne and Raymond 1994]). Aggregation of seabirds development of a broad geographic range and helps
in densely populated breeding colonies, fish in the species to cope with interspecific competition. Ifa
schools, and large ungulates in herds achieve protec- population is too small, then inbreeding will lead to
tion from predators because the members of the group reduced genetic diversity and increased risk of popula-
are individually less susceptible to predation than are tion extinction.
individuals on their own. The reduction of genetic diversity in small popula-
tions may have led to the extinction in the United
Reproduction. Many species have a critical lower States of the passenger pigeon and the heath hen. The
population size (numbers) and density (dispersion)
latter was formerly abundant in Massachusetts and
that is necessary for successful reproduction and sur- probably also from Maine to Delaware (Gross, 1928).
vival. Animals of the open ocean, such as whales and By 1880, the species was restricted to Martha’s Vine-
deep-sea fish, obviously have a serious reproduction yard, an island off the east coast of Massachusetts, and
problem if they become separated from other individ- a large reservation was established to preserve the
uals of the same species; they cannot find a mate. species. Numbers increased to 2,000 in 1916, but a fire,
Some animals require a certain minimum aggregation a gale, and a hard winter combined with an influx of
4

of individuals before breeding will begin. A population avian predators (goshawks) decimated the population,
density ofless than one pair of muskrats per 1.6 km of leaving fewer than 50 breeding pairs. Numbers de-
stream or per 35 ha of marshland is not commensurate clined thereafter until in 1927 only 20 birds were left.
with successful breeding (Errington, 1963). Crowding The last bird seen was reported in 1932. Apparently,
in aphid populations triggers a switch from the pro- the natural disasters of 1916 reduced the population
duction of wingless individuals to the production of below some critical level, after which it was doomed.
winged females, which are then able to disperse to new
host plants. Intraspecific Competition. Although it can have
The stimulation of reproduction by crowding is adverse effects, intraspecific competition (between
also seen in birds that breed in colonies (Darling, members of the same species) often acts to keep a pop-
1938), and it has been suggested that a minimum of ulation healthy and well adapted to its physical envi-
10,000 birds at a mean density of three nests per ronment. Sick and poorly adapted individuals do not
square meter is necessary for the establishment of a survive the competition. By selection for high fitness
successful breeding colony of the guano-producing in the intraspecific sense, such competition may also
cormorant on the islands off Peru (Hutchinson, 1950). render the population more successful in interactions
Apparently, the stimulation provided by the noise and with other species.
movement of other birds is necessary to prepare them
for mating, nest building, incubation, and feeding the Division of Labor and Cooperation. Colonial
young. As a result of this requirement, large concen- and social behavior has proved to be a successful evo-
trations of birds promote synchrony of reproduction, lutionary development, as evidenced by ants, wasps,
which is thought to confer benefits to the individuals bees, and humans. All these animals have gained ad-
in the population by lowering the risk of predation. By vantages by the development of cooperation and of di-
synchronizing the production ofseeds, eggs, or young, vision of labor. Many individuals working together can
organisms can “swamp the market.” Predators are achieve more than that number working alone, be-
often intimidated by or cannot cope with large num- cause cooperation facilitates the development of spe-
bers of their prey. Alternatively, the predators simply cialization, with its attendant increased efficiency.
cannot “keep up,” and therefore most of the offspring Worker bees and ants, freed from the time and energy
survive (see the section Functional and Numerical demands of reproduction, can invest all their efforts in
Predator Responses). food gathering, nest building, and maintenance. Spe-
 Introduction 371

cialized soldier ants and termites lessen the need for

defensive adaptations in and activities by the workers,
with consequent energy savings. Cooperation among
predators greatly increases their power and efficiency.
A single wolf could never bring down a healthy, ma-
ture moose, whereas a large pack of determined
wolves might eventually prevail.

Disadvantages. Against the advantages of aggrega-

tion must be set a number of disadvantages.

Intraspecific Competition. Because natural selec-

tion operates to reduce competition between species,
individual organisms often suffer more adverse effects
Figure 14-2 Chronically overstocked lodgepole pine in
from competition with individuals of the same species central British Columbia. Large numbers of seeds were re-
than with individuals of some other species. Although leased from serotinous cones by a fire, and the failure of the
intraspecific competition can have the beneficial effect stand to undergo natural thinning has resulted in stagnation of
of weeding out those individuals that are sick or that growth. Numbers of trees as high as 500,000 per hectare are
have physiological, anatomical, or behavioral abnor- not uncommon.
malities that prevent them from competing success-
fully, generally the immediate effects of intense Increase in Levels of Stress. “Vhe physical prox-
competition within a population are adverse for most imity of other members of the population may in-
members of that population. As Darwin (1859) put it, crease the stress to which an individual is exposed.
“The struggle will almost invariably be most severe Stress produces morphological and_ physiological
between individuals of the same species, for they fre- changes (e.g., thinner plant cell walls and altered
quent the same districts, require the same food and are height/diameter ratios in plants in dense shade;
exposed to the same dangers.” changes in the production of animal hormones, e.g.,
Overpopulation can produce a variety of types of adrenalin) that may alter behavior and susceptibility to
competition. Irees in an overstocked forest compete pathogenic factors such as disease or environmental
for space to spread their branches; for light, soil mois- extremes. For example, it is generally believed that
ture, and soil nutrients; and for the rooting volume humans are much more likely to become sick after a
needed to provide stability. Overstocking in dry envi- period of stress than when they are unstressed. Stress-
ronments can result in the stagnation of the entire induced changes in hormone levels can induce marked
population because none of the individuals obtains behavioral changes, which may have adverse conse-
sufficient resources to be able to outcompete its quences for the individual and the population.
neighbors and causes natural thinning. ‘This is a com-
mon situation in Douglas-fir, lodgepole pine, and Alteration of the Environment. ‘The population
ponderosa pine stands in dry areas of western North density of arctic lemmings periodically increases to a
America (Figure 14-2). The effects of reduction in level at which they browse the vegetation so heavily that
competition after removal of large overstory domi- they virtually eliminate the cover needed for escape
nants or after the thinning of overstocked stands can from predators. Some animals foul their environment
be seen clearly in Figure 14-3. with their wastes or metabolic byproducts if the popula-
Overpopulation in animals can lead to severe com- tion density gets too high (e.g., human populations in
petition for food. If a few individuals exert dominance overcrowded cities). This can so degrade the environ-
over the others, then this competition may simply re- ment for that species that its numbers are reduced.
duce numbers to the point at which all remaining ani-
mals receive sufficient food. In the absence of such Disease Transmission. Parasites, predators, and
dominance, all of the population will suffer deficien- disease organisms are generally benefited by high den-
cies and all may die of starvation. sities of their host organisms. More prey and more
372 CHAPTER 14 = Population Ecology

Newre 14-3 Cross-section of a Douglas-fir tree showing the effects of moisture competi-
tion and release therefrom. The first 80 or more years of growth was as a suppressed sapling, com-
peting for moisture with a scattered overstory of large, mature trees. When these were removed,
thereby reducing competition for moisture, the trees grew rapidly. The stem disc was approximately
7 cm diameter at age 80 and 14 cm diameter 8 years after release. (Courtesy A. Vyse, British Columbia
Ministry of Forests)

frequent encounters between prey facilitate the spread 14.2. Population Growth and
and growth of disease and parasitic organisms. The
Demographic Characteristics
spread of insect pests and diseases through mixed-
species forests may be slower than through single-
of Populations
species stands.
Most of the research on population growth until fairly
Physical Interference. If animal populations be- recently was done by zoologists working with animals
come too dense, then encounters with other individu- and using the number of individuals as an index of
als become so frequent that normal patterns of population. However, the first significant paper in
behavior are interfered with. This can interrupt feed- population biology was not written by a zoologist but
ing and breeding activities. by a botanist in 1874, and Malthus had earlier (1798)
hese adverse aspects of intraspecific interactions all made it clear that the fundamental characteristics of
serve as a negative feedback on population growth. We populations are found equally in both plants and ani-
examine some of them in more detail later in the chapter. mals (Harper, 1977). The early focus on animals prob-
 Population Growth and Demographic Characteristics of Populations — 373

ably occurred because many animal species lend them- Before we can discuss the central question of pop-
selves to population census better than many plant ulation ecology (what regulates the abundance and
species. For many animal species, it may be relatively distribution of a species), we must consider how popu-
easy to monitor the change in number of adults in a lations develop over time from a single reproductive
population as it increases over time, whereas for unit (a male-female pair for sexual reproduction, or a
plants, it is not clear exactly what should be measured. single individual for asexual or self-fertilizing repro-
Does one count seeds, seedlings, or mature plants? duction).
For annual plants, a census may in fact be a reasonably
easy and realistic measure of the population. For
perennials, particularly for long-lived woody tree Geometric Growth
species, numbers may not be a good measure of the If a reproductive unit of a species with a short genera-
local abundance of a species. Biomass or basal area tion time is placed in an environment that provides all
may be better. requisites for life in excess of needs, then the abun-
Simple, free-living algae or free-floating aquatic dance of the organism will increase over time as indi-
plants are analogous to individual insects or mammals. cated in Figure 14-4. This pattern of population
They can be investigated experimentally in much the change is called geometric or exponential growth. The
same manner as the simple animals that were used by abundance of the population shows a curvilinear in-
researchers such as Gause (discussed later in this chap- crease with time as does the rate of change in abun-
ter) to develop the basic concepts of population ecol- dance from generation to generation.
ogy. Unfortunately, the analogy does not hold for If the organism in question has discrete, nonover-
large plants such as trees, which are more analogous to lapping generations (e.g., annual plants or an insect with
a termite colony or a bee’s nest. Functionally, a tree is a single generation per year in a seasonal environment),
a population of leaves or of buds connected by woody then the curve in Figure 144 can be described by
structures, with a high degree of central organization
and integration. Each leaf on the plant is in some ways Nea RoN;
like an individual organism in that it has a life history,
it competes with other leaves, and it is a member of a
population. A study of the change in number of com-
plex plants in an area over time is therefore more anal-
ogous to a study of the number of bees nests or ant
colonies in an area than to a study of the number of in-
dividual bees or ants. We have already seen that the
leaf area index of a particular ecosystem is a character-
istic of that system and that maximum leaf area is at-
tained rapidly and is more or less sustained for long
periods. The numbers of leaves or the leaf area index
grows in much the same way that a population of ani- Population
(N)
size
mals increases toward and fluctuates around its carry-
ing capacity. Because of these considerations, it is not
always possible to apply population theories developed
for animal populations to plant populations, and plant
population dynamics has developed as a branch of
ecology in its own right (Harper, 1977), and the gen-
0 10 20 30 40
erality of population ecology for plants, animals, and Time in generations
microbes has been stressed (Begon and Mortimer,
1981). Despite the differences between plant and ani- Figure 14-4 Geometric increase in abundance of organ-
mal populations, many of the population processes are isms that have discrete generations starting with 10 indi-
common to both life forms. We return later to a dis- viduals and various net reproductive rates (Ro). (dfter Krebs,
cussion of the characteristics of plant populations. For 1972. Copyright C.J. Krebs. Used with permission of author and
now, we focus primarily on animal populations. Harper & Row, Publishers.)
374 CHAPTER 14 Population Ecology

where N, is the population size at generation t, N,,; is when the quantity of food (and/or nutrients), space,
the population size in the subsequent generation, and and the numbers of other individuals of the same
Rj is the net reproductive rate, or the number of female species are kept at optimum values and other species
offspring produced per female per generation (assum- are excluded (Andrewartha and Birch, 1954). The
ing sexual reproduction and a constant ratio of males to term ,, thus is a parameter that is obtained under and
females). If Ry remains constant at a value greater than refers to ideal laboratory conditions. Its value with ref-
1, then the population increases. The higher the value erence to field populations is that it provides a stan-
of Ro, the more rapid the population growth. dard with which to compare the performance of field
If, however, the organism in question breeds con- populations and how close they are able to come to
tinuously throughout the growing season of the year obtaining their maximum theoretical rate of increase.
and if the population is made up of overlapping gener-
ations (e.g., humans or a population of aphids on a
rose bush with many generations during one growing Logistic Growth
season), then population growth must be described Geometric population increase is characteristic of the
using differential equations. Under these circum-
early phases of population growth or the growth of a
stances, the population curve in Figure 14-4 would be
population that has recently escaped from some tradi-
described (Lotka, 1922) as
tional population control (e.g., the human population
dN re as described in Chapter 1). Most populations exhibit
=> T.

dt relatively little change in numbers from year to year.

In 1944, it was estimated that there were about 125
Rate of change in __ natural rate number in
number per unit time of increase ‘ the population million ducks in the United States, which produced 10
to 16 eggs per pair. Had all the adults and offspring
The size of the population at any particular time survived, there would have been approximately 900
(N,) is then calculated from the population at some million ducks the following year. In reality, the num-
earlier time (Np) as bers remained approximately the same. Queen ter-
mites are said to be capable of laying tens of millions
N,
ee aT,
sce = N o&
rt
of eggs, and oysters reportedly can discharge 500 mil-
lion ripe eggs in one spawning. Even these astronomi-
where e = 2.718 (the base of natural logarithms), and t cal numbers are dwarfed by the reproductive excesses
is the time elapsed. of fungi, which can release billions of spores per night
The term 7, the natural rate ofincrease (or biotic po- from a relatively small area of substrate (references in
tential) of the species, is a parameter that incorporates Clarke, 1967), yet despite these enormous reproduc-
both additions to the population (births) and losses tive potentials, populations of such organisms grow at
from the population (deaths). It is the rate of popula- a geometric rate only for a very brief period, and ex-
tion growth per capita and expresses the result as cept after some major disturbance, they generally do
growth rate per individual. Birth rate on its own is not not show sustained growth over many generations.
enough to describe population growth, because The reason for this lack of sustained growth is that
growth will vary with different death rates even if the as a population grows in numbers, so does the negative
birth rate remains constant. feedback caused by competition, diseases, stress, and
Because additions to and losses from a population so on. This negative feedback is referred to as the
vary widely according to the size and condition of the environmental resistance (Chapman, 1928). As a popula-
population, the age and stage of growth of the repro- tion starts to increase from a single breeding unit, this
ducing individuals, and the nature of the environment, resistance is low and the population expands geomet-
r is variable for any species. To permit comparisons rically. So does the environmental resistance, which
between r values for different populations, ecologists steadily reduces the rate of population growth until
use the parameter 7;,,. This is called the innate capacity the resistance becomes so great that no further in-
for increase of the species. It is defined as the maximum crease is possible. Such population growth has an S or
rate of increase attained by a species at a particular sigmoidal shape and is called a Jogistic growth curve
combination of temperature, humidity, and so on, (Figure 14-5). The population grows slowly at first,
 Population Growth and Demographic Characteristics ofPopulations — 375
|
Carrying capacity (k) dt

Population size oy,

AN
ake
dN dt
__ Rate of change in
population size

Population
(N)
numbers
size:
biomass.(kg)
or

of
Rate
size:
population
the
in
change
Time

Figure 14-5 Idealized logistic growth curve. The graph shows the sigmoidal increase in num-
bers over time and the asymptotic approach to K, the carrying capacity. The rate of change in num-
bers (or biomass) is also shown.

followed by a period of rapid increase. The population continued to grow geometrically but since then has fi-
growth rate then slows down as numbers approach an nally begun to conform to the expectations of the lo-
upper limit (K, the carrying capacity) asymptotically. gistic growth model.
When the carrying capacity has been reached, the en- Numerous studies have compared actual popula-
vironment is “saturated” with that particular species. tion growth with the idealized logistic model. Growth
Expressing the effect of environmental resistance of yeast and bacteria cultures or simple organisms
mathematically, the simple geometric growth equa- such as protozoans raised under constant conditions
tion (the “logistic” equation) with a continual supply of food does approximate the
logistic curve. Laboratory populations of more com-
plex organisms, such as flour beetles, initially conform
to the logistic, increasing sigmoidally to an asymptote
(Gause, 1934), but if the experiment is prolonged,
becomes
then the population is found to fluctuate widely
DN aay
iar
around the asymptote and eventually to decline (Park
et al., 1964).
A search of the literature reveals that there are no
where [1 — (N/K)] is the proportion of the carrying ca- documented cases in which the population of an or-
pacity that remains unused. As N approaches K, [1 — ganism with a complex life cycle has maintained a
(N/K)] approaches 0 and rN [1 — (N/K)] becomes very steady population at the upper asymptote of the logis-
small. The population ceases growth. If N is greater tic curve for more than a brief period and that popula-
than K, then population “growth” becomes negative tion growth often fails to conform to the shape of the
and the population will decline. logistic curve (Krebs, 1978). This is probably because
the logistic growth curve is based on various assump-
tions that are frequently not met in either experimen-
Logistic Growth Equation and tal or natural conditions. We examine four aspects of
Actual Populations populations that may interfere with the attainment of
The logistic equation of population growth was pro- logistic growth: lack of response to K, time lags in re-
posed for the human population as early as 1838 by sponse to K, changes in K induced by high populations
Verhulst and again in 1920 by Pearl and Reed in a or other factors, and the age class structure of the pop-
study of the U.S. population. Until the mid 1970s, it ulation.
376 CHAPTER 14 Population Ecology

Lack of Response to K. Some species do not ex- quirements of the individual, so that the population
hibit a negative feedback between population density never exceeds K and never degrades the carrying ca-
and growth rate. Their population processes are not pacity of the resource (see the discussion of territorial
sensitive or are only weakly sensitive to increasing behavior in grouse later in the chapter).
population density; they are density-independent or For nonterritorial herbivores, K can easily be de-
they undercompensate for density increases. In a sur- pressed if populations get too high. Reindeer have
vey of 64 studies of population growth in which there been introduced into many parts of Alaska during the
was a significant directional trend in numbers with past century to replace dwindling caribou herds. ‘Two
time, evidence for a negative feedback was found in 47 of the Pribilof Islands in the Bering Sea received such
species, whereas no such evidence was found in 16 introductions in 1911. Four males with 21 females
species (Tanner, 1966). In only one species was there were released on St. Paul’s Island (106 km’), and 3
any evidence for a positive feedback, where the rate of males with 12 females were released on St. George’s
population growth was positively correlated with pop- Island (91 km’), where the populations were unmo-
ulation size (density). The species was Homo sapiens. lested by predators, including humans. The St.
This analysis suggested that 73% of the populations George’s population grew to a peak of 222 animals
studied exhibited the negative feedback necessary for early in the 1920s and then declined to a reasonably
population stability. However, there is some question stable herd of 40 to 60 animals. The St. Paul popula-
concerning the statistical validity of this conclusion, tion initially increased at a similar rate to the St.
and several population ecologists believe that there are George population, but instead of declining, it started
few cases in which there is strong evidence of an effec- growing geometrically in the mid-1920s, reaching a
tive negative feedback component in population peak level of approximately 2,000 animals in 1938.
growth. This high population overgrazed its food supply (slow-
growing lichens) and then declined rapidly to a level of
Time Lags. If a population is so responsive to in- only 8 animals in 1950 (Figure 14-6). The experience
traspecific interactions that there is an immediate neg- on St. Paul’s Island was reproduced when reindeer
ative feedback between an increase in population and
the population growth rate, then one might expect
population growth to be sigmoid. If, however, there is 2000
a delay in such a feedback, then one would expect pop-
ulations to overshoot the carrying capacity and then St. Paul Island
fluctuate around it: the greater the time lag, the 1500
greater the overshoot and the greater the population
fluctuations.
1000
Reduction in K by High Population Density. For
. No data
some species, the attainment of high populations is ac-
reindeer
of
Number
companied by a decrease in the carrying capacity and
Nni)ee)
an accompanying decline in the population. This is
not true for organisms such as trees, which respond to
St GooredsIsland
the attainment of K (in terms of volume or basal area) ~~ NL 260 teem,

by reducing population density in favor of increased 1910 1920 1930 1940 1950
size of the surviving individuals. When the carrying
capacity for volume or leaf area is attained, some indi- Figure 14-6 Reindeer introduced onto St. Paul Island in
the Bering Sea without any natural enemies increased to
viduals are able to continue to grow by outcompeting
levels of abundance at which they overgrazed their food
and killing other individuals. Some species of animal supply (lichens), thereby lowering K and almost driving the
are able to avoid overshooting and degrading their population to extinction. A similar population introduced to
carrying capacity by dividing the resource among the St. George Island declined before it had much effect on K. (After
individuals in the population. Territories are estab- Scheffer, 1951. Copyright American Association for the Advancement
lished, the size of which depends on the resource re- of Science. Used with permission of AAAS and the author.)
 Population Growth and Demographic Characteristics of Populations — 377

were introduced to St. Matthew’s Island (332 km?) in found in an expanding population. This obviously has
1944. An initial population of 5 males with 24 females important implications for social organization. Fewer
had increased to 1,350 by 1957 and reached a peak of schools will be required, more geriatric hospitals will
6,000, after which it crashed to 42 in 1966 after over- be needed, and there will be fewer people in the work
grazing (Klein, 1968). Similar events have occurred in force to support the enlarged older age classes. Fewer
New Zealand after the introduction of the Himalayan people will be creating the capital necessary to provide
thar (a hollow-horned ungulate animal resembling a the pensions of large numbers of older people and to
goat) and in several other parts of the world (Caughley, provide the medical and other social services that they
1970). Such population eruptions seem to be initiated will need. The human age distribution obviously has
by an improvement in food supply (an increase in important implications for the human population.
food-regulated carrying capacity) or introduction to an Similarly, the age distribution is an important deter-
unexploited food supply and terminated by overgraz- minant of the dynamics of other populations.
ing, which lowers the carrying capacity and prevents a A life table provides a summary of the schedule of
sigmoid pattern of population growth. deaths in a population and may list the causes of mor-
tality in each age group. A decline in numbers between
Age Structure of the Population. ‘The logistic two adjacent age classes generally means that deaths
growth equation is based on the assumption that a have occurred, although it can also indicate emigra-
population will eventually attain a stationary age distri- tion. The life-table technique of summarizing the na-
bution. By this we mean that the ratio of immature, ture and occurrence of mortality and the survivorship
prereproductive individuals to mature, reproductive (percentage survival from one age class to the next) of
individuals remains constant over time and that natali- a population was first developed by students of human
ty equals mortality. It is the distribution of ages populations (human demographers), with major contri-
in a population that has ceased to grow and is at equi- butions from life insurance companies, which are in-
librium with its total environment. If the population terested in knowing when people are likely to die.
fails to achieve or sustain this type of age structure, then An example of a life table’s age-specific summary of
the population will not follow the logistic growth curve. mortality is given in Tables 14-1 and 14-2. They show
The importance of age structure in the dynamics the numbers of survivors at the start of each age inter-
of populations requires that we examine it in some de- val (age class), the number dying in each age interval,
tail. Taking the human census data for Canada as an the rate of mortality (numbers dying per 1,000 individ-
example, one finds tables (called /ife tables) that list the uals), and the mean expectation of further life for indi-
numbers of people in each 4-year age group from viduals alive at the start of each interval. The shorter
birth to an age of 85 years or older. These life-table the time interval, the greater the detail of information
data can be displayed graphically in the form of an age- on when mortality occurs, which helps to identify the
distribution diagram. Figure 14-7 compares such dia- causes of mortality. Each of the columns of a life table
grams for Canada, the United States, and India for can be calculated from any of the other columns. The
various dates, and one can see that there has been a various columns merely represent different methods of
considerable change in the shape of the distribution. presenting the basic survivorship data.
Early in this century, while the Canadian and U.S. Perhaps the most commonly used information in a
populations were expanding rapidly, most of the popu- life table is the number of survivors at the start of each
lation was in the younger age classes, representing age interval. This can be presented graphically in the
both a greater rate of reproduction and a greater rate form of a survivorship curve, of which there are two cat-
of immigration of young people than occurred in the egories: the cohort, or dynamic, type and the stationary
1970s. Also shown is a graph of the expected distribu- age distribution, or static, type, each of which represents
tion of age classes for North America around the year a different type of life table. Cohort life tables and sur-
2000 and for when the population eventually stops vivorship curves are constructed by following the sur-
growing (the predicted stationary age distribution). vival of a particular group of individuals (a cohort,
Similar data are presented for India. It can be seen that e.g., all Douglas-fir seedlings on a 1-ha plot that ger-
there are far fewer young people and far more older minated in 1996 until they all are dead). Static or
people in a population that has ceased to grow than is instantaneous life tables and survivorship curves are
 1971
90* 1881
80
70
60 fof 2 fo} 2 [_] Post-reproductive

> +50
(_] Reproductive
% J 40
<=
30
Prereproductive
20
10
0
1% 0 6% 6% 0 6%
(a)

Ultimate stationary
1900 1970 2000 age distribution
90"
80
70
60 ) 2 é 2 3 4 3

50

Age,
yr =
30
20
10
0
6% 0) 6% 6% 0 4% 4% 0 4% 4%
(b)

1970 2000 2045

3 o fo} g

3}
ob
<x

8% 0) 8% 4% 0 4%
(c)

Figure 14-7 Historical and predicted patterns of change in the age distribution in the
human populations of Canada (A), the United States (B), and India (C). The horizontal scale is
the percentage of the population in each 5-year age class by sex. It seems that in the move toward a
stationary age distribution, India was approximately 100 years behind the two North American
countries in the 1970s but that it will be only approximately 75 years behind in the eventual attain-
ment of this demographic condition. See also Figure 1-1 for the projected age-class distribution of
the world population in 2030 according to various assumptions about fertility and mortality. (Data
for A from Statistics Canada, 1976. Used by permission. Data for B from C.F. Westoff, The population ofde-
veloped countries. Copyright © 1974 by Scientific American, Inc. All rights reserved. Data for C from T. Fre-
Jka, 1974. Used by permission.)

378
Table 14-1 Life Table for Red Deer Stags on the Isle of Rhum, Scotland

Mean Expectation
Number of Survivors Number of Deaths Mortality Rate of Further Life at
Age at the Beginning During Age per 1,000 Beginning of
Class (x) of Age Class x (I,) Class x (d,) Individuals (1,000 q,) Age Class (e,)

1 1,000 (1,000) 282 (84) 282.2 (84.0) 5.81 (4.70)

2 718 (916) if (19) 9.8 (20.7) 6.89 (4.15)
3 714 (897) 7 (0) 9.8 (0.0) 5.95 (3.25)
4 704 (897) 7 (150) 9.9 (167.2) 5.01 (2.23)
5 697 (747) 7 (321) 10.0 (430.0) 4.05 (1.58)
6 690 (426) i (218) 10.1 (512.0) 3.09 (1.39)
ii 684 (208) 182 (58) 266.0 (278.8) 2A (1.31)
8 502 (150) 253 (130) 504.0 (866.5) 1.70 (0.63)
fe) 249 (20) 157 (20) 630.6 (1,000.0) 1.91 (0.50)
10 92 (0) 14 (0) 152.1 (0.0) 3.31 (0.0)
11 78 14 179.4 2.81
WZ 64 14 218.7 2.31
13 50 14 279.9 1.82
14 36 14 388.9 1.33
Ws) 22 14 636.3 0.86
16 8 1,000.0 0.50
17 0 0) 0.0 0.0

The table was constructed on the basis of a census in 1957 (Lowe, 1969). Such a table is called a time-specific, or static life, table as opposed
to a dynamic, or cohort, life table in which the survival of an initial group or cohort of young is monitored until all individuals are dead. Figures in
brackets show cohort life table data for the first 9 years for the 1957 cohort.

Table 14-2 Life Table for a Hypothetical Striped Maple Population

Mean Expectation
Number of Survivors of Further Life at
at the Beginning Number of Deaths Mortality Rate per Beginning of Age Class
Age Class of Age Class x During Age Class x 1,000 Individuals Cause of
(x) (Ix) (dx) (1,000 qx) (ex) Death

1 10,000 8,750 875 4.1 43% eaten

16% winter kill

41% other

ZAG 1,250 low low 28.2 little mortality

16-40 1,250 1,205 38.4 (42 PARZ

41-100 45 43 15.6 33.6 Crown closure,

physical damage

+100 2 _ low _

Hibbs, 1979.
380 CHAPTER 14 Population Ecology

made by examining the age distribution of the popula- removal of predators will significantly affect the curve
tion at the time of a census (e.g., the age distribution for a deer population. The curve for light-demanding
of trees in a 1-ha Douglas-fir forest at the time of mea- tree seedlings will change if defoliating insects cause
surement). heavy mortality in the overstory, thereby reducing
Although a wide variety of survivorship curves can seedling mortality. As mortality and natality vary from
be found in nature, four major types have been identi- time to time, so does the shape of the age distribution.
fied (Figure 14-8). In type I, the population experi- However, a population in which mortality and natality
ences very little mortality until near the end of the rates remain constant for a considerable period has a
physiological life span; most dying organisms are old. regular and constant age distribution. In steadily in-
Mortality in type II populations is distributed evenly creasing and in steadily declining populations, the age
across all age classes (there is a constant number of distribution is said to be stable. A stable age distribution
deaths per unit time), whereas in type II, a constant is one in which the shape of the survivorship does not
percentage of the survivors die in each time interval. change but the total number of individuals in the pop-
In type IV, heavy juvenile mortality is followed by low ulation may be changing. Where mortality exactly
and fairly constant mortality for the rest of the life equals natality, the population will remain constant
span. These four curves are hypothetical, and few and will eventually achieve a stationary age distribution.
populations conform to them exactly. Homo sapiens in This is equivalent to the survivorship curve for a pop-
developed countries conform fairly closely to type I, ulation that is neither growing nor decreasing and that
except for a brief period of infant mortality in the first is at equilibrium with its environment.
few days of life. Many bird species conform approxi- Populations that have a stable age distribution but
mately to type II, whereas type IV is characteristic of that have higher numbers in the younger age classes
many fish and marine invertebrates. Plants may con- than the stationary age distribution will grow geomet-
form to any one of these types. Many species have rically. If the age distribution changes with time to-
heavy juvenile mortality followed by constant adult ward the stationary form, then a sigmoidal growth
mortality intermediate between types II and II. Oth- curve will result. Thus, one cannot interpret the prob-
ers are intermediate between types I and II. able future growth of a population from the age distri-
The survivorship curve is not a fixed characteristic bution alone. One also needs to know the
of a population. It changes if the action of the agents characteristic or stationary survivorship curve for the
of mortality changes. Introduction of hunting or the species.
From this discussion, it can be seen that in many
cases the logistic growth equation is too simplistic to
Type I represent accurately what happens in nature. Never-
theless, some populations do conform to this type of
growth, and the idea of environmental resistance con-
straining the biotic potential of a species is useful.

14.3. Major Determinants of

Population Size
~
es The number of organisms in a population results
original
of
%
surviving
cohort from the combined action of four major population
parameters: natality, mortality, immigration, and emi-
gration. Many factors influence these four parame-
ters, and much time and energy have been spent
Time or Physiological Life Span
trying to decide on their relative importance in regu-
lating population size. We consider each in turn and
Figure 14-8 The four major types of survivorship curve. then discuss the various theories concerning popula-
(After Slobodkin, 1961. Used with permission ofL.B. Slobodkin.) tion regulation.
 Major Determinants ofPopulation Size 381

Natality they first reproduce (Table 14-3), as do animals

Natality refers to the process of producing new indi- (Table 14-4). The longer reproduction is delayed,
viduals, whether this is by sexual (birth, hatching, or the slower the rate of population growth, and delay-
germination) or by asexual reproduction. It is a rate ing marriage and family formation is often proposed
that can refer to the actual number of new offspring as an alternative to reduced family size as a way of
regulating the rate of growth of human populations.
produced per female or per reproductive individual
For example, compare the contribution of a couple
per unit time (the fertility of the female or individ-
that produces a family of four children in their late
ual) or the maximum potential rate of which the fe-
teens and early 20s with couples that produce four
males or reproductive individuals of a particular
children when they are in their late 20s and early 30s
species are capable (the fecundity of the species). The
or their late 30s and early 40s. If their offspring be-
fecundity of Homo sapiens is one birth per 9 to 11
haved reproductively the same way as their parents
months per female during the reproductive years.
and had a similar life span of approximately 80 years
Two offspring per female is becoming common in
and if half the children were female, then the number
developed countries, and some women choose to
of offspring would be approximately 238, 44, and 28,
have no children, thereby offsetting families of more
respectively, after 120 years.
than 2 children. Gradually, human fertility is ap-
The number of offspring per reproduction
proaching replacement fertility: the level of fertility at
(brood, clutch, or litter size when referring to animals)
which each breeding unit (a couple in the case of hu-
varies greatly between different types of plants and
mans) on average produces only enough offspring to
animals. Brood size is under genetic control and is
replace themselves.
believed to represent the optimum allocation of re-
Measurement of natality is complicated by the vari-
sources for the particular species in its particular en-
ety of breeding patterns found in organisms. Some
vironment. For example, tropical birds tend to lay
breed once a year, some several times a year, and some
smaller clutches of eggs than do temperate and
continuously. Sometimes the fertility rate will be re-
northern birds (Cody, 1966; Skutch, 1967). Appar-
ported on an annual basis, sometimes on a generation
ently, it is better to lay fewer larger eggs in the trop-
basis, and sometimes on the basis of some arbitrary age ics and a larger number of smaller eggs at the higher
class. latitudes. At any one latitude, clutch size varies with
Natality varies during the life of an organism. species; gulls normally lay three eggs, whereas the
There is always a prereproductive period, and some Canada goose lays four to six eggs. Brood size is also
organisms survive to a postreproductive stage. Trees
influenced by the availability of food. It has been re-
from temperate latitudes vary in the age at which ported that the great tit (Parus major) in Holland has
a smaller average clutch size, and fewer pairs raise
second broods in years when the birds are abundant
Immigration Natality than when they scarce. This was tested experimen-
tally by tripling the population by providing nesting
boxes, thereby increasing competition for food. This
reduced average clutch size by two eggs (from ap-
proximately 10) and the proportion of pairs raising
Population second broods from 63% to 16% (Kluij-ver, 1951).
increasers
Klomp (1970) reviewed the question of clutch-size
determination in detail.
[mre | Mammals also show a considerable degree of envi-
ronmental control on brood size. Arctic foxes produce
Population larger litters in years when lemmings are abundant
decreasers than when lemmings are scarce. Lions in South Africa
were reported to have litters of four to five cubs when
game was plentiful compared with two to three cubs
per litter when game was scarce (references in Lack,
Emigration Mortality 1954). In a study of North American elk, 20 to 25% of
382 CHAPTER 14 Population Ecology

Table 14-3 Age at Which Forest Trees Begin to Reproduce Abundantly

Early,* 10-20 yr Intermediate, 20-40 yr Late, 40-60 yr
a

Conifers Jack, pitch, lodgepole, Red, eastern Sugar, western white,

Pines knobcone, Virginian, sand, white, shortleaf ponderosa, limber,
Monterey, bishop, whitebark
slash, loblolly
Other Tamarack, black spruce, Red and white Spruce, true firs
northern white cedar, spruce, balsam
Port Orford cedar, fir, Douglas-fir
southern white cedar,
cypress
Hardwoods Willow, cottonwood, aspen, alder, Hickories, maple, Beech, oak
gray birch, paper birch, pin basswood, ash,
cherry, red maple, bigleaf maple, elm, sycamore,
box elder, scrub oak, and other chestnut, buckeye
fast-growing short-lived trees that
produce small seeds

Source: After Daniel et al., 1979. Copyright McGraw-Hill Book Co., NY. Used with permission of publisher and authors.
@Many of these species may produce appreciable quantities of seed after as few as 5 years.

the hinds bore twins in an area with ample food, for the species is obtained by summing (/,7,) for all
whereas almost no twins were produced on a range age classes.
that had been heavily overgrazed. The same thing has The number of offspring produced by an organ-
been reported for deer. ism is believed to represent a balance between the
Fecundity estimates are often included in life ta- energy costs of producing more offspring and the
bles. Table 14-5 presents a fecundity schedule for benefit of the increased contribution of those off-
the hinds in the red deer population on the Isle of spring to the genetic composition of the next genera-
Rhum, referred to in Table 14-1. For each age class, tion. If a20% increase in offspring reared means that
it shows the sex ratio, the number of female off- each offspring receives 20% less food, then fewer off-
spring per female (7,), and the fertility of the age spring may survive and contribute their genes to the
class, obtained by multiplying m, by the number of next generation. In this case, there would be strong
hinds in the age class (/,). The net production rate selective pressure to reduce the number of offspring
to the point at which each receives adequate food.
Conversely, if a 20% rise in the number of offspring
increased the contribution of the genotype to the
next generation, then selection would favor more off-
Table 14-4 Age at Which Animals Start to Reproduce spring.
From the discussion above, it is obvious that a
Age of First
species has a “choice”! to make. It can allocate its
Type of Animal Reproduction
available energy and nutrient resources to a large
Planktonic organisms Few days number of small offspring, as is done by many tree
European voles Few weeks species and aquatic invertebrates, or it can produce
Tropical sparrow 6-8 months
Nontropical songbirds 1-3 years
Beaver, wolf, lion, and whale 2 years
Deer, bison, bear 3 or more years ‘The use of these and other anthropomorphisms is justified by their utility
Elephant 8-16 years in the learning process. The reader is reminded that there is no such thing
Rhinoceros 20 years as a strategy (implying advanced planning) in evolution, and other species
do not make evolutionary choices in the sense that humans do. “Choices”
Kendeigh, 1961. and “strategies” as used here refer to evolutionary trade-offs.
 Major Determinants ofPopulation Size 383

Table 14-5 Schedule of Fecundity for Female Red Deer in the Isle of Rhum Population (Lowe, 1969)

Percentage Female Offspring Net Production

Age Class, yr of Offspring Per Female Rate
(x) That Are Female (m,) (Ro or 1.m,)
1 es = =
o) = ss =
3 50 0.311 0.242
4 50 0.278 0.193
5 50 0.302 0.184
6 61.9 0.400 0.210
i 61.9 0.476 0.210
8 61.9 0.358 0.128
9 61.9 0.447 0.081
10 61.9 0.289 0.017
11 50.0 0.283 0.014
12 50.0 0.285 0.012
13 50.0 0.283 0.010
14 50.0 0.282 0.007
lis 50.0 0.285 0.005
16 50.0 0.284 0.003

fewer, larger offspring that are better equipped for ber of individuals and the same mortality rate, can
survival. Where the environment is very patchy and have totally different temporal changes in abundance
unpredictable, where the risk of mortality to any one according to when the mortality occurs: before or
individual is high, or where there is very little compe- after the reproductive period.
tition from other organisms, the optimum evolution- Mortality can occur as the result of many different
ary “strategy” is to produce large numbers of small processes, including disease, predation, parasitism, an-
offspring. Such species are called strategists and have tibiosis, physical conflict, starvation, malnutrition,
been produced by 7-selection. Species in more pre- temperature effects, and dehydration.
dictable and uniform environments with high levels of
inter- and intraspecific competition are selected to
produce fewer but larger offspring with greater com-
Immigration and Emigration
petitive abilities. Such species are called K-strategists. For some species, the arrival of individuals from or
Some characteristic features of r- and k-strategists are their departure to remote areas constitutes the major
listed in Table 14-6. factor regulating the actual number of organisms in a
locality. A well known example in which dispersal is
the major determinant of animal numbers in an area
Mortality is immigration of African locusts johnson, 1963).
The topic of mortality was introduced in the discus- Locusts appear periodically in plague proportions
sion of life tables and survivorship curves, and it was through much of Africa, destroying the vegetation
noted that two populations with the same total abun- wherever they settle (Albrecht, 1967). The desert lo-
dance can have different growth patterns because of cust was at plague levels for 37 of the 56 years from
differences in age structure and, hence, mortality. One 1908 to 1964 and can cover areas as large as 44 mil-
can be expanding, if most of the individuals are young. lion km? (Waloff, 1966). Individual swarms can cover
The other can be on the point of collapse because areas as large as 1,000 km’, contain 1.6 x 10!” locusts,
most of the individuals are old and it has no juveniles. and weigh approximately 50,000 Mg (tonnes) (Gunn,
Similarly, two populations, both with the same num- 1960).
384 CHAPTER 14 Population Ecology

Tuble 14-6 Some Characteristic Features* of 7- and K-Strategists and Their Environments
Feature r-Strategists K-Strategists
a eee

Climate Variable and/or unpredictable Fairly constant and predictable

Mortality Often catastrophic, density Density dependent

independent

Survivorship curve Often type III° (Figure 14-11) Usually types | and II (Figure 14-11)

Population size Variable, nonequilibrium: Fairly constant, equilibrium

usually well below carrying value: at or near carrying capacity
capacity; unsaturated com- of the environment; saturated
munities; recolonization communities; annual recolonization
often occurs annually not usually necessary

Intensity of competition Variable and often low Usually high

Life span Short, usually less than Usually more than 1 year and often
1 year many years
Selection favors Rapid growth Slower growth
High r,, Greater competitive ability
Early reproduction Delayed reproduction
Small body size Larger body size, often very large
Single reproduction/life cycle Repeated reproductions/life cycle

Source: After Pianka, 1970. Copyright The University of Chicago Press. Used with permission.
*Obvious exceptions will be found. For example, some trees are r-strategists (e.g., hemlock, western redcedar, eucalypts, alder) and some are
K-strategists (e.g., true firs, oaks), yet they may be similar in size and longevity.
>Deevey, 1967.

Before the work of a Russian entomologist tion to the one in which the insects are flying (John-
(Uvarov, 1961), who developed the phase theory of lo- son, 1963; Rainey, 1963).
custs, the sudden, almost magical appearance of vast In the case of the African locust, immigration is an
swarms of locusts and their equally rapid disappear- important component of the initial rise in population
ance had baffled researchers. Uvarov showed that lo- density that is needed to trigger the switch from soli-
custs have two phases, which differ in behavior, color, tary to gregarious phase, and emigration is important
physiology, and morphology. In the solitary phase, lo- in moving the swarms from area to area. The size of
custs behave much like any other grasshopper; a local the locust population in many areas therefore is more
nuisance but never aggregating into swarms of plague closely related to immigration and emigration than to
proportions. Rainfall in areas inhabited by solitary lo- any other population parameter. Wind dispersal of the
custs is followed by plant growth, improved locust nu- spruce budworm is similarly an important determi-
trition, and an increase in population density; wind nant of budworm population dynamics in eastern
movements can increase the density of the locust pop- Canada (Ludwig et al., 1978).
ulations even more by concentrating the insects in a Many animals produce offspring in excess of the
particular locality. When a sufficient concentration carrying capacity of the environment. These excess in-
has developed, the behavior of the insects changes and dividuals often disperse to new areas and thereby avoid
they actively aggregate. The resulting increase in fre- populations in excess of the carrying capacity in the
quency ofinteractions between individuals leads to the home area. Human history is replete with examples of
formation of the gregarious phase, in which swarm for- Europeans and Scandinavians leaving overcrowded
mation and emigration occur (Key, 1950). Although home countries to seek new territories, and in many
active flight is undertaken, the subsequent distribution other overcrowded countries around the world, emi-
of the swarms to new areas is largely by wind, the lo- gration has historically been an important mechanism
cust swarms often being blown in the opposite direc- to ease population pressures. The resulting immigra-
 Theories About the Natural Regulation ofPopulation Size 385

tion into recipient countries has often had a greater nomic”) entomologists interested in the control of in-
short-term effect on population increase in those sect pests.
countries than has natality in the existing population. There are two major properties of populations that
In a study of the regulation of field mice, Krebs need to be explained: variation in mean abundance be-
and Myers (1974) established that dispersal was a crit- tween different environments and temporal variation
ical component of population regulation in these small in the numbers of individuals about the mean in any
mammals. By fencing an area and preventing the dis- one environment. Some of the theories described next
persal that normally occurs during a period of popula- refer mainly to mean abundance, whereas others refer
tion growth, mouse numbers increased to three times mainly to temporal fluctuations.
the density in an unfenced control area. This led to The various theories about population regulation
overgrazing and starvation, which reduced the popula- in animals that have been developed over the past 70
tion to approximately half that in the unfenced area; years can be grouped into several schools of thought:
56% of the males and 69% of the females produced the biotic (or density-dependent) school, the abiotic (or
during a population increase were lost from the un- density-independent) school (these two schools can be
fenced control area by dispersal. combined as the extrinsic regulation school), the se/f-
Dispersal is also important in territorial animals. regulation (or intrinsic regulation) school, and the
Once all the choice territories are occupied, remaining comprehensive school. ‘The application of these theories
animals disperse to marginal habitats, thus preventing to plants is discussed later.
the population from reaching unsustainably high den-
sities. More is said about this later.
Biotic, or Density-Dependent, School
The modern debate over the regulation of animal
populations began with the concept of the balance of
14.4 Theories About the Natural nature and the assertion that this could only be the re-
Regulation of Population Size sult of density-dependent or facultative mortality factors
(Howard and Fiske, 1911; Nicholson, 1993; Smith,
The basic elements of population growth have been 1935). For these factors, the effect intensifies as popu-
known for a long time. Greek philosophers such as lation increases, and relaxes as population declines.
Plato understood that population growth could be Only by having such a density-dependent feedback
stimulated by compulsory and early marriage and by mechanism could populations avoid increasing to lev-
large family size and that overpopulation could be els at which they destroyed their habitat or supply of
avoided by infanticide, abortion, celibacy, and emigra- resources, or decreasing to extinction. There was dis-
tion. However, the first clear statement that popula- agreement over which was the most perfect density-
tions do not grow indefinitely but reach some upper dependent mechanism. Parasites and predators were
level awaited the writings of an Italian by the name of nominated as the key density-dependent mechanism
Botero in 1588. He noted that despite an undimin- until it was pointed out that time lags in the response
ished power of increase, population growth is halted of both predator/parasite and prey populations can re-
by such events as famine, disease, wars, earthquake, sult in oscillations of increasing amplitude rather than
and floods and that the most fundamental limitation balance; competition for food or other resources was
was lack of food. Two centuries later, Malthus (1798) recognized as a more perfectly density-dependent
enunciated his famous principles of population mechanism (Lack, 1954; Milne, 1957, 1962).
growth: that a population has the potential to increase The scientists who supported these ideas recog-
geometrically but is restrained by the finite carrying nized that whereas mortality factors such as climate
capacity of its environment. He noted that increased periodically cause major mortality, these density-
population density was accompanied by increased independent factors can never produce the moderate
mortality from disease and increased violence. Many fluctuations of population abundance around a fairly
others both before and after Malthus have contributed stable long-term mean value that was said to represent
to a growing knowledge of population regulation, but the balance of nature. Much of this early debate was
the foundations of the modern theories about animal led by entomologists who were trying to understand
populations were developed largely by applied (“eco- what controlled the abundance ofinsects.
386 CHAPTER 14 — Population Ecology

Abiotic, or Density-Independent, School changes in population size result in changes from se-
lection for aggression and competitive ability to selec-
In contrast to the density-dependent view of popula- tion for survival. The former, which occurs as
tion regulation, another group of entomologists population density reaches high levels, reduces repro-
claimed that the distribution and abundance of insects ductive success and survival, whereas the latter, which
were much more closely related to climatic factors occurs when the population has declined to lower lev-
that operate in a density-independent manner. els, has the opposite effect, thereby generating the
They believed that many insects killed by parasites or cycle. Because of this density-related change in the ge-
predators would have been killed eventually by climat- netic quality of the population, abiotic factors such as
ic factors anyway (a phenomenon referred to as weather may operate in a density-dependent manner.
compensatory mortality) (Bodenheimer, 1928; Uvarov, Other examples of intrinsic mechanisms include
1931). The importance of abiotic factors was sup- behavior and social organization (Wyne-Edwards,
ported by a second group of insect ecologists who re- 1965). Although food is ultimately the most critical
jected the strict definition of mortality factors into limiting resource for most animals, many limit their
density-dependent or density-independent categories populations through social structure or by establishing
(Andrewartha and Birch, 1954; Andrewartha and
territories. In the former, the socially dominant ani-
Browning, 1961; Browning, 1962). They suggested
mals always get access to adequate resources. In times
that because mortality affects individual organisms of ample resources, subdominant animals also get the
rather than populations, investigations of animal dis- resources that they need, but in times of resource
tribution and abundance should focus on the individ-
scarcity, they are resource-limited and do not survive
ual. They identified four factors that influence the
or do not reproduce. ‘Territoriality works in a similar
survival and reproduction of individuals: weather,
manner. Dominant individuals set up breeding territo-
food, other organisms, and shelter (a place in which to
ries, the size of which ensures that they can get suffi-
live). They concluded that of these four, climate was
cient food to raise their young. Individuals who fail to
the most important as it affects food, other organisms,
establish territories in good habitat disperse to mar-
and the availability of shelter, in addition to acting as a
ginal habitat, where most fail to survive. In a study of
direct mortality factor. However, as these ecologists
the cyclical variation in numbers of red grouse on the
recognized the importance of biotic factors, their the-
Scottish moors, it was found that there was an inverse
ory is not exclusively abiotic.
relationship between territory size and population
density: territories were largest when populations
were lowest. The size of the territory was reported to
Self-Regulation, or Intrinsic, School
be related to the aggressiveness of the male defending
As the debate between the supporters of the two ex- the territory, which was thought to be related to how
trinsic theories of population developed, another well nourished that male was as it was growing. The
group of ecologists looked to mechanisms that operate fluctuation of grouse numbers was interpreted to be
within the population itselfas the explanation for pop- due to changes in territory size as a result of fluctua-
ulation regulation. The earliest of these considered tions in the quantity of nutritious food available to the
that change in the genetic constitution of apopulation grouse (Jenkins et al., 1963; Miller et al., 1966; Moss,
occurs over time and that this determines the response 1972). Another example of how behavior and the
of individuals in the population to the external mortal- “quality” of the individuals in the population can be
ity agents (Ford, 1931, 1975). Change in the propor- affected by nutrition and how this can drive popula-
tion of different genotypes in the population occurs as tion fluctuations is given by Wellington (1965) in his
the selective pressures of either abiotic or density- study of the cyclical outbreaks and declines of the
related factors external to the population change. western tent caterpillar.
These genetic shifts are associated with changing re-
productive success and mortality and, therefore, in
population increases and declines. One of the best Comprehensive, or Compromise, School
known advocates of the genetic theory of regulation Although several researchers have proposed that pop-
was Chitty (1960, 1971). Working on the causes of ulations are regulated by a single factor, a much larger
population cycles in small mammals, he suggested that number of population ecologists have subscribed to
 Theories About the Natural Regulation of Population Size 387

the idea that the abundance and distribution of most high. Governing mechanisms are defined as “the ac-
populations are determined by a combination of ex- tions of repressive environmental factors, collectively
trinsic and intrinsic, biotic and abiotic, and density- or singly, which intensify as the population density in-
dependent and density-independent mechanisms. The creases and relax as the density falls.”
particular combination of regulating factors will vary Conditioning factors tend to be more important
from case to case. Homeothermic animals may be reg- than governing factors in determining animal abun-
ulated differently than poikilotherms. Organisms that dance in physically unfavorable or extreme environ-
live in physically extreme, widely varying, or unpre- ments, and the converse is true for favorable or
dictable environments may be regulated differently moderate environments (Figure 14-9). This is be-
than organisms that live in physically moderate, rela- cause in environments in which the conditioning
tively constant, and highly predictable environments. mechanisms set a low population potential, the popu-
The mechanisms of regulation may vary at different lation generally does not attain densities at which
stages of the life cycle and from year to year. The reg- density-dependent factors become effective. Govern-
ulation of populations is seen to be the result of a com- ing mechanisms are therefore relatively unimportant
plex of mechanisms and factors. contributors to the setting of population levels, and
Huffaker and Messenger (1964) proposed the use the population responds directly to variations in the
of the terms conditioning mechanisms and governing low carrying capacity. In environments in which con-
mechanisms as a replacement for the controversial ditioning factors set a high population potential,
terms density-independent and density-dependent. Condi- populations increase to the point at which the density-
tioning mechanisms are defined as “environmental dependent governing factors act strongly on the
factors or agents that, uninfluenced by density, con- population, and these factors are preeminent in deter-
tribute to the setting or fixing of a framework of po- mining population numbers (Figure 14—-9). Where the
tential environmental carrying capacity or affect population potential set by conditioning factors is sub-
interim population regulation when capacity is not at- stantially above the level set by governing factors, the
tained.” Conditioning factors determine the general former will seem to play little or no role in population
levels that a population may attain in an environment: regulation. The relative importance of governing and
whether the population is generally very low or very conditioning factors for a given species will vary in

Importance of conditioning mechanisms

6ise}
S>
I

High
2o 4i=
ig (2
Ss
o
A
>
[=|
o
=!
° Governing
a mechanism
Py oO
Spe Mas
i}
(e)
es : Correlation
density
with
change
i Low High Favorable Unfavorable

Importance of governing mechanisms Type of environment

(a) (b)

Figure 14-9 Variation in the importance of density-dependent and density-independent

mechanisms of population regulation in different types of environment. (A) Importance of
conditioning (i.e., density-independent) and governing (..e., density-dependent) mechanisms in dif-
ferent environments. By scanning across the graph at any point on the vertical scale, you can see the
relative importance of the two mechanisms for that particular type of environment. (B) Correlation
between population density change and conditioning and governing mechanisms. (Modified after
Huffaker and Messenger, 1964. Copyright Chapman and Hall Ltd., London. Used with permission of the
publisher and the authors.)
388 CHAPTER 14 — Population Ecology

different parts of its range and at different times as the overlook that in reality some populations may be reg-
species range expands and contracts. Near the center ulated by a complex set of factors. It may be necessary
of the range, governing factors will dominate, whereas to have complex theories to account for multiple de-
conditioning factors will play the major role at the pe- termined phenomena.
riphery of the range. Although the scientific method may work ad-
The comprehensive approach to population regu- mirably in mature sciences such as physics and chem-
lation suffers from one major weakness as far as tradi- istry, it may be inappropriate for the juvenile stages of
tional scientists are concerned: it is very difficult to a young science such as ecology. Many of the branches
prove a comprehensive theory wrong. Science ad- of ecology are still in the early descriptive stages in
vances by a process of logical reasoning. A scientist ex- which the deductive experimental scientific method
amines the known facts about a phenomenon and by may not always be useful. It must also be realized that
inductive reasoning formulates an initial theory about it: although the scientific method may be well suited to
a statement of what he or she thinks are the reasons examining subcomponents of ecosystems, this does
for the known facts. This theory is then subject to ex- not necessarily qualify the method as the ideal or only
perimental testing in an attempt to prove it wrong. This tool for advancing knowledge about entire ecosys-
is done by making predictions (hypotheses) from the tems, at least not with our present incomplete knowl-
theory (by deductive reasoning’), which are then tested edge of these ecosystems. A better approach may be to
in carefully designed experiments. The experimental design methods for the scientific evaluation of ecosys-
observations (results) are then used in conjunction tem-level theories that can deal with greater com-
with inductive reasoning to evaluate the initial theory. plexity. Such methods are to be found in multivariate
If proved wrong, then the theory can be reformulated statistics and computer simulation (see Chapter 21).
to account for the experimental evidence or replaced Comprehensive theories vary greatly in the num-
by a more plausible theory and then retested. If, after ber and type of factors that are included. This varia-
repeated attempts to disprove it, the theory still fits tion reflects the experience of the theorist, the species
the facts, then it may be accepted as a reasonable ap- involved, and the type of environment. Climatic ef-
proximation of reality. If the theory remains after fects may be invoked directly and/or indirectly by
many years of testing, then it may be elevated to the their action through the soil, plants, or other animals.
status of a principle or law. This process, the scientific For herbivores, the effects of parasites, predators, dis-
method,’ requires that a theory generate testable, eases, competition, and food availability and quality
refutable hypotheses. The problem with comprehen- may be included. For parasites and predators, the role
sive theories of population regulation is that they of climate, disease, and the availability and nutritional
rarely generate easily refutable hypotheses; because of value of the host may be important. A good example of
their inherent complexity, they tend to generate com- how population theories are increasing in complexity
plex hypotheses that are not easily tested in simple ex- is provided by Price et al. (1980), who argued that re-
periments. This shortcoming has been used by some alistic progress in understanding the population dy-
to reject the comprehensive approach (e.g., Chitty, namics of insect herbivores will not be made unless
1967) in favor of theories based predominantly on the we consider the interactions among three trophic
action of a single factor. levels—plants (and implicitly between plants and soil),
There can be little disagreement with the desire to herbivores and carnivores—and incorporate into pop-
advance our understanding of natural phenomena or ulation theories the interactions between plants and
with the claim that science is advanced by rigorous herbivores, and between herbivores and their parasites
testing of inductively derived theories. However, to and predators. They note that it may not be possible
force our ideas about the regulation of a natural popu- to understand herbivore-parasite relationships unless
lation into the scientific straightjacket of a simple, we first understand herbivore—plant interactions. This
testable hypothesis for the convenience of science may includes the important question of herbivore nutrition
(e.g., Mattson, 1980). The conclusion that population
regulation theory should be framed at the ecosystem
‘Inductive reasoning: from the specific to the general. Deductive reasoning: level rather than just at the population level is sup-
from the general to the specific. ported by Krebs et al. (1995) and the ideas presented
‘The scientific method is discussed again in Chapter 17. in Chapter 3.
 Cyclical Population Fluctuations 389

14.5 Cyclical Population Fluctuations tality that initiates the cyclical decline, combined
with predation, which sustains the decline in hare
An example of the need for a comprehensive approach numbers to low levels (Keith et al., 1984). This is
to explaining animal population dynamics is the his- referred to as the Keith hypothesis, or the winter
tory of attempts to explain the fascinating phenome- food-predation hypothesis.
non of cyclic population fluctuations that characterize 2. The cycle is driven by changes in the chemistry of
many species of herbivores that live at high latitudes. the hares’ food plants in response to herbivory. This
Population cycles are fluctuations of relatively constant is referred to as the Bryant hypothesis (Bryant,
period (significantly different from random) around a 1981, 1987), or the plant chemistry hypothesis.
long-term mean population density. Cyclic population
3. Predation acting alone drives the cycle. The combi-
fluctuations are seen in various parts of the world in dif-
nation of specialist and generalist predators in com-
ferent kinds of animal populations, including mammals,
bination with their reproductive time lags generate
birds, and insects. They have always fascinated humans
the hare cycle; starvation reduces the predator abun-
because in a variable and unpredictable world, pre-
dance after the hare decline (Hanskii et al., 1991).
dictable, regular change in animal abundance is unex-
pected. Perhaps the best known example of cyclic 4. The polymorphic behavior hypothesis (Chitty,
population fluctuations is that seen in Canada, where 1967), which suggests that changes in behavior as-
many different species of wildlife, including snowshoe sociated with density-dependent changes in the
hare, lynx, muskrat, and various species of tetraonid genetic make-up of the population is adequate to
birds (e.g., grouse), exhibit an approximate 10-year cycle explain the cycle. This is the Chitty hypothesis.
of abundance with a marked degree of synchrony be-
tween species and between widely separated areas Of these theories, predation, acting either alone or
(Keith, 1963). Shorter cycles are also well known among in concert with other factors, is currently seen to be the
small mammals (Krebs and Myers, 1974). most important population mechanism in the cycle.
A wide variety of explanations have been offered Thus, the debate is now centered on two alternative
for the 10-year cycle of snowshoe hare and other ani- explanations: (1) The cycle is caused by time delays in
mals in the boreal forests of Canada and Alaska. The the interaction of three trophic levels (predators, hares,
first critical evaluations of the phenomenon were by winter food supply); or (2) The cycle results from the
Elton (1924), Huntington (1932), MacLulich (1937), interaction between predation by birds (e.g., the great
and Elton and Nicholson (1942), who examined fur horned owl) and mammals (e.g., the lynx) on the hare.
trapping records of the Hudson’s Bay Company back A recent study of the snowshoe hare cycle in the
to the 1790s. This led to 30 years of theories based on Yukon (e.g., Hik, 1994; Krebs et al., 2001) suggested
cosmic influences (e.g., the 10-year sunspot cycle), cli- that there is a complex interaction among food, preda-
matic variation, parasites and predators, physiological tion, and cover. When predator numbers are low and
“shock” variation, food supply, food quality, and ge- there is an abundance of food, hare numbers increase;
netic/behavior variations (Keith, 1963). Most of these foraging for high-quality food poses low risk of preda-
were “causal” theories: they tried to explain the phe- tion. As predator numbers increase, the hares alter
nomenon in terms of one or a small number of popu- their food-searching behavior to try, unsuccessfully, to
lation processes, in marked contrast to Elton’s (1924) avoid the increasing risk of predation, and this re-
original ecosystem-level concept that invoked the ac- stricts them to areas of low food quality. The physio-
tion of a complex of biotic and abiotic factors. More logical stress of poor nutrition and predation risk
recently, there has been a trend back toward more contributes to reduced fecundity and lower survival.
comprehensive, ecosystem-level theories (Kimmins, We still do not have a completely satisfactory ex-
1970; Krebs et al., 2001; Sinclair et al., 1993) or theo- planation for the 10-year cycle. Renewed interest in
ries that combine several mechanisms (e.g., Krebs et the influence of the 10-year sunspot cycle (Sinclair et
al., 1995). Currently, there are four competing theo- al., 1993) has demonstrated a solar cycle—climate—hare
ries to explain this cycle (Krebs et al., 1992). population cycle link that may explain the widespread
regional synchrony that has largely eluded attempts to
1. Winter food shortage, which reduces reproduction explain it so far (but see Kimmins, 1970; Watt, 1968;
at the peak of the cycle and causes starvation mor- Wellington, 1952).
390 CHAPTER 14 — Population Ecology

It seems that the further we go in researching this

phenomenon, the closer we get to where we started: a
comprehensive, ecosystem-level theory that is difficult
to test in totality by conventional scientific research, becomes
that nevertheless may approximate what is really hap- dN
pening, and that can generate testable subhypotheses. a = (r; nas k,P)N

Ultimately, the only way of evaluating such complex

ecosystem theories may be by computer simulation where N is the numbers of prey, P is the numbers of
modeling. However, we cannot construct believable predators, 7 is the innate capacity for increase for the
models until we understand the major features of all prey, and &; is a constant that measures the ability of
the key processes that are involved. A comprehensive the prey to escape predators.
long-term study of this phenomenon in Kluane Park The equation for the predator population in the
in the Yukon has provided such a knowledge base that absence of prey (assuming a geometric decline in the
has permitted the development of a credible concep- absence of food) is
tual model (i.e., theory) of how the 10-year cycle
works there (Krebs et al., 2001). dP
— =
Sarit
t
In closing this section on cyclic population fluctu- dt ‘
ations, it should be noted that several species of forest
Lepidoptera exhibit an 8- to 10-year cycle analogous where 7) is the death rate of predators in the absence of
to the hare cycle. Studies of the cycle of the western prey. This is modified as follows for the presence of prey:
tent caterpillar near Vancouver, British Columbia,
have shown that it is difficult to disrupt the cycle, ore.
isa (eo + 2N) k,N)P
which is regionally synchronized on a variety of food
plants. Although there is still no satisfactory explana-
tion, the role of viral diseases and a cyclical variation where k is a measure of the ability of the predator to
catch prey.
in resistance to the disease may be involved (Myers,
1988, 1990). It seems that the need for a comprehen-
Solving both of these equations from some initial
sive theory for these cyclical insects is as great as it is starting value produces oscillations of the two popula-
for the hare. tions that are slightly out of phase with each other
(Figure 14-10). The magnitude of the oscillation de-
pends on the starting densities of the populations.
14.6 Role of Predation in Although mathematical and graphical analyses of
Population Regulation interspecific interactions such as predation have their
uses, they have rarely proved to be an accurate descrip-
Interest in predator—prey interactions was stimulated tion of the temporal dynamics of populations in their
largely by Elton’s (1924) descriptions of cyclic animal natural environments. Too many antecedent determi-
population fluctuations, and, as suggested by the dis- nants of population growth are left unaccounted. The
cussion of population cycles, it is clear that predation equations are helpful in that they give us a basic pat-
is still believed to be a dominant form of mortality and tern of processes, but we must go to the field if we are
of major importance in determining the abundance really to understand what is happening. With this in
and distribution of many plant and animal species. mind, we can ask the question, “How realistic are the
As with population growth in general, the topic of predictions of these simple predator-prey equations?”
predation soon became the subject of mathematical
analysis. In the mid-1920s, simple equations were de-
rived to describe, for populations with overlapping Experimental Investigation of
generations and continuous breeding, the effects of Predator—Prey Relationships
prey—population size on the dependent predator popula- The first test of the predictions was made by Gause
tion, and vice versa (Lotka, 1925; Volterra, 1926). These (1934) in a laboratory study in which he attempted to
involve a modification of the geometric growth curves raise a population of protozoans (Paramecium cauda-
for prey as a result of the presence of the predator: tum) together with a predator (Didinium nasutum) in a
 Role of Predation in Population Regulation ey |

Prey
——— Predator

size
Population

Figure 14-10 Classical predator-prey population oscillations as predicted by the Lotka-Volterra equa-
tions described in the text, with 7; = 1.0, k, = 0.1, 7, = 0.5 and k, = 0.02. (Afier Krebs, 1972. Copyright CF.
Krebs. Used with permission ofthe author and Harper ¢ Row Publishers.)

liquid nutrient medium made by boiling oats in water talis), which feeds on Eotetranychus. The main feature
and using the clear supernatant liquid with an inocula- of the experiments was a progressive increase in the
tion of bacteria as food for the Paramecium. In a series degree of complexity and spatial heterogeneity of the
of experiments in which he varied the starting popula- mites’ environment as the series progressed.
tion size and the size of the container, Gause experi- ‘To start with, the predator was placed on a single
enced nothing but failure. An initial increase in orange infested with six-spotted mites, which were
Paramecium \ed to an increase in Didinium, which ate rapidly eliminated, followed by the death by starvation
all the Paramecium and then starved to death. Only of the predatory mite. The experiment was then re-
when he introduced a refuge where Paramecium could peated with 40 oranges, some of which were partly
hide from Didinium could he change this pattern. covered with paraffin wax barriers or paper to act as
When he replaced the clear liquid medium with the refuges for the prey and barriers to the dispersal of the
entire boiled oats medium (which resulted in sediment predator. In other 40-orange experiments, Huffaker
of oat fragments in the bottom of the container), some replaced some of the oranges by balls so that he could
of the Paramecium entered the sediment and thereby vary the dispersion of feeding sites. As with the single
escaped predation. The Didinium eliminated the orange, the 40-orange experiments all led to a final
Paramecium from the liquid medium and starved to elimination of both predators and prey. However, as
death as usual, but then the Paramecium emerged from the complexity of the experiment was increased in suc-
the sediment and entered a period of geometric in- ceeding experiments, something approximating a
crease in the liquid. Gause then examined the effect of Lotka—Volterra oscillation was eventually obtained in
repeated additions of predator and prey to a simple an experiment involving 252 oranges separated by a
nutrient medium. This stratagem finally resulted in complex system of barriers and vertical sticks that fa-
something that approximated the Lotka—Volterra pre- cilitated the distribution of the six-spotted mite.
dictions (Figure 14-11). Gause concluded that such Whenever a predator discovered an orange with a prey
oscillations cannot be produced by the predator and population on it, the prey was eliminated, but suffi-
prey populations themselves, that they are not an in- cient numbers of the prey mite dispersed and founded
herent characteristic of predator-prey systems, and new populations on uncolonized oranges that its over-
that they depend on outside influences (in this case, all population survived. When the predator reduced
immigration). the prey populations markedly, the prey became very
A more sophisticated laboratory test of these ideas hard to find and the predator population declined.
was performed by a Californian entomologist, Huf- The conclusion drawn from this experiment was
faker (1958). To evaluate Gause’s conclusions, Huf- that if the experimental environment is heterogeneous
faker undertook a series of experiments in which he enough and large enough to permit the normal local
examined the relationship between two mites: the six- movements of predator and prey, then reasonably sta-
spotted mite (Eotetranychus sexmaculatus), which feeds ble predator-prey oscillations can be sustained for
on oranges, and a predatory mite (Zyphlodromus occiden- considerable periods. There is evidence that if such
392 CHAPTER 14 Population Ecology

50 Prey
——-— Predator

—— Prey
120 ——— Predator

individuals
of
Number

Time (days) Time (days)

(a) (b)

Additions of 1 prey and | predator

[mec Prt Piety SHfitaeseabl

Prey
——— Predator

Time (days)
(c)

Figure 14-11 The interaction of two ciliated protozoans in a laboratory culture. (A) The
prey (Paramecium caudatum) is eliminated by its predator (Didinium nasutum), which then starves. (B)
The addition of a refuge for the prey in the bottom of the container resulted in the elimination of
the predator and the subsequent increase in prey numbers. (C) A more permanent predator-prey os-
cillation in the absence of a prey refuge was attained by periodic additions (immigration) of one prey
and one predator. (After Gause, 1934.)

experiments are successfully maintained for many local population crash. The metapopulation, composed
generations, then adaptations that reduce the ampli- of many local subpopulations that are sustained by dis-
tude of the oscillations and thereby reduce the risk of persal and colonization processes, will exhibit differ-
the populations’ being eliminated may occur (Pi- ent population dynamics from these subpopulations.
mentel et al., 1963). In many cases, our understanding of populations is
Huffaker’s study laid the basis for two important based on studies of the processes occurring in subpop-
ideas: (1) the importance of habitat diversity and “safe ulations. Recognition that these occur in the context
sites” for the maintenance ofa balance between preda- of the larger metapopulation may help us to under-
tor and prey populations and (2) the idea that many stand species populations better and will assist in the
populations are stable only when examined at a large design of rational species conservation strategies.
spatial scale. At a more local scale, there may be re- Apparently, the classical predator-prey oscillation
peated short-term geometric growth followed by a can occur in physically diverse laboratory habitats.
 Role ofPredation in Population Regulation — 393

Does this mean that predators play a major role in reg- tion grows until saturation density is reached, when
ulating populations in nature? The answer to this is, the proportion starts to decline.
“It depends on the species.” The biological control of In type 2, the increase in the number of prey taken
prickly pear in Australia, Lantana in Hawaii, St. John’s per predator for each increase in prey numbers de-
wort in the United States, and many other successful clines steadily until it levels off and the predator be-
cases of biological control of weeds by introduced phy- comes insensitive to prey density. Type 2 is found in
tophagous insects shows the power of herbivory. Many many invertebrate predators and is largely the result of
plant species in the world are maintained at low popu- handling time. It takes a finite amount of time to locate
lations by herbivores that are scarcely ever even no- a prey, but this decreases as prey density increases. In
ticed, so low is their population. Biological control of contrast, handling time (the time to overpower and
insect pests has been equally successful. Perhaps the consume the prey) remains constant irrespective of
best known example of successful biological control is density. Consequently, even when the predator can eat
the case of the vedalia beetle (Rodolia cardinalis), which continuously, an upper level of prey consumption is set
eliminated the cottony cushion scale (Icerya purchasi) as by how quickly the predator can consume each prey. In
a major pest problem after the vedalia was introduced this type, the proportion of the population taken by
to California fruit orchards in 1888. Whether a partic- the predator is initially high but declines steadily.
ular predator is able to control a particular prey can be Type 3 functional response follows a sigmoid curve
predicted if we know the details of how the predator and is typical of vertebrate predators that are polyphagous
can respond to increasing numbers of prey. This is the (prey on several different species). The response of the
topic of the next subsection. predator to increases in the population of a particular
prey species is small at low densities of that species be-
cause the predator would gain little by transferring its
Functional and Numerical Predator Responses search from other prey to this species. However, as the
One of the most significant contributions to our un- population density of the prey species continues to grow,
derstanding of how predators work and why they are predators start switching to this improved feeding op-
sometimes able to control populations and sometimes portunity and form a search image for the species (Tin-
not was made by Holling (1959, 1965), then an ento- bergen, 1960). They learn to recognize the particular
mologist working on European pine sawfly popula- prey and where and when it may be found, which in-
tions in Canada. He investigated the ability of a creases their rate of predation on this species. At higher
predator of this sawfly to respond to changes in the prey densities, satiation and handling time act to level
abundance of its food. The masked shrew (Sorex sp.) off the response curve as in the other two curves.
feeds on sawfly cocoons, which are hidden in the for- An interesting example of predator switching is
est floor of pine plantations. By examining the basic provided by Sullivan (1979). Deer mice prey on
components of predation (searching time, capture conifer seed in western North America, and this can
time, handling time, appetite satiation, and recovery contribute to the failure of natural regeneration after
from satiation), Holling was able to describe the timber harvesting. This problem can be reduced by
functional response of a predator to its prey population. providing alternative seeds (sunflower seeds and oats),
This is defined as the change in number of prey eaten which cause the deer mice to switch to these from the
per predator per unit time as the population of its prey conifer seed.
increases. These three types of functional response are
Investigations revealed three basic types of func- shown in Figure 14-12.
tional response. In type 1, the number of prey con- The slopes of the three response curves vary ac-
sumed per predator per unit time increases linearly cording to the efficiency of the predator at finding the
with increase in prey density (as predicted by the prey. In type 3, it also depends on the degree of
Lotka—Volterra equations) until some saturation value polyphagy in the predator. A predator that prefers a
is reached, at which the number of prey taken per particular prey or is opportunistic (it recognizes a
predator becomes independent of prey density; the good opportunity when it sees it) will show a much
rate of ingestion of prey equals the maximum diges- more rapid functional response than will a species that
tion rate of the predator. In this type, the predator also feeds on other prey, whose numbers may not be
takes a constant proportion of the prey as the popula- increasing. Holling found that the short-tailed shrew
394 CHAPTER 14 Population Ecology

Type | Type 2 Type 3

time
of
unit
predator
per
Number
of
eaten
per
prey
Prey Density

Figure 14-12 Three major patterns of functional response of predators to increasing prey
populations as proposed by Holling (1959). In type 1, there is a linear increase in number of prey
consumed per predator per unit time, up to some upper limit. In type 2, there is an asymptotic ap-
proach; and in type 3, there is a sigmoid approach to the upper limit. (Copyright Entomological Society
of Canada. Used with permission of Entomological Society of Canada and the author.)

(Blarina sp.), which is a comparatively uncommon but Numerical and functional response information is
opportunistic feeder, responded to the density of pine interesting, but it is the total response (functional X nu-
sawfly cocoons much more rapidly than did the deer merical) that is of real interest in assessing the impor-
mouse (Peromyscus sp.) or masked shrew (Sorex sp.) tance of predation in population regulation. Figure
(Figure 14-13A). Because of the rapid nature ofits nu- 14-13C shows the total response of the three species of
merical response, Blarina is an important component small mammals preying on European pine sawfly. Note
of the contribution of predators to population regula- that below a certain density of sawfly cocoons, each of
tion as long as the prey population does not get close these predators responded effectively to prey density.
to the upper asymptote of the response curve. However, for each species and for their combined ac-
Predators can also respond to greater prey numbers tivity, there was a prey density above which they be-
by an increase in their population density (numerical re- came less and less effective as a population control
sponse) (Figure 14-13B). This is obviously a longer- factor. This critical prey density is called escape density.
term phenomenon than the functional response, and
because of the inherent time lags, numerical responses
tend to generate classical predator-prey oscillations. 14.7 Plant Population Ecology
There can be three major types of numerical response:
(1) predator populations may increase, (2) they may Despite that the first significant paper in population
show no response, or (3) they may decrease as prey ecology was written by a botanist, Nageli (in 1874;
populations increase. For many species of predator, Harper, 1977), this branch of ecology was developed
immigration, greater natality, and reduced juvenile almost entirely by animal ecologists until the 1970s.
mortality accompany an increase in prey population Malthus (1798) had earlier made it clear that the basic
density, and this leads to a growth in the predator pop- characteristics of populations were similar in both
ulation (type 1). The predator numbers do not grow plants and animals, but it was not until the work of
linearly, however, because food is not the only determi- Harper (1977) and his students that plant population
nant of predator carrying capacity. If the predator feeds ecology attracted much scientific attention. In some
on other species, then it may respond to an increase in ways, plants are easier to study than animals because
one prey species simply by switching its feeding to that they are stationary, because they are easy to enumer-
species with relatively little change in either total pre- ate, and because behavior does not complicate mat-
dation or predator population density (type 2). As the ters. However, from the population dynamics
prey species declines, the predator switches back to al- standpoint, plants can present problems because in
ternative foods. Increase in prey abundance may result some respects a plant is more analogous to a colony of
in increased abundance of and competition from other animals than an individual animal. Each leaf on the
species of predator, and this may lead to a decline in plant competes with other leaves for the basic necessi-
numbers of the first predator (type 3). ties of life: water, nutrients, light, and space. Each leaf
 Plant Population Ecology 395

x
60 x CS = Sorex
iS 300 Blarina
o
i oe ¥
so}
Q,
om e AS Va
& /
E Ps
= te wz — i rE

35 200 7 Peromyscus
o
OT a x 30 /
ite a 2 /
2s :
53 100 -—
7 «— Sorex S OL oe ae —.-—-—.- —e Peromyscus
=] 5 S
ZO
Ae
ig IIS)
seers
Blarina
() at eon lll . fella |e all Le el lames
0 500 1000 1500 2000 2500 0 500 1000. =1500 §=©2000)=62500 )=63000 = 3500
Number of cocoons per hectare (X 1000)
(a) (b)

50
oe Ns Total

5
5) 40 :S
5 Se
= Ss
S 30 oes
8 \_ Sorex as
es
en20)
S Se
a ss
5 ie SS.

2 10 Peromyscus”

0 500 1000 1500 2000 2500 3000 3500 4000 4500

Number of cocoons per hectare ( 1000)
(c)

Figure 14-13 Functional (A), numerical (B), and total (C) response of three species of small
mammalian predators to changes in the density of one of their foods: cocoons of the Euro-
pean pine sawfly (after Holling, 1959). Blarina had the greatest and most rapid functional response,
but because of its negligible numerical response, it is an effective predator only at lower population
densities. Sorex was the most effective predator because although it had only a modest functional re-
sponse, it had the highest numerical response. The vertical dotted lines in C show prey escape densi-
ties, above which the individual predators or their combined activity becomes progressively less ef-
fective in controlling the prey. (Copyright Entomological Society of Canada. Used with permission of
Entomological Society of Canada and the author.)

has its own life history from birth (as a primordium on years. In fact, a plant really is analogous to a popula-
a meristem) through growth to maturity and eventual tion of buds or of leaves.
senescence, unless it succumbs to some mortality fac-
tor first. Populations of leaves have a survivorship
Recruitment of New Plants Into
curve and an age-class structure, and they undergo
population growth from a small number of cotyledons
the Population
on the germinating seedling to the large population Recruitment can occur in several ways, as a result either
on the fully mature plant. The leaf population may of sexual reproduction or of vegetative reproduction.
have discrete generations, as in deciduous perennial or
annual plants, or overlapping generations, as in ever- 1. Seed rain: the production at and dispersal of seeds
green conifers that maintain their leaves for many to a site. This is analogous to the two population
396 CHAPTER 14 Population Ecology

500
Edge of clearing

400

Clearcut area

|ss,
S

200

seed
(sound
Seed
rain*yr
m

0 50 100 150
Distance from forest edge (m)

Figure 14-14 The seed rain from Picea engelmannii in three forest clearings in Utah as a
function of distance from the stand edge. (Data from Roe, 1967.)

processes of natality and dispersal. Plants have a ground organs. The meristems of these buds re-
genetically determined fecundity (potential maxi- main dormant under the influence of hormones
mum number of reproductive propagules per plant produced by the live aboveground biomass. When
per year or per generation) and an environmen- this biomass is killed, the meristems become active
tally determined fertility (number of reproductive and new shoots are produced. This is an important
propagules actually produced per plant per year or adaptation in trees that live in areas prone to
per generation). The seed rain varies as a function stand-replacing fire (Chapter 12, Effects of Fire on
of distance from the plant (Figure 14-14), wind, Plants), and in many forests, the bud bank can be
and any other factor that influences seed dispersal. more important in determining the density and
. The seed rain contributes to the seed bank: the pop-
ulation of living but ungerminated seeds contained Table 14-7 Number of Seeds in the Seed Bank of
Various Different Forests in Maine
in the soil. The seed bank of forests can be as large
as 3,000 seeds per m, (Table 14-7). Vegetation Seeds m”
Most of the viable seeds in the seed bank are lo-
White pine (70-yr stand) 1,000 (173)
cated in the surface layers of soil, and the removal
of the forest floor by fire or mechanical means will White pine (80-yr stand) 320 _

generally remove most of the viable seeds. The Spruce, fir (30-yr stand) 2,850 (49)
ratio of viable seeds of different species changes Red pine plantation (24-yr stand) 532 (5)
between the soil surface and deeper soil layers, so Beech, yellow birch, sugar maple 1,000 (91)
removal of different depths of forest floor or sur- (110-yr stand)
face soil will alter the species composition of ger-
Beech, yellow birch, sugar maple 218 (42)
minants recruited from the seed bank into the
(50-yr stand)
plant community as well as reduce the number
(Figure 14-15). Sugar maple (150-yr stand) 122 (11)

In addition to the seedbank, there is often a bud

Data from Olmsted and Curtis, 1947.
bank: a population of buds on the lower stem; the Values without parentheses are direct counts by microscopic exami-
root crown; or roots, rhizomes, or other under- nation. Values in parentheses are the result of germination tests.
 Plant Population Ecology 397

Number of viable seeds (X 10° m °) to break seed dormancy; (2) the conditions and re-
5 10 ils) 20 sources required for germination are present; and (3)
specific mortality agents such as animals, diseases,
toxic or lethal soil conditions, and competition for
light and soil resources are tolerable (Harper, 1977).
The absence of seedlings of a species from a site may
result from a lack either of seeds or of safe sites in
which they can become established. The availability
:N
LA
CLL
and characteristics of the safe sites in an area act as an
MWSVRee environmental filter that results in the species composi-
N
tion of the community of germinants being different
a N
from that of the seed rain and/or seed bank.

in
Depth
soil,
the
cm
i)(=) Survival
Seeds that germinate to become seedlings, buds that
develop shoots (ramets), and/or seedlings that are re-
leased from the seedling bank by disturbance form a
new cohort of plants. As with a cohort of animals, this
30 _—
EE plant cohort will exhibit a characteristic survivorship
curve, the shape of which varies between different
Figure 14-15 The seed bank in a rich hay meadow in plant species and between different environments for
Wales. The graph shows the number of viable seeds of various any particular species.
species at various depths in the soil. (Data from Chippindale and As in many animal populations, there is heavy ju-
Milton, 1934; after Harper 1977. Used with permission ofAcademic venile mortality in most plant populations. Figure
Press and the author.)
14-16 shows the causes of mortality among seedlings
of Douglas-fir on six different substrates under three
light regimes in a clearcut area in Oregon. It is inter-
species composition of the postdisturbance plant esting that the total mortality did not vary greatly by
community than the seed rain or seed bank. Dif- substrate or light condition even though the type of
ferent species have bud banks at different depths. mortality agent or the relative importance thereof var-
As a result, fires of differing severity, which heat ied. Compensatory mortality seems to have been im-
the soil to lethal levels to differing depths, will re- portant in this study.
sult in very different postfire communities devel- Tree populations have survivorship curves and
oping from bud banks. age-class distributions that vary greatly between
4. The seedling bank. Seedlings of shade-tolerant tree species and the stage of development ofthe plant com-
species can remain in the understory as suppressed munity. Light-demanding pioneer species tend to
seedlings or saplings for decades or, in extreme have a small number of age classes (i.e., are even-aged)
cases, for centuries. These may be capable of re- that pass up and out of the age-class distribution as
sponding to increased light and reduced competi- the species is eliminated in the course of forest aging
tion for soil resources after disturbance to the and successional development. Shade-tolerant climax
overstory canopy. Where there is a well-estab- species in old forests have a more stable age distribu-
lished seedling bank that survives a stand-replacing tion (Figure 14-17). Figure 14-18 presents survivor-
disturbance, it may dominate the seed rain, seed ship curves for various tree species on the eastern
bank, and bud bank in determining the composi- slope of the Rocky Mountains near Boulder, Col-
tion of the postdisturbance community of trees. orado. The different shapes of the curves match the
successional status of the trees. Lodgepole pine is a pi-
Recruitment from the seed rain and the seed bank oneer species that, in the absence of fire, is normally
depends on the existence of so-called safe sites. These excluded in 100 years, Ponderosa pine is a midseral
are areas in which (1) there is an appropriate stimulus species, and the mature (climax) forest is dominated by
398 CHAPTER 14 Population Ecology

No shade Light shade Heavy shade

80

60

40
%
Mortality,
20

ChB 5 IMS; “ees Ch HB LB MS L S$

Substrate type
(a) (b) (c)

Burn type/substrate Mortality factor:

Ch = Charcoal MS = Mineral soil Heat Damp-off (fungi)
HB = Hard burn L = litter
Animals [ ] Other factors
LB = Light burn S = sawdust

Figure 14-16 Causes of Douglas-fir seedling mortality on six different substrates under
three degrees of shading in a clearcut in Oregon. (After Hermann and Chilcote, 1965. Used with
permission of R.K. Hermann.)

Engelmann spruce (Knowles and Grant, 1983). Sur- manner. Tree stems in crowded stands are slender and
vivorship curves and life table analysis were used by more susceptible to damage by heavy snowfalls than
Yarie (1981) to determine the historical frequency of open-grown trees. Root systems in dense stands are
forest fires in Alaska. smaller than in comparable low-density stands, in-
creasing the risk of windthrow. Physiological stress
Density-Independent Regulation associated with high interspecific competition may re-
of Plant Populations duce resistance to insects and diseases, and fire type,
behavior, and severity will be influenced by stand den-
Tree height, leaf area, maximum biomass, and maxi-
sity and structure.
mum basal area of trees per hectare are largely deter-
mined by climate, soil moisture, and nutrients. This is
the basis for site index: the classification of the pro- Density-Dependent Regulation
ductive potential of a site for tree growth by the height Competition is a major factor that determines the size
of the dominant trees at an index age. These variables of individual plants and the number of plants in the
are fairly constant for a particular species on a particu- population. An even-aged cohort of plants undergoes
lar site over a wide range of stand density (Oliver and competition-induced mortality that reduces popula-
Larson, 1990) and therefore are relatively density in- tion density as the average size of the individual plants
dependent. They vary for a given species on different increases. The relationship between stem number and
sites, between different species, and at extremes of mean size exhibits a remarkably constant relationship,
density. An open-grown forest in which trees have not which is referred to as the se/f-thinning rule.
formed a closed canopy or a forest at very high density The self-thinning rule (Harper and McNaughton,
may have different tree heights, foliage mass, basal 1962; Yoda et al., 1963) says that if you plot the loga-
area, and total biomass than a “normally” stocked rithm of mean total mass of individual plants against
stand ofthe same age. the logarithm of the number of plants per unit area
Density-independent mortality may occur as a re- (stand density) for fully stocked stands, then you will
sult of insect outbreaks, wind, fire, snow, or diseases, get a straight line with a slope of -3/2 (Figure 14-19).
but these can also operate in a density-dependent The conceptual basis for this relationship is that any
 !)
ha™
(stems
Density

Figure 14-17 Change in the frequency (stems ha") of stems of different diameters (dbh) in a
chronosequence of postfire stands in the sub-boreal spruce forests of central British Columbia
(redrawn from Clark, 1994, with permission). The figure shows the numbers of stems in each size class in
68- (A), 111- (B), 210- (C), 311- (D), 370- (), and 436-year-old (F) stands. The pioneer pine domi-
nates the young stand but is replaced over time by spruce, which is long-lived and reaches large sizes.
In the long-term absence (>250 years) of stand-replacing disturbance, the abundance of spruce declines
and the stand becomes dominated by subalpine fir, which exhibits a typical reverse-J size distribution.
num-
Stands older than 230 years did not exhibit much variation in species size-class structure: smal]
bers of spruce, with a few large old trees, and large numbers of fir reaching only medium size.

399
400 CHAPTER 14 Population Ecology

100

Englemann spruce
O—O Ponderosa pine
80 o—o Lodgepole pine
x—x Limber pine

60

40
survival
Percent

0 100 200 300 400 500

Age class (years)

Figure 14-18 Survivorship curves for four species of conifer in the Rocky Mountains of
Colorado. Engelmann spruce and ponderosa pine are climax dominants on the appropriate sites,
whereas lodgepole pine is a pioneer species that colonizes burned areas. The limber pine is neither
climax nor pioneer; it can play either role but is often found in patches at a variety of seral stages.
(After Knowles and Grant, 1983. Used with permission ofthe Ecological Society ofAmerica and P. Knowles.)

site has a maximum plant biomass-carrying capacity. were fewer of them. On a log-log scale, there would be
As the plant population approaches this limit, individ- an inverse relationship between number of objects and
ual tree growth can continue only if the number ofin- their size (when the volume was filled), and a graph of
dividuals is reduced. If the trees were inanimate this relationship would have a slope of -1. Plants that
objects all of the same size filling a given volume, then grow from low initial density to fully occupy a site do
the individual objects could get bigger only if there show this relationship at the point at which they have
just reached the carrying capacity. However, plants in
dense stands where stand self-thinning is occurring
generally exhibit a -3/2 slope: mean tree size is larger
for a given stand density than expected by the simple
relationship between tree size and number expected
where all the trees are the same size. This reflects that
log mean tree populations have a distribution of stem size rather
plant mass than a single size and that because of time lags in trees
dying from competition, individual tree size increases
faster than stand density is declining in stands that are
undergoing density-dependent self-thinning.
The -3/2 power law has attracted a lot of attention
from applied ecologists. It is the basis of stand-density
log stand density
management guidelines that suggest planting densities
Figure 14-19 The self-thinning rule or power law show- and thinning regimes for agricultural and forest crops
ing the self-thinning line with a slope of -3/2 and the tra- to achieve a particular crop plant size before competi-
jectory of various hypothetical stands developing from dif- tion begins to affect individual plant growth (e.g., Drew
ferent initial stand densities (dotted lines). and Flewelling, 1977, 1979). However, the idea that the
 Plant Population Ecology 401

self-thinning rule is one of the more general principles species and mixed species stands. Most understory
of plant population biology (White, 1980) and that it is vascular plant species are shaded out in this phase,
universally applicable has received some criticism and mosses may be smothered by the heavy litter-
(Lonsdale, 1990; Weller, 1987; Westoby, 1984, Zeide, fall. The stands are dark, and the ground is bare of
1987). The conclusion seems to be that there is far too vegetation.
much variation in the slope of the self-thinning line to 3. Understory reinitiation. As stand self-thinning con-
support the claim that there is a single ideal relation- tinues and trees get larger, canopy gaps fill in more
ship that holds true for all species at all densities on all slowly, and crown damage caused by the swaying
sites. Nevertheless, the concept of a self-thinning rule of tall trees in the wind or by snow may prevent
is attractive and has proved useful as a qualitative ap- complete reclosure of the canopy after gap forma-
proach to the management of stand density. tion. The increase in light levels received at the
ground and the decrease in aboveground litterfall
permit the development of understory vegetation
Stand Development Phases as a Result of and the recruitment of a seedling bank.
Competition and Self-Thinning 4. Old growth. As the stand ages, canopy gaps caused
by individual tree death do not close, permitting
There are four main phases in the development of an
shade-tolerant seedlings and saplings in the under-
initially even-aged cohort or population of trees
story to be recruited into the canopy. This leads to
(Figure 14-20) (Oliver and Larson, 1990). The devel-
a mixed age, multicanopy stand with standing dead
opment of these phases can be accelerated or retarded
trees (snags). As this stand structure develops, the
by appropriate stand management:
forest enters the old-growth phase, which can de-
velop in any seral stage, not only the late-seral
1. Stand initiation. The early development of a cohort of
stages (Chapter 17) that many people assume are
trees before canopy closure during which tree mor-
required for “old growth.”
tality may occur as a result of density-independent
mortality factors or interspecific plant competition,
and recruitment of additional individuals may occur.
Change in Size Distribution
2. Stem exclusion. After the closure of the canopy, ad-
ditional recruitment ceases and density-dependent
of ‘Irees Over Time
mortality is caused mainly by competition for light Figure 14~17 illustrates how the size-class distribution
(moisture competition may be important in dry of a forest changes over time as it undergoes succession.
environments). The stand undergoes self-thin- Such diagrams have a broad overall relationship to the
ning, the details of which will vary between single age-class distribution of a developing forest, but be-

a,

ed
Am = a7 8kS
sv

iil ull

Stand initiation phase

ae Stem exclusion phase
Understory isan
Old growth phase
reinitiation phase

Figure 14-20 The four phases of stand development, from stand initiation to old growth
(based on Oliver and Larson, 1990).
402 CHAPTER 14 — Population Ecology

cause tree size is a function of stand density as well as of survivorship curve of that genotype under the particu-
tree age, the relationship of age to size is often inexact. lar conditions. In the nursery, the forester must ac-
After a stand-replacing disturbance that removes count for percentage viability and percentage seedling
all trees, both large and small (e.g., a fire), an area may mortality if she or he is to sow enough seeds to achieve
be invaded by herbs and shrubs and subsequently (or production goals. The forester must know about in-
at the same time) by a relatively even-aged population traspecific competition in the seedbed and its effects
of early seral tree species (the Pinus in Figure 14-17). on seedling morphology if he or she is to get the right
The size- (or age-) class distribution of this initial tree type of seedling for planting out. When direct seeding
population is usually narrow and more or less bell- of clearcuts is used, the forester should have a life table
shaped (Figure 14-17A). As time passes, these trees for those seeds in the application area, or experience,
(the Pinus) grow older and larger, the tree size distri- to know how many seeds to apply to achieve a desired
bution moving to the right on the graph (Figure population of seedlings. If predation by small mam-
14-17B). There will not normally be any recruitment mals is an important mortality factor, then alternative
of additional seedlings of the shade-intolerant early foods can be provided (e.g., sunflower seeds) to cause
species, but seedlings of more shade-tolerant mid- and predator switching (Sullivan, 1979). During stand
late-seral species will invade the stand if a seed source management, the forester should know the natural
is available. The shade-tolerant species will develop a mortality rate of the trees so that she or he can plan
reverse-f size- (and age-) class distribution (Abies in the intensity and frequency of thinning (subject to
Figures 14-17B-F). Moderate shade-tolerant, mid- economic and human resource constraints). This type
seral species (Picea in Figures 14-17B—F) will develop of information is sometimes provided to foresters in
a relatively flat size- and age-class distribution as a re- the form of natural stocking curves, but unfortunately
sult of continuous slow recruitment into canopy gaps the population information needed for sophisticated
and small-scale disturbances (safe sites). Where such population manipulation is frequently lacking. Stand
species (the Picea) live longer and have a better sur- density diagrams, based on the relationship shown in
vivorship than the more shade-tolerant climax species Figure 14-19, are useful in this respect.
(the Abies), they will develop a flat distribution from Wildlife management, range management, and
seedlings up to a few large and old individuals, where- fish habitat management, which all are variously asso-
as the shade-tolerant climax species will continue to ciated with timber management, are areas where there
exhibit a reverse-J distribution up to intermediate is a great need for improved knowledge of population
sizes and ages. If the stand-replacing disturbance had ecology. Much more information is required concern-
left a seedling bank of late-seral trees (e.g., a wind- ing carrying capacity, natality, mortality, dispersal, sur-
throw event), then the early size-class structure would vivorship curves, stationary age-class distributions,
be different. and the nature of population regulation (e.g., the im-
The typical low-elevation coastal sequence in portance of weather, territoriality and other self-
British Columbia of red alder—Douglas-fir—-western regulating mechanisms, parasites, predators, disease,
hemlock would have a similar chronosequence pattern food availability and quality) if we are to be more suc-
of size distribution to the pine-spruce-fir sequence cessful in managing animal populations in forest
shown in Figure 14-17. ecosystems.
Control of insects and diseases is a vitally impor-
tant area of forest management. Enormous quantities
14.8 Importance of Population Ecology of potentially harvestable biomass are transferred to
in Forest Management decomposer food webs annually by these biotic mor-
tality agents. Their control is expensive and not always
Timber management is, to a large degree, applied biologically or economically warranted. They may
population ecology. Foresters attempt to establish a merely kill trees that would otherwise be killed by
population of a particular genotype of a particular some other mortality agent. Control is often not initi-
species with the objective of attaining a particular ated until a point in the outbreak at which the “pest”
number of mature individuals of a particular size at a population is declining naturally, and high pest popu-
particular age. This involves manipulation of the age- lation levels early in the pest’s life history may not nec-
class structure to achieve a predetermined age-class essarily mean economic damage if some severe
distribution, which in turn requires knowledge of the mortality agent operates on later stages of the pest
 Take-Home Message 403

population. In the absence of adequate population in- some circumstances, especially at low to moderate prey
formation, the forester may be committed to expen- population, predation can be a major factor preventing
sive, unnecessary, and ineffective control measures. If increase in prey abundance but that above a certain den-
one is to be able to predict the course of pest popula- sity it becomes relatively unimportant. This threshold
tion trends accurately enough to be able to apply con- density is specific to each particular predator-prey sys-
trol measures in an ecologically and economically tem. Knowledge of the functional, numerical, and total
response of predators to varying prey abundance is of
sound manner, then the major determinants of their
critical importance to the design of programs of biologi-
populations must be known. cal control of pests based on predation.
Population ecology has traditionally been the domain
SUMMARY of animal ecologists, but in the past few decades, a school of
plant population ecology has been developed. There are
For many ecologists, understanding what controls the
strong parallels but also significant differences between
distribution and abundance of organisms is the focus of
plant and animal populations that require slightly different
their science. These population ecologists are concerned
approaches for the two groups. Despite these differences,
with the four major factors that determine population
considerably improved insights into the best way of manag-
size and distribution: natality (the rate at which individu-
ing plant populations can be obtained by applying concepts
als enter the population through reproduction), mortality
and techniques developed for animal population ecology.
(the rate at which individuals die), and immigration and
Forest resource managers are concerned with ecosys-
emigration (the rates at which individuals enter and/or
tems, and the ecosystem approach must be the basis on
leave the population through movement). These four
which they conduct their activities. However, their im-
processes are in turn regulated by a wide variety of phys-
mediate concern is often with populations; therefore,
ical and biological factors. their knowledge of ecosystem ecology must be comple-
Populations have a set of attributes that are unique to mented by a sound appreciation of population ecology.
this level of biological organization. They have a pattern As noted in Chapter 3, a focus on populations is com-
of dispersion in their environment, there are characteris- pletely compatible with the ecosystem concept as long as
tic patterns of change in abundance over time (either in- there is explicit recognition of the ecosystem framework
creasing or decreasing), there is a characteristic schedule within which populations exist.
of mortality over the life of a particular cohort, and there
is an age-class structure to the population. Understand-
ing populations involves understanding how these attrib- TAKE-HOME MESSAGE
utes are determined.
Development of theories about the regulation of the The abundance and spatial distribution of a particular
abundance of animal populations has engaged the minds species are strongly affected by the physical environment
of population ecologists for more than three quarters of a and by other species, but processes that are unique to the
century. The early theories were causal in nature; they population are also of prime importance. Understanding
sought to explain population dynamics in terms of a single why the number of a particular wildlife species varies
factor or a limited set of factors. Great controversies from year to year and from place to place or being able to
raged for many decades about whether density-dependent predict how a particular stand of trees will change in
biotic factors, especially competition, were preeminent or stand density, tree height, and biomass over time requires
whether density-independent factors, especially physical knowledge of population ecology, as well as of other as-
factors of the environment, were primarily responsible for pects of the ecosystem.
controlling abundance and distribution. Most of these An understanding of the determinants of population
early theories were based on the action of factors external growth and regulation is essential to any attempt to man-
to the population, but over the past three decades, addi- age a population, whether you are working with the
tional theories have been developed, based on behavioral, human population, a stand of trees, a population of a
physiological, and genetic mechanisms—factors internal wildlife species, or an outbreak of a “pest” insect or “weed”
to the population. A more recent school of thought has plant species. Success in species conservation, preventing
taken a position midway between these competing ideas. overpopulation, and achieving desired population levels or
The comprehensive school suggests that many different patterns of change over time depends, to a considerable
population and ecosystem processes are involved but that extent, on the application of this type of knowledge.
their relative importance varies for different types of or- Community ecology is concerned with the inter-
ganism and in different types of environment. action of species. To understand fully the biotic com-
Predation is a mechanism of population control that munity, we must understand the population ecology of
was long believed to be important in regulating prey pop- the individual species in the community. A knowledge of
ulations. Analytical investigations have shown that in population processes in the plant community is also es-
404 CHAPTER 14 — Population Ecology

sential for an understanding of ecological succession. . What determines the size of a population?
Clearly, sustainable forest management requires a basic . Which of the theories of population regulation do
knowledge of population ecology. you think fits the observed patterns of population
growth and change over time best?
. Can predators regulate populations of prey?
STUDY QUESTIONS
. What are seed rain, seed bank, and bud bank? Why
if How do populations develop? does a forester need to know about them?
Ds Is it better to exist as an individual organism or to be . What does the age- (or size-) class structure of a forest
a member of a population? tell you about the future development of that forest?
. How does the abundance of organisms in a popula- . What is the relationship between the number of trees
tion change over time? in a stand and the average size of the trees?
. Is there a “balance of nature”? Do populations in- . What are the different stages of stand develop-
crease to a stable level that represents this “balance”? ment called, and how do these relate to stand self-
. Why is knowledge of the age-class structure ofa pop- thinning?
ulation important? 1. How can a forester use knowledge of plant popula-
. How does age-class structure relate to survivorship tion ecology to predict whether a “weed” competi-
curves? tion problem will develop after harvesting?
 Community Ecology

15.1 Introduction ogists who studies the relatively species-poor beech

forests of Switzerland concluded that this type of com-
The survival, abundance, and distribution of a species munity contains approximately 10,500 species of
depend on its adaptation to the physical environment plants, animals, and microbes (Daubenmire, 1968).
and to the other living organisms with which it shares Even in dense managed monocultures of conifers, one
that environment. In Chapters 13 and 14, we exam- is still confronted by an amazing diversity of organ-
ined the role of the physical environment and in- isms that coexist and interact in still incompletely un-
traspecific interactions in determining the abundance, derstood ways. Decomposer organisms in the soil,
distribution, and population dynamics of a species, as fungal and insect pathogens, various species of lichens
well as certain interactions between species in differ- and mosses, a scattered ground flora of herbs and
ent trophic levels (predator-prey interactions). ‘This shrubs, and the bird and animal populations that are
chapter focuses on the structure and organization of inevitably associated with any plant community to-
the complex mixture of species populations that exist gether form an integrated, interacting system that is
together in what we call the biotic community and on re- responsible for the observed patterns of energy flow,
lationships between different species that occupy the biomass accumulation, and nutrient cycling.
same or different trophic levels. The biotic community has been defined as “an as-
Populations do not exist in isolation, except in arti- semblage of plants, animals, bacteria and fungi that
ficial experimental environments. They also do not live in an environment and interact with one another,
exist in nature in association with only a population of forming a distinctive living system with its own
a single species of host, competitor, or natural enemy. composition, structure, environmental relations, de-
Even in a wheat field or a dense young plantation of velopment and function” (Whittaker, 1975a). Each
Douglas-fir trees, there will be other species of plants community is characterized by a particular species
(agricultural weeds, or mosses, lichens, and other composition, vertical structure, patterns of change
minor forest vegetation, respectively), the number and over time, biomass, energy flow, and nutrient cycling.
variety of which will tend to increase as the age of the It is the biotic component of the ecosystem. In com-
stand increases. mon with the term ecosystem, community has no im-
Dense monocultures (communities dominated by plicit definition of spatial extent or boundaries,
one species of plant) do occur in unmanaged forests or although its geographical extent can be defined by the
native grasslands, but in many ecological zones, they plant association.
are much less common than are mixed communities. The community can be split into three subdivi-
The dominant plant is usually accompanied by a sur- sions for purposes of study and description: the plant
prising variety of other organisms. Botanists and zool- community, the animal community, and the microbial

405
406 CHAPTER15 Community Ecology

community. It is often convenient to make this subdi- The concept of the community as a supra-organ-
vision, but it must be remembered that the intimate ism has never received widespread acceptance and is
degree of association and interaction among these not popular among ecologists today because of the nu-
three components requires that they are never consid- merous fundamental differences between a commun-
ered in isolation. We have already seen that although a ity and an individual organism. For instance, the death
clear understanding of intraspecific mechanisms of of an organism is not really comparable to the replace-
population increase and decline is essential to an over- ment of a mature or climax community by the com-
all understanding of ecosystem structure and function, munity of an earlier stage of ecosystem development.
only when such an understanding is placed within an Similarly, the successive replacement of communities
ecosystem framework will it confer reliable powers of through successional development cannot be com-
prediction about the ecosystem. Similarly, knowledge pared with the growth of an individual. Certain suc-
of the plant community is frequently insufficient on its cessional stages can be omitted from certain types of
own as the basis for reliable predictions about the bi- successional sequence with little effect on the final
otic community or ecosystem as a whole (see. also the outcome, and the species composition of a given com-
discussion in Chapter 6). munity can vary significantly from time to time. In
Early interest in communities focused on the plant contrast, it is impossible for any of the stages of on-
component. From the time of Theophrastus until the togeny to be omitted from the development of an in-
middle of the 19th century, little or no recognition was dividual organism. Above all, communities lack the
given to the important role of animals or microbes. strict genetic definition that is inherent in the individ-
Karl Mobius, a marine biologist who studied oyster ual. Despite these reservations, it is generally recog-
communities in the late 19th century, has been cred- nized that communities are organic entities in which
ited with being the first to emphasize the importance there is a considerable degree of internal intercon-
of studying the entire biotic community. He recog- nectedness, and many ecologists still accept the con-
nized the intimate association among oysters (the ob- cept of the community as a quasi-organism. ‘This
ject of his study), the algae on which they feed, the simply says that there are certain similarities and par-
parasites that prey on them, and other organisms (e.g., allels between these two different levels of biological
sponges) that compete with the oysters for space. organization.
Recognition of the integration of the components of a The early focus on plant communities probably
biotic community into a functional system led to the occurred because plants account for the majority of
idea of the community as a supra-organism, which in the organic biomass in most terrestrial ecosystems and
turn led the English ecologist Tansley to develop the because plants are frequently the physically dominant
ecosystem concept in 1935. More recent studies in the and visually most impressive members of the com-
ecological subdisciplines of energetics and biogeo- munity. Although the animal community often plays
chemistry have confirmed many of the early ideas con- an important role in regulating the composition
cerning the holistic and integrated nature of the biotic and structure of the plant community, the former is
community. largely dependent on the latter for food and shelter.
Development of the analogy between the commu- Consequently, animal communities are generally
nity and the individual organism is largely attributed more determined by the plant community than vice
to the American ecologist Clements. He compared the versa, notable exceptions notwithstanding. Animal and
developmental stages of a community as it progresses microbial communities are also more difficult to study
from a group of pioneer organisms that invades an un- than is the more static plant community.
occupied area to a final, mature, self-replacing, or This chapter starts with the vertical structure and
climax community (see Chapter 17), with the develop- diversity of growth forms in plant communities and
mental stages of an individual from birth to maturity. then examines the major types of interaction that
He believed that a community, like an organism, is occur between the different species in a biotic com-
born, grows, matures, reproduces, and dies. That a munity. The spatial variability of plant communities
forest community on many sites may redevelop and their distributions along environmental gradients
through a fairly regular pattern of stages after succes- was covered in Chapter 13. This chapter concludes
sive disturbances was used to support the analogy (but with a discourse on the diversity of communities: the
see the elaboration of this topic in Chapter 17). complex and important issue of biological diversity.
 Structure and Growth Forms of Plant Communities 407

15.2 Structure and Growth plants with large woody, perennial, aboveground
Forms of Plant Communities stems, generally taller than 3 m; (2) shrwbs—plants with
medium to small, woody, perennial aboveground
Plant communities have several characteristics by which stems, mostly less than 3 m tall; (3) berbs—plants with-
they can be described: structure, life form, spatial pat- out a perennial aboveground stem; (4) thallophytes—
tern, species composition, successional stage, biomass, nonvascular plants without perennial underground or
and functional processes (energy flow and nutrient cy- aboveground stems; and (5) epiphytes—plants that grow
cling). We start by examining structure and life form. wholly aboveground on other plants. An additional
layer found mostly in tropical but also some temperate
forests is (6) Jianas—woody climbers or vines.
Community Structure as a Result
Most of these layers can be subdivided on the basis
of Plant Life Form
of height. In the description of plant communities,
The structure of a plant community refers to the vertical ecologists may divide the tree layer into three sublay-
arrangement and spatial organization of the plants. It ers and the shrubs into two sublayers. Some foresters
does not require many visits to a forested area to give divide the tree layer into several crown classes (e.¢.,
one the impression that vegetation occurs in distinct dominant, codominant, intermediate, and suppressed)
layers (subdivisions of the vegetation based on plant on the basis of the competitive status of the crown.
height) and that each layer is characterized by one or These classes are used both for stand description and
more distinct groups, or synusiae, of plants. In a tropical for prescribing thinning treatments for overcrowded
forest, the tree layer consists of a tree synusia and a plantations or natural stands. For a summary of domi-
woody climber synusia. The herb layer of a temperate nance classifications, see Daniel et al. (1979).
forest may include a perennial herbaceous grass synusia, Most of the major growth forms have several sub-
an annual herb synusia, and a bulbous perennial herb divisions. Herbs can be split into grasses, ferns, and forbs
synusia. Visits to forests in different localities within a (herbs that are neither ferns nor grasses). Trees can be
climatic region and to forests in different climatic re- split into needle-leaved trees, broad-leaved evergreen
gions will expand this impression to include that the trees, broad-leaved deciduous trees, rosette trees, and
structure or layering of forests varies between different so on. Thallophytes can be divided into those that
sites within a region as well as between different re- grow on the ground, those that grow on rock (epiliths),
gions. In fact, community structure is a characteristic and those that grow on rotting wood (epixyles). Epi-
feature of different sites and regions. The structure of phytes can be classified according to whether they
the plant community on a dry, rocky ridge is different grow on foliage, branches, or stems. A list of the major
from that in the valley below (see Figure 6-10). The terrestrial growth forms is presented in ‘Table 15-1.
structure of the boreal forest varies from that of tem- Plants that occupy a particular layer exhibit adapta-
perate deciduous forests, and both differ from the struc- tions to the conditions that they experience in that
ture of a tropical rain forest (Figure 15-1). layer. ‘Trees that are exposed to the physical stresses of
The vertical structure of plant communities is in wind, to high light intensities in the upper crown, and
part the consequence of variations in the growth form to rapid transpiration stress have strong woody stems
or gross morphology of the plants (cf. the life forms of and branches, have large spreading or deep root sys-
Raunkiaer, Figure 8-13). The overall growth form of tems, and bear foliage with xeromorphic characteristics.
communities is referred to as their physiognomy. A for- Forest forbs, however, have little need for the strength
est is an ecosystem characterized by a plant commun- of woody stems, but they must be able to photosynthe-
ity dominated by plants with a tree physiognomy. A size efficiently at low light intensities. They typically
prairie is an ecosystem dominated by plants with a have higher levels of foliar nutrients and often have
herbaceous, graminoid physiognomy. The physiog- broad, thin leaves, features that improve photosynthetic
nomy of the plant community is defined by the domi- efficiency and permit survival, growth, and reproduc-
nant plant species, but most communities also include tion at the low light intensities that exist below the tree
species that represent several other growth forms or- canopy. The lack of woody tissues means that virtually
ganized into one or more subordinate layers. For all net photosynthates can be used for feeding roots,
example, forest plant communities are generally com- leaves, and reproduction, but it places greater impor-
posed of five major layers (Figure 15-2): (1) trees— tance on cellular turgidity for the maintenance of plant
ar

Sea Chea
Figure 15-1 Structure of forest communities (opposite). (A) Black spruce stand in Quebec ex-
hibiting the characteristic structure of many boreal forests on mesic sites: dense, rather uniform
stocking with trees of uniform size, low tree species diversity, and an understory limited mainly to a
thick carpet of mosses. (B) Temperate deciduous hardwood forest: a second-growth stand on the Al-
legheny Plateau of northwestern Pennsylvania. Sugar and red maples, beech, black cherry, and yellow
poplar are the major species in an overstory of moderate diversity. Such stands are often multiaged
and multistoried, with a considerable understory. (Photo courtesy of D.A. Marquis; see Marquis, 1975,
1981.) (C) Left. Tropical rain forest near Manaus, Amazonas, showing the vertical complexity of the
vegetation (several canopy layers). Right. When the large emergent trees die or blow down, large gaps
are created, which creates the characteristic heterogeneity of the structure and composition of tropi-
cal rain forests.

© Epiphytic
lichens

@® Tree
layer

oy
G) Epiphytic 2a
mosses

@) Shrub
layer
bree
@) Herb i te YE,
layer. NG V sé 1
WV S
re 7,

ae 7 <a
Oe ZAM tend RR BSE Or en
AT GyVYteirccces TARO
as ly boyOT
. Sr
WY GWGHORA
"Wy. aioe
77777), Uy s LY
floor Yy
Yyy Uni YYVNR Up Y ji eistage
Mineral / IAW /XI i AX \
a y) ““COS
Y y Wy) v7) Y YYiy Y “WY OK)
soil
/ ¢ )

plant community. Each layer

Figure 15-2 The five major layers of live plants of a forested
climbers, may also be found.
can be subdivided into sublayers, or synusiae. A sixth layer, woody
here) may be an important nonliving
Standing dead trees and large decomposing logs (not shown
ies. Standing dead trees and large decomposing logs
structural component of some forest communit
t nonlivin g structura l compone nt of some forest communities.
(not shown here) may be an importan
409
410 CHAPTER15 Community Ecology

Table 15-1 Major Growth Forms of Terrestrial Plants often have only one or two layers: a layer of perennial
Trees (larger woody plants, mostly well above 3 m tall)
herbs in the grassland and a perennial shrub layer plus
Needle-leaved (mainly conifers—pine, spruce, larch, an ephemeral herb layer in the desert. As one moves
redwood, etc.) into more humid environments, a tree layer is added,
Broad-leaved evergreen (many tropical and subtropical and as moisture becomes increasingly abundant,
trees, mostly with medium-sized leaves) perennial herb and epiphyte layers are added. Thallo-
Evergreen-sclerophyll (with smaller, tough, evergreen
phytes may be present in almost any environment, al-
leaves)
Broad-leaved deciduous (leaves shed in temperate though they tend to become increasingly abundant
zone winter or in tropical dry season) along a transect from hot and dry to cool and moist
Thorn trees (armed with spines, in many cases with environments.
compound, deciduous leaves) Structural change along environmental gradients
Rosette trees (unbranched, with a crown of large
leaves—palms and tree-ferns)
results in characteristic combinations of growth
forms (cf. life-form spectra, Chapter 8) at different
Lianas (woody climbers or vines) points along the gradient. This occurs on all conti-
nents and results in broad divisions of a continental
Shrubs (smaller woody plants, mostly below 3 m in flora that are called plant formations and that are
height)
Needle-leaved
found on most of the large continents. Because major
Broad-leaved evergreen gradients such as temperature and moisture tend to
Broad-leaved deciduous run north-south on large continents, plant forma-
Evergreen-sclerophyll tions tend to occur as east-west bands across the con-
Rosette shrubs (yucca, agave, aloe, palmetto, etc.)
tinents. However, the east-west orientation is greatly
Stem succulents (cacti, certain euphorbias, etc.)
Thorn-shrubs
modified by mountains and deserts and also by the
Semishrubs (suffrutescent, i.e., with the upper parts of variation in climate between coastal and interior lo-
the stems and branches dying back in unfavorable cations on large continents. The temperate decidu-
seasons) ous forest of Europe is a different formation from the
Subshrubs or dwarf shrubs (low shrubs spreading near
temperate deciduous forest of North America, al-
the ground surface, less than 25 cm high)
Epiphytes (plants wholly above the ground surface, on though both belong to the same formation type (all
other plants) similar formations around the world are grouped in
Herbs (plants without perennial aboveground woody one formation type).
stems) A plant formation on a particular continent to-
Ferns
Graminoids (grasses, sedges, and other grass-like
gether with its associated animal and microbial com-
plants) munity and physical environment is called a biome, a
Forbs (herbs other than ferns and graminoids) group of ecosystems in which the primary producers
have similar growth forms and consumers have
Thallophytes broadly similar feeding habits. Similar biomes (those
Lichens
Mosses
with the same plant formation type) around the world
Liverworts form a biome type, such as the temperate deciduous for-
est biome type or the boreal forest biome type. The
Beard, 1973. physical environment (climate, topography, geology,
soils) of a biome is called a Jife zone.

shape and orientation. Consequently, most forbs are re- Community Structure as a Result of
stricted to moist soils, where they can maintain their Disturbance and Successional Development
turgidity. Good nutrition is necessary to permit net Community structure is not a static, unchanging charac-
photosynthesis to occur at low light intensities, and this teristic of the forest ecosystem. As we will see in Chapter
restricts most forbs to fertile sites. 17, forests are continually being subjected to distur-
The number oflayers of vegetation in a plant com- bance and are continually undergoing processes of
munity reflects the character of the physical environ- ecosystem recovery (called autogenic succession). Some-
ment. Dry environments such as grasslands or deserts times the disturbance is severe and large scale, removing
 Interactions Between Species ina Community 411

virtually all of the existing community over large areas. mice, salamanders, and worms. Snags and CWD may
Sometimes it is simply the death of individual large play a significant functional role in some forest ecosys-
canopy trees. Some disturbances occur frequently, and tems as well as their structural role, serving as a site for
others are “unusual” events. All combinations and low levels of nonsymbiotic nitrogen fixation and con-
intermediate disturbance regimes can be found. The di- tributing to the long-term supply of soil organic mat-
versity of natural disturbance regimes in the world’s ter. The ecological role of snags and CWD is not the
forests results in a wide diversity of community struc- same in all forest ecosystem types, of course; it must be
tures, from even-age, single-species stands of low plant evaluated for each different type of forest. Snag or
species and structural diversity to multiage, multi- CWD retention strategies must be site and stand type
species, and multicanopy-level forest structures. specific and relate to clearly identified objectives for the
Ecosystem recovery from disturbance is a major future structural characteristics of the forest. Large
cause of change in community structure over time. quantities of CWD in humid forests can accelerate the
Both the life forms and the structural arrangement of processes of soil acidification and podzolization and in
these life forms in the community change from the terms of soil chemistry and long-term site fertility
stand initiation, through the stem exclusion, to the might, therefore, be considered undesirable.
understory reinitiation phases of the development of a
stand after a stand-replacing disturbance (Figure
14-20). Structural change continues if the forest develops Importance of Community Structure Versus
through these phases in each of several tree-dominated Ecosystem Function
stages (seral stages) to the final or late successional stage.
As a response to forest management practices that sig-
The change in community structure as a result of
nificantly altered the structural attributes of forest
disturbance and recovery is vitally important in creat-
ecosystems and forested landscapes, forest ecology
ing habitat for a variety of animals. Each particular
and the “environmental debate” in the 1980s and
structure offers its own unique habitat features, and as
1990s became preoccupied with community structure,
we shall see later in this chapter, maintenance of a di-
sometimes to the point of excluding concern over
versity of community structures across the landscape is
other aspects of forest ecology. Having achieved the
important for the maintenance of high landscape-level
objective of focusing attention on the importance of
wildlife diversity.
structure for a variety of forest ecosystem values, this
debate has now moved on to a more ecosystem-based
Standing Dead Trees (Snags) and Large approach and an emphasis on ecosystem function. It is
Decomposing Logs as Components of generally recognized that if the function and succes-
Community Structure sional processes of ecosystems are sustained, then
other values, such as community structure, will be
Although community structure is largely determined
provided for. The lessons learned by a decade of pre-
by living plants, snags and decomposing logs on the occupation with forest structure should not be forgot-
ground (coarse woody debris [CWD)) also contribute
ten, however. They should simply be viewed in the
to community structure. Early, exploitative logging in dynamic ecosystem context in which structure exists.
North America generally left this dead structural mate-
rial in place, but the “tidier” and less “wasteful” timber
harvesting of the past 50 years has depleted many
forests of this structural component. Both snags and 15.3 Interactions Between
CWD provide a variety of wildlife habitat values. Snags Species in a Community
provide a source of wood-boring insects for woodpeck-
ers to feed on, a shelter or nesting site for various Certain types of relationships that exist between dif-
cavity-nesting birds and mammals (including bats), and ferent species were dealt with in Chapter 14 because
a perch for large birds such as eagles: CWD may pro- they have traditionally been thought of as major deter-
vide a “safe site” for the establishment of tree seedlings minants of population dynamics. There are many
on “brushy” sites, a display site for birds such as male other important types of interspecific relationships
grouse, and either winter or summer shelter for a wide that have not yet been mentioned, and they are exam-
variety of small animals, from winter wrens to deer ined in this section.
412 CHAPTER 15 Community Ecology

The biotic component of the environment of any review of species interactions and the consequent co-
individual organism is extremely important to it. In evolution, see Futuyama and Slatkin (1983) and
some environments, physical factors are dominant in Thompson (1982).
determining the characteristics of the biotic commu-
nity, but in most ecosystems, the organisms them-
selves and the way in which they interact are equally Symbiotic Interactions
important. The presence of another species may be The term symbiosis was coined in 1879 by a German
vital for food and/or shelter, or it may constitute a botanist, DeBarry, to describe the relationship be-
major threat in terms of disease, predation, parasitism, tween certain species of algae and fungi that live to-
or competition. gether to form lichens. He defined it as the living
Beneficial or harmful relationships may exist be- together of two dissimilar organisms in close associa-
tween organisms of similar size, as in the case of a tion or union. Sometimes the term is applied to all in-
Douglas-fir tree competing for light with a western terspecific interactions, as originally intended by
hemlock tree. Such relationships also occur between DeBarry (Whittaker, 1975), but it is commonly re-
organisms of vastly different size, as in the case of the stricted to those interactions that are predominantly
mycorrhizal relationship between a giant tree and a beneficial or that lack a negative effect on either of the
microscopic fungus or between a whale and the partners. The term is used in the latter sense in this
minute aquatic animal that it eats. Interspecific rela- book. Symbiosis defined in this manner can be subdi-
tionships that are not always readily apparent may be vided into two subtypes: mutualism and commensalism.
the major factor controlling the performance or even
the presence of a species in a particular ecosystem.
Interspecific interactions can be classified accord- Mutualism. Mutualism includes all interactions in
ing to whether they are harmful or beneficial, whether which both partners benefit. It includes a wide variety
they involve a continuous or an intermittent inter- of interactions, from permanent intimate contact be-
action, and whether they are vital (obligate) or not tween partners to situations in which there is no ac-
(facultative or opportunistic). The major types of in- tual physical contact.
teractions are shown in ‘Table 15—2. Frequently, a rela-
tionship may be intermediate between these classes, With Continuous Contact. ‘There are numerous
which merely represent a convenient division of what examples of continuous contact that will be well
is in reality a continuum of relationships. Also, rela- known to most readers. One of the best known is the
tionships can change from one class to another as time intimate association that develops when the cortex of
passes and conditions change. For example, mutualism the smallest order of secondary roots is invaded by
can grade into parasitism, and commensalism can specific fungi during periods of active root growth.
grade into physical exploitation or parasitism. For a Where the invasion of the cortex is intercellular, the as-
sociation is called an ectomycorrhiza; where it is
intracellular, the term endomycorrhiza is used. Where
both inter- and intracellular infection occurs, the rela-
tionship is called an ectendomycorrhiza. These associa-
Table 15-2 Types of Interspecific Interactions tions increase the solubility of minerals, improve
Effect on uptake of nutrients by the host plant, transport water
Category of Type of
to the root over ecologically significant distances, pro-
Interaction Interaction Species A Species B
tect the roots against pathogens, produce plant hor-
Symbiosis Mutualism + + mones, and move carbohydrates and nutrients from
Commensalism + oO one plant to another (Simard, 1997). As noted in
Antagonism Exploitation Chapters 5 and 11, mycorrhizal roots are the norm in
physical most plants. A mycorrhiza is not a permanent relation-
parasitism |
ship, most mycorrhizae being re-formed each year.
predation
Antibiosis, including + The mycorrhizal relationship develops only when
allelopathy both partners benefit. In a nutrient-rich soil, many
Competition - ~ trees will develop normal long roots with root hairs
 Interactions Between Species in a Community 413

rather than short roots with mycorrhizae: the tree does ergy supply from the alga. The algae of some lichens
not “need” the association. Trees that are growing ex- are of the nitrogen-fixing blue-green variety, which
tremely poorly may also lack mycorrhizae; the fungus provide the fungi with a supply of nitrogen in addition
cannot obtain sufficient carbohydrate from the root to to carbohydrates. The algae may be able to exist inde-
support the relationship, and it may become parasitic pendently, although this is unlikely because they are
on the root. Mycorrhizae have been called “the ulti- adapted to the lichen symbiosis. The fungus provides
mate in reciprocal parasitism” (Hacskaylo, 1972). If one them with shelter and improved availability of mois-
partner no longer needs the other, then the relationship ture and nutrients. Some lichens are known to repro-
changes from symbiotic to antagonistic. Reduced light duce only sexually in this manner. Others are known
intensity and stem girdling or heavy defoliation caused to reproduce only asexually; small fragments that con-
by insects or foliar diseases can also reduce the inci- tain both fungus and algae break off and are distrib-
dence of mycorrhizae, because these treatments reduce uted by wind. Some reproduce both ways. The lichen
the supply of carbohydrates to the root. association permits these two organisms (fungus and
The formation of mycorrhizae is a complex matter algae) to exist in environments, such as bare rock or
that depends on more than just the level of root carbo- the tops of trees, that are unsuitable for most other life
hydrate: the internal balance of inorganic nutrients forms. The relationship enables them to avoid compe-
and organic metabolites are also thought to be impor- tition and makes them a successful life form.
tant (Slankis, 1971). For some tree species (all pines The difficulty of obtaining nitrogen has led to the
and some species of the genera Quercus and Fagus), the development of many mutualistic relationships be-
relationship is generally obligate, and many attempts tween heterotrophic microbial nitrogen fixers and au-
to establish plantations of nonnative pines on nonfor- totrophic plants not capable of exploiting atmospheric
est soils in various parts of the world failed until the N;. The invasion of roots of leguminous plants by
soil was inoculated with the appropriate mycorrhizal bacteria of the genus Rhizobium and the subsequent
fungi. For example, slash pine introduced as seed into formation of root nodules has already been mentioned
Puerto Rico grew 30 cm in 3 years, was chlorotic, and (Chapter 5). The bacteria gain a shelter and plant car-
had few needles. Plants inoculated with mycorrhizal bohydrates as a source of energy, and the host plant
fungi were 244 cm tall and had a full complement of gains a secure source of nitrogen. This enables the
needles at the same age (Hacskaylo, 1972). It is now plant to grow in nitrogen-deficient environments and
recommended that when new forest nurseries are es- gives it a competitive edge over plants without such
tablished in areas that lack mycorrhizal fungi of the symbiotic associations. ‘he invasion of alder roots by
tree species being grown, the nursery soil be inocu- nitrogen-fixing actinomycetes and the formation of
lated with the appropriate fungus (Mikola, 1970). leaf nodules were mentioned in earlier chapters.
A mycorrhiza is a mutualistic relationship that ex- Another important example from the microbial
ists only when there is continuous contact, although world is the relationship between intestinal bacteria
both partners can, under certain circumstances, exist and the host animal (Howard, 1967). Ruminant herbi-
alone. This is also true of the mutualistic relationship vores are able to exist on woody plant species because
that we call a /ichen. These life forms do not represent bacteria in their ruminant stomachs alter the cellulose
individual genetic packages as do life forms such as a to chemical forms that the host can digest. By convert-
ponderosa pine or a black-tailed deer. Despite that we ing a lot of the organic matter to CO, and methane
classify lichens with generic and specific names as we (the flatulence for which such animals are renowned
do plants and animals, they are in fact nothing more [see Chapter 5]), these bacteria increase the concen-
than a mutualistic relationship between particular tration of nitrogen in the nongaseous material passing
species of fungus and algae. When crustose lichens re- through the animals’ intestine, thereby facilitating its
produce, they release two different types of propagule: utilization. Insects also depend on gut flora to enable
fungal spores and algal spores. If the former land on a them to survive on nutritionally marginal material.
suitable substrate, then the spores germinate and de- This is especially important in wood-boring species.
velop a small mat of mycelium. This mycelium will Their gut flora are able to synthesize amino acids and
perish unless spores of an appropriate species of alga other essential organic nutrients, enabling the host in-
land on and become entrapped in the mycelium; the sects to survive on a diet that is lacking a number ofes-
fungus is a heterotroph and is adapted to obtain its en- sential constituents.
414 CHAPTER15 Community Ecology

Without Continuous Contact. Many examples of spread through the wood, softening it, and enriching
a relationship without continuous contact can be it with organic nitrogen. By the time the beetle larvae
found in the animal kingdom, but it also occurs be- hatch, they are surrounded by a zone of wood that
tween other life forms. contains fungal mycelia, which also lines the tunnels in
The literature is replete with fascinating examples of which the larvae feed. This is an ideal environment
mutualistic relationships between animals. For example, and food supply in which to mature, and the presence
many large ungulate herbivores permit certain species of of fungal spores in their food and their environment
birds to land on them to feed on external parasites such ensures that the mycangia of the developing larvae be-
as ticks. The cowbird of North America and the little come suitably charged with spores so that the next
white heron in Africa both feed on such external para- generation will be equally well supplied.
sites to the mutual benefit of both bird and ungulate. Janzen (1966) described a fascinating case of
Many different symbiotic relationships have been ant—plant mutualism between the ant Pseudomyrex fer-
reported between ants and various other life forms. ruginea and the Bull’s-horn acacia (Acacia cornigera) in
Most gardeners are familiar with the solicitous atten- Central America. The ants protect the acacia plants
tion given to aphids by ants on rose bushes. The ants from insect pests and reduce competition by other
protect the aphids from parasites and predators and in plants; in exchange, the ants obtain food and a nesting
turn obtain a supply of “honeydew” from the aphids. site from the plant. To start a colony, the queen ant
This is the liquid excreta of the aphids: essentially bores a hole in one of the swollen acacia thorns and
plant sap plus aphid excreta less the nutrients that have hollows out its soft interior to make space for eggs and
been removed from the sap by the aphids’ digestive larvae. When the first thorn is filled, the next thorn is
system. The degree of contact between the ants and used. Food is “provided” by the plant in the form of
aphids varies. In some species, the aphids are kept nectaries at the base of the leaves and nutritious nod-
right in the ants’ nest, feeding on plant roots. In oth- ules that form at the tips of the leaves.
ers, the aphids may feed aboveground but are carried In some ant-acacia associations, both partners
to the nest and to other plants by the ants. Sometimes, can exist independently, although both benefit from
the ants merely attend the aphid population where the relationship. The Pseudomyrex/Acacia symbiosis,
they find it on the plant. Aphid eggs may be carried to however, is obligate and borders on being a case of
the nest for the winter, and the young aphids that continuous-contact mutualism. ‘To sustain the intensity
hatch in the spring are used to repopulate host plants. of the relationship, both partners have developed spe-
The parallel between the human farmer/domestic cial adaptations. For their part, the ants are active 24
livestock relationship and the ant/aphid relationship is hours a day to provide continuous protection; this is
remarkable. No less remarkable is the parallel between unusual for ants, which normally have a diurnal behav-
the methods used by farmers of arable land and the ior pattern. The acacia is evergreen, which ensures a
cultivation of fungi by certain beetles, ants, and ter- continuous supply of food for the ants; most closely re-
mites (Batra and Batra, 1967). Tropical leaf-cutting lated species of acacia are deciduous in the dry season.
ants (Atta sp.) cut up green leaves and carry the pieces To demonstrate the degree of protection pro-
to their nests. There they are chewed up and the re- vided by the ants, Janzen prevented ant colonization
sulting pulp is spread out in special underground of acacia shoots by means of ant poison and then
chambers to form a substrate on which spores of a par- measured growth and survival of shoots with and
ticular species of fungus are planted. The developing without ants. Table 15-3 demonstrates that the sym-
mycelium is carefully cultivated by the ants, which re- biosis is highly beneficial to both growth and survival
move other species of fungi and prevent the fruiting of of the acacia shoots.
the desired fungus because they feed on the mycelium Insect pollination is another well-known example
(Brues, 1946). of mutualism without continuous contact. Flowering
Many wood-boring beetles have a special morpho- plants have evolved nectaries with showy flowers that
logical adaptation (called a mycangium) located at are often equipped with ultraviolet reflecting lines to
some part of their body in which wood-rotting fungal guide insects to them. These plants invest a consider-
spores accumulate (Francke-Grosmann, 1967). When able part of their energy budget in the production of
the beetles invade a new tree to breed, they excavate a flowers, pollen, and nectar to gain the advantage of
tunnel in which to lay their eggs, infecting the area between-plant transfer of pollen. The most highly de-
with fungal spores as they work. These germinate and veloped plant-pollinator relationships are found be-
 Interactions Between Species in a Community 415

Table 15-3 Effect of the Ant-Acacia Symbiosis on the tionship evolves over a long period, then the affair will
Growth and Survival of Bull’s-Horn Acacia in Mexico develop into mutualism.
Parameter With Ants Without Ants A common example of commensalism is the non-
parasitic growth of one type of plant on another. Some
Growth of shoots types of forest are characterized by tree crowns fes-
May 25-Aug. 3, cm 104 16
tooned with epiphytic mosses and/or lichens. By
Survival of stumps growing in the crowns, these life forms avoid competi-
over 10-mo period, % 72 44 tion from herbs and shrubs on the forest floor, benefit
Mean weight of shoots from the nutrients in the throughfall of the host, and
per stump after 10 mo
are held up to the light. The host experiences virtually
of growth, g 579 44
no loss where the growth of the epiphytes is modest or
Mean number of leaves where it occurs on the stem. If, however, the develop-
per stump after 10 mo
of growth 108 52
ment of crown epiphytes becomes excessive, then a
tree may experience some loss of light to its foliage,
Mean number of swollen
thorns per stump after
and one sometimes observes a tree in obviously poor
10 mo of growth 104 39 health that is smothered in lichens. The conclusion
that the epiphyte has become antagonistic may not be
Data from Janzen, 1966. warranted, however. It is just as likely that the epi-
phyte is responding to increased availability of light as
a result of an independently caused pathological con-
tween the orchid family and euglossine bees. So dition in its host. Also, some epiphytes are nitrogen
highly developed is this relationship that the flowers fixers, and as they die and become litterfall, they con-
no longer have nectaries and they are visited only by tribute to the nitrogen economy of the trees. What
male bees. The flowers have such specialized perfumes seems to be a commensal relationship thus may in re-
that in many cases only one species of bee is attracted ality be mutualistic. It may well be that many examples
(Dodson et al., 1969; Dressler, 1968). Similarly, the of commensalism are merely examples of mutualism
peculiarly enclosed flowers of the commercial fig are that we do not yet fully understand.
pollinated only by wasps of the genus Blastophaga, Examples of apparent commensalism are common
which lay their eggs in special floral structures. The among animals. Scavenging hyenas and vultures that
complexities of the fig—wasp mutualism have been re- follow the large African predators benefit from the
viewed by Janzen (1979) and Wiebes (1979). Yet an- leftovers of a kill but contribute little to the predators’
other obligate and specific symbiosis is found between economy. Similarly, remora fish, which attach them-
the yucca plant and the yucca moth (Hartzell, 1967). selves by a sucker to sharks and whales, benefit in
The female moth visits the yucca flower in the evening terms of transportation and protection and may obtain
and collects a ball of pollen from the anthers. Then, scraps of food left over as the host feeds. They are
holding the pollen ball in specially adapted mouth small enough that their presence does not seem to
parts, she flies to another plant, pierces the ovary of hinder their powerful hosts.
the flower with her ovipositor, lays her eggs, and then Many animals seek shelter in the nests of others.
deposits the pollen ball into the stigma. As with so Some species of birds inhabit the abandoned nests of
many mutualistic symbioses, the evolutionary devel- others, and there are numerous examples of small ani-
opment of such a relationship taxes the imagination. mals sharing the abodes of larger animals. An example
of a more intimate shelter relationship is seen in the
Commensalism. When the relationship between case of the intestinal flora of humans (including
two different species benefits one partner (guest) but Escherichia coli), which serves no known beneficial
neither benefits nor harms the other (host), it is function but neither does it do much harm in healthy
termed a commensal symbiosis, or commensalism. As with individuals. Such flora may be an evolutionary “left-
mutualism, the commensal relationship is not a fixed over” from our recent evolutionary past when our diet
one. If the guest partner becomes too prolific, then it contained large proportions of plant material of low
may adversely affect the host partner and the relation- nutritional value. Under conditions of disease, stress,
ship may become antagonistic. Alternatively, if some or injury in infants or in older debilitated individuals,
advantage to the host develops as the commensal rela- E. coli can produce intestinal diseases or enteritis.
416 CHAPTERI15 Community Ecology

E. coli thus is usually nonpathogenic but opportunistic; temperate ecosystems, where plants such as ivy (Hedera
it will produce disease if it gains access to susceptible helix) and Clematis sp. exploit trees and bushes (Figure
tissues or organs, such as the bladder, where it causes 15-3). Initially, the climbers may act as commensals,
cystitis. In many cases, these shelter arrangements are but sometimes they grow over and kill their host.
probably truly commensal, but in some, the host may Physical exploitation is seen in several species of
benefit by the scavenging activities of the guest, which birds. The antarctic penguin will steal nest-building
serves to “keep the house clean.” materials and may even steal unattended eggs. Pirati-
Vegetation in general participates in an important cal behavior is seen in birds such as eagles, skuas, and
commensal relationship with animals by providing jaegers, which will attack lesser birds of prey that are
shelter. Many animals require shelter to feel secure, carrying fish, forcing them to release their capture and
and they obtain this from plants without affecting then consuming it themselves. Hijacking of food is
them in any other way. Documented examples of such also seen in the insect world. Tropical flies of the
a symbiosis are far too numerous to mention here. genus Bengalia will lie in wait for ants returning from a
The interested reader is referred to Cheng (1970), hunting raid on a termite nest. They attack ants carry-
Henry (1967), and Jennings and Lee (1975). ing dead termites, causing them to drop their booty,
which the fly promptly consumes (Clarke, 1976).
That harbinger of springtime in the English
Antagonistic Interactions woodland, the cuckoo bird (Cuculus canorus), exploits
smaller birds in a particularly insidious way. After mat-
All relationships in which at least one ofthe partners is
ing, the female cuckoo seeks out the nest of a smaller
adversely affected are included in the general category
bird in which it lays an egg. This generally occurs after
of antagonism. Such relationships play a major role in
the nest owner has already laid a clutch, possibly be-
determining the abundance and distribution of species
populations and the diversity of species in a communi-
cause the large cuckoo must feed for longer in the
ty. They are also important in the evolution of species
spring to generate its comparatively large eggs (Lack,
characteristics. 1966). The host species may recognize the cuckoo’s
Antagonistic relationships can be subdivided into egg as an interloper and eject it from the nest, but fre-
(1) nonconsumptive physical exploitation, (2) consump- quently this does not happen and the cuckoo egg is in-
tive physical exploitation (including parasitism and pre- cubated and hatched. The host bird dutifully feeds the
dation), (3) antibiosis (including allelopathy), and (4)
resulting chick, which may ungratefully dispatch its
competition. The first three types generally involve nest mates by tipping them out of the nest or may con-
benefit to one partner and harm to the other. The tribute to their starvation because the smaller chicks of
fourth type involves mutually adverse effects. the host species cannot compete with the vociferous
cuckoo chick for the attention of the parent bird at
feeding time. The cuckoo chick quickly outgrows the
Physical Exploitation
hardworking but unwitting foster parents, which
Nonconsumptive Physical Exploitation. Many forest faithfully feed it until it is ready to fly.
plants invest a considerable portion of their energy This type of relationship, which is also found in
budget in building a strong stem with which to hold up North American cowbirds (Molothrus sp.), is often
their foliage to the light and to resist the damaging ef- called brood parasitism and could just as easily have
fects of wind. If it were not for this evolutionary strate- been discussed below under parasitism (consumptive
gy, we would have neither forests nor a forest industry, physical exploitation). The cuckoo certainly diverts
and the human species might never have evolved. Not energy from the host population, weakening and
all plants solve their problems this way, however. killing some of that population. However, brood para-
Shade-tolerant herbs get by with lower light intensities sitism is included here because parasites are normally
by having large shade leaves, by operating seasonally, thought of as organisms that actually consume part of
and by saving energy by dispensing with woody peren- the living body tissues of their host.
nial stems. Climbing plants also dispense with strong A similar type of relationship but with an interest-
woody stems by using other plants for physical support. ing twist to it exists in the cowbird in Panama. In
This adaptation is well developed in tropical forests, this region, the giant cowbird is a brood parasite of
where the trees are often draped with lianas and other another bird, the chestnut-headed oropendula, which
climbing plants (Figure 15—1C), but it is also seen in builds a sack-like nest that hangs from the branches of
 Interactions Between Species in a Community 417

Ms

Figure 15-3 Physical exploitation. (A) Ivy (Hedera helix) climbing on and smothering ash trees
(Fraxinus excellsior) in southern England. (B) Tropical climbers smothering trees in Hawaii.

large trees. Natural selection has acted on species that bot flies from laying eggs on them), and open their
are subject to brood parasitism so that they are able to eyes within 48 hours. The active cowbird nestlings
recognize the difference between their own eggs and snap at anything small that moves in the nest, includ-
those of a brood parasite. However, selection has also ing the adult bot flies, and they also remove larvae
acted on the parasite to produce eggs that mimic those from the skin of the oropendulas (Table 15-4). The
of the host so well that they avoid detection. In a brood parasite in this case has both a mutualistic rela-
study of the oropendula—cowbird relationship in the tionship and an exploitive relationship, which appar-
Panama Canal zone, a curious phenomenon was dis- ently compensate for each other in natural selection.
covered (Smith, 1968). It was found that in some
oropendula colonies, the cowbird eggs were mimics, Table 15—4 Relationship Between the Incidence of
whereas in others, the eggs of the two species were Bot Fly Parasitism of Oropendula Nestlings and the
distinctly different. The explanation for this enigma Presence of Cowbird Chicks in the Nest
was found in two additional relationships: a parasitic Number (%) of
one and a commensal one. Oropendula Nestlings
Oropendula chicks are subject to parasitism in the
With Bot Fly Without Bot Fly
nest by bot flies, which lay their eggs on the naked Cowbird Nestlings Parasites Parasites
skin of the newly hatched birds. The chicks are unable
to groom themselves to remove these parasites be- Present 57 (8.4) 619 (91.6)
cause their eyes do not open for 6 to 9 days after Absent 382 (90.1) 42 (9.9)
hatching. The cowbird chicks hatch 5 to 7 days earlier,
are born with a covering of down (which deters the Data from Smith, 1968.
418 | CHAPTER15 Community Ecology

Consequently, there is no selective advantage to the natural selection develops host resistance and reduces
oropendulas in discriminating against cowbird eggs parasite virulence (coevolution; Chapter 16) to the point
where there are bot flies, and in such areas, egg mim- at which a stable relationship is established or restored
icry has not been selected for. (the endemic phase). An example of this coevolution in
If this is the correct explanation, then why has se- parasitic relationships is the malarial parasite, which,
lection favored cowbird egg mimics in other oropen- in the absence of medication, can be fatal to people
dula colonies? Smith discovered that all oropendula from nonmalarial areas. Human populations exposed
colonies in which the birds discriminated between to malaria for many generations have a high incidence
their own and other eggs and ejected nonconforming of sickle cells in their blood, which are thought to con-
objects were close to nests of bees or wasps. Bot flies fer some resistance to the parasite.
rarely approach close to the nests of these insects, the There are several interesting examples of this type of
proximity of which confers on the oropendulas an im- interaction from the world of forestry. Before 1900, the
munity against bot fly parasitism (i.e., there is a com- American chestnut was a major species in the eastern
mensal relationship between the bees/wasps and the hardwood forest of the United States. In the early 1900s,
oropendulas). This immunity in turn eliminates the a fungus (Endothica parasitica, the chestnut blight) was
selective advantage of having a cowbird in the nest and accidentally introduced from Asia, where it existed as an
makes its presence exploitive rather than mutualistic. endemic parasite. Within 20 years, most of the mature
The study of the cowbird—oropendula relationship chestnut trees in New England were dead, and by the
is an elegant demonstration of the need to recognize 1940s, the same fate had befallen the chestnut all the
variations in intraspecific relationships if we are to un- way to the southern end of its distribution in the south-
derstand the form and function of ecosystems. ern Appalachians. Chestnut still sprouts from the old
root stocks but never survives long enough to become a
Consumptive Physical Exploitation. Consump- canopy tree. It is likely that if the species can continue to
tive physical exploitation includes relationships in survive as a shrub, then a more stable endemic relation-
which (1) one organism consumes part of the blood or ship that will result in the return of this species as a
tissues of a host organism, often weakening and some- major tree in the region will eventually evolve.
times killing it (parasitism); (2) one animal kills and A similar example is found in the history of the
consumes all or part of another animal (predation); Dutch elm disease, caused by the fungus Ceratocystis
and (3) an herbivore consumes all or part of a plant, ulmi. Vhis fungus, which is endemic on elms in Eu-
possibly killing or weakening it. rope, where it causes only occasional branch mortality,
Essentially, this category of antagonism accelerates is spread by a bark beetle with which it has a mutualis-
the loss of energy and biomass from the individual or- tic relationship. Logs that contain live bark beetles
ganism. were shipped to the United States in about 1930. The
beetles rapidly spread the disease to the native North
Parasitism. As with many interspecific interactions, American elms, which proved to be highly susceptible
there is no clearcut distinction between parasites and and which have been virtually eliminated from much of
predators. Parasites are normally very much smaller the eastern United States and Canada. During this pe-
than their hosts and generally exist in more or less riod of epidemic conditions, the fungus apparently
continuous contact with them. When a host—parasite evolved a more virulent strain, which was then acciden-
relationship has existed for many thousands of genera- tally introduced back into England in the early 1970s.
tions of the host, the parasite is rarely lethal, although There, the U.S. experience was repeated. Spreading
if its population becomes too high, it may be so debil- out in concentric rings from the points of introduction
itating that its host succumbs to other mortality fac- (ports into which beetle-infested logs were imported),
tors. The situation is different when a new the virulent disease eliminated most of the 5 million
host—parasite relationship is established. When a plant hedgerow and ornamental elms in southern England
parasite has recently been introduced to an area previ- (Figure 15—4). Eventually, the elm population will re-
ously outside its range or a new, more virulent geno- cover as an endemic relationship evolves.
type of a pathogen has evolved, the relationship may
well be fatal. This early stage of a parasite relationship, Predation. In contrast to typical parasites, typical
which is called the epidemic phase, is transient because predators are often larger than their prey (or at least
 Interactions Between Species ina Community 419

ondary carnivores. Soil fauna prey on soil fungi in

much the same way that herbivores prey on plants, but
here again the hunter can become the hunted. Certain
species of soil fungi are predaceous on soil fauna much
larger than themselves. One species of fungus pro-
duces a chain of cells in the form of a loop. Should a
nematode or other soil microorganism insert part of its
body through the loop, the cells contract, tightening
the loop sufficiently to kill the victim (Bessey, 1950).
The general public often views predators as “red in
tooth and claw”: as vicious, savage, and undesirable. In
most cases, this is a misconception. Predators gen-
erally act to remove the sick, the maimed, and the dying
Figure 15-4 Elm trees (Ulmus sp.) in southeastern United from the prey population. They help to keep the popu-
Kingdom killed by Dutch elm disease after the introduc- lation below its carrying capacity and thereby prevent
tion of a virulent strain from the United States. The photo- the carrying capacity from being degraded. Certainly,
graph was taken in midsummer. there are examples of wanton killing by predators, but
this usually represents situations in which the normal
adaptations between predator and prey have been upset
not many orders of magnitude smaller), are free living, by human action or some other environmental factor.
and generally have only a single successful encounter Killing of domestic livestock by large predators or of
with their “host” (prey or victim would be a more ap- deer by packs of semi-wild domestic dogs represent un-
propriate epithet). The predator is sometimes said to natural situations that cannot be compared with natural
live on “the capital” of its food resource, whereas a predator-prey relationships. However, there are cases
parasite utilizes “the interest.” Although _ this in which prey populations have been depressed by some
metaphor has some reality in terms of the individual mortality factor to the point at which the population is
host, it loses its currency when applied to host-prey prevented from increasing again by its predators once
populations in general. Parasites can so weaken host the action of this initial factor is removed. ‘The popula-
individuals that the population of which they are tion is below its escape density (Chapter 14) in terms of
members goes into decline, whereas predators may its predators—a situation referred to as a predator pit.
take only diseased and old individuals or those that are Over-fishing of ocean fish stocks that are normally
in excess of the carrying capacity. By so doing, preda- above their predation escape density can push the fish
tors may actually sustain the health and abundance of into a predator pit from which recovery is very low even
the prey population. if fishing is stopped.
As with a parasitic relationship, a recently developed
predator-prey relationship can be unstable. When a Antibiosis. Many if not all organisms interact with
predator population is introduced to a prey population each other chemically as well as physically. Such
that has been without predators for some time, its activ- chemical interactions occur between microbial life
ities can virtually decimate its prey (this is generally the forms, between plants and animals, between different
basis for biological control programs). If both popula- species of animals, and between different species of
tions survive through several predator-prey cycles, then plants. Such interactions are broadly referred to as
a stable relationship may be established. antibiosis. Vhere are many well-known examples of in-
Predation can occur between a wide variety of or- teraction between microbes. Perhaps one of the best
ganisms. Herbivores are normally considered to be examples is the production of penicillin by fungi (be-
predators and plants are the prey, but the tables are longing to the genus Penicillium), which inhibits the
turned in plants such as sundew (Drosera sp.) and the erowth of bacteria in the vicinity of these fungi. There
pitcher plant (Sarracenia sp.), which are insectivorous; is frequently intense chemical competition between
the plants are the predator and insects are the prey. different species of crustose lichens invading rock sur-
Herbivores of various types are the prey for primary faces and between the fungi and other microbes oc-
carnivores, which in turn become the prey for sec- cupying the soil. These organisms use antibiotic
420 CHAPTER15 Community Ecology

substances to influence the growth of competing Allelopathy (Antibiosis Between Plants). Another
species. For example, in a study of interactions be- well-known case of antibiosis is that which occurs be-
tween actinomycetes and fungi isolated from soil and tween plants. The metabolism of plants is extremely
tree seedlings, antagonistic effects were found in the complex, involving a very large number of organic
majority of more than 200 different combinations that molecules involved in a very large number of pro-
were examined (Vaartaja and Salisbury, 1965). Many cesses. With such chemical complexity, it should not
plant species produce antibiotics that inhibit the be surprising that evolution has failed to fine-tune
growth of microbes (Stoessl, 1970). Members of the plant biochemical systems to the point at which there
family Cupressaceace, which are well known for their are no waste products or metabolic “leftovers.” Many
durability and decay resistance, produce a substance, of these secondary plant chemicals are of no apparent
f-thujaplicin, which inhibits the growth of a wide vari- value to the plant, but some are because they modify
ety of fungal species and is fungicidal to several species the growth and behavior of other organisms to the
(Trust and Coombs, 1973). benefit of the plant. Chemicals that inhibit the germi-
Many plants contain chemicals that make them un- nation, growth, or occurrence of other plants are re-
palatable to herbivores. Larvae of the winter moth can ferred to as allelochemicals and the phenomenon as
eat only young oak leaves because old oak leaves contain allelochemical antibiosis, or a//elopathy.
high levels of tannin, which render the leaves unpalat- Examples of allelopathy have been reported since
able (Feeny, 1968, 1970). Trees that are injured by insect the 19th century. Perhaps the best known is the exam-
feeding may produce polyphenols that can inhibit the ple of walnut trees (Juglans sp.). It has long been ob-
population growth of the insects. Altered polyphenol served that very few species of herbs and shrubs will
metabolism was noted in Scots pine attacked by the Eu- grow beneath a canopy of walnut trees and that this is
ropean sawfly (Thielges, 1968), and there is a growing not the result of mere competition for light, moisture,
body of evidence that this may be a general phenome- and nutrients. The walnut tree produces an allelo-
non (e.g., Haukioja, 1980). Willows that are fed on by chemical called juglone, which occurs in a water-solu-
snowshoe hares in Alaska become unpalatable by pro- ble, nontoxic form (hydroxyjuglone) in leaves, fruits,
ducing secondary chemicals, and this may be a part of and other tissues. This is washed into the soil by rain,
the mechanism of the 10-year cycle (Bryant, 1986; where it is oxidized to juglone, which inhibits the ger-
Bryant et al., 1983; Bryant and Kuropat, 1980). Where mination and growth of many other plant species.
activities of one herbivore stimulate plant defense mech- A somewhat different case of allelopathy is found
anisms, other herbivores that feed on the same plant in the hot, dry climate of southern California. Grass-
may be adversely affected, leading to some degree of lands in this area are invaded by a community of low
synchronization in their population — fluctuations shrubs known as soft chaparral. This community is
(Haukioja, 1980; Kimmins, 1970). It has been suggested dominated by a sage brush (Artemesia californica) and a
that when some species of plant are attacked by herbi- mint (Salvia leucophylla), both of which produce
vores, they may produce volatile chemicals that stimu- volatile allelochemicals that fill the air above the com-
late the chemical defenses of neighboring plants: a plant munity with a characteristic fragrance. Experiments
“early warning system” (Baldwin and Schultz, 1983). have shown that the characteristic absence of grasses
Interaction and chemical coevolution between and herbs from the areas occupied by and in a belt 1 to
plants and herbivores may have been a major factor in 2 m surrounding the shrub clumps is not the result of
the evolution of high species diversity in terrestrial competition, water-soluble allelochemicals, or the ef-
ecosystems (Erlich and Raven, 1967). Any explanation fects of animals but that it is the result of the
of the great diversity of species in tropical rain forests volatilization of terpenes (camphor and cineole) from
must include chemical adaptations of plants to herbi- the shrubs into the surrounding air. These terpenes
vores, and evolution of plant diversity in response to have a marked inhibitory effect on germination and
the chemical effects of plants on each other is proba- growth of seedlings and also on soil bacteria. The ter-
bly one of the major explanations for the evolution of penes are absorbed onto clay particles in the soil sur-
species diversity higher in the food web. For a treat- face, rendering the soil unsuitable for the growth of
ment of the interaction between herbivores and plant other plants until the chemicals are either driven off
secondary chemicals, see Rosenthal and Janzen (1982). by fire or washed away by rain. In wet years, there is an
 Interactions Between Species ina Community 421

increased abundance of other plants in the belts ability inhibit nitrogen-fixing bacteria in the soil,
around the shrubs, whereas the wet season after a fire which in turn retards the invasion of the area by nitro-
in chaparral is characterized by the conspicuous gen-demanding species (Rice, 1964, 1965). Pioneer
blooming of annual plants that were mostly absent be- plants may also interfere with symbiotic nitrogen fixa-
fore the fire (Muller, 1966, 1969; Muller et al., 1968). tion (Rice, 1967, 1968). This effectively reduces com-
Allelochemicals act in a wide variety of ways. They petition and prolongs the occupancy of the area by the
may inhibit the germination of seeds, inhibit the activ- pioneer species.
ity of nitrogen-fixing bacteria, or interfere with the It is probable that allelopathy has been selected for
formation of mycorrhizae. For example, heather in many pioneer species. Crustose lichens that colo-
(Calluna vulgaris), which makes the Scottish moors fa- nize bare rock compete with each other using sec-
mous, releases substances that can inhibit the forma- ondary chemicals, and the persistence of some pioneer
tion of mycorrhizae on the roots of some trees. Early mosses may have some chemical basis. Climax species
attempts to afforest these moors with Sitka spruce met may also prevent the invasion of pioneer species by al-
with failure because the planted seedlings grew ex- lelopathy. Root growth in yellow birch seedlings (pio-
ceedingly slowly. Physical removal of the heather re- neer species) has been shown to be inhibited by the
sulted in the appearance of mycorrhizal fungi and the presence of maple seedlings (climax species), the inhi-
development of increased vigor in the trees (Handley, bition being caused by substances leached out of the
1963). Plowing, which turns over a sod and kills the maple root tips. The difficulty of regenerating yellow
underlying heather, had a similar beneficial effect. In birch in areas occupied by sugar maple may well be the
contrast to the effect on the introduced spruce, result of these allelopathic effects (Tubbs, 1973).
heather has much less effect on the native Scots pine The allelopathic effects of a species are not fixed in
(Malcolm, 1975), and the presence of pines improves time. Other species adapt, and eventually the allelo-
the growth of planted spruce. This difference has been pathic effect may become less potent. For example,
related to differences in mycorrhizal fungi between stands of a given species of Eucalyptus in the United
pine and spruce when they are grown as monoculture States sometimes contain almost no other plants,
plantations on these moorlands and between spruce whereas stands of the same species in its native Aus-
grown in monoculture and spruce grown in mixture tralia have a well-developed understory. The inference
with pine or birch. is that there has been sufficient evolutionary time in
Allelopathy has been invoked in many parts of the Australia for minor vegetation species to develop tol-
world to explain the widespread observation that erance to the eucalypt allelochemicals, whereas U.S.
growth of coniferous evergreen trees is often poor in plants have not had time to adapt to this recently in-
the presence of a vigorous community of ericaceous troduced species. Similarly, a species of buttercup
shrubs (de Montigny, 1992; de Montigny and Weet- (Ranunculus testiculatus), a native of southeastern Eu-
man, 1990). This has been reported for eastern Can- rope and central Asia, excluded all other species over
ada (Damann, 1971; Titus et al., 1995), western large areas when it was introduced to dry areas of
Canada (Fraser, 1993; Messier and Kimmins, 1990; Utah and the interior of Oregon and Washington.
Prescott and Weetman, 1994), and the heathlands of ‘Tests showed that allelochemicals from the buttercup
Europe (Malcolm, 1975). It is believed that this poor strongly inhibited the germination of native forbs and
growth is related to low levels of availability of nitro- grasses. [The effect was greatest on sandy soil and least
gen and possibly also of phosphorus or to the failure of on fine-textured soils, in which the allelochemicals
the conifers to access those nutrients in the soil. How- are presumably absorbed and rendered ineffective
ever, in most cases, it has been difficult to quantify the (Buchanan et al., 1978).
separate contributions of allelopathy, root competi- Allelopathic effects of overstory dominants on un-
tion (ericaceous shrubs generally produce a dense fi- derstory plants are thought to contribute to the patchy
brous fine root system), and low quality of litterfall in distribution of minor vegetation that is characteristic
producing this low nutrient availability. of forests of mixed-species dominance. Understory
Interference with nitrogen-fixing bacteria by alle- species also experience allelochemicals produced by
lochemicals has been observed. The decay products of other understory species. It has been reported that a
some grasses that are tolerant of low nitrogen avail- composite, Hyoseris scabra, grows in the presence of a
422 CHAPTER 15 Community Ecology

wild onion (Allium chamaemoly) but only when a third This equation must be further modified when we have
species, Bellis annua, is also present. The latter miti- populations of two species competing for the same re-
gates for itself and the Hyoseris the allelopathic effects sources. For species 1, we can rewrite the equation to in-
of the onion (Whittaker, 1970). clude —N,, the negative effect of individuals of species 2
Allelopathy is not a peculiarity of a few species but on the carrying capacity of the environment for species
a widespread and normal, although generally incon- 1, either because of competition for a common resource
spicuous, phenomenon of natural communities. Dra- or because of interference in the use of a resource:
matic cases of allelopathy are easy to recognize dN K,-N-N
(Muller, 1968). In a much wider range of cases, allelo- Teo oe a ae
chemical effects are less obvious but may, nevertheless,
make significant contributions to plant competition. or

Allelopathic effects are probably far more impor- dN Ny as

Sap anes, 1 RCA eIER
tant in both forestry and agriculture (Tukey, 1969)
than is generally realized. Most home gardeners will
where 1; is the per capita rate of population increase of
recognize the adverse effects of tomato plants on or-
species 1, N, is the number of individuals of species 1,
namental plants, of ivy if it is allowed to climb over
K, is the carrying capacity of the environment for
other plants, and of various species of conifers in the
species 1, N, is the number of individuals of species 2,
area beneath their crowns, effects that cannot be ex-
N,/K, is the portion of the carrying capacity for
plained in terms of simple competition for space, light,
species 1 used up by Nj, and N,/K;, is the portion of
water, nutrients, and so on. Similarly, understanding
the carrying capacity for species 1 used up by N).
how forest ecosystems operate must involve recogni-
Thus, the greater the number of individuals of species
tion of the chemical interactions of species. However,
as noted above, caution is needed. Some of the re-
2 competing with species 1, the lower the value of
(K, — N, — N,)/K, and the lower the rate of population
ported examples of allelopathy may also be explainable
in terms of competition or of allelopathy acting in increase for species 1. Similarly, for species 2,
combination with competition and other mechanisms. dN _ N. Kk, -N, = INE

ie Oe K
Competition. Interspecific competition occurs
or
wherever two different species attempt to utilize the
same resource when that resource is in limited supply. dN; N, N,
= 75 N35 Shs eS a oe
There is no competition when they share a common dt iv Kosirnniiss
resource that exceeds their combined demands on it.
The greater the number of N, individuals, the less the
Competition can also occur when the resource is not
remaining unused carrying capacity for species 2 and
in short supply but the two species interfere with each
the lower its population growth rate.
other’s use of it.
This formulation is reasonable as long as both
We saw in Chapter 14 how growth equations have
species place the same demands on the resource and
been used to describe population growth, including
the ability to exploit the resource is the same for indi-
intraspecific competition. The Lotka—Volterra logistic
viduals of both species. If they differ, then a further
equations that were discussed can be extended to de-
modification must be made. Rather than reducing the
scribe interspecific competition and used to predict the
carrying capacity for species 1 by the number of species
probable outcome of such interactions.
2, we must reduce it by the number of species 2 ex-
When we recognize that environments have a car-
pressed in terms of the equivalent number of species 1.
rying capacity and that individuals in the population
For example, if species 2 eats twice as much as species
compete for the resources that determine that capac-
1 (.e., N, = 2Nj), then we would reduce K, by -2N
ity, the simple growth equation
rather than by —N). To generalize, we can say that
dN
N, = aN, and N, — BN,
becomes

where B = N,/Nj, or the ratio of resource use by

species | to resource use by species 2, and B = N,/N,,
 Interactions Between Species ina Community 423

Change in abundance of species | Change in abundance of species 2

| stability line stability line

4
Strong Weak Absent Strong Weak Absent

Level of interspecific competition

(a) (b)

Figure 15-5 Graphical representation of the change in numbers of two species that are
competing with each other. (A) Change in abundance of species | as the abundance of a competi-
tor, species 2, varies. When the abundance of species | falls in the shaded area, species 1 increases
until the stability line value is reached. When there are K,/f8 or more individuals of species 1, species
2 is eliminated. When there are K,/f or more individuals of species 2, species 1 is eliminated. If the
abundance of species | falls in the unshaded area, then numbers will decline until the stability line is
reached. (B) Change in abundance of species 2 as the abundance of a competitor, species 1, varies.
The same comments apply as for A. The small graphs show the position of the stability line for
strong, weak, or no interspecific competition.

or the ratio of resource use by species 2 to resource use other species (Gause, 1934, 1935). This type of exper-
by species 1. The terms B and B are called the iment has been repeated many times, including the
competition coefficients. If two species are competing in benchmark work of Park (1962) and his colleagues
an area, then the number of species 1 that the area can with the flour beetles Tribolium castaneum and T? con-
carry is reduced by the number of species 2 present fusum. When these two species were grown together
multiplied by species 2 competition coefficient, and in a bottle of flour, one species always eliminated the
vice versa. In other words, K, — N, — N,/B = 0, and K, other. The successful species depended on the condi-
— N, — N,/® = 0. The relative numbers of the two tions of the experiment. T° castaneum prevailed
species will depend on the value of 8 and 8. This can in warm, moist conditions, whereas T) confusum
be described graphically (Figure 15-5). succeeded under drier, cooler conditions, although
Interspecific competition attracted the attention of when grown alone, both species did best under warm,
several biologists in the early decades of the 20th cen- moist conditions. This type of experiment has also been
tury, including a Russian biologist, Gause, who exam- performed with plants (e.g., Harper, 1961) with essen-
ined the outcome of competition under laboratory tially the same results: where two species were forced to
conditions between two closely related species of pro- compete in a simplified laboratory environment, only
tozoa (Paramecium sp.) utilizing the same food re- one species was able to persist (Figure 15-6).
source. His experiments all ended the same way: one The results of these experiments led to the
species became dominant and finally eliminated the competitive exclusion principle (Hardin, 1960), which
424 CHAPTER15 Community Ecology

Lemna
polyrrhiza
Paramecium
auralia Alone
400
_e,. *— A
+ s
oe pel e In mixture
200
In mixture

0
P. caudatum Alone 400

of
Abundance
protozoans

200

0 JO") 20° *30 407 Sue 60

Days from start of experiment Days from start of experiment
(a) (b)

Figure 15-6 Classical studies of competition between closely related species under labora-
tory culture conditions. (A) The original experiments by Gause (1934, 1935) that led to the devel-
opment of the competitive exclusion principle. Competition between Paramecium aurelia and P. cau-
datum leads to the elimination of P. caudatum, although P. caudatum can equal P. aurelia in population
growth when grown alone. (B) Essentially the same results were obtained by Harper (1961) for two
closely related species of duckweed (Lemna), a small floating plant of calm freshwater ecosystems.
Lemna gibba excludes L. polyrrbiza even though the latter has a greater productivity than L. gibba
when grown alone. (After Gause, 1934, and Harper, 1961. Used by permission ofthe Society of Experi-
mental Biology and F.L. Harper.)

states that two species competing for the same re- nutrients. Why do so many species persist in the face
source cannot coexist indefinitely. One of the two of this competition? One could also ask why so many
species will eventually dominate. This principle, different species of insect herbivore feed on oak trees.
which is sometimes referred to as Gause’ hypothesis, Why are there so many species of birds that prey on a
has attracted a lot of attention over the past half cen- population of insects when it is in outbreak? These
tury. It led to the development of the concept of eco- questions have fascinated ecologists for years, and nu-
logical niche. merous studies have helped us to arrive at some an-
swers and to understand the important role that
interspecific competition has played in evolution. We
15.4 The Competitive Exclusion shall examine three examples of competitive exclusion
Principle and the Ecological in natural ecosystems and then discuss the theory in
Niche Concept general terms.
Forests generally have a variety of bird species
The competitive exclusion principle works well in feeding on canopy insects. For example, the boreal
simple laboratory experiments with two competing forests of New England are inhabited by five species
species, but in natural ecosystems, one can find nu- of warbler (Dendroica sp.), all of which are about the
merous examples that initially seem to contradict the same size and have a similar insect diet. How can these
theory. Consider, for example, the numerous species apparently competing species coexist? The answer was
of phytoplankton in the surface few centimeters of revealed in studies that showed that coexistence was
ocean water, all of which are competing for limited made possible by a physical and temporal division of
 The Competitive Exclusion Principle and the Ecological Niche Concept 425

A)Bay-breasted being a more general feeder and by concentration on

AN\ warbler the lower crown area, which is underutilized by the
other three species. The fifth species can coexist in
anything more than very low numbers only when a su-
perabundance of foods reduces the competitive pres-
sure from the other species. This study of warbler
coexistence revealed a complex of adaptations by
Black-throated which these closely related species are able to share
reen warbler
the same environment.
A somewhat similar example of coexistence was
observed in a study of six closely related species of
leafhopper (Erythroneura sp.) on sycamore trees
(Pseudoplatanus sp.) in Illinois (Ross, 1957). All six
species seem to have similar habits and life cycles, and
it was concluded that this invalidated the competitive
Maximum feeding zone Midde branch zone
exclusion principle. However, it is not clear that severe
© Inner branch zone LJ Outer branch zone interspecific competition did in fact occur between
these species. The leafhoppers differed slightly in pre-
Figure 15-7 Distribution of feeding activity of five species ferred feeding position, and this type of sap-sucking
of warbler in the crowns of conifers in the northeastern insect has modest food requirements; a large, vigorous
United States. (After MacArthur, 1958. Used with permission of tree is able to support very large numbers of them.
the Ecological Society ofAmerica.)
Also, the action of other population regulators may
well have limited the total numbers of leafhoppers to
below the levels at which interspecific competition be-
the food resource that served to reduce interspecific came significant. All six species were found together
competition to a lower value than intraspecific compe- only on trees in moist, fertile locations, where the
tition (MacArthur, 1958). The division was achieved trees would probably be able to produce enough pho-
by differences in feeding behavior and in the timing of tosynthate to support their own minimum require-
breeding, which varies the time of peak food require- ments and those of the hoppers. In contrast, only two
ment (Figure 15-7). The Cape May warbler feeds only species were found on trees in droughty areas, where
on the outer branches in the upper crowns of trees and the trees would produce less photosynthate. Large
is abundant only during outbreaks of species such as populations were found in the fertile locations and
the spruce budworm. The bay-breasted warbler feeds small populations in the droughty areas. It may well be
mainly on the middle parts of the crown, with limited that interspecific competition is unimportant on the
additional feeding in the inner parts of the upper and fertile sites but that it prevents four of the six species
lower crown. It is also most successful during periods from coinhabiting the trees on droughty sites. Com-
of superabundance of food. The blackburnian warbler petitive exclusion can take place only where competi-
feeds mainly in the outer part of the upper two thirds tion is severe, which probably never occurred on the
of the crown. The black-throated green warbler feeds fertile site.
in the middle and outer sections of the branches in the The enigma of high diversity in populations of
middle of the crown and also on the branch ends of oceanic plankton may be explained by the fact that
the upper crown. The myrtle warbler is less abundant competition between these microscopic autotrophs of
than the other species and less specialized in its feed- the open ocean never continues long enough to favor
ing behavior, but it exhibits a tendency to feed in the permanently any one species. It was noted earlier in the
lower crown. It depends on having a very large terri- discussion that the outcome of competition between
tory and is able to maintain a fairly constant popula- two species in the laboratory depends on the condi-
tion through periods of low insect abundance. tions of the experiment. If these conditions are contin-
Three of the five species of warbler seem to be able ually changing, then neither species will hold the
to coexist by a combination of some degree of spatial competitive edge long enough to dominate the other.
and temporal separation of their feeding and by differ- The surface of the ocean is being continually stirred, so
ences in feeding behavior. One species can coexist by that the conditions of light, temperature, and water
426 CHAPTER 15 — Community Ecology

density to which the plankton community are exposed the Hawaiian Islands (Amadon, 1947). Beak length
are continually or at least frequently changed. Under varies significantly among each of these two groups of
these conditions of environmental diversity and insta- closely related bird species. The different beak size re-
bility, many competing species can coexist. stricts feeding by the birds to different types of food,
In contrast to these examples, one can find cases in thereby avoiding competition.
which competitive exclusion has been clearly demon- An interesting example of competitive displace-
strated, Numerous examples occur where a species of ment was described among pocket gophers (Geomyi-
plant or animal has been introduced from another dae) in Colorado (Miller, 1964). Four species occur:
country without its normal complement of parasites the Plains, the yellow-faced, the Botta’s, and the
and predators. Uncontrolled, such species multiply northern pocket gopher. The four species always oc-
rapidly and occupy areas previously occupied by native cupy different areas, although a study revealed that
species that are competitively excluded. For example, frequently more than one species could have lived in
when the prickly pear (Opuntia sp.) was introduced to any one area in the absence of the other species. The
Australia in the 19th century, it spread over large por- most critical factors determining the observed distri-
tions of the continent, excluding native vegetation. By bution of the gophers were interspecific competition
1900, six species had become established on more than and tolerance of soil conditions.
4 inillion hectares. By 1925, the cactus occupied an The relative competitive abilities of the four
area of nearly 25 million hectares; on half of this area, species were observed to be in the order Plains yellow-
the cactus had become so dense that it was practically faced Botta’s northern. That is, the Plains pocket go-
impenetrable by humans or large animals. In a similar pher could outcompete the yellow-faced pocket
example, the Klamath weed (Hypericum perforatum), gopher, and so on. The Plains gopher is found only
which was first reported in the United States in Cali- over a narrow range of soil conditions: deep sandy
fornia in 1900, increased until by 1944 it dominated loams. The yellow-faced gopher also prefers deep
nearly | million hectares of rangeland in the western sandy loams but cannot compete with the Plains go-
United States. pher, which excludes it from these preferred habitats.
A case of competitive exclusion reported from Cal- The yellow-faced gopher can tolerate heavier, more
ifornia resulted from attempts to control an important compact soils and drier sites, which is where it is found
pest species on orange trees, the California red scale when the two species are sympatric (their ranges over-
(Aonidiella aurantii). This pest became heavily para- lap). The Botta’s gopher will occupy a wide range of
sitized by a parasitic wasp, Aphytis chrysomphali, which soils, being absent only from hard clays and very
was accidentally introduced into southern California coarse-textured materials, but because it is unable to
around 1900, In 1948, another species of Aphytis (A. compete with either the Plains or the yellow-faced go-
/ingnanensis) was introduced from China. Sharing the pher, it is usually excluded from the more favorable
same food source, it spread rapidly and succeeded in medium-textured soils. Finally, the northern gopher,
displacing A. chrysomphali from an area of approxi- which is adapted to a wider range of soils and topogra-
mately 1,000 km* in approximately 10 years (DeBach phy than any of the other three species, prefers and is
and Sundby, 1963). In the late 1950s, a third species most abundant in deep, easily worked soils. However,
(4. melinus) was introduced from India. Upon release, because it can tolerate very shallow and coarse soils and
it began to displace A. /ignanensis from hotter interior very compacted soils that are beyond the tolerance of
areas but not from the cooler coastal areas. The mech- the other three species and because of its low competi-
anism of displacement was not known, but it was tive abilities, it tends to be restricted to these marginal
shown that none of the species could coexist in labora- sites when the other three species are present. Figure
tory populations even in the presence of excess sup- 15-8 summarizes the fundamental and realized niches
plies of food. (see below for explanation) of these four species.
A common means by which closely related species Species avoid competition by evolving different
are able to coexist is the evolution of morphological adaptations, tolerances, requirements, behavior, and
differences that enable them to utilize different re- so on. As Gause predicted, no two species that have
sources or slightly different types of environment. the same set of characteristics can compete indefi-
Iwo well-known examples of this are the cases of Dar- nitely. That species do coexist means that they do not
win's finches (Geospizinae) from the Galapagos Islands compete intolerably: that they differ in their set of re-
(Lack, 1947) and the honeycreepers (Drepanididae) on quirements. This observation led to the development
 The Competitive Exclusion Principle and the Ecological Niche Concept 427

GZ Preferred part
Realized niche
Fundamental WY Bierccencetr of fundamental
niche i i
Y), three other species biche species of
pocket gopher

YYy Y Low

ZL) ee YY Northern

: | al
Botta’s =
aa)

||
os]

g=
vo
| Yellow faced =
S
|
|Plain’s
| jan ¢
Northern ; Botta's | Helton Plains | velon | ee Northern | Species High

Compacted Sandy Shallow

clay loam gravel

Heavier Coarser

More compacted Shallower

Figure 15-8 Fundamental and realized niches of four species of pocket gopher in Colorado
expressed in terms of soil texture. All four species achieve peak abundance on loose sandy-loam
soils, but there is a progressive exclusion to increasingly marginal soil types as competitive ability de-
clines. (Based on Miller, 1964.)

of the concept of ecological niche: a term that express- enemies. “The ecologist should cultivate the habit
es the total role of an organism in the environment. of looking at animals from this point of view as
The term was first used by an animal ecologist (Grin- well as the ordinary standpoints of appearance,
nel, 1917) to refer to the habitat, the geographical range, names, affinities and past history. When an ecolo-
and the adaptations of the bird that he was studying, gist says ‘there goes a badger,’ he should include in
the California thrasher. The use of the term has his thoughts some definite idea of the animal’s
changed since 1917. Its current usage encompasses place in the community to which it belongs, just as
three major concepts: if he had said ‘there goes the vicar.” The “profes-
sion” of an animal defines how it “makes its liv-
1. Niche refers to the functional role of a species in an ing,” when it makes its living, and how it interacts
ecosystem. It stresses the entire complex of charac- with other species. When we define a person as a
teristics exhibited by that species. Is it a plant, ani- forester, a wildlife manager, a teacher, or a miner,
mal, or microbe? Is it large or small? Is it we have defined much of how that person makes a
autotrophic or heterotrophic? What is its produc- living and the role ofthat person in the local econ-
tivity and role in nutrient cycling? In other words, omy and community. We have defined the func-
how does the species fit into the complex func- tional aspects of the person’s niche in the
tional processes of the ecosystem? An analogy has community.
been drawn between this aspect of niche and the 2 Niche refers also to the Aabitat of a species: the
profession or occupation of a human. How does the type or range of environments in which it lives. By
organism “make its living”? The English ecologist defining the habitat, the range of physical con-
Elton (1927) had this in mind when he defined ditions to which the species is exposed is also par-
niche as the relationship of an animal to food and tially defined. Thus, definition of a species niche
428 CHAPTER15 Community Ecology

includes its adaptations to light, temperature, cal range all are variable over time. The niche of an in-
moisture, nutrients, soil, fire, and so on and the dividual when it is young may differ from its niche
amplitude of these factors to which it is exposed at when it is old.
various times. The niche of a species has been likened to a vol-
3. Definition of niche involves a statement of the ume within which the species is competitively
geographic area or range over which a species is supreme. For example, if we could define the niche of
found. Geographic area is different from habitat. three species of trees in terms of light requirements,
The latter is concerned with the factors that deter- nutrient requirements, and moisture requirements,
mine where a species is found locally, whereas the then we could represent their niches graphically as
former is concerned with the geographical extent in Figure 15-9. The American ecologist Evelyn
and location of its range. Hutchinson (1957) introduced this idea when he de-
scribed niche as an -dimensional hypervolume (n = the
Niche thus refers to the functional, adaptational, number of parameters being used to characterize the
and distributional characteristics of a species. Defini- niche). He also noted that it is necessary to identify
tion of the niche of a species is obviously complex, in- two hypervolumes: a fundamental niche, which contains
volving a large number of parameters. Also, the a smaller realized niche (Figure 15-8). The fundamen-
definition cannot be fixed, in that a species’ interspe- tal niche represents the maximum niche that the
cific interactions, habitat adaptations, and geographi- species could occupy in the absence of competition

High

Light

Low

A Marginal
conditions

= B
“eb
—]
Moisture Moisture
€ ;
Optimum
conditions
Nutrients Light Nutrients

Figure 15-9 Diagram of the realized niche volumes of three tree species, defined in terms
of light, moisture, and nutrients. The niches defined in terms of two parameters at a time are
shown below. The concentric volumes (circles) represent various portions of the realized niches,
from optimum to marginal. Species A is light demanding but tolerant of low nutrient and moisture
conditions (e.g., a species of pine). Species B is moderately light demanding but requires high mois-
ture and nutrient levels (e.g., a spruce). Species C is tolerant of low light and nutrients but requires
moderate to high levels of moisture (e.g., a hemlock).
 Biological Diversity 429

from other species. The realized niche is that portion dation for sustainable forest management. Within the
of the fundamental niche occupied by the species in potential set by the environmental framework, popu-
the face of competition. In an ecosystem that has been lation and community processes and the ecosystem
relatively undisturbed for a long time, the geographi- processes of ecological succession (Chapter 17) oper-
cal range of a species will reflect its realized niche. ating in conjunction with ecosystem disturbance act to
After disturbances such as fire, extensive logging, or create spatial patterns and different levels of the vari-
removal of some component of the community by bi- ous measures of biological diversity.
otic processes, a species may spread to occupy much or
all of its fundamental niche (Figure 15-8).
Douglas-fir has been planted in western North Different Measures of Biological Diversity
America over a much wider elevational and soil mois- One of the important attributes of a biological com-
ture range than is normal in old-growth forests, where munity is its diversity. Biological diversity, or
it is restricted to a much narrower realized niche. biodiversity, as it is commonly called, emerged as one of
Some of these plantings have failed because the the key global environmental concerns during the de-
species has been planted outside its fundamental niche bate over the world-wide depletion of tropical rain-
(i.e., outside the range of physical environments to forests. Since then, biodiversity has become a matter
which the species is adapted). Others have failed be- of scientific interest and public concern throughout
cause although the plantings were done within the the world, irrespective of the type of ecosystem in-
fundamental niche, competition from other species volved. The discussion of biodiversity has been made
was too high: the species was planted outside its real- difficult by the many different measures of diversity
ized niche. Artificial manipulation of competition by that can be used, the persistent failure to identify
manual, mechanical, biological, or chemical weeding which measures are being used when the topic is being
can permit a species to expand greatly its realized discussed, the different spatial scales over which it is
niche compared with that in an unmanaged system. evaluated, and the change in all measures over time.
Also, where a species is introduced to a new environ-
ment without the natural enemies and competitors Genetic Diversity Within a Species. As is discussed
with which it has evolved, its realized niche may be- in Chapter 16, genetic diversity is a vitally important
come extensive (e.g., prickly pear in Australia, radiata attribute of a species and a population. The presence of
pine in New Zealand and Australia). different ecotypes permits a species to survive in a vari-
ety of different physical and biotic environments,
whereas genetic variation allows a population of a
15.5 Biological Diversity species to adapt to changing environmental conditions.
The resilience of biotic communities in the face of nat-
Ecological Diversity ural and human-induced change is in large part due
As noted in the early chapters and in Chapter 6, the to the adaptation of the member species of the com-
potential for biotic communities is set by the factors of munity made possible by genetic diversity.
the physical environment. These define the ecological
stage, which in turn defines the ecological play (the Taxonomic Diversity. Taxonomic diversity de-
change in the biotic communities and many of the scribes the number of different taxonomic groupings
physical conditions over time—ecological succession found in the area of interest: genera, families, orders,
[Chapter 17]) and the ecological actors (species diversity) etc. An area with a certain number of species but all in
that can occur in the ecosystem. As we go through the same genus will have lower taxonomic diversity
the various measures of biological diversity and the than an area with the same number of species distrib-
various spatial scales at which these measures can be uted between many different genera—similarly for the
evaluated, you will recognize that climatic and topo- distribution of the genera in an area between different
graphic variability, together with local variations in families, etc.
soil and physical disturbance factors such as fire, wind,
and slope instability—collectively referred to as Species Diversity. Species diversity is the most
ecological diversity—define the template for biological commonly considered aspect of biological diversity.
diversity. This is why ecological site classification (or When people express concern about biodiversity, they
gradient analysis) is such an essential ecological foun- are usually referring to the loss of species, especially
430 CHAPTER15 Community Ecology

rare and endangered species. Species diversity is mea- ence of standing dead trees (snags) and decaying logs
sured by species richness (the number of species present on the ground (CWD) of various species, sizes, and
in a sample of a particular community) and species even- degrees of decomposition constitutes another very im-
ness (the relative abundance of the different species). portant measure of the biological diversity of an
Species richness is often used to refer to the species ecosystem—structural diversity. The biological diver-
diversity of particular groups, or guilds, of species sity of animals that depend on the plant community is
rather than all the species in the ecosystem in question. closely related to the species richness and evenness of
We talk about the tree species richness, or the diver- that community, but it is also influenced by the spatial
sity of bird species, or the diversity of terrestrial verte- arrangement of the plants, both horizontally and ver-
brates in a particular community. A boreal spruce tically; the structure of the tree canopies; the presence
forest will have low species richness in trees, vascular of canopy gaps; and the presence of snags and CWD.
plants, and vertebrates but may have a very high diver- These structural characteristics influence the number
sity of soil invertebrate animals and of microbes. The of potential animal niches provided by a forest plant
word “may” is used here deliberately, because rarely do community. The example of the diversity of warbler
we have information about the species richness of the species in a low-structural-diversity spruce forest
last two groups of organisms. Whether one considers given above suggests that animal species diversity can
the boreal forest a low-diversity ecosystem thus de- and does exist even in structurally simple forest com-
pends on how many groups, or guilds, of organisms munities. Similarly, some “old growth” or “climax”
are included in the evaluation. In contrast, many trop- forests that have high structural diversity have low
ical rainforests that have received minimal disturbance species diversity for some or many of the groups of or-
from human activities have a large number of different ganisms that make up that biological community. The
tree species per hectare of forest, and their high species increase in vascular plant species diversity that often
diversity of insects and other animal life forms is leg- accompanies the structural simplification of the plant
endary. Such forests are usually considered to have community after forest disturbance by fire, wind, or
high biological diversity, but knowledge of microbial logging also suggests that there is no simple, universal
diversity (which may in fact represent the majority of relationship between structural diversity and species
the species biological diversity in forests) of tropical richness. However, it is generally believed that within
and other forests is generally not yet sufficient to per- a particular type, age, and condition of a forest, in-
mit confident statements to be made about the total creased structural diversity will be associated with in-
biological diversity of tropical versus boreal forests. creased species richness and greater species evenness.
Species richness is an incomplete description of
species biological diversity because it does not account Functional and Life History Diversity. Another
for differences in the relative abundance of the differ- important measure of biological diversity is the varia-
ent species in the community. Imagine a forest com- tion in the functional and life history attributes of the
munity with 1,000 trees belonging to 10 different species that make up the community: deciduous versus
species. If 9of the species were represented by only 10 evergreen plants; shrubs versus herbs versus trees;
individuals and the 10th species by 910 individuals, herbivores versus saprotrophs versus carnivores versus
then the forest would have the appearance of a single- omnivores; annuals versus perennials; rhizomatous
species forest (a monoculture) with a few individuals versus nonrhizomatous plants; seed banking versus
of the other species scattered through it. In contrast, if non-seed banking species; plants with high nutrient
all 10 species each had 100 individuals and all were uptake demands versus low nutrient demands; varia-
distributed evenly or randomly across the area, then tions in shade tolerance; variations in longevity and
the forest would have the appearance of a diverse, survivorship curves, etc.
mixed-species forest. Clearly, two forests with the
same species richness can have different biological di- Temporal Diversity. One of the most fundamental
versity if the species evenness, or relative frequencies attributes of biological diversity is that it changes over
of different species, is very different. time. All the above measures of biological diversity
vary as forest grow older, ecological succession pro-
Structural Diversity. The many different canopy ceeds, and the forests are disturbed. The change oc-
layers, the variation in plant life forms, and the pres- curs at very different spatial and temporal scales in
 Biological Diversity 431

different kinds of forest ecosystem. In some types of dry gravel ridges, sandy areas, mid-slope areas with
old forests in wet climates where stand-replacing dis- medium-texture soils, and poorly drained lower areas.
turbance events are rare, the change may occur at The beta diversity can also result from patchy natural
small spatial scales (some fraction of a hectare), with (fire, insects, wind) or logging disturbance. Con-
little change at larger scales within time periods of one versely, the frequently high alpha diversity of tropical
to several centuries. In forests that are characterized rainforests can exist in a landscape with low beta diver-
by frequent and severe ecosystem disturbance, various sity—the alpha measures of diversity do not vary much
measures of biological diversity may change over time across the local landscape if the topography is flat and
scales as short as one or a few decades. Temporal diver- there are uniform soil conditions. However, if there
sity is as important a measure of the biodiversity char- has been a history of wind or small-scale human dis-
acteristics of a forest as are the previous five measures. turbance in this tropical environment or if there is
Clearly, there are many different ways in which we local soil and topographic variability, then the beta di-
can describe the biological diversity of a given com- versity may also be moderate to high. Temporal beta
munity in addition to the simple measure of the length diversity can be low in landscapes where the spatial
of the species list, which is what people usually mean scale of disturbance is small. The landscape is a mosaic
when they talk about biodiversity. of stands of different age, structure, and composition,
and the overall character of the mosaic may be fairly
Different Spatial Scales at Which constant except that the location of stands of different
These Measures Are Evaluated ages and condition shifts over time within the mosaic.
Stand Level, or Alpha, Diversity. When the above
measures of biological diversity are assessed at the Regional Landscape, or Gamma, Diversity.
stand level—an area of perhaps 1 to 100 ha—they col- Major topographic features such as mountains, varia-
lectively describe alpha diversity. This is the diversity tion in the distance to oceans or large lakes, and signif-
that one can see and hear standing in a particular for- icant changes in latitude are associated with changes in
est ecosystem. It is the spatial scale of biodiversity that climate that result in major changes in the life-form
most people relate to because it is the spatial scale and species composition of the vegetation and associ-
of diversity that their senses can evaluate. Temporal ated organisms (see earlier discussion of biomes and
diversity is particularly important at the alpha level plant formation). This creates the ecological zones
because succession is continuously changing the biodi- that define regional- and continental-level biological
versity characteristics of individual stands (Chapter 17). diversity. The variation in both alpha and beta diver-
sity measures as one proceeds along climatic gradients
Local Landscape, or Beta, Diversity. As one pro- across larger landscapes constitutes gamma diversity.
ceeds across a local landscape within a particular cli- ‘Temporal diversity at the gamma scale is generally
matic area, it is common to encounter a series of very low. The overall variation in forest character be-
different forest stands and biological communities that tween different climatic areas (cf. biogeoclimatic
vary in soil, slope, aspect, and other physical character- zones and subzones; Chapter 6) is constant unless
istics. They will also vary in their age and history of major climatic change occurs. Otherwise, it increases
disturbance. These variations result in changes in the only where very large disturbance events such as major
various measures of stand-level biodiversity discussed fires or insect epidemics result in major changes in the
above. Beta diversity is measured by the degree of vari- forests within one of the climatic areas encompassed
ation in alpha measures of diversity across this land- by the landscape in question. Changes in location of
scape. A landscape that has low measures of alpha timber harvesting across climatic gradients over time
biodiversity in the individual stands can have either can also increase gamma diversity.
high or low levels of beta diversity as a result of past There are many reasons that conservation of bio-
ecosystem disturbance, or high beta diversity as a re- logical diversity should be an important objective of
sult of high ecological diversity: a high spatial diversity in forest management and conservation policy (Burton et
soil and topographic conditions. Lowland boreal al., 1992; Kimmins, 1992). However, simply saying
forests that have low alpha diversity for several groups that biodiversity is important does not provide an ade-
of organisms can have high beta diversity as a result of quate basis for achieving this aspect of conservation. It
undulating topography that gives a complex mosaic of is necessary to identify specific biodiversity objectives
432 CHAPTER 15 Community Ecology

in terms of specific measures of biodiversity, specific given in Figure 15-10. This shows the number of tree
groups of organisms, and specific spatial and time species recorded in 20 X 20-km grid cells in British
scales. In most forests, it is also essential to define ob- Columbia. The greatest diversity is found in moun-
jectives with respect to temporal patterns of biological tainous regions in the southern part of the province at
diversity. Maintenance of a particular measure of alpha the interface between coastal and interior climatic
diversity at some constant desired level in a forest that conditions where the ecological diversity is very high,
naturally exhibits change in that measure over time re- the grid cells covering several biogeoclimatic (BEC)
quires active management intervention to eliminate zones, and where the climate is milder. It is lowest in
the natural temporal biodiversity of that ecosystem. the northern area that has lower ecological diversity—
Conversely, accepting natural temporal biodiversity much flatter topography and fewer BEC zones—and
generally precludes the maintenance of any particular colder climates.
measure of alpha diversity at a constant level. Many explanations have been advanced to explain
Natural forests exhibit a wide range oflevels of the the global pattern of latitudinal variation in diversity:
different measures of biological diversity. Some forests time, rate of speciation, predation, environmental sta-
have low tree species diversity but may have high di- bility, environmental heterogeneity, and size and spa-
versity of mosses, lichens, herbs, and/or shrubs. Some tial isolation of “islands.”
forests that are much richer in tree species diversity
may have lower levels of diversity in these other plant
Time Hypothesis. One of the earliest suggestions
life forms. Relatively little is known about the diver-
was based on the greater age of tropical ecosystems,
sity of soil invertebrates and microbes in most forests,
where evolution has supposedly continued with rela-
except that it is generally high, that it varies between
tively little interruption for millions of years, giving
different forest ecosystem types, that it varies over
the opportunity for new species to evolve and for eco-
time in response to disturbance and the processes of
logical niche diversification: the division of the ecologi-
ecological succession, and that there is high temporal
cal roles and opportunities within an ecosystem
diversity at very small spatial scales and high spatial di-
between an increasing number of species. Glaciation
versity over small areas.
in the recent past makes many temperate areas very
young by comparison, giving less time for high species
diversity to evolve. Certainly, the fossil record, al-
Global Patterns of Biological Diversity though incomplete, does suggest a latitudinal gradient
in species diversity, and it has been pointed out that
Perhaps the only reliable generalization that can be
the diversity of insect species that feed on trees is re-
made about global patterns of biological diversity in
terrestrial ecosystems is that alpha diversity for many
lated to the time over which the insect—host relation-
groups of organisms generally decreases from lowland
ship has developed (Southwood, 1961). For example,
sessile oak, which is native to Britain, is associated
tropical to high-latitude or high-altitude ecosystems.
A tropical rain forest in Malaysia can contain up to 227
with 284 species of insect, whereas evergreen oak, in-
tree species on approximately 2 ha (Richards, 1969),
troduced in 1580, has only two associated insects. A
whereas the central hardwood forests of the eastern graph of the present-day number of associated insect
United States may have 10 to 40 overstory species species plotted against the length of the fossil record
(Clark, 1976) present in an area, and the boreal forest
of the host tree species in Britain shows a reasonably
linear trend (Figure 15-11). The validity of the time
in Canada may have only 1 to 4 species. The number
hypothesis is difficult to assess because of the incom-
of ant species varies from 222 in Brazil to 63 in Iowa
pleteness of the fossil record and because unequivo-
and 3 in arctic Alaska (Fischer, 1960). There are ap-
cal proof would require actual observation, which is
proximately 600 species of breeding land birds in
nearly impossible considering the time scale. Conse-
Panama, 150 in Oregon, and 26 in the Alaskan north
quently, there has been more interest in the alternative
coast (MacArthur and Wilson, 1967). The number of
hypotheses.
land mammal species increases from 15 in northern
Canada to more than 150 in Central America (Simp-
son, 1964). An example of how the diversity of tree Rate of Speciation Hypothesis. It is widely held
species can vary latitudinally and longitudinally is that the rate of speciation (evolution of new species) is
 Biological Diversity 433

60°N

aS h Co
TAY
Nes

Pacific
Ocean
Number of species

30-34 United States

Figure 15-10 ‘Tree species diversity in British Columbia. The highest diversity occurs in the
area of climatic transition from coastal to interior climatic conditions. The area is mountainous, and
there is a high diversity of physical environments based on variation in elevation, aspect, soil, and cli-
mate within the transition area. (After Krajina et al., 1982. Used by permission of f.G. Worrall.)

faster in the tropics than in temperate or arctic envi- ates a more favorable environment for speciation.
ronments. The longer growing season and relative Many animals and birds in the tropics are more seden-
lack of seasonal environmental extremes in the tropics tary than at higher latitudes and altitudes. This char-
permits many generations per year and maintenance acteristic reduces the flow of genes between
of populations close to their Kit value (carrying capac- populations that are diverging genetically and thereby
ity). This intensifies interspecific competition and cre- facilitates speciation. High mountains in tropical
434 CHAPTER15 Community Ecology

300
@ Oak
e Willow

200

¢ Hawthorn

100

species
each
with
tree
associated
species
insect
of
Number e Hazel

4 Hornbeam
@ Common maple
° Holly
0) Yew 100 200
Number of fossil records over the past million years

Figure 15-11 The relationship between the length of the fossil record of several British tree
species and the number of insects known to be associated with them. The number of fossil
records of tree remains for the Quaternary (Pleistocene and recent geological periods, approxi-
mately the last million years) is a measure of how long the tree species has occurred in Britain and
how long insect-tree coevolution has been occurring in this area. (After Southwood, 1961. Used with
permission ofBlackwell Scientific Publications.)

countries tend to be a greater barrier to bird and ani- these serve to hold down prey populations to such low
mal movement than do the mountains of temperate levels that competition among prey species is reduced
zones; high mountain passes are climatically much (Paine, 1966). This permits the addition of more prey
more different from lowland habitats in the tropics species, which in turn supports new predators. The
than in temperate zones (Janzen, 1967). addition of more prey species tends to reestablish in-
Natural selection in tropical environments is pre- terspecific competition, which in turn promotes fur-
dominantly in response to biotic factors (interspecific ther speciation. There is some evidence in favor of this
interactions), whereas at higher latitudes and altitudes, idea. It has been shown that in some communities, the
physical factors play a major role (Dobzansky, 1950). removal of predators lowers the diversity of prey
The release of adapting systems from the physiologi- species because it permits the development of domi-
cal constraints imposed by the demands of the physical nance by a few species.
environment allows a more rapid evolutionary re- Janzen (1970) suggested that seed predators con-
sponse to biotic factors in the tropics. This encourages tribute to the low density of adult trees of any one
the development of a positive feedback between the species in species-rich, lowland tropical forests and to
development of diversity in different components of the regular spacing of such trees. Many tropical
the community; increased diversity of animals leads to seedeaters are host specific and tend to be aggregated
a diversification of the vegetation, which creates the around the host tree. This produces a “seed shadow”
potential for a greater diversity of animals. around each tree and a steady increase in seedling es-
tablishment with increasing distance from the seed
Predation Hypothesis. Another theory about trop- source, until one approaches maximum dispersal
ical diversity suggests that there are more predators range (Figure 15—12). Survival of a tree’s offspring is
and parasites in the tropics than elsewhere and that reduced within its own seed shadow, permitting other
 Biological Diversity 435

Number of
seeds
Probability
of survival — ats
Net = oft

ea
eas

Number of seedlings
of
Number
falling
seeds
unit
per
area established per unit Probability
survive
will
seed
thatgerminate
and
maturity
to

No reproduction Successful 1 No reproduction

| reproduction |

Distance from the parent tree

Figure 15-12 Graph of the suggested explanation for the lack of self-reproduction in the
vicinity of tropical trees (“seed shadow”). This phenomenon is thought to contribute to the high
diversity of tropical rain forests. (After Janzen, 1970. Copyright University of Chicago Press. Used with
permission of the publisher and the author.)

species to grow there and producing a high alpha di- sity is greater in deep-sea communities than in very
versity. Janzen’s hypothesis requires that seed preda- shallow water communities has been used to challenge
tion is less efficient in the less diverse temperate and the theory that diversity is related to temperature: that
high latitude/altitude environments. In support of the high temperatures encourage the development of high
theory is the report by Harper (1969) that the herba- diversity and low temperatures induce low diversity
ceous vegetation on an island off the coast of England (deep-sea temperatures are low). It has been proposed
was more diverse when heavily browsed by rabbits that diversity is not related to temperature per se but
than after removal of the rabbits by myxomatosis (a to the stability of that temperature and the pre-
virus disease). dictability of its changes. Stable climates permit the
The exact role of predation in population regula- development of more finely adapted species and there-
tion is still unclear and so also is the role of predation fore more potential niches than erratic climates in
in determining diversity. There is a lack of data to test which there is a strong selective advantage to having
this hypothesis critically, and it may be that predation broad adaptations. Against this argument is that there
is more important in explaining alpha diversity in the are many hot desert and semidesert environments that
tropics than in explaining the global latitudinal gradi- are markedly unstable in terms of moisture and yet
ent in species diversity. have high plant diversity. This would support the idea
that temperature, rather than either moisture or the
Environmental Stability Hypothesis. An impor- stability of environmental conditions, has a key influ-
tant difference between tropical and polar environ- ence on terrestrial vascular plant diversity (Whittaker,
ments is the degree of environmental stability. 1975).
Environments with low biotic diversity tend to be se-
vere or unpredictable or both, whereas those with Environmental Heterogeneity Hypothesis. “There
high biotic diversity tend to be either stable or pre- seems to be a relationship between diversity and envi-
dictable or both. The surprising fact that biotic diver- ronmental heterogeneity (ecological diversity). “The
436 CHAPTER15 Community Ecology

more heterogeneous and complex the physical environ- many species; therefore, the extinction rate is higher
ment, the more complex the plant and animal commu- than on a large island with many resources. It is
nities. Mountainous areas are associated with higher thought (MacArthur and Wilson, 1967) that islands
species diversity than areas of gentle topography. that are distant from the source of immigrants will
Mountainous Columbia has more species of birds than have a lower rate of immigration and therefore will
Brazil, which has a more uniform topography. The Pa- support a lower equilibrium diversity than islands that
cific coast and Sierra Nevada mountains in the United are near to a mainland source of immigrants. The
States have a higher diversity of mammals than do the species-area curve (the relationship between the size
high plains or Mississippi Valley. In Canada, conifer of an area and the number of species that it supports)
forests of topographically and climatically diverse should rise more rapidly on islands that are close to
British Columbia have in total a much greater diversity the mainland than on remote islands. Many ecologists
than coniferous forests in the less variable physical envi- have investigated this idea, which is referred to as
ronments of provinces such as Saskatchewan, Mani- Island Biogeography Theory, and is discussed below.
toba, and Ontario. The greatest diversity in British
Columbia is found in the coastal mountains in the tran-
Island Biogeography Theory
sition zone between interior and coastal climatic re-
gions (Figure 15-10). This area has a particularly high
and Forest Fragmentation
diversity of physical environments. Mountainous relief Large areas of tropical forest are being converted to
is characterized by marked variations in elevation, as- agriculture, leaving isolated patches of remaining for-
pect, slope, climate, and soils, and this heterogeneity is est. Urbanization, agriculture, and other land uses
associated with a considerable diversity of plants and have created similar isolated patches of mature or sec-
animals. ondary forest in a matrix of permanent non-forest
An interesting aspect of spatial heterogeneity is ecosystem conditions in countries at other latitudes
found among birds. A study of bird diversity in the de- over the past few centuries. This has raised the ques-
ciduous forests of the eastern United States revealed tion as to how many of the species native to the origi-
that there was a stronger relationship between bird nal forest will survive in these isolated forest
species diversity and foliage-height diversity than be- fragments. Many people have invoked the concepts of
tween bird diversity and plant species diversity. island biogeography theory to examine the problem of
Foliage-height diversity is a measure of the stratifica- forest fragmentation (e.g., Harris, 1984).
tion of the plant crowns. High foliage-height diversity The relationship between species diversity and size
occurs where foliage extends continuously from of “island” suggested by island biogeography theory
ground to upper canopy, whether this is because of a seems to be generally correct where there is a biologi-
variety of species or a variety of age classes of a single cally significant and relatively permanent “hard edge”
species (MacArthur and MacArthur, 1961). between the “island” and its surroundings. The
ocean-island interface constitutes a significant and
Size and Spatial Isolation of “Islands.” The di- permanent impediment to immigration and emigra-
versity of species on islands, peninsulas, or isolated tion, which play important roles in determining the is-
patches of vegetation (e.g., “islands” of forest in grass- land size-species richness relationship. Forest parks
lands, agricultural or urban areas, or the vegetation on embedded in large urban areas or patches of forest
a tall, isolated peak) presents a special case. It has been surrounded by permanent agriculture similarly exhibit
observed that the number of species on oceanic islands relatively hard “ecological edges” that inhibit plant,
depends on the size of the island (Figure 15-13) mod- animal, and microbial immigration from the sur-
ified by the distance of the island from the mainland rounding forest metapopulation (Chapter 14). The
and that the shape of the relationship is reasonably larger the forest patch, the greater the chance of de-
constant. The number depends on two factors: the veloping a metapopulation within the patch from
rate at which species become extinct and the rate at which local population extinctions can be replaced and
which extinction losses are replaced by immigration to the greater the probability that organisms that move
the island. The explanation for reduced diversity on from the surrounding regional metapopulation will
small islands is that the small supply of resources on reach the patch. For those species that for physical or
small islands cannot support large populations and/or biological reasons cannot survive close to a “hard
 Biological Diversity 437

400
200
3g 1000 100 Hispaniola
100
a 50 Puerto Rico Guba
ce KOO)
°
Jamaica
2S| 10 10
=) Montserrat
Z,
Redonda
|
10 100 1000 10,000 100,000 1 10 1000 1 10 100 1000 10,000 100,000
Size of the island (miles?)
(a) (b) (c)

Figure 15-13 Species diversity and the area of islands in various parts of the world. The
graphs show how the number of species increases as the size of an island increases. (A) Terrestrial
and freshwater birds on the Sanda Islands, the Philippines, and New Guinea (after MacArthur and
Wilson, 1967). (B) Terrestrial plant species on the Galapagos Islands (after Preston, 1962). (C) Am-
phibians and reptiles in the West Indies (after MacArthur and Wilson, 1967). (A and C from Robert H.
MacArthur and Edward O. Wilson, The Theory of Island Biogeography. Copyright © 1967 by Princeton
University Press. Reprinted by permission of Princeton University Press. B from Preston, 1962. Used with
permussion of the Ecological Society ofAmerica and FW. Preston.)

edge,” the larger the patch, the greater the area of The size and shape of forest reserves and the pat-
habitat they can occupy (i.e., the greater the area of tern and type of forest harvesting across a landscape
“interior habitat”). Finally, the larger the patch, the have important implications for various measures of
greater the chance that it will encompass environmen- biological diversity. Site-, landscape-, and situation-
tal diversity. A small forest patch that includes signifi- specific biodiversity objectives must be formulated,
cant topographic and soils diversity may support and equally specific plans must be developed to
higher species richness than a much larger patch with achieve them. Once specific biodiversity objectives
a low level of environmental heterogeneity. have been decided, landscape patterns of ecosystem
Island biogeography theory clearly is a useful con- disturbance that will achieve these objectives should
cept for the design of parks and ecological or wildlife be designed. Success in this regard is unlikely if one
reserves where ecological differences between the re- single concept, paradigm, or theory is applied to all
serve and the surrounding areas create a relatively per- landscapes, irrespective of their ecological difference
sistent “hard edge” for species of concern. However, and the differences in the ecology of the different or-
where the “hardness” of the edge softens with time be- ganisms in the different areas. Although there are
cause of ecological succession or where at least part of some principles and concepts that apply broadly across
the surrounding environment is suitable for species most forests, in most cases the successful conservation
migration (i.e., there is a connection, or “bridge,” to of biodiversity, however this is defined, must be based
other suitable habitat), the theory may not provide a on the local species, ecosystem, and landscape ecology
good prediction of species richness. This seems to be and must not be based on generalized theory that ig-
the case for many forest reserves embedded within a nores the diversity of the forest ecosystems in which
matrix of managed forest. Where the surrounding we seek to “conserve biodiversity.”
forest is managed to maintain patterns of forest age
classes, seral stages (Chapter 17), and community
Biological Diversity and the Stability of
structures that facilitate species dispersal and prevent a
Communities and Ecosystems
“hard” and permanent edge from existing around the
entire reserve, the diversity-patch size relationship The important topic of biological diversity and the
will become much less reliable. The relationship will stability of communities and ecosystems is discussed in
also vary according to the degree of ecological diver- Chapter 17. It is important because we must under-
sity within a given patch. stand the implications for the stability, resilience, and
438 CHAPTER15 Community Ecology

functioning of forest ecosystems of altering natural solely or partially on the plant community. Develop-
levels and natural temporal patterns of change in bio- ment of a practical plant community-based system of
diversity. It is sufficient to note here that the often forest classification involves a recognition of how
sought-after simple relationship between “diversity” species are distributed along environmental gradients
and “stability,” however these are defined, remains (Chapter 13). A system that assumes discrete associa-
elusive. Although there are examples from particular tions will not work well where there is a continuum of
forest ecosystem types where specific measures of bio- vegetation change, whereas the assumption of a con-
logical diversity do seem to be related to certain spe- tinuum ignores the advantages that can be gained in
cific measures of stability, no generally applicable regions where there are mutually exclusive associa-
relationship has emerged yet from the studies that tions that reflect significant variations in the produc-
have been conducted. It continues to seem that a wide tive potential of the physical environment. Knowledge
variety of diversity—stability and diversity—ecosystem of species distributions along environmental gradients
function relationships exist in unmanaged forests and can help in predicting such things as the correct tree
that an equal range of relationships can occur in man- species for the site, the brush hazard, and the appro-
aged forests. Io resolve this question so that we can priate silvicultural regime.
devise sustainable systems of forest management One of the most important aspects of community
seems to require that we give up the search for gener- ecology is the diversity of interspecific interactions.
alized relationships that can be applied everywhere Most foresters are well aware of antagonistic interac-
and turn our attention to understanding how these re- tions between their tree crops and some of the other
lationships vary in different forest ecosystem types and species in the community, but the full significance of
conditions. interspecific interactions for the composition and pro-
ductivity of the community is not always recognized.
The geographical range of a species may be deter-
15.6 Community Ecology and mined as much by its susceptibility to an herbivore or
Forest Management a pathogen as its ability to tolerate some climatic or
edaphic factor (note that biotic and abiotic factors in-
Foresters are concerned with the structure and func- teract). The growth of a tree species may not be possi-
tioning of forested ecosystems. Although this concern ble in the absence of some mutualistic interaction
requires that due consideration be given to both the (e.g., mycorrhizae or N-fixing microbes), whereas the
abiotic and biotic components, it is the latter part of presence of a noncommercial but deeply rooted
the ecosystem that is often of greatest immediate in- species with nutrient-rich, highly decomposable litter-
terest. Chapter 3 stressed the need to consider the fall may significantly increase the production of a shal-
biotic community in an ecosystem and landscape con- low-rooted crop species.
text. However, there are several aspects of community Recognition of the importance of interspecific in-
ecology that have value for forest management, with teractions and how the forester can influence them
or without a broader ecosystem framework. creates the possibility of managing these interactions
In Chapter 6, we saw that knowledge of life forms to our advantage. Improved mycorrhizal and sym-
and their relative frequencies (life-form spectra) and biotic nitrogen associations offer the possibility of sig-
the vertical layering and percentage cover of the vari- nificant increases in tree growth. Understanding the
ous layers in plant communities can be very useful in nature of competition can facilitate wise investment of
the description, classification, and identification of time and energy in alleviating competition or can lead
forest ecosystems. Stand structure is important in tim- to the use of competition as a biological method of
ber, wildlife, water, range, and recreation, so foresters weed control.
should have a basic knowledge of community struc- Community ecology considered in an ecosystem
ture and its relationship to the resource values that context leads to the concept of ecological niche.
they are managing. Knowledge of the fundamental niche of a tree species
Classification of forest ecosystems (Chapter 6) in- can indicate the potential geographical and habitat
volves division of the landscape into classes. This can range over which the species might be grown, but suc-
be done in a number of ways, but it is often based cessful extension of a species outside its realized niche
 Take-Home Message 439

involves an understanding of the biotic interactions cussed in the next chapter. Communities vary in the num-
that act to exclude the species from those areas of its ber of layers of vegetation that they have, which depends
fundamental niche that it does not normally occupy. on the variety of life forms represented in the community,
Only when the forester is able to modify these interac- which in turn reflects soil and climatic conditions and the
tions appropriately will crop establishment outside a actions of biotic factors. The floristic composition and
species’ realized niche be successful. Harvesting and structure change as these two physical features change,
which gives rise to the spatial variation in communities.
postharvest site treatment can so alter the physical
Irrespective of exactly how species arrange them-
characteristics of a site that the area can become part selves in space, there is little question that the distribu-
of the realized niche of a species that is normally ex- tion is related to the ecological niches, both fundamental
cluded from that area. By this means, the forester may and realized, of the species concerned. These niches are
be able to expand the range of a desirable species but defined both in terms of the physical requirements and
may also expand the range of a pest species. However, tolerances of the organisms and in terms of the rich di-
it must be recognized that unless the physical changes versity of interspecific interactions that exist in any com-
are maintained, antagonistic biotic interactions that munity. As with population regulation, the relative
normally exclude the desired species may intensify and importance of physical and biotic factors in determining
result in the eventual failure of this species. The com- community structure, composition, and variation along
environmental gradients will differ between different cli-
plex interplay between biotic and abiotic factors that
matic, physiographic, and biotic regions.
determines a species range is seen most dramatically in Comiunity diversity has been the subject of much
tree lines. A consideration of this type of ecotone is discussion. Earlier ideas that there was a positive relation-
helpful in appreciating the complexity of ecosystems. ship between the diversity and the stability of the commu-
Much has been written about the relationship be- nity have had to be reevaluated, partly because no simple
tween species diversity and community productivity relationship seems to exist and partly because it is so diffi-
and stability. Few practically useful generalizations cult to decide just what we mean by stability, and there are
have emerged, but it does seem that diversity can be a so many different measures of diversity. All that can be
useful parameter in defining ecosystems. More re- said at this point is that stability, however this term is
search is required before we are able to assess for most used, is determined by far more than biological diversity.
sites the relative desirability of low and high levels of The diversity of climates and soils around the world
has permitted the evolution of diverse forms of terres-
the various measures of diversity. trial life, which are arranged in broad biotic assemblages
It should be apparent from this brief review that over the surface of the continents. These assemblages are
community ecology involves a variety of topics that are characterized by the life form of their vegetation (i.e., the
of value to foresters. Much of the value relates to an plant formation) but include all the animals and microbes
understanding of ecosystem change and the applica- with which the plants are associated. The term biome
tion of community ecology in ecological classification. refers to these broad regional biotic assemblages and
their associated abiotic environments.
Community ecology is important in forest resource
SUMMARY management because in attempting to favor particular
species, the manager must understand the species’ niche,
There are two major components in the environment of
both fundamental and realized, and the variety of inter-
any organism: abiotic factors (physical and chemical) and
specific interactions that will determine, in large part, the
biotic factors. The latter can be further subdivided into
success or failure of his or her activities. Interactions that
organisms of the same type (other members of the same
are adverse for the desired species must be controlled,
species population) and organisms of some other type.
and those that are beneficial should be promoted.
These two subcomponents give rise to intraspecific and
interspecific interactions, respectively, both of which are
important in determining the abundance and distribution TAKE-HOME MESSAGE
of organisms. This chapter has been concerned with the
second of these two types of interaction. The community is the biotic component of the ecosys-
Mixed-species communities have three dimensions to tem. Many of the things that we value about forests are
their structure, two of which were considered in this chap- related to this living component: wood products, wildlife,
ter: the vertical structure and the horizontal structure. the beauty of the forest, the modified microclimate, the
The third dimension, change in structure over time, is dis- stabilization of steep slopes and snow packs, the control
440 CHAPTER 15 = Community Ecology

of water flows, and the creation and maintenance of ecological system of which the community is a part.
stream habitat for aquatic animals. Forestry is concerned Community ecology—the study of biotic communities—
with managing this community to create and sustain cer- should not be considered as an end in itself but as a
tain desired ecosystem conditions and desired patterns of vitally important contribution to the understanding of
change in these conditions over time. To achieve this, we forest ecosystems: how they function, how they change
must understand what forest communities are—how they over time, and how they should be managed.
are structured and how this structure varies along envi-
ronmental gradients—and how their component organ-
isms interact with each other. With increasing public
concern about loss of species, we must also understand STUDY QUESTIONS
the complex topic of biological diversity: what it is, how 1. What kinds of interspecific interactions are particu-
we measure it, and how we manage the biotic community larly important in forests?
of the forest ecosystem to achieve desired levels of the
various different measures of diversity. 2. Explain the concept of ecological niche.
To understand and interpret communities, we must Why are there so many examples that seem to contra-
understand the autecology and population ecology of the dict the competitive exclusion principle?
individual species that make up the community. We must 4. What is biological diversity? What are the different
understand their interactions with the physical environ- measures that are used to describe it?
ment and their individual contributions to ecosystem
5. What determines the global patterns of species rich-
function. We must recognize that the interactions be-
ness?
tween the member species and their response to the non-
living factors of the ecosystem are strongly influenced by 6. How does the “island biogeography theory” apply in
the genetic make-up of the population of each of the the management of forests and in conservation?
species in the community. In short, we must approach 7. How can a forester use a knowledge of community
both the study and the management of the forest com- ecology to design appropriate “vegetation manage-
munity in the context of the complex, highly interrelated ment” (i.e., “weed” control) strategies?
 C H E R

16
Genetic and Evolutionary
Aspects of Ecosystems
Adaptation and Evolution'

16.1 Introduction be borne in mind when manipulating either the biotic

or the abiotic components of ecosystems.
Our discussion of ecosystems and their functional All living organisms share the need for energy and
processes and the responses of populations and com- for an appropriate supply of chemical elements with
munities to biotic interactions has so far largely ig- which to obtain, store, and utilize that energy. That
nored the genetic variation in the character of the they are able to do this in almost all of the enor-
individual members of a population. In contrast, the mously varied environments present on Earth is a re-
chapters on the physical environment repeatedly re- flection of the tremendous diversity of living organ-
ferred to ecotypes—genetic subpopulations adapted to isms. This diversity is not merely a response to
specific physical environmental factors. Understand- variation in the physical environment, for if this were
ing and managing forest ecosystems requires knowl- so, we might expect that the biota of any particular
edge of the genetically determined behavioral, ecosystem would consist of only one or at most a very
physiological, and morphological adaptations that per- few species at each trophic level (low alpha diversity’).
mit a particular organism to survive and prosper in Although flora and fauna of very low diversity are
particular biotic and physical environments. Such found in some physically extreme environments, it is
knowledge is just as important as knowledge of the far more common to find a large number of different
ecosystem processes, the physical environment and the species participating in ecosystem processes. So great
biotic interactions in which that organism is involved. is the variety of species in most ecosystems that one is
This chapter is intended to remind the reader that prompted to ask, as did Hutchinson (1959), why there
biotic responses to physical and biological conditions are so many kinds of plants and animals. Part of the
and events are largely determined by the genetic char- answer to this question is to be found in one of the
acteristics of the individual organisms involved, that most important processes of ecology and of biology in
all populations of organisms exhibit a variation in general: genetic adaptation and evolution.
these characteristics, and that organisms can acclimate Biological evolution is the change in the genetic
and/or adapt to changing conditions. However, for makeup of a species or population over time. It occurs
any particular organism, there is a maximum rate at because the individuals of any population differ from
which these two processes can occur and a maximum
degree to which an organism can acclimate. This must
*Alpha species adversity—the number ofspecies (the species richness) with-
in a sampling unit (e.g., a local ecosystem or forest stand). Beta species di-
versity—the variation in species composition between two different
‘Much of this chapter was contributed by Dr. 7.G. Worrall, Dendrologist, sampling units. These two measures of biodiversity also refer to species
Faculty ofForestry, University ofBritish Columbia, Vancouver. evenness—the relative abundance ofthe different species.

441
+42 CHAPTER 16 Genetic and Evolutionary Aspects of Ecosystems

each other in their genotype (genetic constitution) fined by the mean and the standard deviation (SD),
and, consequently, in many cases, also in their pheno- which is a measure of the spread of the data, Approxi-
type (behavioral, physiological, and morphological mately 68% of values lie within | SD of the mean, ap-
characteristics). Differences in survival and reproduc- proximately 95% lie within 2 SD, and 99% lie within
tion within the population are associated with varia- 3 SD. A population with an SD of 5 and a mean of 100
tions in the genotype, and those genotypes that is obviously less variable than one in which the SD is 5
produce individuals with phenotypes that are better and the mean 10, so the amount of variability is often
adapted to their physical and biotic environment will expressed as a ratio of the two (ie., 5/100 X 100 = 5%
contribute more offspring to the next generation than in the first case, 5/10 X 100 = 50% in the second), This
will less well-adapted genotypes. Such successful ratio is called the coefficient ofvariation.
genotypes have greater fitness. If the environment The data in Figure 16-1 are from Morgenstern
changes with time, then different genotypes will prove (1969), who investigated the height growth of black
to be best adapted to, or fittest in, these new environ- spruce seedlings of many different geographic origins.
ments. The genetic makeup of the population will His results are shown in the first two columns of ‘Table
change (i.e., evolution will occur), and better adapted 16-1, whereas the third column presents values from
phenotypes will be produced. The process by which the theoretical normal frequency distribution derived
this genetic change occurs is called natural selection. from the raw data. The mean is 11.31 em, and the SD is
In this chapter, the genetic variability of popula- 2.82 cm. The x’ value’ is 9.14, indicating a fairly good
tions and the importance of this variation for adapta- fit between the observed and the normal distribution,
tion are examined, as are the importance of genetic because x7 0.05 with 12 degrees of freedom is 21.0.
variability to forest management and the opportuni- Morgenstern gave values for means and SDs of
ties and problems that it presents. The questions of other morphological and physiological characters for
adaptation and of acclimation are considered here these seedlings. In the case of morphological charac-
only briefly; no attempt is made to cover quantitative teristics, the SDs are better expressed as a percentage
population genetics in this book. The treatment of the of the mean (i.e., the coefficient of variation). For
topic here is intended only to remind you of the im- height growth, this was 24.9%; tor number of cotyle-
portance of genetics in understanding ecology. dons, 7.1%; and for the amount of lammas growth,’
44%. Coefficients of variation for other parameters
generally fell between such limits and averaged ap-
16.2. Genetic Variation in Populations: proximately 25%. For phenological variables, such as
One of the Important Measures of date of bud burst, the SD was only 3.6 days, whereas
Biological Diversity and the Raw for data on growth cessation, it was 18.7 days. It might
Material for Natural Selection be inferred that the date of bud burst is more critical
to successful survival (fitness) than the date of growth
The character of individual organisms in a population cessation and so is more strongly controlled (i.e., there
is varied. In a plantation of Douglas-fir (Pseudotsuga is less variation).
menziesii [Mirb.] Franco), some trees grow faster in All such phenotypic characteristics are ultimately
height and diameter than others, and there may be controlled by the genetic makeup of the individual
similar variation in resistance to attack by insects and seedlings, modified to a greater or lesser extent by the
disease, the number and size of branches, bark thick- environmental conditions. Where does this genetic
ness, wood specific gravity, the frequency and size of variation come from? Before this question is an-
seed crops, and so on. The pattern of variation that is
often observed is illustrated by the variation in height
growth of black spruce (Picea mariana BSP) after two
‘Chi-square (x2) is a distribution of ratios ofvariances (standard devia-
growing seasons in a forest nursery. Figure 16-1
tions squared), often used to test the goodness of fit of data to an assumed
shows the frequency of occurrence of seedlings in var- distribution (a normal distribution in this case). Large x° values indicate
ious height classes in this seeding population. poor fit.
Many variables show this sort of (normal) distribu- ‘Height growth in many temperate zone species is completed quickly and
tion, where most values are clustered symmetrically quite early. A second flush of growth in late summer is called lammas
about the mean. The distribution is completely de- growth, the bread-baking festival “lammas day” falling on August 1,
 Genetic Variation in Populations 443

DB)

j=) |
ie)
—n

S) |

in
seedlings
of
Number
height
each
class

m4

Height class, cm

Figure 16-1 Actual and expected distribution of heights of black spruce seedlings based on
the data from columns 1 to 3 in Table 16-1.

swered, a brief review of some basic genetic concepts mosomes to the zygote (the fertilized egg), which is the
is given for readers who are unfamiliar with them. first cell of the next sporophyte or diploid generation.
Recognition of genetically controlled variation
Sources of Natural Variation among individuals and the relationship between this
Most living cells contain nuclei, the contents of which variation and the evolution of different races and
are clearly separated during cell division into rod-like species dates from the work of the great biologists
structures. Having an affinity for particular dyes, these Darwin, Wallace, and Mendel in the 1850s and 1860s.
are called chromosomes. They consist of deoxyribonu- Much of the knowledge of how this variation arises
cleic acids (DNA), ribonucleic acids (RNA), and pro- has come from much more recent work, and the mod-
teins. In the vegetative cells of higher plants, the ern theory of genetics dates from only the 1930s.
chromosomes occur in (homologous) pairs: for exam- Around 1900, it was realized that the inheritance pat-
ple, 13 pairs in Douglas-fir, 12 in beech (Fagus L.), and terns for various traits described by Mendel could be
a higher and variable number in birch (Betula L.), for explained by chromosome behavior during sexual re-
example, 14 or 28. In mitotic (vegetative) cell division, production if it could be assumed that discrete parts of
the chromosomes are duplicated so that each of the chromosomes controlled specific phenotypic charac-
new cells contains the same number of chromosomes ters. DNA molecules in the chromosomes consist of
as the original cell—i.e., the diploid number—but in two spiral strands, each consisting of multiples of the
reproductive cell division via meiosis, only one mem- organic bases adenine, cytosine, guanine, and thymine
ber of each pair of homologous chromosomes is repre- bonded to pentose sugars, which are connected by
sented in each resulting cell (gametes). These cells are phosphate radicals. Each spiral is held to the other by
haploid. Each parent contributes a haploid set of chro- hydrogen bonds between adenine and thymine and
444 CHAPTER 16 Genetic and Evolutionary Aspects of Ecosystems

Table 16-1 Observed and Expected Frequencies of Black Spruce Seedling Phenotypes and Genotypes

F ee Binomial Expectations
Observed requency in
Phenotype Frequency Height Class Zain 1st Generation 2nd Generation
Height in Height (Normal Parent of Selection of Selection
cm Class Distribution) T t Population Down Up Up

1 0 24 2
2 4 1 23 1.4
3 ) 2 22 D2
4 6 3 21 4 ahle7é
5 2 afea: 4 20 6 18.7
6 3 2.9 5 19 1.8 23.0
7 3 ys, 6 18 4.3 22.2 1
8 10 8.6 TK 17 8.3 Weds) 2
9 14 12.6 8 16 ose 10.8 iG
10 14 glers, 9 15 CAAA 6.1 1.8
11 20 1 10 14 19.9 PAT 4.1
12 1 16.7 11 13 19.1 1.0 7.8
13 14 14.7 12 12 15:5 4 12.4 a2
14 10 10.9 13 11 10.8 ct ilfe2 6
alee: 5 7Gis: 14 10 6.4 19.9 1.6
16 8 4.3 15 9 3.2 19.8 4.0
ie 3 2.4 16 8 1.4 16.4 8.3
18 1.0 17 7 5 11.4 14.5
19 0 18 6 1 6.6 20.9
20 4 19 5 Sia 244
21 20 4 1.1 22.4
22 21 3 3 15.7
23 22 2 1 7.8
24 23 1 PLS
25 24 0 4

Data from Morgenstern, 1969.

between guanine and cytosine to the inside. The base nents of DNA and the existence of different (allelic)
plus the sugar is called a nucleotide. forms of these combinations that is the basis of the
The four bases along a strand can be arranged into natural variation in populations.
triplets in 4° = 64 ways. One or more of these 64
triplets can be arranged in 64N ways to form a gene or
unit of inheritance. Hall (1974) estimated that there is Qualitative Variation
enough DNA in the nucleus of jack pine cells (Pinus As an example of qualitative variation, consider Fagus
banksiana Lamb.) to provide more than 10’ genes or sylvatica L.., European beech, and one of its common
genetic messages! In the simplest case, two genes gov- ornamental forms, var. /arciniata, the cutleaf beech. In-
ern a phenotypic trait, each at a similar position in the stead of having the normal ovate leaf, this form has
homologous chromosomes. These genes may be in deeply lobed, almost cleft leaves, as shown in Figure
several alternative forms (alleles), but often there are 16-2. If the normal and the cutleaf types are cross-
just two alleles of each gene. It is the enormously large bred, then the offspring (F;) may be either all normal
number of possible combinations of the subcompo- or half normal and half cutleaf. Further breeding pro-
 Genetic Variation in Populations 445

Phenotype: Normal var. laciniata var. asplenifolia

Genotype: Ll, Ll, or LG
Aly hl
Figure 16-2 Four leaf phenotypes of beech. The probable genotypes of each phenotype are
indicated. (Based on Lamprecht, 1966; drawings from F.G. Worrall.)

duces the following results. If the normal trees are pigment gene (G) is dominant over no pigment (g).
self-fertilized, then they yield offspring (F,) that are Such a condition is seen less often in plants because
50:50 normal : cutleaf. If two of the F,; normal off- (g¢) for no chlorophyll is lethal, and white seedlings
spring are crossed with each other, then their F, prog- live only a few weeks until food stored in the seed is
eny are in the ratio 3:1 normal: cutleaf progeny. exhausted. Heterozygous individuals, of normal phe-
Further crosses between progeny have to wait for sex- notype, give a 3:1 ratio of normal: white progeny.
ual maturity (approximately 40 years in beech, which The fastigiate’ crown form, as seen in lombardy
illustrates why progress in tree genetics and tree poplar (Populus nigra var. Italica Muench.) and in culti-
breeding is relatively slow and why Mendel’s choice of vars of such species as black locust (Robinia pseudoacacia
peas for his studies was a wise one). However, we may fastigiata Dieck.), seems to be recessive to the normal
assume that the inheritance patterns would be exactly oval crown forms. In the former case, the fastigiate
the same as those found by Mendel, which are shown form has erroneously been assigned varietal status,
in Figure 16-3, where L is the gene for normal leaves, whereas actually it is merely a clone (or clones).
which is dominant to the recessive allele (/) for cut- There is variation in the resin color in slash pine
leaves. This type of diagram is called a Punnett square. (Kraus and Squillace, 1964). Individuals occur rarely that
Many other traits in trees presumably are of this have yellow resin, and when these are bred with each
type, where a single dominant allelic form of a gene other, they produce progeny with yellow resin. When
masks the recessive allelic form of that gene and there crossed with normal trees that have colorless resin, the
are only two phenotypes: the dominant and the reces- progeny usually have normal resin. Yellow resin is the re-
sive. The latter phenotype is seen only when the re- sult of the homozygous recessive condition.
cessive allele is present on both chromosomes. There Segaard (1969) studied disease resistance in Thuja
are numerous examples of this. Leaves vary in color, L. (arbor-vitae) species. Western redcedar (or giant ar-
such as in the very dark-leaved forms of European borvitae; Thuja plicata D. Don.) from western North
beech. In this “copper” beech, the gene (C) for antho- America is susceptible to the serious cedar leaf blight
cyanin production is dominant over (¢), no antho- (Didymascella [Kiethia] thujina (Dur.] Maire), whereas
cyanin, so that both CC and Cc genotypes have
copper-colored foliage, whereas cc genotypes have
normal green foliage (Blinkenberg et al., 1958). Most Fastigate: an upright crown form, where laterals grow almost vertically,
people are familiar with albinos (gg) in animals; the and are almost as long as the leading shoot.
446 CHAPTER 16 Genetic and Evolutionary Aspects ofEcosystems

Parental phenotypes:

/ Selfed

All normal 1:1 normal: cutleaf 3:1 normal: cutleaf

F, phenotypes

Figure 16-3 Inheritance patterns of the two beech leaf phenotypes, normal and cutleaf (va-
riety laciniata).

the closely related Thuja standishii (Gord.) Carr. produced indistinguishable phenotypes. With this
(Japanese arbor-vitae) is not. Despite that these type of inheritance, the phenotype is a direct reflec-
species are separated by the Pacific Ocean and have tion of the genotype in all but the most unusual envi-
been distinct for some time, they hybridize, and these ronments and does not reveal variations in the
hybrids are fertile. The hybrid is resistant to the intensity of action of a gene. For example, the (G)
disease. The self-fertilized (“selfed”) hybrid yields gene controls the production of a catalyst (an enzyme)
progeny with a 3:1 ratio of resistant : susceptible. that is involved in the synthesis of chlorophyll. Varia-
The backcross’ to T. standishii is resistant; the back- tions in the abundance of the catalyst make no differ-
cross to 7? plicata has a 1 : 1 ratio of resistant to suscep- ence to the rate of production of chlorophyll: it is all
tible plants. Evidently, this trait is governed by a single or nothing. Under special conditions, phenotypes may
pair of genes, with the homozygous recessive being not reflect the genotype. If one were to grow seeds
disease susceptible. with genetic makeup GG, Gg, and gg (no catalyst) in
The European beech also serves to illustrate an- the dark, all would produce white seedlings that would
other point. There is a variety that has straplike be phenotypically indistinguishable. This situation is,
leaves—var. asplenifolia (Figure 16-2). This is also due of course, a little artificial.
to a double-recessive condition, but the gene responsi- In natural populations, examples of all-or-nothing
ble is a third allelic form of the leaf-shape gene. Some action of genes are relatively uncommon, but they
genes exist in a whole array of allelic forms, three in have been widely studied because of their simplicity
this case. and have helped to elucidate the basic mechanisms of
genetic inheritance. The more common situation is
Quantitative Variation one in which phenotypic traits show a whole range of
values. Many traits are quantitative.
In the examples given above, phenotypes governed by
a pair of genes were of only two kinds, because the ho- Copper beech is not the only tree to have abnor-
mally pigmented foliage. Many ornamental varieties
mozygous dominant and the heterozygous genotypes
of other tree species have unusual and attractive col-
"Backcross: a hybrid crossed with a parent. ors. Yellow foliage may result from a less-than-normal
 Genetic Variation in Populations 447

level of chlorophyll, so the carotene pigments are not coding of a particular allele. For example, the G gene
masked as much as they usually are. Such individuals for chlorophyll production constantly mutates to g for
are not normally very fit in natural environments, and albino, and vice versa, but the g gene is continuously
they generally survive and reproduce only by the in- removed from the population because of its zero fit-
tervention of horticulturalists. However, occasionally ness in the homozygous state. On a larger scale, chro-
trees with yellow foliage are found in the wild, for ex- mosomes may break and rejoin in different order.
ample, the “golden” sitka spruces (Picea sitchensis They may be lost or gained in whole or in part. Whole
(Bong.) Carr.) of Queen Charlotte Island off the north sets of chromosomes may be gained when, during du-
coast of British Columbia. There is also an aurea vari- plication, cell walls fail to form. This yields polyploid
ety of Norway spruce, Picea abies (L.) (Karst) (Langner, cells, as has been mentioned for birch, where the chro-
1954). In both these cases, as before, only two genes mosome number may be 27 = 28, or 2n = 56 (ie.,
are involved, but the three genotypes give rise to three tetraploid) or even hexaploid or octoploid. Such ge-
instead of two phenotypes. The G gene for chloro- netic variation is constantly arising, and many of mod-
phyll is not dominant over g, albino, so the heterozy- ern technology’s chemicals seem to accelerate it in a
gote (Gg) does have some chlorophyll but not as much disastrous way (i.e., they are powerful mutagens).
as the homozygote (GG). In this case, the effects of The genetic variation present in a population or an
each G gene are additive, the amount of chlorophyll individual may be expressed immediately as pheno-
depending on the number of G genes present. GG is typic variation. ‘Irees that contain a segment with dif-
normal, Gg is aurea, and gg, as before, is albino. Self- ferent leaf color or form or different phenology or
ing of the aurea phenotype yields all three phenotypes, branching pattern are seen commonly, this being the
in the ratio 1:2:1 green: aurea: albino, as can be result of a somatic mutation in a meristem. More
checked by drawing a Punnett square of the type in often, however, the variation is not expressed until the
Figure 16-3. This particular genetic system is well next generation(s), when independent segregation and
known, because Langner was able to trace the distance assortment of genes produces new genotypes that ex-
and amounts of pollen flight from isolated aurea trees, press themselves phenotypically (e.g., Figure 16-3, or
because the progeny with adjacent normal trees will be the aurea spruce mentioned above, which when selfed
in the 1 : 1 ratio of green to aurea. This is the simplest yields three phenotypes). ‘Thus, constantly accumulat-
case of a quantitative trait. Another is the heterozy- ing genetic variation is expressed via sexual reproduc-
gote /,/, of beech, which has leaves intermediate to /,/; tion, and on this resulting phenotypic variation, natural
cutleaf and 1, asplenifolia (the fourth leaf in Figure (and artificial) selection can act. It has been estimated
16-3). that without sexual reproduction, evolution would pro-
This simple case can be taken a step further. For ceed at less than 1% of its present rate.
instance, in domestic fowl, there is a phenotype “friz- There is plenty of scope for selection, because nei-
zle” that has curly feathers. This is homozygous (FF). ther the seed or the seedlings resulting from sexual re-
When crossed with a normal fowl (ff), all the progeny production in plants nor the offspring of animals all
are mildly frizzled (Ff), so there are the three pheno- survive. At equilibrium, each mature organism is re-
types, each produced by a different genotype. The F placed by only one of its progeny. In slash pine stands,
gene adds a certain amount of frizzle to the Ff het- each tree produces approximately 1,500 seeds per
erozygote. However, there are other deviant feather year, and approximately half a million might be pro-
types, from very mild frizzle all the way up to absurdly duced altogether during the life of a tree (Florence
frizzled. Two pairs of genes govern this trait, so there and McWilliam, 1956). In small fruited species such as
can be five genotypes and five phenotypes: fff, nor- red alder (A/nus rubra Bong.), there are close to 10
mal; Ffff, mild frizzle; FFff, frizzle; FFFf, very frizzled; million potential progeny throughout the life of the
and FFFF, grotesque (Hutt, 1949). parent. Few seeds will germinate, because most will
The examples of qualitative and quantitative in- fall on unsuitable sites or succumb to rodents or fungi
heritance described above show that there is plenty of and insects. Those that do germinate are not likely to
variation of both phenotype and genotype. Where survive a full season; very few will grow to reproduc-
does the variation in genetic structure come from? It is tive age; still fewer will reach maturity. At every stage
the result of mutation, caused perhaps by radiation or of their life, organisms have many physical and biotic
by miscopying during chromosome duplication dur- factors working against their survival, and only those
ing cell division. This involves a change in the actual individuals that have the phenotypic characteristics
448 CHAPTER 16 Genetic and Evolutionary Aspects of Ecosystems

that make them well adapted to their environment tion. Selection that produces diverging means is called
(and some good luck) will survive and reproduce. directional selection. It is interesting that new types are
In summary, there is variation, some of which is appearing, types that did not exist in the original pop-
genetic and is inherited. Not all progeny survive, so it ulation. For example, there were no plants as tall as 24
seems that the stage is set for selection, adaptation, cm in the original population, but these are present
and evolution. Have they happened? after selecting for tallness for two generations. In the-
ory, at least, it seems that a variable population can, as
a result of selection, yield forms that could not have
16.3. Examples of Selection, been predicted.
Adaptation, and Evolution The selection gain just described is theoretical
only. In practice, the selection gain expected from the
Human-Caused Selection Pressures selection differential is rarely if ever obtained. This is
for many reasons, perhaps the most important of
The Theoretical Case. If selection pressures based
on tree height were applied to the black spruce which is that the individual effects of genes are not
seedlings of Table 16-1 and Figure 16-1, then some strictly additive. For instance, in the aurea spruce vari-
trees would be eliminated because they proved to be eties, the phenotypes corresponding to the genotypes
unfit under these particular pressures. As a result, the that have two genes for chlorophyll (GG), 1 gene for
proportion of T genes in the remaining population chlorophyll (Gg), or no genes for chlorophyll (gg) do
would change. For example, if all seedlings greater not contain chlorophyll in amounts 2: 1 : 0. Instead,
than 9 cm tall were destroyed, then the proportion of the ratio is closer to 7: 1:0. The realized gain ex-
T genes, p(I), in the remaining population of 18 pressed as a proportion of the selection differential is
seedlings would become <0.2338, and their mean called the heritability, and for many traits in trees it is
height would be 6.37 cm. The selection differential approximately 25%.
(the original population mean minus the mean of the
selected individuals) is 11.31 — 6.37, or 4.94 cm. If they Actual Examples of Selection Pressures
were of reproductive age (actually approximately 15 Intentional. One has only to look at the products
years in spruce), then these selected seedlings would of many generations of selection by agriculturalists to
produce a population of offspring that had a height see that the theoretical prediction made above (that
distribution (at the same age) as shown in column 7 of selection in a variable population can yield unexpected
Table 16-1; they would retain the same mean as the phenotypes) is also true in practice. Domestic animals
parent population (6.37 cm). This distribution was cal- and plants can hardly be recognized as being the same
culated using the binomial’, as was the theoretical dis- species as their wild ancestors. Compare, for example,
tribution for the original population. Similarly, we the huge, red, shiny, juicy (and almost totally tasteless)
could select against small size and promote tallness in fruit of McIntosh apples, with those of its wild ances-
the population by selecting the tallest: the top 16 indi- tor, Malus pumila Mill., the common crabapple of Eu-
viduals in the original population. These have a p(T) rope. Darwin was fascinated by the variability of
value of 0.5972 and a mean height of 14.3 cm. The pigeons and bred them to investigate the heritability
theoretical distribution of their progeny is shown in of the variation in these birds. The modern “fancy”
column 8 of Table 16-1. If selection for tallness were breeds that have resulted from over a century of selec-
to be continued by selecting only those individuals in tive breeding bear little resemblance to the ancestral
the second generation that were taller than 18 cm (the- rock dove (Columbia livia Gmelin) and in fact make
oretically, there would be 11.2 individuals with a p/T] one hope that we do not ever initiate selective breed-
value of 0.786), then the resulting progeny would have ing programs in humans. Who knows what the results
a distribution of height growth as shown in column 9. might be? Such spectacular results of selection are also
These distributions are shown in Figure 16-4. The the aim of tree breeders.
means are diverging from that of the original popula- Figure 16-5 shows the results of selection, both for
and against, DDT resistance in the fruit fly (Drosophila
melanogaster Meigen). Within 15 generations, approx-
In the binomal distribution, an event has the probability p of success, imately 25-fold changes were produced in this charac-
ql! — p) offailure. Ifrepeated n times, the number ofsuccesses r = (Geeks teristic, both up and down. This indicates the
 Examples of Selection, Adaptation, and Evolution 449
Downward selection Upward selection
<~— Original population mean
25

Original population
Population after one generation
20 | \
| \ / \ ~~ 0 of downward selection
ie) \ \ a—-—a_ Population after one
| \ generation of upward selection
|
| \ e----@ Population after two
generations of upward selection

of
Number
in
seedings
class
height
each

10 12 14
Height class, cm

Figure 16-4 ‘Theoretical distributions of seedling height after one generation of downward
selection and one and two generations of upward selection. The data are from columns 6 to 9 of
Table 16-1.

tremendous potential that insect populations have for selection. In accidental as opposed to deliberate
adaptation against human efforts to control them and human-induced selection, there is a free exchange be-
why we will probably always have to share the envi- tween the selected and original populations.
ronment with “pest” insects. The classic case of human activities causing differ-
It is disturbing to consider the same sort of entials between populations is that of industrial
“progress” that has been made in disease-causing bac- melanism in the peppered moth (Biston betularia Gue-
teria, many of which are now resistant to antibiotics. nee), which lives in the woodlands in England. There
Such resistance can be induced rapidly merely by ex- is a good record of the phenotypic appearance of this
posing the parent bacteria to low doses of such antibi- moth over the past 150 years because specimens of this
otics in the process of screening out the susceptible species are to be found in moth and butterfly collec-
ones (see the coevolution example below for an expla- tions made periodically throughout the period. Most
nation of the possible mechanism involved). of the specimens dating from the beginning of the
19th century are of the “typical” variety, which is
Accidental. In natural environments, popula- whitish, speckled with gray and black—hence the
tions undergoing selection are not artificially kept sep- name “peppered.” A dark-colored, or “melanic,” vari-
arate from one another as is the case in experimental ant occurred in the early collections, but apparently it
 450 CHAPTER 16 Genetic and Evolutionary Aspects of Ecosystems

however. Pollution control programs have signifi-

Selection for high resistance cantly reduced air pollution in some areas of Great
125.0 Britain, with parallel increases in the frequency of the
typical form (Cook et al., 1970).
= 25.0 Similar accidental selection has resulted in mos-
quito populations that are resistant to DDT, which is
i]

5
? 5.0 weet Cx Selection for low resistance no longer effective in the control of malaria in areas in
E
la
a \
which DDT has been used extensively. The produc-
at 1,0 \
Gaue /
tion of the resistant strain followed much the same
/
path as depicted in Figure 16-5 for the fruit fly. A sim-
0.2 be ilar phenomenon, which is of considerable interest to
foresters and wood users, is the evolution, after 50 or
0.04 Te=vsie litt earl Gc ee so years of selection, of forms of shipworm (Teredo ner-
Ole 2. 34 eG 1S SO L0H IO aS
valis) that are not affected by creosote. This means that
Number of generations
wooden marine installations must now be protected by
Figure 16-5 Variation in resistance to DDT in popula- some other method, such as pentachlorophenol.
tions of the fruit fly, Drosophila melanogaster, subject to se- There are comparable examples in plants of evolu-
lection over 15 successive generations either for or against tion induced by our efforts to control pests and by the
resistance to DDT. Resistance was increased by a factor of ap- widespread pollution of the environment. It has been
proximately 40 in the former and reduced by a factor of ap- shown (Marriage and Warwick, 1980) that in response to
proximately 6 in the latter. The technique involved selection of herbicide application, pigweed (Chenopodium album)
genetic variation that was already present in the original popu-
types that are resistant to the chemicals involved have
lation (sib-selection). (After Bennett, 1960. Reproduced from
evolved. Bradshaw (1952) showed that races of grasses
Heredity by permission of Genetical Society of Great Britain and F.
Bennett.) such as Agnostis tenuis that are so tolerant of high concen-
trations of heavy metal that they can grow on mine tail-
was rare. Selection pressure from predators would ings rich in such elements as zinc and lead have evolved.
have favored the typical form because of its better
concealment as it rested on the lichen-covered bark of
tree species such as birch (Figure 16-6). The melanic
Natural Selection Pressures
form did not remain rare, however. Its frequency in- A species is a genetic grouping of organisms (a gene
creased steadily over the past century until in many pool) whose genetic constitution (genome) is sufficiently
parts of England it became the more common form. homogeneous to allow sexual reproduction (gene flow)
Only in the less industrialized parts of Britain did the between individuals of the group (species) but differ-
typical form retain its earlier frequency. ent enough from that of any other species to preclude
The change in the relative frequencies of the typi- or severely limit gene exchange between the groups.
cal and melanic form paralleled the increase in air pol- In fact, this definition often does not work well, be-
lution that occurred in Great Britain over the last cause individuals from opposite ends of the range of a
century. Increasing air pollution led to two important widely distributed species will not breed with each
changes in the environment of the peppered moth: the other and so could be called separate species. How-
killing of the pale-colored lichens that covered dark ever, gene exchange between these remotely located
tree trunks, lichens being extremely sensitive to SO), individuals can occur by breeding them with individu-
and darkening of the pale tree trunks by the deposi- als from the middle parts of the range.
tion of soot (carbon particles). These changes in the Because most species have a geographical range
moth’s habitat improved the concealment of the that includes considerable variation in the character of
melanic form from avian predators as the moths rest the environment, different local populations of a
on the tree trunks but virtually eliminated conceal- species will experience either slight or significant dif-
ment for the typical form. The resulting changes in ferences in selection pressure. The continuing opera-
selection pressure altered the relative fitness of the tion of this natural selection will result in the survival
two forms and led to a change in their frequency in the of those individuals and their offspring whose genetic
population. This change has not been permanent, makeup best suits them for life and reproduction in
 Examples of Selection, Adaptation, and Evolution 451

Unpolluted Area
Polluted Area
Moths on a tree trunk in
: : : Moths on a tree trunk in
an area without air pollution an area with heavy air pollution
Variants of Moth

O typica form
@ carbonaria form >
©) insularia form Gy

Peppered
% form %
100 100 Dark form

80

60 : aad
% frequency of dark and peppered forms
40 in local populations

20 Dark
form
0
%
30

20 20
% recapture of marked dark and peppered
forms released in woods
10 10

0 0
% %
100 100

80 80

60 % of dark and peppered forms in observed 60

cases of predation by birds
40 40
20 20

0 0

Figure 16-6 The effect of changing selection pressure on the frequency of dark (melanic or carbonaria) and peppered (typ-
ica) forms of the peppered moth. The photo on the left shows adults of this species at rest on the lichen-covered bark of a tree
trunk in an unpolluted area. The photo on the right shows two individuals on a tree trunk from which lichens have been eliminated by
air pollution and that has been darkened by the deposition of air-borne soot. The change in the frequency of the two forms is proba-
bly related to predation by birds that use visual clues in searching. (Data from Kettlewell, 1956, 1958; photos from the experiments of Dr.
H.B.D. Kettlewell, Oxford University. Reproduced with permission from Heredity.)
452 CHAPTER 16 — Genetic and Evolutionary Aspects ofEcosystems

their particular environment (i.e., gives them a high Clausen et al. (1948) (see also the review of the physi-
fitness). As Darwin put it, life involves a “struggle for ology of races and species by Hiesey and Milner
existence,” leading to “survival of the fittest,” and this [1965]). This and other examples are discussed by
results in the development of locally adapted geno- Jones and Luchsinger (1979). Yarrow is a herbaceous
types or ecotypes, a genetic subdivision of a species in plant that grows widely across California from sea
response to local environmental conditions. level to elevations over 3,000 m in the Sierra Nevada.
The classic example of the presence of genetic The tallness of the plant varies greatly over its range
variation within a species between different parts of its (Figure 16-7), with a general decrease in size from the
range is found in the studies of yarrow (Achillea sp.) by San Joaquin Valley toward the Pacific Ocean and with

=B Frequency of different
150 4 3 Bnitin local populations

Frequency

population,
cm
504
natural
in
plants
of
Height

0 WN mm
‘
Bodega San Gregorio
ee ae
Clayton
tt Xe
Selma Knights Groveland
WAVAspen
Mather
2 VEXtz ~
N wv
==
WZ
~y
Yosemite Tenaya Tuolumne Big Horn Timberline Conway Leevining
Ferry Valley Creek Lake Meadows Lake; Cos
3600 =

> 2400 | | |
Pacific ; Sierra Nevada Mts.
Coast Mts.
1200 4 ee San Joaquin Valley
level,
msea
above
Elevation 0
(a)

Knights
Ferry Groveland Mather
Grown at Grown at Grown at
80 4
' q /.,&
(?
Mather
b
— Stanford
elev_90 m
7 f <4
f
Mather
elev. 1400 m
| Timberline
elev. 3050 m
| \
oR nie Groveland; x
i Ee:
60 4 / = q SESS :
y ioe Knights WY? i. Re Tuolumne =
ee) E V. Tuolumne Ferry iy ‘Tuolumne Meadows
40 4 ( Meadows r _Meadows i Groveland Mather
ia /
ap |

gardens,
cm ‘ a Big Horn Y | \ Big Horn} Knights 4
mol { Lake — Bh ea Lake,
plants
of
Height
in
grown
0 - ; Na NY NZ, :
Sea Level 900 1400 2600 3300 SeaLevel 900 14 2600 3300 SeaLevel 900 1400
Elevation from which seed was collected, m
(b)

Figure 16-7 Evidence for the existence of ecotypes: the classical study of variation in height
growth of yarrow (Achillea millefolium) (after Clausen et al., 1948). (A) Variation in the average
height growth of ecotypes of yarrow along a west-east transect across California. The variation in
height growth around the average for these local populations is also shown. (B) Height growth of
five ecotypes from across the elevational range when they were grown together in gardens at three
elevations. A stylized representation of the plants is used. (Reproduced by permission of Carnegie Insti-
tute of Washington).
 Examples of Selection, Adaptation, and Evolution 453

increasing elevation eastward. This phenotypic varia- coastal provenances fail completely when grown in the
tion does not prove the existence of ecotypes (i.e., interior of the province because of frost injury. The
local genotypic variation) because many species have explanation is the same in both cases. Coastal prove-
the capacity to acclimate (i.e., produce different phe- nances cease growth and acclimate to the coming win-
notypes from a single genotype) when grown in dif- ter when the length of the shortening daylight period
ferent environments. Acceptance of the existence of is approximately an hour longer than it will be when
yarrow ecotypes requires demonstration that these the first killing frost will occur. They still do this when
differences (or others) are heritable and that they will grown in the interior, but unfortunately the first frost
appear when plants from across the range are grown in this region occurs approximately a week before the
together in a single environment (i.e., in a provenance critical coastal day length occurs. Because of their
test). adaptation to local conditions, interior provenances
‘To test for ecotypes, Clausen et al. collected seed cease growth and become frost hardy earlier than the
from several locations (provenances*) along the altitudi- coast provenances, when the day length is still rela-
nal range of yarrow and planted them in prepared tively long, and so are not damaged. When grown on
seedbeds at sea level (Stanford) and at medium- the coast, these provenances cease growth unnecessar-
(Mather) and high-elevation (Timberline) sites. These ily early and so do not grow as well as the coastal
experiments showed that height growth is indeed in- provenances. After growth cessation, the coastal and
herited (that in this case, the different provenances are interior provenances show little difference in absolute
different ecotypes) but that the genetically controlled frost hardiness. Their differences are associated with
variation in height growth is greatly modified by the fitting their period of activity into the growing season
environment. The ecotypes performed best (if tallness to which they have become adapted. Similar differ-
is considered “best”) on sites closest to the elevation of ences occur between north and south provenances.
their origin (in many species, this is not the case). Such photoperiodic’ ecotypes were originally shown
Moving ecotypes from low to high elevation was gen- in grasses and trees by Vaartaja (1959).
erally more deleterious than the reverse, indicating Other examples of ecotypic variation in trees in-
that it is easier for an ecotype that is adapted to a se- clude variation in height growth of ponderosa pine
vere environment to exploit a moderate one than vice (Pinus ponderosa Laws) from different elevations and
versa. The survival of the yarrow ecotypes in environ- differences in photosynthesis, respiration, and leaf
ments that were markedly different from their loca- characteristics of sugar maple (Acer saccharum Marsh)
tion of origin indicates that this species has a growing at different elevations but separated by a dis-
considerable ability to acclimate. However, this was a tance of only 0.8 km (Ledig and Korbobo, 1983).
controlled experiment under ideal conditions (gar- We have seen that local differences in the physical
dens, where competing vegetation was eliminated). environinent give rise to locally adapted genetic sub-
In natural conditions, the “foreign” ecotypes would populations called ecotypes. Where there is a sudden
likely be eliminated by competition from other change in environmental conditions, such as in steep
species (see the discussion of ecological niches in mountain country, or where there is a sudden change
Chapter 14). in the chemical and physical properties of soil, local
A similar example for trees occurs in species, such ecotypes may be clearly differentiated. An example of
as Douglas-fir and black cottonwood (Populus tri- the latter is the local ecotypes of various plants grow-
chocarpa Yorr and Gray), that grow in areas that differ ing on serpentine soils (rich in magnesium) (Krucke-
widely in the length of the growing season (the num- berg, 1954). Similarly, if there is a sudden change in
ber of frost-free days). Provenances of these species (in the biological significance of a physical factor along a
this case, ecotypes) from the continental climate east gradient of gradual change in that factor (e.g., 0°C
of the coast range in British Columbia survive when along a gradient of temperature), then clearly differen-
grown on the coast but perform poorly. Conversely, tiated ecotypes will occur. However, where the spatial
variation in the physical environment is very gradual,
there will be a similar gradual change in the genetic
‘The term provenance differs from the term ecotype: the former refers to
the geographical location in which a genotype has evolved; the latter is the
genotype that is adapted to the conditions ofthat location. ’Photoperiodicity is discussed in Chapter 7.
454 CHAPTER 16 Genetic and Evolutionary Aspects of Ecosystems

constitution of the population. Such a continuous and Dougl.) pines seems likely, but in this case, the jack
gradual change in genotype along an environmental pine spread west on the north side of the prairies and
gradient is called a cline. Objective delineation of the is now sympatric with lodgepole in two locations. In
geographical distribution of different ecotypes along a these areas, it hybridizes with lodgepole pine, produc-
cline is difficult or even impossible. ing fertile hybrids. Presumably, evolution to true
Unlike different species that are normally geneti- species would have occurred if the populations had re-
cally isolated from each other, different ectoypes are mained separated for a longer period.
capable of interbreeding. Sometimes the offspring There are also striking pairs of similar species on the
that result are larger and more vigorous than either west coast of North America and the east coast of Asia,
parent. This phenomenon is known as hybrid vigor and such as Japanese (Acer japonicum Thunb.) and vine (Acer
is widely used in the breeding of improved strains of circinatum Pursh.) maples. Did these originate from a
agricultural and horticultural crops and also in tree species that inhabited the Bering area before the conti-
breeding. More frequent, the hybrid offspring have nents became separated by the straits? Similarly sugges-
characteristics that are intermediate between those of tive are the North Pacific distribution of the Psewdotsuga
the two parents. Hybrid populations generally contain genus and the existence of only two Liriodendron species,
more variation than the parent populations. This in- yellow poplar (Liriodendron tulipifera L.) in North
creased variability increases the possibilities for select- America and L. chinense Sarg. in China.
ing new and more fit combinations of alleles. The average conditions of the physical environ-
Hybridization thus increases the potential for evolu- ment are reasonably stable. Changes in regional pre-
tion and adaptation. cipitation or temperature regimes and changes in soil
Do species and ecotypes both evolve the same conditions generally occur so slowly that even long-
way? [he circumstantial evidence from plant geogra- lived organisms have ample opportunity to adapt to
phy is very strong, because there is a remarkable num- them. Year-to-year variations in weather pose greater
ber of pairs of similar species that seem to have a evolutionary difficulties. A period of 10 years of mild,
common ancestor and whose range was split into two moist weather will favor genotypes that may be poorly
parts by a geographic feature or climatic event. In adapted to the 2 years of drought and late spring frosts
North America, we have eastern white pine—western that follow. Such unpredictable periodic fluctuations
white pine (Pinus strobus L.—Pinus monticola Doug].), in weather tend to maintain a high degree of genetic
eastern hemlock—western hemlock (Tiuga canadensis diversity in populations that have an annual life histo-
[L.] Carr—Tsuga heterophylla (Rafn.] Sarg.), Canada ry and which are entirely replaced every year. Such cli-
yew-Pacific yew (Taxus canadensis Marsh.—Taxus brevi- matic fluctuations promote phenotypic adaptations
folia Nutt.), and so on. It is thought that each pair had such as acclimation in long-lived organisms (e.g.,
a common species as an ancestor, whose original trees), which are generally less variable genetically
transcontinental range was split by the formation of than the annual species.
the prairies. This split prevented gene exchange, and
the different selection pressures in the two subpopula-
tions, acting on varying phenotypes caused by differ- 16.4 Coevolution and
ent randomly occurring mutations, eventually the Biotic Environment
produced separate species. Fraser fir (Abies fraseri
Poir.) is very similar to balsam fir (Abies balsamea Mill.) Factors of the physical environment cannot on their
and perhaps originated when populations adapted to own explain evolution. Natural selection is also guided
low temperature migrated north or up the mountains by environmental differences that are caused by living
of the eastern central United States after glacial peri- organisms. Consider, for example, the evolution of
ods to stay within the temperature range to which they species of land plant that do not require much light.
were best adapted. The mountain populations eventu- Before there were large land plants, there were virtu-
ally became separated from the main population by ally no shady environments for our present shade-
low-altitude warm areas and hence developed into a tolerant plants to exploit. They evolved when shady
separate species. The same sort of origin for jack environments were created by other plants, but cre-
(Pinus banksiana Lamb.) and lodgepole (Pinus contorta ation of new physical environments is not the only way
 Coevolution and the Biotic Environment 455

in which the biota have influenced evolution. Interac- disastrous effects of introduction of nonnative organ-
tions between individuals of the same species and be- isms of all kinds. The examples of chestnut blight and
tween individuals of different species have also been Dutch elm disease in the eastern United States are
extremely important. Species evolve in a manner that well known, as is the recent case of the reintroduction
promotes beneficial interactions and/or favors a par- of a more virulent form of the latter disease back into
ticular species in some antagonistic interaction such as England. It seems likely that few of the present Eng-
competition or predation. This topic was explored in lish elms will survive the current scourge (this example
Chapter 15. was discussed in Chapter 15).
Consider, for example, the relationship between a The alternating evolution of two species of organ-
population of herbivores and the population of plants isms in response to mutual selection pressures is called
that they are eating. Because of natural variation in coevolution. Sometimes coevolution results from at-
plant populations, some of the plants will be less at- tempts by one species to reduce the impacts of anoth-
tractive, less palatable, or less nutritious to the herbi- er species, as in the example just given. Sometimes it
vores and will be less utilized by the herbivores than involves attempts to strengthen relationships between
will the other members of the population. These two species. For example, the development of a plant
plants will tend to survive better and produce more characteristic (e.g., nectar production) that is benefi-
seed than will the more heavily utilized individuals. cial to a species of animal (e.g., a nectar-feeding insect)
Over many generations of herbivore pressure, the fre- may lead to improved pollination of that plant. This
quency of these more fit individuals in the plant popu- will generate selection pressures in the plant popula-
lation will increase. This will lead to a general tion favoring the genotypes that produce the most
decrease in the suitability of the vegetation for the nectar and in the insect population for those geno-
herbivores whose abundance will gradually decline. types that select this genotype in preference to others.
However, this will generate selection pressure on the Such a sequence of mutually beneficial developments
herbivores, leading to above-average survival and in- can lead to a high degree of specificity between certain
creased fitness of those herbivore genotypes that can species of plants and animals. The importance of this
successfully utilize the less desirable plant form. After type of evolutionary process is seen in the many amaz-
many generations, the frequency of these tolerant her- ing adaptations of plants and insects to each other, ex-
bivore genotypes will increase, leading to a gradual re- amples of which were examined in Chapter 15.
covery in their abundance and a renewal of Coevolution is an important concept. Because
herbivore-induced selection pressure on the plant most of the species in a particular biotic community
population, and so on. have coexisted for thousands or even millions of years,
This sort of evolution is very important in forest it is most likely that some degree of coevolution will
pathology (forest diseases), for instance, the evolution have occurred, whether this has been to favor or to
of white pines and the fungal disease white pine blister impair relationships between species. The addition or
rust (Cronartium ribicola Fisher). This Asian disease removal of species to or from the ecosystem therefore
has historically applied pressure to its white pine can be expected to influence both the performance
hosts, which resulted in more resistant Asian white and the evolutionary development of the remaining
pines. These resistant hosts obviously applied pressure biota. Good survival and high productivity in its native
to the blister rust, which resulted in a more virulent habitat is no guarantee that an organism will perform
disease. After many generations, a very virulent form similarly if transferred to a completely different biotic
of the disease developed but did not do a great deal of environment. Similarly, the rarity or unimportance of
damage to its now highly resistant host. However, in an organism in its native environment is no guarantee
North America, white pines have not been selected for that it will not become a pest species or have great
resistance to blister rust, which is not native, and it is economic potential when introduced into a new envi-
now apparent that very few, if any, of the North Amer- ronment. A change in the biotic environment of an in-
ican white pines have any resistance to the disease. dividual can have as great or even greater effect on a
Large-scale destruction of white pine populations oc- population than can a change in the physical environ-
curred when the disease was introduced into this con- ment. The character of an organism, in a particular
tinent. There are numerous other examples of physical environment, is to a considerable extent a
456 CHAPTER 16 Genetic and Evolutionary Aspects ofEcosystems

function of its adaptation to the other biotic compo- a new pest genotype that is not affected by the control
nents of that physical environment. measure. Perhaps the best examples of management-
Coevolution has undoubtedly played a major role induced counterevolution is the evolution of resis-
in the evolution of many different kinds of organisms. tance to chemical poisons in populations of insect
The continuing pressure of competition for energy pests. It is clear that chemical control measures that
and material resources and of exploitation by other or- must be applied frequently to a pest that has a rapid
ganisms has selected from among the natural variation potential rate of evolution constitute at most a very
within populations those genotypes with the greatest temporary solution to the problem. Figure 16-5
fitness. Divergence in morphological, physiological, showed how DDT resistance in a population of fruit
and behavioral characteristics has resulted, leading ul- flies can be raised or lowered by artificial selection.
timately to the formation of genetically isolated Rates of evolution of crop species can be speeded
species. The combination of adaptations to the enor- up artificially. By selecting seed from trees with desir-
mously variable physical environment and adaptation able properties such as wood strength; growth rate; or
to other organisms that struggle to survive in that en- resistance to frost, disease, or insects, the forest ge-
vironment is responsible for the enormous diversity of neticist can alter the genetic makeup and thus the phe-
life that exists today. notype of the population to a condition that has
improved economic value. In terms of improved tech-
nical characteristics and in terms of improved adapta-
16.5 Evolutionary Time Scales tion to climatic factors (which have a long time scale
of directional change), such accelerated evolution is
The rate at which evolution can occur is highly vari- likely to have a beneficial outcome. In terms of adapta-
able. It depends on several factors and has been dis- tions to factors of the biotic environment that are
cussed by Pollard (1979). Ultimately, new variation in themselves evolving, the outcome of artificial manipu-
a population is the result of mutation, and major evo- lation of evolution is less predictable. Breeding resis-
lutionary change is limited by the rate at which new tance in trees to diseases and insects (accelerated
mutations enter the population. However, smaller counterevolution) undoubtedly offers a potential solu-
evolutionary adjustments to the genetic makeup and tion to some insect and disease problems, but it may
the fitness of a population are the result of the varia- merely lead to counterevolution by the pest and only
tion that results from the rearrangement of existing temporary relief from the problem. In agricultural
genetic information during sexual reproduction. The crops where the generation time of the crop is not
potential rate of evolution and adaptation is therefore greatly different from that of the pest, the problems of
closely related to the frequency and abundance of sex- coevolution may be manageable. In forest crops,
ual reproduction. An insect species with 10 genera- which have generation times on the order of decades
tions a year will tend to be capable of faster genetic or centuries, the possibility of counterevolution by
change than will a tree with one generation per centu- pests with annual generations poses potential difficul-
ry, but it will exhibit slower genetic change than will a ties to long-term genetic solutions for pest problems.
species of bacterium or virus that may have hundreds
of generations a year. A plant species that produces
many thousands of seeds per year will generally be ca- 16.6 Implications of Genetic
pable of faster genetic change than will a plant species Variability, Evolution, and
with an annual production of a dozen seeds. Adaptation for Forestry
The potential rate of evolution of an organism has
important consequences for ecosystem management, Recognition of genetic variability and adaptation has
because humans act as a potent force of natural selec- proved to be as important for forestry as it has for bi-
tion and our activities may deliberately or accidentally ology in general. All too often, poor survival, low pro-
accelerate rates of evolution. Action by farmers or ductivity, and undesirable growth form have been the
foresters to reduce the losses of their crops to pest dominant characteristics of tree plantations grown
species places selection pressure on the pest popula- from seed collected in areas that differed from the
tions. This has frequently resulted in the evolution of plantation in elevation, latitude, climate, and/or soil.
 Implications of Genetic Variablilty, Evolution, and Adaptation for Forestry 457

These expensive demonstrations of the limitations im- an insect or disease attack because at least some of the
posed by the genetic constitution of a plant on where population will be resistant to the unfavorable condi-
and under what conditions it may be grown success- tions. Maintaining genetic variation is one of the most
fully have led to planting restrictions in many coun- important responses to the risk of climate change.
tries. For example, the British Columbia Forest Genetic tree-improvement programs are generally
Service in Canada has a requirement that the seedlings a form of stabilizing selection in which an attempt is
to be planted on a particular site must be grown from made to regenerate the forest with genotypes that are
seed originating within 150 m elevation and 2 degrees superior in some respect. Seed is collected from a re-
of latitude and longitude of the location of planting. stricted range of parent genotypes, and nursery culling
All seedlings must be grown from seed collected in the of excessively large or small individuals may further
same seed collection zone as the area in which they are reduce variability. This is fine as long as the resulting
to be planted. It is hoped that these restrictions will population retains sufficient variability or has suffi-
ensure that the thousands of years of selection that cient powers of acclimation to resist unpredicted vari-
have resulted in our present-day heritage of locally ation in the environment. However, if the genetic
adapted ecotypes are not squandered by regeneration improvement selects an ecotype with relatively little
with nonadapted genotypes. variability in adaptation and with little ability to accli-
Where natural selection has not produced com- mate to varying conditions, then the benefits of im-
mercially desirable attributes of growth and tree mor- proved growth and form could be outweighed by lack
phology, programs of artificial selection and breeding of resistance to unpredictable variations in environ-
offer the possibility of augmenting the natural process mental factors, both physical and biotic. Because trees
of selection. Care is required in the conduct of such are long lived, we must maintain sufficient variation in
programs, however. We have already seen that the tree populations to deal with changes that might occur
genome of natural ecotypes generally represents a over the next 50 to 100 years.
compromise among several opposing forces of selec- The native organisms that inhabit a particular bio-
tion. Competition from other trees will promote geocoenose may be well adapted to the problems of
stabilizing selection, which reduces the variety of geno- energy acquisition in their environment but do not
types by favoring the most competitive type. Con- necessarily represent the best possible or the best liv-
versely, the spatial variability of the physical ing adaptation to that environment. Mountain and
environment within any locality will tend to promote ocean barriers to the dispersal of organisms may have
disruptive selection, which increases the variety of geno- excluded species or ecotypes that are better adapted
types within the local population. than the native biota. Time lags in evolution, time lags
A population of very similar, well-adapted geno- in the invasion of new genotypes into a rapidly chang-
types has the advantage that all the individuals will ing environment, and the constraints that develop
grow, compete, and reproduce successfully under through coevolution may mean that imported species
“normal” conditions (the conditions to which they are can sometimes perform better than native species, at
adapted). It will have the disadvantage of being sus- least initially. This has been demonstrated both by the
ceptible to unfavorable conditions. If random fluctua- successful planting of “exotic” (1.e., from a different
tions of weather produce climatic conditions to which region) tree species in afforestation and reforestation
these genotypes are not adapted or if an insect or dis- programs and by epidemics of accidentally introduced
ease organism that finds these genotypes to be a desir- insect pests and diseases.
able food supply arrives, then a high percentage of the Having made this point, one can say that, in gen-
population may be killed or adversely affected. A pop- eral, a locally adapted ecotype will be better able to
ulation with a wide variation in genotypes may be less survive and produce biomass in a given environment
successful in terms of overall growth and survival than will an imported ecotype. Its genotype has been
under stable environmental conditions because only a molded by long experience to give the balance of
small proportion of the population may be really well phenotypic characteristics that allow it to survive the
adapted to the environment at any one time. Such a variety of stresses that it will experience in that partic-
population is generally well adapted to survive under ular environment. Consequently, it is in the best inter-
fluctuating environmental conditions or in the face of est of forestry to create and conserve reserves (gene
458 CHAPTER 16 Genetic and Evolutionary Aspects ofEcosystems

pools) of local ecotypes, even if they do not offer im- developments during the first 3 billion years of the
mediate commercial values. Adaptive phenotypes are earth’s existence, of which only one basic pattern sur-
not always highly productive or of desirable morphol- vived and evolved to the point of leaving fossil evi-
ogy, but should the more commercially desirable dence of its existence.
genotypes prove to be poorly adapted, the forester can Evolution during the subsequent 500 to 1,000 mil-
_go back to using the original genotype or introduce lion years has been a steady process of adaptation of a
some of the original genes into the crop gene pool. few surviving early life forms to exploit new opportu-
Perhaps the best strategy for the forester is to marry nities for the acquisition and storage of energy. As all
the local ecotype to a genotype of desirable physical the opportunities to obtain energy by one method
characteristics, thus gaining the best of both worlds. were used up in a particular type of environment, the
As forests become more intensively managed and as resulting selection pressure generated by competition
genetically improved planting stock is used more produced organisms with new methods of energy ac-
widely, it will become increasingly important to create quisition. As photosynthesizing organisms (photoau-
ecological reserves: examples of the unmanaged condi- totrophs) completely exploited the light energy in an
tion containing the local ecotypes. These must be environment, some of these organisms adapted to be-
conserved as gene banks from which the forester can come herbivores, and selection subsequently led some
make withdrawals should his or her initial investments of these to become carnivores. When all the trophic
in genetic improvement prove to be unwise. possibilities of one environment were exhausted, se-
For a recent discussion of the evolutionary and lection pressure led to the adaptation of some of the
ecological significance of genetic variation in conifers, organisms to the utilization of energy sources in a new
see Mitton (1995). environment. In this way, land areas were colonized by
the life forms that had evolved in the water. The or-
ganisms that we observe today represent the outcome
16.7 Evolution of Ecosystems of billions of years of adaptation to solve the most
basic need of life: the need for energy.
The beginnings of life are thought to have involved
the chance combinations of chemical elements and en-
ergy into various organic molecules that were suffi- SUMMARY
ciently stable to accumulate in __ sheltered Ecosystems are biogeochemical systems that have
environments. Life itself probably began when, evolved to trap, concentrate, and accumulate energy.
through some chance event, combinations of mole- The biological components of this system all utilize the
cules occurred in which some of the molecules were same basic mechanism for energy accumulation and
able to degrade chemically some of the other mole- transfer, but evolution has produced a remarkable
cules and to use the liberated energy and atoms to syn- number of variations on the basic theme of life. Thou-
sands of millions of years of natural selection induced
thesize duplicates of themselves. Arrival at this stage
by unfavorable physical conditions, competition for en-
would have taken a great deal of time, because of the
ergy and other resources, and exploitation by other or-
low probability of the appropriate molecular combina- ganisms has diversified life into a myriad of different
tion. This probability has been likened to the proba- types of organism. Each type has become specialized in
bility of assembling letters of the alphabet at random competing, surviving, and reproducing itselfin particu-
and producing a book on forest ecology. For example, lar types of physical and biotic environment but is also
the probability of combining groups of four letters of capable of adaptation to changing conditions. This
the alphabet at random into the word “tree” is 1 in ability arises from the natural variation in morphology,
456,976 (i.e., 26"). The probability of assembling mol- physiology, and behavior that is present in all natural
ecules at random into self-replicating aggregations populations of organisms. As conditions change, differ-
was perhaps approximately the same as assembling let- ent genotypes within the population become the best
adapted and are favored by natural selection. In this
ters at random and producing an intelligible sentence,
way species evolve.
and the probability of the evolution of complex single- Organisms have adapted to the physical environ-
celled organisms might have been equivalent to the ment: their needs and tolerances are attuned by natural
probability of producing a good short story (Ricklefs, selection to the physical conditions and resource avail-
1973). There were probably innumerable chemical abilities experienced by their recent predecessors. They
 Study Questions 459

have also coevolved with other organisms. This has re- at which some of the relationships are essential to the
sulted in the development of mutual dependencies be- normal functioning of the other parts. The forester is
tween many organisms as well as the development of wise to remember this before making major changes in
chemical, physical, and behavioral mechanisms of de- the ecosystem. She or he must remember that forest
fense and offense. The interdependencies that have management can act as a powerful force of genetic selec-
evolved between the plant and animal members of biotic tion and must ensure that the results of all such selection
communities give an internal structure and character to pressures direct evolution in ways that are in the long-
many communities. term interests of society.
Adaptation and natural variation pose both problems
and possibilities to the ecosystem manager. Adaptation
places constraints on which organisms can be grown in a TAKE-HOME MESSAGE
particular environment, and coevolution may determine
the mixture of organisms that are necessary for the pro- Because of the long time scales of tree crop production
duction of crops of one of them. Adaptation also limits and forest ecosystem change and the complexity of
some of the possible methods of dealing with animal and forests, sustainable forestry must depend much more on
plant pests. Natural variation provides the possibilities natural processes and attributes than is the case in agri-
for the artificial production of desired genotypes: the ge- culture. In most cases, we neither want nor are able to
netic composition of the plant and animal community manage forest ecosystems the way agroecosystems are
can be changed. In managing a forest ecosystem, the managed. Consequently, we must rely on the ability of
forester must recognize local adaptation in the form of the living organisms to undertake the processes that are
ecotypes or provenances. In attempting to improve the responsible for ecosystem function and recovery from
production of the forest, care must be taken to conserve disturbance. A vitally important determinant of this abil-
the natural gene bank and wherever possible to incorpo- ity is genetically controlled adaptation and the ability to
rate natural adaptations in any improved genotypes used acclimate to changing circumstances.
in regeneration programs. The genetics of forest organisms is the most funda-
The individual organisms in the ecosystem have mental aspect of forest biodiversity, and genetic variation
evolved with the physical environment and with each is essential if ecosystems are to be successful in adjusting
other. The phenomenon of coevolution has resulted in to climate change. Respect for the role of genetics and
many of the organisms in the ecosystem becoming so de- genetic variation in controlling the response of ecosys-
pendent, directly or indirectly, on each other that they tems to management must be one of the fundamental
are able to accumulate energy efficiently only in each issues in certification of forest practices as being sustain-
other’s presence. Consider, for example, the fungal and able or not.
tree partners in the mycorrhizal association or the rela-
tionship between moss and spruce trees described in
Chapter 5. The marked degree of interaction and inter- STUDY QUESTIONS
relationship that characterizes the biota of many eco-
1. Why should we maintain genetic variability in forest
systems led to the idea of the ecosystems as a “supraor-
ecosystems?
ganism” in which the various species are compared
with the various organs of an individual organism. This 2. What are ecotypes?
idea contributed to the development of the eco- 3. Why are there strict controls on the provenance of
system concept by Tansley, but there are so many differ- planted tree seedlings?
ences between an individual organism and an ecosystem 4, What are the genetic considerations in the design of
that the idea has fallen into disrepute. It would be unwise forest pest control strategies?
to reject the analogy completely, however. An ecosystem
is a highly interdependent biotic/physical system, the 5. What are the implications of coevolution for the way
components of which have evolved together to the point in which we harvest and manage forests?
 ned Teese Fahoa
‘(yo
; yard a nn Ae Se>i hewayan
wa!
tt 2 a

ha Ve 4 uted bie | is diy dia

fid=nshivany oAT enya) int
thw
Suibes Tne! ella hdd
~~. Vir telnet ts AY
= oa wc \ ol Pa
e- Te \inneart del @
mi vere ald vividly ce geogloa
baer | wadersOy SRY,as

=
el =o5 )i-ie
Rs RIES
Thal “oh
Boa we
' = o 3: : i

Wi
y) be ri

Be
=
F0 SPE AGiey a)
heey = i
oP
a
+
arene i Lo pe aero Whe
— e. ee ? m4‘

+! - anet q! adi tow

20m = Of oe Pau otreek acres
i, Semen api ecsdeia yh nee

A
Pea Perum: =.
ey te Lariat i
: “it ' le any
: A Wein
is) bn, ‘
Aireaher wt a
=~
itiFi
|

ae
bo

oh
Y Lyre
~igli as
’ wih! da hehe
rs) el eerie
; > te | heathen
ves! eee Deen
Slings ~~ tts ap@eg@
oe iy ’ ig wey ity al |
iv * ter rl ti
: + og aeons .
yh aut
of 26 vw ae =
ae aioe @y Caen
 ‘Temporal Diversity
The Ecological Play in Which
All Aspects of Ecosystems
Change Over Time

I Parts I to IV, we have examined the overall characteristics of the

forest ecosystem, how it functions, its physical and biotic compo-
nents, and the importance of the genetic constitution of the biota in
determining its function and response to physical and biotic factors.
We have also seen that change is one of the fundamental attributes of
all ecosystems and that temporal diversity associated with ecosystem
change is one of the most important measures of biodiversity.
In the two chapters of Part V, we consider change in ecosys-
tems over time, how such change influences ecosystem structure
and function, and how knowledge of historical patterns of change
can contribute to the design of sustainable forestry. In the first
chapter, we focus on the relatively rapid change that occurs in the
structure and composition of the biotic community in an area
after some disturbance of the original community—a type of
change called ecological succession. This type of change occurs
“naturally” and is also produced, either intentionally or otherwise,
by forest management. Because much of a forest manager's life is
spent manipulating succession, it is extremely important that
foresters have a sound grasp of this topic. The second chapter
deals with how knowledge of historical disturbance regimes and
the natural range of variation that results can be used as a tem-
plate for the design of sustainable forest management.
The first chapter is long, but it constitutes the central concept in
forest ecology: how and why do ecosystem structure and function
change over time, and what are the types and ecological roles of
disturbance in forest ecosystems? The main justification for all of
the previous chapters of this book is to prepare you for this chapter.
46]
 » eS > =a

ie '

\\
a

Sue ; =~)

7 So

’ % - {ye '
- | > =} | A

~ : =
2
‘ 4
“—.. ‘

7 > = 4
~. we

——— ~
<<
—

jd nat = a

‘<= * :
< ~2 Me om

— oe a
~ ee ae

~ \ cnr 1 é
= =a

\ = <i a > a!

7 ; > —
- ni

:
i= = ' ! i

Po < L b
> a ~~
he ss

? ™ : fi

2 ; Sy !
~ : b

1
6 ia
j
: liye vo

- tS] Oe OY

iS fol

u ' yeaa
4 ia
— ’ vT?
el .

me =_ a -

- r 7} >oie

J “a _ ried

=a al wo cade penate peghs

= : hig

} if wid «
ne
- (ines t 5 :

Rete en
- od Ct paceose al
’ eyes op " @ aa
7
_ =
 C H E R

17
Ecological Succession
Processes of Change in Ecosystems

17.1. Introduction consistent directional changes in the biota. Rather,

they produce medium-term variations around
Change is one of the most fundamental characteristics the average condition; the position of the alpine
of ecosystems. Just as individual organisms change meadow-forest or grassland—forest ecotone fluctu-
through the various stages of their life history, so the ates about a long-term average.
characteristics of ecosystems alter as time passes. Changes in the genetic constitution of organisms as the
Three major categories of ecosystem change can be result of natural selection. ‘This type of change is oc-
identified. curring continually and is called evolution
(Chapter 16). It can occur rapidly in response to
1. Long-term changes in the physical environment. Ice rapidly changing physical or biotic selection pres-
ages come and go (Pielou, 1991), soils develop or sures, but it also occurs on a longer time scale in re-
they may be eroded, and lakes become shallower sponse to slow but directional changes in climate,
and may eventually disappear as they are filled in soil conditions, and other organisms. Natural se-
with sediments. This type of change normally oc- lection is constantly altering the genetic constitu-
curs very slowly, and generally we cannot observe tion of populations in a manner that increases their
the consequences for biotic communities within genetic fitness. With continual changes in the
our lifetime. The resulting changes in the physical physical and biotic environment, there is continual
environment tend to be directional over long peri- genetic adjustment to maintain fitness.
ods; those in the vegetation can be seen in the Changes in the types, numbers, and groupings oforgan-
composition of pollen found at various depths in isms that occupy an area and concomitant changes in
the sediments or peat of lakes and bogs (cf. Figure certain features of the physical microenvironment. This
17-7). Sometimes, the change occurs more rapidly type of change occurs in both newly exposed, pre-
and we can observe the effects in decade-to-decade viously uncolonized physical environments and in
changes in plant and animal populations. For ex- previously colonized areas after disturbance to the
ample, several decades of weather that is warmer existing community. The changes in the biota are
and drier than average may result in alpine mead- accompanied by changes in the microclimate and
ows being invaded by trees. Several decades of soil. Sometimes these physical changes result from
weather that is cooler and wetter than average may the changes in the biota; sometimes vice versa.
result in the invasion of dry grasslands by forest.
Such medium-term climatic changes, however, are
rarely directional and generally do not produce ‘Chapter 17 was written jointly with D. Mailley.

463
464 CHAPTER17 = Ecological Succession

Of these three categories of change, we are most fa- make up a sere are called seral stages. Throughout this
miliar with the last one: the temporal development of chapter, succession is used to refer to the process of
and change in ecosystem structure and function. If we change, and sere or seral stage is used to refer to the
give up the fight with weeds in our garden or if an product of succession.
economically marginal farm is abandoned, then the Ecological succession, the process of ecosystem
freshly exposed soil will soon be colonized by a variety development, occurs in virtually every type of envi-
of plants, most of which are annuals. Within a few ronment found on Earth, although the details vary ac-
years, these will have been joined by perennial herbs cording to the type of ecosystem. It is called primary
and, unless we are in a very arid climate, it will not be succession when it begins in environments that lack or-
too long before woody plants make their appearance: ganic matter and that have not yet been altered in any
either shrubs or trees. In many areas, the first trees to way by living organisms. A fresh rock surface exposed
become established (the pioneers) are broadleaved hard- by a landslide, a layer of till deposited by a retreating
woods (angiosperms); in others, shade-intolerant glacier, a new lake formed by the construction of a
conifers are the pioneer trees. There is no single pat- dam, or a new island created by volcanic eruption
tern that is true everywhere. The early seral or pioneer would undergo primary succession. Where succession
tree species may be replaced over time by shade-toler- begins in an environment that has already been more
ant hardwood species, but in many parts of the non- or less modified by a period of occupancy by living or-
tropical world (e.g., northern temperate and boreal ganisms, it is called secondary succession. Forest clearcuts
forests), young foresters will probably live to see the day and abandoned agricultural fields both undergo sec-
when shade-tolerant coniferous tree species initiate ei- ondary succession.
ther a slow or a rapid replacement of the existing trees. Succession in very dry (xeric) environments is
This type of change is of enormous importance to called xerarch succession; that in moist (mesic) and very
humans. It lies behind the need for farmers to plow wet (hydric) environments is called mesarch and
fields and to use selective herbicides. It is the ecologi- hydrarch succession, respectively. The resulting seres
cal rationale for the practices of clearcutting and pre- under these three different moisture conditions are
scribed burning in the management of certain types of called xeroseres, mesoseres, and hydroseres. Psammoseres
forest. It is why we spend so much of our summer result when succession occurs in sand dune environ-
weekends on our knees in our gardens. It is called ments, and the sequence of communities in salty envi-
ecological succession (or simply succession) and is the topic ronments such as salt marshes is called a halosere.
of this chapter. Succession on rock surfaces produces a /ithosere.
A further distinction is sometimes made on the
basis of the fertility status of the environment.
17.2. ‘Terminology and Concepts Oligotrophic succession occurs in nutrient-poor environ-
ments; comparable terms for moderately fertile and
The term succession can be used in two ways. It can rich environments are mesotrophic succession and
refer to the sequence of plant, animal, and microbial eutrophic succession, respectively. These three adjectives
communities that successively occupy an area over a have traditionally been used to describe different
period of time, such as the changes that can be ob- stages of a hydrosere, but there is no a priori reason
served over the hundred years after the abandonment that they should not also be used to describe variations
ofa plowed field (Figure 17-1). It can also refer to the in succession between any environments that vary in
process of change by which these biotic communities re- nutrient status. The successional pattern of plant com-
place each other and by which the physical environ- munities on an oligotrophic soil is generally very dif-
ment becomes altered over a period of time. When the ferent from that on a eutrophic soil.
term is used in this latter sense, the product of succes- The driving force behind succession, the reason
sion is called a sere: the characteristic sequence of that change occurs, is not always the same. When clas-
biotic communities that successively occupy and re- sified according to the driving force, three main cate-
place each other in a particular environment over time gories of succession can be identified. In many cases,
after either disturbance of the original community or the replacement of one community by the next results
the formation of a new, previously uncolonized envi- from changes in the physical, chemical and biotic en-
ronment. The various communities that together vironment that have been produced by the resident
 Terminology and Concepts 465

Coniferous trees
S060 0 OO

Importance
value a g —_—
So SS “Nudum” stage
“

Time since abandonment (years) ———-—————>

(d)

Figure 17-1 ‘Typical sequence of plant communities that might be observed occupying a
plowed field or scarified clearcut in coastal British Columbia over the subsequent 100-year
period. The photographs show representative communities for three stages: herb-shrub-young de-
ciduous tree (A), deciduous tree (B), and maturing conifer (C). (D) A drawing showing the variations
over the period in the importance values of the different vegetation forms. In similar “old-field” suc-
cessions in other regions, a conifer forest stage (frequently consisting of pines) may follow the shrub
stage, to be replaced in time by a final deciduous stage.

organisms. These changes often render the site less physical processes that are more or less independent
optimal for the organisms that produce the change of and external to the biotic community cause changes
and more optimal for those organisms that replace in the physical environment, which in turn lead to
them. Such a process is called autogenic succession, in changes in the biota. The filling in ofa lake with sedi-
contrast to allogenic succession, which occurs when ment to form a bog and the accompanying changes in
466 CHAPTER17 — Ecological Succession

the biota is an example of allogenic succession, where- that imply fixed, directional patterns of change, where-
as the subsequent biotically controlled conversion as vegetation dynamics lacks this connotation and allows
from bog community to forest is an example of auto- for a more flexible and variable concept of change.
genic succession. A third and somewhat less common Succession, involving change in species over time,
type, biogenic succession, occurs when there is sudden in- is different from the changes in population density and
terference with an autogenic or allogenic succession community structure that occur within a stand as it
by a living organism that becomes the major agent of ages. Stand self-thinning and other causes of mortality
successional change, at least temporarily. A sudden that affect individual tree size and the vertical layering
change in herbivore pressure and the sudden removal of the community result in a series of phases in the de-
of a segment of the plant community by a pathogen velopment of a stand that may or may not result in
are two examples of biogenic succession. The mainte- succession. Such changes are referred to as stand dy-
nance of grasslands in some areas of the world has namics (Oliver and Larson, 1990; Figure 14-20).
been attributed to the activity of browsing and grazing Having introduced the basic terminology, we now
herbivores. Successional change toward a conifer for- need to consider what is meant by mechanisms,
est in northern temperate and boreal areas can be has- pathways, and models of succession. There has been a
tened by the removal of an earlier hardwood seral lot of confusion in the discussion of succession be-
stage by insects or disease organisms. These are also cause of inconsistent usage of these terms (Pickett et
examples of biogenic succession. al., 1987a):
Although change in the composition of the biota
over time is a fundamental characteristic ofall ecosys- 1. A mechanism of succession is a process or interac-
tems, the rates of change vary widely in different seres tion that contributes to successional change.
and between the different stages of a single sere. In Mechanisms of autogenic succession include com-
most areas, change does not continue indefinitely. petition, predation, and seed rain, or the accumu-
Communities develop in which rates of change be- lation of an organic forest floor and a change in
come exceedingly slow or in which the composition of soil moisture and nutrient availability.
the biota remains approximately constant for a long i) A successional pathway is the temporal pattern of
period of time. Such relatively stable communities, ecosystem change: the sequence of plant commu-
which represent either the final or an indefinitely pro- nities and changes in the physical and other biotic
longed intermediate stage of a sere, are called climax components of the ecosystem over time. It refers
communities—or simply climax. The climax can be con- to a particular sere.
sidered both in terms of physiognomy and in terms of 3. A model of succession is some abstract conceptual
structure and floristic composition. Thus, the term representation used to describe and/or explain suc-
can be used to refer both to plant formations or cessional pathways in terms of the successional
biomes and to plant associations. mechanisms that result in successional change.
Most of the above successional terminology refers Models can be word-, picture-, diagram-, or com-
to a progressive, forward development of the ecosys- puter-based (see Chapter 21).
tem toward a climax condition. Successional retrogres-
sion, however, refers to the effects of disturbance in Ecologists have been struggling to find a general
altering the seral condition of an ecosystem back to an explanation of the mechanisms and pathways of eco-
earlier stage, for example, when fire destroys a climax logical succession since the early part of the 20th cen-
community and leads to the development of a primary tury. Numerous conceptual models have been
or secondary pioneer successional community, de- proposed, but none has proved satisfactory. They ei-
pending on the severity of the fire. ther have lacked generality or have been unable to ex-
The term succession is deeply ingrained in the eco- plain the frequently variable details of community
logical literature, and it is likely that its use will con- succession in similar ecosystems after similar distur-
tinue. However, some researchers (e.g., Oliver and bances, or the overall similarity in the pathways of suc-
Larson, 1990) prefer the term vegetation dynamics. cession on these sites despite variations in the details.
They believe that “succession” is too closely linked Many ecologists are now concluding that the
with early theories of ecosystem change (see the sec- search for the “holy grail” of ecological succession—
tion Classical Concepts and Models of Succession) one simple theory that explains it all—is in vain: a uni-
 Classical Concepts and Models of Succession 467

versal, general cause for succession is deemed to be tain caribou or of black-tailed deer, it is necessary to
unlikely (McCook, 1994). A key problem in this understand the vegetation potential and patterns of
search has been the lack of integration of the extensive vegetation development over time that characterize
body of knowledge about succession and that theories the area. In this chapter, we review the history of con-
about succession are often untestable by traditional re- cepts about succession and consider why species and
search approaches. The problem is too complex—too communities replace each other over a period of time.
multidetermined—and the time scales of succession We consider the three major types of sere and how
are too long, often by several orders of magnitude, to rates of succession vary in different environments. A
be addressed in conventional experimental research. discussion of the concept of climax leads to a consider-
Computer models of succession are proving to be a ation of changes in the structure and function of the
useful means of providing this integration and a way of ecosystem during succession, and the chapter con-
dealing with complexity and long time scales. The cludes with a brief comment on the importance of suc-
“gap” models of Botkin et al. (1972, 1993), Pastor and cession to humans.
Post (1986, 1988), Shugart (1984), Smith and Huston
(1989), and others and the model of Pacala et al.
(1993, 1996) have proved to be helpful in advancing 17.3 Classical Concepts and
our understanding of succession. However, consider- Models of Succession
able further development of ecosystem models, in
conjunction with rigorously conducted field studies, Few topics in ecology have been debated more than
will be needed to make the next major advance in this succession and the related concepts of diversity and
branch of ecology (McCook, 1994). In most cases, stability. Descriptions of succession to be found in
these models are too simple to provide the level of many earlier ecology textbooks did not correspond
predictive accuracy needed in forest management closely to the succession that can be observed in
planning. They lack many of the key determinants of forested ecosystems, and there continues to be discus-
production ecology and biotic interactions that are sion concerning the mechanisms of succession and ex-
needed to predict the change and development of for- actly what constitutes a climax. However, significant
est ecosystems over time. advances in our understanding of succession have oc-
As we will see by the end of the chapter, the sim- curred during the past two decades. These advances,
ple, progressive development of a sere from a bare, together with the shortcomings of earlier theories, are
unvegetated area through a predictable series of stages discussed in the following reviews that are recom-
to a stable climax is not an accurate description of mended to the interested reader: Botkin, 1980; Drury
what really happens in many kinds of ecosystem. and Nisbet, 1973; Horn, 1974, 1976; Glen-Lewin et
Many ecologists whose experience is limited to an- al., 1992; MacMahon, 1980; McCook, 1994; McIn-
thropogenically disturbed environments do not find tosh, 1980; Pickett and White, 1985; Pickett et al.,
the concept of a linear development to a predictable 1987a,b; Shugart, 1984; and West et al., 1981.
climax community useful, and it is now recognized Many of the shortcomings of conventional theo-
that the pattern of seral stages is often variable. How- ries about ecological succession have resulted from the
ever, the linear concept of successional development uncritical extrapolation of the earliest studies to a vari-
will serve as a useful starting point for our discussion. ety of different types of ecosystem and from a preoc-
Ecological succession is one of the most important cupation among North American ecologists during
topics of both basic and applied ecology. It is one of the 1950s and 1960s with two particular cases of suc-
the oldest, most basic, yet most controversial of eco- cession: abandoned farmland (“old-field” succession)
logical concepts and perhaps the one least understood and prairie succession (MacMahon, 1980). There was
by the public. Understanding the processes, rates, and a conspicuous failure to differentiate adequately the
patterns of ecosystem development and how these major types of succession, and the interpretations of
vary in different types of ecosystems is a key factor in many earlier authors were heavily influenced by their
successful management of ecosystems. Whether one’s own field experience. Ecologists from the eastern de-
goal is the management of a wilderness or a city park, ciduous hardwood forest region of the United States
a valley-bottom farm or a mountain rangeland, a sea- have tended to present a view of succession based on
level plantation or a subalpine forest, a herd of moun- studies of eastern forest ecosystems that is not always
468 | CHAPTER17 — Ecological Succession

accurate or appropriate for northern or western North ecologists: Cowles and Clements. Their ideas, which
American environments, and the opposite is also true. were developed while studying vegetation sequences
Similarly, the traditional North American view of suc- in sand dune areas on the shores of Lake Michigan,
cession has not always been useful in other parts of the developed the earlier work of European ecologists, in-
world under very different climatic and natural distur- cluding a Danish botanist, Warming, and led to the
bance regimes. first theory of succession: the monoclimax theory (see
The problem of parochialism has occurred in reviews in Clements, 1916; Drury and Nisbet, 1973;
many branches of ecology, and many energetic argu- MacMahon, 1980; and Whittaker, 1974).
ments in the ecological literature can be related in part
to the different field experiences of the participants.
One of the easiest mistakes that can be made in ecol-
The Monoclimax Theory
ogy is to attempt to apply a general theory developed The monoclimax theory states that the species com-
under a restricted range of ecological conditions to a position and structure of the terminal community—
very different type of ecosystem. Although it is one of the climax—of all seres is determined by the regional
the aims of ecology to develop broadly applicable the- macroclimate. The climax consists of plants that can
ories about the patterns in nature, it must never be reproduce successfully beneath their own shade and
forgotten that the theory must fit the facts rather than therefore maintain the community indefinitely under
the other way around. The reader must remember the prevailing climatic conditions. The theory does
that the author of this book also has a limited experi- not require that all areas in a given region support the
ence of the vast range of ecosystem types in the world climatic climax at any given time, because many areas
and that the view of succession presented here is un- will be occupied by earlier seral stages that are pro-
doubtedly influenced by that experience. However, a duced by disturbance and that are actively developing
conscious attempt has been made to consider the vari- toward the climax. In fact, the theory allows that none
ety of successional conditions that can be encoun- of the vegetation in a region may be in the climatic cli-
tered, and hopefully the ensuing discussion applies to max condition and that development in some areas
most forested environments. may be arrested and attainment of the climatic climax
conditions greatly delayed or permanently prevented
because of the action of soil, animal, fire, or other en-
Early References to Succession
vironmental factors that override the effect of climatic
Recognition that vegetation undergoes change over factors. All the theory requires is that there be a defi-
time is recorded in the earliest of botanical writings. nite linear development toward the climatic climax.
Theophrastus wrote about it in 300 B.C., and the pat- Proponents of the monoclimax theory have produced
tern of development of vegetation in bogs in Ireland a series of terms with which to describe these “atypi-
was described in print in 1685. A French botanist, cal” conditions.
Buffon, noted the succession of tree species in French Where succession is arrested early in the sere, such
forests and recognized the process of autogenic suc- as a spring-fed bog where the vegetation is unable to
cession as early as 1747. Many descriptions of vegeta- alter the moisture conditions or a lichen community on
tion change were published in the 19th century, a a steep exposed rock face, a serclimax results. Such seral
period when many of the earlier empirical observa- communities may persist indefinitely. Where develop-
tions and experiences of foresters were formalized into ment proceeds but is arrested permanently or for a
published statements concerning vegetation change in very long period at the stage immediately preceding
managed forests. Early use of the succession concept is the climax, a swbclimax results. For example, fire, wind,
reviewed by Pignatti and Savoia (1989). or logging can maintain in perpetuity a forest of pio-
In 1863, the American philosopher and naturalist neer tree species such as birch, pine, or Douglas-fir,
Thoreau described the succession in which even-aged rather than climatic climax species such as beech, hem-
hardwood stands replace pine stands after logging in lock, or cedar. A disclimax, (disturbance climax) occurs
the northeastern United States. He coined the term where communities are held in a stable early succes-
“forest succession” to describe the change. It was not sional condition by the activities of humans or other
until the turn of the century, however, that Thoreau’s animals. Farmland generally represents a disclimax, as
early statement was developed into a formal concept do those subalpine meadows that are maintained in
of plant succession by the work of two American plant their recreationally valuable condition by the activities
 Classical Concepts and Models of Succession 469

of burrowing animals. Removal of these animals sites (oligotrophic) may become enriched. According
would, in many cases, result in the reinvasion of these to the theory, this convergence in the soil and micro-
meadows by subalpine forest communities. A classic climatic conditions eventually permits the same biotic
case of a disclimax is the calcareous downlands (low, community to occupy all sites irrespective of the
rolling chalk hills) of southern England. It was once original condition of the physical environment. Figure
thought that soil conditions were responsible for the 17-2 presents an idealized example of ecological con-
grassland community that had characterized the downs vergence for low-elevation forests in coastal British
for centuries, which were therefore regarded as a ser- Columbia. Ecological convergence may also be re-
climax. However, drastic reduction of rabbit popula- ferred to as successional convergence.
tions by disease in the 1950s was followed by prompt Since the monoclimax theory was first advanced by
succession toward climax hardwood forest, demon- Clements in 1916, it has generated both intense argu-
strating that the grasslands were really a disclimax. ment and strong support. The main point of con-
Early discussion of the monoclimax theory linked tention initially was the concept of the climax as a
it with the idea that the climax was analogous to a “complex supraorganism,” which suggests an invari-
“supraorganism.” After the removal of a mature cli- able sequence of stages within any type of sere, leading
max community (analogous to the death of an old or- inevitably to a single climax type: the climatic climax.
ganism), a new succession Is initiated (analogous to the Studies of secondary succession on power rights of
birth of an organism), which leads through a fixed se- way, abandoned farmland, and forest clearcuts have
ries of seral stages (analogous to stages in the life his- revealed that expected seral stages may frequently be
tory of an organism) to culminate in the omitted, that early or midseral stages can sometimes
reestablishment of the climax (analogous to the attain- be surprisingly stable, and that the floristic composi-
ment of maturity in an organism). A corollary of the tion of particular seral communities can vary greatly
supraorganism concept is that succession can no more from time to time and from place to place in similar
fail to result in the ultimate reestablishment of the cli- physical environments. This obviously invalidates the
max than germination can fail to lead to another ma- comparison between a sere and the developmental
ture individual plant of the same type. Although the stages of an individual organism. Another major point
monoclimax theory recognizes the existence of of contention is the idea that vegetation will continue
relatively stable communities that do not yet reflect the to develop toward the same end point over periods of
regional climate, these are thought of as temporary several centuries. This point is challenged by the evi-
delays in the ultimate and inevitable progress toward dence of major changes in climate and vegetation in
the climatic climax: the principal, prevailing un- North America over periods considerably shorter than
disturbed plant community of a region that best ex- that required for complete ecological convergence
presses the regional climate. (Davis, 1986, 1987, 1989; Webb et al., 1985).
Another corollary of the monoclimax theory is the
principle of ecological convergence. This states that be-
cause succession is regulated primarily by climate, the
The Polyclimax Theory
biotic communities that occupy all site types within a Although the concept of the climax as a supraorganism
given climatic region will converge in structure and has not been without its supporters (e.g., Phillips,
species composition to arrive eventually at a single cli- 1931, 1934, 1935a,b), it has been criticized by propo-
max community type: the climatic climax. Within any nents of the second major theory of succession: the
region, there is always a variety of physical environ- polyclimax theory (Tansley, 1920, 1935, 1939, 1941).
ments that differ in moisture status, fertility, type of This theory notes that many factors can intervene to
soil, and so on. Early seral communities in these di- prevent an area from reaching the climatic climax con-
verse environments differ greatly in structure, floristic dition. In many parts of the world, frequent natural
composition, and function. However, over a period of fires serve to maintain grasslands or forests of mid-
time, which may be very prolonged (centuries or even seral tree species in areas in which the climatic condi-
millennia), the combined activities of the plants, ani- tions would indicate progression to an open forest or a
mals, microbes, and physical processes gradually mod- forest of later-successional tree species, respectively.
ify the extremes of the physical environment. Dry sites The presence of outcrops of serpentine soils rich in
become wetter, and wet sites become drier. Very fertile magnesium or of limestone soils rich in calcium in an
(eutrophic) sites may become less fertile, and infertile area dominated by acid igneous rocks will result in
470 CHAPTER 17 Ecological Succession

e
Stage Moss
stage Shrub
stage
MOIST ENVIRONMENT
Mesarch succession bem
mat
———
ip
de
‘
Red
alder Shrub =
stage stage
Herb
stage

Bare
mineral
soil
stage
Shrub
WET ENVIRONMENT conifer
Hydrarch succession Compacted
glacial
Forest flog Time srs
bog of variable (100-S00 years?)
Bottom _stage
-rooted
Forest depth
floor
aquatic
plant Sphagnum
Stage peat

Open
water
stage

Figure 17-2 Idealized representation of ecological succession and convergence in the three
major types of environment of the Coastal Western Hemlock Zone of British Columbia. The
seral stages illustrated represent typical but not necessarily inevitable components of the conver-
gence. The time scales are not linear and are not comparable for the three types of sere. Estimates of
the probable ranges in time scales for the three seres are given.

local climaxes that are floristically different from the different climax communities determined by the inter-
regional climax. The activities of grazing animals, al- acting mosaic of habitats and disturbance regimes.
ready referred to, can maintain grassland communities In defense of the monoclimax theory, it should be
in a climax condition, preventing the invasion of noted that the supraorganism analogy is not a neces-
woody species. Areas of thin soil and steep topography sary part of the theory, that in some types of environ-
will support stable communities that differ from those ment the overall sequence of broad seral stages is
of flat, valley-bottom areas. Thus, according to the en- remarkably invariable, and that in some climates
vironmental factor acting to prevent convergence to ecological convergence does occur. As with so many
the climatic climax, we can have a pyral climax (fire), ecological theories, the monoclimax theory is a rea-
edaphic climax (soil), or a biotic climax (animal influ- sonable approximation of reality in some types of
ence) in addition to the climatic climax. ecosystem in some parts of the world, whereas in oth-
In the polyclimax theory, the vegetation of a re- ers, the polyclimax theory fits the facts better.
gion is viewed as a mosaic of communities at different
stages of succession, some of which may reach cli-
matic climax relatively rapidly (a few centuries or less). The Climax Pattern Hypothesis
Some of the communities may be developing slowly Both the monoclimax and the polyclimax theories as-
toward a climatic climax but so slowly that the climate sume the existence of discrete, mutually exclusive
may have changed before the climax is reached. For plant communities, an assumption that is rejected by
these areas, it becomes academic to talk about a cli- some plant ecologists on the grounds that vegetation
matic climax. Other areas may never approach the cli- forms a continuum in which species are distributed
matic climax condition for the region because factors and replace each other independently along environ-
other than climate are dominant in determining the mental gradients (see Chapter 14). Individual species
structure and composition of the community. Thus, are combined in many different ways into communi-
according to the theory, succession in a region does ties, and a single species may be shared by many com-
not lead toward a single climax but toward a mosaic of munities (Gleason, 1926, 1939). Vegetation is seen as a
 Review ofRecent Theories and Models of Succession 471

complex pattern of integrating communities rather the mild, moist climate, succession proceeds rapidly in
than a mosaic of distinct communities. Succession much of the humid areas of the coastal forest, and dis-
within this complex pattern will result in a corre- turbance by fire and insects that causes major succes-
sponding complex pattern of individualistic climax sional retrogression is infrequent. The major agent of
communities. This model of succession, called the disturbance is wind, and this generally recycles climax
climax pattern hypothesis (Whittaker, 1953), has not or late-seral stages or advances succession from mid-
proved to be as popular and persistent as either the seral forests toward the climatic climax rather than
monoclimax or the polyclimax models. setting the ecosystem back to an earlier seral stage.
Under these conditions, there is a high probability
that successional processes will carry ecosystem condi-
Do These Classical Theories Contribute to tion to the climatic climax stage over much of the
Our Present Understanding of Succession? landscape well before the next major disturbance, and
These early theories and models of succession have when this occurs, it will likely not set the ecosystem
been criticized as being too rigid, being too theoreti- very far back in the sere. In the interior forests, how-
cal, lacking in detail about mechanisms, and unable to ever, climatic limitations restrict the rate of autogenic
account for and explain the observed variability in suc- succession and create a relatively high frequency of
cessional pathways. However, many of the critics fire and insect disturbance. These disturbances, which
failed to look at the details of early theories. Recent generally occur well before the ecosystem reaches a
reanalysis suggests that these early concepts do in fact late seral condition, are often severe, setting the
have much in common with presently held views ecosystem back to an early seral stage. The result is a
about succession (McCook, 1994). mosaic of different seral stages, few of which approach
A major criticism of the monoclimax theory is that, the climax condition, and there is often little evidence
in most areas, successional convergence to the cli- of convergence in seral conditions among xerarch,
matic climax requires much longer than the natural mesarch, and hydarch pathways.
frequency of major natural disturbance events. In The conclusion to be reached is that it is highly
some cases, it requires longer than the time over unlikely that a single theory about succession will
apply equally well everywhere. Differences in rates,
which significant, directional climate change occurs.
Despite this, one can find ecosystems in the climatic mechanisms, and pathways of succession between
climax seral stage in at least some parts of many un- areas with different climate, topography, geology,
soils, and disturbance histories require models that
managed forest landscapes and over much of the land-
can account for these differences. Clearly, an under-
scape in some ecological zones.
standing of these differences in succession requires an
In the cool temperate rain forest climate of the wet-
understanding of the ecological processes that are re-
ter, cooler parts of British Columbia’s Coastal Western
sponsible for ecosystem change. The many succes-
Hemlock zone, it is not uncommon to find extensive
sional theories and models that have been proposed
areas of “old growth” forest in essentially similar cli-
over the past two decades focus on these processes.
matic climax condition. Successional convergence has
occurred. In contrast, in the subcontinental climate of
the interior of British Columbia, the forest landscape is
more typically a mosaic of forests of different ages and 17.4 Review of Recent Theories
seral stages. “Old growth” forest is much less common, and Models of Succession
and there is much less evidence of convergence of dif-
ferent parts of the landscape to a single vegetation type The earlier theories were broad generalizations based
or ecosystem condition. The monoclimax model of on the observation that (1) natural vegetation within a
succession seems to be supported by the evidence from climatic region tends to follow a characteristic pattern
the coast, whereas the polyclimax model is a better de- of change in species abundance after disturbance and
scription of the forests in the interior. (2) these changes often involve a succession of com-
The explanation for this conclusion is to be found munities dominated by species with progressively
in the differences in rates of autogenic succession and greater maximum size, age, and shade tolerance and
the frequency and intensity of disturbance between progressively lower maximum growth rates and dis-
humid coastal forests and interior forests. Because of persal abilities (McCook, 1994). The usefulness of
472 CHAPTER17 — Ecological Succession

these generalizations in ecology and resource manage- genic succession is often artificial—that in many suc-
ment is reduced, however, because they overlook the cessional pathways, both causes of change are operat-
great variation in the details of succession from place ing. They asserted that succession is related to the
to place and time to time within a climatic region. Re- correlation between various different life history at-
cent theories focus on understanding, explaining, and tributes (see also Odum, 1969). Early successional
predicting this variation. plant species often have widespread and rapid disper-
sal of seed, rapid growth, shorter life spans, low toler-
ance of light competition, and, sometimes, tolerance
Recognition of the Importance of low availability of soil nutrients, especially nitrogen.
of Individual Species In contrast, later successional species are often toler-
The details of vegetation change depend on which ant of light competition, are intolerant of low supplies
species are currently present and which species are in- of nitrogen in mineralizable organic matter, have rela-
vading. This is a basic theme that runs through most tively slow growth rates, have only local distribution
of the recent theories of succession. This individua- of seed, and may be longer lived than early seral
listic, or stochastic (i.e., unpredictable), view of succes- species. This categorization of the life history “strate-
sion was first advanced by Gleason (1926, 1927, 1929) gies” of species of different seral stages has been criti-
in his criticism of the overgeneralizations of the mon- cized because of the many exceptions that have been
oclimax theory. His ideas were developed by Egler observed, especially with regard to the “strategies” of
(1954), who proposed that the succession of plant tree species; much variation around this basic pattern
species in abandoned agricultural fields depends can be found. Nevertheless, as a broad generalization,
mostly on the relative growth rates and shade toler- it does have some validity and has proved to be useful
ance of the assemblage of species in the fields at the in understanding successional patterns.
time of abandonment or arriving shortly thereafter The idea of plant strategies as an important deter-
(the “initial floristic composition” model). He believed minant of successional sequences was examined by
that in many successions, the sequence of species ar- Grime (1974, 1979). He classified plants according to
riving later and the alteration of the site by the early their ability to tolerate stress (low levels of resource
colonizers (the “relay floristics” model that is implicit in availability: light, nutrients, water). He proposed three
the monoclimax theory of succession) were much less major categories of adaptation (ruderal, competitor, and
important. Egler saw succession in old fields as de- stress tolerator) to three major types of environment
pending mainly on “who gets there first” and their rel- (low disturbance-low stress; high disturbance-low
ative growth rates, but he stated categorically that stress; and low disturbance-high stress, respectively).
most successional pathways involve elements of both During succession, there is a change in the environ-
relay floristics and initial floristic composition ment from high disturbance-low stress (high re-
processes. Egler’s models, which are discussed further sources) to low disturbance—-high stress (low
in the section on Colonization, introduced the idea resources), with an accompanying change in plant
that succession could be explained to a considerable species from ruderals to stress tolerators. In more pro-
extent by variation in the life histories of the species ductive ecosystems, mid-succession may be character-
that characterize different seral stages. ized by low stress and low disturbance conditions
under which “competitor” species will be dominant.
Grime’s theory predicts that species with the greatest
Importance of Life History Characteristics capacity for acquiring resources under a particular set
In an important review, Drury and Nesbitt (1973) sup- of resource conditions will be superior competitors,
ported the argument that succession frequently devi- and it supposes that a species that is superior in com-
ates from the inevitable linear progression envisaged peting for one resource will also be superior in com-
by the monoclimax theory. They suggested that most peting for other resources. The best competitor is the
of the details of succession are the result of differences species that is able to capture the most resources,
in colonizing ability, growth rates, survival, and abili- thereby growing faster and dominating less competi-
ties to tolerate stress associated with competition for tive species. This approach to describing successional
resources or changing resource availability. They pathways based on a classification of broad categories
noted that the distinction between autogenic and allo- of plant life history strategies lacks any explanation of
 Review ofRecent Theories and Models of Succession 4/3

the mechanisms of change in species composition, ground (soil nutrients and water) resources and for re-
other than the implicit equilibrium response to unex- production and defense against herbivory (Tilman,
plained changes in resource levels on the basis of 1990b). Allocation to increase access to soil resources
which species is the best competitor for resources will reduce access to light, and vice versa. A change in
(McCook, 1994). resource supply over time results in a change in
The life history strategy approach to explaining species because of their different resource-equilibrium
succession was further developed by Noble and Slatyer values and allocation strategies. The variation in re-
(1977, 1980a,b). Referring to the life history charac- source supply is partly the result of the plants them-
teristics of plants as “vital attributes,” they developed a selves (i.e., it is the result of autogenic processes), but
model of community development based on how these he allows that these changes may also reflect postdis-
attributes determine species responses to different dis- turbance changes in supply (i.e., allogenic processes).
turbance regimes. Their qualitative model has been However, his model does not explain these changes. A
used successfully to predict patterns of vegetation de- weakness of his approach is that it assumes that the
velopment in landscapes with varying fire disturbance rate of change in resource availability is slow com-
regimes (see discussion of the multiple pathway model pared with the rate of change in plant species, and it
in the section Changes in Ecosystem Function During also assumes the attainment of equilibrium conditions.
Succession). However, it lacks detail of the actual Both these assumptions have been challenged (refer-
mechanisms of succession and is more a predictive ences in McCook, 1994). Nonequilibrium models are
model of “vegetation dynamics” (i.e., change over increasingly seen as the way to explain and predict
time) than an explanatory model of the processes and succession (Botkin, 1990, 1993). However, most of the
pathways of succession (McCook, 1994). existing nonequilibrium computer models of succes-
sion ignore changing levels of soil resources and
An Explanation of Succession by Both Plant changes in plant allocation strategies in response to
these changes that were discussed in Chapters 4 and 5.
Strategies and Changing Resource Levels: Clearly, there is a need to combine all of these features
The Ideas of Tilman in such computer models (see the section Computer
Tilman’s approach is similar to that of Grime in that Models of Succession ).
he argues for plant strategies with respect to resource
availability as the key mechanism of succession. It is
different in that he focuses on the role of competition
Connell and Slatyer’s Three-Pathway Model
along gradients of resource availability. He emphasizes In contrast to the emphasis on the role of individual
the species that has the lowest resource demands and plant species’ strategies in directing successional de-
is therefore most competitive under conditions of re- velopment, Connell and Slatyer (1977; also Connell
source competition (Tilman, 1982, 1985, 1988, et al., 1987) proposed that succession should be
1990a,b). Species that are able to colonize, grow, and viewed as following one of three broad pathways
reproduce fastest at a particular level of the resource (Figure 17-3). Their concept was that early seral
(e.g., nutrient, light, moisture) that is most limiting species or communities assist or benefit (facilita-
will be competitively dominant. By occupying a site, tion), have little effect on (tolerance), or inhibit or
species will reduce the availability of resources to an constrain (inhibition) the development of later seral
equilibrium level at which the population demand for species or communities. This is a useful conceptual
resources is in equilibrium with the resource supply. framework in which to incorporate many of the
The species that achieves its equilibrium level at the ideas of the monoclimax, polyclimax, and Egler’s
lowest level of resource supply will outcompete the ideas, but it has been widely criticized. As presented,
other species. the three-pathway model implies the exclusive role
One of the strengths of Tilman’s approach, which of any one of the three pathways and associated
he calls the resource-ratio hypothesis, is that he goes be- mechanisms in any particular succession, whereas in
yond whole-plant or whole-population strategies; his reality, as noted by Egler (1954), most successional
model also considers resource allocation strategies sequences involve a mixture of the different pathway
(resource tradeoffs) within a plant in response to the mechanisms. For example, facilitation may occur in
competing needs for aboveground (light) and below- the early stages of primary succession, to be replaced
474 CHAPTER 17 Ecological Succession

disturbance opens a relatively)

arge space, releasing resources.
|
|
FACILITATION
| PATHWAY
| The environment is favorable
| The environment is favorable
for the establishment of any
for the establishment of only
certain “early successional” species. Early successional
| species. species usually appear first
| because of their life history
| characteristics.
| Plants modify environment so
| that it becomes less suitable TOLERANCE INHIBITION
for subsequent recruitment of PATHWAY PATHWAY
| early successional species, but
Plants modify environment Plants modify environment
: more suitable for later
so that it becomes less so that it becomes less
successional species. suitable for subsequent
Le suitable for subsequent
recruitment of early recruitment of both early
| Later successional species
successional species, but and late successional species.
this has little or no effect
| grow to maturity. Early on subsequent recruitment

|C
successional species are of late successional species. As long as individuals of
gradually eliminated. the initial colonists persist
undamaged and/or continue
Juveniles of later successional to regenerate vegetatively,
| Successional sequence species that manage to invade they exclude or suppress
| continues until resident or are already present grow to subsequent colonists of all
| species no longer facilitate maturity in spite of presence species.
| the establishment and of early successional species.
In time,
ae these earlier species
| growth of other species.
are eliminated.
|
| Successional sequence
| continues until there are
| no longer any species that
| can invade and grow in
| the presence of the resident
| species.
|
|
| At this stage, further invasion and/or growth to maturity can occur only when a
| resident individual is damaged or killed, releasing space and resources. Continued
| change in species composition depends on site conditions and the characteristics
| of species that are available as replacements.
|

Figure 17-3 The three-pathway model of succession of Connell and Slatyer (1977).
(Copyright The University of Chicago Press. Used with permission of the publisher and the authors.)

by either tolerance or inhibition mechanisms as suc- standing, the idea of facilitation, tolerance, and inhi-
cession proceeds. Inhibition of one species by an- bition is useful.
other in an early seral stage may speed the invasion Forest managers spend much of their time plan-
oflater seral species or communities. What is inhibi- ning the manipulation of succession. An important as-
tion for one species may be facilitation for another. pect of this manipulation in some forest ecosystems is
Real life is more complex than suggested by a three- management of the interactions between crop and
pathway mosaic. McCook (1994) and Pickett et al. noncrop vegetation. Traditionally, this was called
(1987a) provided a detailed evaluation of Connell “weed control”; now it is referred to as “vegetation
and Slatyer’s approach. These criticisms notwith- management” in recognition of the many benefits
 Review ofRecent Theories and Models of Succession 475

provided by noncrop vegetation in addition to their sal and/or the size of the propagule pool (e.g., seed
sometimes negative effects on crop species. bank, bud bank) (mechanisms).
In some ecosystems, noncrop vegetation definitely
3. Variations in species performance (a “cause”). The
facilitates crop trees: nitrogen fixation; capturing and
performance is defined in terms of a species re-
holding nutrients released during the “assart” period
sponse to resource availability and its ecophysiol-
(the “nutrient sponge” effect); protecting seedlings
ogy and life history strategies, environmental
from overheating, desiccation, herbivory and frost stress, competition, allelopathy, and natural ene-
heaving; and building up soil organic matter. Removal mies (mechanisms).
of all noncrop vegetation in such facilitation pathways
would be completely inappropriate. These authors believe that such an ecosystem-
In other cases, early-seral, noncrop vegetation may level, analytical approach, which explicitly incorpo-
reduce the early growth of crop trees but not prevent rates community, population, and __ individual
their forming a closed canopy after some delay. Over a organism-level mechanisms within an ecosystem-level
long rotation, this loss of growth may be negligible conceptual model, can provide the basis for develop-
compared with the cost of vegetation management, a ing an integrated explanation of succession.
cost that must be carried over the whole rotation. Be-
cause the crop trees will eventually outgrow the com-
petition (a case of tolerance), the best management Computer Models of Succession
decision may simply be to accept slow early growth. The debate about the causes and mechanisms of suc-
However, this loss of crop growth may be serious over cession has followed two broad paths during the past
a short rotation, and where slope stability, hydrologi- two decades: (1) relatively simple conceptual models
cal, wildlife, aesthetics, and timber supply considera- and hypothesis testing (described above) and (2) devel-
tions argue for prompt reestablishment of closed opment of process-level computer models of succes-
forest, vegetation management may be called for. sion. There is a continuing need for more descriptive
Where early seral stages threaten to form serclimaxes and experimental field studies of succession, and sim-
(i.e., an inhibition pathway), management objectives plistic mathematical models of succession may offer
may not be achievable without vegetation manage- little insight into the process of successional change
ment. Tree crops may not be possible within desired (Pielou, 1981). However, the recent generation of
time spans unless the noncrop vegetation is con- computer-based models has proved to be a vital com-
trolled. Thus, although the three-pathway model has plement to field research (see reviews by Huston,
several limitations from a theoretical perspective, it is 1992; Huston et al., 1988; Judson, 1994; and Urban
still a useful concept for forest management. and Shugart, 1992).
Computer models have the great advantage that
they can incorporate a wide range of successional
A Hierarchy of Successional Causes mechanisms and processes and project the conse-
One of the more recent contributions to the discus- quences of these over relevant time scales. Such mod-
sion of successional theory focuses on the develop- els constitute complex hypotheses, and a comparison
ment of a “comprehensive causal framework” within between their predictions and field observations can
which the various competing ideas about succession provide a basis, albeit incomplete, for evaluating the
can be resolved and integrated (Pickett et al., 1987a). hypothesis, something that cannot be done in conven-
This framework is presented as a “hierarchy of succes- tional experimental research. These models have the
sional causes and mechanisms of succession” (Pickett disadvantages, shared by many of the conceptual mod-
et al., 1987b). The approach is based on the explana- els described above, of (1) being difficult to validate
tion of succession in terms of (Table 17-1): (test the predictions against reality) over the time
scales that they address because of lack of an available
1. The availability of sites for invasion (a “cause”). “reality” for comparison and (2) being an incomplete
The availability of sites is attributed to disturbance representation of the ecosystems that they purport to
(the mechanism). mimic. Nevertheless, the models of Botkin (1972,
2. Differences in the availability of species (a 1993), Huston and Smith (1987), Pacala et al. (1993,
“cause”). This is attributed to variations in disper- 1996), Pastor and Post (1985), Shugart (1984), and
476 CHAPTER 17 Ecological Succession

Table 17-1 A Hierarchy of Causes of Succession*

Contributing
General Causes Processes or
of Succession Conditions Defining Factors
8S

Site availability Coarse-scale disturbance Size, severity, time, dispersion

Differential species Dispersal Landscape configuration

availability Propagule pool Dispersal agents, time since disturbance, land use

Differential species Resource availability Soil conditions, topography, microclimate, site history
performance
Ecophysiology Germination requirements, assimilation rates, growth
rates, population differentiation
Life history strategy Allocation pattern, reproductive timing, reproductive mode
Stochastic environmental stress Climate cycles, site history, previous occupants

Competition Presence of competitors, identity of competitors, within-

community disturbance, predators and herbivores,
resource base
Allelopathy Soil characteristics, microbes, neighboring plants
Herbivory, Climate cycles, consumer cycles, plant vigor, plant defense, /

Disease, and predation community composition, patchiness

“The highest level of the hierarchy (general causes) represents the broadest defining phenomena. The intermediate level (contributing processes
or conditions) contains the mechanisms of change or “causation” of the highest level. The lowest level consists of the particular factors that de-
termine the outcome of the intermediate-level processes and are discernible or quantifiable at specific sites. Whether a particular process or fac-
tor advances or slows succession must be determined experimentally in specific instances or by generalization among comparable cases. Other
processes or factors may be recognized in specific situations. For simplicity, interactions among factors at each level are not shown. (Pickett et
al., 1987b, used with permission of authors and Dr. W. Junk, Publishers.)

others have proved to be useful in advancing our un- and make our understanding of succession useful in
derstanding of certain aspects of succession. They can the management of natural resources, then we have to
provide a convenient way of testing theories (e.g., ask the additional question, “How rapidly does succes-
Huston and Smith, 1987; Pacala et al., 1996) and of sion occur?” In answering these questions, we will dis-
predicting succession under changing environmental cover that the driving force and rate of succession vary
(e.g., Pastor and Post, 1986, 1988) and resource man- from site to site, from region to region, and between
agement regimes (Botkin et al., 1972; Kimmins, 1990, different seral stages on a given site.
1993b; Sachs and Trofymow, 1991; Seely et al., 1999). Succession occurs as the result of autogenic or bio-
McCook (1994) sees an increasing role for such mod- genic processes (associated with the living com-
els, especially if they are developed to include a wider munity), allogenic processes (associated with the phys-
range of successional mechanisms. Because of their ical environment), or some combination thereof.
importance, a selection of these models is discussed in Many successions are initiated by an allogenic process
more detail in Chapter 21. (e.g., wind, fire, landslide) but are then determined
largely by autogenic processes. In a few cases, allo-
genic processes may continue to play an important
17.5 Mechanisms of role throughout the succession, for example, the suc-
Successional Change cession that occurs along the banks of a meandering
river in a valley floodplain, which starts on recently
If we are to understand vegetation dynamics and suc- deposited sandbars and is directed initially by subse-
cessional change, we must first answer the question, quent alluvial deposits (e.g., Van Cleve and Viereck,
“Why does change occur?” If we wish to go further 1981). Such examples are less common than succession
 Mechanisms of Successional Change 477

that is largely the result of autogenic processes. These effects of the accumulation of undecomposed grass lit-
can be grouped into three major mechanisms: (1) colo- ter on unburned areas (Old, 1969).
nization, (2) alteration of the physical characteristics of If colonizers produce short-lived reproductive
the site, and (3) displacement of species by competition propagules, then they must produce very large num-
or antibiosis. This grouping of mechanisms is broadly bers unless they have an efficient means of dispersal to
similar to that proposed by Pickett et al. (1987b). suitable new habitats. Many plants depend on wind for
dispersal and produce abundant quantities of small,
relatively short-lived seeds to compensate for the fact
Colonization
that wind is not always a reliable means of reaching the
Colonization is a process with two components: inva- appropriate type of habitat. Alternative strategies have
sion and survival. The rate at which a site is colonized evolved in some plants, such as those that produce
depends on both the rate (numbers per unit time) at fewer but larger seeds that are dispersed to suitable
which individual organisms (seeds, spores, immature sites by birds or small mammals or those that produce
or mature individuals) arrive at the site and their suc- long-lived seeds. Many forest plants seem to exhibit
cess at becoming established and surviving. Success in the latter adaptation, and viable seeds of pioneer
colonization depends to a great extent on there being species can be found in large numbers in some forest
a site available for colonization—a safe site where dis- floors. For example, as many as 1,125 viable seeds per
turbance has either removed competing species or re- square meter were found in a 100-year-old Douglas-
duced levels of competition and other negative biotic fir-western hemlock forest in coastal British Colum-
interactions to a level at which the invading species bia. Nearly all the seeds that germinated from this seed
can become established. For a given rate of invasion, bank were from early successional species (Kellman,
colonization of a moist, fertile site is likely to be much 1970). The rapid colonization of such sites after dis-
more rapid than that of a dry, infertile site because of turbance by clearcutting is undoubtedly in part a re-
poor survival on the latter. A fertile, plowed field is flection of the large seed bank in the forest floor.
rapidly invaded by a large variety of weeds, whereas a Another adaptation exhibited by many early colo-
neighboring construction site from which the soil has nizers is a symbiotic association with nitrogen-fixing
been compacted or removed to expose a coarse, infer- organisms. Lichens, typical colonizers of bare rock or
tile parent material may remain virtually free of vege- dry, nutrient-poor soils, may be partially composed of
tation for many months or even years despite nitrogen-fixing algae, whereas plant species such as
receiving the same input of seeds as the plowed field. red alder and lupine have symbiotic nitrogen-fixing
Both the rate of invasion and the rate of extinction microorganisms within nodules on their root systems.
vary greatly among different plant species. Pioneer or Nonnodulated symbiotic nitrogen fixation has been
fugitive species (those that occur only in the earliest attributed to several early seral tree species. This is as-
seral stages) tend to have high rates of invasion be- sociated with rhizosphere bacteria or bacteria in the
cause they produce very large numbers of reproduc- heartwood: e.g., pine (Bormann et al., 1993), aspen
tive propagules and because they have an efficient (Hendrickson, 1988), birch (Nohrstedt, 1988), and
means of dispersal (normally, wind). For example, a cottonwood (van der Kamp, 1986). Bormann et al.
study of succession in African grasslands revealed that (1993) suggested that pine forests may fix as much as
the average number of seeds produced per plant was 50 kg N ha! yr7!. Lack of available nitrogen is a char-
20,700 for pioneer colonizing species, 6,200 for the acteristic feature of early stages of primary seres, and
initial grass species, 272 for secondary grass species, this results in very low survival of plants that are not
and only 27 for the climax species (Jones, 1968). Pro- equipped to overcome this problem.
duction of reproductive rather than vegetative bio- A further adaptation that is well developed in colo-
mass also varies with time within a seral stage. Burning nizing species is a high degree of variation (polymor-
a tall grass prairie in Illinois was reported to increase phism) in germination behavior. Seeds of a given
the percentage of the annual biomass production in- species exhibit a wide range of germination dates
vested in flower stalk production to 53% from the 7% (Cavers and Harper, 1966; Salisbury, 1970), increasing
value found on unburned plots. The reduction of the probability that at least some of the seeds will ger-
flowering with time after burning was ascribed to the minate during a period of favorable environmental
478 CHAPTER17 — Ecological Succession

conditions. This is particularly important for species it is apparent that seeds or plants of later successional
that colonize an abiotic environment where there is no stages are present immediately after the disturbance
existing vegetation to ameliorate microclimatic ex- that initiates the succession. Different seral stages
tremes and in which there may be great microclimatic merely represent different growth rates and competi-
heterogeneity. Frost heaving of germinants of early tive abilities and variable periods of occupancy before
seral colonizers on exposed mineral soil is an impor- the species becomes dominant. The seeds or diminu-
tant process in determining which species get estab- tive plants of shrub species may be present during the
lished first. Having a deep taproot and variation in herb stage but remain ungerminated or unnoticed by
germination date are two important adaptations that the casual observer. Similarly, tree species may be
facilitate to colonization where frost heaving is a prob- present from the outset but take many years before
lem. If there is a significant year-to-year or decade- they germinate and/or overcome the shrubs and be-
to-decade variation in the amount and timing of pre- come dominant (Figure 17-4B). Undoubtedly, both
cipitation, then germination polymorphism will in- relay floristics and initial floristic composition are im-
crease the population’s ability to adapt to this. portant in succession, but their relative importance
Variation in rates of invasion and growth plays an will vary. In mesarch succession (e.g., old-field suc-
important role in determining patterns of succession, cession), initial floristic composition will often be the
especially secondary succession. Early seral species are dominant pattern, whereas in hydrarch and xerarch
those that produce abundant seed that is distributed succession, relay floristics will dominate. Similarly,
successfully to new sites. Such species generally grow initial floristic composition will generally be more
rapidly and quickly dominate such sites, excluding important in secondary succession than in primary
other species with lower invasion and growth rates succession.
(Drury and Nisbet, 1973). The first community that Knowledge of the relative contributions of these
occupies a disturbed area therefore may be composed two pathways to secondary succession in clearcuts and
of species with the highest rate of invasion, whereas other harvested areas is important for the design of
the community of the subsequent seral stage may con- strategies for the management of noncrop vegetation.
sist of plants with similar survival rates but lower inva- Control of vegetation that is competing with planted
sion rates. This may be the reason that on mesic sites, trees will differ if the problem arises from a seed
species from almost any seral stage can grow in almost and/or bud bank already on the site (initial floristic
any other seral stage if established before competition composition) or from current colonization of the
becomes severe. However, growth under such circum- ecosystem (relay floristics).
stances is generally much less than that achieved under
the seral conditions with which the species is normally
associated, and environmental alteration (facilitation)
Alteration of the Physical Characteristics
by pioneer species is frequently a prerequisite for suc- of the Ecosystem
cessful establishment of later successional species. It is Survival of a species that has invaded a site is a measure
probable that invasion rate is a more important deter- of its adaptation to and tolerance of the physical and bi-
minant of successional patterns in mesic than in xeric otic conditions of the site. However, by occupying the
and hydric seres (see below) and more important in site, a species inevitably changes the site conditions, and
secondary than in primary succession. the changes are frequently not favorable to the contin-
Before leaving the topic of colonization, we must ued occupancy of the site by that species or its ability to
reconsider the relative importance of what has been establish new seedlings. The changes may either reduce
termed the re/ay floristics and the initial floristic compo- the competitive/regeneration abilities of the resident
sition pathways of succession (Egler, 1954; see the sec- species or increase those of the invading species, or
tion Recognition of the Importance of Individual both. The net result is the replacement of one group of
Species). The conventional (monoclimax and polycli- species by another group. This is reflected in the
max) view of colonization has envisaged waves of resource-ratio approach to explaining succession
plants successively colonizing and occupying an area (Tilman, 1985) and in the structure of various comput-
and being replaced by the subsequent wave of new er models (e.g., Kimmins, 1993b; Kimmins et al., 1999;
species. Each seral stage prepares the site for the next Pacala et al., 1993, 1996; Smith and Huston, 1989). For
stage (Figure 17-4A). However, in some types of sere, example, shade-intolerant pioneer species create so
 Mechanisms of Successional Chan ge 479

Agricultural field
abandoned
SS
SSN
RN
ox S NI \
ENS Late

°CesSsiona]
Stages
Success;

Relative
of
dominance
plant
species

Agricultural field
abandoned

———— SSN o

. ee AQ
ee SS
WN
faa ay LEER RRRNS
Saree SRS
framers =] = a
Se PR AV’. s
[arene ae EEN s
Herbaceous ESS NY
weed 3

Relative
dominance
plant
of
species
Time ————>
(b)

Figure 17-4 Diagrammatic representation of relay floristic (A) and initial floristic (B) com-
position types of old-field succession as envisaged by Egler (1954). Most successions will ex-
hibit some combination of these two patterns. (Reproduced by permission ofthe International Society for
Vegetation Science and the author.)

much shade as their community develops that their own erally favors the nutrition of climax tree species over
seedlings either cannot survive or grow poorly, whereas early or mid-seral herb, shrub, and tree species. An in-
shade-tolerant seedlings of invading species flourish. teresting example of this is the acidification of soil be-
Failing to reproduce themselves, the pioneers are re- neath early seral stands of red alder because of nitrogen
placed by the subsequent seral community. Similarly, fixation and nitrification. This reduces the availability
the change in soil pH accompanying the accumulation of soil phosphorus, which in turn reduces the growth of
of tree litter and the development of a forest floor gen- alder. The growth of mid-seral Douglas-fir after the
480 CHAPTER17 — Ecological Succession

alder is not reduced because the mycorrhizal fungi on munity and the subsequent seral stage of shrubby wil-
its roots are able to access the phosphorus; on the con- lows. The shrub (willow thicket) stage does not last
trary, it is benefited by the increase in nitrogen avail- long, for it is invaded by red alder (also a nitrogen-fix-
ability caused by the alder (van Miegroet et al., 1990). ing species) and occasional cottonwood trees, which
In some types of primary succession, such alter- form a dense thicket that resists invasion for many
ation of the physical environment is an absolute pre- decades but is eventually replaced by a pure Sitka
requisite for the establishment of later seral stages. In spruce forest. The spruce is not climax, however. In
this type of situation, environmental modification is the absence of disturbance, spruce stands are invaded
probably the major driving force of succession, and by mountain and western hemlocks to give a mixed
the relay floristics model of succession would apply. In spruce-hemlock climatic climax. The rate of succes-
other types of primary succession and in many sec- sion is not uniform over the entire area, and some bare
ondary successions, this mechanism contributes to the ground remains uncolonized for a long time. Howev-
successional process but is not necessarily the major er, the transition from the pioneer community to the
driving force. willow-alder thicket stage generally begins approxi-
Some of the best-documented cases of environ- mately 15 to 20 years after the exposure of the surface,
mental change during a primary succession can be and complete cover by alder thickets is achieved after
found in studies of soil and vegetation development in another 35 to 40 years. The alder holds its own for ap-
the wake of retreating glaciers. Perhaps the best- proximately 50 years, but by approximately 120 years
known example of such a study is that conducted at after initial colonization, the succession to spruce
Glacier Bay, Alaska (Crocker and Major, 1955). There stands is essentially complete (Figure 17—5SA).
have been numerous advances and recessions of glaci- Accompanying the changes in vegetation are some
ers since the Pleistocene ice age (e.g., Harris and Farr, major changes in soil conditions. These are studied by
1974; Lawrence, 1950, 1958), the most recent retreat examining sites of progressively increasing age
at Glacier Bay starting in the mid-18th century at the (chronosequences) since the retreat of the ice. In the
end of the post-Pleistocene “Little Ice Age.” The area early stages, the soils are merely disorganized accumu-
that was exposed during the current retreat of approx- lations of morainal debris and till exhibiting little reg-
imately 96 km (60 miles) has undergone succession re- ular change in properties with depth. After invasion,
sulting in an age sequence of communities of up to 180 there is a progressive modification of many of the soil’s
years old. Succession commences with the invasion of properties, starting at the surface and gradually pro-
recently exposed surfaces by a pioneer community that gressing down through the soil profile. Bulk density of
can include mosses (Rhacomitrium species), herbs (fire- the fine fraction of the soil, which averages approxi-
weed, horsetail, sedges, and rushes), shrubs (mountain mately 1.4 g mL! in uncolonized areas, is reduced
avens [Dryas sp.] and dwarf willows), and trees (cotton- somewhat by frost heaving and then exhibits a contin-
woods). Almost all of the major species of the entire uing decrease to approximately 0.7 to 0.8 g mL! in
succession are physiologically capable of colonizing the spruce stage. These changes are largely confined
the bare surfaces, the observed sequence of communi- to the upper 15 cm, the zone that contains most of the
ties being due partly to differences in seed and spore roots. The values shown in Figure 17—5B are for the
dispersal and partly to differences in growth rates of upper 5 cm, with a progressive increase with depth to
the species with or without environmental modifica- 15 cm. All sites had a value of approximately 1.4 below
tion—especially the addition of organic nitrogen. 15 cm, irrespective of time since glaciation.
Plants in the pioneer community generally grow Soil pH, which is initially high (8.0 to 8.4) because
poorly unless the shrubs mountain avens (Dryas drum- of the presence of calcium carbonate-rich rocks, de-
mondit) and soapberry (Sheperdia canadensis) are present creases with the colonization of the surface at a rate
(Lawrence et al., 1967). Both ofthese species have root that depends on the type of vegetation. Whereas un-
nodules that contain nitrogen-fixing microorganisms. colonized surfaces exhibit virtually no drop in pH over
The nitrogen-fixing ability of the mountain avens (a 30 years, 20 years of colonization by cottonwood, wil-
creeping shrub) gives it a strong competitive advantage low, or mountain avens is associated with a reduction
in the nitrogen-poor environment exposed by the re- in the pH to approximately 7.8. The alder thicket
treating glacier, and it generally develops to form a stage is associated with a more dramatic drop to values
continuous mat between the initial pioneer com- in the surface soil of approximately 5.0 after 35 to 40
 Mechanisms of Successional Change 481

2 PAN 4)
er he
0 25 50 100 125 150 175 200
Time since reretreat of the glacial ice (years)
(a)

Upper 5 cm of Upper 5 cm of
mineral soil mineral soil

[ony
pH
Soil

fraction
soil
fine
of
density
Bulk
(g/cc)

Total soil Total soil

Forest floor
~
Forest floor
D= VFS \
/ \ .. Mineral soil

Organic
carbon
46
mineral
in
soil
to
cm
(g/m?)
nitrogen
Total
46
soil
mineral
in
to
cm

Figure 17-5 Changes in soil characteristics accompanying primary mesarch succession at

Glacier Bay, Alaska (after Crocker and Major, 1955). (A) Vegetation sequence over 200 years after
the retreat of the glacial ice. (B) Changes in mineral soil bulk density. (C) Changes in soil pH. (D)
Changes in soil organic carbon. (E) Changes in soil nitrogen. Note the rapid accretion of nitrogen in
the forest floor during the alder stage, followed by a decline during the spruce stage. Much of this
apparent nitrogen loss may be incorporation into spruce biomass. (Copyright Blackwell Scientific
Publications Ltd. Used with permission.)

years of occupancy. This reflects a litter pH of 5.6 to litter being no more acidic than the alder litter.
6.1 in the early stages and 4.2 to 4.6 in the late stages Changes in pH were largely restricted to the upper 25
of alder dominance (Figure 17-5C). The rapid decline cm, although some changes could be detected as deep
in pH ceases after the alder thicket stage, the spruce as 41 cm. Similar drops in soil pH accompanying post-
482 CHAPTER17 — Ecological Succession

glacial retreat succession have been recorded in many munities of pines or balsam fir. These communities
other studies: 2.3 pH units over 250 years in central give way to a climatic climax beech-maple community,
Alaska, 2.8 units over 500 years in Switzerland, 4.5 but many areas are characterized by prolonged occu-
units over 50 years in northern Sweden, and 3.0 units pancy by a subclimax black oak—blueberry community.
over 200 years in Australia. Primary succession on Surface soil pH changes during this succession from
sand dunes is also accompanied by pH declines: 2.7 7.7 to 4.0 over an estimated 10,000 years. A bleached
units over 280 years in western England and 0.9 unit Ae layer (the eluvial surface mineral horizon common
over 235 years in eastern England (Major, 1974). to many podzol soils) with a pH of approximately 4.5
Accumulation of organic carbon in both the min- forms after 1,000 years (a similar period to that re-
eral and organic soil layers increased at a fairly con- ported for Ae development in southwest Alaska
stant rate during the alder stage but leveled off during [Chandler, 1942], although Alaskan podzols have been
the spruce stage, at approximately 130 years after the reported to develop in as little as 150 years [Ugolini,
retreat of the ice. The rate of accumulation in the 1966]), decreasing to 3.6 units after 10,000 years. Or-
upper 46 cm of the mineral soil (most of it in the ganic carbon in the surface soil increases from approx-
upper 15 cm) was similar to that in the surface organic imately 0.05% in a 5-year-old dune to approximately
layers: approximately 15 g m ° yr | (Figure 17-5D). 1% at 80 years beneath a pine-bunchgrass community
The biomass of the forest floor increased most rapidly and to more than 2% beneath a black oak—blueberry
during the alder thicket stage, but accumulation con- community at 10,000 years. The persistence of the
tinued throughout the 180-year age sequence of com- subclimax black oak is interpreted to be the result of
munities to reach a maximum mean depth of 13 cm the acidity of the accumulating organic layer, which
and a biomass of 90 to 100 tha '. This is similar to the inhibits invasion by maple and beech.
accumulation of 212 t ha! forest floor material with a A third type of environment in which an age se-
mean depth of 16 cm under a spruce—hemlock stand quence of successional communities can be studied is
developed over 300 years on a landslide area elsewhere river terraces. Fonda (1974) examined the communi-
in Alaska (Gregory, 1960). ties of the successively older river terraces to be found
The continued accumulation of organic carbon is with increasing distance on each side of the Hoh River
believed to be related to the pattern of nitrogen accu- in western Washington. A pioneer community of wil-
mulation. Initial nitrogen accumulation has been as- lows and herbs colonizing gravel bars is replaced by
cribed to fixation by the algal crust that develops on alder as the bars develop into alluvial flats (80 to 100
bare areas and to the root nodules of the soapberry years). The alder is replaced by a Sitka spruce—bigleaf
and mountain avens in the pioneer shrub community. maple—cottonwood community on the first river ter-
Far greater accumulation occurs during the alder race (400 years). This gives way to a Sitka spruce—west-
stage, which has an estimated net fixation rate of 62 kg ern hemlock community on the second terrace (750
ha! yr-'. Nitrogen accumulation in the forest floor years), whereas climax hemlock stands are found on
levels offin the alder-spruce transition stage and then the older third terrace, areas first exposed by the re-
declines substantially, probably as a result of reduced treating Pleistocene alpine glaciers. Table 17-2 shows
nitrogen fixation and of uptake and incorporation in the variation in soil physical properties observed along
spruce biomass. The decreasing nitrogen and constant this chronological sequence. Nitrogen concentrations
carbon levels in the forest floor result in an increase in decline over the first three stages; organic matter,
the carbon:nitrogen ratio from approximately 13 in phosphorus, and moisture-holding capacity all tend to
the alder stage to 33 to 36 for the spruce forest floor. increase; and pH shows a sustained decline.
Undoubtedly, this reduces rates of litter decomposi- Studies of succession on north-facing slopes in inte-
tion and results in the continued accumulation of for- rior Alaska have revealed a progressive change in soil ni-
est floor biomass (Figure 17—5E). trogen and phosphorus and the development of
Other studies of changes in soil accompanying sphagnum bogs on sites formerly occupied by produc-
succession have reported similar results. Olson (1958) tive forest (Heilman, 1966, 1968; also, van Cleve and
documented soil changes over a 10,000-year succes- Viereck, 1981; van Cleve et al., 1986). The succession
sional sequence on sand dunes on the shores of Lake starts with a productive birch-alder community, which is
Michigan. This psammosere develops from bare sand gradually invaded by white and black spruce. This leads
surfaces through grass and shrub stages to seral com- to the formation of a black spruce-moss community
 Mechanisms of Successional Change 483

Table 17-2 Variation in Soil Properties of the Upper 4 cm of the Mineral Soil in a Chronosequence of Communities
on the Banks of the Hoh River, Washington
Seral Stage
Soil Spruce-Maple- Spruce-
Property Alder Cottonwood Hemlock Hemlock
% nitrogen 0.77 0.20 0.14 0.47
% phosphorus 0.58 0.70 0.47 1.08
% organic matter 4.5 Ue 5: 18.6
pH 5 4.9 4.4 4.0
Moisture retention 7.4 11.6 19.7 24.8

Data from Fonda, 1974.

characterized by Hylocomium, Pleurozium, Polytrichum, coastal beach deposits, through a mature Sitka spruce
and Dicranum mosses. The buildup of the moss layer in- stand, into a spruce-sphagnum community and even-
sulates the soil, lowers soil temperatures, and raises the tually a sphagnum bog. The spruce stand that devel-
level of permafrost in the soil. Under these conditions, ops on the freely drained mineral beach materials
the above moss species are replaced by the peat-forming develops a thick, acidic forest floor. Under the cool,
moss Sphagnum. Vhe accumulation of Sphagnum peat humid, high-precipitation climate of the region, this
further reduces soil temperatures and increases water- leads to podzolization of the mineral soil. Iron is
logging of the upper layers of the soil, which leads to a leached from the upper mineral horizons to be de-
decrease in the density of the black spruce stand and the posited as a B; horizon. After centuries of this process,
formation of an open Sphagnum-—black spruce bog or the iron accumulation becomes so great that an iron
muskeg (Bonan, 1992a,b). pan is formed: a layer of mineral soil in which the min-
‘Temperature and moisture changes beneath the eral particles are cemented together by iron and alu-
developing vegetation are undoubtedly driving mech- minum oxides. The iron pan develops to the point at
anisms in the succession, but it has been suggested which it impedes drainage and the soil becomes boggy.
that the concomitant change in soil nitrogen and This leads to the development of sphagnum moss and
phosphorus is also very important (Figure 17-6). As a layer of sphagnum peat, which, in conjunction with
succession proceeds, the concentrations of available the impeded drainage, leads to the death of the trees.
nitrogen and phosphorus in the rooting zone (the The result is a sphagnum bog.
upper soil layers) decline because of lower soil temper- Ugolini and Mann’s theory seems plausible until
atures, increasing moisture content, and the formation one realizes that few of these coastal sphagnum bogs
of low-bulk-density moss peat that has very low con- are presently underlain by an iron pan, but there is an
centrations of nutrients. The highest soil nutrient explanation. As the sphagnum bog builds up, organic
concentrations are found in progressively lower and acids that form organic complexes with the iron are
colder layers of the soil, out of reach of the trees, produced. Over a long period of time, the result is the
which consequently experience increasing nutrient dissolution and disappearance of the pan. This does
deficiency. The more-nutrient-demanding birch and not lead to a reversion of the area back to forest, how-
alder are replaced by the less-demanding black spruce, ever. The thick layer of sphagnum peat that has devel-
but even this species is adversely affected as the avail- oped by the time the pan disappears is more than
ability of nutrients continues to decrease. The produc- capable of reproducing the hydrological effects of the
tivity of the spruce decreases, and eventually even this pan. A similar change in vegetation and soil conditions
species is excluded, leaving a sphagnum bog. is reflected in a pollen profile from a bog forest in cen-
A theory concerning the succession that leads from tral coastal British Columbia (Figure 17-7).
forest to bog in coastal Alaska was advanced by Ugo- In addition to the important effects of succession
lini and Mann (1979). They examined a sequence of on soils, the successional development of vegetation
communities from a recently established forest on produces significant alterations in microclimate. Be-
484 CHAPTERI7 — Ecological Succession

Sphagnum- ,
Whi Black spruce- Sphagnum- bog Soil
sate moss Black spruce- black spruce depth (cms)
: sphagnum
Birch-alder Ay Aue ie b A 4} IS
Vas Organic
Was a
wake wae nasa
eevee yuy wee Soil
— 0
6°c /
Inorganic

()°e ee saYs Ws 150

Time
(a)

Soil ee ae c oncentration
ion of of available
available phosphoru
phosphorus
depth Concentration of nitrogen (mg/cc) depth (mg/cc)
g
(cms) 0 | 2 3 (cms) 0 01 .02 .03

Birch-alder
; Birch-alder

Pa
< Black spruce-sphagnum 20

50 Are
AG 6 Black spruce-sphagnum

Sphagnum-
black spruce

100 60

September soil
temperature 80 a Sphagnum-black spruce
soeeseceo- >4°C
24°c
150 HHH (0-2°c
# permafrost

(b) (c)

Figure 17-6 Variation in the distribution of nitrogen and phosphorus with depth in the soil
during a successional sequence on a north-facing slope in interior Alaska (after Heilman, 1966,
1968). The process whereby soil nitrogen and phosphorus become located in deeper and colder lay-
ers of the soil as succession proceeds as the result of the accumulation of an increasingly deep layer of
moss is readily apparent. (A) Vegetation sequence and changes in the depth of the organic layer and in
September soil temperature (tree height and soil depth not drawn to the same scale). (B, C) Variation
in the concentration of total nitrogen (B) and available phosphorus (C) with depth in the soil.
(Copyright Ecological Society ofAmerica. Used with permission ofESA and the author.)

cause the effect of vegetation on microclimate was dis- creased, and the evaporative power of the air beneath
cussed in Chapters 7 to 11, the topic will not be elabo- the vegetation canopy is reduced. For example, Cain
rated here. It will suffice to remind the reader that and Friesner (1929) found that loss by evaporation
light intensity and wind speed are reduced, tempera- from porous ceramic bulbs — through succes-
ture extremes are moderated, relative humidity is in- sive seral stages from 14.6 mL day! in an old-field to
 Mechanisms of Successional Change 485

Pacific
Lodgepole ‘ibe Sitka silver Red Skunk Sweet Red and Ericaceous
ae eee pine emlock spruce fir alder cabbage oie gale yellow cedars _ plants ee
y J \ N
Me \
Z SS NN AN

\ \ 7 Y gear
N ‘ C
\ J ta NS m

50 cm ( é
1 \ ie

:
oO
Spt
?
ee
eC
}
7
~ ans
N pas
=z Nias \ Y Wis %
‘2 100 cm N
IN BS
\ipa< ent es
a

\ ‘} \ y P\E
6eb
p—
\
N
Ay tet“ae
N a =

Neri
\yatees N\ ra
i=
150 cm
\eresseneh
\\ +\ elNN ‘
\— sete
jee lS N ae
. yf Ne
wag aa
\
\N\A en
y 3
Oe/ 8710 + 210
200 cm 5 1 ; Years before
e _ 20 0 20 40 OTOKO 20 AOR CO Oi 20 present
~
a 2 % composition of the pollen
Gs}

Figure 17-7 Vegetation change in central coastal British Columbia since the retreat of glac-
iers as reflected in the pollen profile from a bog-forest near Prince Rupert. The change in
vegetation certainly reflects changes in climate over the past 9000 years, but it also reflects changes
from a community growing on fertile mineral soil to a community growing on 2 m of slowly decom-
posing, nutrient-poor peat, a change that has resulted from the processes of succession (Banner et
al., 1983). The area beneath the dotted line represents a 10x expansion of the striped area beneath
the solid line. (Copyright National Research Council of Canada. Used with permission of NRC and the au-
thors.)

4.7 mL day| in a climax forest. Such changes in mi- Displacement of Species by Antibiosis,
croclimate are of great significance for plants, animals,
Autotoxicity, and Competition
and microbes and normally contribute to successional
change. However, microclimatic effects of early seral Not only do plants alter the microclimate and the
vegetation can also slow succession. physical and inorganic chemical characteristics of the
Blue-joint grass (Calamagrostis canadensis) often in- soil, they also alter their organic chemical environ-
vades clearcuts on moist mineral soils in western ment. In the discussion of allelopathy (Chapter 15), it
Canadian boreal forest. Dense grass communities not was seen that plants produce a wide variety of chemi-
only compete with invading conifer seedlings for light cals to inhibit germination and/or growth of other
and nutrients and may cause winter “snow press” but species. This adaptation plays a significant role in suc-
also may lower soil temperature. Slow decomposition cession. In some cases, allelopathy serves to accelerate
of the grass litter forms a light-colored mulch that in- succession, whereas in others, it impedes it.
sulates the soil and reflects sunlight. This reduces soil The order in which different species of plant in-
temperature, which results in reduced tree seedling vade abandoned fields in Oklahoma has been shown to
root growth, water uptake, stomatal conductance, and be related to the plants’ nitrogen requirements, and
photosynthesis rates, thereby delaying tree growth the low level of soil nitrogen that is characteristic of
and canopy closure that would shade out the grasses such fields seems to be a major factor in determining
(Hogg and Lieffers, 1991). the successional sequence (Rice, 1964, 1967, 1968).
486 CHAPTER17 = Ecological Succession

Any factor that influences the availability of nitrogen conditions in which tree regeneration, especially cer-
thus would affect the rate of succession. Several of the tain conifer species, is slow to establish, is chlorotic
pioneer plants in this old-field succession that are tol- (i.e., yellow rather than green, which usually reflects
erant of low nitrogen availability have adapted to pro- poor nutrition), and grows poorly. Trees may even be
long their occupancy of the site by producing entirely excluded and shrub climaxes that are very sta-
allelochemicals that inhibit nitrogen-fixing and _nitri- ble over time scales of centuries may form, thereby ar-
fying bacteria and thereby impair the growth of later resting succession at an early stage. Conditions that
seral plants. This may be a fairly common adaptation favor the development of ericaceous heaths include
among early seral plants. For example, clover plants, arctic and subarctic climates; the removal of forest and
which exhibit extremely poor root nodulation and prevention of reforestation in cool, humid temperate
have poor survival in the presence of certain species of regions; and humid climates in alpine and subalpine
grass in New Zealand, failed to respond to even heavy climates (Gimingham, 1972). Dominance of erica-
fertilization, with or without trace elements. Killing of ceous species is not only an early seral phenomenon,
the grasses by herbicide, conversely, permitted normal however. In the absence of periodic disturbance, such
growth of and nitrogen fixation by the clover by elim- as fire, closed-canopy forests in these areas may be re-
inating the nodule-inhibiting influence of the grasses placed by ericaceous communities: ericaceous wood-
(Beggs, 1964). However, the success of a formalin land or ericaceous heaths (heath refers to a plant
treatment of the soil in overcoming the inhibition of community dominated by ericaceous shrubs) that lack
the clover and of nodulation suggests that soil mi- trees. Disturbance of forest that removes trees without
croflora may also have been involved. promoting their regeneration (certain types of natural
In contrast to the findings in Oklahoma, an early mortality, certain types of fire, or partial forest har-
seral species in old-field succession in North Carolina vesting that fails to disturb the soil and/or an existing
has been shown to produce a substance that hastens its ericaceous understory) can also convert forest to erica-
own demise and speeds the transition to the subse- ceous woodland or heath.
quent community. Horseweed (Aster canadensis), a Initially it was thought that successional inhibition
winter annual, is the first species to colonize aban- by ericaceous shrubs was a result of competition for
doned fields, but it rapidly disappears from the com- moisture and nutrients. Many of these shrubs have ex-
munity because growth of additional horseweed tensive rhizomes (underground stems) and _ prolific
seedlings is inhibited by chemicals released from the fine-root systems, and it is clear that they compete
decaying roots of the parent plants (Keever, 1950). strongly for nutrients (Messier, 1993; Messier and
There are numerous other examples of such Kimmins, 1990). Fertilization, which temporarily alle-
autotoxicity among pioneers in old-field and other suc- viates competition for nutrients, can promote tree
cessions (see Whittaker, 1970). For example, the alle- growth in the presence of ericaceous shrubs. However,
lochemical habit of chaparral can turn to autotoxicity in studies of the negative effects of heather (Calluna
and kill the center of old shrub patches if such heavy vulgaris) on Sitka spruce growth in Scotland, it was
accumulations of terpenes (which reduce competition found that earlier successional tree species such as
from grasses) occur that they become toxic to the pro- pine, larch, and birch are much less affected than later
ducing plants (Muller, 1969, 1970). successional species such as spruce, hemlock, and true
Sometimes, the loss of vigor or the death of plants firs (Abies sp.) (Weatherell, 1953), and removal of the
in the center of shrub patches is more the result of old ericaceous shrubs does not always alleviate the inhibi-
age than autotoxicity, as is the case in heather (Ca//una tion (Malcolm, 1975). Research showed that in the
vulgaris) in Scotland (Barclay-Estrup and Gimingham, case of heather, there may be an alleopathic effect that
1969; Watt, 1947, 1955). interferes with the formation of mycorrhiza in spruce
There has been much debate over the mechanisms that can access nitrogen and phosphorus contained in
involved in the widely observed inhibitory effects of recalcitrant (slowly decomposing) organic matter
ericaceous shrubs on trees and on succession. Species (Handley, 1963; Harley, 1952; Robinson, 1972). Pine
of the family Ericaceae are one of the most ecologi- and other early seral species are able to form mycor-
cally successful taxonomic group of plants in the rhizae with fungi that can access the nutrients in this
northern hemisphere (de Montigny and Weetman, material. When spruce is grown in mixture with early
1990). They are commonly associated with heathland seral tree species, it is also able to form appropriate
 Mechanisms of Successional Change 487

mycorrhizae and the inhibitory effect of the heather is soil associated with salal could reduce phosphorus up-
removed. When grown with pine or larch, on these take by conifer seedling roots, and seedlings watered
sites, spruce shows vigorous growth (Malcolm, 1975). with salal leachate had drastically reduced phosphorus
An interesting example of an ericaceous shrub ser- uptake and significantly reduced growth (Prescott and
climax (a very stable and persistent shrub seral stage) Weetman, 1994). It was also shown that the ericoid
that may possibly involve allelopathic mechanisms mycorrhizae of salal, which can use organic forms of
is that of the ericaceous shrub Kalmia angustifolium nitrogen in pure culture, are antagonistic to the fungi
in eastern Canada. If this species becomes well es- that form ectomycorrhizae with western hemlock
tablished, such as after a fire, then it can form a (Xiao, 1994). Although the suspected allelopathic role
dense shrub community that resists tree invasion for ex- of salal does not seem to be as dominant as in the case
tremely prolonged periods. Damman (1971) studied of Kalmia and heather, salal clearly plays an important
balsam fir, black spruce, and a Kalmia heath growing on role in the succession of these coastal forests.
very similar mineral substrates in western Newfound- Allelochemicals serve to modify the competitive
land. He found that the Ka/mia heath had a raw humus relationships of species, but competition itself, particu-
layer mass of 293 t ha“', in comparison with 87 t ha ! in larly for light, plays a dominant role. Early succes-
the Picea forest and 65 t ha! in the Abies forest. These sional plants are generally shade intolerant and small
accumulations represented 78, 21, and 14 times the an- in stature, whereas later successional species are
nual litterfall, respectively, suggesting very slow litter generally shade tolerant and taller in stature. Shaded
decomposition on the Kalmia heath. Mineralization of seedlings are more susceptible to attack by fungi (Vaar-
nitrogen and phosphorus showed similar trends, and taja, 1952), and the ability to resist such attacks is im-
calcium mineralization was much slower in the heath portant in determining the role of a species in
than in the forest. It is interesting that a net increase in succession. As noted in Chapter 7, ability to survive in
nitrogen of 23 kg ha 'yr_'was calculated for the heath. the shade is related to seed weight and respiration rate.
Only half of this could be accounted for by precipita- Seeds of shade-tolerant, late successional species are
tion, and, in the absence of known nitrogen-fixing often larger and the seedlings often have lower growth
plants, the difference was ascribed to dust, pollen, or and death rates in the shade than shade-intolerant,
the direct absorption of ammonia from the air by plants early successional species. However, early successional
or the soil. This study shows that Ka/mia heaths inhibit species have a much higher growth potential than
mineralization and accumulate nutrients in an unavail- shade-tolerant, late successional species when they all
able form. Whether the resulting low availability of nu- are fully illuminated (Grime, 1966; Grime and Jeffrey,
trients is sufficient to explain the lack of tree invasion or 1965). It has been suggested that successional replace-
whether the chemicals that inhibit decomposer organ- ment is closely involved with competition for light and
isms also inhibit tree mycorrhizae and/or tree root de- that light adaptation represents a compromise between
velopment is unclear. It would be interesting to the ability to grow rapidly in full light and the ability
investigate whether Ka/mia resists tree invasion by the to survive fungal attack under shaded conditions.
same mechanisms as Ca//una (see Titus et al., 1995). Competition for water often parallels competition
A final example of the role of an ericaceous species for light, as both shading and moisture loss by transpi-
in directing succession is that of salal (Gau/theria shal- ration are closely related to leaf area. Herbicide re-
Jon), a broad-leaved, evergreen shrub common in the search has repeatedly shown dramatic increases in tree
coastal forests of British Columbia and the U.S. Pacif- growth after removal of competing early seral vegeta-
ic Northwest. Salal can form dense cover and compete tion (Walstad and Kuch, 1987; Woods et al., 1992), and
strongly for nutrients in humid areas and for both nu- this has frequently been attributed to a reduction in
trients and water in drier areas. As with the Scottish water competition (e.g., Squire, 1983). However, water
example with heather, the growth of Sitka spruce and nutrient availability are so intimately related that it
planted in salal communities is severely inhibited. Fer- is difficult to evaluate them separately in investigations
tilization or salal removal alleviates the effect, and it of competition (Nambiar et al., 1984; Nambiar and
has been shown that much of the effect can be ex- Sands, 1993), and on moist sites, competition for nutri-
plained by salal competition, mainly for nitrogen ents may be a major factor in delaying the replacement
(Messier, 1993; Messier and Kimmins, 1990), How- of early seral communities by closed forest cover (e.g.,
ever, it was shown that phenolic acids and tannins in Neary et al., 1990). Yield decline of Chinese fir
488 | CHAPTER17 — Ecological Succession

(Cunninghamia lanceolata) in areas of southeast China is 1. The climate and the soil conditions: climate affects
related both to declines in soil fertility caused by inap- the biological and ecological processes that are the
propriate forest management and to nutrient competi- mechanisms of successional change, and the avail-
tion caused by alangalang grass (Imperata cylindrica) and ability of soil resources determines the growth and
other early seral herb and shrub species that invade de- productivity of the plants that are responsible for
graded sites and form dense communities. autogenic succession.
Although competition for water and nutrients is 2. The degree of environmental change that must
not included in many of the successional models being occur before the community of one seral stage can
used at present, this is clearly an important succes- be replaced by that of another: the greater the
sional mechanism, and these models should be modi- change, the more prolonged the stage.
fied to incorporate it if they are to be applied in 3. The productivity of the organisms and the efficiency
ecosystems where moisture and nutrients are limiting.
with which they produce environmental change: the
The focus on light competition and the omission of
more productive and efficient the organisms, the
moisture and nutrient competition is clearly an unac-
more rapidly they are able to affect environmental
ceptable oversimplification in the simulation of many
change and the greater their ability to compete with
intensively managed ecosystems, especially on poor
and displace the existing community, in both cases
and dry sites. A useful review of many aspects of plant
speeding the process of successional replacement.
competition is provided in Grace and Tilman (1990).
These examples will suffice to demonstrate that 4. The longevity of the organisms that dominate each
plants do not occur in successional sequence merely as seral stage: the longer-lived the organisms, the
a result of the physical conditions of their microenvi- longer the stage may last.
ronment. Interactions between plant species with dif- 5. The degree to which communities at any particu-
ferent life history characteristics and between plants lar stage occupy and dominate the site and resist
and microbes and plants and herbivores contribute invasion by other species: the better developed the
significantly to the order and rate with which species community, the more it is controlling access to the
replace each other. site’s resources, the more resistant to invasion, and
As noted at the beginning of this section, the ex- the longer lasting it will be.
planation for the successional replacement of species 6. The frequency and severity of disturbance that
differs from one situation to the next. Various combi- causes successional retrogression: if the environ-
nations of invasion rate; shade tolerance; allelopathy mental alterations produced by a seral stage (e.g.,
and autotoxicity; environmental modification; compe- accumulation of nutrients and an organic forest
tition for light, nutrients, and moisture; old age and floor) are repeatedly removed by disturbances such
senescence; seed longevity and availability of seed as fire or erosion, then successional progress may
sources; and ability to sprout from subterranean or- be very slow.
gans are responsible for the general patterns of species
replacement. These mechanisms vary considerably
from site to site and from one time to another. Under- Climatic and Soil Conditions
standing plant succession obviously requires a good Climatically related conditions such as short growing
knowledge of physiological ecology: the physiological seasons, cold soil, drought, and hot summers that slow
adaptations of plants that determine their responses to nutrient cycling, reduce leaf area, limit net primary
the physical environment and the other plant and ani- production (NPP), and reduce seed production will
mal species (Bazzaz, 1979; Bazzaz and Pickett, 1980; slow many of the processes that are responsible for au-
Givnish, 1988; and Rice 1984; for a more recent treat- togenic succession. Succession is normally much slower
ment, see Smith and Hinkley, 1995a,b). in hot, dry, cold, and climatically extreme environ-
ments than in warm, humid, and equitable climates.
Nutrient-poor and dry soils will generally limit
17.6 Rates of Successional Change rates of succession by limiting leaf area development
and aboveground production. Moist, fertile soils that
The rate at which succession proceeds is highly vari- promote high leaf area, NPP, and aboveground alloca-
able. It depends on tion will generally have much faster rates.
 Rates of Successional Change 489

Degree of Environmental Change high biomass production that characterizes mosses in

During a Succession later stages of mesoseres in humid climates. Phyto-
When one considers the total set of ecological condi- plankton in early hydrarch succession may have high
tions at the beginning and end of a xerosere, a individual photosynthetic efficiencies (if not limited
mesosere, and a hydrosere, it becomes obvious that a
by low levels of nutrients), but they do not accumulate
much biomass because they are so short lived and
far greater change occurs during xerarch and hydrarch
heavily grazed and most of their remains decomposes
than during mesarch succession. In a typical primary
fairly rapidly. In common with the roots of higher
lithic xerosere, one starts with a completely unmodified
plants, lichens and mosses produce organic acids that
microclimate and little soil or unconsolidated soil par-
accelerate the weathering of rock. This biochemical
ent material. There is little or no soil moisture storage
weathering is important, but in pioneer xerarch stages,
capacity, nutrient retention capacity, or available nutri-
it is limited by the lack of moisture and the low biotic
ent capital and little in the way of a rooting medium to
productivity, and it is certainly much less significant
provide stability for higher plants. At the climax, one
than the more extensive chemical breakdown of rock
finds a forest microclimate, a well-developed organic
that occurs under well-developed, moist organic layers
layer, the beginnings of a mineral soil, a relatively large
later in xerarch succession. Thus, the extensive eco-
site nutrient capital, an active nutrient cycle, a signifi- logical changes that are necessary early in xerarch and
cant soil moisture storage capacity, and reasonably hydrarch succession are associated with diminutive
good physical support for the trees (Figure 17-2). and frequently slow-growing life forms or low rates of
The typical primary hydrosere also involves exten- organic matter accumulation. Consequently, rates of
sive change in the physical and chemical environment. autogenic succession are very slow in these seral
From a nutrient-poor aquatic environment, one pro-
stages. The allogenic processes of lake sedimentation
ceeds through a semiterrestrial condition to a forested and bedrock weathering are also extremely slow, so
condition that will eventually bear a close resemblance these early stages are generally very long—often as
to the climatic climax of the xerosere and mesosere. much as many hundreds or thousands of years.
The ecological conditions between the start and the The situation in mesarch succession is different,
finish of the sere are enormously different. with very much less environmental alteration involved
The typical mesosere, however, involves much less and a productive, rapidly growing community of
environmental change. Soil moisture conditions gen- plants involved in producing it. Relatively little in the
erally do not change radically, and with the important way of allogenic processes is involved, and the auto-
exceptions of the accumulation of an organic layer, the genic changes generally proceed just as fast as the veg-
development of soil horizons, and a change in site nu- etation is able to develop. Most mesarch seral stages
trient capital and availability, there may not be a great involve a single generation of plants, whereas most
change in the soil. The major changes are in microcli- xerarch and hydrarch seral stages involve several to
mate and surface soil conditions, both of which can be many generations within a seral stage. Consequently,
changed rapidly by plant growth. As a result of these the early stages of mesarch succession develop rapidly
differences, xerarch and hydrarch successions tend to and are associated more with the next two factors than
be slow, especially in the early stages, whereas mesarch with the preceding two.
succession tends to be much faster.

Longevity of the Organisms That

Rate at Which the Biota Alter Dominate the Site
the Environment In mesarch succession, the length of a seral stage is
The living community in early xerarch and hydrarch partly determined by the longevity of the organisms
succession is composed of small organisms that either involved, especially in midseral and subclimax stages.
grow very slowly or are short-lived and accumulate The duration of the alder—thicket stage at Glacier Bay
very little biomass. Lichens grow so slowly that there is largely determined by the physiological longevity of
is even a science of /ichenometry, which measures time the alder, which resists invasion until it begins to
since historical disturbance events in terms of lichen senesce. The shrub stage on subhygric and hygric sites
growth. Pioneer dryland mosses similarly lack the in coastal British Columbia can last for several decades
490 CHAPTER17 — Ecological Succession

until the salmonberry thickets that frequently develop above ground that tree seedlings are unable to invade,
on such sites start to senesce. The subclimax Douglas- even after low-severity, stand-replacing natural distur-
fir stage in the western hemlock zone of British Co- bance or clearcutting. Only significant soil disturbance
lumbia can last for many centuries before it is replaced and disruption of herb and shrub roots and rhizomes
by the climax western hemlock—-western redcedar will result in the reestablishment of closed forest. ‘This
stage, simply because of the great age reached by is a major reason that foresters generally advocate
Douglas-fir. Similarly, the subclimax Sitka spruce prompt regeneration of harvested sites, by planting
stage on the B.C. and Alaska coast may last for as long where natural regeneration is slow or unreliable.
as half a millennium or more before being replaced by
the climatic climax hemlock stage simply because of its
longevity. The short duration of early mesosere stages Frequency and Severity of Disturbance
may simply relate to the high percentage of annuals or Succession will not proceed or will proceed very
biennials in the early plant communities of this sere. slowly when there is frequent disturbance of sufficient
severity to permit reinvasion by plants of the same or
Degree to Which Communities Dominate earlier seral stages or the environmental modification
that has been achieved is reduced or removed. Distur-
the Site and Resist Invasion
bance that merely serves to remove the dominants of a
The climax community, where it develops, persists be- seral stage (e.g., insect defoliation and death of early
cause it fully occupies and dominates the site and re- seral hardwoods; windthrow of canopy dominants)
sists the invasion of nonclimax species. It may do this without significant soil disturbance or loss of microcli-
by competing efficiently for light, by making nutrients matic conditions may accelerate succession. However,
unavailable to all but appropriately adapted climax in fire- and insect-dominated successions, ecosystems
species (cf. direct nutrient cycling), by competing for may rarely get past mid-seral conditions before being
moisture, or by inhibiting germination of seeds of redisturbed to an earlier stage. In such environments,
other species by allelochemical mechanisms. Similarly, progress toward late-seral stages will be very slow or
earlier seral communities can become very long last- will not occur.
ing if they ever become so well established that they Various combinations of these six determinants of
are able to dominate the site and resist invasion. Shrub the rate of succession and the duration of particular
stages in particular, once well established, may become seral stages precludes any reliable general statement
extremely resistant to invasion and colonization by the about rates and durations. The following points are
tree species of the subsequent seral stage. Ordinarily, probably as far as it is reasonable to go at present. The
seedlings of such trees become established beneath the reader is reminded that as with all generalizations,
shrub community soon after it gains control of the they will not always hold true.
site, and after a relatively brief period of competition,
the trees grow above the shrubs and shade them out
(initial vegetative composition). However, if invasion
1. Overall rates of succession are generally much
by the trees is delayed until the shrubs have become slower in primary than in secondary succession be-
large and completely occupy the site both above- cause of the greater degree of environmental alter-
ground and belowground, then it may be many ation that is involved. For ecologically comparable
decades before the trees can enter the community. seral stages in primary and secondary succession
This phenomenon has been applied on a limited scale (i.e., middle stages of primary succession compared
in the establishment of virtually maintenance-free, with early stages of secondary succession), the dif-
low-stature vegetative cover below power lines as an ference in rates is less than for the entire sere.
alternative to continual mechanical or herbicidal con- 2. Rates of succession are much faster in mesarch
trol of tree growth (Bramble and Byrnes 1972; Egler, than in xerarch and hydrarch succession.
1953; Niering and Egler, 1955). It is a common prob- 3. Rates of succession in the earlier stages of xe-
lem where regeneration is delayed after timber har- roseres and hydroseres are slower than in later
vesting and herbs or shrubs capture the site. Similarly, stages. The opposite is true for mesarch succes-
herbs and shrubs in gaps in high elevation, snow- sion, in which earlier stages are succeeded more
dominated forests may so dominate both below and rapidly than later stages.
 The Three Major Types of Sere 491

4. ‘The duration of any particular stage will be greatly useful framework for the discussion of succession, and
influenced by the timing and rate of invasion of the the following descriptions of examples of these three
site by reproductive propagules of individuals of the seres serve as a model with which real examples of suc-
subsequent seral stage. Where such invasion is slow cession in other areas can be compared. The examples
or delayed, a seral plant community may become given below are from one of the dry subzones of the
very well established, resist invasion, and conse- western hemlock biogeoclimatic zone of coastal
quently last very much longer than where such in- British Columbia (Krajina, 1965): a humid region with
vasion is rapid and immediate (cf. relay floristics vs. warm summers, mild winters, and annual precipitation
initial vegetative composition). The relative dura- of 110 to 439 cm. Different species will be involved in
tion of different stages consequently is variable. other climatic areas, and different patterns of succes-
5. Succession will be much faster in climates that sion can be found. However, the overall pattern and
promote high rates of NPP and biomass accumu- processes of community changes described are com-
lation than in climates that limit plant growth. mon to the primary successions of many temperate
forest regions.

17.7 ‘he Three Major Types of Sere Primary Xerarch Succession

Succession and the resulting seres occur on a wide va- and the Xerosere
riety of terrestrial substrates: fine-textured alluvial de- ‘Typical primary xerarch succession (see Figures 17—2
posits; glacial tills; and coarse, gravelly river terraces, and 17—8A) can be observed on a rock surface exposed
sand dunes, and rock surfaces, all of which can have by a landslide or a road cut or on a rocky ridge after a
originated in parent material varying in chemistry severe wildfire or escaped slashburn. Such succession
from limestone to siliceous sandstone. They also results in a primary lithosere. The physical environ-
occur in climates that vary from equatorial to polar ment at the start of the sere is inhospitable for most life
and from alpine to maritime. It is to be expected, forms. There is little or no water storage capacity on
therefore, that although succession generally follows the site, and water is limiting to plant growth, espe-
one of a few broad patterns, the details vary greatly cially during the growing season. Daytime tempera-
from place to place. tures can be very high as a result of full illumination,
The major classification of seres is not based di- and radiation cooling at night can result in a large diur-
rectly on soil texture, type, depth, or chemistry but on nal temperature fluctuation. Unless the rock is well
variations in the availability and ecological effective- fractured or fissured, the lack of soil for physical sup-
ness of water (see Chapter 10). The moisture status of port precludes the growth of plants with an upright
a site is often closely related to other soil characteris- growth form, and there is virtually no capital of nutri-
tics, and such a classification implicitly reflects many ents available on the site for plant growth. Nitrogen in
other soil physical and chemical properties. particular is limiting, and although mineral nutrients
Seres are classified as xeric, mesic, or hydric. These are released from the weathering surface of the rock,
terms were introduced in Chapters 6, 10, and 11, and most of them are washed away because of the lack of
the reader is reminded that they can be used in either cation exchange sites and plant roots. The major source
an absolute or a relative sense. The former refers to of many nutrients may be precipitation and dust.
the absolute water content of the soil, whereas the lat- Seeds and spores of many plants are continually ar-
ter refers to the availability of soil water to organisms riving at such sites. Most are blown or washed away or
relative to the regional climate (see Chapter 6). In suc- fail to germinate, or the young plants die shortly after
cessional terminology, these terms are used to imply germination for lack of the basic prerequisites of plant
very dry, moist, and very wet environments. life. Thus, the pioneer plant community is limited
Xeric, mesic, and hydric sites represent the two ex- to crustose lichens and pioneer mosses (e.g.,
tremes and the modal conditions in what is in reality a Rhacomitrium sp.) that are tolerant of drought, infertil-
continuous variation in moisture status. Most sites ity, and lack of physical support. The former grow
represent an intermediate situation (¢.g., submesic, very slowly but gradually cover the surface of the rock
between xeric and mesic; hygric, between mesic and with a thin layer of organic material that contains ni-
hydric). However, these three seral types serve as a trogen (often fixed by the algae in the lichens) and
492 CHAPTER 17 — Ecological Succession

Figure 17-8 Diagram of the successional

consequences of different levels of distur-
bance during harvesting of the penultimate
successional stage of three different types
of sere in coastal British Columbia. Arrow |
represents minimal disturbance, which permits
Moss layer
Litter layer
succession to proceed to climax, over the sub-
TRANS =s———s— Mineral soil sequent 100 years. Arrow 2 represents the de-
*————_ Parent material
Rock or
sc yeant Lichens and Shrubs (e.g., 4 Climax association of gree of disturbance necessary with a 100-year
parent material | Pioneer mosses| salal) 1 | hemlock and cedar frequency to maintain the penultimate succes-
Degree of successional retrogression produced by harvesting sion stage in perpetuity. Arrows 3 and 4 repre-
sent increasing levels of disturbance. The varia-
tion in the successional consequences of severe
disturbance between the three seres can readily
be seen. (A) In the xeric environment, succes-
Extent of successional development occurring over (e.g.) 100 years postharvesting sional changes that accompany disturbance in-
(a) HYPOTHETICAL XERARCH SUCCESSION clude loss of nutrients and of nutrient and
moisture storage capacity. Recovery from ex-
cessive disturbance is slow and cannot readily
be accelerated. Patterns of recovery are pre-
dictable. (B) In the mesic environment, overall
successional changes that accompany distur-
bance are modest and recovery is generally
rapid. Recovery can be accelerated fairly easily
when it is undesirably slow. Note that both the
* | " } degree and the pathway of successional recov-
a
= SSS
ZZ SEER RES Organic layer
ery are variable and difficult to predict. (C) In
Deep, moist,
IZA oo nutrient-rich the hydric environment, the major successional
soil
Bare soil | Herbs, Red alder or Douglas-fir ZZ
| association of
changes that accompany disturbance are in-
(e.g., after | grasses, Salmonberry | hemlock and cedar creasing wetness and loss of soluble nutrients.
scarification) | and pioneer Degree of successional retrogression produced by harvesting
| mosses Recovery tends to be slow but can be acceler-
|
ated by such techniques as drainage. As with
=,
c= the xeric environment, the pattern is very pre-
7 8
dictable. (After Kimmins, 1972. Copyright Cana-
dian Institute of Forestry. Used with permission.)
Extent of successional development occurring over (e.g.) 100 years postharvesting

(b) HYPOTHETICAL MESARCH SUCCESSION

—— Bed rock
Open | eek: Floating , gnum peat marsh , Cedar swamp Climax association of
water | plants sedge hemlock and cedar
peat Degree of successional retrogression produced by harvesting

Extent of successional development occurring over (e.g.) 100 years postharv esting

(c) HYPOTHETICAL HYDRARCH SUCCESSION

 The Three Major Types of Sere 493

other nutrients and provides a very limited but ecolog- Douglas-fir forest to a climatic climax cedar-hemlock
ically significant storage of water. community is very slow, generally requiring many
The organic layer produced by the lichens un- centuries for completion, but in the prolonged ab-
doubtedly gives an advantage to invading mosses, sence of disturbance by wind, fire, erosion, disease, in-
although there may be allelopathic mechanisms af- sects, or logging and of significant climatic change,
fording protection to the lichens against their ultimate the climax forest community will eventually develop.
fate of being overgrown by a carpet of mosses. The In the drier parts of coastal British Columbia, these
mosses grow faster than lichens and accelerate the late seral species may invade the site and persist as
process of humus accumulation, but depending on the small trees in the understory but never become a part
climate and the species involved, their growth in an of the main canopy. They die of heat and drought as
early xerosere is generally slow compared with that of they move from the sheltered microclimate of the un-
the higher plants that occupy the area in later seral derstory into the harsher microclimate of the main
stages. In the drier parts of coastal British Columbia, it canopy. This has implications for the silvicultural sys-
may take many decades or even centuries before the tem used in summer hot areas.
mosses develop sufficient nutrient capital, moisture
storage capacity, and physical support to permit per-
manent occupancy and growth of dryland shrubs such
Primary Mesarch Succession
as salal (Gaultheria shallon) and certain species of and the Mesosere
Vaccinium. In high rainfall areas, the process may re- ‘Typical mesarch succession (see Figures 17-2 and
quire only a few decades. 17-8B) occurs on deep (usually >1 m), medium-tex-
The rate of organic accumulation increases as the tured soil (which is often of moderate fertility) or soil
shrub seral stage develops. The aerial cover of the parent material. If the topography is sloping, then
shrubs greatly modifies the microclimate at the soil there must be sufficient upslope drainage area to com-
surface, and the area is invaded by species of moss and pensate for the downslope loss of water from the site,
lichen that are not found in the pioneer stage. After a and this type of sere frequently occurs midslope. It
variable period of the shrub stage, tree seeds that have also occurs on ridge tops if the soil is moderately thick
been reaching the site all along will successfully ger- and has good soil moisture retention characteristics,
minate, and seedlings of species that are tolerant of the and in flat areas where soil drainage is moderate but
xeric and oligotrophic conditions of the site (e.g., there is no upslope drainage area to provide additional
lodgepole pine, Douglas-fir) will survive. Growth of moisture. Primary succession is much less common on
trees at this stage is characteristically slow, and many mesic sites than on xeric sites because the probability
trees die when they reach a certain size because the of any disturbance removing all the products of the
site can no longer satisfy their increasing moisture and previous succession (organic matter, a well-developed
nutrient requirements (especially if there is an unusu- soil, soil fauna and flora, and an accumulation of plant
ally hot, dry summer), or they may be blown over. spores and seeds) is much lower for a mesic site than
However, their litterfall contributes to the buildup of for a xeric site. Consequently, most successions on
organic matter, and eventually the site reaches a con- mesic sites are secondary. Primary succession occurs if
dition that can support a continuous cover of slowly logging or fire is followed by extensive sheet erosion
growing closed forest. that removes everything above the C soil horizon.
The establishment of a closed forest is followed by Areas subject to mechanical land-clearing operations,
a long period in which there is a slow but continuing landslides, or deposition of alluvial materials would
buildup of the organic layer, which in the prolonged also undergo primary mesarch succession.
absence of fire can eventually reach depths of 1 m or The pioneer stage of this sere is frequently domi-
more if the climate is humid. As the layer builds up, nated by herbaceous plants, unless the availability of
the moisture storage capacity and the nutrient capital nitrogen is low (as in primary succession, in which
of the site steadily increase and may eventually permit case plants associated with symbiotic nitrogen-fixing
invasion of the site by shade-tolerant tree species such microbes or with very low nitrogen requirements may
as western redcedar and western hemlock. The dominate). In coastal British Columbia, for example,
process of conversion from a lodgepole pine and/or the herbaceous pioneer stage is frequently missing or
494 CHAPTER17 — Ecological Succession

poorly developed on sites that are very poor in nitro- interest in the hydrosere is confined largely to the
gen; such sites are rapidly colonized by dense stands of later, semiterrestrial and terrestrial stages.
red alder, which has nitrogen-fixing root nodules. If All lakes undergo a progressive process of filling
the site is somewhat more fertile, then invasion by red in. For large lakes, this is generally a process measured
alder may occur with or after a stage dominated by on a geological rather than a biological time scale, al-
herbs such as bracken fern (Pteridium aquilinum) and though even large lakes can fill with sediments fairly
shrubs such as salal, species of Vaccinium, elderberry rapidly. For example, Lake Mead, formed by Boulder
(Sambucus sp.), and possibly salmonberry (Rubus Dam on the Colorado River in the southwestern
spectabilis), although the last species is more typical of United States, is estimated to have a life of only 150
wetter and richer sites (hygric to subhydric). If the dis- years because of deposition of the enormous load of
turbance initiating the succession does not expose silt carried by the Colorado River (Sawyer, 1966). In-
mineral soil, then alder may fail to colonize the area, puts of sediments in drainage water, wind-carried
which may regenerate directly to a mid-seral conifer dust, and the accumulation of organic material from
stage or even to the climax species western hemlock. autotrophic and heterotrophic production in such
Douglas-fir generally requires some disturbance ofthe lakes all serve to reduce the depth of water. Initially,
climax or late seral forest floor if it is to colonize the organic production in the lakes is limited to food webs
area and compete effectively with hemlock. If the dis- based on phytoplankton. In oligotrophic (immature,
turbance includes fire, however, then Douglas-fir re- nutrient-poor) lakes, such production is very low, and
generation is often promoted. lake fill-in is largely a geological process.
If it is successful in colonizing the sites, then red As the lake matures and becomes more shallow,
alder, which is the fastest-growing tree on such sites in bottom-rooted aquatic plants and lakeshore vegetation
its early years, may occupy the site for 30 to 60 years, contribute increasingly to the accumulation of organic
gradually giving way to either Douglas-fir or western sediments. Where the land shelves gently into the lake
hemlock and western redcedar, depending on aspect, and where the lake is mesotrophic to eutrophic, marsh
seed source, and local climate. If a well-stocked Dou- and swamp plants such as sedges, rushes, and shrubs
glas-fir seral stage does occur, then it may be very pro- form a distinct zone around the margin of the lake. In
longed because the species is long lived. However, oligotrophic lakes (as shown in Figures 17-2 and
because Douglas-fir is shade intolerant on this site in 17-8C) the lake-edge environment may be dominated
the more humid low elevation areas of coastal British by Sphagnum mosses and shrubs tolerant of acidic bog
Columbia, regeneration beneath the Douglas-fir will conditions, the dead remains of which decompose
be climatic climax hemlock and cedar. These will slowly and produce an ever-thickening, floating layer
eventually replace and exclude the Douglas-fir unless or shelf of peaty material that may extend out many
the canopy is sufficiently opened by disturbance (e.g., meters over the surface of the lake. Eventually, the
windthrow) to increase light penetration to the point layer becomes so thick that its lower surface rests on
at which some Douglas-fir seedlings can survive. If the bottom of the lake, converting that portion of the
such disturbance does occur, then seral Douglas-fir area from an aquatic to a semiterrestrial ecosystem.
can exist more or less permanently as scattered indi- Once a continuous layer of sphagnum peat has
viduals or small groups in the shade-tolerant climax been established or the lake sediment level has reached
community. The succession can go directly from alder the water surface, further plant litter accumulation
to shade-tolerant cedar and hemlock. serves to raise the surface of the organic litter above the
water table. This produces a drier layer that is invaded
by a variety of shrubs, which, under acidic, oligotroph-
Primary Hydrarch Succession
ic conditions in coastal British Columbia, include
and the Hydrosere Labrador tea (Ledum groenlandicum), sweet gale (Myrica
Primary hydrarch succession (Figures 17-2 and gale), swamp laurel (Kalmia polifolia), and hardhack
17-8C) starts in newly formed lakes such as those that (Spirea douglasii). Continued litter production and
develop in the wake of a retreating glacier or when a evapotranspiration by this shrub stage raise and dry the
river is dammed by a landslide or through deliberate surface layers of the soil to the point at which trees
human activity. Aquatic succession is a major topic in such as lodgepole pine, western white pine, western
its own right but is beyond the scope of this book. Our redcedar, or western hemlock and larger shrubs such as
 Linear and Cyclical Succession: The Problem with the Concept of Climax 495

crabapple (Malus diversifolia), ninebark (Physiocarpus 17.8 Linear and Cyclical Succession:
capitatus), and high-bush cranberry (Viburnum edule)
can become established. Tree growth is slow initially,
The Problem with the Concept of Climax
but after a prolonged period of accumulation of leaves,
needles, branches, and logs, the surface of the organic Linear Versus Cyclical Succession
soil becomes sufficiently elevated above the water table The pattern of succession that has been described so
to permit moderately productive growth of all the far has been based on the implicit assumption that
conifers listed above, although the light-demanding there is a linear development from an early-seral
pines are frequently ousted by the shade-tolerant hem- postdisturbance condition to a final, self-perpetuat-
lock and cedar. The result of hydrarch succession is ing, and therefore “stable” climax that persists for a
concentric bands of vegetation radiating outward from very long time. This view of linear, or “directional,”
the water body, eventually developing into climatic cli- succession (Knapp, 1974) accepts that the climax
max western hemlock stands. The progression through community may eventually become senescent, lead-
this sequence is generally very slow. ing to its removal in a catastrophic manner by fire,
In comparing the descriptions of these three par- wind, insects, or disease. This initiates a secondary
ticular seres, there is one point that should have be- linear succession leading back to the climax. Howev-
come apparent. In environments characterized by er, this concept of succession assumes that the more
moisture extremes, a certain sequence of seral com- common pattern is one of a “permanently” self-per-
munities is more or less obligatory and predictable, petuating climax (i.e., over many generations), with
especially in the earlier seral stages. The shrub seral seedlings of the same species replacing individuals as
stage in the primary xerosere cannot occur before the they die of old age. The linear concept of succession
moss stage. The tree stage generally cannot occur does not allow for the situation in which the sequence
before the shrub stage, and the climatic climax stage of seral stages may eventually lead back to the “ini-
cannot occur until the subclimax forest has wrought tial” or somewhat similar stage without some cata-
very considerable changes in the moisture and nutri- strophic disturbance event.
ent status of the site. Similarly, the shrub seral stage A number of vegetation studies have reported that
in the hydrosere cannot occur until the combination in certain types of ecosystems, one can observe a
of allogenic and autogenic processes has ameliorated patchwork of communities that are successively re-
the moisture status of the substrate sufficiently, and placing each other in a cyclical sequence. This cycle
the closed forest seral stage cannot occur until after may involve only the final stages of the sere, or there
the shrub stage. In the mesosere, however, climax may be a cyclic repetition of the entire seral sequence.
tree species can participate in the pioneer com- The former case has been referred to as a regeneration
munity, especially if the microclimate is cool and complex or stand cycle; the latter has been called cyclic
humid (higher elevations, northerly aspects, and the succession (Watt, 1947, 1955).
windward side of coastal mountains). Almost any of In studies of the patchwork of microcommunities
the intermediate seral stages can be omitted or great- in a British heathland, Watt (1947, 1955) identified
ly truncated, although greatest productivity of later four phases (pioneer, building, mature, and degener-
seral communities generally occurs after the environ- ate, which leads to the reestablishment of the pioneer
mental modification (largely the accumulation of or- phase) and described these for a variety of plant com-
ganic matter and nitrogen) that accompanies earlier munities. For example, vigorous young heather
seral stages, especially if these have nitrogen-fixing (Calluna vulgaris) growing on an English moorland
species. Arguments about whether succession must ages, becomes senescent, and dies, leaving a patch of
follow a fixed pattern have generally failed to recog- bare organic soil that is rapidly invaded by young
nize the different nature of these three types of heather (Figure 17-9A). The height and number of
succession. They have also frequently failed to rec- shoots per plant and the depth of the mor humus be-
ognize that the need for environmental modification neath the heather exhibit a cyclical variation. Left
before a particular seral community can colonize an undisturbed (by fire) for many years, a heather moor
area applies far more in primary succession than in becomes a mosaic of small areas (a few square meters)
the much more common but more variable sec- of the four phases. At any one time, the majority of the
ondary succession. area is in the building phase.
496 CHAPTER 17 — Ecological Succession

& Pioneer phase ——*

-
Building phase ————» Mature phase ——————_ Degenerate phase
ne
Ne i *

ASE NA cee
SAISNZ
A PRE te Saas
SoS
f
hesSTSG ee
\
re aS ER
cmlZ ERENT 4
Time
Mean height of plants (cm) 2-5 29 48 a=

Mean no. shoots per plant Few 52 17 255

Thickness of mor humus (cm) _ Sparse 2S 3.1 2.8

% of area in phase 15 49 26 10
(a)

Mature phase —* Degenerate phase —> Hollow phase ——» ~~ Building phase

Senescent grass Vigorous grass

Lichens
= a /
a M] if
( ‘| rh HY)

Depth of soil > iy re i} WHOA

< Yj2 _(crustaceous) Grass seedlings » vy
above layer < WW
of stones | YU. . WH WW
(cm) 0 fy Yy PyUSM SETA
SS cs
SA 5ee.KKKERR
ecacecestes Sete Mineral soil X89LARA Le RIES
100 Layer of stones
Crass (erosion pavement)

80 Bare soil and stones

D
&60\>
Y

34
= 40

20

Figure 17-9 Two examples of the regeneration complex form of cyclical succession (after
Watt, 1947, 1955). (A) Heather growing on an English moorland. Senescence of the vegetation is
followed by a period of absence of the dominant plant and then by recolonization. (B) Regeneration
complex in a sheep’s fescue (Festuca ovina) grassland in England. (Copyright Blackwell Scientific Pub-
lications Ltd. Used with permission of the publisher.)

In the second example (Figure 17-9B), a tussock- building, mature, and degenerate. Grass seedlings in-
forming species of moorland grass, Festuca ovina, cre- vade areas of exposed mineral soil or erosion pave-
ates a sequence of microtopographic conditions that ment. The activity of ants and earthworms together
are associated with four vegetation phases: hollow, with the deposition of soil by wind and splash erosion
 Linear and Cyclical Succession: The Problem with the Concept of Climax 497

from adjacent degenerating or hollow phases builds up frequency of approximately one wave every 60 years
the mineral soil around these seedlings, whose roots (Figure 17-10).
stabilize the accumulating soil and thereby build a Examination of the climax in many types of ecosys-
“tussock.” As the grass ages, it becomes senescent and tem has revealed that the regeneration complex—con-
is replaced by various species of lichens and sisting of a patchwork of communities or community
bryophytes. These fail to stabilize the soil, which is conditions, each patch depending on its neighbors and
gradually eroded back down to the erosion pavement, developing under conditions partly imposed by
at which time grass seedlings become reestablished them—is common in many types of environment. This
and initiate a new tussock-building phase. As with the observation suggests that the climax should be consid-
heather, the four phases of the tussock grass commu- ered as a “steady state” of repeated short-term cyclic
nity form a mosaic of small patches. variations in the composition and/or structure of small
A similar regeneration cycle (wave regeneration; patches of vegetation around a mean community con-
Sprugel, 1975) has been documented in the Adiron- dition, rather than a stable, unvarying condition in
dack Mountains of the northeastern United States. which there is an individual-by-individual replace-
The shade-tolerant, climax-dominant balsam fir (Abies ment. Although the latter does occur in some situa-
balsamea) becomes susceptible to a variety of environ- tions, either the frequency and intensity of natural
mental stresses, notably wind, as it approaches maturi- disturbance (see the references in Vitousek and Rein-
ty at approximately 60 years of age. The death of ers, 1975) or the normal pattern of senescence and
individual trees exposes the crowns of downwind death more commonly produces patch replacement:
trees, which are damaged and then killed, exposing the the regeneration complex. Recognition of the impor-
next group of trees. Prolific and regular production of tance of such small-scale disturbance has resulted in a
seed crops by the fir ensures prompt regeneration of growing interest in gap dynamics in forest ecosystems
this species below the dead and dying trees, and the (e.g., Doyle, 1981; Pickett and White, 1985; Shugart et
net result is a series of crescent-shaped bands of dead al., 1981; Shugart and West, 1980). There is now a very
and dying trees moving slowly across the forest with a extensive literature on forest gaps.

Direction of prevailing wind and

wave movement

Figure 17-10 Wave regeneration in a climax balsam fir forest on Whiteface Mountain in the
Adirondack Mountains, New York State (after Sprugel, 1976). Wavelength varied from 50 to 150
m. Tree age at the wave edge was as great as 79 years (based on the study of 11 waves). The diagram
shows the variation in stand structure and tree age across one wave (+1 standard deviation).
(Copyright Blackwell Scientific Publications Ltd. Used with permission of the publisher and the author.)
498 CHAPTER17 — Ecological Succession

Cyclical succession involving the entire sere occurs These are able to colonize the site because of the de-
over a much longer time scale than the regeneration cline in overstory leaf area that is a characteristic fea-
ture of stand development and the formation of more
mst
Sap
4.
t
complex or stand cycle. The best example is seen in
the development of forests and bogs on northern permanent canopy gaps as a result of the death of in-
slopes in Alaska (Heilman, 1966, 1968; Neiland, 1971; creasing large trees. The fourth phase—old growth—
Viereck, 1970). This has already been partially de- develops as the canopy dominants approach
scribed (Figure 17-6) but was portrayed as a linear physiological old age, lose vigor, suffer increasing dam-
succession. However, if the area is fairly flat and the age and disease, and experience accelerated mortality.
succession is considered over a much longer period, Tree seedlings of the subsequent seral stage species
then it can develop into an example of cyclical succes- become established as part of the understory in the
sion. As the accumulation of sphagnum peat contin- third and fourth phases unless they are inhibited by
ues, it may produce a lens-shaped bog whose center is the herbs and shrubs. On many sites in northern; sub-
higher than its edges (called an ombrotrophic bog, in alpine; or cool, humid, temperate forests, there is a
which the only source of water and nutrients for plants tendency for an increasing component of ericaceous
is precipitation). This may eventually lead to the for- shrubs in the understory as succession proceeds, and,
mation of streams draining the bog. As these streams as noted above, these can have a very negative effect
erode their way down into the peat, the area may be- on tree regeneration. Thus, there may be a progres-
come drier and warmer, the permafrost level recedes, sive reduction in the density of tree regeneration in
the peat starts to decompose more rapidly, and the the understory reinitiation phase as succession pro-
area may be invaded by birch and alder, reinitiating ceeds. Severe disturbance by fire, windthrow, or ero-
the sequence already described. The process of self- sion removes this ericaceous community or disrupts it
drainage is a very slow one, and fire usually intercedes sufficiently to permit trees to reestablish. In contrast,
before the drainage is complete, burning off the sur- less severe windthrow, some types of fire, natural mor-
face of the accumulated peat, lowering the permafrost, tality, or winter selection logging that creates only
and reinitiating the succession. small canopy openings and no disturbance to soil or
shrub layer can result in ericaceous shrubs replacing
trees. In dry areas of the northern boreal, this can lead
Is Closed Forest Always the Climax
to the conversion of closed forest to open shrub/lichen
Community of Forested Landscapes? woodland. In humid subalpine forests, it can result in
Many of the published descriptions of successional the break-up of closed forest into tree islands sur-
pathways are similar to those in Figures 17-2 and 17-8; rounded by ericaceous vegetation. In cool, humid
succession proceeds from nonforest communities to a coastal areas and humid boreal sites, it can lead to
forested climax stage that is self-replacing over several open bog forest or sphagnum bog or to ericaceous
or many generations of trees. The implicit assumption shrub woodland. It is disturbance of an appropriate in-
is that forest is the end stage of succession. In many tensity, frequency, and scale that maintains the land-
cool, humid temperate forests and cold, northern or scape in a closed forest condition in these areas.
subalpine forests, this seems not to be the case.
Each forest seral stage goes through a series of de-
velopmental phases (Oliver, 1981; Oliver and Larson, Arguments for and Against the
1990): (1) stand initiation, (2) stem exclusion (under- Concept of Climax
story suppression), (3) understory reinitiation, and (4)
old growth. These correspond to the regeneration Many ecologists have concluded that the concept of
phase; the closed, dark, “nudum,” phase that generally climax has outlived its usefulness. Others maintain
lacks understory plants and during which there is ac- that the concept continues to have value. Its contribu-
tive stand self-thinning (mainly as a result of light tions include
competition but also wind and snow damage); and the
reestablishment of understory phase in which herbs, 1. An expression of a real and important difference
shrubs, and mosses that are characteristic of the site in stability (in the sense of persistence) between
reestablish, together with seedlings of more shade- successional and self-maintaining (i.e., climax)
tolerant tree species of the subsequent seral stage. communities.
 Linear and Cyclical Succession: The Problem with the Concept of Climax 499

2. An indication of the direction and degree of vege- generations of dominant plant species—relative to
tation development in a region. other seral stages). As we shall see, there are problems
3. Provision of a basis for comparative studies of the with this definition because the early seral stages of
composition, structure, function, and management both primary and secondary seres may outlast the cli-
of successional communities. By selecting areas max, and seral communities may be self-replacing for
with a common climax, one can make more mean- many generations before they are in turn replaced.
ingful comparisons between seral stages. However, such early stages are subject to directional
change (albeit slow), whereas in the absence of
4. A means of identifying particular types of physical
catastrophic disturbance, the climax may exhibit only
environment and of defining local macroclimates
minor fluctuations around an average composition
and variations in soil conditions.
and structure over long periods of time (centuries), as
Undoubtedly, the concept of climax is more useful long as there is no directional change in regional cli-
in some types of environment than in others and for mate. By examining a series of areas that vary in age
some branches of ecology than others. In physically since disturbance (a chronosequence), it is possible to
extreme environments, seres may be telescoped into construct the probable seral sequence (or range of al-
two or even a single stage in which pioneer and climax ternative sequences) for any region, and an examina-
organisms are one and the same species. In such envi- tion of the population age-class and/or size-class
ronments, the concept is less useful than where suc- structure of the dominant vegetation will confirm a
cession involves a sequence of seral stages leading to a climax or seral interpretation. If the dominant species
distinct climax. exhibit a reverse J-shaped age or size frequency curve,
How does one recognize that a particular commu- then one may perhaps conclude that the area is climax,
nity is or is not climax? The definition of climax is a whereas if the curve is bell-shaped, then the species is
self-replacing seral stage that is relatively stable (in the probably not replacing itself and may be seral (Figure
sense that it persists for a long time—several to many 17-11; see also Figure 14-17).

All tree species combined

All trees > 10 cm dbh

frequency
Relative

rN
A S
ee
gee
ae Beech NS
Spruce... a

80 100 120 0 30 60 90

Diameter at Breast Height (em) ——————————>

(a) (b) (c)

Figure 17-11 Size frequency curves for the dominant trees in three forests. (A) Temperate
seral forest. A successional forest in Arizona in which aspen is failing to reproduce itself and is being
replaced by white fir (Abies concolor). (B) Temperate climax forest. An apparently climax forest in the
Carpathian Mountains. The trees do not exhibit a perfect reverse-J shape, probably because regen-
eration and mortality of larger trees occur sporadically rather than continuously. (C) A climax tropi-
cal rain forest. (After Cousins, 1974, and Whittaker, 1975.) See also Figure 13-17.
500 CHAPTER17 = Ecological Succession

Before leaving the discussion of climax it is worth climax forest community, then it may be replaced
reiterating that a climax community can occur for one by shrub communities that in the long-term ab-
or more of the following reasons: sence of disturbance form the true climax vegeta-
tion. Very old climatic climax forests may simply
1. There are no rapid allogenic changes in the envi- become senescent/decadent, which sets the stage
ronment, and the living community is not capable for a natural disturbance event, such as fire, insects,
of inducing any further autogenic change. Conse- or wind, that initiates a new secondary succession.
quently, the community becomes self-replacing
because it represents the group of organisms best
adapted to compete for and utilize resources on
Is There Such a Thing as a Climax Forest?
that site under those particular conditions. So, is there such a thing as a climax forest? As in most
2. Although the climax community may be capable of cases in ecology, the answer is “it depends”. It depends
inducing further autogenic change in the environ- on the following:
ment that would eventually lead to the replace-
1. Your definition of climax. If a climax is a stable,
ment of that community by another, the natural
self-replacing community over several genera-
incidence of fire, wind, disease, or the action of
tions, then the answer is clearly “yes.” Early seral
herbivores prevents these changes from occurring.
pine or eucalypt fire climaxes are self-replacing
For example, the invasion of pine forests by later
over thousands of years in fire-dominated environ-
successional species may depend on amelioration
ments. Shrub serclimaxes can be stable for many
of microclimate and soil nutrient and moisture
centuries or millennia, as can edaphic climaxes.
conditions by the pine trees. Frequent ground fires
Some climatic climax forests are self-replacing
may prevent the accumulation of the forest floor
over many generations, but some late seral forest
material needed to produce amelioration of soil
stages may last only one or two generations before
conditions and thereby maintain an open stand
they become ecologically senescent (e.g., because
structure. This in turn maintains the site in a con-
of the build up of diseases and insect pathogens) or
dition that permits regeneration of the resident
functionally decadent (i.e., in a state of functional
species and is unfavorable for regeneration of
decline) because of the stagnation of nutrient cy-
species of the subsequent seral stage.
cling and decline in NPP and prone to disturbance
3. The invading species of a subsequent seral stage that initiates a new succession. Alternatively, such
are constantly removed by fire, herbivores, disease, forests may be replaced by shrub or bryophyte
or climatic factors that operate to prevent colo- communities (e.g., heath or muskeg, respectively),
nization. This is somewhat similar to the preced- which may become the climax. Our perception
ing reason, but it differs in that the environment is that old forests are self-replacing climax forests
capable of supporting the next seral stage but colo- may in many of the world’s forests have more to do
nization is prevented. For example, many grass- with the short human life span relative to that of
lands that could sustain a climatic climax forest trees than with the existence of a truly stable, self-
remain as grasslands because of the action of fire replacing ecosystem condition.
or herbivores or both. Many of the Eucalyptus
forests of Australia are prevented from being re- 2. The frequency and severity of natural disturbance.
placed by the local climatic climax species by a fre- This is the topic of the next chapter.
quency of fire approximately once in 350 years A final thought about climax forests. There is con-
(Beadle, 1966). fusion among some people about the relationship be-
4. Allogenic or autogenic changes in the environment tween “old growth” forests and climax forests. There
may continue, albeit at a reduced rate, but there are is apparently the feeling that for a forest to be climax it
no organisms in the region that are better adapted must be old and uneven aged. This is incorrect. A
to occupy the site or capable of competing with the young, even-aged forest composed of the climax tree
climax organisms. However, if continuing allo- and understory species for that site, with a climax soil
genic changes in very old climax forests eventually condition and microclimate, is every bit as much a cli-
cause soil conditions that are unfavorable to the max forest as an “old growth” forest on the same site.
 “Vital Attributes” Model of Succession 501

The concept of the climax forest is a self-replacing plant population ecology with so-called J/ife history
community that persists over several generations. Im- strategies:
plicit in the definition is that mature or old forest
communities of the climax seral stage will periodically 1. Method of persistence. The method of invasion and
be disturbed and will be replaced by younger individu- colonization, or of persistence of propagules after
als of the same species. This replacement may be on a disturbance (four classes).
tree-by-tree basis (“gap dynamics”), resulting in an a. Persistence by the arrival of seeds from remote
uneven-aged, multicanopy-level forest, or it may be by sources (D species: dispersed seed or seed rain).
large scale disturbance such as windthrow that facili- b. Persistence by seeds with long viability stored
tates the regeneration of an even-aged stand of the cli- in the soil. This pool can persist through sev-
max species and maintains a climax forest floor and eral disturbances without replenishment, be-
soil condition. cause not all the seeds germinate after the first
There is similar confusion over the term “old disturbance (S species: stored seed or seed bank).
growth” forest. This is frequently assumed to be an- c. Persistence of seed surviving on the site in pro-
cient climax forest. In fact “old growth” is a phase of tective cones or fruits held in the vegetation
stand structural development that occurs in every seral canopy; seeds available only if sexually mature
stage. Old growth can and does occur in early, middle, plants are available before the disturbance (C
and late seral stages. The later is a relatively perma- species: canopy seed).
nent and stable ecosystem condition. By comparison, d. Persistence by means of a protected vegetative
old growth in earlier seral stages is a relatively tempo- organ that resprouts promptly after distur-
rary condition (Kimmins, 2003). bance (Vspecies: vegetative recovery).
2. Conditions for establishment. The conditions re-
quired (tolerated) for establishment and growth to
17.9 “Vital Attributes” maturity (three classes).
Model of Succession a. Species able to establish at any time, in the
presence of mature individuals of the same or
Although the successional models discussed in the sec- different species; able to tolerate competition
tion Review of Recent Theories and Models of Suc- (T species: tolerant).
cession all represent some degree of improvement b. Species able to establish only immediately after
over the original Clementsian model, they have been disturbance when competition is usually reduced;
criticized (Cattelino et al., 1979; Noble and Slatyer, intolerant of competition (J species: intolerant).
1977) because of their inapplicability to highly dis- c. Species able to establish only after the envi-
turbed (e.g., fire-dominated) ecosystems, their failure ronment has been modified by the presence of
to quantify the effects of disturbances of different in- mature individuals of the same or other species
tensity and frequency, and because they do not ex- (R species).
plicitly incorporate the adaptations of species that 3. Critical life history events. The time taken to reach
determine the model’s response to different types of critical stages in the life history (e.g., age of repro-
disturbance. To address the high diversity of species duction, longevity) (four classes).
adaptations and variations in the intensity, type, and
frequency of disturbance, these authors developed a a. Replenishment of the supply of propagules suf-
multiple-pathway model of succession. This model in- ficient to replace a species after another distur-
corporates classical concepts and those of Egler (1954) bance (event p).
and of Drury and Nisbet (1973). b. Attainment of reproductive maturity; production
For predicting the direction and patterns of suc- of sufficient propagules to enable a species to
cession in a disturbed ecosystem, the multiple-path- persist through another disturbance (event 77).
way model requires three types of knowledge, referred c. Senescence and loss ofa species from the com-
to as “vital attributes,” for each of the species that ex- munity (event /).
erts a dominant role in the successional process. ‘This d. Loss of propagules and elimination of a
is a model of succession that combines elements of species (event ¢).
502. CHAPTER17 = Ecological Succession

Using this information, diagrams that predict and decline somewhat as transition to the next stage occurs.
summarize secondary succession can be prepared. For Various combinations of incomplete occupancy of the
example, patterns of succession in two different forest site by the invading community, competition between
types in northwestern Montana are shown in Figure the invading community and the remnants of the pre-
17-12; a further discussion of this successional model vious community, and occupancy of part of the site by
can be found in Noble and Slatyer (1980a). senescent individuals of the previous community all
Succession is such an important but complex and will serve to prevent maximum primary productivity at
variable aspect of ecology that there will undoubtedly the beginning of a seral stage or during the transition
be continued development of successional models. to a new seral stage. As the seral community fully occu-
This is discussed further in Chapter 21. pies the site and develops toward maturity, net produc-
tivity will peak again, only to decline as the ratio of
photosynthesizing to respiring biomass decreases, as
17.10 Changes in Ecosystem competition within the community increases, and as
Function During Succession the individuals in the community become senescent.
Net productivity in a mesosere in a moderate cli-
As an area proceeds from early to late seral stages of a mate will rise rapidly from the pioneer community
linear succession or from one point to another of a and peak during one of the intermediate seral stages
cyclic succession, there is a marked change in several (Figure 17-13B). This may occur in the shrub stage if
characteristics of the community. The life form and it is succeeded directly by conifers, in an early hard-
longevity of the dominant plants, the structure of the wood tree stage (e.g., red alder in coastal British Co-
community and its productivity, and the ratio of green lumbia), or in an early conifer tree stage (e.g., pine or
to woody biomass all undergo change. In this section, Douglas-fir). Net productivity may decline somewhat
we examine the changes in ecosystem function that ac- as the climax community develops, because later suc-
company changes in ecosystem structure: ecological cessional species have been selected during their evo-
energetics, nutrient cycling, and ecosystem stability. lution more for survival than for rapid growth.
Aspects of this were presented in Chapters 4 and 5 However, it is more difficult to make generalizations
(e.g., Figure 5-17). about mesoseres than about hydroseres and xeroseres
because, as we have seen, mesoseres tend to follow a
less consistent pattern of succession. It may well be
Ecological Energetics that in cases in which an area reforests directly with
It is difficult to generalize about the variation in en- climax species after disturbance, productivity in the
ergy flow at different stages of succession because this young climax community would be just as high as that
will differ so much among xeric, mesic, and hydric en- of earlier seral communities had they been occupying
vironments and between primary and secondary suc- the site. Also, some shade-tolerant, mid- and late seral
cessions. No comprehensive, comparative analysis of tree species can be very productive on moist and fer-
these patterns has been made (largely for lack of ade- tile sites after a period of slow early growth.
quate data); therefore, the following model must be Living biomass may approach its maximum value
considered hypothetical. There is growing interest in in the middle stages of the mesosere, with only modest
such comparisons; among other reasons, there is con- increases after either the deciduous or the pioneer
cern about carbon storage in forests. The next few conifer stage. However, total biomass accumulation
decades of research should provide a much improved may continue until the next major ecosystem distur-
understanding of the topic. Because of the long time bance in cool humid climates because tree stems may
scales involved and the complexity of the topic, such be incorporated into the forest floor in a largely unde-
research will probably depend heavily on computer composed or slowly decomposing condition as they
modeling to extrapolate from our existing experience die and are replaced.
and knowledge. The pattern of production for xeroseres and hy-
At the beginning of each seral stage, net productiv- droseres is considerably different. In a xerosere, gross
ity (excess of photosynthesis over respiration) will gen- production, net production, and biomass are all very
erally increase until perhaps the middle or later part of small during the pioneer lichen stage (Figure 17-134).
the stage, then level off (cf. Figure 4-10), and finally Production and biomass increase slightly during the
 Stand age (yrs.)
Specific
Species type 0 10 20 130 150 300
Aspen VI
Lodgepole pine Ell ress Peete
Larch DI se hil
:
Spruce DT : ay iy A
Douglas-fir DT Pp my

—_
DI \arch dominant
DT spruce-Douglas-fir
(aspen and lodgepole pine extinct) oe ee araie eenrer st 300

rage
eR ae
NY
‘ ) 20-130
a
VI aspen with 130 DT spruce-
CI lodgepole pine Douglas-fir with 150
CI lodgepole pine
(aspen extinct)
<20

VI aspen 130 DT spruce-Douglas-fir

: “a
(lodgepole pine extinct) (aspen and lodgepole pine extinct)

Species
Species type 0 20 25 150 250 ~

Lodgepole pine Cl 2 —— le
Western hemlock JOME Pp m le
Western larch DI ee =

x ys \ }a \
CI lodgepole 25 CI lodgepole pine Pee LESSOR DT hemlock dominant kn oe re DT hemlock
pine dominant DI \arch
Xs DT hemlock and
~ DI \arch

aod DI \arch dominant and

“~~ DT hemlock

Figure 17-12 Life history characteristics and species replacement sequences in two commu-
nity types in northwestern Montana. (A) Aspen community type. (B) Hemlock community type
(after Cattelino et al., 1979). Numbers on solid lines indicate the time (years) for transition in the
absence of disturbance. Numbers on dashed lines indicate frequency of disturbance by fire.
(Copyright Springer-Verlag New York Inc. Used with permission ofthe publisher and the authors.)
504 CHAPTER 17 Ecological Succession

Xerosere moss stage and again in the shrub stage. Increases con-
tinue throughout the seral conifer stage, reaching a
peak only when the area has converged to the climatic
climax community. In the absence of complete con-
vergence, biomass and production will peak in the ter-
minal seral stage.
Respiration
In a hydrosere, productivity increases from the
Lichen stage Moss stage Shrub stage Seral conifer Climax conife oligotrophic to the eutrophic lake stage, but very little
stage stage living biomass accumulates. Productivity may drop
(a) somewhat as one passes through the transition from
aquatic to wet terrestrial stages (bogs), although marsh
| Mesosere
leah aa Gross production —_
ea aan er
ecosystems are sometimes associated with very high
Ga
as ae
—_— productivity (Chapter 4). Productivity may then either
y er...-.
-----
SS ret a
aetnaccnwnesos~—

a drop or increase, according to the productivity of the

/ wa Biomass
yo -
semiterrestrial stage (depending largely on nutrient
&
Respiration status), rising further as succession approaches the cli-
matic climax (Figure 17—13C).
It has been suggested that the ratio of total ecosys-
=p i a SRE WOSLQLLILLLLLILLLLLLLLLLL
LL tem production to total ecosystem respiration (the
Herb Shrub Deciduous Seral conifer Climax conifer stage
Stage Stage tree stage stage
P/R ratio) is a good index of the degree of succes-
(b)
sional maturity. Where the ratio approaches | (no net
biomass accumulation), the ecosystem is said to be ap-
proaching a climax condition, whereas a high value in-
Marsh Gross production
@ ee dicates an early seral condition. Less than 1 should
|
Alternative stages
‘7
i/ 7
Leo
y
> indicate a senescent climax condition (Whittaker and
\ / Bong
|
|
:
Eutrophic stage y 1‘| -y
iomass Woodwell, 1968). Although this may be a useful ap-
Mesotrophic stage \ lf proach to an overall evaluation of ecosystem maturity
Respiration
|
Oligotrophic stage
a ae S. in some types of ecosystem, it is to be expected that
aoe \E there will be variation in P/R ratio within any single
Bae =
Aquatic stages
= aoe ———
Marsh or Seral shurb
a LLILLTLLT
LLL LZ
Climax conifer
stage as the seral community progresses from juve-
bog stage and conifer stage nility to maturity to senescence (cf. Figure 14-10).
stage Also, we do not have enough information from stands
LIne is of varying successional conditions to be completely
(c)
confident that the P/R ratio does vary as suggested. An
Figure 17-13 WHypothetical model of the overall varia- additional problem lies in the implicit assumption that
tion in autotrophic production and biomass accumula- respiration equals gross production at climax. Experi-
tion during primary succession in three types of envi- ence in forestry suggests that biomass can continue to
ronment. Time scales for the three graphs differ, and none accumulate for very prolonged periods (hundreds and
of the scales is linear. The mesosere is much shorter than perhaps even thousands of years) in climax communi-
the other two seres. No actual time scales are given because ties in humid areas where fire is infrequent. In the Pa-
the duration of any stage can vary greatly. The curves are cific Northwest, ratios may exceed 1 in 1,000-year-old
smoothed to show overall trends. In reality, each stage stands, and yield is sustained in 300-year-old stands in
probably has an initial increase in biomass and production Utah (MacMahon, 1980). Only in overmature, senes-
followed by a slight decline late in the stage. Respiration is
cent climax communities will net biomass accumula-
shown to exceed gross production late in the climax stage,
tion approach 0 or become negative (cf. Table 4-10).
resulting in a biomass decline. In some forest ecosystems,
biomass may continue to accumulate in the climax commu-
Even where biomass accumulation ceases, ecosys-
nity for a very long time. tem production may still be relatively high. Many
forests and some shrub and herb communities that
grow in humid areas on nutrient-poor soils produce a
lot of organic acids as their litter decomposes. Under
 Changes in Ecosystem Function During Succession 505

the acidic conditions found in the soils of these ecosys- mineral nutrient input will be higher during succes-
tems, these acids are leached from the soil as dissolved sion than before it starts (cf. lithoponics). The rate of
organic matter (DOC). The streams that drain such nitrogen input will also vary through succession as the
areas are characteristically stained brown or even activity of nitrogen-fixing organisms varies. This is
black by the high levels of DOC. The Rio Negro still an incompletely understood topic, but nitrogen
(Black River), which drains part of the northwestern inputs will generally peak during seral stages domi-
Amazon basin, streams draining heather moors in nated by plants such as alder, which have a symbiotic
Scotland and buttongrass communities in southwest- relationship with nitrogen-fixing microorganisms
ern ‘Tasmania, and many streams draining bogs and (Chatarpaul and Carlisle, 1983). However, nonsymbi-
climax forest in the humid coastal areas of western otic nitrogen fixation occurs in decomposing litter and
Canada are examples of such organic matter exports logs, and this may be a significant input to ecosystems
that prevent continued biomass accumulation despite where these materials are abundant.
continued production. Clearly, one cannot always During periods of high biomass accumulation,
evaluate ecosystem productivity solely on the basis of output of essential nutrients from the ecosystem will
organic matter accumulations. This has significant im- be less than inputs because of their incorporation in
plications for the preparation of carbon budgets for the accumulating biomass. Table 17-3 shows the aver-
forests in humid areas. age growing season concentrations of six ions in
streams that drain nine late-successional ecosystems
and five mid-seral ecosystems that resulted from log-
Biogeochemistry
ging. The lower values for the latter reflect the rapid
Not only does the magnitude of nutrient turnover biomass accumulation in seral forests. As net biomass
vary during succession but so also does the relative im- accumulation decreases, the rate of nutrient accumu-
portance of the three pathways of nutrient movement: lation also slows until outputs once again equal inputs
the geochemical cycle, the biogeochemical cycle, and at the climax stage, as shown in Figure 17-14 for a
the biochemical cycle. Some aspects of the variation in mesosere. The input/output balance for xeroseres and
these three cycles within a stand (seral stage) cycle hydroseres will also reflect biomass accumulation, and
were shown in Figure 5-17. one could say that balance in the geocycle is largely a

Geochemical Cycle. At the start of the pioneer Table 17-3 Average Growing Season (June 1-Septem-
stage of primary succession, the addition of nutrients ber 30, 1973 and 1974) Concentrations of Five Ions* in
to the ecosystem in dust, precipitation, seepage water, Streamwater Draining Five Mid-Successional and Nine
biotic imports, biological fixation of gases, and the Late-Successional Ecosystems, New Hampshire
weathering of rock and soil minerals is most probably Streamwater
balanced by outputs in biotic exports, water and wind Concentrations,” eq L“'
erosion, and soil leachates. The extent of leaching loss
Late- Mid-
is determined by the cation and anion exchange capac- Successional Successional Ratio of
ities of and the production of anions in the mineral lon Watersheds Watersheds Concentrations
soil (Cole et al., 1975; cf. Figure 11-6).
As succession proceeds, the unconsolidated layers NO, 53 (5) 8 (1.3) 6.6
of surficial deposits (where they exist) are occupied by K* 13 (1) 7 (0.5) 1.8
plant roots and their mycorrhizal associates, and a soil Mg?* 40 (4.9) 24 (1.6) 157
organic layer is formed, which greatly increases the
Ca2- 56 (4.5) 36 (2.5) 1.6
ability of the abiotic environment to conserve nutri-
Cl” 15 (0.3) 13 (0.3) ie
ents. Nutrient output from the system is also substan-
tially reduced by the uptake of nutrients by plants and Na* 29 (2.6) 28 (0.9) 1.0
their incorporation in plant, animal, and microbial
Source: Vitousek and Reiners, 1975. Copyright the American Insti-
biomass. Weathering of primary minerals and rocks is tute of Biological Sciences. Used with permission.
accelerated by the organic acids produced by decom- aNote the relative absence of any effect of logging on sodium and
posing organic matter, by roots (Voigt, 1968), and by chloride ions: nonnutrient elements.
soil fungi (Silverman and Munoz, 1970), so the rate of Standard error of the means is shown in parentheses.
506 CHAPTER 17 Ecological Succession

87,
Boe
he

pbsae —ie= J
| Disturbance
Es
3.2) Bose
% 5 as

Mid
Mature
seral seral
forest
stages stages

H|:
jl
|:
~ . |
a. | Brief period of high loss
Z ir : iY at start of secondary
O's lj succession following
a. |; disturbance
& Hu
Poe saa Ij
S :
Nutrient input = nutrient output
% ee a
See yl, /
ee a !
2E=%
357. 3 =3 Essential
Mmaot = nutrient

(Output
<

C——— >

Figure 17-14 Model of the response of an ecosystem to disturbance. (A) Variation in biomass
accumulation. (B) Balance between nutrient inputs and outputs. In a mature forest before distur-
bance, nutrient outputs will generally equal or exceed inputs. As succession starts, there will be a
brief period of high nutrient loss, but as biomass accumulates, outputs will fall below inputs for es-
sential nutrients; nonnutrient elements remain unaffected (after Vitousek and Reiners, 1975). This
example is probably valid for mesoseres, but xeroseres and hydroseres may exhibit somewhat differ-
ent patterns. (Copyright American Institute of Biological Science. Used with permission of the publisher
and the authors.)

function of patterns of primary production and bio- Biogeochemical Cycle. This cycle varies according
mass accumulation (Figure 17-13). Disturbances such to biomass turnover (litterfall) and leaching losses
as fire or logging, which temporarily devegetate or re- from plants. The high turnover of relatively nutrient-
duce live vegetation biomass in an area and speed de- rich biomass in pioneer, shrub, and seral hardwood
composition, result in a sudden increase in output stages can exceed the turnover in late successional
from the system as organic matter is lost. This net loss conifer stages, so in mesoseres, the biogeochemical
is temporary and is reversed to become a gain as soon cycle can peak in an early seral stage. In xeroseres and
as biomass accumulation begins. hydroseres, it is more likely that nutrient turnover will
Our knowledge of the variation in the balance be- peak in the immature climax or late subclimax stages
tween geochemical inputs and outputs during succes- because the organic production of these stages is
sion is far from complete, and there is need for a lot higher than that of earlier stages. Biogeochemical cy-
more research in this field (Gorham et al., 1979). This cling in the mature climax stage is probably largely
is an important topic. Humans are influencing the confined to direct nutrient cycling (tree > forest floor
geochemical balance of the earth in as yet incom- ~ tree), and because in many types of forest decompo-
pletely understood ways (e.g., acid rain), which may sition tends to slow at the climax stage because of
have significant effects on ecosystem function. changes in the organic and inorganic chemistry of the
 Relationship Among Ecosystem Stability, Biodiversity, and Successional Stage 507

litter, nutrient turnover is probably reduced. The ap- that the diversity of the biota increases during succes-
parent adaptation of “hoarding” nutrients by promot- sion and that there is a concomitant increase in stabil-
ing direct nutrient cycling and_ inhibiting ity, but it is now recognized that there are probably no
mineralization probably helps the climax community simple relationships between diversity and stability
to perpetuate itself and to compete with other species (see reviews by Orians, 1975, and others in Attiwill,
even if it does reduce rates of nutrient turnover and 1994a; Botkin, 1980; Grubb, 1977; Simberlof, 1996;
productivity. van der Maarel, 1993; van Dobben and Lowe-
McConnell, 1975; and Veblen, 1992).
Biochemical or Internal Cycle. Internal nutrient The concept of stability usually refers to the ten-
cycling is an important conservation mechanism ex- dency of a system to remain in its present condition or
hibited by many, if not all, plants. Not enough is return to that condition after a disturbance. The term
known yet about the variation in internal cycling dur- has been used with a number of connotations (Orians,
ing different stages of succession to present a clear pic- 1975), and therein lies part of the difficulty. The fol-
ture, but accompanying the hypothesized trend lowing meanings of the term “stability” have been
toward “nutrient hoarding” within the ecosystem as used (Figure 17-15).
succession proceeds, one might expect a trend toward
a greater degree of internal cycling. By removing a 1. Constancy. The lack of change in some
large proportion of the nutrients from its old foliage parameter(s) of the ecosystem such as the number
before abscission, a plant can “hoard” nutrients effi- of species, the life form of the community, or phys-
ciently, thereby rendering them less available to com- ical features of the environment.
petitors. One might expect, therefore, that internal 2. Persistence. ‘The length of time over which the
cycling will become more efficient as succession pro- ecosystem is constant or maintains a particular
ceeds. Additional studies are necessary to test the va- condition within specified bounds.
lidity of this hypothesis, and it is certainly not 3. Inertia. The ability of the ecosystem to remain
universally true. Eucalyptus species, which are often constant or to persist in the face of disturbing fac-
subclimax, are outstandingly good at internal cycling tors such as wind, fire, disease, herbivore out-
of phosphorus (Attiwill, 1980; Ferreira et al., 1984). break, and so on.
Early seral pine and larch species also have efficient
4. Resilience. Vhis term has been used as a synonym
internal recycling. The low nutrient availability in the
for elastic stability.
soil early in some successions and the high growth rate
potential of many pioneer plants may result in greater 5. Elasticity. Vhe speed with which the ecosystem re-
internal cycling early than late in succession. Also, turns to its original condition after disturbance.
early seral species may be more adapted to utilizing 6. Amplitude. The extent to which an ecosystem
nutrients in soil solution (e.g., during the assart can be changed and still return rapidly to its
period), whereas in later seral stages, species may be original condition.
more adapted to nutrient uptake form decomposing 7. Cyclic stability. The property of an ecosystem to
litter. Where this is the case, maintenance of an ade- change through a sequence of conditions that
quate supply of litter nutrients and litter of a quality bring it back to the original condition (i.e., the re-
that is favorable for decomposer organisms may have generation complex of a heather moor; even the
resulted in selection for reduced internal cycling. Alaskan forest—bog cyclic succession could be said
to have a form of stability if considered over a long
enough period of time).
17.11 Relationship Among 8. Trajectory stability. Vhe tendency of ecosystems to re-
Ecosystem Stability, Biodiversity, turn to a single final condition after disturbance has
and Successional Stage altered the initial condition to a variety of new con-
ditions. Ecological convergence after successional
Ecologists have spent a great deal of time discussing retrogression is an example of trajectory stability.
the stability of ecosystems. Lack of agreement over
just what is meant by stability has contributed to the Of these terms, constancy and persistence can be
longevity of the debate. Some people have claimed used to refer to the stability of individual seral stages.
508 CHAPTER 17 — Ecological Succession

High resilience
High constancy

% of time
ecosystem
exists in each Low constancy
condition hh
~

Range of possible ecosystem

conditions
(a)

High elasticity,
fast return

Low elasticity,
slow

Figure 17-15 Diagram of various concepts of ecosystem stability (modified after Orians,
1975). In B to F, the condition of an undisturbed ecosystem is represented by the position of a ball
in a hollow. Horizontal displacement of the ball from its resting position indicates a change in
ecosystem condition. The shape of the hollow reflects the various aspects of ecosystem stability be-
cause the shape determines the ease and direction of the movement of the ball when displaced (i.e.,
how the condition of the ecosystem changes when disturbed). Although it remains within the lip of
the hollow (the limit of stability of the ecosystem), the bail always returns to its initial condition
when released from disturbance. If pushed over the lip, then the ball will continue to move until it
reaches some new resting condition (i.e., another hollow). (The ball-hollow analogy was borrowed
from Holling, 1973.) (A) Constancy—lack of change in ecosystem condition (defined by one or more
parameters). (B) Inertia or resilience—degree of resistance of the ecosystem to change. Vertical rise
to the lip is a measure of the ecosystem resilience. (C) Elasticity—speed with which a disturbed
ecosystem returns to its initial condition. (D) Amplitude—degree to which ecosystem can be
changed but still return to its initial condition after termination of the disturbance. (E) Cyclic stabil-
ity—the condition of the ecosystem changes through a cyclic sequence of conditions. (F) Trajectory
stability—ecosystem will always return to the same condition irrespective of condition when re-
leased from disturbance. (Copyright Dr. W. Junk Publisher, Netherlands. Used with permission ofthe
publisher and the author.)
 Relationship Among Ecosystem Stability, Biodiversity, and Successional Stage 509

Cyclic and trajectory stability can be used to refer to mates is an example of the lower stability of these early
the overall stability of successional patterns, the for- seral stages. However, an early seral lichen community
mer referring to cyclical succession and the latter to on a rock surface may have high persistence and iner-
ecological convergence. Inertia, resilience, elasticity, tial stability; it also has low biodiversity.
and amplitude can be used to refer to the degree of Much of the discussion about ecosystem stability
successional retrogression after a disturbance and the focused on ecosystem structure: the species composi-
subsequent rates of succession back to the initial con- tion and spatial arrangement, the depth and composi-
dition. These are important concepts for the idea of tion of the forest floor, and so on. The evaluation of
ecological rotations, discussed in Chapter 18. stability has been broadened to include ecosystem
The problem with the quest for a predictable rela- function (O’Neill, 1976; O’Neill et al., 1975; O’Neill
tionship between diversity and stability arises from the and Reichle, 1980; Reichle et al., 1975; Webster et al.,
fact that a given ecosystem may be stable according to 1975). Ecosystems that have a large live biomass and a
one of the above definitions and unstable according to slow turnover of that biomass (i.e., large, long-lived
another. For example, diverse (species-rich) tropical organisms) generally have a high inertial stability. Re-
forests may have a high inertial stability but low elastic silience stability—the ability of ecosystems to recover
and amplitude stability. Conversely, many of the less from change—is often associated with a small biomass
diverse temperate forests may have a lower inertial that is turning over rapidly (short-lived organisms).
stability but a higher elastic and amplitude stability. However, a large capital of organic matter provides a
Ecosystem stability will vary between different reservoir of energy and nutrients that drive the het-
types of sere, between different seral stages, and be- erotrophic and autotrophic activities that return the
tween primary and secondary successions. ‘The inertial ecosystem to its initial condition. Consequently,
stability of a mesosere will be greater than that of a xe- ecosystems that are rich in organic matter and nutri-
rosere or hydrosere (it takes a greater degree of distur- ents tend to have both higher inertial and higher re-
bance to produce a major change), and mesoseres are silience stability than ecosystems that lack such
more elastic (they return to the climax more rapidly). reserves. Disturbances that deplete organic matter and
They also tend to have a greater amplitude stability nutrient reserves (e.g., fire) are potentially far more
(they can be pushed much further back in succession disturbing to long-term ecosystem function than dis-
before the return to climax is significantly delayed) turbances (e.g., windthrow) that merely rearrange or-
than xeroseres and hydroseres. Early stages of primary ganic matter and nutrient pools within the system.
seres will have lower values for several of these stabil- Different components of terrestrial ecosystems
ity indices than will early stages of secondary seres. may confer different types of stability on the ecosys-
Early seral stages may or may not be more stable tem. Forests usually have an understory of shorter-
than late seral stages or climax according to these def- lived plants that confer a degree of resilience to the
initions. Whereas a low-intensity surface fire may system, whereas the long-lived overstory provides in-
destroy an early seral stage, a climax forest may ertial stability.
be relatively unaltered by it. The opposite may be The inertial and resilience stability of forest
equally true, depending on the species involved. How- ecosystems are sometimes related to spatial and ge-
ever, whereas a windstorm may destroy a climax forest netic heterogeneity (i.e., diversity) of the biomass, but
seral stage, it may have no effect on a shrub stage. It this relationship is extremely complex and is difficult
has been argued that early stages are more stable than to generalize. The ability of an ecosystem to recover
climax stages, because it obviously takes longer to re- from perturbation is probably more closely related to
turn to the climax after severe disturbance than to re- its ability to process energy than to its diversity per se.
turn to a much earlier stage (Horn, 1974). This idea One of the reasons that a simple diversity—stability
may be justifiable in terms of elastic stability but over- relationship cannot exist is that in many ecosystems,
looks that such early stages have low constancy and maintenance of species diversity depends on distur-
persistence stability compared with the climax and bance. In the long-term absence of disturbance, only a
have lower amplitude stability, and both stages may few species may remain: species that do not require re-
merely be parts of the same overall trajectory stability. lief from competition or a special seedbed for regener-
The greater susceptibility of agricultural ecosystems ation (Grubb, 1977; Veblen, 1992) or species that have
to erosion by wind and flood compared with that of the lowest resource requirements at the species re-
the regional climax ecosystems in all but the driest cli- source equilibrium condition (Tilman, 1985). In many
510 CHAPTER17 — Ecological Succession

of the world’s forests, maintenance of both alpha and and autogenic change than low-diversity communities
beta diversity requires periodic disturbance (Attiwill, (e.g., Tilman and Downing, 1994) but may be more
1994a). However, this relationship may depend on the subject to disruption by allogenic change than low-
historic role of disturbance in that community. Forests diversity, disturbance-adapted communities.
that have an evolutionary history of disturbance may
achieve highest levels of many of the measures of di-
versity under periodic, intermediate levels of distur- 17.12 Succession in Animal
bance (Grime, 1979; Huston, 1979); sometimes the Communities
levels of many measures is highest in earlier stages
(Denslow, 1985). Conversely, ecosystems that are less So far we have talked about succession as though it in-
frequently subject to large-scale disturbance may volved only plants, but succession happens to ecosys-
achieve highest diversity in later seral stages (Denslow, tems, not merely to plant communities. Animal and
1985; van der Maarel, 1993). The veracity of these microbial communities each undergo a sequence of
theories depends on the type of ecosystem, the mea- changes that parallel the seral stages of the plants.
sure of biodiversity being considered, and the type and Most animals are habitat specific, and most microbes
severity of disturbance. However, there seems to be are substrate or host specific. Consequently, both
little doubt that in many instances, disturbance and groups change in response to the allogenic or plant-
ecosystem change are necessary prerequisites for the induced autogenic changes in the ecosystem. Some-
maintenance of at least some of the measures of biodi- times they are themselves the agents of successional
versity. Freedman et al. (1994) reviewed the relation- progression, retrogression, or stagnation (e.g., animal-
ship between biodiversity and forest management. induced climax). Insect defoliation of seral hardwoods
The scale at which stability is considered consti- can speed their replacement by conifers by as much as
tutes another problem in establishing these relation- half a century, and defoliator-induced mortality of cli-
ships. Winter snow break of branches may not max conifers can initiate a secondary succession begin-
constitute “disturbance” or a threat to stability at the ning with a hardwood stage. Heavy browsing by deer,
stand level, but it does to the epiphytic community liv- rabbits, or similar animals can maintain grassland “per-
ing on the branches. Gaps created by the death or manently” in areas that would normally proceed
windthrow of individual large trees may create signifi- through one or more additional seral stages. Figure
cant local disturbance but maintain a stable climax 17-16 shows the effect of wildlife browsing on tree re-
condition at the stand level. Death of a 20- to 60-ha generation (plants younger than 10 years) in a German
stand by fire, insects, wind, or logging may be viewed forest and the composition of a grassland before and 10
as a major disturbance but merely be part of a stable, years after protection from moderate wildlife browsing.
shifting mosaic of age classes or seral stages at the Each seral plant community is associated with a
landscape level. Before we can talk about ecosystem particular group of birds and other animals, a fact well
stability, we must clearly define the spatial and tempo- known to hunters. Early seral stages, in which most of
ral scales involved. the biomass is green, digestible, and within reach of
The relationship between biodiversity and stability ground-dwelling herbivores, tend to support a higher
is also complex: no single relationship applies every- secondary productivity and biomass of herbivores and
where (Simberlof, 1996). Many low-diversity northern associated carnivores than do late seral and climax
forests are very stable in that the same condition has stages. The climax is far from devoid of animal life,
persisted over many centuries or millennia as a result however. The many species of insect that feed on tree
of disturbance that results in regeneration of the same seed, foliage, and decaying wood provide food for a
species. However, these forests have low inertial resis- variety of insect-feeding birds. Many species of small
tance to disturbance by fire and insects. High-diversity mammals also feed on climax tree seed. Specialized
tropical forests have high resistance to insect damage herbivores, such as mountain caribou, feed on the
because only a small percentage of the trees will be af- lichens that grow in climax or mature late seral conifer
fected by an epidemic of a particular insect. However, forests. Many species of large ungulate herbivore feed
the forest has low resistance to fire and will take much in early seral stages in the summer but require mature
longer to recover to predisturbance condition than an late seral forest stages in the winter both for protec-
early seral, fire-adapted, low-diversity forest. Multi- tion against deep snow and for food (arboreal lichens
species communities may be more resistant to biogenic and green-leaf litterfall).
 Effect ofForest Management on Succession 511
o
o
45)
120 4 jaa)
Xy Areas subject to moderate Areas protected from
N browsing by wildlife browsing by wildlife
100
Deschampsia
sp.
80 ash
Mountain 30

20 weed
|Fire

raspberry
Red
%
Cover, Poa
sp.

Hornbeam

of
m2)!
plants
10
(100
old,
number
Average
<yr
No Browsing Browsing
(a) (b)

Figure 17-16 Effects of protection from browsing on the composition of the vegetation in
two areas in Germany. (A) The number of tree seedlings in a forest with and without browsing. (B)
The abundance of two species of grass (Pao chaixii and Deschampsia flexuosa) and other components
of a grassland with and without browsing (after Knapp, 1974). Browsed and unbrowsed areas were
on the same site type in both studies. The effects of such wildlife browsing on succession are obvi-
ous. (Copyright Dr. W.Funk Publisher, Netherlands. Used with permission of the publisher and the au-
thor.)

Figure 17-17 shows a sequence of bird and animal is obviously desirable to disturb the forest sufficiently
species associated with old-field succession in Illinois. to create physical and biotic conditions that favor the
Although the species sequences shown represent only growth of that species. This requires the creation of an
the study area, this diagram demonstrates that some an- early seral-midseral microclimate, tolerable levels of
imal species have narrow seral ranges, that others occur competition, and an early or midseral forest floor—soil
in almost all stages, and that the overall pattern of bird condition. Clearcutting may achieve the first two re-
and animal species replacement with time is very similar quirements but may not satisfy the third. If there is a
to that of plant species succession. The explanation for thick, late-seral (climax) forest floor condition that re-
this similarity lies in the dependence of many animal mains largely undisturbed after clearcutting, then
species on particular seral stages. As the plant commu- seedlings of the early seral-midseral crop species may
nity changes, so does the animal community. not grow well, especially if there are high surface tem-
peratures and low surface soil moisture conditions.
They have the appropriate microclimate but an inap-
17.13 Effect of Forest Management propriate soil condition. Late seral species will not do
on Succession well either because the early seral microclimate is in-
appropriate or its effects on the otherwise suitable soil
Effects of Clearcutting surface are intolerable. Essentially, the two major
As discussed above, a seral stage is defined in terms of physical components of the seral stage have got “out
microclimatic, soil, vegetation, animal, and microbial of step” with each other (cf. the use of bark mulch on
conditions. The living community that characterizes a flowerbeds to inhibit the invasion of weeds). Few
particular stage of a particular sere consists of organ- plants are able to cope with such a combination of
isms that are adapted to the physical and biotic condi- conditions, and such sites tend to revegetate slowly
tions of that stage. Clearcutting alters all of these after clearcutting. Either forest floor reduction
ecosystem parameters to a greater or lesser extent. through decomposition, fire, or disturbance must
Where the objectives of management require that occur, or there must be some modification of the mi-
an early or midseral tree species form the next crop, it croclimate by some hardy plant that can tolerate the
wn —_i) CHAPTER 17 Ecological Succession

Short-tailed
shrew

White-tailed Red
Cottontail deer squirrel
rabbit
White-footed
mouse
TEZEATZT————

~ Black-
throated
green warbler

Nashville ©

Field
sparrow
Grasshopper
sparrow

Figure 17-17 Sequence of animals associated with the stand cycle of pine plantations in
central New York (after Smith, 1980). Species vary in the number of stages with which they are as-
sociated.A succession of wildlife is found accompanying all plant successions. (Copyright R.L.
Smith. Used with permission ofthe publisher and the author.)

“unusual” combination of conditions. Thus, although ity characteristics of the area. If clearcutting poses
concern is often expressed over the degree of soil dis- problems of parameter desynchronization on such
turbance that accompanies logging, in some cases the sites, then the answer lies in not creating the marked
disturbance may be desirable. It may serve to “resyn- change in the microclimate. In fact, as pointed out
chronize” the microclimatic and forest floor compo- earlier, maximum production is generally attained by
nents of the seral stage. In fact, where accidental the late or climax seral stage of such seres (Figure
disturbance is inadequate, the site may be deliberately 17-13); therefore, major successional retrogression
disturbed by soil scarification. is usually undesirable anyway. Consequently, the
Resynchronization of successional conditions by scale of harvest disturbance should generally be re-
disturbance of the forest floor may be a reasonable stricted on such sites. Harvesting should be by small
thing to do in well-structured loamy soils on gentle patch, strip shelterwood, or individual tree removal
topography. It is less reasonable on very wet or very that maintains a permanent forest influence on the
dry sites, the extremes of which would be a hy- site. This conclusion will be modified by the climate
drosere and a lithic xerosere, respectively. Distur- of the area. Loss of forest influence is of much
bance in a hydrosere tends to raise the water table, greater consequence in hot and dry climates than in
which has already been raised by the reduction in mesothermal climates, and in winter-cold climates
transpiration loss. Disturbance in a lithic xerosere the soil warming caused by clearcutting may be ben-
may degrade the moisture, nutrient, and tree stabil- eficial to ecosystem function.
 Effect ofForest Management on Succession 513

Another way of judging the desirability of harvest- turbance. It may lead to soil erosion and mass wasting
ing disturbance is to consider the disturbance in rela- that, instead of merely initiating an earlier stage of
tion to rate of ecosystem recovery from that secondary succession, may push the area back to the
disturbance. Figure 17-8 shows three hypothetical start of a primary succession and even cause a change
primary seres that are similar to the seral development from one type of sere to another (e.g., mesosere to xe-
that can typically be observed in hydric, mesic, and rosere). Thus, clearcutting is a powerful and positive
lithic-xeric environments at low to middle elevations tool of successional manipulation, but it can be abused
in southern coastal British Columbia (cf. Figure 17-2). if the user is not aware of its full range of environmen-
Beneath each sere are two families of arrows. The tal consequences and how they vary in different sites
upper family shows the degree of successional retro- and in different climates.
gression (hypothetical) that would accompany various
intensities of harvest-induced disturbance in these
three types of environment. The lower family of ar- Effects of Slashburning
rows indicates the hypothetical degree of successional The successional consequences of slashburning de-
development of the area over the subsequent 100-year pend on the aggregate of all the other environmental
period for each level of disturbance. effects of slashburning, on the type of fire, and on the
Although there are no quantitative data to substan- type of site. A hot burn on a late seral xerosere might
tiate this hypothetical model of retrogression and eliminate all the surface organic accumulation and
recovery, it is supported by qualitative field observa- lead to the erosion of much of whatever mineral soil
tions. Figure 17-8 suggests that significant retrogres- has developed. This would transform the area to the
sion in xeroseres and hydroseres is followed by very beginning of a primary xerosere. Alternatively, a slash-
slow rates of recovery and that such retrogression may burn on a moist, fertile site might act to accelerate the
be considered to be semipermanent in terms of forest development of a secondary sere. By suppressing
management. Conversely, retrogression to the very shrub growth that might otherwise form a shrub ser-
beginning of the sere in mesic environments generally climax and by promoting the reestablishment of a tree
will not prevent succession from replacing at least the crop, slashburning can hasten the redevelopment of a
midseral or even the late seral stage over the subse- climax stand. Obviously, reliable generalizations about
quent 100 years. Moderate retrogression may permit the successional effects of slashburning are impossible.
the redevelopment of near-climax conditions in this Although clearcutting often results in successional
time period. The reader should be able to make an retrogression, it often has a greater effect on microcli-
analysis of this type for the forests with which he or mate than on forest floors. As discussed above, this can
she is familiar. render different parameters of successional condition
In summary, clearcutting inevitably creates succes- out of phase with each other. Mechanical scarification
sional change. Sometimes it pushes an area back to an can sometimes rectify the situation, but where topog-
earlier seral condition, which may or may not be desir- raphy and/or stumps and slash make this impossible,
able. In other cases, it does not create sufficient distur- fire can be used. By reducing the depth of low-bulk-
bance to achieve management objectives. In such density L-F layers and by altering forest floor chem-
cases, it must be combined with soil scarification or istry, prescribed fire can return climax forest floors to
slashburning to synchronize the soil surface and the an earlier seral condition that matches the early seral
microclimatic conditions or simply to create an earlier microclimate. On sites where this has unacceptable re-
seral stage. In yet other cases, clearcutting may accel- sults in terms of ecosystem function, one might con-
erate succession by removing the present dominant clude that radical changes in microclimate as a result
tree species and releasing understory seedlings and/or of clearcutting should not be permitted. Small patch
saplings (advance regeneration) of shade-tolerant cutting or using the shelterwood method would per-
species of the next seral stage. Unfortunately, the ef- mit the maintenance of a forest microclimate, reduc-
fects of clearcutting are not always so benign. In xe- ing the need for radical alteration of forest floor
roseres and hydroseres, it can produce combinations condition. In dry, hot environments, these alternative
of soil and microclimatic conditions that are unsuit- harvesting methods will often be preferable to
able for the desired species of plant, at least initially, clearcutting. In cooler and more humid environments,
and this cannot normally be rectified by further dis- especially with very old and overmature forests,
514. CHAPTER 17 Ecological Succession

clearcutting, sometimes accompanied by slashburn- can also be important. The various processes of succes-
ing, will frequently be the best management strategy, sion occur in a wide variety of environments with the re-
at least in the first rotation. Other methods may be sult that there is an enormous variety of seres. This
both possible and desirable in subsequent rotations. variety can be classified, one of the most useful ways of
doing so being according to the moisture status of the site.
We will return to this question in the next chapter.
The variation in mechanisms and physical environ-
ments involved in succession produces great variation in
a
N
Oa
my

rates of succession and the duration of different seral

SUMMARY stages. Rates and durations tend to be characteristic for
One of the characteristics of ecosystems is change. Soils each major type of sere, but there is a considerable range
develop or are eroded. Weather varies seasonally, and cli- within each type, and generalizations must be made with
mates exhibit long-term change. Individual organisms caution. There can also be a great variation in the pattern
develop, mature, reproduce, and die. of seral stages in some types of sere.
Throughout these continual changes in the physical, Changes in the structure and species composition of
chemical, and biological environment, the processes of the community are accompanied by changes in ecosys-
energy capture, energy storage in biomass, and the cy- tem function. Productivity, biomass, and the exchange of
cling of nutrients continue. The community of organ- energy between different components of the ecosystem
isms that is responsible for these processes changes, the all vary between different seral stages and between differ-
pattern of change reflecting the type of physical and ent types of sere. Nutrient cycling similarly shows a pat-
chemical environment and the successive modification of tern of change as succession proceeds.
that environment by both the biota itself and various Much of the theory and discussion of succession is
combinations of geological and climatic processes. It is concerned primarily with plant succession. However,
this process of change in ecosystems that we call ecological succession refers to ecosystem events, and there are con-
succession, whereas the pattern of ecosystem conditions comitant successions in the animal and microbial com-
observed over time in a particular location is called a sere. munities. Because these organisms are largely dependent
Early attempts to describe general patterns of vegeta- on autotrophic production and the plant-controlled mi-
tion change after disturbance resulted in the theory croenvironment, such successions tend to follow in the
(monoclimax) that there is a fixed sequence of communi- wake of plant succession. There are, however, numerous
ties that inevitably succeed each other in a region until a examples in which animals and microbes play a major
stable, self-replacing climax in equilibrium with the re- role in determining the course, the speed, and the out-
gional climate is produced. The attractive simplicity of come of plant succession. Succession thus is an ecosystem
this theory resulted in its survival for many decades until phenomenon, and it is not the special domain of any par-
its preeminence was challenged by a second theory (poly- ticular branch of ecology.
climax), which claimed that community composition and As a component of ecosystems, humans’ day-to-day
stability are under the influence of many factors, any one activities are intimately affected by succession. Both
of which may be dominant at different times and places. farming and forestry, two fundamental human activities,
Succession will not necessarily always follow the same are constantly working to manipulate successions to opti-
pattern and stabilize with the same climax community. mize the availability of certain seral conditions.
The result is a patchwork of different types of climax Without an adequate theoretical and/or empirical
communities that are at equilibrium with a number of knowledge of succession, successful land management is
different factors. A third theory (climax pattern) consid- generally a matter of luck, and many of the common ex-
ers that because communities integrate with one another, amples of unsuccessful management could have been
because of the individualistic distribution of individual avoided by applying a knowledge of succession. Our cur-
organisms along environmental gradients, the climax rent environmental predicament and that, globally, we
community will similarly exhibit an integrating pattern. still seem to be losing ground, literally and metaphori-
The climax community that occupies the largest propor- cally, are in no small degree due to a failure to recognize
tion of habitats that are not edaphically or microclimati- and understand succession as it occurs in the wide variety
cally special or extreme for the area is considered to be of terrestrial and aquatic environments on Earth.
the prevailing climatic climax and to express the climate As our understanding of succession has evolved, there
of the area. has been a switch from broad theories about successional
Succession occurs for a variety of reasons that vary in pathways to consideration of processes of species replace-
their importance from place to place and from time to ments. This has emphasized the importance of the eco-
time. Biotically induced change is probably the major dri- logical attributes of individual species—their life history
ving force of most successions, but geological processes attributes and “strategies”—as well as population and
 Study Questions Di)

community processes. This is not enough, however. of natural disturbances that we no longer permit, and we
Many life history attributes are related to resource acqui- must balance the combined frequency and severity of dis-
sition, and resource availability is determined by allo- turbance with the ecological requirements of the species
genic ecosystem processes as well as autogenic com- and seral stage(s) that we wish to sustain. This requires
munity processes. This has led to the increasing popu- the matching of disturbance regimes to the ecology of
larity of computer models of succession and to the grow- the local site and landscape. One level and frequency of
ing realization that these models must be ecosystem-level management-induced disturbance everywhere, irrespec-
models rather than population or community models— tive of the ecological diversity of the landscape, is simply
that they must incorporate representations of many dif- unacceptable.
ferent processes, not only competition for light between
trees.
STUDY QUESTIONS
TAKE-HOME MESSAGE 1. Why does ecosystem change occur?
ho. In which type of environment might the monoclimax
Change is an inevitable and natural component of theory be a reasonable approximation of the pattern
ecosystems (Botkin, 1994). In many ecosystems, the of ecosystem change?
change that results from disturbance is necessary to
maintain recent historical levels of ecosystem function 3. What is the importance of understanding succes-
sional facilitation, tolerance, and inhibition in forest
and biodiversity; in others, disturbance can impair these
ecosystem characteristics, depending on its frequency,
management?
type, and severity. 4. How have ideas about succession changed over the
Except in the earliest stages of cultural evolution, past 80 years?
humans have modified successional processes through- 5. What are “vital attributes,” and how does knowledge
out history: to promote hunting, for food production, of them help us to predict vegetation dynamics in fre-
and to provide timber and other forest values. All too quently disturbed environments?
frequently, this modification has been made in the ab- 6. What are Tilman’s major contributions to the devel-
sence of an adequate understanding of succession and opment of our understanding of succession?
ecosystems. This did not matter when human numbers
were small and there was little technology, but as the 7. What are the major mechanisms of autogenic succes-
human population increased and technology became sion?
powerful, successional manipulations in the absence of 8. What determines the rate of autogenic succession?
appropriate ecological knowledge often produced seral 9. Do you think that there is one simple theory of
conditions that reduced values for humans and many succession?
other species. Development of a sustainable relation- 10. Why have ecologists turned to computer models as a
ship between the humans and both local and global en- way of evaluating theories about succession?
vironments requires that environmental modifications
be made with explicit knowledge of their successional 11. What is the role of disturbance in ecosystems?
consequences. 12. What is the relationship between ecosystem stability
If forestry is to be sustainable, then management- and disturbance and among biological diversity, sta-
induced ecosystem disturbance must emulate the effects bility, and disturbance?
 Understanding and Emulating
Natural Forest Disturbance
A Template for the Management of Succession
and Temporal Diversity?

18.1 Introduction desired values are sometimes provided by species

adapted to mid- or even late seral species and condi-
Chapter 17 described how change is initiated in forest tions. Forestry and agriculture thus both are deeply
ecosystems by disturbance that alters the existing involved in the management of disturbance and of au-
biotic community and the physical conditions of the togenic succession to create and sustain the seral con-
ecosystem. After disturbance, change is continued by ditions or the sequence of seral stages required by the
the processes of autogenic succession. This allo- desired values.
genic/biogenic disturbance—autogenic ecosystem de- Success in intensive agriculture and forestry is usu-
velopment cycle is responsible for maintaining the ally achieved as a result of significant investment in
sequence of biotic communities that successively oc- site preparation (site disturbance), weed control,
cupy a particular physical environment over time. It planting, and fertilization to create the early seral mi-
thus is closely related to the maintenance of high lev- croclimate, assart period of elevated soil fertility, and
els of various measures of beta and gamma biodiver- reduced vegetation competition required by early
sity, including temporal diversity, and a variety of seral crop species. Where the social value of harvested
other values. In some types of forest, it is closely relat- resources is high, such as most food crops and some
ed to the long-term maintenance of ecosystem pro- timber crops, the economic costs of emulating
ductivity. through intensive management the ecological effects
The successional disturbance-recovery cycle in of repeated natural disturbance can be justified. In
forests creates habitats for a wide variety of organisms, contrast, the economic value of wood products in
each different seral stage providing for the life necessi- many forests cannot sustain such investments. Also, al-
ties of adifferent assemblage of species. Intensive agri- though well-managed agriculture may sustain soils,
culture (as opposed to food gathering in unmanaged hydrology and water quality, and values related to
ecosystems) is often based on early seral, disturbance- food crops, few other values may be sustained. In con-
related food crop species, both animals and plants, be- trast, much of the forested landscape must be man-
cause human dietary needs are generally met by a aged to supply many values in addition to those
balance of herbivory on early seral plant species and associated with timber crops. This usually requires a
carnivory on early seral herbivores. Similarly, forestry much wider range of seral conditions and therefore
often manages early to midseral tree species because of much longer crop rotations, which can raise signifi-
their fast early growth, their tendency to have some cance economic barriers.
form of nitrogen fixation ability, their adaptation to The historical range of values in unmanaged
the disturbance caused by harvesting, and their wood forests has been provided in the past by the interaction
properties. However, in both agriculture and forestry, of natural disturbance events and_postdisturbance

516
 Paradigms for a New Approach to Forestry = 517

ecosystem recovery. Where the historical disturbance Emulation of natural disturbance is becoming a
regimes no longer occur as a result of human action or popular concept for sustainable management of multi-
natural environmental change, we may be able to re- ple forest values. However, careful consideration must
produce and sustain much of this variety of values be given to defining and understanding the ecological
through management. However, where our knowl- role of natural disturbance if we wish to design man-
edge of how to do this is limited, where we lack an in- agement to emulate it. In this chapter, we briefly ex-
ventory of all the values that the forest provides, or amine a variety of paradigms that could be used to
where economics prevent an intensive management guide the development of forestry into the 21st centu-
approach to their maintenance, we may be able to sus- ry and then focus on how natural disturbance could be
tain the full range of historical values by emulating used as a template for this development.
through management the ecological effects of past
natural disturbance regimes. To be successful in this,
both ecologically and economically, it is clear that we
must understand the ecosystem effects of the past fre-
quency, scale, severity, and types of natural distur-
18.2 Paradigms for a New
bance that have been associated with the values that
we wish to sustain.
Approach to Forestry
Disturbance is recognized as a ubiquitous phenom-
In designing changes to how forests should be man-
enon in the world’s forests, and much has been written
aged, there are several different approaches or para-
about disturbance and its role in maintaining ecosys-
digms that could be used (Kimmins, 2002). These are
tem productivity, biodiversity, and other ecosystem
not mutually exclusive—many of them will be used in
conditions (Attiwill, 1994a,b; Cairns, 1980; Luken,
combination.
1990; Mooney and Godron, 1983; Pickett and White,
1985; Sousa, 1984; White, 1979). In many forests, it is
so pervasive that classical theories of autogenic succes-
sion do not apply (Rowe, 1961b). The well-established
Ecosystem Management
dependence of biological diversity on disturbance in Although there are many different interpretations and
many forests (references in Attiwill, 1994a) requires definitions (Boyce and Haney, 1997), ecosystem man-
that ecologists and resource managers gain a much agement (EM) means managing forests as whole, inte-
better understanding of the role of disturbance in for- grated systems rather than the frequently fragmented
est ecosystems. It also requires that policies designed and uncoordinated management for individual ecosys-
to achieve desired levels of particular measures of bio- tem components and values—trees, wildlife, water,
diversity account for the relationship between distur- aesthetics, and so on. EM requires a single EM plan
bance and the desired biodiversity conditions. for all desired values implemented by a single agency
Disturbance in forests comes in many forms: fire, or institution on a defined forest area over ecologi-
wind, insects, disease, landslides, floods, ice storms, cally appropriate spatial and temporal scales. It re-
wildlife herbivory and livestock grazing, timber har- quires a clear definition of a desired future forest
vesting, road construction, and conversion to nonforest condition, and there must be a respect for all forms of
land use such as agriculture. These disturbances cannot diversity (including temporal diversity) and for the
easily be characterized because each can vary in fre- ecology of sustainability and natural disturbance. EM
quency, spatial scale, and severity; disturbance in forests requires a focus on which trees and other ecosystem
forms a continuum from individual tree replacement components to leave, which ecosystem processes to sus-
(small gaps) to hundreds of thousands of hectares killed tain, and which sequence of ecosystem (successional)
by fire or insects (Bormann and Likens, 1979). Both conditions to create after harvest, rather than on which
natural and human-caused disturbances form a spec- trees to take and regulations as to what not to do.
trum of disturbance regimes and effects, with a broad Because of the complexity of forest ecosystems,
overlap between the two (Attiwill, 1994a). Some natur- planning for EM requires the uses of multivalue EM
al disturbances have a counterpart in human-caused models that are capable of spanning the temporal and
disturbance, and vice versa, but some natural and spatial scales involved. EM is explored further in
human-caused disturbances have no such analogue. Chapter 19 as we discuss landscape ecology.
518 CHAPTER 18 Understanding and Emulating Natural Forest Disturbance

Adaptive Management Suitable models for this purpose are discussed in

Chapter 21.
Most management practices have not been applied for
long enough for ecologists and resource managers to
Zonation
be confident about the long-term results, so all forest ag
i
ma
en
+
Oe
es
><
management is essentially an experiment from which Using different management practices in different
we should learn by monitoring the outcomes (Walters, types of forest ecosystem or to sustain different sets of
1986). The results of monitoring are compared with values in any particular type of ecosystem is a corner-
anticipated (hypothesized) outcomes. Unless this stone of the zonation paradigm. Examples of multiple-
comparison is made, there is little or no basis on which use and integrated resource management approaches
to judge the success of management, and monitoring from the past provide ample evidence of the need for -

may constitute little more than ecological and social spatial and/or temporal separation of certain forest
“stamp collecting.” Management that accepts uncer- values—it is ecologically impossible to provide all val-
tainty and risk, tries different methods, and learns by ues all of the time from a particular local forest stand.
comparison between the results of management and Zonation provides spatial separation of different types
predictions is called adaptive management (AM). and intensities of forest management and for different
AM involves the synthesis of existing knowledge timber and nontimber values across the forest land-
and experience into predictive forecasting systems, scape. It provides space-for-time substitution. All of
trying different management strategies in different the values provided over time in any one ecosystem as
places, monitoring the results and comparing them it develops through all the seral stages of an entire sere
with our forecast of outcomes, and improvement or can be provided by managing different parts of the
change of practices and/or forecasting systems when landscape to sustain different seral stages. How large
outcomes are not as predicted. The forecasting sys- each of the zones should be depends on the spatial
tems also play a role in selection of policies and prac- scale of the ecological processes that are responsible
tices that are to be tested and in exploration of for the characteristics of a particular management
alternative possible and desired forest futures that zone. This issue lies at the heart of landscape ecology
generally cannot be imagined without such tools. (Chapter 19).
AM is particularly suitable for and was developed Theoretical considerations suggest that the world’s
for application in resource management systems in demand for wood could be satisfied from the intensive
which the results of change in policy and practice can management of 20% or less of the world’s forests
be measured over relatively short time periods, such as (Binkley, 1997; Sedjo, 2001; Sedjo and Botkin, 1997).
agriculture or fisheries. In forestry, we frequently can- There are many reasons that the “20-80” solution
not measure the full impact of changes in management (20% forests managed intensively for timber; 80%
for many decades (e.g., Lundmark, 1977, 1986) or managed for a balance of timber and nontimber values
even several tree crop rotations (which may involve or for nontimber values or not managed) would prob-
centuries). As a consequence, AM without adequate ably fail to satisfy all sectors of society. These range
forecasting systems may not solve critical problems in from social and cultural considerations to political and
forest management over time scales of many decades. sovereignty issues. However, the concept makes the
Forecasting for AM in fisheries or agriculture has in- point that there is not enough global demand to jus-
volved relatively simple forecasting models, and al- tify intensive timber management on all or even the
though more complex models may give improved majority of the world’s forest land and that on some for-
accuracy of prediction, some of the simple models est land, nontimber values should be the focus of man-
have proved useful. In contrast, the forecasting needs agement. Zonation of the “working forest” into more
of AM in forestry today require the use of relatively intensive and less intensive timber management areas
complex, multiple-value, EM models that span spatial and areas managed for nontimber values makes good
scales from tens to hundreds of thousands of hectares economic sense and can be good for nontimber values.
and from a few years to decades or centuries; simple The design of zonation systems requires a sound
models are not adequate (Seely et al., 1999). These knowledge of successional processes, especially for
models must be able to address the effects of distur- zones in which management intensity will be low.
bance on forest ecosystem function and succession. Planning zonation systems is best achieved using EM
 Paradigms for a New Approach to Forestry 519

models that can span from the stand to the landscape bance-based paradigm is maintenance of the natural
or forest estate level. (or historical) range of variation (NRV or HRV)—the
range in ecosystem conditions that is the outcome of
past natural disturbance. NRV can be useful as a tem-
Variable Retention plate for management of forests, such as the boreal,
The different values in a forest have different ecologi- that have experienced relatively frequent stand-replac-
cal rotations (Kimmins, 1974)—they require different ing natural disturbance events, because the natural
periods for their renewal. For example, trees that are frequency of disturbance matches fairly closely the de-
suitable for eagles’ nests or bear winter dens and the sired management frequency of disturbance (tree crop
rotations). NRV is less useful in ecosystems that have
supply of large decomposing logs require longer for
experienced infrequent and/or very large-scale or very
their renewal than does commercial wood volume.
severe disturbance in the past. Society is so unlikely to
These different values can be sustained in different
accept the emulation of such disturbance regimes as to
areas by zonation, but there is often merit in sustaining
severely limit the utility of NRV in such forests. Simi-
values with different renewal periods in the same stand
larly, if NRV is applied to forests where natural distur-
by retaining elements of older forest within a matrix of
bance tends to be frequent and small scale with very
younger trees. This is achieved by retaining individual
infrequent stand-replacing disturbance, then NRV
“wildlife trees,” wildlife tree patches and snags within a
may imply a level and frequency of management dis-
more intensively managed timber stand. Where there
turbance that is inconsistent with tree disease, soil
is a spatially variable retention strategy that allows for
compaction, road density and other management con-
different levels of retention for different values, site
siderations, species composition, desired levels of
types, economics of management, worker safety issues,
wildlife, and various measures of biodiversity.
and so forth, this paradigm is referred to as variable re-
NRV describes the range of disturbance patch sizes,
tention (VR) forestry (Franklin et al., 1997). VR is a
the landscape pattern of disturbance patches, and the
subset of structure management silviculture that is based
range in severity of ecosystem disturbance (and thus the
on conceptual models of natural disturbance and stand
variability in plant community composition and struc-
development (Franklin et al., 2002). Although it was
ture) that have occurred historically over some defined
developed with wildlife and biodiversity issues in
period. Social, cultural, and economic considerations
mind, VR is applied as much for aesthetic and public
are then used to select that portion of this range that is
acceptance reasons as for these conservation purposes.
acceptable. A comparison of NRV with the socially se-
As with the other paradigms, it can be used in combi-
lected range of desired conditions alerts managers to
nation with EM, AM, and zonation.
the portion of NRV and associated environmental and
Despite the many merits of VR forestry systems,
other values that are being foregone.
there can be significant insect, disease, windfirmness, A problem with ENFD and NRV/HRV is that
tree regeneration, biodiversity, worker safety and eco- long-term, natural climatic variations, changes in the
nomic constraints on their design, and these may limit frequency of forest fire, invasion of nonnative species,
the application of VR in some forest ecosystems. Plan- air pollution, and the current threat of human-
ning for VR should involve the use of EM models that induced rapid climate change render some past distur-
can represent the spatial heterogeneity in ecosystem bance regimes and the resultant NRV inapplicable as a
conditions, processes, and growth resources created template for the future. It also poses the question,
by such systems. “Which historical period should be the baseline for
NRV and HVR?” These issues have led many to reject
the NRV paradigm in favor of managing for a desired
Emulation of Natural Forest Disturbance future forest condition and pattern, which may or may
A paradigm that is popular with some environmental not be similar to HRV.
groups and many natural disturbance ecologists is As noted above, one of the ways in which forests
emulation of natural forest disturbance (ENFD) (Angel- could be managed to maintain conditions and values
stam, 1998; Attiwill, 1994; Bergeron et al., 1999; Lief- close to those that have prevailed historically, if this is
fers et al., 1996; McRae et al., 2001; Spence et al., desired, is to design management-related disturbance
1999). Another way of looking at this natural distur- or combined “natural” and management disturbance
520 CHAPTER I8 Understanding and Emulating Natural Forest Disturbance

regimes so that their effects on forest ecosystem struc- (a leaf, branch, tree, stand, or landscape); the time over
tures and processes are similar to those of disturbance which the definition is to be applied (hour, day, year,
regimes during the historical period selected as the decade, etc.; White and Pickett, 1985); and the object,
comparison baseline (e.g., Bengston et al., 2000). value, condition, or process that is being disturbed.
Management to sustain individual wildlife species or These authors define disturbance as “any relatively
ecosystem conditions—the fine filter approach—has discrete event in time that disrupts ecosystems, com-
proved to be problematic because for every species or munity or population structure and changes resources,
group of species that is benefited, there may be nega- substrate availability, or the physical environment.”
tive effects for another species or group of species. They note that semantic problems persist in the de-
Maintaining the NRV in forest age, seral stage, and bate over the meaning of disturbance and that it is al-
ecosystem conditions—the coarse filter approach—has a ways important to define how it is being used.
much higher probability of sustaining all species and Before the working definition of disturbance that
values across the landscape, especially where our is used in this chapter is presented, it is worth examin-
knowledge of their ecology is limited. However, be- ing the meaning of individual words.
fore ENFD is accepted as the template for designing
sustainable forest management, much more detail is
required to clarify what “disturbance” is; the NRV in Disturbance
disturbance and resultant ecosystem conditions in par-
There are many different interpretations of disturbance
ticular forest types; which portion of this natural range
but few agreed-on definitions. Webster’s New Twenti-
is socially acceptable; and which approach to forest
eth Century Dictionary, 2nd Edition (1975) gives it as
management would result in a desired, socially accept-
“interruption of a settled state of things.” However,
able, and nondeclining pattern of change in forests.
because for most forest ecosystems there is no “settled
Throughout this section, it has been suggested re-
order” (Edwards et al., 2000), White and Pickett’s
peatedly that the complexity of these questions and
(1985) definition given above may be more appropri-
the unacceptably long time required to answer them
ate. Rogers (1995) noted that most ecologists focus on
from experience alone requires the harnessing of
the disruption of the functions of ecosystems when they
available knowledge and experience in ecosystem-level
think about disturbance. However, it is unclear what
simulation models. These can be used to create virtual
“disrupts” (“to break or burst apart” [Webster’s dic-
forest ecosystems in which alternative paradigms and
tionary]) means in the context of an ecosystem that is
management systems can be given preliminary assess-
constantly changing by nondisruptive processes and
ment before being field-tested.
that continues to function after disturbance.

18.3 Definitions of Disturbance Natural

and ENFD Webster's dictionary defines natural as “forming part
of, or arising from, nature; what is found or expected
Betore we can discuss ENFD, we need to agree on what in nature”; “in a state provided by nature, without
“disturbance” is. Notwithstanding that disturbance is a man-made changes; wild, uncultivated,” but these def-
spatial and temporal continuum and involves a con- initions are confounded by Webster’s definitions of
tinuum of severities, some definition of disturbance is nature: “the sum total of all things in time and space,”
necessary if we are to be able to discuss, plan, and apply “the primitive state of man,” or “not cultivated or
forest practices that emulate natural disturbance and tamed; wild; savage,” which either do not deny that
produce a natural or historical range of ecosystem vari- humans are part of nature or explicitly include hu-
ation or a socially acceptable subset thereof. mans. Evenden, in Paehlke (1995), asserted that “na-
The literature on this topic makes it clear that the ture” is a term of such dangerous ambiguity to make it
definition. will depend on the level of organization one of the most obscure terms in the English lan-
(suborganism, organism, population, community, and guage. He noted that for many in the environmental
ecosystem) being considered; the spatial scale involved movement, nature has become synonymous with
 Management Strategies with Respect to Succession 521

wilderness, an etymological shift that would substan- to change the prevailing rate, direction, and pathway ofau-
tially change the traditional meaning of the word. togenic succession and either return the ecosystem to an ear-
One of the best treatments of natural is given by lier seral stage, reinitiate the present seral stage (thereby
Peterken (1996). He noted that Henry David Tho- arresting autogenic succession), or accelerate the transition
reau, that “apostle of the wild,” believed that to a subsequent seral stage in comparison with the rate ex-
“mankind and nature form an ecological unity.” Pe- pected from autogenic succession; or alter the successional
terken believes that definition of natural depends on pathway (the sequence ofseral stages) that the autogenic suc-
personal philosophies and belief systems but that strict cession was following. The normal autogenic processes
adherence to definitions that exclude humans would of invasion, colonization, competition-related mortal-
exclude most terrestrial ecosystems from membership ity, environmental alteration, and displacement of ex-
in the class natural, a view supported by Botkin (1995). isting species (cf. the stand development phases of
Oliver and Larsen [1990] or Franklin et al. [2002] and
their associated processes) are not considered ecosys-
Emulation
tem disturbance. This definition is proposed in the full
The Webster’s dictionary definition of emulation is “to knowledge that disturbance is a continuum of scale,
try to equal or exceed,” whereas mimic is “to copy severity, and frequency, and, as a consequence, no sin-
closely, to imitate accurately, to simulate,” and copy is gle definition will satisfactorily serve all discussions of
“to duplicate, to fully reproduce, to imitate.” There the management of disturbance.
has been considerable discussion over whether forest
management impacts mimic or copy “natural distur-
bance” (e.g., McRae et al., 2001), and the word 18.4 Management Strategies with
emulation was introduced as a more neutral term that Respect to Succession
recognizes to a greater extent than copy or duplicate the
inevitable differences between management-induced Because ecosystem change is inevitable and because
and nonhuman disturbance effects. The essence of many desired values are sustained only under an ap-
emulation, then, is to try to equal the overall effects of propriate regime of disturbance and change, much of
natural disturbance through well-designed manage- forest management is concerned with management of
ment disturbance (Fiihrer, 2000). From this discussion disturbance and_ succession. Silvicultural systems
of definitions, it seems that although we have a good (Figure 18—1)—systems of stand treatments designed
grasp of the meaning of emulation, it is far from clear to regenerate a new stand of trees after timber harvest-
what “natural” means (depending on personal philoso- ing and to manage the new stand to sustain desired
phies and the purpose for which “natural” is being timber and nontimber values until the subsequent har-
used), and there is an urgent need to understand what vest—constitute a continuum of disturbance from that
“disturbance” means. caused by harvest ofindividual trees (the selection har-
As a working definition for this chapter, ENFD is vesting system) to the removal of all the trees in a sin-
defined as “the attempt to emulate through manage- gle harvest from an area sufficiently large that the forest
ment interventions, over ecologically significant time influence is removed from more than half of the harvest
and space scales, the ecosystem effects of physical (al- area (the clearcutting system). These different silvicul-
logenic) or biotic (biogenic) events that human action tural techniques can be compared with natural distur-
has changed but that have historically altered the ex- bance effects in terms of the removal of trees and the
pected pathways, patterns, and rates of autogenic suc- associated loss of forest influence (Figure 18-2).
cessional development in the ecosystem in question. What is immediately obvious from this comparison
Such emulation has as its objective the maintenance of is that most types of natural disturbance do not consti-
the historical NRV or a socially acceptable subset tute a single severity class the way that individual silvi-
thereof in desired ecosystem conditions and functions cultural systems do. Fire can have disturbance effects
over defined spatial and temporal scales.” that range in severity from that associated with individ-
The rest of the chapter explores this definition of ual tree or group selection to the ecosystem distur-
ENED, which is based on the definition of disturbance bance created by clearcutting plus slashburning
as allogenic (physical) or biogenic (biological) events that act (although the removal of the tree stems in harvesting
Wwbhbh CHAPTER 18 Understanding and Emulating Natural Forest Disturbance

Single tree selection and their retention in most natural disturbance consti-
tutes an inevitable and significant difference between
harvesting and most natural disturbances). Wind may
simply thin out a stand or can cause complete stand re-
placement over large areas. Landslides cause a nar-
rower range of disturbance severity that is normally far
greater than the most severe silvicultural disturbance—
it initiates new primary succession, whereas silviculture
aims to initiate new secondary successions. Thus, emu-
lation of the NRV in forest ecosystems involves not
only applying the full range of silvicultural systems re-
quired to span the natural range of disturbance inten-
sity but also varying the scale and disturbance intensity
associated with each silvicultural system.
The upper part of Figure 18-3 presents a diagram-
matic model of “natural” disturbance of moderate to
high severity interacting with autogenic succession in
a hypothetical mesosere that has a deciduous
broadleaved pioneer forest community; early, middle,
and late seral conifer communities; and a_ final
shrub/herb woodland climax community. In such a
Shelterwood
mesosere, where Egler’s “initial floristic composition”
pathway (Egler 1954) or the tolerance pathway of
Noble and Slatyer (1977) is the expected pattern of
ecosystem development, autogenic succession can un-
dergo any of a variety of postdisturbance development
sequences. The diversity of possible successional con-
sequences of allogenic/biogenic disturbance shown in
Figure 18-3 reflects differences in type and severity of
Seed tree
disturbance and the different seral stages in which the
disturbance can occur. The lower part of Figure 18-3
shows possible successional consequences of allo-
genic/biogenic disturbance of low to moderate sever-
ity in this hypothetical mesosere. In contrast to the
regressive effects of moderate to severe disturbance,
low- to medium-severity disturbances can accelerate
Clearcutting
autogenic succession. It does so by prematurely thin-
ning or removing the overstory of the existing seral
stage, releasing advance regeneration (the understory)
of seedlings/saplings of the next successional stage, or
facilitating their early invasion of the site. The wide
diversity of possible regressive and progressive effects
of disturbance on ecosystem conditions and dynamics
Figure 1S-1 Diagram of the degree of treé removal and
change in forest influence associated with “classical” silvi-
lies at the root of the difficulty in making useful gen-
cultural systems. Recent adaptations of these systems as part eral statements about the effects of disturbance in for-
of VR or structure management silviculture involve permanent est ecosystems.
retention of individual trees (similar to the shelterwood or seed Figure 18-4 provides a diagrammatic analysis of
tree systems but with permanent retention) or groups of trees management-induced (biogenic) disturbance de-
dispersed throughout the harvested area. signed to emulate allogenic or nonmanagement bio-
 Ecological Rotation Concept as a Framework for ENFD 523

% trees harvested % trees retained

100

dH
————
Clearcutting

LANDSLIDE ===>
Shelterwood Patch Cut
BS Selection Reserve

Figure 18-2 Comparison between the degree of tree removal in classical silvicultural sys-
tems and in various types of natural disturbance. Note that for most natural disturbance types,
there is a wide degree of tree removal, so a variety of silvicultural systems can be used to emulate dif-
ferent degrees of severity of natural disturbance.

genic disturbance to achieve a variety of seral objec- historical range of variation in forest ecosystem con-
tives. Figure 18—4A presents successional diagrams ditions (cf. Figure 18-2).
for six different timber management-disturbance ob-
jectives: conversion of nonproductive early seral (by
chemical, mechanical, fire, biotic, or manual “vegeta- 18.5 Ecological Rotation Concept
tion management” techniques) or late seral (by fire or as a Framework for ENFD
mechanical disturbance) stages to a forest community
that is productive of desired values and disturbance- An ecological rotation (Kimmins, 1974) is the time re-
recovery sequences that maintain four different seral quired for an ecosystem or a particular ecosystem at-
conditions. Figure 18-4B depicts the use of silvicul- tribute (e.g., composition, structure, function,
tural strategies using disturbance to convert (acceler- complexity/diversity) to return to predisturbance
ate) various seral stages into late-seral, closed-forest conditions or some desired new condition after dis-
conditions and then maintain these. Figure 18-4+C turbance. It is a function of the degree of disturbance-
shows a variety of disturbances of late-seral forest to induced change (the disturbance severity) interacting
initiate new seral sequences of various lengths, fol- with the rate of autogenic recovery from that change
lowed by management-induced low-severity distur- (a measure of ecosystem resilience; Figure 18-5).
bance to accelerate stand and seral development. Forestry based on the ecological rotation concept will
Figure 18-4D contrasts silvicultural disturbance result in nondeclining patterns of change in ecosys-
regimes that sustain (1) early or (2) mid to late seral tems over the period being considered.
sequences. Achievement of these different succes- The ecological rotation concept can provide a con-
sional pathways and outcomes requires the use of ceptual framework for the design of ENFD regimes.
mixtures of several traditional silvicultural systems It provides a way of taking the successional models
(Figure 18-1), modified in various ways (e.g., VR sys- shown in Figures 18-3 and 18-4 and converting them
tems; Franklin et al., 1997) to emulate natural distur- into a management scenario. Figure 18-6 presents
bance and re-create the desired portion of the a hypothetical example in which a conventional
524 CHAPTER 18 Understanding and Emulating Natural Forest Disturbance

Climax herb/
Early seral | Pioneer Early seral Mid seral | Late seral shrub/byrophyte
herb/shrub | hardwood | conifer conifer | conifer | woodland

es , sildike
Fed
¢

\WAAARCS
(ne Al]orenic/biogenic disturbance

———p AU LOgenic succession

Figure 18-3 Possible pathways of successional retrogression resulting from ecosystem dis-
turbance of various types and spatial scales and of moderate to severe severity (upper part of
figure) or low to moderate severity (lower part of figure), occurring in various different seral
stages of a hypothetical mesosere. (Copyright 2002, .P. Kimmins.)

approach of fixed frequency and severity of distur- The conventional clearcutting system on a 60-year
bance (repeated clearcutting and planting on a 60-year rotation with no thinning (line 1) sustains early to
rotation length, line 1; or a short, 30-year rotation midseral conditions with a fairly narrow range of seral
length, line 2) is contrasted with a “soft-touch,” low- variation. The size of this range and the seral stage
severity disturbance regime (line 3) and one possible that is maintained will depend on the rotation length
ENFD scenario (line 4). In this ENFD scenario, a (compare lines 1 and 2), which will also determine
high-severity, stand-replacing disturbance is followed how many phases of stand development (Oliver and
by a period of autogenic stand development and suc- Larson, 1990) occur in this seral stage before the sub-
cession accelerated by periodic lower severity distur- sequent clearcut. The 30-year rotation scenario would
bance (stand density reduction—thinning—followed maintain an earlier seral condition and a lower range
by several partial harvests and underplanting) based on of stand development phases than the 60-year rota-
VR and stand structure management silviculture. tion. If the stands in these two regimes were thinned,
 Ecological Rotation Concept as a Framework forENFD — 525

Early seral Pioneer Climax herb/

Early seral Mid seral Late seral
herb/shrub | hardwood shrub/byrophyte
conifer ' conifer conifer
i} 1 | woodland

ELS are ow Mafeoe

a OO

Conversion to a i}
PT ETT :PES | Conversion to a
i} |
productive tree- i} Mid seral conifer | Late seral conifer productive tree-
dominated seral Early seral conifer dominated seral
Pioneer hardwood F . with clearcutting,
I
with shelterwood,
stage with clearcutting 1
stage
with clearcutting ae ae patch cut, seed tree I patch cut or
| |
i}
or shelterwood i selection
| |
| |
| 1

| i
| |
Protection or
Conversion of even age, early and mid seral stages to conversion to a
uneven age, late seral stage, maintained by selection harvesting. late seral tree stage

$ 4 ps x ty,
f ip < \

ey) | | I | |
i
|
Management-induced disturbance of a late seral stage to initiate a new autogenic successional i}

sequence. Different severities of disturbance initiate the next sequence from a different seral stage. 1

Stand fertilization, thinning and manipulation of species composition accelerate succession. |

en — = —_—— \

I
I
I
1 ' 1
| Acceleration of successional development through stand management. |
dottins | | |
|
1

|
|
| M anagement-induced successional retrogression and acceleration
: through stand management to sustain early (1) or mid to late (2) seral stages.
! ‘
|
|

Allogenic/biogenic disturbance
Autogenic succession
Time SS Ss SE |

Figure 18-4 Diagrammatic summary of a variety of strategies to manage ecosystem disturbance to achieve a wide variety of
successional objectives. (A) Ecosystem disturbance to convert either early seral herb/shrub or late seral herb/shrub woodland to
productive closed forest and disturbance/recovery sequences to maintain particular desired seral stages. (B) Acceleration of autogenic
succession by selective harvesting combined with natural regeneration or underplanting to create late seral stand composition and
structures. (C) Management disturbance of late seral conditions to create different seral stages and management to produce seral stage
sequences on a management rather than successional time scale. (D) Management-induced disturbance/accelerated recovery se-
quences to sustain either early (1) or mid- to late (2) seral sequences. (Copyright 2002, 7.P. Kimmuns.)
526 CHAPTER 18 — Understanding and Emulating Natural Forest Disturbance

{ u t + Sustainable {

owl :
joe 1 Non-sustainable,
es

= Ba ae

5 (b)

Zz } { | { Sustainable {
yr

1 Non-sustainable
ae

Various Combinations
—— Time ———>
(d)

Figure 18-5 Relationship between ecosystem disturbance events (indicated by the arrows),
the frequency of disturbance, and the rate of ecosystem recovery (the slope of the recovery
line). (A-C) Sustainable/nonsustainable trends as a result of variation in frequency, rate of recovery,
and severity, respectively (based on the concept of ecological rotations; Kimmins, 1974). (D) Varia-
tions over time in both severity and frequency of disturbance for three different disturbance regimes
that might represent ENFD. Although the three scenarios differ greatly, they all return to the same
starting condition at the end. (Copyright 2002, 7.P. Kimmins.)

then the range of stand phases within a particular seral the conventional clearcut system, line 2) accelerates
stage represented in each rotation length would in- autogenic succession toward late successional condi-
crease. Without thinning, these regimes would sustain tions. The stand would exhibit many characteristics of
early and midseral species, respectively, in the stand those conditions by the end of the first 60-year period
initiation and stem exclusion phases of stand develop- if stand structure management silviculture were used
ment. The understory reinitiation phase would occur and would sustain a late seral stage thereafter, assum-
only late in the longer rotation unless thinning is un- ing a seed source or underplanting of shade-tolerant,
dertaken to accelerate stand structural development. late seral tree species. This system would support only
The “soft-touch” (frequent entry, low-disturbance mid- to late seral species with a very narrow range of
severity), partial harvest system depicted by line 3 (a seral conditions after the first 60 to 80 years. Most
harvest every 10 years—six entries to one rotation of early and many midseral species would be excluded.
 Ecological Rotation Concept as a Framework for ENFD 527
————
T T

Early seral | Pioneer | Earlyseral ' Midseral '! Late seral H| Climax herb/
herb/shrub '! hardwood ! if
conifer ! I
conifer ! ifer
conifer ie eeu Dyscpeyie
ian

| | | wk ae ae
i i]

i] i} : Q

dy L, ev xe 7 :

1 ie =, 1
979 sey 24 T |! L a at |
| ‘ \ Gy | 1 : ot
Silvicultural 1
ae ! ! mn oor - system
i) 1 i]

\ 1 9
0 le CLR wEeresene ; : SCL ELLs Ra = ! ’ _—

ng OEIl Ye
i}

%,
D
Y

1
Doe
1
i}
1
1
:
\ “8 \ % \
1 1 S 1 1 1
EK TH Yeo ao oo \ ' ! 2 x\ 6 Xx
1
!a ! ‘x !
| 1
Cher
i 1
Se 3

per
ere
TS
|
! . %
a |
5
|
:

woos %, fy —
%, | |
1
; =< Cl
60+ TH ene. yest ——

*. \
| “ey i ;

~ PH oe
i} % 2
\
1 ‘ i}
| i
| &%, \ ay | 1
__ PH | a ee a neo : - } \2 X\ 6 Xx
in \ \ \ f CL | PH
1 1 1
1 yy
1 G— 1 1 1
- PH \
\ \ %
1 ! % i} 1

“ GE
120} CLR ~

I- TH

\2SIG Xx
— TH (iL) 1st

= Jue

OE;
180+ TH
i}

_ PH i\

uaewak 1 \ \2 x\6 x
_ PH i Gui Pe
1 *. | \

| PH ' ®
\ on 1 1
1 & ! :
1 Me, \ |

240 —_____—_—_—
conventional (line 1) and short-rotation
Figure 18-6 Seral sequences in a hypothetical mesosere that might result from
clearcutti ng and planting; a low-disturbance, frequent-entry, par-
(line 2) even-age silviculture with repeated rotations with
on or underplan ting of shade-tole rant species (line 3); and an ENED system that
tial harvest system with natural regenerati
of partial harvesting and incorporates VR
alternates moderate to severe harvest disturbance every 120 years with periods
sustained are far greater for the ENFD sce-
and stand structure management silviculture (line 4). The ranges of seral conditions
l example are probably unreasonably short for many northern
nario than for the others. The time scales shown in this hypothetica
ecosystem to develop through such a sequence of seral stages. ‘The patterns are
forests and shorter than it would take an unmanaged
thought to be valid, however. (Copyright 2002, 7.P. Kimmins.)
528 CHAPTER 18 Understanding and Emulating Natural Forest Disturbance

The ENFD scenario produces the widest range of that “small is always beautiful” (see Kimmins, 1999),
successional conditions and supports the widest range there has been a concomitant rise in popularity of
of species, including early, mid-, and late seral species. small-scale “soft touch,” “continuous forest cover”
It would support the greatest range of forest products forestry. Such small-scale harvest disturbance emu-
to allow for uncertain future market demands; would lates the creation of gaps in the forest canopy by
be the most adaptable in the face of accelerated global small-scale natural disturbance processes. In natural
climate change; and would likely have the lowest dis- forests in which the frequency of large-scale, stand-re-
ease, insect, and soil nutritional problems. placing disturbance is low relative to the longevity of
Figure 18-6 is a hypothetical example only. For the tree species, replacement of trees generally occurs
many northern forests, the time scale suggested in this on a tree-by-tree or tree group-by-tree group basis
figure is unreasonably short, development of such through a process that has been referred to as “gap” or
ecosystems through the seral sequence requiring longer “patch” dynamics. This concept, which originated in
periods even with thinning and underplanting. Howev- research on beech forests in southern England (Watt,
er, the comparison of ranges of seral conditions sup- 1919, 1934, 1947) and on tropical forests (Aubréville,
ported by the different systems are thought to be valid, 1938), has spawned much empirical research on the
and the seral development under silvicultural manage- ecological characteristics of gaps and their dynamics
ment would be much faster than in unmanaged stands. and directly underlies the development of the gap
As noted above, all of the successional management models discussed in Chapter 21. According to this
diagrams presented here have focused on stand struc- concept, canopy gaps that originate from treefalls cre-
ture, species composition, and seral stage. They have ate opportunities for establishment of new trees by re-
not addressed the important issues of landscape pat- leasing the resources utilized by the former canopy
terns of seral stages, the character of disturbance patch dominants. If the individual gaps are large enough to
edges, and the many impacts of the road systems that create an earlier seral microclimate and soil condition,
are an inevitable consequence of forest management. then the forest may be a mosaic of small patches, each
Traditionally, dispersed small-patch harvesting and of which is undergoing its own successional trajectory.
frequent-entry partial harvest systems have been asso- In many situations, the gaps are large enough only to
ciated with a higher density of permanent roads than permit regeneration of shade-tolerant late seral
large-patch and extensive clearcutting systems. Where species. The “patch dynamics” concept is similar to
roads are the source of the major environmental im- the “shifting mosaic” concept of Bormann and Likens
pact of timber harvesting, the benefits of smaller scale, (1979), although the latter also includes the idea of a
dispersed harvesting must be balanced against the uniformity of patch distribution in time and space so
greater environmental costs of more roads, more that there is an overall landscape-level equilibrium of
human access, and more sustained mechanical activity patches (White and Pickett, 1985).
in the forest that normally results from the so-called The formation of canopy gaps of various sizes and
“soft touch,” dispersed cutblock silvicultural systems. the subsequent recruitment and growth of plants in
Helicopter and long-line cable harvesting can reduce the gaps are key processes in autogenic forest succes-
road densities, and new techniques have rendered these sion and have become a major focus of plant popula-
technologies capable of some types of partial harvest- tion and community ecologists. To enter the literature
ing. Although they pose their own set of environmen- on “gap ecology” and “gap dynamics,” see Brokaw and
tal impacts, technological developments in timber Scheiner (1989), Canham et al. (1990), Denslow
harvesting are making the variety of successional man- (1987), Krasny and Whitmore (1992), Lertzman
agement options just examined and ENFD in general (1992), Pickett and White (1985), Runkle (1981,
much more feasible than in the past. The landscape 1992), Spies and Franklin (1989), and Spies et al.
pattern aspects of ENFD are discussed in Chapter 19. (1990). Gap formation is a particularly important
process in forests where large-scale periodic distur-
bance (caused mainly by allogenic processes) is not a
18.6 Small-Scale or Gap Disturbance major factor in determining the succession or is very
infrequent. Gap dynamics also plays a significant role
With the increasing public rejection of visually dis- in the details of change in allogenically or biogenically
tasteful clearcutting and large harvest patch size and driven ecosystems (e.g., those dominated by fire, wind,
the assertion by many in the environmental movement and large-scale insect outbreaks) when the frequency
 Small-Scale or Gap Disturbance 529

of disturbance is low, but it is subordinate to these disturbances (Spies et al., 1990). There cannot be a
larger scale disturbance processes in determining the single theory of gap dynamics that is equally valid for
overall, long-term character of the ecosystem. all forests.
A gap can be defined as the vertical projection Much of the discussion about treefall gaps has fo-
downward of any canopy opening in the forest that ex- cused on light availability as the most important
tends through the vegetation strata down to an aver- change in availability of resources. However, there is
age height of 2 m above ground (Brokaw, 1982). To also a belowground gap that is generally much smaller
account for the additional area that is indirectly af- than the canopy gap because roots extend much fur-
fected by the canopy opening, an expanded gap is de- ther from the stem than branches. This is reflected in
fined as the area formed by the canopy gap plus the the finding that soil nutrient availability in recent gaps
adjacent area described by the stems of the canopy smaller than approximately 300 m’ is not significantly
trees surrounding the canopy gap (Runkle, 1982). The higher than in the adjacent intact stand (Denslow,
response of vegetation after gap formation is closely 1987; Vitousek and Denslow, 1986). More recently,
related to gap size: differences in sizes of gaps result in however, the formation of a “belowground” gap in
differences in species composition over the next gap association with the creation of a canopy gap was re-
cycle (Whitmore, 1989a). For instance, pioneer ported in a Pennsylvanian forest dominated by Acer
species (sensu; Swaine and Whitmore, 1988) only re- saccharum and Prunus serotina (Wilczynski and Pickett,
generate in large gaps because seeds of these species 1993). Less fine-root biomass was found within
generally germinate only in open microclimates, and canopy gaps than within the intact forest, a trend at-
their seedlings do not tolerate shade or root competi- tributed to root death after gap formation that has im-
tion. In contrast, mid- or late-seral (i.e., nonpioneer) plications for availability of soil resources to invading
species are able to establish in the forest shade in small seedlings.
gaps (Whitmore, 1989a). The ability of these species Many models of forest ecosystems are based pri-
to survive in these gaps is related to their ability to tol- marily on light and ignore the belowground availabil-
erate periods of suppression (Canham, 1989). Thus ity of and competition for soil resources. Where soils
late-seral and climate species can be divided into (1) are moist and fertile, light will often be the key limit-
species whose seedlings are very persistent in deep ing factor for regeneration and plant growth, but
shade and whose growth is stimulated with only small where moisture and nutrient availability is less than
increases in light (i.e., in a very small canopy gap) and the total plant demand, this omission of soils may ren-
(2) species whose seedlings will persist only under der the predictions of light-driven successional models
light to moderate shade or will persist in shade only less accurate. As noted above, the much greater exten-
temporarily and whose growth is released only by sub- sion of fine roots below ground than branches above
stantial increases in light (Whitmore, 1989a). ground means that the belowground influence of edge
Although species respond differently to gap cre- trees surrounding a gap will generally extend much
ation depending on their adaptations to resource further into gaps than the microclimatic modifications
availability (e.g., pioneer or climax species), the overall that they cause. Below ground gaps will always be
pattern of gap-phase replacement is largely a function more or less directly beneath the canopy gap, but the
of the interaction of canopy height and latitude with increase in light at ground level will not necessarily
gap size. Canham et al. (1990) compared light pene- mirror the opening of the canopy. At higher latitudes,
tration through idealized single-tree gaps in different light increase at ground level will be displaced pole-
forest types. Canopy gaps characterized by a very high ward into the adjacent stand because of the low angle
ratio of canopy height to gap radius (i.e., gaps with a of the sun in the sky. The southern edge of a gap will
small “view factor,” e.g., in Douglas-fir-hemlock receive much less light than the northern edge at
forests, have little effect on understory light regimes). higher northern latitudes, and vice versa at higher
Gap formation rates and vegetative responses to gaps northern latitudes. Simulation models of gaps may
in these forests are low relative to those of other forest need to represent this degree of spatial separation be-
types (Spies et al., 1990). Clearly, the “gap paradigm” tween the light gap and the soil resources gap.
must be broadened to incorporate details of forest There can be little doubt that even in the absence
structure; the frequency, size, and severity of natural of treefalls, many “closed” forests are heterogeneous
disturbances; within-gap heterogeneity; and the inter- in terms of species composition, height, canopy depth,
action between large catastrophic and small canopy and foliage density. These factors influence under-
530. CHAPTER 18 Understanding and Emulating Natural Forest Disturbance

story composition and subsequent responses to Changes in indigenous peoples and their traditional
treefalls (Veblen, 1992). The gap paradigm, which has practices caused by invasions, war, and colonization by
been formulated as a gap versus nongap dichotomy of other human cultures have resulted in substantial
forests, thus is arbitrary, given the continuum of varia- changes in human disturbance to forests over the past
several thousand years. Forests thus have been continual-
tion in light levels within a forest (Lieberman and ly changing in extent and character. Against this back-
Lieberman, 1989), Nevertheless, it has proved to be a drop of change, societies in many countries have now
useful tool for studying and understanding regenera- decided to try to return forests to some earlier extent and
tion dynamics. condition. However, unless this effort is based on an un-
Large-scale disturbance—stand replacing and derstanding of the relationship of past forests to environ-
landscape-scale disturbance—is discussed in the next mental conditions and disturbance regimes, this effort
chapter. will fail.
Forest management is the management of forest dis-
turbance to achieve a set of management objectives. It in-
cludes both the management of natural disturbance
SUMMARY processes and the management of human interactions
Disturbance is a natural, frequently desirable, and seem- with forests, As a consequence, management must be
ingly inevitable component of forest ecosystems, despite based on an understanding of ecological succession and
the best efforts of humans to prevent it. Both large- and the ecological role of disturbance. Because of the com-
small-scale forest fires, insect outbreaks, wind damage, plexity of the processes involved, the long time scales of
and disease epidemics will continue to occur in forests, these processes, and the frequently large spatial scale
but in many forests, their historical frequency, severity, over which they operate, successful management of suc-
and scale have been changed. Large disturbances will cession and successful emulation of natural forest distur-
probably continue as they are generally beyond our abil- bance require the use of complex (because ecosystems are
ity to influence them. Small disturbances will also occur, complex) EM simulation models. These must be capable
but the large range in disturbance events of intermediate of representing the effects on ecosystem processes of
size and severity that occurred in the past generally do both natural and management-induced disturbance.
not occur in managed forests. Added to this natural dis- They must be based on the conceptual models of ecosys-
turbance is human introduction of diseases, pests, nonna- tem structure, function, and change that have been elab-
tive wildlife, forest harvesting, air pollution, roads, and, orated on in the previous 17 chapters. A discussion of
apparently, accelerated climate change. modeling approaches is presented in Chapter 21.
The most reliable basis on which to re-create and ——
lS

sustain past forest conditions is to understand the ecolog-

ical role of past natural disturbance regimes in forests and
to try to emulate these effects through the combination STUDY QUESTIONS
of anticipated continuing natural disturbance events and 1. Is the death of an individual tree a disturbance?
management-induced disturbance. This must be based
on a sound understanding of succession and the relation- 2. Qver what ume scale should one compare the ecosys-
ship between desired values and the interplay between tem effects of “natural” and management-induced dis-
allogenic and biogenic disturbance and autogenic turbance?
processes of ecosystem development. 3, Is clearcutting similar to stand-replacing natural dis-
turbances?
+. Are the effects of partial harvesting similar to those of
TAKE-HOME MESSAGE surface fires or partial windthrow?
Ss- Which criteria would one use to assess the compara-
The historical character of forests and the values that bility of natural disturbances and management-in-
they provide to human society reflect the historical dis- duced disturbances?
turbance regimes to which they have been exposed. Nei-
6. Would the concept of ecological rotations help to im-
ther this character nor the disturbance regimes have been
constant. Ice ages have come and gone, and warm and plement an ENFD regime?
dry periods have been replaced by cold, wet periods; cli- - How flexible would forest management be under an
mate is always changing. Many high-latitude species are ecological rotation paradigm?
sull changing their geographical ranges in the wake of 8. What are some possible constraints on implementa-
the retreat of glaciers approximately 10,000 years ago. tion of an ENFD regime?
 Spatial Diversity
The Landscape Revisited

D irectly after our discussion of forests as ecosystems, we ex-

amined the description and classification of ecological diversity
across large landscapes (Chapter 6). Ecological diversity is the
physical framework (the ecological stage) within which temporal
diversity (the ecological play—disturbance/ecosystem recovery se-
quences) and the associated biological diversity (the ecological ac-
tors—the species) develop. Ecological site classification captures
both ecological and existing and potential biological diversity. In
Chapter 13 we then examined the variation in biotic communities
along environmental gradients. However, neither of these Chap-
ters addressed the spatial patterns of biotic communities of differ-
ent ages and seral stages across the landscape imposed on the
underlying ecological diversity by disturbance; nor were the im-
plications of these spatial patterns for the functioning of the land-
scape and the individual stands therein considered.
The single chapter in Part VI presents a very brief introduc-
tion to what is becoming a major focus in forest ecology: ecosys-
tem management and landscape ecology. These important and
closely related topics deserve a much more detailed treatment, but
that must be done elsewhere.
 Th

yy

fi
 Ecosystem Management and
Landscape Ecology
The Ultimate Focus in Forest Ecology’

19.1 Introduction scales of individual plant growth, physiological and

ecophysiological functions, between-plant interac-
At the outset of this book, it was asserted that ecosys- tions, soil development, and nutrient cycling. The sci-
tems should be the ultimate focus of ecology and, ence had a definite botanical and plant-centered
therefore, that forest ecosystems should be the ulti- ecosystem focus. When animal—plant interactions
mate context for forest ecology. Furthermore, it was were considered, the larger spatial scales of animal
claimed that forestry should have forest ecology as its population dynamics and interactions were largely left
biophysical foundation. If this argument is accepted, to wildlife managers, entomologists, and zoological
then the biophysical component of forestry should be ecologists. Ecological site classification addresses large
forest ecosystem management. It seems from all the spatial scales as it relates plant geography to climatic,
signs that after decades of preoccupation in ecology topographic, and edaphic diversity, but the classifica-
with individual ecosystem processes and in forestry tions are generally static and do not address larger spa-
with stand-level or estate-level timber management, tial scales in the context of ecosystem function. Only
both forest ecology and forestry are now converging relatively recently has the science of forest ecology ex-
on the scientific basis for and the practice of manage- panded its preoccupation with stand-level ecology to
ment of forest ecosystems, which is as it should be. To embrace patterns and processes operating at much
some, the term forest ecosystem management may seem larger spatial scales—the landscape.
redundant. Forest are ecosystems, and therefore forest Interest in the landscape is not as new in forestry as
management /s ecosystem management. However, the it is in forest ecology. Although the focus of silvicul-
history of emphasis on timber management justifies tural forestry has traditionally been on populations of
the implicit redundancy. trees at the stand level (areas of generally less than 100
In the discussion of ecosystems in the early chap- ha), the overall forestry enterprise has generally been
ters, it was pointed out that the term ecosystem lacks concerned with the management of forested land-
spatial boundaries; it is a complex, functional system scapes. This was certainly true in some of the early de-
of interacting biological and physical components that velopments in forestry in Europe, where forest
is constantly changing over time. This definition ap- management was often as much about maintaining a
plies equally to a few square meters of vegetated soil supply of game species for hunting as about timber
and to a million hectares of forested landscape. Much supply. A sustained flow of a variety of social goods and
of the early research and thinking in forest ecology services, such as timber supply and the social benefits
was focused at relatively small spatial scales—the of employment and wealth creation, cannot be sup-
plied by individual stands. It requires the management
'This chapter was written with the assistance ofEliot McIntire. of relatively large segments of the forest landscape,

Sek
534 CHAPTER 19 Ecosystem Management and Landscape Ecology

and many of the planning tools used by foresters relate bance that were responsible for diversity of wildlife
to entire forest estates—timber supply models that in- habitat and to satisfy the biological and ecological
tegrate up from stand-level growth and yield models, needs of larger and far-ranging species of mammal
streamflow models that integrate local hydrological (e.g., grizzly bears). By 1970, a call had been made for
balances over whole landscape watersheds, and wildlife the management of ecosystems as the basis for public
models that look at habitat supply or models of the land policy in the United States, a change that it was
grazing carrying capacity of forest rangeland across recognized would require a major restructuring of
the landscape. Foresters have used large area maps, government organizations and land tenure systems
aerial photography, satellite images, and the tools of (see historical review in Grumbine, 1994).
geographical information systems (GIS) to support The development of EM in the 1980s and 1990s
this landscape management. However, these have gen- was driven by the development of the science of con-
erally been used as “snapshot” or inventory tools and servation biology and the increasing public and scien-
have less commonly been used to relate to key ecosys- tific concern about loss of species and other biological
tem processes at these larger spatial scales. Many of diversity issues. This has led to a division of the debate
the landscape models were not spatially explicit, and over into EM into (1) those who see it as the founda-
they rarely addressed landscape processes. They had tion for the sustainable management of ecosystems at
more to do with managing landscape inventories of the stand and landscape scale to support a desired flow
timber than managing for the sustainability of multiple of resources, environmental conditions, and services,
resource values at the landscape scale. including both social and environmental values, and
Development of interest in ecosystem manage- (2) those who see EM as the protection of nature
ment has been a major motivating force for the expan- against humans, in which the needs of other species
sion of the spatial domain of forest ecology. It is now (“nonhumans”) should take precedence over the needs
realized that sustainable management for multiple val- of humans, with 50% of all regions being set aside for
ues requires management of whole ecosystems and nonhumans (Noss, 1992, in Grumbine, 1994). This
that this involves landscape units that are large enough reflects the philosophy of the “deep ecology” move-
to provide the manager with the ability to regulate ment (Devall and Sessions, 1985; Fox, 1990). Biodiver-
ecosystem processes that cover extensive areas. ‘This sity conservation and, thus, EM are seen as requiring a
led inevitably to the recognition that many issues in rejection of humanism and anthropocentrism, accep-
forestry, conservation, and sustainability are more re- tance of nature “on its own terms,” and the preven-
lated to the landscape than to individual stands. The tion of human uses that are incompatible with nature
combination of the landscape planning tools from (Keiter and Boyce, 1991; Noss and Cooperrider, 1994)
forestry and landscape ecological models has facili- (refer to the discussion of nature and natural in
tated this spatial transition. Chapter 18 as you evaluate these positions).
This chapter examines what is meant by ecosystem A major contribution to the debate on ecosystem
management and landscape ecology and examines some management is the book of Agee and Johnson (1988).
of the issues involved. It also briefly introduces some of Other recent contributions are Boyce and Haney (1997),
the tools and concepts that are being developed for this Christensen et al. (1996), and D’Eon et al. (2000).
very important emerging area of forest ecology. There are several different lists of the key themes
in ecosystem management. The following is a distilla-
tion of those lists:
19.2. Ecosystem Management
1. Establishment of a clear set of management goals (ob-
As noted in Chapter 18, there are many different in- Jectives) that respect the ecological tolerances and re-
terpretations of ecosystem management (EM). The ori- silience of the ecosystems, species, and values of interest.
gins of the term and the concept are to be found in the These objectives must recognize the role of hu-
discussions of ecologists and conservationists in the mans as part of the ecosystem and respect the val-
1920s and 1930s about the need for natural area re- ues that they desire while respecting the ecological
serves ofsufficient size to conserve wildlife species and roles and needs of other species. Forestry is about
biological diversity in general. It was soon recognized people; management of the interactions among
that to serve their purpose, reserves had to be large 6.3 billion people (headed for 9 to 11 billion) and
enough to encompass the natural processes of distur- forest ecosystems to sustain a desired set of values
 Ecosystem Management 535

and ecosystem conditions. Forest ecosystem man- use commitments; and/or crosses international
agement is about ecosystems: management of boundaries.
their key processes and their interactions with
4. Assign the management offorest ecosystems at both the
their component species, including humans. The
stand and landscape scales to a single agency that has
objectives of ecosystem management may give
the mandate to manage for a balance of values across
greater emphasis to other species; in comparison,
the landscape and the temporal variation in these val-
forest management seeks to balance the needs and
ues in any one stand. The assignment of manage-
desires of people with maintenance of ecosystem
ment responsibility must cover scales that are long
structure and function within historical or desired
enough to plan for and achieve nondeclining pat-
ranges. The most fundamental problem with the
terns of change in ecosystem conditions. This re-
development and application of EM as the tem-
quirement addresses the common institutional
plate for forestry emanates from this fundamental inadequacies that often serve to prevent the appli-
difference in emphasis. Thus, the major hurdle for cation of EM as a management paradigm. Resis-
EM is social, political, philosophical, and eco- tance of governments to change in institutional
nomic rather than biological, ecological, and sci- arrangements and lack of public trust in the man-
entific issues. The challenge is to find the balance agement of multiple forest values by a single
between the two. It is a tough balancing act (Kim- agency or organization result in great difficulties
mins, 1997). in achieving this requirement.
. Manage forests as ecosystems, crossing levels of biologi- . Maintain ecosystem function, seral sequences, and
cal organization and integration. ‘Vhis requires the species composition within historical ranges of variation
management and conservation of genetic, species, (emulation of natural forest disturbance (ENFD);
and biotic community diversity, based on our Chapter 18). This is sometimes referred to as
knowledge of individual species biology and aute- maintaining ecosystem integrity. Public rejection
cology. It involves management of population and of large-scale and/or moderate to severe ecosystem
community ecology and ecosystem function and disturbance, a preoccupation in the public debate
dynamics. Where we lack the knowledge to man- with “small is beautiful,” and the persistent linkage
age for individual species (the fine filter approach) between the aesthetics of stands and landscapes
and where the management for any one species and evaluation of their ecological condition and
conflicts with management of another species, “integrity” (Kimmins, 1999) stand as significant
management to maintain the entire landscape impediments to the attainment of ENFD and the
ecosystem within a historical range of variation historical or even desired range of ecosystem and
that is technically and socially acceptable (the biological variation.
coarse filter approach) is appropriate. EM is a
. Manage adaptively. When appropriate experience is
“systems” approach to management. Lack of in- unavailable or fails to provide a basis for sustain-
stitutional structures and the knowledge needed
able forest management, management should be
to institute this approach and the lack of public conducted as an experiment in which different
trust and acceptance of what must be done to management approaches are tried and their results
achieve EM are major barriers to its successful are monitored and compared with predicted out-
implementation. comes and future management is guided by what
Define management areas ecologically. Areas covered has been learned. This requires that existing
by ecosystem management should be defined eco- knowledge and experience be synthesized into pre-
logically rather than politically and legally. The dictive models that guide the choice of alternative
ecologically defined boundaries should reflect management practices and policies and provide a
major landscape processes, such as watersheds and basis for comparison of the results of monitoring.
the geographical movements of major species. Success with adaptive management in forestry is
Where much of the land is in public ownership, far more dependent on predictive ecosystem mod-
this may be possible. There are generally signifi- els than is the case in fields such as agriculture and
cant legal and political challenges where much of fisheries, where either short-term experience or
the forest landscape is in private ownership; has simple population models may provide the needs
binding long-term, legal resources tenure or land of adaptive management. This is because we often
536 CHAPTER 19 Ecosystem Management and Landscape Ecology

cannot measure the significant long-term conse- The origins of landscape ecology are rooted in
quences of forest management for many decades biogeography (Troll, 1939)—the study of the effects of
or even centuries. the physical environment on broad vegetation pat- ly
sn
a
a

terns (Chapter 6). Early development of the science as

Ecosystem management is a very new field and, in it is today was based on the recognition that land-
common with other new scientific fields, has had to scapes are a mosaic of patches of vegetation of differ-
use generalized theories and concepts in the absence ent types and in different stages of postdisturbance
of appropriate experience and empirical data. Many of development. Sometimes the landscape mosaic is rela-
these theories have been found valid for the systems tively fixed as a result of land use patterns and physical
and circumstances for which they were developed but features—forestry, agriculture, urbanization, water
inadequate for ecosystems and situations that differ bodies, and so on. In other cases, the pattern reflects a
significantly from those that provided the basis for shifting mosaic of disturbances in continuous vegeta-
theory development. Many concepts that were the tion communities. Where sufficiently large areas are
mantra of the early days of EM are now being found considered and where disturbance regimes and vari-
to be in need of modification and adaptation when ap- ability in climatic conditions remain within the histor-
plied more widely in ecologically different forest ical range, the overall character of a landscape mosaic
types. Some are simply wrong in some ecosystems. We may remain constant despite change at the patch
can expect to see a long period on evolution of this level—the shifting steady-state mosaic (Watt, 1947).
concept, with many earlier ideas being modified as However, long-term climate change, postglacial
they are tested and found wanting. This is normal and spread of species ranges, and human-induced spread
to be expected. What will remain difficult in the evo- of diseases and competitive species may mean that
lution of EM is that it is interpreted very differently by some landscape mosaics never do remain constant
groups that hold very different views about the rela- over very long periods (Turner et al., 1993).
tionship of humans and their needs to forest ecosys- The stability of landscapes is an important issue.
tems and “nature.” This fundamental difference, The forests of Canada would have been a lot older
which has echoes of the fundamental differences in during the wetter and cooler periods such as the little
belief systems between different human cultures that ice age (Chapter 8), when there would have been
are the source of ongoing international conflict, can fewer fires, than during the warmer, drier, and more
be expected to make the debate over EM both vigor- fire-prone period that followed. However, vegetation
ous and interesting (Kimmins, 1993a). leaves a biological legacy in terms of soil chemistry,
humus forms, minor vegetation patchiness, and repro-
ductive propagules after many disturbances. This
19.3. Landscape Ecology “ecological fingerprint” on the landscape may result in
the predisturbance landscape pattern being reestab-
The goals of landscape ecology are to describe, under- lished, unless the disturbance was severe enough to
stand the causes of, and interpret the ecological impli- “wipe away the fingerprint.” Postfire boreal forests in
cations that arise from landscape pattern. A major Canada may regenerate to a similar composition and
focus is the search for a link between landscape pattern spatial pattern because of serotinous cones and stump
and landscape function. It considers the development sprouting or root suckering. The roots and rhizomes
and change over time in spatial heterogeneity, the of dense shrub and herb vegetation in canopy gaps fre-
interactions and processes that occur between the quently persist after disturbance. Their resprouting
patches across a heterogeneous landscape, the influ- often prevents tree regeneration, thus perpetuating
ence of spatial pattern on biotic and abiotic processes, the gap. Thus, landscape pattern may sometimes be
and the management of spatial heterogeneity (Turner, much more stable than previously thought, even in
1989). Review of landscape ecology can be found in frequently disturbed landscapes.
D’Eon et al. (2000), Gergel and Truner (2002), Liu The relationship between landscape patterns and
and ‘Taylor (2002), Sanderson and Harris (2000), and landscape processes continues to attract a lot of atten-
Turner et al. (2001). Spatial modeling of landscapes, tion from landscape ecologists. A fine-grain landscape
discussed again in Chapter 21, is reviewed by Mlade- pattern, such as that produced by natural gap forma-
noff and Baker (1999). tion or partial harvesting systems, will have a different
 Fragmentation, Connectivity, and Edges = 537

effect on the dispersal of seeds, spores, diseases, ani- a lot of ecological research to short time and small
mals, the spread of fires, wind damage, and insect epi- spatial scales; and the persistent use of measurement
demics compared with a coarse-grain landscape techniques or sampling strategies that are inconsistent
pattern produced by some fires, insect outbreaks, with the pattern, scale, and temporal dynamics of the
wind, and clearcutting. The different microclimates object or process of study have proved to be a signifi-
and degrees of soil disturbance associated with small- cant impediment to progress in ecology.
scale and large-scale disturbance patterns result in dif- Landscape diversity provides a space-for-time sub-
ferent seral stages and different successional pathways. stitution if one does indeed have a shifting steady-state
These different successional outcomes in turn influ- mosaic. Rather than provide a sequence of forest values,
ence the type and character of subsequent distur- habitat conditions, and biotic communities over time
bances. Different patterns result in different measures from any one ecosystem (the stand-level perspective),
of biological diversity and temporal diversity. Thus, steady-state forest landscapes provide a continuing sup-
landscape pattern is very much linked to process ply of all values at all times in a shifting mosaic. The size
(Pickett and Cadenasso, 1995). Although a landscape of the landscape required to do this relates to the spatial
is in fact simply a mosaic of stands, the effect of land- scale of disturbance and the spatial scale of species biol-
scape pattern on landscape and stand processes means ogy and ecology. The linkage between space and time
that one cannot predict landscape patterns, dynamics, was recognized early in the development of landscape
and processes solely on the basis of understanding the ecology and is one of the foundations for the preoccu-
patterns and processes in all the individual stands that pation with the relationship between spatial pattern and
make up the landscape. the processes of temporal change.
Observation and understanding of the ecological Recognition that there is a level of ecological inte-
consequences of landscape pattern is scale dependent: gration within watersheds that requires their study
it is very sensitive to the scale at which pattern is ex- and management as landscape units and the rise of
amined (Wiens, 1989), as is the case for ecological suc- concern about biological diversity have brought land-
cession and landscape dynamics (Frelich, 2002). The scape ecology to the forefront of the development of
perception of the pattern of patches in the boreal for- forest management. Because maintenance of land-
est gained from a satellite, from a commercial airliner scape biological diversity is related to the maintenance
at 10,000 m elevation, from a 1 to 10,000 scale air of a mosaic of forest ages and seral stages across the
photo, and from an on-the-ground inspection of a landscape in a spatial pattern that matches the ecology
stand varies enormously, as does the understanding of of the species involved, landscape management has be-
the interactions of the processes within and between come a major component of land use planning and de-
patches. What is seen as a repeated pattern and stable, cision making and of studies of human geography
sustainable mosaic on a satellite image may appear as (Turner, 1989). It is becoming a major focus of forest
an incomplete sequence of seral stages on an air pho- resource management.
tograph and as a single and disruptive disturbance
event when standing in a single patch. Processes that
vary at a very local scale, such as those determined by 19.4 Fragmentation,
changes in soil resources along a local topographic Connectivity, and Edges
gradient, may be overlooked in regional comparisons
across different climatic zones. A considerable propor- Forest fragmentation occurs when an area of continu-
tion of ecological theory is based on research con- ous forest cover is broken up into isolated fragments
ducted at a scale of a few square meters to a few by land use changes, such as agriculture or urbaniza-
hectares, and controversy often arises when such theo- tion, which surrounds the remaining patches of for-
ry is applied to ecosystems in which the spatial scale of est—forest fragments—by nonforested ecosystems.
the processes that affect the different species in the The term has also been applied to areas of previously
community varies greatly (Wiens, 1989). The debate continuous forest cover of similar age and condition
between ecosystem classifiers and gradient analyzers that is broken up into patches of younger forest
(the continuum theorists) discussed in Chapters 6 and and different seral stage by forest harvesting (Harris,
13 is sometimes related to issues of scale of sampling. 1984). Such fragmentation reduces the total area of
The failure to address issues of scale; the restriction of the previous forest condition and exposes the
538 CHAPTER 19 Ecosystem Management and Landscape Ecology

organisms in the fragments to an increase in the area geography theory, discussed in Chapter 15, which as-
of ecotone—edge—and a reduction of “forest interior.” serts that a pattern of habitat that has “hard” and per-
It increases the area over which forest influence has manent ecological edges has profound implications for e
eee
ee

been reduced. Species that are adapted to ecotones or species dispersal and equilibrium species richness, has
edge conditions will be favored, whereas those that are often been invoked in a “snapshot” assessment of the
adapted to the forest interior will be negatively af- implications of forest landscape pattern for wildlife and
fected. Concern about fragmentation focuses on the biodiversity without adequate recognition of the dy-
effect of fragmentation on movement of animals be- namic shifting mosaic and the temporary and relatively
tween patches—the issue of connectivity, echoing the is- “soft” edges that exist in a disturbed forest landscape.
sues raised in island biogeography theory—and on the During the 1960 to 1980 period, forest edges were
penetration of nonforest microclimates, edge preda- generally considered a positive feature for wildlife be-
tors, and species of open areas into the forest patch, cause they were more favorable for at least part of the
thereby threatening the species of the previously con- annual habitat needs of several game species than inte-
tiguous forest (Laurance et al., 1998; Murcia, 1995). rior forest or open areas. This encouraged a switch
The issue of connectivity is part of the effect of from large clearcuts to small dispersed patch logging
pattern on process. Movement of animals across a in the U.S. Pacific Northwest. The same harvested
landscape between summer range and winter range, area in small patches has much more ecotone area
thermal cover and feeding areas, breeding and rearing than a single large patch. However, as this patch har-
areas, and so on may require the linkage of different vesting continued across the landscape, with its atten-
components of a species range or habitat by particular dant network of roads, concerns about fragmentation
types of vegetation or seral stages. Depending on the and its effects on nongame organisms developed. It
type, size, and behavior of the animal and the climate was recognized that landscapes harvested in this way
conditions (e.g., heavy snowpack areas versus non- have relatively little forest interior, in contrast to large
snowpack areas), connectivity between habitat and clearcuts that result, several decades later, in extensive
range elements may be critical. Linkage of isolated areas of forest interior. These concerns stimulated a
woodlots by treed strips or hedgerows in an agricul- period of intense interest in the ecological character of
tural landscape or of parks in a city by urban green- disturbance patch edges.
ways or street trees may have a great influence on the Most studies of edges have been short term and
fauna that can be supported. Similarly, the size of have focused on finding simple, general relationships.
clearcut will affect movement and use of the area by They have been “snapshot” assessments that, although
animals that require proximity to standing mature for- valid for the study of permanently isolated forest
est until the harvested area has regrown to an appro- patches in agricultural or urban area, have lacked the
priate stand phase or seral stage. all-important dynamic component needed to evaluate
Legitimate concerns about connectivity have fragmentation issues in forests. Forest edges change
sometime been overgeneralized. For example, in land- over time as forests regrow and as wind and other
scapes where natural disturbance has historically iso- sources of edge tree mortality “soften” the edge. Re-
lated patches of undisturbed vegetation, the local generation of new branches on edge trees and tree re-
fauna, flora, and microbes can be expected to have a generation inside the forest along the edge reduce the
different connectivity requirement than species in penetration of open microclimatic conditions into the
landscapes where natural disturbance has not histori- remnant patch.
cally caused such fragmentation. In some landscapes, The microclimatic character of edges depends on
isolation of forest patches may be important as a way of aspect, slope, cloudiness, and windiness; the height,
limiting the spread of disease and the activities of deciduousness, and crown characteristics of the trees;
predators and in maintaining metapopulation genetic the type of disturbance that created the edge; and at
diversity. Fixed connectivity corridors may become which side of the disturbance patch/remnant patch
predator traps—the predators learn that prey species you are looking (e.g., north or south). The biological
use them to move through the landscape. Creating a characteristics of edges will depend on the type of for-
more diverse and changing series of travel routes est and the native organisms in the area in question.
across the landscape my result in a more sustainable Many studies of edges have documented the character
predator-prey relationship (D’Eon, 2002). Island bio- of the edge without providing an understanding of the
 Patch Size and Event Size Frequency Distributions 539

effects of edges and their change over time on ecolog- forest. They also vary according to the organism in-
ical processes. Many studies have focused on recently volved and their dispersal and movement abilities and
created edges, on short-term microclimate changes, requirements. Some animal and bird species are able
and on increased predation on birds (Lahti, 2001). to cross substantial open areas; others are not. Plant
Some studies have shown little or no edge influence. species with wind dispersal seed and spores will react
Others have revealed significant if temporary effects. very differently to landscape pattern compared with
The substantial discrepancy between edge studies re- species that rely on bird and animal dispersal. For ex-
flects the site- and situation-specific nature of the ample, bats are important in seed dispersal in certain
edges that were studied, and no general theory of tropical humid forests, and their restriction to 50 m
edges has yet emerged (Murcia, 1995). from the forest edge limits regeneration in large dis-
A key variable in the study of edges is distance of turbance patches to plant species with longer distance
edge influence (DEI)—essentially the width of the edge seed dispersal mechanisms.
ecotone. DEI values have been calculated for many Landscape ecology is a very broad and complex
types of northern temperate forest. In a survey of liter- topic. Much remains to be discovered about it. The
ature, McIntire (unpublished) found that the average treatment here is very much introductory.
width of the edge in terms of measurable vegetation
and microclimate was as follows: east, 54 m; north, 29
m, west, +7 m; south, 38 m; undefined aspect, 49 m.
These values depend very much on the height of the
195 Patch size andiventesize
forest. They will be too low for very tall forests and Frequency Distributions
too high for shorter forests. As a result, edge effects
An important aspect of both landscape ecology and
are often expressed as number of tree heights rather
the emulation of natural forest disturbance is the rela-
than actual distances. Estimates of edge distance vary
tive frequency of disturbance “patches” of different
from 1.5 to 4 tree heights (McIntire, unpublished).
sizes (Figure 19-1) and how these disturbance patches
Tree mortality in Amazonian rain forest as a result of
are aggregated into disturbance “events” of various
edge microclimatic effects has been reported to pene-
sizes (Figure 19-2). A disturbance patch is an area over
trate 60 to 100 m into the stand and possibly farther
which the disturbance has operated and has a defin-
(Laumme et al., 1998).
able edge and within which there may or may not be
A final comment on edges is that they are really
remnants of the predisturbance forest condition. A
boundaries, not edges. The word edge has the connota-
disturbance event in this context is an aggregation of
tion of a line, with one condition to one side and an-
patches, separated by strips or patches of undisturbed
other condition on the other. Ecotones are an area of
transition—a gradient—from the open to the closed
forest condition. This gradient involves both the gra-
dient of physical conditions—including light, wind, number of patches |
temperature and humidity—and the gradient of biotic =e = % cover by size t
conditions—live and dead trees, biotic community
structure and composition, litterfall, roots-related
processes, wildlife habitat values, and so on. The
steepness of the gradient depends on the character of
the “edge” and of the remaining forest. We map dis-
turbances as areas within the edge of the disturbance.
In reality, much of a disturbed area is within the eco-
tone or boundary area. These disturbance boundaries Disturbance
of
Number
Patches
Size
Patch
by
Area
Disturbed
of
%
have a distinctive ecological character, have distinctive
successional pathways, and may affect future distur- small —————————SOPatch Size large
bance events differently than the communities that de-
velop to either side of the boundary. Figure 19-1 Generalized frequency distribution of distur-
Fragmentation and connectivity are issues that bance patch sizes and percentage of total disturbed area in
vary considerably in importance in different types of different patch sizes.
540 CHAPTER 19 Ecosystem Management and Landscape Ecology

tively few large disturbances. Different forest land-

Event 1.5000 ha Event 2.1000 ha scapes with different natural disturbance histories will
have different shapes to these curves, but the general
relationship has been reported for many forests.
Forest harvesting in the past has generally pro-
duced a limited range of patch sizes (Figure 19-3).
Partial harvesting silvicultural systems tend to homog-
enize the landscape by producing a narrow distribu-
tion of small disturbance patches. Exploitative
progressive clearcutting (each year’s harvest was locat-
ed adjacent to the previous year’s harvest, resulting in
very large disturbance patches as the entire valley or
Event 3.3000 ha management unit was harvested) has frequently pro-
duced only large disturbance patches. Partially har-
Figure 19-2 Yo emulate large landscape patterns of nat- vested (selection harvesting or “continuous forest
ural disturbance may require a concentration of harvest cover” forestry) and progressive clearcut harvested
patches into “events,” each of which has a patch size fre- landscapes generally lack intermediate-sized distur-
quency distribution. (Based on D.W. Andison, personal communi- bance patches. As size restrictions on harvest patches
cation; SERM, 2002.) were imposed by public pressure, economic consider-
ations, and other factors, the landscape became a pat-
forests, into an area of the landscape that is clearly dif-
tern of medium patches of very constant size and
ferentiated in space from adjacent disturbance events
shape (Figure 19-3). To emulate natural disturbance,
(aggregations of patches). Wildfire, major windstorm,
there is need to create a few large patches if large-scale
or insect epidemics will characteristically move across
natural disturbance no longer does this. We also need
a landscape creating individual patches of disturbance
to create a lot of small disturbance patches if these are
of various sizes and degrees of tree kill, separated by
forest patches over which the disturbance has skipped.
an important feature of the unmanaged landscape and
Each major disturbance event (each major fire, epi-
one wishes to emulate this. Many small natural patch-
demic, or wind storm) is restricted to a region of the es were created by small forest fires that are now put
out or small-scale insect, wind, and disease distur-
landscape defined by the start, spread, and termination
of that disturbance. Events may affect only a few tens bances, which are lost when damaged stands are sal-
or hundreds of hectares, consisting of a single or small vaged to capture the social values of the timber.
number of small- to medium-sized patches. Alterna-
tively, a single event can affect more than a million Conventional

hectares and consist of a large number of disturbance

patches of widely varying size and severity (this deter-
mines the number of retention patches of undisturbed
forest, groups of trees, or individual live trees within
the patch).
Figure 19-1 shows a hypothetical example of the
frequency of disturbance patches of different sizes in a
naturally disturbed forest landscape and the relation-
ship between the patch size frequency and the propor-
tion of the total disturbed area that is accounted for by
different patch sizes. Generally, there are very many
Figure 19-3 Conventional clearcutting has produced a
small disturbance patches and very few very large regular pattern of rectangular openings (disturbance
ones, with a curvilinear decline in frequency of patch- patches) across the landscape. ENFD involves a variety of
es of increasing size. A relatively small proportion of disturbance patch sizes and shapes, with green retention
the total disturbed area is accounted for by small dis- patches or individual trees within larger patches. (Copyright F.P.
turbances, and most of the disturbed area is in a rela- Kimmins, 2002.)
 Take-Home Message 541

19.6 ‘Tools in Landscape Ecology: Large-scale computer models that can encompass
Landscape Models, Remote Sensing, landscapes have become possible only recently
and Gis (Mladenoff and Baker, 1999). Despite advances in the
speed and power of computers, these large landscape
A major reason for the delay in the development of models are still unable to incorporate details of stand-
landscape ecology was the lack of tools with which to level processes that play a vital role in determining
observe, describe, and quantify landscape patterns and landscape-level processes. Meta-modeling, in which
processes. The development of computerized geo- mechanistic stand-level models that incorporate key
graphical data bases—GIS—did much to solve this stand-level processes are “soft-linked” to the large
problem. Development of computer software to ana- landscape models, may be the way to overcome the
lyze spatial patterns presented by GIS gave ecologists problem. This is discussed further in Chapter 21.
a large number of /andscape metrics, or measures of spa-
tial relationships. FRAGSTATS is perhaps the best
known of these (McGarigal and Marks, 1995). It pro- SUMMARY
vides a plethora of measures of spatial relationships. Many of the key aspects in forest stewardship, conserva-
These metrics have been calculated for a wide variety tion, biodiversity, and natural disturbance emulation are
of landscapes and disturbance regimes. However, de- now recognized to be landscape- rather than stand-level
scriptions of the ecological meaning of these metrics issues. Forest ecology has responded to this by strength-
has lagged behind the development of the metrics and ening in the area of landscape ecology. As a new branch
spatial statistics, and the lack of biological and ecolog- of ecology, much remains to be done to understand the
ical interpretations of the metrics has identified an im- processes and the link between pattern and process.
Some of the early theories and ideas are being found to
portant area for research.
be less generalizable than had been hoped, and some
Aerial photography has been an important method have been applied out of context. As this branch of forest
by which to monitor and describe landscape patterns. ecology matures, the incorrect and inapplicable theories
It has been widely used in forestry since its develop- and ideas will be discarded. The use of remote sensing
ment in World War I. However, in landscapes char- and landscape modeling will undoubtedly play an impor-
acterized by very large-scale disturbance events, the tant role in this evolution of understanding. However,
scale of photographs from airplanes has often been too these tools must be linked to our knowledge of smaller
small. Satellite imagery has solved this problem and is scale ecological processes. We must find ways of linking
seeing increasing usage. The linkage of such images to knowledge across scales of space as well as time. To do
large-scale landscape models is a useful development this, our study of landscapes must recognize the dynamic
nature of the shifting landscape mosaic and the probabil-
(Waring and Running, 1998).
ity that many landscapes may change in character over
An example of the power of satellite images in
periods of centuries. The idea of landscape constancy
gaining an understanding of landscape processes is may be as elusive as the concept of stand constancy is
provided by Albani (2002). In examining the question generally inconsistent with our present understanding of
of whether white spruce regeneration in disturbed bo- ecosystems.
real mixed-wood landscapes in western Canada is lim-
ited more by distance to seed source than by site and
topographic features, he used satellite imagery to ana- TAKE-HOME MESSAGE
lyze the distance of regenerating spruce patches from
spruce seed sources and the relationship of both seed Biodiversity, fragmentation, connectivity, landscape pat-
tern, and the linkage of pattern to process are complex is-
sources and regenerating patches to site topographic
sues that are still very much under development. As
features. His results supported the distance-to-seed research proceeds to evaluate correct and incorrect con-
source hypothesis. Field sampling to assess this ques- cepts and theories, forest resource managers and conser-
tion would have been prohibitively expensive because vation biologists must keep an open mind on these issues.
of access issues, the great variability of mixed woods, Where our knowledge is lacking, it is always best to go
the confounding between distance to seed source and back and look at nature. This is the rationale for the
environmental variability, and the vast scale of the bo- coarse filter, emulation of natural disturbance approach
real forest. to biodiversity conservation and maintaining “ecosystem
542 CHAPTER 19 Ecosystem Management and Landscape Ecology

integrity.” We must not be prevented by our preoccupa- STUDY QUESTIONS

tion with local and stand issues and ecology from seeing
the bigger picture at the landscape. Conversely, we must 1.What is a landscape?
not lose our understanding of stand-level ecosystem 2. What is a gap?
processes by focusing only on landscape pattern. Both are
3. What is ecosystem management?
important. Landscapes are mosaics of stands, so we must
learn how to scale up from our stand knowledge to the 4. How can we use stand-level knowledge to understand
landscape. However, we must also recognize that there landscapes?
are landscape processes that arise from the interactions of 5. Why is scale important?
local processes that make landscapes more than just the
sum ofits stand-level parts.
Application of Ecological Knowledge in
the Management of Forest Ecosystems

lee are thinking animals; therefore, knowledge has a value for

human societies that is independent of its practical application. However, the
value of knowledge is much greater if it can be put to use in improving the
human condition, in understanding how to conduct our activities without de-
grading the environment, and in learning how to conserve the evolutionary
legacy of biological diversity.
Part VII examines the ways in which principles of forest ecology (the infor-
mation contained in Parts IH to VI) can be used to increase the success of deal-
ing with complex issues in both renewable-resource management and the
conservation of biological diversity and environmental conditions and ser-
vices. Ecological classification, already described in Chapter 6, uses knowl-
edge of ecosystem structure and function to identify and describe different
types of ecosystem, each of which has a characteristic response to manage-
ment. Identification of ecosystem types greatly improves our ability to predict
the response of our forests to management and to make more sustainable pol-
icy and management practice decisions. A second application of such knowl-
edge is in analysis of environmental issues in forestry (Chapter 20). By
undertaking an ecological analysis of issues that may have a strong social and
political component, as well as issues that are largely biophysical, we are much
more likely to find the basis for a solution to them. A final application of eco-
logical knowledge is in the development of conceptual and computer models
of forest ecosystems and the impacts of management thereon (Chapter 21).
Models provide a mechanism by which to organize our knowledge about a
particular ecosystem type and use it to make predictions about the future con-
dition of that ecosystem under alternative management regimes. Chapter 22
considers the issues of renewability and sustainability in forestry, and the book
ends in Chapter 23 with a philosophical retrospective on the link between the
ethical responsibilities of both forestry and environmentalism, and the appli-
cation of knowledge of forest ecosystem ecology.
543
 ATGNORE Ss
> ' .

Ol
. a , - av pie

10 tsi ‘sare oe

° ‘rofl =
i pic roan wegtviath,

a Ti chien
Sngape |
=#.

F d =
ea .
> ti) ithe oe
_
— ) Pa a | fon :
=
Ties a, Paw,
get — "
NAO
> T ea
oi | PY ths a 7 ;
> | rhe 0
’ Ties ier ~ :
Seer ae
ie Fay

ViGAD
5 ds ' of eh:
> MP4 ied Yi :
- : 7. ony!reer -
ones sid eee
=ae% ef ‘ ohh ir) jratly Mi ;
5 valerst
(Pa as Waele ‘-
as xl Athen 2 i
oa 9 < raged emalag
rPuN iin yyy
Wis <1) noted ps ie
 Environmental Issues
in Forestry
The Role of Ecology in the Management
of the Ecological Play
“A thing is right when it tends to preserve the integrity, stability and beauty of the bi-
otic community. It is wrong when it tends otherwise.”
but
“The evolution of a land ethic is an intellectual as well as emotional process. Conserva-
tion 1s paved with good intentions which prove to be futile, or even dangerous, because
they are devoid of critical understanding either of the land, or of economic land use.”
ALDO LEOPOLD, “THE LAND ETHIC,” INA SAND COUNTY ALMANAC (1966)

20.1 Introduction society, the same ecologist will almost certainly con-
sider these to be negative and unacceptable distur-
Forest ecology is a science. It is concerned with the bance events, because the sum of the ecological
form and functioning of forest ecosystems. As a sci- conditions that result will be judged to have lower
ence, it makes no judgments about the “goodness” or value for humans and certain other species than the
“badness” of any particular ecosystem condition, predisturbance conditions. What, then, is the role of
ecosystem disturbance, or pattern of ecosystem the science of ecology in resolving environmental is-
change. It recognizes that forest ecosystems vary in sues in forestry?
their ecological characteristics from place to place and We begin this penultimate chapter with a discus-
that the forest in any particular place is continually sion of the interface between ecology and environ-
changing—sometimes rapidly, sometimes slowly, but mentalism. We then briefly explore two of the current
nevertheless always changing. environmental issues in forestry to show the need for
For many people, science is not enough. In its dis- scientific analysis as the foundation for policy and reg-
passionate analysis of the way things are and of what ulation: the concepts of forest ecosystem health and
they might become, human values and preferences are integrity and the relationship of forests and forestry to
essentially excluded. Outside the ivory towers of aca- atmospheric carbon budgets and the threat of climate
demia and basic research, science is confronted with change. A broader discussion of environmental issues
the value systems and belief systems of contemporary is given in Freedman (1994) and in Kimmins (1997).
society that are often fundamentally different from the
value and belief systems of the scientist. As a scientist,
an ecologist may describe and attempt to predict the 20.2 Environmental Issues in Forestry
ecological consequences of “pollution,” “overlogging,”
“overfishing,” loss of endangered species, and soil ero- Faced with a 2003 population of approximately 6.3
sion without making judgments about whether these billion and the prospect of another 3 billion people
things should or should not happen. As a member of by the end of the 21st century, humans must learn

Shay
546 CHAPTER 20 Environmental Issues in Forestry

how to live sustainably with their forest environ- and of the ecology of the values that we wish to sus-
ment. To do so, decisions must be made, based on tain. They will also fail if they have social and eco-
current value systems, about which kind of forest en- nomic consequences that the voting public is not
vironment we want to sustain. Science should then ready to accept or if they fail to recognize the institu-
be used to understand, predict, and formulate strate- tional and policy issues that lie behind so many envi-
gies by which to achieve our social and environmen- ronmental problems in forestry (e.g., Adamowicz et
tal objectives. Unfortunately, there is frequently ER Syi
confusion over these two functions. Ecologists are Belief systems about nature (“green religion”;
often expected to be arbitrators of the morality of Kimmins, 1993a) play as important a role in the polit-
human-induced environmental change, and _belief ical phase of change in conservation policy and forest
systems about the environment are frequently substi- management as religion plays in establishing moral
tuted for ecological science in the design of environ- values, behavior systems, and ethical standards in
mental management. This confusion must be other aspects of society. However, historical experi-
resolved if the science of ecology is to make its full ence generally shows that fundamentalist belief sys-
contribution to the achievement of sustainable use tems are not appropriate as the sole basis for
and conservation of forests. This point is made elo- organizing and governing a society. Government that
quently by Botkin (1990, 1995). is based solely thereon seems to lead to intolerance,
During the 1980s and 1990s, there was a profound inflexibility, conflict, and violence. This experience
change in public interest in and concern about the has been the basis for separating state from church and
world’s forests. This revolution in attitudes must be secular from spiritual matters in the governance of ~
credited largely to the efforts of the “environmental many countries. Belief systems contribute importantly
movement.” Many scientists and foresters expressed to setting objectives in forest management and conser-
concern about the condition and management of vation, but it is actual experience of forests and
forests from time to time over the past century,’ but forestry and the critical analysis of biophysical and so-
recognition of these concerns by politicians awaited ciological facts that provide the most effective basis for
the bold statements and often dramatic actions of ac- achieving one’s objectives.
tivists in “environmental nongovernment organiza- The sustainable relationship that humans must de-
tions” (ENGOs). velop with the world’s forests must be based on a re-
ENGOs created public and political awareness of spect for nature (Kimmins, 2000). There are two
environmental problems associated with forestry. By interpretations of the word “respect”: (1) to revere, es-
establishing the possibility for change, those individu- teem, and be deferential toward and (2) to notice with at-
als and organizations have successfully executed the tention and to pay due regard to the characteristics of the
essential first stage in the process of improving forest object of respect. Sustainable forestry must be based on
management and conservation. To do this, they made both aspects of respect. The former is important in
use of the media by simplifying complex environmen- defining the objectives of management, and the latter
tal issues to the level at which the media would report is important for the design of policy and practice that
them. This strategy is both effective and often neces- will achieve these objectives.
sary at the “political” stage of conservation and the We are an emotional species that depends heav-
improvement of forest management. Unfortunately, it ily on our visual senses. This can lead to confusion
is frequently unsuitable as a basis for designing the de- between the aesthetics of an ecosystem and its eco-
tails of the changes that will actually achieve particular logical condition, as was noted by Aldo Leopold in
management and conservation goals. The regulations, The Land Ethic (Kimmins, 2000b; Sheppard and
policies, and laws that determine how forests are to be Harshaw, 2001). This raises the question, “Are beau-
managed will not succeed in achieving their objectives tiful stands and landscapes always sustainable, are
if they are based on rhetoric that fails to reflect our sustainable stands and landscapes always beautiful,
current understanding of how ecosystems function and is small-scale, low-severity disturbance always
more appropriate than larger scale, more severe dis-
turbance?” (Kimmins, 1999). The following discus-
‘Concerns expressed at turn-of-the-century forestry conferences read much sion should help to illuminate some aspects of this
like concerns expressed at environmental conferences in the 1990s. question.
 Forest and Ecosystem Health = 547

20.3 Forest and Ecosystem Health? — . Trees are vigorous and productive.
2. There is rapid biomass accumulation and nutri-
It is perhaps inevitable that as the public became in- ent cycling.
creasingly concerned about the environment they
3. Insects and diseases have little negative effect on
would use human health as a metaphor for ecosystem
the development of tree populations and plant
condition. Overwhelmed by visual images of defor-
communities.
ested, eroding, and/or drought-ravaged landscapes
and the accompanying scenes of human starvation and 4. Processes of ecological succession and stand devel-
illness, the concept of “dying ecosystems” became opment are proceeding at rates that are characteristic
firmly entrenched in the minds of the public. Scenes of the current seral stage of the particular ecosystem.
of polluted air and poisoned rivers and lakes, together 5. The ecosystem has high “vitality” and is able to
with pictures of dead and dying plants and animals, withstand stress.
made the idea of “sick ecosystems” easy to embrace.
As the media increasingly portrayed “sick” and The utilitarian definition of forest health empha-
“dying” ecosystems as being worthy of the same con- sizes noninterference with management objectives
cern as sick and dying people, the question of forest (Wagner, 1994). For example, forest health is a condi-
and ecosystem health became an important issue in tion in which biotic and abiotic influences on the forest
the evolving debate about the environment. Forest (i.e., pests, silvicultural treatments, and harvesting
legislation in Canada (e.g., the 1993 Forest Practices practices) do not threaten resource management objec-
Code of British Columbia) and the United States (ref- tives now or in the future (USDA forest service, 1993).
erences in Kolb et al., 1994) now requires that main- This view is based on the analogy between a
taining forest health be a fundamental responsibility healthy organism (e.g., a human) and a healthy forest
of forestry. European governments, concerned about or healthy ecosystem—one that is free of diseases and
the impact of acid rain and air pollution on forests, parasites that deform, deface, and/or reduce vigor,
have invested large sums of money in research to pro- productivity, and resistance to other diseases or
vide technical explanations for and solutions to exam- stresses. It is a vision of a healthy forest as an early or
midseral forest or a young, late-seral stage forest or
ples of forest decline (e.g., Innes, 1993). In many
countries, maintaining ecosystem health has become a any healthy, young productive forest.
legal requirement and a major preoccupation in the In contrast to the traditional and utilitarian para-
digm of forest health, the “environmental” literature
debate over the environment (e.g., Constanza et al.,
1992; Covington and DeBano, 1994; Everett, 1994,
suggests that a healthy forest is one in which
Johnson et al., 1994; Joseph et al., 1991; Monning and 1. The forest has reached its peak of development: a
Byler, 1992; SAF, 1993; Sampson and Adams, 1994; late seral stage or climax condition.
USDA Forest Service, 1993). Ecosystem health has
2. The canopy is multilayered with many large gaps
become a major preoccupation in the new field of
and the trees are uneven-aged.
landscape ecology (Chapter 19).
3. There are many large, live trees, some of which are
broken and/or partly decayed; numerous large
Definitions dead trees (snags); and an abundance of large de-
caying logs; these collectively provide habitat for
Attitudes toward ecosystem health have been catego- the animals that use these “structures.”
rized as either utilitarian or holistic (Kolb et al., 1994;
4. Rare and “specialist” species, which are consid-
Wagner, 1994). In fact, there are three distinct ap-
ered to be indicators of a healthy forest ecosystem,
proaches to the definition of ecosystem health: utili-
are present.
tarian, “environmental,” and ecosystem-centered.
The utilitarian view is essentially the traditional Implicit in these ideas is the vision of ahealthy for-
view of a healthy forest: est as a mature or old forest, a late-seral or climax
stage, or an old-growth phase of any particular seral
stage that has developed in the long-term absence of
Based on Kimmins (1995a,b). significant ecosystem disturbance.
548 CHAPTER 20 Environmental Issues in Forestry

Within the scientific and conservation community, 2. Resistance to catastrophic change and/or the abil-
there have been several definitions that reflect an ity to recover from catastrophic change at the
ecosystem-centered view: stand and landscape scales
3. A functional equilibrium between supply and de-
1. Ecosystem health is “the capacity of the land for mand of essential resources (water, nutrients, light,
self-renewal” (Leopold, 1949). growing space) for major portions of the vegetation
2. “A healthy forest is one that is resilient to changes 4. A diversity of seral stages and stand structures that
and characterized by tree species and landscape di- provides habitat for the historical list of native
versity that provides sustained habitat for fish, species and all essential ecosystem processes
wildlife, and humans” (Joseph et al., 1991).
3. “A healthy forest is a... productive, resilient, and This view of ecosystem health echoes the ideas of
Leopold (1949), Botkin (1990), and many others: that
diverse ecosystem: a forest with a future” (Wilson,
1991, in USDA Forest Service, 1993).
ecosystems are repeatedly being disturbed and are re-
peatedly changing in response to this disturbance.
4. “A forest in good health is a fully functional com- This disturbance may or may not constitute a threat to
munity of plants and animals and their physical en- or cause a change in ecosystem health, which is, there-
vironment. A healthy forest is an ecosystem in fore, not a steady-state or equilibrium concept.
balance” (Monning and Byler, 1992). In an attempt to summarize and simplify this diver-
sity of definitions of ecosystem health, I propose the fol-
Neither the traditional timber management nor lowing definition. A stand-level forest is healthy when
the “environmental” views of forest health address ad-
equately the characteristic variation in ecosystem 1. The stand-level structure, species composition,
structure and function, in tree/parasite or tree/disease ecosystem processes, and patterns of change there-
relationships, and between different seral stages or be- in all are within the historical range exhibited by
tween different forest ecosystem types across the land- that ecosystem over temporal sequences of seral
scape. Implicit in both views is the suggestion that for stages that are characteristic for that ecosystem.
a forest to be healthy, a similar “health” condition 2. The landscape pattern of forest ages and seral
should exist in all ecosystems across the landscape. As stages and the temporal changes in that pattern are
a result, neither of them allows for the development of within the range that is characteristic for that land-
management systems that recognize a diversity of re- scape and to which the biota are adapted.
source values whose relationship to vigor or disease
and “pests” can vary greatly. As cautioned by Botkin If one were managing a forest to sustain “old
(1990), paradigms of nature must truly reflect the way growth” conditions, then dead, diseased, broken, and
nature is and how nature varies from time to time and damaged trees; lots of canopy gaps; incomplete stock-
place to place and must not be based on some precon- ing; and lots of decaying logs might constitute a
ceived notion about what nature should be like. This healthy ecosystem condition but only if this condition
echoes Leopold’s warning presented above. were characteristic of the late seral stage of that forest
By combining our understanding of ecological ecosystem. In contrast, if one were managing that
succession, ecosystem function, the physical environ- same ecosystem as a productive, even-aged, second-
ment, and population and community ecology, we can growth forest mainly for timber, then the presence of
develop a definition of ecosystem health that reflects the very same features would indicate a forest in poor
more closely the variation of ecosystem conditions and health. These characteristics are not normally found
characteristics that exist across the landscape. Using in a young, “healthy” midseral forest. If one were
this approach, Kolb et al. (1994) proposed that a managing this second-growth forest to maintain some
healthy forest ecosystem should have the following late-seral (i.e., “old growth”) wildlife habitat values as
characteristics: well as the midseral timber values, then the presence
of some snags and a few misshapen and diseased trees
1. The physical environment, biotic resources, and would be entirely consistent with the objectives of
trophic networks needed to support productive management, and the stand condition might be con-
forests during at least some seral stage sidered healthy. Any of these three conditions can
 Forest and Ecosystem Health BaD

occur naturally in a particular ecosystem according to ate an imbalance in the historic disease—host relation-
the seral stage it is in and its disturbance history; any ship. The result is an incidence of disease much high-
one of them may be considered “healthy” within one er than that found in most naturally disturbed,
of the three different management objectives. unmanaged forests but possibly similar to the inci-
This view of forest ecosystem health allows that a dence in similar ecosystems that have escaped distur-
wide variety of ecosystem productivity, vigor, and bance for a long time. Examples of this can be found
host/parasite or host/pathogen relationships can be in the mistletoe-lodgepole pine and mistletoe—west-
consistent with the concept of a healthy ecosystem if ern hemlock relationships in western Canada (Unger,
those conditions/relationships are within the range 1992), described briefly in Chapter 21.
that is characteristic for the particular ecosystem type A major difficulty with the forest health issue is
in question. Each seral stage of a particular forest that it is strongly influenced by the human health per-
ecosystem type is characterized by different “health” spective, which focuses on the individual human.
conditions, just as each age class of humans—infants, Ecosystems and individuals both are true levels of bio-
adolescents, adults, and seniors—experiences a differ- logical integration (Rowe, 1961); consequently, these
ent set of health challenges. two levels of biological organization share some com-
mon characteristics. However, there are also funda-
Discussion mental differences that reduce the accuracy of the
analogy between stand-level ecosystems as “supraor-
In many forest types, forest pathogens play an impor-
ganisms” and individual organisms. Many aspects of
tant role in stand thinning and the creation of habitat
ecosystem health relate to landscape pattern and the
for wildlife; a certain level of disease-related mortality
ranges of ecological conditions across the landscape,
in these forests is consistent with a “healthy” condi- rather than individual stands. There are few useful
tion. However, epidemics of insects or diseases, and
parallels between the individual human health issue
species composition, stand structure, and stand den- and forest health at the landscape spatial scale.
sity that are outside the “normal” condition for a par- There is increasing recognition that the landscape
ticular ecosystem could be considered “unhealthy.” scale is ultimately more important than the stand scale
Previously fire-dominated forests on the rain shadow in defining the critical environmental issues in
side of the Cascade mountains in Oregon and Wash- forestry. The rate and landscape pattern of forest har-
ington that have been protected from fire for many vesting, the proportion of a watershed that is in a “hy-
decades have developed stand structure, density, and drologically altered” condition, the age-class structure
species composition characteristics that render them of forests across the landscape, and the spatial pattern
highly susceptible to fire and insects and that cause and patch size of forests of different seral stages at the
“gnnatural” levels of moisture competition (e.g., landscape scale are emerging as more important con-
Everett, 1994; Johnson et al., 1994; Wickman, 1992). servation issues than the ecological condition of any
These forests would have to be considered unhealthy. one local ecosystem, important though this may be at
Their condition is not within the range normally ex- the stand-level scale.
hibited by the forests of that region because once the As noted in Chapter 19, the science of landscape
forest reaches this condition, it has historically been ecology is very young, and there is as yet no well-
removed by fire and/or insects, returning it to a established body of theory. There is great uncertainty
“healthy” condition. about the generality of issues such as fragmentation
Managing naturally even-aged forests in which and connectivity (Harris, 1984). Although these issues
diseases have historically been controlled by large- are undoubtedly of importance for some species in
scale, stand-replacing disturbances (e.g., fire) as some forests, much remains to be discovered about
uneven-aged forests by partial harvesting can also cre- how applicable the principles of island biogeography
theory are in disturbance-driven forests that may be
naturally fragmented from time to time. There is as
‘Note that in some forests, stand-replacing insect infestations are a normal
part ofthe ecology. Although a forest that is killed by insects may be consid-
yet no agreed-on paradigm (no implicit theory or con-
ered “unhealthy” by some people when considered over a short time scale, ceptual framework) for the health of a landscape. Is a
over larger time and space scales, such events may be part of the normal landscape that is almost entirely an even-aged mono-
functioning of that ecosystem and therefore “healthy.” culture resulting from a large stand-replacing wildfire
550 CHAPTER 20 Environmental Issues in Forestry

more or less healthy than a landscape covered by a low With growing public concern about the environ-
tree species diversity climax forest, and is either of ment, many people have come to think that any dis-
these more or less healthy than a landscape with a mo- turbance that creates visual ugliness, that disturbs the
saic of stand ages, seral stages, variable species compo- “perfect condition” of a beautiful old-growth forest,
sition, and species richness? Any of these conditions or that seems to have violated the “completeness” of —
can occur naturally in a given landscape under differ- the forest ecosystem has damaged ecosystem integrity
ent natural disturbance regimes. They will be associ- (Sheppard and Harshaw, 2000). Partially harvested
ated with different wildlife habitat conditions, they will forests with little or no soil disturbance look complete,
have different measures of biological diversity, and whole, and unbroken, in contrast to clearcut areas or
they will have different levels of ecosystem and eco- partially harvested areas with soil disturbance. The
nomic productivity, but do these different landscape latter, by comparison, look broken, violated, and in-
conditions differ in their “health” and “integrity”? complete, as though their integrity has been impaired.
This is not a question that science can answer out- A frequent consequence of these responses to visual
side the context of what society wants the forest land- stimuli is a rejection of ecosystem disturbance on the
scape to be (Kimmins, 2002c). There is no basis for basis that it destroys or damages ecosystem integrity:
resolving this question based on the science of ecology ergo, ecosystems should be protected from distur-
alone. However, ecological science can evaluate whether bance, their “delicate integrity” conserved.
the landscape condition is within the range that has oc- As was the case with ecosystem health, there is leg-
curred historically in that landscape and whether the islation in several countries that requires “mainte-
landscape-level ecological processes are within historical nance of ecological integrity” as well as maintenance
rates. Where natural or human-caused disturbance has of ecosystem health and biodiversity (see references in
taken the landscape out of its historical range but the re- Woodley et al., 1993). As with ecosystem health, such
sulting condition is not a threat to the existence of any legislation requires a clear and accurate understanding
species or important local genetic populations of a of what ecological and ecosystem integrity is before it
species and has not impaired the function of any key can achieve its intended objectives (Kay, 1993).
ecological processes, it is society, not science, that must
define whether the new landscape condition is more or
less healthy and has more or less integrity than some Definition
former condition that was within the historical range. There is a curious redundancy in the idea of “ecosys-
The threat of climate change poses the risk that tem integrity.” An ecosystem is defined as an ecological
many stands and landscapes may become unhealthy system that exhibits the criteria of structure, function,
because they are no longer adapted to the climate and complexity, interconnectedness, and change over time.
the associated disturbance regimes. Major human- By definition, it is a complex, structural, functional sys-
caused climate change is one of the major threats to tem in which the parts and processes are highly linked
the health of global forests. and that changes over time. By definition, it has in-
tegrity as a dynamic, resilient system. If an ecological
system is a functioning ecosystem, then it has integrity.
20.4 Ecosystem Integrity A particular landscape unit will retain its integrity as
long as its structure, function, complexity, interactions,
Integrity is the state of being complete, unbroken, and pattern and rate of change over time are consistent
unimpaired: a perfect condition, wholeness, entireness with the historical range in these variables. The follow-
(Webster’s New 20th Century Dictionary, 2nd Edi- ing definition is based on these thoughts.
tion, 1978). A forest ecosystem has integrity if its structure and
Images of eroding hillsides, rivers choked with species composition, the rate of its ecological process-
logging debris, river beds filled with sand and silt from es, and its ability to resist change in the face of distur-
landslides, and rocky hillsides from which organic bance or stress are within the characteristic range
matter and soil have been removed by fire or over- exhibited historically by that ecosystem.
grazing are the antipodes of our feelings about the “in- The historic reference period is that period during
tegrity” of an ecosystem or landscape. Recently which regional climate has not changed enough to
clearcut forest areas often elicit the same reaction. cause a significant directional change in ecosystem po-
 Ecosystem Integrity 551

tential or processes of change: it is developing by auto- ecosystem has not been damaged. Thus, the integrity
genic successional processes toward a later succession- of a particular seral stage is a very different question
al condition at rates and by pathways that are from the integrity of the ecosystem of which that seral
characteristic of that ecosystem, or the ecosystem is al- stage is merely one of several alternative conditions.
ready at a late seral stage. A particular seral stage of a particular ecosystem
This definition of ecosystem integrity views has integrity if its structure, species composition, eco-
ecosystems as continually changing. “Integrity” can- logical condition, and process rates are within the
not be defined by “snapshot” evaluations or be based characteristic range exhibited historically by that seral
on steady-state or equilibrium concepts. stage of that particular type of ecosystem.
One frequently hears the assertion that “clearcut The historical reference period is the same as for
harvesting destroys the integrity of forest ecosystems ecosystem integrity.
and of old-growth forests,” and the corollary that “a for- The integrity of a particular seral stage of a partic-
est that regrows after clearcutting is not a forest ecosys- ular forest ecosystem type is lost as the processes of
tem,” because the integrity of the ecosystem has been succession or disturbance-driven successional acceler-
destroyed. If forest harvesting is conducted in a way that ation or retrogression lead to the replacement of that
impairs the processes of ecosystem recovery from dis- stage by another stage. This change in successional
turbance, then the integrity of that forest ecosystem has conditions does not threaten the integrity of the
indeed been compromised. However, if the harvesting ecosystem if the condition of the new seral stage satis-
simply converts the ecosystem condition from one seral fies the criteria of ecosystem and seral stage integrity.
stage or condition to another without impairing the If the objective of management is to maintain the
processes of ecological succession, then its integrity as conditions of a particular seral stage, then manage-
an ecosystem has not been damaged. If the harvesting ment must be limited to those practices that maintain
alters ecosystem conditions to beyond the historical the ecosystem within the historical range of conditions
range, then the integrity of the ecosystem will have been observed for that seral stage. If, however, the objective
altered and a different type of ecosystem created. It may is to move the ecosystem to and then maintain it in a
convert a forest ecosystem temporarily to a nonforest different seral stage or to allow the ecosystem to pass
ecosystem. The integrity of a forest ecosystem is re- through a series of stages between successive harvest-
placed by the integrity of a nonforest ecosystem. related disturbances, then a different set of manage-
ment regimes would be consistent with maintaining
ecosystem integrity (Figure 18-6).
Ecosystem Integrity Versus
Seral Stage Integrity
Much of the debate over ecosystem integrity and about
Integrity of the Forest Landscape
clearcutting results from the confusion between ecosys- The focus on linking pattern with process in landscape
tem integrity and the integrity ofa particular seral stage. ecology (Chapter 19) has led to the idea that landscape
Clearcutting destroys the integrity of the old integrity can be linked to landscape pattern and the
growth condition and may destroy the climax/late mosaic of stand ages and seral stages. Some ecologists
seral stage of west coast old-growth forests (or any and many ENGOs have asserted that a landscape lacks
other late-seral or climax forest). Wildfire or other integrity unless it has a particular ratio of stand ages
stand-replacing natural disturbance similarly destroys and seral stages arranged in a pattern that provides
the integrity of this ecosystem condition by creating connectivity and minimizes fragmentation. Research
an earlier seral stage. However, unless poor road loca- on the nature of disturbance patch edges (or bound-
tion, construction, maintenance, and/or debuilding aries) has noted the frequent difference between an
cause slope instability and soil erosion, or where loss edge created by timber harvesting and edges created
of root strength causes slope instability, or where in- by natural disturbances, and there is discussion as to
sufficient soil disturbance permits ericaceous or other whether a landscape that lacks natural edges will func-
shrubs to dominate the site after logging, the prompt tion the same way as one with such edges.
regeneration that normally occurs in these west coast Because natural edges are highly variable in char-
forests after timber harvesting ensures a rapid return acter and change over time, no single generalization
to forest conditions. The integrity of the forest about natural disturbance edges has emerged and is
552 CHAPTER 20 Environmental Issues in Forestry

unlikely to do so. In contrast, human-caused edges are curs between periodic, large-scale, allogenic, or bio-
often relatively “hard” and less variable—the bound- genic stand-replacing disturbances (e.g., Figure 18-6).
ary is frequently abrupt, and the ecotones are rela- In late seral or climax forests, small-scale disturbance
tively narrow. However, they change and often tends to maintain the integrity of both the seral stage
“soften” with time, leading to a degree of convergence and the ecosystem, but in midseral forests, gap forma-
between natural and human-caused edges. Thus, the tion may be the mechanism by which one seral stage is
question as to whether landscapes with natural or replaced by another. Gap formation in these forests
human-caused edges will function the same way and maintains the integrity of the ecosystem but not that of
have similar “integrity” is time dependent. particular midseral stages. Larger-scale disturbance
Landscape patterns created by the history of natural generally maintains ecosystem integrity but destroys
disturbance are often considered temporary. Future dis- the integrity of the predisturbance seral stage by accel-
turbances will create new patterns. Only when sufficient- erating change to the next seral stage or regressing the
ly large landscapes are considered can one find relatively forest to an earlier stage (Figure 18-3). However, se-
stable ratios of age classes, seral stages, and disturbance vere large-scale disturbance sometimes damages the
patch sizes. It is a shifting mosaic, the character of which integrity of a forest ecosystem, creating early seral herb
is stable only above some critical landscape size. If this in- or shrub seral stages until successional recovery has re-
terpretation is correct and if landscape integrity is related stored a forest seral stage.
to constancy of these ratios, then the question of land- A significant area of British Columbia (the wetter
scape integrity is spatial scale dependent. subzones of the Coastal Western Hemlock zone, many
Contrary to the shifting mosaic concept, pattern in areas of the Black and White Boreal Zone, humid
some landscapes may be relatively permanent. Insect, areas in the Engelmann Spruce-Subalpine Fir zone,
fire, and wind disturbances are often strongly influ- and parts of the Interior Cedar Hemlock zone) has
enced by the structure, species composition, physical closed-forest communities only because of periodic
site features, and certain functional aspects of the for- large-scale, stand-replacing disturbances such as fire
est (e.g., decomposition rates that determine fuel load- (mainly), insects (commonly in the interior), and wind
ing), causing subsequent disturbances to create similar (less common but the major factor in some humid
patterns and edges. The biological legacies of organic coastal areas). By reducing shrub and herb competi-
matter, soil fertility, and reproductive propagules may tion, providing suitable seedbed conditions for trees,
result in postdisturbance development of a similar for- and improving soil physical and chemical conditions
est, thereby increasing the potential for a repeat of the for tree growth, these large-scale disturbances pro-
past disturbance and the landscape pattern. Where mote abundant tree regeneration. In the /ong-term ab-
this is true, landscape pattern may remain similar for sence (several centuries or even millennia) of
prolonged periods even in highly disturbed land- significant, natural, stand-replacing disturbance, un-
scapes. The “footprint” or “ghost” of the predistur- managed forests in these areas slowly lose their closed-
bance forest is reflected in the new forest. Only when forest condition until their structure is open
the new disturbance is very severe and extensive is the woodland, bog forest, muskeg, or treed shrub heath,
“footprint” removed and the “ghost” exorcised. rather than closed forest. Where this is true, the in-
tegrity of the closed-forest ecosystem may require pe-
riodic medium- to large-scale and moderate to severe
Discussion
disturbance rather than such a disturbance being a
Ecosystems are dynamic. There is continual change. In threat to its integrity.
some ecosystems or at some times in the “life” of apar- The above discussion approaches the definition of
ticular ecosystem, this is mainly on a tree-by-tree or ecosystem and ecological integrity from an entirely
small patch basis that creates small- to medium-sized biophysical perspective. Unfortunately, life is not so
canopy gaps that do not significantly diminish the “for- simple. ‘There is widespread recognition that a discus-
est influence.” In other ecosystems or at other times, it sion of these issues outside the context of prevailing
is mainly large-scale, stand-replacing disturbance that social, economic, political, and policy concerns nei-
removes the “forest influence” over large areas. In ther is useful nor provides a basis for designing policy
many ecosystems, gap-level disturbance is a continual and regulation that will achieve our environmental ob-
and important autogenic successional process that oc- jectives (Kay, 1993; others in Woodley et al., 1993). As
 Role ofForests and Forestry in Global Carbon Cycles and Climate Change — 553

was the case with forest health, ecosystem integrity of greenhouse gases (GHGs) by human activity is cur-
will inevitably be discussed and defined in the context rently under way. The main GHGs are CO;, methane,
of what society wants for its forested environments. nitrogen oxides (mainly N,O), and chlorofluorocar-
However, society will undoubtedly not get what it bons (CFCs). Their contributions to global warming
wants if it imposes forest management policies and are a function of their ability to absorb long-wave-
practices that are inconsistent with the ecology of the length radiant energy (e.g., methane is apparently ap-
values that are to be sustained. The role of ecology is proximately 25 times more efficient at absorbing
in defining whether the public’s desires are achievable infrared radiation than CO,) and how long they re-
in the face of the ecological characteristics of the for- main in the atmosphere. Table 20-1 presents estimates
est in question and in identifying ecological con- of the contributions of different GHGs to atmos-
straints on policies and practices that are intended to pheric warming, both instantaneously (i.e., the relative
satisfy these desires. contribution of current emissions) and after their per-
sistence in the atmosphere is accounted for. Clearly,
CO, is the GHG of major concern. Figure 20-1 shows
20.5 Role of Forests and Forestry the history of CO, release from fossil-fuel burning,
in Global Carbon Cycles and cement production, and gas flaring. Liquid fossil fuels
have dominated the releases over the past 70 years, but
Climate Change natural gas and cement production are showing
greater percentage increases; releases from cement
Control of Atmospheric Co, production have increased 20-fold since the 1920s.
The issue of greenhouse gases and the risk of global The global carbon cycle is summarized in Figure
climate change was introduced in Chapter 8. In this 20-2. Living organisms as a group contain the six
section, we briefly review the global carbon cycle, the major nutrients in the following proportions: hydro-
recent history of carbon release to the atmosphere, gen, 2,960; oxygen, 1,480; carbon, 1,480; nitrogen, 16;
and the rise in the atmospheric concentrations of CO, phosphorus, 1.8; and sulfur, 1.0 (Deevey, 1970). Car-
and consider the role of forests, forestry, and defor- bon therefore is one of the major constituents (25%)
estation in determining atmospheric concentrations of of organic matter. It enters the biogeochemical cycle
CO,. Extensive treatments of this important topic can of terrestrial ecosystems primarily as gaseous CO,
be found in Watson et al. (2001) and in a wide variety (50% of organic matter), absorbed from the atmos-
of Internet sites (e.g., cdiac.ornl.gov). For a descrip- phere by photosynthesizing plants. A small amount is
tion of the climatic record from ice cores, see also added as bicarbonate ions in rainwater and from
Mayewski and White (2002). the weathering of carbonaceous rocks. Carbon is lost
Scientists are in general although not complete from plants as respiratory CO), or it may enter graz-
agreement that global warming caused by the release ing and detritus trophic webs. It is released from these

Table 20-1 Contributions of Different GHGs to Global Warming

Relative Contribution to Atmospheric Warming
% Instantaneous % Contribution When
Atmospheric Lifetimes Contribution of Atmospheric Residence
Greenhouse Gas (years) Current Emissions Time Is Accounted for

CO, 50-100? Be 2 80.3

Methane 10 hc8) 2.2

CFCs 65-1305 21.4 8.8

N,O 150 8.1 8.7

Modified from van Kooten et al., 1993.

aReflects the uncertainty over absorption of CO, by oceans and plants.
Different CFCs have different atmospheric residence times.
554 CHAPTER 20 Environmental Issues in Forestry

7000
eel

Solids
6000 —-—- Liquids

5000

4000

3000

2000
metric
Million
carbon
of
tons

1000

[meee
1850 1875 1900 1950 1975 2000

Figure 20-1 Global CO, emissions from fossil-fuel burning, cement production, and gas
flaring for 1850 to 1999 (modified after Marland et al., 2002).

primarily as the respiratory CO, of saprophytic organ- ecosystems such as grasslands. Estimates of the quan-
isms in the detritus trophic web, but a substantial ad- tities of carbon stored in different global ecosystem
ditional amount passes back to the atmosphere as components and rates of transfer within the carbon
methane gas (CH4). Some of the carbon released as cycle are shown in Figure 20-2. The uncertainties in
CO, is promptly reabsorbed by photosynthetic organs arriving at these estimates are considerable, and the
or is dissolved in soil water and taken up by roots as bi- numbers should be used with appropriate caution, al-
carbonate ions, but most of it returns to the atmos- though the great public and scientific concern over the
phere or is carried off in drainage waters as particulate continuing rise in atmospheric CO, levels has led to a
or dissolved organic matter or bicarbonate ions. A tremendous improvement in our knowledge of the
smaller amount is returned to the soil from decompos- carbon cycle over the past two decades.
ing organic matter in the form of organic molecules, At the start of the Industrial Revolution, the con-
such as amino acids; some of these can be absorbed by centration of CO, in the atmosphere is thought to
plant roots, but this pathway is relatively unimportant. have been approximately 290 ppm. It has risen stead-
Most carbon atoms spend only a small proportion ily since then, reaching levels of approximately 315
of their time in biogeochemical cycles, the majority of ppm in the late 1950s, 340 ppm in the early 1980s, and
global carbon dynamics occurring in geochemical cy- 370 ppm by 2002. This increase is a dramatic change
cles. The residence time of carbon in the biosphere from the relatively constant levels that seem to have
varies, of course, and we have already seen that the prevailed for the past 12,500 years: 260 to 280 ppm.
trapping and daytime reabsorption of night time res- Evidence from bubbles of atmospheric gasses trapped
piratory CO, is more efficient in forests than in more in ancient ice sheets in Antarctica, Greenland, and
diminutive plant communities in which most respira- elsewhere indicates four cycles of atmospheric CO,
tory CQ) is lost back to the atmosphere. Also, carbon concentrations between approximately 180 and 300
that is fixed in permanent plant structures will remain ppm over the past 400,000 years. Estimates of even
for a longer time in an ecosystem with large, long- earlier atmospheric concentrations (based on infer-
lived organisms than in one with small and/or short- ences from geochemical data) suggest that this range
lived organisms and longer in ecosystems with low of variation may have persisted for 25 million years
plant utilization by herbivores than in heavily grazed but that over the 500 million years before that, con-
ecosystems. Carbon therefore will tend to remain centrations may sometimes have risen as high as 6,000
longer in the biogeochemical cycles of forests than of ppm. (Figure 20-3).
 Atmospheric carbon
CHy, CO, and CO,

CO, Exchange
Bicarbonate Termectial
ions (HCO, in rain esta
ie) ol photosynthesis ie
WERE =n. eames Dilation Combustion of Volcanism Carbon in detritus
= oer eG eee
——
— fossil fuels trophic webs
Carbon in
terrestial : '
] Carbon in organic ; : Ht
| _____} and sedimentary /!
Sere inorganic deposits POPE ws fe
7 | U
| Y
Carbon in detritus} |
trophic webs : Carbon in calcium
LB if carbonate on the plants
a 4 Le
Carbon in E==~ coon Respiration
the soil a 6 e
= Precipitation Aanatio
ap eabomsn Eee hia nthesis
river water solution P »
CO, exchange
Carbon in ocean waters
Dissolved CO, + H,0 = H,CO; = H* + HCO; = H* + COng

(a)

Atmospheric C
750
Respiratior Uptake of CO,

| Anima] [Pam aye

60 -s into solution in

Soil < 0.5 ; 120 Release of CO, 92

respiration " 2 : ; from solution in
60 Combustion of Sea
YY) fossil fuel ; 90

NY YY Yip | Respiration
Nise Ul y ais a
yy yy
Yj
yyLf eras
Aquatic +
~ animals Fe 9)

L/h, //; =05 :

Organic and inorganic carbon
MDG TTT] TR
in river water
//7

Extractable
Fossil Fuels
4,000

(b)

Figure 20-2 The carbon cycle. (A) Distribution and transfers of carbon in the biosphere. Double
lines indicate the biogeochemical cycle; single lines indicate the geochemical cycle; solid lines indi-
cate major transfers; dashed lines indicate transfers of secondary importance. (B) Estimates of quan-
tities and rates of transfer of carbon in the biosphere. Units are billions (10’) of metric tons of car-
bon (10! g). Transfer rates are expressed on an annual basis. (B based on Bolin et al., 1979; Clark et al.,
1982, with updated estimates from Houghton and Hackler, 2002; Marland et al., 2002; Watson et al.,
2001).

Al anal
556 CHAPTER 20

CO,
360
320
280
240
200
concentration
(ppm) 1950 1960
\
Environmental Issues in Forestry

1970 1980 1990 2000

CO,
(ppm)
360
320
280
240
200
concentration
800 1000 1200 1400 1600 1800 2000
Year Year

(a) (b)

ppm)360
320
280
240 |
200 IM
concentration
(CO, 12500 10000 concentration
CO,
(ppm)
7500 S000 2500 0 400 300 200 100 0
Age (yr BP) Age (kyr BP)
(c) (d)

=Qa, 7500
ppm) == 6000
S
= 4500
:
3 3000
co)
3 1500
isl
ind== Oo
concentration
CO,
( .o) 0
tNnn 20 15 10 5 0 500 400 300 200 100 oO

Age (Myr BP) Age (Myr BP)

(e) (f)
Figure 20-3 Variations in atmospheric CO, concentrations over various historical periods
(based on Watson et al., 2001). (A) Late 1950s to 1998 measurements at Mauna Loa, Hawaii. (B)
Variations over the past millennium, from ice core data and actual measurements. (C) The ice core
data from Taylor Dome, Antarctica, for the past 12,500 years. (D) Data from the Vostock ice core,
Antarctica, for the past 400,000 years. (E and F) Inferred concentrations from geochemical data for
25 million and 500 million years before present. (Modified after Watson et al., 2001.)

Predictions of future trends in atmospheric CO, rious threat to life as we know it, including forest
concentrations are highly variable according to as- ecosystems.
sumptions about population increase, trends in the per Human-caused increases in atmospheric carbon lev-
capita use of fossil fuels, and land-use changes, but els before the Industrial Revolution are believed to have
they range from approximately double to more than resulted largely from land-use changes (compare 1850s
three times the recent historical levels by the end of data in Figures 20-1 and 20-4). In contrast, atmospher-
the 21st century (Watson et al., 2001): from approxi- ic carbon increases in the 1850 to 1980 period have
mately 500 ppm to as high as 1,000 ppm in year 2100. been attributed 62% to fossil-fuel burning and 38% to
Clearly such an increase is without precedent within land-use changes—largely deforestation and conversion
the recent evolution of the human and most other cur- of old-growth to second-growth forest (Dale et al.,
rent species. The climate change implication of such 1991, and other references in Can F For Res 1991, Vol.
an increase in one of the major GHGs poses a very se- 21). During the 1980 to 1989 period, the contributions
 Role of Forests and Forestry in Global Carbon Cycles and Climate Change oy!

Total Flux
Latin America
Tropical Asia
Tropical Africa
China
Former Soviet Union
- Pacific Developed Region
Canada
North Africa and Middle East
Europe
United States
Flux
C/yr)
(Tg

“. oe
00 Ss IS

1850 1860 1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
Year

Figure 20-4 Carbon emissions to the atmosphere as a result of land-use changes, 1850 to
2000. In 2000, these amounted to approximately 24% of total anthropogenic emissions and came al-
most entirely from tropical developing countries. Contributions from China and the former Soviet
Union were significant in the 1920 to 1980 period but have since declined. (Houghton and Hackler;
2002.)

of these two sources were 77 and 23%, respectively clearance in the United States have been reversed as
(Dixon et al., 1994), ratios that were reported again in abandoned agricultural land reforested and forest pro-
2000 after an increase in land-use contributions in 1990 tection from fire led to increases in forest carbon. This
to 26% asa result of large forest clearances in tropical national perspective masks local reductions of carbon
Asia (Houghton and Hackler, 2002). Annual emissions storage in some forests as they were converted from
of carbon as CO, in 1980 were 5.4 + 0.5 Gt (10” g or old- to second-growth forest (Harmon et al., 1990,
billion t) from fossil-fuel burning and 1.6 + 1.0 Gt from 1996, 2002). Since the 1930s, land-use changes in
land-use change; in 1999 they were approximately 6.5 tropical Asia, Latin America, and, increasingly, tropi-
Gt and 2.1 Gt, respectively. cal Africa have dominated emissions related to land-
A comparison between Figures 20-1 and 20-4 use changes.
shows that the relative contributions of these two If the world trend of urbanization of the human
emission sources continue to change, with land-use population continues and is accompanied by increased
change becoming progressively less important than reliance on steel and concrete in buildings and roads,
fossil-fuel burning. Land-use change emissions then we can expect increases in the contribution of ce-
peaked in approximately 1990 and have since declined ment and steel production. A building made of concrete
somewhat, whereas emissions from fossil-fuel burning requires 1.5 times more fossil-fuel energy than an
have continued to increase. The major source from equivalent wooden building; if only the aboveground
land-use change in the 1800s was in the United States, portion of the building is considered, then the differ-
where forest land was cleared extensively for agricul- ence is 1.7 times (because wooden buildings have con-
ture. Since the 1930s, the net emissions from forest crete foundations). If the building is composed largely
558 CHAPTER 20 Environmental Issues in Forestry

of steel structural elements, then it requires 1.9 times Energy consumption

more fossil-fuel energy than a wooden building; 2.4 Energy use less common
Total energy use concrete basement
times as much if just the aboveground structure is eval- 8 6
(Manufacturing [J Extraction =
uated. Clearly, as the housing needs of another 3 billion
people, largely in cities, is satisfied, the need to use
$7 $5
x 6 x4
wood rather than steel or concrete will become
environmentally urgent. (www.cwc.ca/environmental/ g°
94
S3
sustainable_buildings/green_by_design/3) (Figure 20-5). =|
= 3
=|
2 2
The observed increase in atmospheric CO, can- a)
=
5
=|
not account for the estimated releases of CO, from m | i aa]
fossil-fuel burning and other sources. During the 0 Wood Steel Concrete Wood Steel Concrete
1980 to 1989 period, the global oceans were esti-
Air impacts
mated to have absorbed 29% (2.0 + 0.8 Gt) of the re-
leases, whereas the observed increase in atmospheric Greenhouse gas index Air pollution index
140
carbon was 46% of the releases (3.2 + 0.1 Gt). This
2 120 =~ 10
leaves 25% of the releases (1.8 + 1.4 Gt) unac-
= 100 =.
counted for. It is widely believed that forests are the 2 x
5 80 S
repository for the missing carbon (Dixon et al., *_
Ss
oro
=)
= 60 ‘s
1994). Currently, the observed increases in the at-
mosphere seem to account for approximately 50% of 3ey 40 ae:
the known releases. If this suggests a declining trend ©e) 20 G2
Ss

in the proportion of CQ, releases that the world sys- Wood Steel Concrete
f-)
Wood Steel Concrete
tem can adsorb, then the climate change threat from
GHGs may increase faster than the increase in fossil- Water and land impacts
fuel combustion. Water pollution index Solid waste index
Because photosynthesis is generally limited by the 180 160
_ 160 -~ 140
availability of CO,, the world’s vegetation, especially
S 140 S 120
the world’s forests, acts as a negative feedback regula-
Seal x 100
tor that helps to moderate changes in atmospheric > 100 rs
3 $0 4 80
CO, concentrations. There is an estimated 5% in-
crease in net primary production (NPP) for each 10% x 60 x 60
Z 40 3 40
increase in CO, concentrations when water, tempera- Be 9:1) a4)
ture, and other nutrients are not limiting. Also, ele-
Wood Steel Concrete Wood Steel Concrete
vated CO, levels increase the water use efficiency of
plants, thereby increasing NPP in dry climates Figure 20-5 Comparisons between the energy and thus
(Wong, 1980). fossil-fuel requirements and the air impacts, water pollution
The second major regulator of atmospheric CO, index, and solid waste index of the structural components of
levels is the world’s oceans. CO, dissolves in water to buildings in which these are made of wood, steel, or con-
a degree that is controlled by temperature, pH, the crete. Clearly, wood has a strong environmental advantage as a
atmospheric pressure exerted by CO, and the con- building material. The energy requirements of recycling a
centration of other nutrients in the water. Atmos- wooden building are also much lower than for the other materi-
pheric CO, in contact with water forms the als. (Based on McFarlane, 2003, and www.cwe.ca/environmental/
following series of chemical equilibria, which serve sustainable_buildings/green_by_design/3; used with permission.)
to maintain the air above the ocean at fairly constant
CO, levels:
+H,0 .
= = = — =. Rainif
Op > co, = H,CO, =H’ + HCO, = ee CO — CaCO,
Atmospheric Dissolved in Dissolved
seawater

* on the
Sediments igo
Fallic oun
Gi
biogeochemical
cycles
 Role ofForests and Forestry in Global Carbon Cycles and Climate Change 559

As the CO, concentration in the air increases, ocean, which causes more dissolved CO, to be re-
more gas goes into solution in the sea. This results in leased (the chemical equilibrium moves to the left),
increased sedimentation of calcium carbonate and also further enhancing the heating effect. This positive
in greater aquatic production if there are adequate feedback mechanism cannot go on indefinitely, how-
supplies of dissolved nitrogen, phosphorus, and other ever, because a major increase in air temperature
limiting nutrients such as iron. Reductions in atmos- would accelerate evaporation of water from land and
pheric CO), however, lead to a release of CO, from sea, leading to increased cloudiness. This would re-
solution in seawater, the loss being made up by a chain flect more of the sun’s short-wavelength radiation,
of chemical adjustments that lead ultimately to the up- leading to a cooling of the atmosphere. (This was
take into solution of calcium carbonate deposits on the discussed in Chapter 8.)
ocean floor. This mechanism is efficient at regulating
atmospheric CO); it is thought to have removed 35%
of what humans have added over the past century. The Contributions of Forests and Forestry
only snag is that the rate at which the mechanism op- If the foregoing interpretation is correct, then the
erates is dependent on the rate of atmospheric and forests of the world have a vital although ultimately
oceanic stirring and on the rate of response of the nutritionally limited role to play in the regulation of
oceanic chemical equilibria (Broecker, 1973). Because atmospheric CO), and in moderating the negative con-
these all are slow processes and because of nutrient sequences of fossil-fuel burning. Conversely, misman-
limitations on mid-ocean NPP, the oceanic CO, con- agement of forests that reduces their role as a sink for
trol mechanism has not been able to keep up with the atmospheric carbon could exacerbate the existing
very rapid rate of CO, release from fossil fuels. Be- threat of global warming (Kurz and Apps, 1993). The
cause of the lag in response of this control mechanism world’s forests are estimated to contain 1146 Gt of
and because of the reduction in the terrestrial bios- carbon, or approximately 212 times as much as the an-
pheric control mechanism, atmospheric CO, concen- nual release from fossil fuel. More than two thirds of
trations have risen and forests have become an the forest total is held in soils and peat.
increasingly important sink for atmospheric CQ). Identifying the carbon storage and its change over
However, the ability of forests to respond to increas- time and the annual fluxes of carbon in the world’s
ing CO, is also nutrient limited. forests is very difficult. Reliable data simply are not
The regulation of atmospheric CO) is a good ex- available, and errors are often made in data interpreta-
ample of the complex processes that have operated tion. For example, Dixon et al. (1994) showed an an-
for millennia to maintain the chemical conditions of nual net deforestation in Canada of 500,000 ha, but
this planet between the narrow limits within which these numbers refer to the annual rate at which log-
life as we know it can exist. It has far-reaching signif- ging creates not-satisfactorily restocked (NSR) land.
icance for humans and for other living organisms, The NSR designation is related to whether the area is
because the concentration of atmospheric CO, con- regenerated to some technical standard and does not
tributes to the regulation of the world’s temperature tell you whether tree biomass is accumulating on the
in addition to its critical role in plant nutrition. As area. Much of this area is covered with rapidly grow-
noted earlier, short-wavelength solar radiation (sun- ing hardwoods that at the time of the data source
light) can pass relatively freely through our atmos- (Honer et al., 1991) were not considered commer-
phere, but the long-wavelength energy (infrared cially valuable but were rapidly storing carbon (Kurz
radiation) that is reradiated from the surface of the and Apps, 1993). After some years, most harvested
earth cannot. It is absorbed by CO, and water vapor, areas in Canada return to productive forest growth. ‘To
and our atmosphere acts much like the glass in a treat these NSR data as equivalent to the cumulative
greenhouse, maintaining a warm layer of air in con- deforestation data from the tropics significantly mis-
tact with the earth (the atmospheric “greenhouse ef- represents the carbon budget of these northern forests.
fect”). As atmospheric concentrations of CO, and These concerns notwithstanding, Table 20-2
other GHGs increase, more of the reradiated energy presents recent studies of the contribution of the
is absorbed by the atmosphere and its temperature world’s forests as a carbon sink (net uptake from the
increases. This, in turn, raises the temperature of the atmosphere) or a carbon source (net release to the
560 CHAPTER 20 Environmental Issues in Forestry

Table 20-2 Estimated Annual Uptake or Release of 1993). However, natural disturbance increased in the
Atmospheric Carbon by the World’s Forests 1970 to 1989 period, reducing the role of these forests
Net Annual
as a carbon sink. Forests that are subject to large-scale
fluctuations in natural disturbance on a time scale
Forest Area Uptake Release
comparable to tree life times do not seem to reach a
High latitude carbon exchange equilibrium over these time scales
Russia 0.3-0.5 _
(Kurz et al., 1995a). Consequently, the carbon budget
in Table 20-2 should be considered as a “snapshot,”
Canada* 0.08" —
not as an accurate statement of the long-term carbon
Mean 0.48 + 1 balance. In particular, fluctuations in climate, by af-
Mid-latitude fecting forest fire and insect outbreaks, which affect
USA (including Alaska) 0.1-0.25 - forest age-class structure, have the potential to cause
Europe (including Scandinavia)0.09-0.12 _ very large changes in the world’s forest carbon balance
China 0.02 —
(Kurz et al., 1995b).
A study of the effects on carbon storage of conver-
Australia Trace —
sion of west coast old-growth forests of the Pacific
Mean 0.26 + 0.09 Northwest United States to young forest (Harmon et
Low latitude al., 1990b) concluded that the conversion of 5 million
Asia - 0.50-0.90 ha of old-growth to younger forests in the last 100
Africa — 0.25-0.45 years has released 1.5 to 1.8 billion tons (10’ Mg) of -
Americas _ 0.50-0.70
carbon into the atmosphere. The quantity of carbon
stored in these old forests is considerably greater than
Mean 1.65 + 0.40
the carbon stored in a managed second-growth forest,
Net balance Release of 0.9 + 0.4
and, through the use of a computer simulation model,
Dixon et al., 1994.
it was estimated that it would take 100 years for the
Units are Gt yr’. secondary forest to achieve 80% of the carbon storage
“See note in text about error in forest area assumptions. Such errors of the old-growth forest and that this forest would
of data interpretation may affect other numbers in the table. have to be left to grow for approximately 250 years to
>van Kooten et al. (1993) suggested a value of 0.157. Kurz et al. equal it. In this study, it was assumed that only 45% of
(1995) estimated a value of 0.2.
the harvested wood becomes wood products in long-
term storage (buildings), the rest being returned
atmosphere). The data suggest a net release of approx- promptly to the atmosphere. It was further assumed
imately 0.9 Gt: the result of a net uptake of 0.7 Gt by that landscapes managed on rotations of 50, 75, or 100
mid- and high-latitude forests and a net release of 1.6 years would store only 38, 44, and 51%, respectively,
Gt from low-latitude forests. of the carbon stored in the old growth. Assuming that
Evaluation of the global carbon budgets of terres- housing structures last only approximately 50 years,
trial ecosystems (e.g., Houghton et al., 1983, 1990) with all the carbon in these structures being returned
suggests that these budgets have overestimated the net to the atmosphere as soon as the building is replaced,
release of CO, from forests because they considered the authors estimated that conversion from old-
only the release from areas modified by human activi- growth to a 60-year rotation second-growth forest re-
ties and, wrongly, assumed that the remaining forest sults in a one-time net loss of 305 t of carbon per
area makes no net contribution to the global carbon hectare from these particular types of forest. There-
cycle. In the case of Canada, which accounts for 10% after, the carbon just cycles back and forth between
of the world’s forests, only a small proportion of the the forest and the atmosphere.
area is disturbed annually. There was a shift toward an Harmon et al. (1996a, 1996b) undertook a more
older average age in Canadian forests during the 1920 thorough analysis of this issue and concluded that
to 1970 period because of a reduction in forest distur- 48% of harvested logs in the Pacific Northwest actu-
bance; consequently, these unmanaged forests were a ally become long-lived wood products, with an aver-
net carbon sink during this period (Kurz and Apps, age annual rate of return of their carbon to the
 Role of Forests and Forestry in Global Carbon Cycles and Climate Change 561

atmosphere of 0.85%. Using the simulation model chain was with the shorter rotation. However, for both
STANDCARB, Harmon and Marks (2002) concluded species, the longer rotations gave the best overall car-
that maximum carbon storage in managed forests re- bon balance. This was accompanied by reduced wood
sults from long rotations without slashburning and harvests and revenues, with associated loss of social
with low to modest levels of utilization of tree biomass values. Li et al. (unpublished observations) simulated
and that maximum ecosystem carbon occurs in un- the effects of management on carbon sequestration
managed forests protected from fire. They concluded and other GHG emissions from forested wetlands.
that partial harvesting will result in a greater storage They reported that the same management scenario
of carbon in the forest than clearcutting with slash- produced very different consequences under different
burning and that regeneration delay has a significant climatic conditions and that in these wet forest ecosys-
effect on within-ecosystem carbon storage. tems, methane and nitrous oxide fluxes were signifi-
These are useful analyses. However, the models cant. This suggests that a preoccupation with CO,
used did not address the issue of soil fertility and its ef- may give an incomplete picture in some forest types
fects on carbon sequestration and storage (Harmon and that caution should be exercised when extrapolat-
and Marks, 2002) and did not report on the implica- ing the results of studies in one type of forest ecosys-
tions of partial harvesting to promote carbon storage tem to another and from one climatic area to another.
on successional sequences, the abundance of early Although Harmon’s group (Harmon and Marks
seral species, and nitrogen fixation, all of which have 2002) and the Canadian Forestry Service carbon
implications for future NPP, carbon sequestration and group (e.g., Kurz et al., 1998) have contributed very
storage, and various measures of biodiversity, sustain- useful analyses of forest carbon budgets, including the
ability and stewardship. Seely et al. (2002) undertook a storage time for harvested carbon, there is generally a
similar simulation study of carbon sequestration in bo- lack of analysis of the carbon implications of not har-
real forests using the FORECAST ecosystem man- vesting wood, which involves having to use concrete,
agement simulation model. They examined rotation steel, and other wood substitutes. Much of the har-
length, carbon budgets, and changes in species com- vested wood that becomes waste during sawmilling or
position (hardwood, conifer, and mixed-wood options) paper making is burned to generate energy (thus re-
at different rotation lengths, with and without fertil- ducing use of fossil fuels), and the wood in buildings
ization. They noted an important difference between when they are replaced is recycled, used for energy
within-ecosystem carbon storage and the rate of car- (displacing fossil fuels), or land filled. Wood buried
bon sequestration by the growing forest; carbon stor- deeply in land fills will generally decompose slowly if
age in the forests was greater with longer rotations, conditions are anaerobic and/or water saturated. In-
but carbon accumulation rates were greatest with creasingly, methane and other combustible green-
shorter rotations. This tradeoff between sequestration house gasses that are generated by microbial action in
and storage requires an examination of the fate of the landfills are being harvested and used as energy
carbon in harvested products. In a modeling investiga- sources that displace fossil fuel carbon releases. Wat-
tion of the effect of rotation length on carbon seques- son et al. (2001) presented a simulation analysis of
tration in Finnish forests, Liski et al. (2001) found that within-forest carbon, carbon in short- and long-lived
shorter rotations increased soil carbon storage but de- harvested products, the fossil-fuel carbon that is not
creased tree carbon storage. However, they concluded released when the unused harvested wood is used for
that a complete picture of the carbon implications of bioenergy, and the fossil fuels that are not used when
forestry could result only from a complete analysis wood is used instead of steel and concrete.
that included the fossil-fuel energy costs of harvesting Figure 20-6 shows the mineral soil, forest floor,
and manufacturing and a consideration of the end use and live tree carbon content for a scenario starting
of the harvested products. This revealed differences with bare soil and no soil carbon (an unreasonable as-
between species. For example, the postharvest manu- sumption because it ignores the very real control that
facturing and utilization chain from Scots pine (Pinus soil fertility has on forest growth) and two and a half
sylvestris L.) stored the highest amount of carbon with 40-year rotations. At the end of the second rotation,
the longest rotation, whereas for Norway spruce (Picea the trees accounted for approximately 50% of the
abies {L.] Karst.), the greatest carbon storage in the budget, the harvested products and carbon displace-
562. CHAPTER20_— Environmental Issues in Forestry

225 Displaced fossil fuels

Landfill
200 Energy for products

175 Short-lived products

Long-lived products

(tC/ha)
carbon
Cumulative

0 10 20 30 40 50 60 70 80 90 100
Time (years)

Figure 20-6 Carbon budget for a hypothetical managed forest. See text for explanation
(modified after Watson et al. 2001).

ment for approximately 36% of the budget, with the tion in the forest. For many of the world’s old forests
balance (14%) in the soil and forest floor. By the end that have a large inventory of stored carbon but little
of the 100-year simulation, the soil and forest floor ac- or no net carbon uptake (ignoring, perhaps wrongly,
counted for approximately 25% of the budget, but the the export of carbon in rivers and its deposit as long-
harvested products and carbon replacement made up term storage in lake and oceanic sediments), harvest-
55% (Watson et al., 2002). This emphasizes first the ing the timber, putting it into long-term storage, and
need for a dynamic rather than a snapshot evaluation reestablishing a productive new forest will actually in-
of carbon budgets and also the need to undertake a full crease these forests as a carbon sink.
carbon accounting. Liski et al. (2001) also stressed the Clearly, the issue of the role of forests in the regu-
importance of accounting for the carbon aspects of lation of atmospheric chemistry and how this role is
harvesting, manufacturing, and the full fossil-fuel car- influenced by forest harvesting and management or by
bon implications of wood use, not only the carbon conversion from forest to other land uses (deforesta-
storage dynamics in the forest. tion) is extremely important environmentally. Because
For forests that accumulate very high levels of car- of this, it has entered the environmental debate and
bon in long-lived trees and dead organic matter, there has been much publicized in the media. The threat of
will generally be a significant one-time net reduction climate change is both real and very worrying. How-
of in-forest carbon storage when these old forests are ever, as is common when an issue is taken up in the
converted to young forests. However, whether there is media and by the public, confusion has developed over
an overall reduction or increase in atmospheric carbon what is happening and what should be done about it.
as a result of such a reduction of in-forest storage is Phrases such as “forests are the lungs of the earth,”
clearly a function of (1) what we do with the harvested implying that loss of forests threatens the atmospheric
wood (how long it is stored as buildings and other inventory of oxygen and thus the very existence of
long-lived wood products versus short-lived paper oxygen-dependent species, have fueled concerns about
products or fuel); (2) whether the use of wood results harvesting temperate old-growth and tropical forests.
in less fossil carbon release than the use of alternative Concern over climate change has been advanced as an
construction materials or directly displaces the burn- argument that no old forests should be logged.
ing of fossil fuels; and (3) whether we raise, lower, or Human society is going to have to make some diffi-
have no effect on the future rate of carbon accumula- cult decisions with respect to fossil-fuel consumption
 Role ofForests and Forestry in Global Carbon Cycles and Climate Change 563

and land-use changes that threaten to so alter the chem- how forests are managed to sustain their function of car-
istry of the atmosphere that the global radiation balance bon storage and how we use the wood that is harvested.
and temperature could be significantly altered. Harvest- Achieving an optimum strategy for forest management
ing and management of forests have small but important with respect to the carbon question, while at the same
roles to play in this issue, and forests that are dedicated time addressing other environmental issues in forestry,
for wood production must be managed in a way that requires a much improved understanding of the controls
minimizes fossil-fuel consumption and maximizes the on production ecology and the dynamics of carbon in
sequestering and storage of atmos-pheric CO, on a sus- forest biomass. Unless policies with respect to carbon
tained basis. Society will have to devise end uses for storage by forestry reflect this ecology and the econom-
wood and paper products that similarly reduces our de- ics and sociology of our use of wood products, concern
pendence on fossil-fuel burning and maximizes the stor- about this issue will not contribute much to the resolu-
age of carbon in wood products. The issue thus is both tion of the potentially serious risk of climate change.
 Pa - wav «

Ven) baw oore’t Vo

a = ae! ii : Aah P

\ ‘ » al af hat Lev Wa ite sacl?'s

/ j ie ary aff af hie

a ‘ = :
_——— «Wy , ewes ie < it sista) it

ihe DS) ie SU ee va
7 ivgaeh Sa Jaya 9 acathony, fi
- izzpe Seen a oes
ee Teyite >
f°"). Soe Se Oy rida, Ps Bah
, ih Aarcs! hw vata
‘4 ima ‘ees ugtegy se
— ' ¢ a Dr wat 4arle
mp ieee(0 oar

~*

| ~~
— > —

=
’ fe

Se:

~
y

fas
Cay (

a)

eas a
2 "9

wa

¢ lo@ Oagieg
POG! i i Soniye
ee
 C H E R

21
Models and Their Role in
Ecology and Resource
Management!
“Pluralitas non est ponenda sine neccesitate”
WILLIAM OF OCCAM (OR OCKHAM) (1284-1347), AN ENGLISH PHILOSOPHER AND THEOLOGIAN

Occam’s razor is the basis for the law ofparsimony and the rule ofsimplicity: do not posit
complexity more than necessary. However, Occam’s razor has two edges: it says as simple as
possible but as complex as necessary.
A problem is an issue that does not get solved. An issue that gets solved quickly is not a prob-
lem. Problem issues often persist because they are complex and only simple solutions are of-
fered. Complexity must be addressed, as required by Occam’s razor.

21.1 Introduction scription of phenomena in terms of simple or causal

hypotheses. In many cases, this may be the only way to
The complexity of ecosystems has been mentioned advance our knowledge of the details of processes, and
several times, as has an important consequence of this it is particularly appropriate in systems about which
complexity: that events and conditions in ecosystems we already know a lot and/or that are simple. It is gen-
are difficult to predict. A detailed knowledge of erally a much less appropriate method on its own for
ecosystem structure and function (Chapters 3 to 19) investigating the overall functioning of little-under-
greatly improves our ability to make accurate predic- stood, complex ecosystems and the impact of manage-
tions, but this knowledge on its own is not always suf- ment thereon. Simple (causal) and testable (rejectable)
ficient. Many ecological phenomena and resource explanations for complex phenomena about which we
management problems involve so many interacting have limited knowledge rarely contribute much to our
factors that predictions are difficult despite our knowl- understanding of those phenomena and our ability to
edge. Classification of ecosystems (Chapter 6) and of manage them. An alternative approach, which is being
our knowledge about them certainly helps, but a major used with increasing frequency, is to incorporate much
limitation remains: the limited mental capacity of the of the complexity into a comprehensive model of the
human brain, which restricts us to thinking at any one phenomenon and to use the model to make predic-
time about only a few of the many interacting compo- tions about the system and its response to disturbance.
nents and processes of a complex system. It is even In the early 1970s, fewer than 5% of the published
more difficult to comprehend the interactions of many articles in the three most prestigious international
ecosystem components and processes. over long peri- journals on forest research were concerned with mod-
ods of time. els and modeling. By the early 2000s, this had risen to
The complexity of natural phenomena has always 15%, with an apparently linear increase over time
been a problem for science. In their attempts to de- (Messier et al., 2003). This trend reflects the increas-
scribe, define, and explain the world, scientists have
traditionally resorted to scientific reductionism: the de- ‘Chapter 21 was written jointly with Kim A. Scoullar and Daniel Mailly.

565
566 CHAPTER 21 Models and Their Role in Ecology and Resource Management

ing recognition that for knowledge to achieve its full els of 30 years ago to a recent ecosystem-level model,
value for society, it must be incorporated into a system is presented. The stages in the development of a com-
of knowledge at the level of complexity of the issues puter model are examined, followed by a description
that face society—a system that incorporates a consid- of the three major approaches to forecasting (model-
eration for all the key components and processes of ing) in forestry. The third of these approaches, hybrid
that system. This is analogous to the value of individ- simulation modeling, is then examined in more detail
ual bricks, pieces of lumber, glass, wire, and other because it is the approach that is expected to dominate
building materials. Their full value is not expressed stand-level modeling in the future. Landscape model-
until they are combined into a functional house that ing is then introduced, followed by a consideration of
does not leak and serves the needs of the occupants. the question, “How complex should a model be?” The
Most scientists are—and should be—creators of scien- chapter, which concludes with a brief treatment of the
tific “bricks and two-by-fours (pieces of lumber)”— relationship between modeling and the traditional sci-
knowledge about the subcomponents and individual entific method, is intended as a brief introduction to
processes of complex systems. However, if all we do is and overview of modeling for beginning students of
put these in warehouses (we call these libraries), then ecology and renewable resource management.
they do not provide full value. This is achieved only No attempt has been made to review the literature
when scientific “architects”(conceptual modelers) de- or provide a detailed description of the wide variety of
sign systems (models) that can incorporate the appro- modeling techniques. Readers who are interested in
priate knowledge to the level of complexity of the pursuing the topic further are referred to Bossel
issue or system in question, and scientific “builders” (1991), Dixon et al. (1990), Glenn-Lewin et al. (1992),
(modelers and computer programmers) actually build Korzukhin et al. (1996), Landsberg (2003), Lemay and
and operate the models. Marshall (2001, 2003), Makela (2003), Messier et al.
A model can take many different forms, but basi- (2003), Mladenoff and Baker (1999), Mohren and
cally it is either an abstract or a physical entity that Kienast (1991), Stage (2003) and Vanclay (1994).
represents in some way the form and/or the function
of real-world entities and processes. A mathematical
equation that summarizes a physiological process is a 21.2 Limitations of Models
model of that process. Pictures and statues are models.
This book is a model of the form and functioning of a When a model is first developed, it is generally an in-
forest ecosystem. Many different types of model have exact representation of the real object or process that
played and will continue to play useful roles in ecology it mimics. However, models can be improved. Some
and resource management, but it is the relatively re- can be improved to the point at which they represent
cent use of a model in conjunction with a computer reality so well that they are no longer a model but the
that makes modeling such a valuable aid in under- reality that they were intended to represent. In other
standing the properties and behavior of complex sys- cases, this cannot happen. The former are models of
tems. It was the advent of computer simulation that human-made objects; the latter are models of natural,
gave modeling in general the attention and the reputa- living systems. A model airplane becomes a real air-
tion that it now has. Simulation is the creation of a dy- plane when it represents the real airplane so closely
namic representation of the external characteristics of that it can do everything that a real airplane can. This
a system by implementing a series of mathematical is possible because model airplanes can be made of the
equations (a mathematical model) that represents that same materials and with the same construction tech-
system on a computer (Swartzman and Kaluzny, 1987). niques as real airplanes. A model bird that mimics the
This Chapter briefly introduces the reader to the flight of a bird cannot, however, become a real bird. It
variety and nature of models. It starts with a caution- cannot mimic all aspects of a bird’s morphology, phys-
ary note on the limitations of models and a discussion iology, and behavior because at present, we cannot
of why people build models despite these limitations. synthesize living tissues and assemble them into living
The various different types of model and their attrib- systems. Similarly, a computer model of a tree or an
utes are reviewed, and as an example of the historical ecosystem can never represent that tree or ecosystem
trend in modeling, a series of models of harvest- with complete accuracy because the computer model
induced site nutrient depletion, from the simple mod- uses different construction materials and techniques.
 Why People Model and the Survivorship ofModels — 567

Because of this ultimate restriction on how well a the past three decades. These vary greatly in type (e.g.,
model can duplicate biological reality, no model of bi- word, picture, physical, mathematical, computer), in
ological systems can ever be completely “correct.” All scope and resolution (e.g., models of physiological
models of forest ecosystems thus are incorrect to some processes, of individual tree growth, of stand develop-
degree. Whether a model is “good” or not must be ment, of ecosystem processes, of entire landscape
judged by whether it provides an acceptable represen- ecosystems), and in purpose (e.g., to describe natural
tation of the reality that it is describing, not on phenomena, to organize existing data, to identify
whether it is “right” or “wrong.” For example, once information gaps and research needs, to be a teach-
while flying from New York to Vancouver, British Co- ing/learning process, to predict future conditions and
lumbia, I noted that the location of Thunder Bay, On- the response of systems to natural disturbances and/or
tario, on the map of North America in the in-flight management).
magazine was approximately 50 km in error. Does this The diverse types of mathematical and computer
render this two-dimensional, pictorial model of geog- models that have been produced have varied notably
raphy of no value? A pilot equipped with only this map in life span. Some models of restricted application
would have been able to navigate on a cloudy day to have had a brief although useful existence and have
within approximately 100 km of Thunder Bay but not evolved. Other models or model types have been
might have been unable to find the city or the airport. more persistent. It is of some interest to consider
A local city map marks the airport and runway with an which characteristics have made these models last.
error of a few hundred meters—sufficient accuracy to The International Biological Program (IBP) of the
locate the airport but not the runway. A map with an late 1960s and 1970s provided a major impetus to
accuracy of +5 m would be needed for a safe landing modeling of ecosystems and ecosystem processes.
in low cloud. Thus, each map has value for one of Large numbers of biologists spent an enormous
these navigation tasks but not for the others. Deciding amount of time producing very detailed conceptual
on the usefulness of a model is always relative to the models of the systems in which they were interested.
use to which it will be put. The computer models that resulted were highly com-
plex, and many of them produced very acceptable sim-
ulations of parts of the real world. The modeling
21.3 Why People Model and activity proved valuable as a learning experience, for
the Survivorship of Models synthesizing complex data sets, and for describing nat-
ural systems. It therefore is of interest to learn that
If you ask someone who has just completed a model- most of the models produced have had a short life
ing project why he or she did it, what he or she got out span. Like a shooting star, they were brilliant but
of it, and of what value the completed model was, you briefly so. Why?
are likely to receive a wide variety of replies. This sim- Part of the explanation may be that many of these
ply reflects the variety of reasons for undertaking models had satisfied the modeling objective as soon as
modeling (and the success or failure of the modeling they were completed. They were not developed for
activity!). the purpose of prediction and gaming. Many of the
Models may be built as predictive tools, with rela- modelers were biologists who were more interested in
tively little new being learned in their development. describing biological systems than in explaining them
Their value lies in what they can do, not in how they or predicting their response to disturbance and man-
were made. Alternatively, models may be built as a agement. Their purpose achieved, the models fell into
purely heuristic exercise, a way of exploring and syn- disuse as the modelers turned to other types of re-
thesizing what we think we know about some system search activity. Another component of the explanation
and of identifying what we do not know but would like may be that adequate data were not available at the
or need to know. The completed model may be of lit- time when the models were being developed. Many
tle use, and many models are never used after their de- biologists involved in the modeling of that era became
velopment has been completed. Their value is in their disillusioned by having to use so many imaginary pa-
development, not in the completed product. rameter values in the models because of lack of
Many different types of model have been produced empirically derived real numbers; they lost faith in the
by ecologists and renewable resource managers over computer modeling method. This problem was
568 CHAPTER21 Models and Their Role in Ecology and Resource Management

exacerbated by the fact that many of the biologists in- management-oriented but ecologically based ecosys-
volved in the modeling were process-oriented scien- tem models will combine to make gaming with com-
tists, who believed that the more detailed the model, puter models a major tool in the conservation and
the better. As a result, many of these detailed process management of ecosystems in the future.
models required very large amounts of data for cali-
bration. So great were the information requirements
of many of the models that the necessary data sets 21.4 Types of Model: Computer
were collected only for the sites for which the model Models Are Only One of Many
was originally produced. This problem of limited Types of Model
model portability has often restricted the application
of this type of model in resource management. Many different types of model have been produced
In contrast to the relatively short-lived but com- and used for a wide variety of purposes. Models can be
plex biological process-oriented ecosystem models, grouped into several broad classes that are described
several of the relatively simple tree and stand growth here in a sequence from the most fundamental to the
models produced in the 1960s and 1970s have per- most technically advanced type: the computer simula-
sisted in either original or modified form. The major tion model. The sequence starts with mental models
reason is that these models were simple, had low data that exist only in the mind (internally represented mod-
requirements, and served a defined and continuing els) and progresses through a variety of externally repre-
purpose: the prediction of stemwood volume produc- sented models (Figure 21-1). When a mathematical
tion. With low-intensity forest management, these model is implemented on a computer, it becomes a
models simulated growth acceptably. However, as the computer model. Computer models may also be de-
need for more accurate predictions of total biomass veloped directly from word models by translating
production under a wider variety of more frequent them directly into computer language, although this
management disturbances continued to increase, there almost inevitably involves the use of some mathemati-
has been a trend toward developing such models into cal notation.
ecosystem models.
There is now a growing use of computer gaming
with models of forest growth and/or succession to ex-
Internally Represented Models
amine the ecosystem consequences of alternative man- or Conceptual Models
agement strategies and climate change, to make Construction of mental models is something that we
predictions about future timber yields under changed all do every day as part of our normal process of think-
environmental conditions and management systems, ing. We perceive the conditions and entities around us
and to investigate the economic and social implica- with our senses of sight, sound, touch, taste, and smell.
tions of different methods of forest management. Each new perception is combined with previous per-
Much of this activity is still largely based on models ceptions to which it relates, and from the sum of these
that are basically mensurational descriptions of past perceptions we develop an understanding of our envi-
and present stemwood biomass accumulation and are ronment. This understanding, which may or may not
short on descriptions of the ecological processes that be an accurate description of that environment, can be
determine tree growth and ecosystem development. called a model because it provides a representation of
This raises the as-yet-unanswered question as to how the form and/or function of the conditions and enti-
useful such models will prove to be in predicting the ties being perceived. Such mental abstractions are
consequences for a wide variety of ecosystem values commonly referred to as conceptual models because they
and conditions of untested new silvicultural systems serve to unify related perceptions and because the
and of forest growth under changing environmental mental images that embody the model exist only in the
conditions (see Kimmins, 1985). thoughts and memory of an individual mind.
Every type of modeling has its own particular set The formulation of conceptual models is ex-
of advantages and limitations. It is unlikely that any tremely important in all branches of science. This is
one type of modeling will dominate all others in the the way in which hypotheses are formulated concern-
future. Greatly increased power of computers, im- ing the structure and processes of ecosystems, and
proved ecological knowledge, and the development of conceptual models are the basis on which all other
 Types ofModel: Computer Models Are Only One ofMany Types ofModel 569

Internally-represented
Perceptions models

Bias

Store models

)
Short-term Long-term memory
Memory storage for
Recall models conceptual models
Thinking

Word models

Mathematical models

Computer simulation
models
Externally-represented
models

Figure 21-1 Diagram of the relationship between internally and externally represented
models, showing various types of the latter. (Kimmmins and Scoullar, 1984, Copyright Academic Press,
Inc. Used with permission of the publisher and the authors.)

models are developed. These other models are only as found other ways to represent conceptual models in
good as the original conceptual model. permanent, external forms. We carved physical models
using available tools and materials. We painted
picture models with less effort but at the expense of
Externally Represented Models That Are three-dimensional reality. We evolved a written form
Developed From Conceptual Models of language so that we could record our verbal com-
The conceptual models of our mind can be stored munications as word models. Science and technology
only in our memory, and the verbal models that we use have since allowed us to develop and apply mathe-
for communication exist only while the words are matical techniques in the building of mathematical
being spoken. Modern acoustical recording equip- models that can be converted into computer simulation
ment allows the verbal model to act as a permanent, models by translating the former into computer lan-
although inaccurate, record of the conceptual model guage.
and thus provides a means for storing conceptual The distinctions between the different types of ex-
models external to the memory. However, long be- ternally represented models just mentioned are not al-
fore the invention of voice recordings, humans had ways clear. Pictures may be painted or etched so
570 CHAPTER 21 Models and Their Role in Ecology and Resource Management

thickly that the forms begin to take physical shape. A adept at running word models is evidenced by the eco-
written language may evolve from pictures simplified nomic success of book publishing companies. The
to the symbols of stick drawings. Mathematical models greater pleasure that some people derive from reading
often use words and letters as variable names, and a good novel rather than seeing a film based on the
computer simulation models are mathematical models novel is often because highly personalized and imagi-
written in a notation recognized by a computer. The native conceptual models can be developed from the
different types of models are frequently used in combi- word model (the novel), whereas the film (a picture
nation, as in the case of a physical model with painted model) permits little or no conceptual modeling.
details, a picture model that uses words for labels, and Mathematical models are run by completing the
a word model that uses numbers to define quantities. series of calculations that constitute the model. The
The conceptual models of our mind may be more final results of all the calculations are themselves a run
realistic than externally represented models because of the model. With a mathematical model translated
they are not subject to constraints imposed by the into the form of a computer simulation model, we
availability of materials and techniques with which to must run the computer program on a digital computer
build or express realistic external models. However, and allow it to do the calculation for us. The final re-
conceptual models are less rigorous than externally sult, or output, from the computer constitutes a run.
represented models because the thoughts that embody The following are types of externally represented
our conceptual models are fleeting and tenuous and models
are not available for scrutiny by other people. For ex-
ample, we cannot evaluate the conceptual model of a 1. Word models: the word model is the most com-
sculptor until he or she converts his or her vision to a monly used method for externally representing
physical model by carving it in solid form. Even our conceptual models. Most scientific articles are
though the sculptor may be disappointed in his or her word models of the systems under study, often ac-
ability to capture the realism that had been envisioned, companied by picture, graph, and flowchart mod-
the finished sculpture is more rigorous than the vision els (see below).
because it is persistent, unchanging, and available as a he Physical models: three-dimensional representa-
perception to other people for evaluation by compari- tions of real objects or systems, which are gen-
son with their own conceptual model of the object erally of limited use in forest ecology other than for
represented in the sculpture. It is this rigorous nature education and display (e.g., a diorama of the seral
of externally represented models that has resulted in stages of postlogging succession on different sites).
their widespread use. 3. Picture models: two-dimensional representations of
The procedure that one must follow to put a model three-dimensional objects are widely used for edu-
to use is commonly termed running the model. Run- cation and demonstration. A picture may also be
ning a conceptual model involves recalling to mind used as a static model of the present state of the sys-
from long-term memory the thoughts that embody tem. Picture models include maps and histograms.
the model. Externally represented models may be run
either in the mind or external to the mind, depending 4. Flowcharts: a flowchart is an externally represented
on the kind of model involved. Both picture models dynamic model, generally with word submodels (la-
and static physical models can be run by perceiving the bels). It can depict the interactions between the com-
details of the model and allowing these perceptions to ponents of a system, the transfer of materials within a
direct the formation of a conceptual model in the system, and/or the sequence of events in a process.
mind. In the case of a dynamic physical model (e.g., a 5. Graphs: externally represented dynamic models
model that works like a machine), the model must be that generally combine pictures, words, and math-
turned on and allowed to run on its own. ematics, graphs are most commonly used to depict
‘To run a word model, we must read or hear the changes in a component or a process over time or
words that embody the model. We must recall from the change in one component or process as an-
memory the thoughts and images that we associate other component or process changes.
with the words and thus be led to form an appropriate 6. Mathematical models: a mathematical model, an
conceptual model. That literate humans have become external representation of a conceptual model or
 Stages in the Development of a Computer Simulation Model 571

of another type of external model using mathemat- necessary to achieve the modeling objectives
ical notation, may be static (describing state vari- (Occam’s razor).
ables: analogous to an inventory) or dynamic
Representation of the conceptual model in exter-
(describing both state variables and system vari- nal form so that it can undergo evaluation and de-
ables, which define how state variables change over velopment. This may involve any combination of
time). Some mathematical models stress the math- word, physical, picture, or other nonmathematical
ematical viewpoint where accuracy of biological external types of model.
description is compromised in favor of using
. Definition of state and system variables. The com-
“powerful” and/or convenient types of mathemati-
ponents of the system must be defined (state vari-
cal techniques. Biologically biased mathematical
ables) as well as the processes (system variables)
models compromise the type of mathematics that
responsible for transfers between state variables.
can be used in favor of accurate description of the
This stage often involves the development of a
biology of the system. Most mathematical models
flowchart that shows how the processes link the
are intermediate between these two extremes.
components.
7. Computer models: mathematical models can be
Selection of the length of the simulation’s time
translated into computer language and imple-
steps. How often are the transfers between state
mented on a computer.
variables calculated? Daily, weekly, annually?
. Formulation of transfer variables—the develop-
21.5 Stages in the Development of a ment of the mathematical equations that define
transfer rates and that are needed to calculate val-
Computer Simulation Model ues for state variables at each time step. This con-
verts the model developed in steps 4 and 5 into a
Because of the growing importance of computer simu-
mathematical model.
lation models, students of forest ecology and resource
management should understand the main stages in the Construction of a flowchart of model execution.
development of this type of model. Construction of Before the mathematical model can be translated
such a model is not difficult if the modeling activities into computer language for implementation on
are ordered in the right sequence. the computer, the order in which the calculations
are to be made must be identified. This is often
done in the form of a flowchart of model execu-
1. Establishing the objectives of the model or model-
tion.
ing activity—the goal to be achieved by develop-
ing the model. Conversion of the mathematical model into a com-
puter simulation model—translation of the mathe-
2. Development of an educated conceptual model.
matical notation into computer language and
The quality of the predictions of the computer
preparation of the necessary input and output files.
model will depend on how well the conceptual
model describes the real system. There is a ten- 10. Running the model on the computer. The com-
dency to rush this critically important stage and to puter language must be translated by the computer
develop the conceptual model too narrowly. Re- into machine language and then run to check for
member, the success of the modeling activity is programming errors.
largely dependent on how well the conceptual LL Calibration of the model. Parameters in the
model is developed. model that define system variables are given val-
3. Selection of the scope (i.e., the boundaries of the ues that result in realistic predicted values in the
model: how much of the world is described in the output of the model.
model) and the resolution (the level of detail rep- . Verification that the model is operating as in-
resented) of the computer model. The model tended, that its calculations and output accu-
should include only those processes and compo- rately reflect the knowledge represented in the
nents that are necessary to meet the objectives. It model, and that the calibration data are used cor-
should be as simple as possible but as complex as rectly.
572 CHAPTER 21 Models and Their Role in Ecology and Resource Management

13. Validation’ or confirmation of the predictions. The the ecological processes of biomass accumulation, nu-
reality of the model’s predictions must be tested trient cycling, and secondary succession and on recog-
against a data set other than that used for calibration. nition of the impact of forest harvesting thereon.
14.Gaming with the model: running the model Extensive forest management generally places only
(“gaming” or “playing”) to investigate properties moderate to low nutrient demands on a forest site.
of the simulated system, the sensitivity of the Harvesting only the largest logs (mainly heartwood,
model’s prediction to the quality of input data and which has low concentrations of most nutrients) at in-
the assumptions that were used in building the frequent intervals (long rotations) removes only a
model (sensitivity analysis), and/or its response to small proportion of the site nutrient capital; such
perturbation. losses are more than replaced over the next rotation by
natural inputs from the geochemical cycle (cf. Chapter
15. Evolution of the model. Steps 2 to 13 are repeated 5). This situation changes if the intensity of manage-
in the light of deficiencies revealed during steps 12 ment and utilization of tree biomass is intensified.
to 14 until the model performs acceptably in its in-
Shorter rotations, intermediate harvests (thinning),
tended application.
and whole-tree (all aboveground biomass) or even
16. Using the model for the purpose for which it was
Y
complete-tree harvesting (both above- and below-
intended. Sometimes this is for making predic- ground biomass) greatly increase nutrient outputs
tions. Sometimes it is for generating new hypothe- from the ecosystem and may result in a long-term net
ses for experimental testing. Sometimes it is for reduction of the site nutrient capital and productive
synthesizing complex sets of data to test our un- potential. To reduce the risk of site degradation, the
derstanding of asystem. In some cases, the value of potential for this problem must be evaluated before
building a computer model lies in the learning any change from extensive to intensive management.
process involved in building it; when it is finished, The forest manager needs to be able to answer this
it has no further use. question: Will the increased nutrient withdrawals that
accompany the switch from extensive to intensive for-
21.6 Application of Different Types of est management and biomass utilization result in a
decline in the productive capacity of the forest ecosys-
Model to the Study of Site Nutrient tem? The complex understanding that is required to
Depletion Through Harvesting of answer this question has been expressed in a variety of
Forest Biomass modeling approaches.

Most of the types of model described above have been

used in both basic and applied forest ecology (i.e., for- Conceptual Models
est resource management). Numerous examples can be The most fundamental type of modeling that has been
found in this book (e.g., the models of ecological suc- used to evaluate this question is the creation of a con-
cession described in Chapter 17), but to illustrate the ceptual model in the mind of the scientist or resource
diversity of types of model that can be applied to a par- manager who is thinking about it. By considering what
ticular topic, we consider models that have been used one knows of tree nutrition and its relationship to bio-
to evaluate the significance for future biomass produc- mass production and by combining this with one’s
tion of nutrient removals in harvested forest products knowledge of the normal geochemical inputs to and
(Kimmins, 1977). Reliable predictions about this effect outputs from the forest in question, the biogeochemi-
of harvesting must be based on an understanding of cal cycling within the ecosystem, and the internal con-
servation of nutrients within the trees by internal
cycling, one can develop a mental image or conception
*There is considerable debate about the use ofthese terms. Some people ob-
ject to the use ofverification and validation because they imply that models
of the problem from which conclusions can be drawn
are “true” or “correct,” rather than merely imperfect representations ofre- about whether intensive biomass harvesting will affect
ality. It is extremely difficult to verify completely a complex mathematical future productivity in the area.
model and impossible to validate models that predict hundreds ofyears into Such a conceptual model of a forest ecosystem is
the future. The term bench-marking has been suggested as a replacement limited by the mental capacity of the modeler. Just
for verification, and confirmation for validation (Oreskes et al., 1994). how many of the varied states and processes involved
 Application of Different Types ofModel — 573

in the problem can be thought of at one time? The the model. Alternatively, pictorial models of nutrient
mental confusion that attends attempts at conceptual depletion can be created directly from the original
modeling of very complex systems results in a lack of conceptual models. The resulting pictorial models can
clear definition of the magnitudes of states and of the be qualitative or quantitative, depending on whether
rates of processes. Consequently, any conclusions are mathematical measurements have been included.
at best broadly qualitative. Because of these deficien- Figure S—1 is an example of a qualitative pictorial
cies, attempts at conceptual modeling of harvest- model of nutrient cycling. Quantitative pictorial flow-
induced site nutrient depletion and yield reduction charts of nutrient cycling in forest ecosystems, such as
have frequently led to the next stage—word modeling. Figure 5-15, can be modified to summarize the prob-
lem of nutrient withdrawals at the time of harvest. Al-
Word Models ternatively, the word model can be summarized in a
By carefully describing the conceptual model in words diagram that compares the nutrient withdrawals at dif-
ferent intensities of harvest with a static budget of
and recording this verbal description in writing or on
total site nutrient reserves, thereby omitting the infor-
tape, word models of the risk of nutrient depletion
mation on nutrient dynamics (Figure 21-2).
have been produced. These models contain far more
This type of pictorial-mathematical model shows
detail than can be called to mind at any moment in a
conditions at one moment in time but fails to describe
conceptual model. Clear definitions of ecosystem
many of the important ecosystem processes. Small
components are given, and detailed descriptions of
quantities of nutrients being rapidly exchanged and
ecosystem processes are provided. Quantitative esti-
circulated may be more important than large quantities
mates are placed on the magnitudes of biomass and
of relatively immobile nutrients, but the rate and mag-
nutrient compartments and on the rates of processes
nitude of nutrient turnover are not represented in
by including mathematical measurements in the word
Figure 21-2. An improvement is obtained by focusing
model. Examples of such word models can be found in
on the rates of nutrient circulation, thereby emphasiz-
the early literature on nutrient removal in harvested
ing the functional relationships in the system rather
products and its effects (observed or predicted) on bio-
than the inventory of nutrient quantities in different
mass production. The first of these is probably the
components. Figure 21-3 shows an example of this
writing of a German forest scientist, Ebermeyer
kind of pictorial model for a Corsican pine stand in
(1869). He published a word model (a scientific de-
eastern Scotland. Diagrams of this sort can be prepared
scription) of the effects of litter raking in German pine
for stands of different ages and subject to different con-
forests on the nutrient status of the site and the rate of
ditions, thereby permitting conclusions to be drawn
growth of the trees. Decades of litter raking and re-
about temporal changes in function and the effects of
moval for use in agriculture resulted in a significant
management thereon. However, this type of model still
reduction in site nutrient status and tree growth. ‘Ter-
falls short of a truly dynamic description of ecosystem
mination of litter removal resulted in a slow recovery
processes. If enough diagrams of this type were made,
of soil fertility and tree growth rates. Similar word each one representing the state of the system each year
models, complete with tables of numbers defining
over a rotation, then an animated film could be made
quantities and amounts (thereby rendering the model
and a dynamic pictorial model could be produced.
a hybrid between word and mathematical models),
were presented by the English forest ecologists Ren-
nie (1955, 1957) and Ovington (1957, 1959), who, Computer Models
building on the concerns of Ebermeyer, stimulated the Neither word nor pictorial models have proved to be
past two decades of interest in this topic. A review- appropriate tools with which to make quantitative pre-
type word model of the problem can be found in Kim- dictions of the risk of management-induced site nutri-
mins (1977). ent depletion. They cannot deal with the complexity
and dynamic character of the problem, and most re-
Pictorial Models searchers concerned with this topic eventually turn to
The word models described above have been summa- computer models. Mathematical equations that de-
rized into pictorial models with word labels, diagrams, scribe rates of transfer of materials between ecosystem
or flowcharts that more clearly display the structure of components are developed, and when the system of
574 CHAPTER 21 Models and Their Role in Ecology and Resource Management

Stand biomass Nutrient content in above

(t ha”!) ground tree biomass (kg ha —)

Tree heights Ca
(m) 100 240
15
ah per oe MIT Nutrients removed in conventional

10
aL 715 stem harvesting
Additional nutrient removals
£At3 30 incurred by whole-tree harvesting

2)
25
aaa ja K PMg
Y
VY Forest floor Y) Forest floor LN Forest floor
10 A horizon 4
20 3 B+C 250
30 _ horizon es) Mineral
40 500 soil

~750 z

1000

Ay. sampling Dry wt. of “Exchangeable” soil Total soil nutrient

depth (cm) sampled soil nutrients (kg ha!) reserves (t ha” !)
(t ha” ')

Figure 21-2 Pictorial model of nutrient withdrawals in harvested products at two different
levels of utilization. Data for a 65-year-old black spruce stand north of Baie Comeau, Quebec.
(After Weetman and Webber, 1972. Copyright National Research Council of Canada. Used with permission
of the National Research Council of Canada and the authors.)

equations is solved (i.e., the model is run) on a com- shrubs, and trees for light, nutrients, and moisture and
puter, estimates can be obtained of future sizes of can represent the role of mosses (or other bryophytes)
various ecosystem components under a variety of cir- in nutrient cycling. There is a simulation of nutrient
cumstances. inputs to and losses from the ecosystem via the geo-
Several ecosystem-level models have been devel- chemical cycle; of nutrient circulation within the
oped to examine the problem of site nutrient deple- ecosystem by litterfall, foliar leaching, reproduction,
tion (e.g., LINKAGES (Pastor and Post, 1985, 1988; herbivory, plant mortality, or forest management activ-
Pastor et al., 1987), which built on earlier work by ities (e.g., thinning, pruning) and by organic matter de-
Aber et al. (1978, 1979, 1982, and Aber and Mellilo, composition; and of nutrient conservation within
1980); FORCYTE (Kimmins, 1993b; Kimmins and plants by internal cycling. All major stand-level forest
Scoullar, 1994); and FORECAST (Kimmins et al., management practices can be simulated: site prepara-
1999, Seely et al., 1999)). Because LINKAGES was tion (by fire or mechanical methods), regeneration (by
developed from the “ gap model” type of forest succes- planting, natural seeding, or vegetative reproduction),
sion model, it is discussed later. FORECAST control of competition between crop and noncrop
(FORestry and Environmental Change ASsessmenT) plants (“weeding”), tree density control (spacing, pre-
is described as an example of a forest ecosystem man- commercial thinning), pruning, fertilization, commer-
agement model that incorporates nutrient cycling and cial thinning, final harvesting (by clearcutting,
nutritional regulation of forest growth and succession. shelterwood, or uniform uneven-age management
The major compartments and transfer processes with partial harvesting), rotation length, utilization
represented in FORECAST are shown in Figure 214. level, coarse woody debris (CWD), and snag and
FORECAST simulates competition among herbs, “Wildlife tree” retention. Natural disturbances such as
 INPUT
2 RR
TREE CROP Aerosol Rain

Accumulation
| Annual
in
increment
> tree crop Foliar absorption
8
al
S| Crown leaching
“_~ rt] :
= Mobile
Se pool in
shoots
Immobilization
Root death

' ae SOIL
tix, Uptake from Accumulation in Annual | Organic
forest floor forest floor increment layers

Uptake from Accumulation Mineral

mineral soil ee layers
mineral soil

Net loss
(soil leaching)
(a)

(1)
se : Fertili
ertilizer Fertilizer
[47
a
ie AT 168
Sahl!
—
| eet

37

(b)

Figure 21-3 Pictorial model of the nitrogen capital and nitrogen cycling in a Corsican pine
stand in eastern Scotland. (After Miller et al., 1979.) (A) Generalized model. (B) Models for two
different rates of nitrogen fertilizer input: (1) 84 kg nitrogen ha! yr! for 3 years and (2) 168 kg ni-
trogen ha"! yr"! for 3 years. These smaller flow diagrams are pictorial-numbers models based on the
pictorial model. (Copyright National Research Council of Canada. Used with permission of the National Re-
search Council of Canada and the authors.)

wa wa
 enenenneeet

FERTILIZATION
% j INPUT
Trees
Z : ACHIEVED
Shrubs NET
; san DU PRIMARY
Herbs moisture PRODUCTION PRECIPITATION,
. j limitation INPUT
ryo- i
phytes =
%5 Pro anaveeseesnensone TNE
Others :
5 \l SLOPE
: : SEEPAGE INPUT
; La ~

sms FOLIAR vovensnnef WE :

Leaf aenfjis. |
NITROGEN = :

otsace Ning | Nr
Stem : 3 Model :

Branch : EFFICIENCY Function Availability

d - yee of other :
nae : nutrients Se
uctive Hy Be

FINE os
ROOT B LOSS
MASS
: “For NO,_, does not
tas differentiate between
pas leaching and denitrification

PLANT BIOMASS AND AVAILABLE i

NUTRIENT CONTENT SOIL NUTRIENTS tia

LOSS

SOIL (FOREST FLOOR, MINERAL SOIL)

LOSS

Figure 21-4 FORECAST flowchart showing the major compartments and_ transfer
processes that are simulated. The model is driven by a representation of canopy function: foliage
nitrogen efficiency. Solid lines indicate material transfer processes; dotted lines indicate feedback
controls or process regulators.
76
wr
 Three Major Approaches to Modeling in Forestry = 577

wind, fire, and insect outbreak can be represented. The 21.7 Three Major Approaches to
user can simplify this complex model at will; the model
does not require that all these compartments and
Modeling in Forestry
processes be represented. The model can be reduced to
There are three major types of computer model that
a light competition model with only a simplified repre- have been used to simulate the growth and develop-
sentation of trees or can represent the full plant com- ment of forests (Kimmins, 1988, 1990): historical
munity competing for light and soil resources and with bioassay, process simulation, and hybrid.
a representation of the limiting effects of soil moisture.
The model can represent one or several age classes or
one or several tree species, with or without other plant Historical Bioassay Models: Prediction
life forms. Based on Experience
Some examples of how FORECAST can be used to Historical bioassay models take data that define how
examine the interactive effects of rotation length, forests have grown in the past on a particular site and
utilization level, and thinning on stemwood mass, stem- use them to define how the forests on that site will
wood harvest, forest floor mass, CWD (decaying grow in the future. This is the approach used in the
wood), ecosystem nitrogen capital (total and available), mensurational models used by foresters to predict fu-
and leaching loss are shown in Figure 21-5. The model ture supplies of timber. It assumes that our experience
produces similar output for up to 100 variables for each of how forests have grown in the past is the best pre-
tree species, up to 60 for each herb or shrub species, and dictor of their future growth.
as many as 90 soil variables, depending on how many Historical bioassay models are reliable and believ-
plant species, nutrients, and litterfall types are simu- able predictors as long as all the ecological conditions
lated. It also predicts stand-level habitat values for sever- that influenced forest growth and development in the
al wildlife species, the economics of stand management, past remain unaltered, and the data they are based on
energy benefit—cost ratios, and carbon budgets. are an accurate record of what actually happened his-
The predictions of computer models over time torically. The former assumption is unlikely to be true,
scales as long as those shown in Figure 21-5 are not and the latter is questionable more often than we
believable in terms of the actual numbers predicted. would like (Kimmins, 1990). These models provide no
Consequently, the main value of models such as explanation of the mechanisms of change, and are
FORECAST is in predicting anticipated trends in poor predictors of forest growth and succession under
ecosystem development and in ranking the possible different management and environmental conditions.
consequences of alternative forest management strate-
gies. We cannot predict exactly what a forest ecosys-
Process Simulation Model: Prediction Based
tem will be like in 100 years as a result of a particular
management regime, but if our ecosystem models are
on Knowledge
reasonably accurate, then we nevertheless should be The inability of historical bioassay models to make be-
able to rank the various management alternatives that lievable predictions for ecosystem development under
are available to us, and predict the possible range of changing environmental conditions and disturbance
forest futures. regimes or to explain ecosystem change over time led
Another major value of such models in heuristic: ecologists to develop process simulation models.
they can help us to explore the complexity of real- These represent our current knowledge of what the
world issues; identify knowledge and data gaps; and modelers believe are the key ecological processes that
guide future research, inventory, and management determine forest growth and succession.
planning. However, the user should be constantly Numerous process models have been developed
aware that models may also support our biases and over the past 30 years. These predict net primary pro-
confirm our frequently incorrect intuition because duction, carbon flows, and stand development on the
they are generally based on an incomplete and fre- basis of the ecophysiological relationships between
quently inaccurate representation of reality. Models growth and environmental conditions. Process-based
also rarely can predict the risks of future disturbance models have the ability to forecast aspects of tree
events and their consequences for ecosystem values as growth under a wide variety of future conditions, as
accurately as we would like. long as the processes that are expected to change are
 —

Neo
Total stemwood harvested Mass of forest floor

Rotation Length (years)

Rotation Length (years)
Utilization Thinning 60 80 120
eee 0) 80 120
Utilization Thinning S
yes _ :

stem-onl
(80% stem)
wi stem-only
yes

oa a
(80% stem)

no

yes
yes
whole tree
initial
of
Percent
value initial
of
Percent
value
(stem + crown)
whole tree
no (stem + crown)
no

(01 M@m@2 Os B4
[Rs ~2 G7: a sae si |
Rotation number
Rotation number
(b)

Mass of coarse woody debris Soil nitrogen capital & availability

Rotation Length (years) Rotation Length (years)

Utilization Thinning 60 80 120 Utilization Thinning 60 80 120

yes yes

stem-only stem-only
(80% stem) (80% stem)
no no

yes
initial
of
Percent
value initial
of
Percent
value
whole tree whole tree
(stem + crown) (stem + crown)
no no

Lee 3) 4) eee | bho

Rotation number Rotation number

(c) (d)
Figure 21-5 Examples of some output from FORECAST. Shown are the interactive effects of
rotation length, utilization level, and thinning on stemwood harvest (A), forest floor mass (B), mass
of CWD (decaying wood) (C), ecosystem nitrogen capital (total and available) (D), and nitrogen leach-
ing loss (E). Simulation based on calibration data for lodgepole pine (Wei and Kimmins, 1996).

|~—fore)
1a
 Three Major Approaches to Modeling in Forestry 579

N loss by leaching Process simulation models are the ultimate “sci-

entific” approach to ecosystem modeling. They at-
Rotation Length (years) tempt to explain change in terms of the processes that
Utilization Thinning are represented: they are both explanatory and de-
scriptive. Process-based ecosystem models that in-
yes clude all significant determinants can also be excellent
predictive models, but they have the limitation that
stem-only
(80% stem) they are very complex, often are hard to understand,
o
= and may be very slow even on fast computers. Most
no S
> process models omit one or more key processes and
S therefore are limited in the ecological and forest
ee management questions for which they can be used to
i)
=!

yes r= address. ‘These limitations have restricted their appli-

2
oO
cation in resource management, but such models nev-
whole tree A ertheless have made an important contribution to the
(stem + crown)
integration of knowledge in forest ecology. There has
no
been little use of such models in applied forestry deci-
sion support applications to date, but it is clear that
incorporation of process simulation into forest deci-
Rotation number sion support tools will increase in the future (Kor-
zukhin et al., 1996). Simplifications of complex
Figure 21-5 (Continued)
process models are becoming increasingly relevant in
forest management as the need to be able to predict
the long-term consequences of untested management
systems under changing climatic conditions becomes
represented within the model. Iwo important limita- more urgent.
tions of these models are the degree of complexity that
can develop as more processes are included and the
lack of ability of forecast future growth under chang-
Hybrid Models: Prediction Based on Both
ing ecosystem conditions if all the key processes that
may change are not included. Well-known stand
Experience and Knowledge
process-based models are BIOMASS (McMurthie and To gain the advantages that both historical bioassay
Landsberg, 1992), FOREST-BGC (Running and and process-based simulation models offer but with-
Gower, 1991), and MAESTRO (Wang and Jarvis, out their major drawbacks (inability to account for
1990). A useful review of some of these models can be changing conditions and excessive complexity, respec-
found in Landsberg and Gower (1997), and Lands- tively), many forest ecologists have used a hybrid ap-
berg and Coops 1999. proach to modeling ecosystems. These models are
MAESTRO is one of the best-known “pure” forest driven by historical bioassay growth equations or em-
process-based models. It simulates net photosynthesis pirical growth data, with the predicted growth modi-
and transpiration at the leaf level within various sec- fied by or based on the simulation of those processes
tions of the crown of individual trees within a stand. It that are believed to be most important. This gives the
also incorporates details of the structure and arrange- models a believable, empirical foundation and adds
ment of leaves and the distribution of light within the sufficient process simulation to provide some flexibil-
crown. MAESTRO has been used mainly as a research ity in the face of change in future growing conditions
tool to investigate the interaction between the climate while avoiding the complexity of a full process model
and various canopy exchanges processes, but it could (see Kimmins [1993b] for a more complete discus-
also be used to evaluate the biomass consequence of sion). The growing recognition of the power of hybrid
various silvicultural treatments in structurally simple simulation models in forestry justifies a more detailed
stands (Landsberg and Gower, 1997). look at this modeling approach.
580 CHAPTER 21 Models and Their Role in Ecology and Resource Management

21.8 Examples of Hybrid gap models represent the canopy as an “opaque blan-
Simulation Models ket”: the foliage of each tree is spread out evenly ty
oor
-.
6

across the entire gap. ZELIG and models derived

There are three groups of hybrid models: the “gap” therefrom provide some spatial representation of the
models of ecological succession; stand-level ecosystem canopy by simulating a grid of square cells and by con-
management simulation models; and an emerging sidering the shading of trees in one cell by the trees in
group of individual tree-growth models that represent the surrounding cells. Some models account for the
the horizontal as well as the vertical spatial distribu- angle of the sun in the sky, showing that larger gaps
tion of plants and, in some cases, soils, microclimatic are required as you move from the equator toward the
conditions, and other site features. poles to achieve a particular gap light regime and
radiation budget. There has also been a recent trend
toward incorporating canopy processes and ecophysi-
ology in some gap models, giving them more biologi-
Gap Models cal reality and moving them closer to true process
In a development that was far ahead ofits time, models.
Botkin et al. (1972, 1993) created a forest succession The family of gap models that originated from
model of northern hardwood forests that was based JABOWA has provided numerous forestry applica-
on a growth equation. Tree growth in this model is a tions (Botkin, 1993). In recent years, several models
function of diameter at breast height, the equation and modeling approaches have been linked together
representing the genetic potential of the species to address various issues. For example, FORSKA is a
under nonlimiting conditions, with the predicted gap model that simulates assimilation and respiration
growth reduced by four modifiers: availability of rates, with annual growth of each individual tree then
light, annual rate of nitrogen mineralization in the calculated as a function of net assimilation in its verti-
soil, availability of soil water, and temperature regime cal leaf area profile scaled by a species-specific growth
(growing season degree-day sum) (Figure 21-6). The rate factor (Leemans, 1992). For evaluating the past
original model, JABOWA, was referred to as a “gap” and possible future impact of management on carbon
model because it simulates the growth and succession sequestration in forests, FORSKA and carbon budget
of tree species in a small plot (e.g., 0.01 to 0.08 ha) models also have been linked (Price et al., 1997). Fi-
that represents the canopy gap created by the death of nally, for evaluating alternative silvicultural regimes in
a single large tree. In this model, trees are established the Pacific Northwest, ZELIG has been used with
as saplings, avoiding the complexities involved in sim- economic models and habitat suitability regressions to
ulating germination, establishment, and seedling explore the impacts of alternative forest harvesting
growth, and they grow according to their diameter- patterns on logging revenues and bird density
related growth equation, limited by simulated light (Hansen et al., 1995),
competition, and input data on soil moisture, soil ni- Gap models have been criticized because they have
trogen, and temperature. to a great extent retained a modeling approach that
The original JABOWA model has been elaborated was dictated as much by the computational power of
into an extensive family of models, including FORET computers in the late 1960s as by the optimum way of
(Shugart, 1984), FORSKA (Prentice and Leemans, representing forest ecosystems. Although modifica-
1990), LINKAGES (Post and Pastor, 1996), and tions have steadily improved these representations,
ZELIG (Smith and Urban, 1988) (Figure 21-7 and some of the gap models being used today still have
Table 21-1; see also Shugart [1998] for a comprehen- very crude and generalized representations, which are
sive review). Some of these have added a simulation of inaccurate for some forests (Pacala et al., 1996). One
nitrogen availability based on a simulation of decom- of the problems is the failure to represent individual
position of simulated litterfall. Others have improved tree canopies: the “opaque blanket” approach poses a
the simulation of canopy representation, from having fundamental limitation. Another problem is that the
all the foliage at the very top of the tree to represent- driving function is an idealized mensurational growth
ing both canopy top and canopy bottom with foliage function that bears no relationship to true canopy
distributed with height in the crown. However, most function. The growth modifier approach of the gap
 Examples ofHybrid Simulation Models — 581

Growth Driving Function

Diameter
Increment
Diameter
Height
Breast
at
Stem Diameter at Breast Height Tree Age
(a)

Growth Modifiers

x2
Shade Drought
S Tolerant Tolerant
& Z
Es
© 0.5 Drought
0.5
Oo Intolerant
oO
nes
= / ‘
= aes Light
me is Demanding
0
Available Light Soil Moisture Index

| Low &: ESS

Cool-adapted ue \ Warm-adapted
g Nutrient ye i /
Species \Species
is Demand vy High
= } Nutrient if \ \
2 OS ; Demand 0.5 “—
\)
oO
a
3
3)
~~

Available Soil Nitrogen Growth Degree-Days

(b)

Figure 21-6 Tree growth driving function (A) and growth (B) modifiers as used in
JABOWA-derived gap models. (Modified from Urban and Shugart, 1992.) Height, leaf area, and
woody biomass are calculated as a function of stem diameter.

model is computationally efficient and was a stroke of omit soil processes. Although some simulate annual
genius at the time it was derived more than 30 years nitrogen availability by mineralization processes, there
ago, but it fails to provide adequate representation of is no simulation of nutrient uptake demand and the
canopy function for today’s applications. Another fea- consequences of nutrition for canopy function, above-
ture of most gap models is that they focus on auto- ground—belowground resource allocation, and inter-
genic processes and are less well equipped to represent nal nutrient cycling strategies. From the work of
large-scale allogenic or biogenic processes. Tilman discussed in Chapter 17 and the discussion of
Most gap models omit herbs and shrubs: a major resource allocation in Chapter 4, these omissions
limitation in simulating succession where there are fa- would seem to limit the application of many of the gap
cilitation and inhibition mechanisms at work. Most models.
 582 CHAPTER 21. Models and Their Role in Ecology and Resource Management
3
: we
KiWi 7:
2000s JABOWAA-III LINKNZ
1999 LETT* BROLLA FM 97.5 oe a
1998 FORMAN DRYADES ZELIG.CBA
1997 FORMAL EDEN FORSKA-M SORTEDC A
1906 MOSEL arBGC Ey wed GIZELA 1X:

aa FORCLIM I | TUNDRA MIOMBO | FORMIX:.

1994
|a M ALINKAGES.BC | — rar HYBRID FORSKA-2 sree : i
1993 JABOWA-I
1992 i
! as I sane KOPIDE SPACE-1 | MOHREN* ' | t FLAM i FORMIX
199]
1990 | MENAUT* OVALIS ZELIG-1 i
LOKI OUTENIQUA A STEPPE FORSKA-1 |
1989 FIRESUM
FORENZ ZELIG aca |
1988
1987 E
ok SPRUCE-FIR |
1986 FORCAT A |
Logs fta FORFLO | | |
SMAFS | | |
198434 ci ACS | |
1983 SJABO
1982 Doan FORNUT | i
198] asA BRIND —FORICO |
1980 KIAMBRAM |
1979 SWAMP | I I
1978 FORAR FORMIS Hy | |

1977 aa mamma ceri cs

1976 | | I
oa
1974
| | |
| |
|
1973
lop an © Bite ee ee 2 Pe ya
JABOWA ee
oe Lbs Sei ae ee

7p
Figure Ey| Evolution of gap models from the original JABOWA model. (After D. Mailly, un-
21-7
published manuscript. Used with permission.) A dashed line represents a slight departure from the origi-
nal JABOWA model. Models without an official name are named for their senior author. Authors
and key references are provided in Table 21-1.

Stand-Level Ecosystem Management Models bioenergy plantations. Descriptions of the develop-

ment of the FORECAST modeling approach can be
We will examine two examples of this class of model:
found in Kimmins (1993b), Kimmins et al. (1990,
FORECAST and TRIPLEX.
1999), and Seely et al. (1999, 2003).
Forecast Although gap models have been used to The FORECAST model uses empirical historical
investigate the response of forest ecosystems to man- bioassay data to derive estimates of the rates of key
agement-related disturbance (e.g., Aber et al., 1978, growth and ecosystem processes. This is a “backcast-
1979; Yarie, 1986a), their major use has been to inves- ing” approach to process modeling that roots the
tigate forest succession and the possible effects of cli- process simulation firmly in the empirical field data.
mate change thereon. This reflects the original These rates, together with empirical data on processes
motivation for their development. In contrast, the that cannot be backcasted, are then used in the ecosys-
FORECAST forest ecosystem management model tem simulation. The overall approach was to develop a
was designed from the outset as a means of analyzing modeling framework rather than a single multipurpose
the consequences of forest management for ecosystem model, a strategy consistent with suggestions for forest
function, sustainability, and harvestable productivity. stand modeling by Makela (2001) and Stage (2001).
The genesis of FORECAST was concern over Net primary production in FORECAST is simu-
maintaining tree growth and timber harvest yields in lated as a function of foliage biomass and light inter-
the face of changing soil organic matter-and nutrient ception. As part of an internal calibration procedure,
inventories. FORECAST also examines the social val- the model calculates a measure of foliage nitrogen ef-
ues of stand-level economics and energy benefit-cost ficiency (net primary production per unit of foliage ni-
ratios, employment, and carbon budgets (carbon se- trogen content) by comparing estimates of total
questering, storage, and release). The model’s origin biomass production with measures of foliage nitrogen
was the threat of an Arab oil embargo in the early and simulated levels of light interception within the
1970s and the need to assess the sustainability of canopy. This estimate of foliar nitrogen efficiency
short-rotation, intensively managed, and harvested under varying light profiles provides the driving func-
Table 21-1 Models, the Main Geographic Location and Forest Type of Their Application, the Author, and Key
Literature Citation for the Family of Gap Models Presented in Figure 21-7
Model Forest Type Location Location Reference

JABOWA Northern hardwood New Hampshire Botkin et al., 1972

FORET Appalachian hardwood E. Tennessee Shugart and West, 1977
FORMIS Floodplains Mississippi Tharp, 1978
FORAR Pine, oak—pine Arkansas Mielke et al., 1978
SWAMP Floodplains Arkansas Phipps, 1979
KIAMBRAM Subtropical rain forest Australia Shugart et al., 1980
BRIND Montane eucalypt Australia Shugart and Noble, 1981
FORENA Deciduous forests E. North America Solomon, 1986; Solomon et al., 1981, 1984
FORICO Rain forest Puerto Rico Doyle, 1981
SILVA Montane conifer California Kercher and Axelrod, 1981, 1984
FORTNITE Northern hardwood New Hampshire Aber and Melillo, 1982
FORNUT Appalachian hardwood North Carolina Weinstein et al., 1982

CLIMACS Pacific Northwestern conifer Oregon Dale and Hemstrom, 1984

SMAFS Acadian forest Nova Scotia El Bayoumi et al., 1984

LINKAGES Deciduous forests Eastern U.S. Pastor and Post, 1985, 1986

FORFLO Floodplains S. Carolina Pearlstine et al., 1985

FORCAT Deciduous Cumberland Plateau Waldrop et al., 1986

FORECE S. Central Europe Kienast, 1987; Kienast and Kuhn, 1989

SPRUCE-FIR Appalachian spruce-fir N. Carolina Busing and Clebsch, 1987

FORSKA Boreal conifer Sweden Leemans and Prentice, 1987,

1989; Prentice et al., 1993

FORENZ Mixed forests New Zealand DeVelice, 1988

ZELIG Appalachian hardwood Tennessee Smith and Urban, 1988; Urban, 1990, 1993

FIRESUM Montane conifer Montana Keane et al., 1989

BONAN* Boreal conifer Alaska, Canada Bonan, 1989

OUTENIQUA Subtropical rain forest South Africa van Daalen and Shugart, 1989

SERRE Semiarid grassland Colorado Coffin and Lauenroth, 1989

OVALIS Appalachian hardwood Virginia Harrison and Shugart, 1990

KELLOMAKI* Boreal forests Finland Kellomaki and Vaisanen, 1991

SPACE Appalachian cove hardwood N. Carolina Busing, 1991, 1995

MOHREN* Dry sandy soils The Netherlands Mohren et al., 1991

FLAM Spruce/pine forests Central Sweden Fulton, 1991

ZELIG.BC Pacific Northwest forests British Columbia Burton and Cumming, 1991;
Cumming and Burton, 1994

FORCLIM European Alps Switzerland Bugmann 1991; Fishlin et al., 1994

JABOWA-II Deciduous forests E. North America Botkin, 1993

LINKAGES.PNW Pacific Northwestern conifer NE Vancouver Island Keenan 1993

FOREST-TUNDRA Forest-tundra transition zone NE Canada Sirois et al., 1994

MIOMBO Tropical dry woodlands Central Africa Desanker and Prentice, 1994

*Unnamed models, ascribed here to the name of its author.

nN Ww
$84 CHAPTER 21 Models and Their Role in Ecology and Resource Management

tion for FORECAST (see Kimmins, 1986; Kimmins FORECAST is a stand-level model. It does not
et al., 1999). Foliar nitrogen content has been shown represent the spatial distribution of individual trees or
to provide a good measure of the photosynthetic ca- individual tree canopies. This poses limitations on the
pacity of foliage (Agren, 1983a,b; Brix, 1971, 1983; accuracy with which it represents stands that vary con-
Yarie, 1997). Its application as a general model driving siderable in structure across the simulated area. Many
function is also consistent with the suggestion of of the important theoretical and applied questions in
Landsberg (2001) that foliar nitrogen efficiency is a forest ecology and forestry require the ability to simu-
useful surrogate for the complex processes underlying late mixed-species, mixed-age stands that are struc-
primary production. turally variable. This capability, together with the
The user of FORECAST has the option to simu- management simulation capabilities of FORECAST
late plant growth using light competition alone or and a user-friendly interface, are combined in an indi-
both light and available nutrients. When the nutrient vidual tree model, FORCEE. Both of these PC-based
option is added, potential growth derived from light- models are linked to user-friendly educational and re-
related net primary production (as described above) is search software: FORTOON and POSSIBLE FOR-
moderated according to whether expected nutrient EST FUTURES. As with all the models in the
demand for this new growth can be satisfied by the nu- lineage, FORECAST is tied to visualization software
trient supply. Nutrient demand is calculated on the that presents the graphical and tabular output in a vi-
basis of this new production and on empirical data on sual format (Figure 21-8). This presents 2-D and 3-D
nutrient concentrations in each biomass component. screen images as snapshots over time or as movies, ac-
Nutrient supply takes account of the availability of nu- companied by graphical and tabular output.
trients within the plant from internal cycling, avail-
ability of nutrients in the soil (from mineralization of Triplex As an example of the linkage of a series of ex-
litter and humus and other sources), nutrient uptake isting models of individual ecosystem components into a
trom the soil (a function of the proportion of soil vol- hybrid ecosystem model, we will briefly examine
ume oocupied by fine roots and of nutrients captured TRIPLEX (Peng et al. 2002). TRIPLEX is a generic
by competitor species), and in the case of nitrogen- hybrid model of forest growth and carbon and nitrogen
fixing plants, biological nitrogen fixation. Nutrient dynamics, developed and based on three well-estab-
input from precipitation and throughfall can also be lished models: a physiological process-based, tree
stmulated if desired and calibration data are available. growth model 3-PG (Landsberg and Waring, 1997), a
There is a simple representation of moisture limita- tree growth and yield model that has a simple soil sub-
tion of growth response to nutrient supply. This is model TREEDYN3.0 (Bossel, 1996) and a soil mois-
being improved as climate change capability is added ture, decomposition and nutrient dynamics model
to the model. CENTURY4.0 (Parton et al., 1993). The model was de-
FORECAST can represent either even-age or signed to be comprehensive without becoming exces-
multiage stands; monoculture or mixed-species stands; sively complex. It minimizes the number of input
and it can represent herbs, shrubs, bryophytes, and parameters required while capturing key processes and
trees of different life form. Like the gap models, it uses important interactions between the carbon and nitrogen
the “opaque blanket” representation of the canopy, cycles of forest ecosystems. TRIPLEX is designed as a
but it simulates the vertical distribution of foliage hybrid of both empirical and mechanistic components
within the canopy and can represent a variety of that can be used for (1) making forest management deci-
canopy layers. FORECAST simulates all the major sions (e.g., growth and yield prediction); (2) quantifying
pathways of nutrient cycling for up to three nutrients, forest carbon budgets; and (3) assessing the effects of cli-
something that is absent from most gap models, even mate change in both the short- and long-term.
those that incorporate litter decomposition and nutri- ‘TRIPLEX simulates average stand characteristics
tional regulation of growth. There is a complete ac- rather than those of individual trees. It is designed as a
counting of the nutrient inputs and outputs to and generic model that can be parameterized for even—or
trom the ecosystem, the circulation of nutrients with- uneven—age coniferous and broad-leaved species at
in the ecosystem, and the internal cycling within any geographical location and for different soil and
plants. The NuCM model (Liu et al., 1992) represents climate conditions. Modelling processes of TRIPLEX
nutrient cycling in even greater detail. are described in detail by Peng et al. (2002). (Details
 Examples ofHybrid Simulation Models 585

Stand visualization system

Stand visualization system

Figure 21-8 A time sequence of 3-dimensional SVS visualizations of FORECAST output il-
lustrating the developmental dynamics of a spruce-aspen boreal mixedwood stand. ‘The stand
ages shown are: (A) 10 years—a mixture of white spruce and aspen of the same age, (B) 30 years—the
faster early height growth of aspen has allowed it to dominate and suppress the spruce, (C) 70
years—aspen height growth has slowed and the spruce have caught up, giving a rather uniform
mixedwood, (D) 130 years—aspen are dying and the continued height growth ofspruce has resulted
in a spruce-dominated mixedwood stand.

can be found at _ flash.lakeheadu.ca/~chpeng/ “opaque blanket” representation of the canopy: all the
TRIPLEX_Download.html.) By combining empirical foliage is spread out evenly across the entire area being
and process modeling, TRIPLEX is in many ways simulated. This works well for spatially uniform forests
similar to FORECAST and constitutes a very useful but is a poor representation of many mixed-age, mixed-
development in ecosystem modeling. species forests with larger canopy gaps. It also fails to
simulate the patchy distribution of herbs, shrubs, and
regeneration that occurs in real canopy gaps—a prob-
Spatially Explicit Individual Tree, lem that is avoided in most gap models because they do
Community, and Ecosystem Models not simulate minor vegetation.
Stand-level models of forest ecosystems, whether of the As public antipathy toward clearcutting and enthu-
“gap” or the FORECAST type, have generally used the siasm for partial harvesting systems increased, there
586 CHAPTER 21 Models and Their Role in Ecology and Resource Management

was a concomitant need to represent forest canopies models. One of the most distinctive features of SOR-
and gaps more accurately. All the gap models grow in- TIE is that the model was designed to be calibrated
dividual trees, but most use an opaque blanket canopy. and initiated using direct input of data from rigorous Tage
e
le

The SPACE model (Busing, 1991, 1995) provides field studies.

some representation (e.g., light competition) of the SORTIE consists of four submodels: (1) seedling
zone of influence around each tree and the effect of recruitment—a function of parent tree proximity
adjacent trees on the target tree. The ZELIG model and seedbed (substrate) suitability; (2) resource
(Smith and Urban, 1988) provides a coarse-scale spa- availabil-ity—understory light dynamics as a func-
tial representation by having a grid of “gaps,” the trees tion of species-specific light extinction coefficients;
in one grid cell affecting the light reaching those in (3) subcanopy tree growth—a function of light avail-
the adjacent cell according to relative tree heights and ability; and (4) tree mortality—a function of recent
the angle of the sun in the sky. However, as the de- growth rates. In addition, the model has a flexible
mands for more detailed simulation of spatial relation- user interface to facilitate the simulation of a range
ships within the forest increased, these representations of harvest strategies (e.g., understory protection, di-
were no longer adequate. ameter limit harvesting, shelterwood, patch reten-
In response to their concerns about the limitations tion, clearcutting). In SORTIE/Boreal, work is
on the use of gap models as successional predictors, under way to add an understory vegetation submod-
Pacala et al. (1993, 1996) developed the SORTIE el and to improve the prediction of growth and mor-
model—a distance-dependent, individual tree model tality of mature canopy trees. These attributes will
based on empirical relationships between light and ra- render the model more suitable for the simulation of
dial growth and a simulation oflight competition. In- stand dynamics and succession in structurally com-
dividual trees are grown using the gap model growth plex mixed- or single-species stands and the implica-
equation approach, but rather than a small gap, the tions thereon of alternative stand management
model simulates a 9-ha plot (much larger plots can strategies. Current model predictions include (1)
also be used) with many thousands of trees. Rather spatial distribution and sizes of all individuals in a
than using mathematical expressions that describe simulated stand; (2) DBH (diameter at breast height,
general relationships for sapling recruitment and mor- 1.3m) and height distributions, by species; (3)
tality applied to all species, SORTIE uses species- changes in basal area and density, by species, over
specific relationships tied directly to field data and time; (4) tables of basal area and densities of both
represents seed production and dispersal as a critical adult and juvenile trees; and (5) distribution of sub-
factor in the development of landscape patterns of canopy light levels.
communities, rather than the random distribution of Spatially explicit representation of light, light
seeds of all species across the gap as used in most gap competition, and seed dispersal, combined with
models. Light that reaches individual trees is com- species-specific survival, mortality, growth rates, and
puted for all trees (this accounts for 98% of the com- response to light, results in very different predictions
putation time of the model). There is no understory of successional pathways by SORTIE in comparison
representation, and soil nutrients and moisture are not with most gap models. The authors of SORTIE ex-
represented in the present version. press concern that the evolving view of succession as
SORTIE was designed as a spatially explicit model depicted by many gap models may be in error, and
of stand-level successional dynamics, with the xy co- they challenge ecosystem modelers to tie their models
ordinates of each individual tree within a plot being more closely to field data that are site specific and
represented. The model was originally developed for species specific. As the need grows for more accurate
mixed forests in eastern North America but has also prediction of ecosystem response to disturbance, this
been applied in the interior cedar-hemlock forests of seems like sound advice. However, the lack of any nu-
northwestern British Columbia and is being adapted tritional representation in SORTIE limits its applica-
for use in the mixed-wood and black spruce boreal tion in ecosystems and for disturbance regimes in
forests of Canada (SORTIE/Boreal: Coates et al., which nutrient availability influences the competitive
2002). It can be considered as a spatially explicit ver- ability of different tree species, leaf area, and light
sion of the same processes simulated in many gap conditions.
 Landscape Models — 587

Reflecting the same concerns as the authors of hydrology, and other landscape-level patterns and
SORTIE, the stand-level FORECAST model has processes. Landscape models can provide an effective
been developed into an individual tree model: way of analyzing the ecological characteristics and sig-
FORCEE. This model is a spatially explicit, individ- nificance of these landscape patterns (Shugart, 1998).
ual-tree growth model that, like SORTIE, is closely Spatial modeling of landscape ecology and landscape-
tied to empirical field data and is site and species spe- level disturbance draws on the tools of forestry, geo-
cific. It differs from SORTIE by including nutrients graphic information systems (GIS), and remote
and moisture as limiting factors, nutrient cycling, and sensing, as reviewed by Mladenoff and Baker (1999).
minor vegetation; by being driven by simulated pho- The following are some examples of landscape mod-
tosynthesis; and by evaluating wildlife habitat values, els, based on Messier et al. (2003).
carbon storage, and the social values of stand-level
economics, energy benefit-cost ratios, and employ-
ment. It also differs in providing a simulation of a wide LANDIS (Landscape Disturbance and
range of stand-level forest management practices and Succession Model)
natural disturbance regimes. FORCEE produces a
LANDIS is a spatially explicit and stochastic model
light shadow of each individual tree and patch of un-
that simulates forest landscape change (disturbance
derstory vegetation and a “footprint” of each tree and
and succession) over long time scales and large, het-
understory patch in terms of litterfall, CWD, forest
erogeneous landscapes (Mladenoff and He, 2000).
floor, soils fertility, and root systems. This creates a
The objective is to investigate the impacts of natural
spatial legacy of past biotic activity that influences the
and human-induced disturbances at large spatial and
future spatial patterns of forest community function
temporal scales, rather than to predict timber vol-
and development (Figure 21-9).
umes. Model calibration requires GIS maps that de-
scribe initial conditions of the forests (e.g., species
21.9 Landscape Models presence/absence, stand age) and a map that classifies
the initial environmental conditions (e.g., soil type,
Natural and human-caused forest disturbances create topography, moisture). The simulation time step is 10
complex forest landscape mosaics, as described in years, and the cell resolution is given by the GIS data.
Chapter 19, that have important implications for LANDIS models spatially explicit disturbances such as
wildlife and various measures of biological diversity, fire, wind, and harvesting (Gustafson et al., 2000) as
well as ecological processes such as seed dispersal.
Landscape disturbance processes are stochastic (ran-
dom). Species response to these spatial processes is
modeled using a limited number ofspecies vital attrib-
utes according to the different land types, such as
species regeneration and longevity, seed dispersal (ef-
fective and maximum), shade tolerance, and fire toler-
ance. LANDIS produces time series maps of forest
composition (presence/absence data) and age as well
as time since disturbance.

SELES (Spatial Explicit Landscape

Event Simulator)
SELES (Fall and Fall, 2001) is a modeling language
that can be used to model key processes at the land-
scape scale. It is a discrete-event simulator that inter-
Figure 21-9 Graphical representation of stand structure
prets and executes models of such events. ‘The SELES
and the soil “footprint” and light shadow of individual
trees as represented in the individual tree, spatially explicit
language is sufficiently general to facilitate the con-
model FORCEE. struction and integration of models and submodels
588 CHAPTER 21 Models and Their Role in Ecology and Resource Management

that are quasicontinuous, periodic, or episodic; are de- search tool, TARDIS is used to evaluate and apply
terministic, process-oriented, or stochastic; and in methods for scaling some capabilities of high-resolution
which processes may operate locally, regionally, or landscape models up to regional extents.
globally and be either spreading or nonspreading.
SELES is useful as a research tool as well as a decision- Linkage of Harvest Schedules Model to
support tool for exploring problems related to conser- Ecological Modelsm
vation and resource management. Time steps in
SELES may vary in size, and processes with different The models just introduced were developed as land-
temporal resolutions can be integrated. Processes scape models. A different approach is to link a stand-
modeled using SELES have included forest succession level ecological model to a GIS-based landscape
and encroachment; natural disturbances such as fires, modeling framework or to a landscape-level timber
insect outbreaks, and flooding; and management activ- supply model. The latter approach has involved the
ities such as timber harvesting and livestock grazing. linkage of FORECAST to the timber-supply model
FPS-ATLAS (Nelson, 2000). A library of volume-age
curves produced by FORECAST is used to replace the
FEENIX (a Forest Ecosystem Emulator) conventional historical bioassay model timber volume
growth curves that had previously been used. By
FEENIX (Cumming et al., 1998) is a spatial and dy-
adding to this library the time series of a wide variety of
namic simulation model of the ecology and manage-
social, environmental, and ecological variables that are
ment of forest landscapes. As with LANDIS and
produced by FORECAST, this converts FPS-ATLAS
models used with SELES, FEENIX is raster based,
into a landscape ecosystem management model. How-
stochastic, and usually initialized from GIS data.
ever, this approach is not interactive. The stand-level
However, FEENIX is somewhat specialized with re-
polygons do not interact, and the approach does not
gard to its target ecosystem and the processes that it
utilize the full predictive power of FORECAST.
simulates. Although its precursors have been used to
model both terrestrial and marine ecosystems,
FEENIX is designed for boreal forests, especially the Developing Landscape-Level Models With
mixed-wood forests of western Canada. Its principal Dynamic Links to FORECAST
ecological submodels are of wildfire, stand dynamics, Linkage of FORECAST at the landscape level in an
and forest bird distributions. interactive model is achieved in POSSIBLE FOREST
FUTURES based on HORIZON, a landscape model
designed for watershed-scale scenario analyses. In
‘Tardis contrast to ATLAS, HORIZON is dynamically linked
TARDIS (Cumming and Armstrong, 2001; Cumming to FORECAST using a cell-based spatial structure.
and Vernier, in press) is a low-spatial-resolution simula- This permits the simulation of spatially dependent
tion model of regional forest dynamics. It represents events between polygons, such as seed dispersal.
forest regions as a regular grid of loosely interacting Hence, the consequences for natural regeneration of
landscape cells, connected by a dynamic network of different patterns of timber harvesting can be ex-
primary roads. TARDIS was developed for a 74,000 plored. Spatial landscape-level wildlife relationships
km? study region of 825 cells with a cell mean size of can also be explored.
9,000 ha but can be applied to much larger regions. For Variable retention (VR) is a harvesting system in
each cell, TARDIS maintains a list of polygons that which scattered individual trees or tree clumps are re-
represent mapped forest stands and other features and tained at the time of final harvest (Arnott and Beese,
static attributes such as mean elevation—and inter- 1997; Burton et al., 1999). An important difference
polated climate variables. To model ecological processes between VR and shelterwood systems is that the re-
that are sensitive to landscape configuration, TARDIS tained trees are not removed during subsequent en-
relies on statistical models (Cumming and Vernier, in tries. The main objective of VR is to satisfy public
press). This abstraction markedly reduces computation- aesthetic preference, but the system also has potential
al complexity. As a scenario analysis tool, TARDIS is de- biodiversity and wildlife habitat benefits because it en-
signed to evaluate strategic or policy alternatives in sures a sustained supply of mature forest habitat fea-
forest management under the risk of large fires. As a re- tures within a cutblock that is being managed for
 How Conmplex Should a Model Be? Occam’s Razor and the Scientific Method 589

timber on a shorter time scale. It emulates the stand The linkage of FORECAST to various other
structure that results from natural disturbance events. models, including the wildlife habitat suitability
One limitation of VR is a lack of growth and yield model SIMFOR, in a meta-modeling framework is
models and ecosystem management decision support shown in Figure 21-10.
tools that are capable of forecasting stand develop-
ment under this silvicultural system. Some landscape
models can represent complex patterns of polygons
across watersheds and larger areas, but usually the 21.10 How Complex Should a
polygons are internally homogeneous and lack any Model Be? Occam’s Razor and the
ecotonal effects. LLEMS has been developed as a Scientific Method
complex cutblock model with a spatial representation
of ecotonal effects, including light competition, seed The first computer model that one builds of some nat-
production and dispersal, seedling establishment, ural object or system is usually very much oversimpli-
and aspects of belowground soil and root processes. fied. As a result, it generally represents reality poorly
The model is to be based on the pixel data structure and will disappoint you. At this point, you either aban-
developed for the HORIZON model and possess a don the model in disgust or set about to improve. The
minimum pixel size of 10 m X 10 m. Individual pixels latter usually results in the model’s becoming larger,
or groups of adjacent homogeneous pixels are repre- more detailed, and more complex.
sented by a version of FORECAST modified to facil- As you add missing components and processes,
itate spatial interactions. LLEMS will be applied at your model may represent reality more closely but at
the scale of an intermediate- to large-size cutblock or the cost of becoming harder to understand, harder to
group of cutblocks (10 to 1,000 ha) to explore the calibrate, slower to run on your computer, and more
long-term consequences of cutblock shape, orienta- difficulty to “benchmark” and “confirm” (1.e., verify
tion, and variable retention strategies on the eco- and validate). There is a danger that as model develop-
nomic, ecological, and social (i.e., visual) indicators ment continues, you will experience the “dinosaur
of sustainable forest management. syndrome” of model development:

Overall System of Ecosystem Management Models

POSSIBLE FOREST
FUTURES: watershed
wildlife landscape
ee
habitat model management model
ATLAS: forest-level model
7

VISUALIZATION:
stand & landscape level

FORECAST: stand-level,
ecosystem management model
| LLEMS: complex
’ cutblock simulator

FORCEE:
individual tree, ma
complex stand model

Figure 21-10 Meta-modeling framework driven by the ecosystem management model

FORECAST.
590 CHAPTER21 Models and Their Role in Ecology and Resource Management

bigger — better — bigger — better — bigger > tree species, and for ecosystems in which there is suf-
less useful — bigger — useless ficient variation in nutrient levels over time (e.g., the
assart period), inclusion of nutrients and soil in a
Occam’s razor warns against this. The principle of model may significantly improve its performance.
parsimony cautions that one should choose the sim- Similarly, whether herbs and shrubs should be in-
plest of explanations or models that will achieve the cluded in a model will depend on whether they play a
objectives. significant role in tree growth and succession in the
There will come some point in the development of ecosystem in question under the disturbance regimes
any model at which improving its representation of re- of interest.
ality will render the model less useful. The “best” Definition of “best” in modeling is defined by
model for a particular purpose is usually not the most what you want your model to do. As a general rule, the
complex model. In fact, simple models are frequently best model is the one that achieves a particular model-
much more useful for certain purposes than more ing objective with the least effort, time, and expense.
complex models. Think of a map, for example. For some purposes, very simple, abstract models of re-
A map is a two-dimensional diagrammatic model ality will be ideal. In other cases, detailed representa-
of the three-dimensional arrangements of objects in a tions of complex systems may be required if the model
landscape, region, or country. A map of the whole of is to achieve its intended purpose.
Canada is a simple representation of all the features of Before leaving the topic of computer modeling, it
the Canadian landscape. It is useful for navigating is necessary to consider briefly the relationship be-
from British Columbia to Newfoundland but would tween this contemporary scientific activity and the
not be at all useful for navigating around the city of more traditional methods of science.
Toronto. However, a map of sufficient detail to be use- The “traditional” scientific method begins with
ful for inhabitants of Toronto would be far too de- observations/measurements of natural phenomena.
tailed for navigating across Canada. From observation of many examples of a phenome-
The worst possible model (e.g., a map) is one with non, one develops an initial theory about it by the
a scale of 1 to 1. Thus, as we build models, the objec- process of induction (specific > general). From this
tive is not to have the model exactly duplicate the theory, one can then deduce (general —> specific) cer-
scale and complexity of the real system that it repre- tain inferences and make various predictions. From
sents; it is to simplify the true complexity of reality these predictions, one can develop hypotheses that can
into a representation that is complex enough to be critically tested in an experiment. The results (ob-
achieve the specific modeling objective but still sim- servations) obtained from experimentation are used
ple enough to be useful. inductively to support or reject the initial theory or to
An example of the issue of complexity is the ques- derive an alternative hypothesis for subsequent test-
tion of whether a light competitive model (e.g., SOR- ing. If after many tests one fails to reject a theory,
TIE) is the “best” level of complexity for a forest then it may be elevated to a scientific principle. After
stand model or whether soil, competition for mois- many years of evaluation, such a principle may be ac-
ture and nutrients, and nutritional limitations of cepted as a scientific law (Figure 21-11).
growth (added complexity) will improve the model It is a feature of scientific experiments that they
(e.g., FORECAST). It can be argued that where there often tend to raise more questions than they answer; as
is light competition, there is also belowground com- a consequence, hypotheses tend to get increasingly
petition; light competition is an adequate surrogate narrow as a science advances. Actually, reducing a phe-
for all major competitive processes. Adding nutrient nomenon to a series of narrow hypotheses is a basic re-
cycling and soil processes may add unnecessary com- quirement of the traditional scientific method. By
plexity for ecosystems in which nutrients and mois- convention, hypotheses must be capable of being falsi-
ture are not major limiting factors. For example, fied in a definitive experiment, and this is possible only
long-term ecological succession may be predicted ad- when the hypothesis is fairly simple. Thus, there is a
equately based on light competition alone. However, momentum in science that tends to keep a scientist in a
for nutrient-limited sites, for disturbance events that reductionist spiral of hypothesis + testing > modified
alter nutrient availability to the level at which it alters or narrower hypothesis — testing, and so on. Ulti-
growth and the competitive relationships between mately, this is the only way of advancing knowledge of
 Example of the Need for Sufficient Complexity in Successional Models: The Two Edges of Occam’s Razor 591

Computer —-——
modeling activities
“Traditional” ————e
scientific activities
Derivation commonsfine
Scientific Laws
of scientific principles

INITIAL
THEORY
Induction

poi Ga) pre

Observations of Z
%
natural phenomena
MASA
4%
Z
w(LELILLELSS SLA ELS SSAA 0g “trettttttde

ee

ps ee |
COMPUTER a
MODEL a
I
Modification
4 of the model

Predictions

v
Testing of the
model's predictions

Figure 21-11 Relationship between the traditional, hypothetical-deductive, reductionist

scientific method and computer modeling.

the details of a natural phenomenon. However, the 21.11 Example of the Need for
“traditional” scientific method has a problem. It may Sufficient Complexity in Successional
tend to encourage analytical reductionism to the exclu- Models: The Iwo Edges of Occam’s Razor
sion of synthesis and integration. Unless there is a
mechanism to integrate the results of reductionist sci- There is no merit in unnecessary complexity in theo-
ence, we may fail to achieve our original objective of ries or models, but there should be sufficient complex-
explaining natural phenomena. Computer modeling ity to provide the desired descriptive, explanatory, or
can provide such a mechanism. Figure 21-11 shows the predictive powers for the particular phenomenon or
complementary role that computer modeling should system in question: the two “edges” of Occam’s razor.
play in the scientific method and how it can prevent a A good example of this is the successional relation-
research program from getting trapped in a reduction- ships observed in the temperate rain forests of north-
ist spiral by integrating the results from many experi- ern Vancouver Island, British Columbia (Prescott and
ments and relating them back to the initial theory. Weetman, 1994). In this area of the Coastal Western
592 CHAPTER 21 Models and Their Role in Ecology and Resource Management

Hemlock biogeoclimatic zone, 8,000 to 9,000 years of mycorrhizal symbionts, the redcedar germinants seem
postglacial succession involved early seral pioneer to be unable to compete for the soil nutrients that they
hardwoods such as red alder and black cottonwood need to develop shade tolerance. However, the high
and conifers such as lodgepole pine. These were ac- light and nutrient availability (the assart period) after
companied or followed by Sitka spruce and replaced in stand-replacing disturbance creates the seedbeds,
most areas over time by western hemlock and amabilis light, and nutrition needed for establishment of
fir. Approximately 3,000 years ago, western redcedar redcedar germinants and their development into
appeared in the pollen record of bogs in the area shade-tolerant seedlings, and VAM are supported by
(Hebda, 1983). Until recently, a Clementsian model invading herbs and shrubs.
had been proposed for this area, in which stand- Young (<100-120 years old) HA stands are gener-
replacing wind storms periodically reestablish dense ally free of the vascular plant parasite dwarf mistletoe
and highly productive western hemlock and amabilis (Archethobium tsugae). However, as the age of the
fir stands (HA). According to this model, where such stands increases, so does the frequency of mistletoe in-
windthrow events did not occur, the postdisturbance fection. The parasite causes branch deformation and
stands are invaded by supposedly shade-tolerant west- stem swellings that greatly increase the probability
ern redcedar and the ericaceous shrub salal and de- that wind storms will cause stem break rather than
velop into a low-productivity, shrub-dominated open windthrow. Over time, this changes HA stand re-
climax stand (CH) (Figure 21-12). Such forests sponse to windstorms from stand-replacing wind-
presently occupy 30,000 ha (9%) of northern Vancou- throw to gap-forming stem break. These gaps are
ver Island. This implies an annual rate of transition of filled with slowly growing hemlock saplings that in
1,000 ha per century. turn become heavily parasitized. Old HA stands with
A difficulty with this model is that it supposes that heavy infections seem to form self-replacing climax
intermediate seral stages, involving HA stands with in- stands with little evidence of transition to CH stand
creasing densities and heights of redcedar, will be en- types. The germinants of redcedar are still unable to
countered and that climax HA stands should not invade because the gaps are so rapidly filled by the
occur. In reality, intermediate stands are very uncom- abundant hemlock seedling bank and seedlings and
mon and old-growth HA can be found. Investigation saplings of redcedar are rare. Where they occur, they
of the successional relationships in these forests has frequently seem to have originated as veglings formed
suggested the need for a much more complex model. when redcedar branches blown into the stand from
Weber et al. (2003) showed that western redcedar, adjacent areas take root.
long considered as a very shade-tolerant species, has Contrary to the Clementsian model, stand-replac-
germinants that are relatively shade intolerant. Their ing wind disturbance seems to be the major opportu-
spike-like primary foliage is not a good interceptor of nity for redcedar establishment from seed. However,
light. Not until they have developed palmately early growth of redcedar is much slower than that of
arranged, shade-tolerant, secondary-scale leaves are hemlock and amabilis fir. Most of the redcedar that es-
the seedlings able to survive and grow in deep shade. tablishes after stand-replacing disturbance of HA
In fact, the seed size, seedbed requirements, and ger- stands are outgrown by the hemlock. Those that sur-
minant light relationships are much more like those of vive produce saplings with a very high height—diame-
pioneer species adapted to disturbance than those of a ter ratio, and most such saplings are eventually
shade-tolerant late seral species. toppled over by snow, wind, or falling dead hemlocks.
The deep shade in young HA stands prevents de- Some reproduce vegetatively when branches of top-
velopment of shade-intolerant redcedar germinants pled trees touch the ground and root to produce a
into seedlings and thus the colonization of HA stands population of veglings, but most of these suffer the
by redcedar. This barrier to invasion is compounded same fate. However, if a subsequent wind storm re-
by the fact that to develop shade-tolerant secondary moves the overstory before these veglings die, then
foliage, the germinants seem to require good nutri- they may release and some may successfully recruit
tion. To get this, they require vesicular-arbuscular my- into the canopy of the next HA stand.
corrhizae (VAM), but the trees in HA stands support At the next major wind event, these canopy red-
ectotrophic mycorrhizae, and there is a lack of under- cedar have a much higher probability of survival than
story species that support VAM. Without appropriate the hemlock and fir because they shed foliage and
 A B
Mature HA

I
First
disturbance

7150-250 yrs-s, > >250 yrs>- ee

A we on eee eee ee et Cine petee eh

Mature HA

II Young HA
After 1-10
“a
additional >

disturbances

22> >250 YIS~=5

Il
After 1-5
additional
disturbances

100

Young CH
IV
After 1-3
additional
disturbances

oe

After 1-2
additional
disturbances

> >250 yrs--,

f High probability of stand- No stand-replacing

replacing wind disturbance wind disturbance Grey trees
: oe Sos
ER aos Serres oa : | are infected
7 ia aN Ree Western Pacific Western | with dwarf
Y Low probability of stand- Stand-maintaining (gap-
replacing wind disturbance phase) wind disturbance ¥& Salal | Hemlock ! silver firfl redcedar ' mistletoe

Figure 21-12 Clementsian model of succession in coastal western hemlock zone forests on
northern Vancouver Island (cf. Prescott and Weetman, 1994). According to this model, the sequence
of seral stages shown will occur in the longterm absence of stand-replacing disturbance.
594 CHAPTER 21 Models and Their Role in Ecology and Resource Management

branchlets during high winds and the remaining pen- resource competition is also a key feature of the suc-
dulous foliage “streams” in the wind, reducing drag cessional relationships in these stands.
and the probability of windthrow (based on wind tun-
nel studies by Steve Mitchell, University of British
SUMMARY
Columbia, Department of Forest Sciences, personal
communication 2003). Some of these redcedar survive The goal of science is to describe and understand our en-
to become large trees in the stand, establishing the vironment, and scientific knowledge can be used to pre-
conditions for recruitment of redcedar seedlings and dict the future conditions in the world’s ecosystems and
the shrub salal. If sufficient trees of this size recruit, the impacts of our management on these ecosystems. As
knowledge expands and our understanding of the compo-
then conditions are set for the transition of HA to CH
nents of the ecosystem improves, it becomes necessary for
(Figure 21-13). individual scientists to focus their attention on increas-
Many of the details of this successional model ingly narrow subcomponents thereof. This tendency is
await critical testing, but the research to date supports encouraged by the scientific method and is the modus
this interpretation. It seems, therefore, that an accu- operandi of the traditional scientific disciplines. Unfortu-
rate explanatory or predictive model of this system nately, although this trend does increase our knowledge
should include representations of seedbed require- of the subcomponents and processes of a system, it does
ments, germinant and mature seedling shade toler- not always contribute to our understanding of the system
ance, nutritional requirements for the development of as a whole. There is a need for a branch of science that fo-
shade tolerance, the role of mycorrhizae in the acquisi- cuses on the system as a whole and facilitates the integra-
tion of nutrients, wind relationships and the frequency tion of the parts. Ecology provides this focus, but there is
a need for a mechanism by which to examine complex
of major wind events, and the role of dwarf mistletoe
ecological systems and to integrate our rapidly increasing
in mediating the response of HA stands to wind. Such
knowledge thereof. Modeling, especially in conjunction
a model needs to be capable of representing the major with computers, provides such a mechanism.
components of Figure 21-13 and the major processes Modeling is a very ordinary activity indulged in, to
of production ecology, biogeochemistry, population some extent, by all humans. There are many different
and community ecology, and allogenic disturbance au- types of model, from our private thoughts, through pic-
togenic succession. The role of salal in light and soil tures, words, and three-dimensional physical representa-

CH/HA

Young HA
NIV! pte
NW alk

oo
3G
~---

NW.
—
Western Grey trees
Stand-replacing Stand-maintaining No stand-replacing hemlock are infected
wind disturbance gap-phase) wind wind disturbance & Pacific Western with dwarf
disturbance 3h Salal redcedar mistletoe
silver fir

Figure 21-13 Conceptual model of northern Vancouver Island succession based on the re-
search of Adrian Weber. (Ieber et a/., 2003; and unpublished.) In the absence of stand-replacing
wind disturbance, redcedar is excluded and a climax, gap-phase HA stand type develops under the
interacting influences of the parasite dwarf mistletoe and wind. According to this model, the major
opportunity for replacement of HA by CH is the stand-replacing disturbance. This replacement oc-
curs with a low probability. This model is believed to conform much more closely to the landscape
of northern Vancouver Island than the Clementsian model in Figure 21-12. See text for details.
 Study Questions = 595

tions, to mathematical expressions. It is the mathematical TAKE-HOME MESSAGE

expression implemented on a computer that most people
think of as modeling: the computer model. Because of the long time scale of forest crop production
Computer models come in two major varieties: (1) and the pressure from continued population growth, we
those that take advantage of the power and convenience do not have the time simply to wait and see whether a
of certain mathematical techniques but thereby compro- new forest management strategy will sustain desired
mise to some extent the accuracy of the biological de- ecosystem values. If we do not already have examples of
scription in the model and (2) those that sacrifice the consequences of such management over at least one
mathematical elegance and computational efficiency to and hopefully several tree crop rotations, then we must
obtain the best description of the biological attributes of use computer models of forest ecosystem response to
the system being modeled. The user must judge which is disturbance to predict these consequences. Well-
the more useful approach for any particular application, constructed models can harness our current, albeit incom-
and a combination may prove to be the best. The recent plete, understanding of forest ecosystem structure and
trend in model evolution has been toward accurate bio- function and project it out over the relevant time scales.
logical representation. We can also use computer models to evaluate complex
Irrespective of the approach taken, models can be used questions about sustainability of multiple values at the
for various purposes in ecology and resource management: landscape scale. There is really no other way of making
short-term assessments of long-term sustainability issues.
1. They can be used to provide a description of an Modeling is equally important in testing the theories
ecosystem, integrating and synthesizing large data sets of forest ecology. Many of these theories relate to time
in a way that cannot be achieved by any other method. and space scales that cannot be addressed by conven-
tional scientific research methods. Theories about long-
2. They can be developed as a learning activity, for the term patterns of ecological succession, in particular, are
purpose of exploring our knowledge of a system, best evaluated using computer models of succession. The
identifying information gaps, and suggesting future widespread popularity of successional models in forest
lines of research. ecology today is evidence of this.
3. They can be used to make predictions about future Computer modeling is a vitally important tool in
states of a system, with and without management both applied and basic forest ecology. Although no model
treatments. of a biological or ecological system can ever mimic it pre-
cisely, well-designed and calibrated models can emulate
In the past decade, many ecosystem models of the natural systems accurately enough to be useful in forest
detailed biological-process type have been produced, ecology and forest management.
largely serving the first two purposes. Building on earlier In conclusion, the reader is reminded to be skeptical of
work in the 1960s, many simple forest growth (i.e., tree the predictions ofmodels. External models are only as good
growth) models were developed in the 1970s to serve the as the conceptual models on which they are based and the
third objective. Few of these provided an explicit descrip- data used in calibrating the model. These in turn depend
tion of the ecosystem, but in the early 1980s, several on the level of our understanding of the system and the
models that predict tree growth within the context of availability of appropriate data. Both of these items are
ecosystem processes and other ecosystem components generally in inadequate supply, and all models must fun-
appeared. These promise to provide a very useful frame- damentally be considered to be to some degree “wrong.”
work for integrating ecological information with man- Only reality is completely “right.” This does not render
agement and economic considerations to investigate the models and modeling valueless. It merely requires that
ecological and economic implications of alternative man- models be used with discretion and an explicit recogni-
agement strategies. As the power of computers grows tion of their strengths and weaknesses.
over the next few years, computer modeling will and cer-
tainly should become a commonplace tool of ecologists
and resource managers. The greatest contributions of
STUDY QUESTIONS
computer modeling in forest ecology may come from 1. What is a model?
those models that are specifically developed for the pur- 2. What is “dinosaur syndrome” in modeling? What
poses of “gaming” ( that is, running computer models does Occam’s razor tell us, and how does this relate to
with different management and environmental condition the complexity attribute of ecosystems?
scenarios, and comparing how well they achieve desired
3. Can a model of a biological system ever be com-
outcomes) to investigate the response of ecosystems to
pletely “right”?
disturbance in general and the effects of our management
in particular. 4. Is there a “best” model? How do we judge models?
596 CHAPTER21 Models and Their Role in Ecology and Resource Management

. Why do we use models? 9, What are “gap” models, and for what have they been
nN. How does computer modeling relate to the tradi- used?
tional “scientific method”? 10. How would you judge whether a light competition
. What kind of models can be useful for the evaluation
—I model such as SORTIE is sufficient and/or the best
of nutrient losses in harvested materials? model to use and whether one might need a more
- How credible are landscape-level models if they do
oO
complete ecosystem model such as FORECAST?
not represent stand-level ecosystem processes? How Can light competition be used as a surrogate for be-
useful are stand-level models on their own if many of lowground competition?
E

the key issues in forestry relate to the landscape scale?

a
 C H E R

Days
Sustainability and
Renewability of Natural
Resources, and Implications
for Forest Management
22.1 Introduction newable Resource Management. These are organiza-
tions concerned with activities such as forestry, fish-
As the human population continues to grow, material eries, agriculture, and wildlife, which deal with
and energy resources that are renewable will become of biological resources that are commonly assumed to be,
paramount importance to society because their utiliza- by their very nature, renewable. In contrast, the term
tion is sustainable. The cost of nonrenewable resources nonrenewable resource is, by tradition, used to refer to
should rise at an increasing rate because of their in- deposits of fossil fuels and mineral materials such as
creasing scarcity, until many of them become uneco- coal and copper: nonliving materials that are consid-
nomic and they are displaced by alternative resources. ered to be irreplaceable. Implicit in the classification
The cost of renewable resources, conversely, should of resources as renewable or nonrenewable is whether
rise more slowly because of their continuing availabil- the resource is living or nonliving; a direct connection
ity. As a consequence, their relative value will continue is made between /iving and renewable.
to increase, as will society’s concern for their sustain- In contrast to the foregoing usages of these terms,
ability. This will require a better definition of just what I submit that the difference between the two types of
we mean by renewable, nonrenewable, and sustainable and resource is mot a direct function of whether the re-
an improved understanding of the determinants of re- source is living or nonliving: that in fact renewable and
source renewability. This chapter presents definitions nonrenewable are socioeconomic terms that do not ac-
of these terms and, by way of example, applies them to curately reflect the biological or nonbiological nature
the management of forest resources. of a resource. A renewable resource is basically one
that can be restored to the point ofreuse after a period
of time that is within our current economic or social
22.2 Renewability of Resources planning time scale or that is renewed at a rate that
renders investment in its renewal economically attrac-
What is the fundamental difference between renew- tive. Resources that do not meet these criteria would
able and nonrenewable resources? These two terms be classified as nonrenewable.
are used commonly, but rarely is their exact meaning
defined.
Traditionally Renewable Resources
Implicit in the contemporary use of the term
renewable resource is the assumption that the resource is Consider first a classically renewable resource: the
inherently renewable: that it will renew itself, come forest. Forests are renewed by the biological process
what may. Governments have Ministries of Renewable of photosynthesis at a rate that generally places the
Resources, and universities have Departments of Re- renewal period well within our contemporary time

Fi
Nn
598 CHAPTER 22 Sustainability and Renewability ofNatural Resources, and Implications for Forest Management

scale. Eucalyptus plantations in some tropical coun-

tries can reach a height of 25 m in 5 years. They can
Traditionally Nonrenewable Resources
be managed on rotations as short as 4 years and are Consider now a classically nonrenewable resource:
therefore not much different from an agricultural mineral deposits. These are generally formed by tec-
crop. Similarly, red alder on the west coast of North tonic or erosion/solution/deposition processes over
America can reach a height of 10 m in approximately periods on a geological time scale. However, not all
5 years on some sites. Where there is a market for deposits of mineral materials are nonrenewable.
red alder, it constitutes an eminently renewable re- Guano phosphate deposits on islands off the west
source, attaining exceptionally high rates of biomass coast of South America at dry, subtropical latitudes
accumulation over its first 10 to 15 years and out- have been estimated to have a renewal rate of 190,000
growing its coniferous competitors for approximate- metric tons per year, containing approximately 8,800
ly 35 years. Tree farming is indeed an appropriate metric tons of phosphorus (Hutchinson, 1950). These
term for such crops. deposits have been mined extensively as a source of in-
Douglas-fir, another western North America tree, dustrial phosphorous, and in this context, they can be
can grow as much as 60 m in height in 100 years on considered analogous to commercial mineral phos-
very good sites, and this species is capable of produc- phate deposits elsewhere in the world, which have a
ing valuable saw logs in 30 to 40 years on the best zero renewal rate. Although the guano deposits have
sites. On such land, Douglas-fir also constitutes a very been heavily exploited, they certainly represent a re-
renewable resource, one in which society is very will- newable phosphorous deposit, which, if managed
ing to invest capital to assist its renewal. On a dry and properly, could be mined in perpetuity on a sustained-
infertile site, conversely, Douglas-fir might take more yield basis.
than 200 years to reach commercially valuable size. Similarly, not all fossil fuels are nonrenewable,
Such a timber resource begins to reach the limits of even though they have traditionally been considered
renewability in the socioeconomic sense, and there is as such. When we think of fossil fuels, we generally
little enthusiasm to invest much in its renewal. Simi- think of oil and coal, which are nonrenewable, but in
larly, neither 1,000-year-old redwoods in California, countries such as Ireland and the former Soviet
valued primarily as a recreational resource, nor 300- Union, peat has historically been very important. As
year-old oaks in France, planted for the production of oil becomes increasingly scarce and/or expensive, peat
very large wine casks, can be considered as renewable may increase in importance as a fossil fuel in several
resources in the commonly accepted sense of this countries, including Canada, although conservation
term. Although these trees are capable, under appro- and environmental aspects of peat mining may restrict
priate conditions, of reproducing the above-men- the extent of its use. In contrast to coal and oil, peat is
tioned values, the time required to do so greatly being formed continuously and at a rate that may in
exceeds society’s current economic and/or social time some cases make it a renewable source of energy. If the
scales. mining were appropriately regulated to balance re-
A value that is not capable of being re-created in movals and the formation of new peat, then it could
less than 1,000 years (e.g., large redwoods) or even become a renewable, locally important, sustained-
300 years (e.g., large-diameter oak trees) is consid- yield organic fuel energy supply.
ered by most people to be nonrenewable. Therein Thus, both living trees and nonliving material de-
lies the general antipathy of the public toward cut- posits may or may not be considered as a renewable
ting down very large and very old trees that they resource, depending on either how long it takes them
perceive to be nonrenewable resources. Once cut, to generate a particular socioeconomic value or how
the social values that they provided are gone, rapidly the future value of the resource is accumu-
“never” to be renewed. Certainly, if we are prepared lated. Forests are certainly not an inherently renew-
to wait for many centuries, these values may well be able resource, and similarly not all mineral or
re-created. Certainly, in a biological sense, the re- fossil-fuel deposits can be considered intrinsically
source value is probably renewable. However, it is nonrenewable. The difference between renewable and
renewable on a geological time scale, not a socioeco- nonrenewable resources lies more in the rate or time
nomic time scale, and therefore we consider it to be of their renewal than in the physical, chemical, or bio-
nonrenewable. logical character of the resource per se.
 The Nature ofa Material Resource 599

22.3 ‘The Nature of sil carbon deposits (fossil fuels), and forests (timber)
a Material Resource (Figure 22-1).
Most rocks contain at least a few atoms of most of
Material Resources as the precious and semiprecious metals, but whether a
particular geological deposit is a copper mine or a gold
Concentration Phenomena
mine depends on whether the concentration is high
Material resources can be defined as concentrations of enough to have economic value. Only rocks that con-
energy or materials that are sufficiently greater than tain several percent of copper are currently of value as
the average concentrations of these materials in the copper ore; rocks that contain 0.005% copper are just
earth’s crust or at the earth’s surface to have value for rocks. The value of low-grade copper ore is much less
society. The greater the difference in concentration than that of high-grade deposits of the same magni-
between the resource and the average concentrations tude because of the greater energy (and therefore ex-
in the environment, the greater the value of that re- pense) required for the recovery of the metal. The
source to society. This definition is explained using exact concentration at which the rock becomes a cop-
three examples: precious and semiprecious metals, fos- per resource depends on the location and the extent of

No Marginal High-value
copper resource value copper resource copper resource
~~’ ae Sia

copper ore

Carbonaceous shale
Anthracite coal Lignite No carbon-energy resource
very valuable resource valuable resource vale

Density of cross-hatching indicates the

quantity of thermal energy per m? of
the deposit.

Marginal Valuable timber

No timber resource values timber resource resource
———— SS

Pullip Asari A LEQ

YW
LT tit PL i iH: eto hn!

Figure 22-1 Material resources considered as concentration phenomena. (A) Copper deposits
in rock. (B) Carbon deposits. (C) Forests as a timber resource. The concentration of a material that
is considered to be a resource depends on socioeconomic conditions and therefore will change from
time to time.
600 CHAPTER22 —Susta Renewability oflity
andinabi Forest Management
forations
Natural Resources, andImplic
the deposit, on the current costs of recovering it, and formation. Although net photosynthetic efficiency is
on the current market value of copper. Thus, as with only a few percent of incident solar energy, this still _
renewability, the very definition of resource is socioeco- amounts toa large flux of energy in areas where the
nomic rather than scientific. The dramatic rise in the photosynthetic apparatus of primary producers is ade-
value of gold im the winter of 1979 to 1980 resulted in quately supplied by moisture and nutrients and where
old gold mines being reopened and old gold mine tail- climatic conditions are suitable. In those circum-
ings piles being reworked. The nonresource ofyester- stances in which the environmental conditions inhibit
day became a valuable gold resource almost overnight. the photosynthetic process, a biological resource may
Among the coals, anthracite coal is more valuable have a very small energy flux and, consequently, a very
than lignite because of the higher energy content per long time for renewal.
ton and per cubic meter of anthracite. However, lig- Similarly, most mineral deposits have a long time
nite is more valuable than carbonaceous shale, which for renewal because the energy flux of the geochemical
is not considered to be a fossil-fuel resource despite processes involved in their formation tends to be low.
the abundant carbonaceous fossils found between the Where both biological and large-scale geochemi-
layers of rock. The concentration of carbon in car- cal mechanisms are involved in resource formation, as
bonaceous shale is simply too low for the rock to be of in the case of guano phosphate deposits, the renewal
value as a fuel, even though the amount of free carbon time and the resource renewability will be intermedi-
in such rocks is much greater than the average for ate. The relatively rapid accumulation of phosphate
rocks in general. Thus, the value of carbonaceous de- on the guano islands results from the combination of
posits as a source of energy is closely related to the phosphorus-rich water upwelling from the deep Pa-
concentration of oxidizable carbon per ton or per cific abyss (a low-energy-flux geochemical mecha-
cubic meter. nism), a high-energy-flux biological food web
Most areas that contain trees are capable of yield- concentration process (beginning with uptake of phos-
ing some lumber. Even scattered small trees can be cut phorus by photosynthetic plankton and ending with
down, and the small pieces of sawn wood that result predaceous seabirds that nest and defecate on the
can be used directly or glued together to yield larger guano islands), and a relative lack of precipitation,
pieces. However, savanna forests or open woodland which allows the phosphorus-rich feces and remains of
are not normally considered to be timber resources the seabirds to accumulate on the dry, rocky islands.
because the volume of wood per hectare (i.e., the con- An analogous, although perhaps tongue-in-cheek,
centration of wood) is too low to make harvesting eco- example comes from the suggestion that if the chronic
nomical, Only forests with volumes per hectare pollution of Lake Erie with phosphorus in the 1960s
greater than some threshold value are considered to be and early 1970s had continued, then the lake would
timber resources. have become a renewable phosphorus resource by the
turn of the century. The high-energy-flux processes of
Rate of Formation of Material Resources as mining, refining, manufacturing, and transportation
a Function of Energy Flux Through the (using fossil-fuel energy) brought large quantities of
phosphorus from far-removed locations to the water-
Resource-Forming System
shed of Lake Erie. There it has been used in the form
A corollary of the second law of thermodynamics is of detergents, fertilizers, and other chemicals in
that the creation and maintenance of order in a system processes that placed it in urban sewage or agricultural
requires the expenditure of energy (Chapter 4). The drainage waters, all of which flow to the lake (low-
creation of a material resource (a concentration of en- energy-flux process). Once in the lake, food chain con-
ergy or materials) requires energy, and the rate of con- centration processes led to its ultimate deposition in the
centration and accumulation of both energy and lake-bottom sediments (medium- to high-energy-flux
chemicals is a function of the rate of flow (or the flux) process). In the early 1970s, when chronic phosphate
of energy through the system of processes that are re- pollution of lakes was common, it was suggested that if
sponsible tor the concentration and accumulation. this situation were to continue, then there would be so
Biological resources tend to have a short time for much phosphorus in the mud on the bottom of the lake
renewal because of the high energy flux frequently as- by the end of the century that it could be mined eco-
sociated with the photasy nthetic mechanism of their nomically. If this pollution were to have continued un-
 Timber Mining Versus Sustained Yield Management Versus Sustainable Ecosystem Management 601

abated, then such mining could have been continued in- resources into renewable ones or prevent the conver-
definitely and Lake Erie could have become a giant, re- sion of renewable into nonrenewable resources. The
newable phosphorus resource. Hopefully, this scenario quality and the intensity of forest management are
will never come to pass. The development of phosphate- therefore the key.
free detergents and the slow but steady reduction in the Managers of renewable forest resources must use
extent of water pollution over the past three decades ecologically sound management techniques if they are
should mean that this great lake will never be used as a to preserve or improve the renewability of the various
phosphorus recycling facility. resource values that forests provide. Unfortunately,
From this discussion we can conclude that the this has not always occurred. Deforested areas of
lower the energy flux, the slower the rate of resource India, Africa, and China and recent conversion of
formation and the less renewable the resource. The tropical rain forest in South America and Asia are clas-
fundamental difference between renewable and non- sic examples of turning renewable forests into nonre-
renewable material resources lies in the magnitude of newable forests. Some contemporary logging in
the energy flux involved in their formation, not in developed countries also falls into this category, al-
whether the resource is biotic or abiotic per se. though the recent rise of public concern and pressure
on forestry to be more sustainable has greatly de-
creased the prevalence of this problem. If forest man-
22.4 How Does Our Use of Material agement (referring here to the actual way the forest
Resources Affect Their Renewability? resource is presently being “managed,” not to the the-
oretical definition of the term “forest management”)
If this concept of renewability is valid, then it is clear ignores the effects of management on the ecological
that resources that have traditionally been called re- mechanisms of forest productivity and on the func-
newable are not necessarily so and that resources that tioning of the resource, then it may be a little different
are generally thought of as being nonrenewable may from the utilization of nonrenewable mineral de-
be capable of renewal within our socioeconomic time posits—an act of exploitation conducted without any
scale. If the efficiency of the energy flux mechanism of thought for the future renewal of the deposit.
a renewable resource is impaired, then the time for its
renewal will be lengthened and its renewability will be
diminished, possibly to a point at which it no longer 22.5 ‘Timber Mining Versus Sustained
constitutes a socioeconomically renewable resource. If Yield Management Versus Sustainable
the energy flux mechanism of a nonrenewable re- Ecosystem Management
source could be changed from a geochemical to a bio-
geochemical basis or if the efficiency of the energy flux The profession of forestry has evolved in one form or
mechanism of a marginally renewable biotic resource another at various times and places in our cultural his-
could be improved, then their renewal time would be tory. In every case, the motivating force behind its de-
reduced and they might become renewable. velopment has been to sustain the supply of a variety
The distinction between renewable and nonre- of desired goods and services from forests. he Ford-
newable forests is not absolute. In many areas, it is Robertson (1971) definition of forestry reflects the
largely dependent on the type of forest management same notion of sustained provision of forest products
used. For example, misapplication of the otherwise le- or nonmaterial values: “generally, a profession em-
gitimate practices of clearcutting and slashburning to bracing the science, business, and art of creating, con-
an area of climax, xeric lithosere forest can render a re- serving, and managing forests and forest lands for the
newable forest condition nonrenewable by retrogress- continuing use of their resources, material or other.”
ing the area back to the early seral stage of mosses and It is not surprising, therefore, to find the classical sus-
lichens on exposed rock. Conversely, careful harvest- tained-yield timber management concept deeply en-
ing of such sites can leave them in a reasonably pro- trenched in the philosophy of forest land management
ductive condition. Fertilization of nutrient-poor soils, in North America and elsewhere in the world.
drainage of excessively wet sites, and careful matching The concept of sustained yield has been the key-
of species and genotypes to site can greatly improve stone of traditional forest management. Implicit in the
forest productivity and convert nonrenewable forest term is the assumption that the forest is a renewable
602 CHAPTER 22 Sustainability and Renewability ofNatural Resources, and Implications for Forest Management

resource. However, this assumption should be made There can be no a priori conclusion that timber
explicit because many forests constitute a patchwork mining is intrinsically a social evil. In some cases, the
of renewable and nonrenewable forest conditions; the benefits from timber mining may outweigh the costs,
validity of the assumption must be evaluated for each giving a net benefit. For example, high-elevation
particular situation, Under conditions of renewability, clearcut logging, which certainly has the potential to
the concept of sustainable management may be legiti- be a good example of timber mining in some subalpine
mate because the socioeconomic values are eminently forest environments, can result in the expansion of
renewable, However, in the nonrenewable condition, subalpine meadows, with significant long-term bene-
timber management is tantamount to “timber min- fits for mountain recreation. It is certainly necessary to
ing”: that is, the use of socioeconomic values accumu- give serious consideration to the implications of such
lated in the standing crop over a long period of time, forest denudation for hydrology, streams, fisheries,
with no prospect of renewal to the point of reuse over avalanche hazard, wildlife, and the future supply of
the contemporary socioeconomic time scale, ‘To pre- timber, but in some situations, a limited extension of
tend that we are practicing sustainable management in alpine and subalpine plant communities might be both
such forests is to delude ourselves, acceptable and desirable. In many other cases, such
Because sustained-yield management generally fo- high-elevation forest removal would undoubtedly be
cused on timber, other forest values were often com- totally unacceptable. Each case would have to be ex-
promised, Forestry has now refocused on ecosystem amined on its own merits after an appropriate analysis.
management (Jensen and Bourgeron, 1994; Kauf-
mann, 1994; Overboy, 1992), This approach takes the
multiple-use concept of the 1960s and puts it on an 22.6 Sustainability and the Concept
ecological foundation of ecological site classification; of Ecological Rotation
of respect for the dynamic, changing character of the
forest ecosystem and forest landscape and for the his- As noted above, a fundamental tenet of sustainability
torical role of disturbance in forests; and of the re- is that the resource be renewable: that we can harvest a
quirement for a balance among ecological, economic, forest with a justifiable expectation that within an ac-
and social considerations, Although the issue of re- ceptable time frame the timber, wildlife, watershed
newability (i.e,, sustainability) was fundamental to sus- protection, recreational, and other values generated
tained-yield management, it is even more important in by forest cover will be renewed. Whether a resource is
the ecosystem management approach to forestry. being sustainably managed depends on our definition
Ecologically based management of forestland for of sustainability and on the balance between the sever-
timber production, in harmony with wildlife, fish, ity, spatial scale, and frequency of harvest disturbance
water, recreation, and other resource and nonresource and the innate or management assisted rate of ecosys-
values, would seem to be an appropriate land use for tem recovery.
the majority of the forests that truly constitute renew- As noted already, sustainability at the stand level is
able resources, Certainly, there will sometimes be con- not an equilibrium, no-change concept. It is a
flicts between these alternative resources, but the nondeclining pattern of change over time in stand-level
major question is how to make timber production structure, composition, and function. Similarly, sustain-
compatible with the other land uses, not whether tim- ability at the landscape scale involves a shifting mosaic
ber production should be one of the objectives of land ofstand-level ecosystem characteristics, the overall character
management. In contrast, the decision to undertake ofwhich is constant ifa sufficiently large landscape area is
timber production in forests that are not renewable in included in the evaluation. Sustainability at both of these
a socioeconomic sense (i.e, timber mining) requires a scales is a function of the rate of autogenic processes
much more critical analysis, The determination of and the frequency and severity of allogenic or biogenic
whether timber mining is permissible must depend on successional retrogression.
a comparison of the long-term socioeconomic and en- Forest crop rotations (the length of time between
vironmental costs and benetits, and the conclusion successive harvests) can be calculated in a number of
may frequently be reached not to harvest some or all ways. A technical rotation is that period required to pro-
of the timber, duce a specified type of product: a certain size of log or
 Sustainability and the Concept ofEcological Rotation 603

Ecological rotation B

Ecological rotation A
=_-=e_—_—_————

Initial
condition
: A
ea
a
ae | aa
Hie |
1 aeaaN | & |
= Wee. 5 | bs
aS) i Disturbance BL .— ee
=
uo)
re
|:
JL .—.—- a
p—-
a
pee
4 Sai ae

2
3
1S) |
|
2 a
|
3
Er a

| Disturbance C ea
Z | SEG
> | oa
LD Sea
g | a
ca) | eee

Time

Figure 22-2 Graphical representation of the concept of ecological rotation—the period re-
quired for an ecosystem to recover to its original or some desired new condition. (Based on
Kimmiuns, 1974.)

a log with the outer layers free of branch knots. An turbance with a moderate degree of logging distur-
economic rotation may be the period over which mean bance, the latter being repeated with a frequency
annual return on investment is maximized. A higher than or equal to that required for an ER.
maximum volume yield rotation is the period over which The earlier stages of lithic xeroseres are frequently
mean annual stemwood increment is maximized. An prolonged and unproductive of trees, so it is easy to
ecological rotation (ER), however, would be the period designate the high degree of disturbance as undesir-
required for a given site managed with a given tech- able. Conversely, a moderate degree of disturbance
nology to return to the predisturbance ecological con- may be both deliberate and desirable in the first ro-
dition or to some other desired seral stage (Figures tation. Unfortunately, if it is repeated at intervals
18—5 and 22-2). We discuss ERs in terms of ecological shorter than the time required for complete succes-
succession and site nutrient capital, but they could sional recovery, such a disturbance can result in a
equally be based on temporal patterns of wildlife habi- gradual retrogression of the area to successively earlier
tat features, hydrological function, or other values. seral conditions, and ultimately this can result in the
same nonproductive condition as is produced by the
single high degree of disturbance. Thus, a technique
Succession and ER that produces moderate disturbance that seems to be
If the severity of disturbance caused by forest harvest- acceptable over the first one or two rotations may
ing is great enough (cf. Figure 18-3) then the site will eventually produce undesirable results if not coupled
be reverted to an earlier stage of the successional se- to a rotation of appropriate length (1.e., a successional-
quence. This is often both economically desirable and recovery ER).
ecologically appropriate in terms of favoring a particu- In terms of the renewability of resources, the sin-
lar species, but if the disturbance is excessive, then it gle severe disturbance depicted in Figure 22—3 imme-
can be damaging in terms of future forest productivity diately impairs the energy flux through the system,
(Figure 17-17). Figure 22-3 compares the succession- rendering the resources less renewable. The moderate
al consequences of a single high degree of logging dis- disturbance with a short rotation initially sustains the
604 CHAPTER 22 Sustainability and Renewability ofNatural Resources, and Implications for Forest Management

Western Western redcedar

redcedar Western hemlock
Bare Hardy
rock Lichens Pioneer mosses, shrubs Lodgepole pine, Douglas-fir |Douglas-fir Douglas-fir

Successional
retrogression Single severe disturbance
Successive moderate disturbance
Initial
condition

« : 7
Successional a pe) OAKS
recovery Time period \_ a Bese Ecological
less than enw uy Sees rotations
ecological ace aes art
TOGO es OE bog aaa
9 —<———
(a)

Initial Ecological rotation for moderate disturbance

* harvest pe a Se
—= © a2 ‘ oi
oo os = eat ripe a ¢

5 oll2e eo — so)
No Sis
sjj css Ores act
noe *
Buss 2 oo
a a Se Eo PAE he
S 3 re Se ra
= 2 ss Sat
E : - o
= |
Oo 2 E kz 3 o
38
— ) IN oO 59

oS ued oo
z
St
na =| ~
¥= ree
as
th Ecosystem altered to
early seral condition

a
LLL
Ecological rotation for severe disturbance

Pioneer
Time
(b)

Figure 22-3 The concept of ecological rotation expressed in terms of successional retro-
gression and recovery. The figure depicts the successional consequences of either a single excessive
disturbance or a series of successive, moderate harvesting disturbances in conjunction with rotations
equal to or less than the ecological rotation. (A) Successional retrogression and post disturbance re-
covery in a hypothetical lithic xerarch succession at low elevation in southern coastal British Colum-
bia. (B) The same events expressed graphically.

energy flux, changing only the species involved, but as certain nutrients in certain proportions to grow. The
successively earlier seral stages are created, energy forest has a certain capital of these nutrients, which
flow is decreased, and this treatment eventually pro- exists as a dynamic equilibrium between a variety of
duces the same effect as the severe disturbance. inputs and outputs. Harvesting inevitably results in
some depletion of the site nutrient capital through
losses in harvested materials, as a result of disruption
Site Nutrient Capital and ER of nutrient retention mechanisms, as a result of in-
As a second example of the concept of ER, consider creased output in drainage waters, and/or as the result
the depletion of the site nutrient capital that can ac- of such postlogging site treatments as slashburning.
company harvesting. Among other things, trees need These losses are gradually replaced, of course, and for
 Nonrenewable Aspects of the Forest Ecosystem 605

a given loss of nutrients on a given site, there will be a practice until one knows the anticipated rotation
given nutrient recovery period, which might be re- length and rate of ecosystem recovery. Similarly, one
ferred to as the nutrient recovery ecological rotation. cannot define a particular rotation length to be sus-
This idea was presented in Figure 5-14, which tainable or nonsustainable until one knows the degree
showed the effect on site nutrient capital of rotations of ecosystem disturbance and the rate of ecosystem re-
that are shorter than the nutrient-recovery rotation covery. The complexity involved in determining ERs
and the effect of a given level of harvest depletion and suggests the need for ecosystem management simula-
rotation length on sites that vary in their rate of recov- tion models as described in Chapter 21.
ery. he length of the recovery period is a function of ER will vary for different forest values as well as
two things: the degree of site nutrient depletion ac- for different forest ecosystems. The ER for timber
companying harvesting and the rate of replacement of volume will be very different than for the natural de-
the losses. On a site that receives nutrients in seepage velopment of bear winter denning habitat, which re-
water or has large reserves of readily weatherable soil quires very large trees that are hollow at the base or
minerals, even substantial losses of these nutrients very large hollow logs. Clearly, ERs need to be esti-
may be replaced rapidly. On a site with very slow re- mated for all the varied values that are to be sustained.
placement and/or poorly developed nutrient accumu- Not until this has been done can one say with confi-
lation mechanisms, even a small loss may require a dence that the management has a high probability of
substantial period for replacement. being sustainable.
Because ERs depend so much on the degree of
ecosystem change caused by a particular disturbance
ERs and the Design of Sustainable and on rates of ecosystem recovery, they cannot be cal-
Forest Management culated without asound knowledge of ecosystem ecol-
The length of an ER is defined by three variables: ogy. Io be useable, this knowledge should be represented
in ecosystem simulation models. Although such models
cannot be expected to predict the future with complete
1. The degree to which ecosystem condition has
accuracy, a model that incorporates all the key ecological
been altered by disturbance, including the severity
processes and reflects the important role of organic and
of the disturbance and its spatial extent
biological legacies in determining sustainability can be
2. The rate at which ecosystem condition recovers the most effective way of assessing ERs and the probabil-
from disturbance ity of nondeclining patterns of change.
3. The frequency with which disturbance is repeated

In designing a sustainable management system 22.7 Nonrenewable Aspects

for a particular forest ecosystem, a forester can vary
the frequency of ecosystem disturbance by altering
of the Forest Ecosystem
the rotation length and, to some extent, by manag-
The discussion so far has focused on the renewability of
ing natural disturbance events. The degree of
resources in terms of rates of accumulation of materi-
management-induced ecosystem change can be varied als, structures, and species assemblages. ‘This is not the
by using different harvesting methods, silvicultural only aspect of the renewability of forest ecosystems.
systems, utilization levels, and/or site preparation and
stand management techniques. The potential rate of
ecosystem recovery is determined by the climate,
Genetic Constitution of the Ecosystem
topography, geology, physical and nutritional charac-
teristics of the soil, and rate of autogenic processes. By One of the most important nonrenewable aspects of a
managing soil organic matter, nutrient cycling, and forest ecosystem is the gene pool. The genotypes of
successional processes, a forester has the ability to vary the plants, animals, and microbes that constitute the
recovery rates. biotic community have been fashioned by millions of
Because of the multiply-determined character of years of natural selection, which has adapted organ-
ERs, it is not possible to interpret the sustainability isms to the wide variety of physical and biotic environ-
implications of a particular harvesting or management ments that exist in any area. As a result, organisms are
606 CHAPTER 22 Sustainability and Renewability ofNatural Resources, and Implications for Forest Management

able to acquire and accumulate biomass in virtually all Timber managers should be particularly con-
types of environment present on Earth. cerned with the conservation of tree genotypes. Seed
If particular genotypes are lost from the gene pool, collection and storage, together with gene banks of
then they will probably never be re-created. although vegetatively reproduced trees in tree orchards, can
other new genotypes will evolve and there is a finite make a significant contribution to this genetic conser-
possibility that a particular genotype could reoccur vation, but these mechanisms cannot substitute for
naturally by means of natural selection, this possibility having significant areas set aside permanently in ge-
is small and the process would require a very long netic-ecological reserves. It is the responsibility of
time, especially for long-lived organisms such as trees. professional foresters to ensure that this is done.
Humans may be able to speed up the process of re-
creating particular genotypes or generating new geno-
The Soil Resource
types by artificial selection and breeding, and as the
new science of genetic engineering develops, the pos- Trees, fish, mushrooms, berries, and terrestrial
sibilities of recreating lost genotypes will improve. wildlife are the major biotic crops yielded by most for-
However, on the basis of our present knowledge and est ecosystems. The basic resource from which these
technologies, it does not seem that it will be feasible crops develop is the soil.
within the foreseeable future to re-create a particular Soil is a complex phenomenon that results from the
genotype if all similar genotypes have been lost. We action of physical, chemical, and biological processes
simply do not know how to re-create a species once it acting on soil parent material over a period of time.
and its DNA have become extinct. Considering this The time required to produce a given type of soil is
and the lack of certainty concerning the re-creation of highly variable, depending as it does on climate, vege-
specific genotypes from similar genetic stocks, the ge- tation, animals, topography, and the nature of the par-
netic constitution of a species must, for the present ent material. Some types of soil form relatively rapidly,
time, be considered to be nonrenewable if it is lost or as for example the conversion of moist, unconsolidated
severely depleted. glacial till parent material to fertile soil in a few decades
Without appropriately adapted genotypes, ecosys- under the influence of nitrogen-fixing pioneer species
tems will not function normally, and possibly not at such as red alder. However, in many or most cases, de-
all. It is of paramount importance that we place top velopment of new soil is much slower. Mass wasting of
priority on conserving our genetic resources whenever soil that leaves only infertile soil parent material, espe-
we are planning the exploitation of biological re- cially in dry, cold, or hot environments, can reduce the
sources. Representative examples of all major types of productivity of a site for a particular tree species for an
ecosystem should be set aside in ecological reserves, extended period (several decades or even several cen-
which must be large enough to encompass the local turies) until pioneer plants and autogenic succession
variation of genotypes and to ensure the survival of ge- has restored appreciable levels of nutrient cycling and
netic diversity. These reserves will serve as future gene net primary production. Where all unconsolidated ma-
banks for plant, animal, and microbial populations and terial is lost and weathering-resistant bedrock is ex-
will act as long-term benchmarks against which we posed, it may be many tens of thousands of years or
can make future measurements of the long-term im- more before the original productivity of the site is re-
pacts of our resource management. Fortunately, pro- stored. Thus, in many situations, soil can be considered
grams to establish ecological reserves were set up in as a nonrenewable resource.
many countries in the 1970s, but many more must be The renewability of the soil resource is highly vari-
created if we are to conserve the world’s genetic inher- able. As noted above, nitrogen-fixing trees such as red
itance. Many countries have agreed to set aside 12% alder can grow very satisfactorily on exposed parent
of each of their major ecological zones, including all material that is devoid of nitrogen but in which weath-
representative site types. Although there is no scien- ering of primary minerals provides an adequate supply
tific basis or ecological justification for 12% and al- of mineral nutrients. In this case, the soil could be
though conservation on a global scale is more closely considered renewable. However, alder is not able to
related to the quality of ecosystem management of the fulfill this function on parent materials that lack
other 88%, there is no question that a comprehensive weatherable minerals. Substrates that are potentially
network of ecological reserves is a fundamental re- fertile if species such as alder are available may be to-
quirement for sustainable forest management. tally unsuitable for a later successional species because
 Study Questions 607

of the lack of nitrogen and organic matter until pio- flow of energy through the ecosystem, but it is also deter-
neer species have prepared the soil. The renewability mined by the population and community processes that
of the soil in terms of the productivity of a particular control pathways and rates succession.
tree species thus will depend on the nutritional and
moisture adaptations of the species involved. Many
tree species grow best in relatively well-developed TAKE-HOME MESSAGE
soils, and for these species, the soils on most sites can
The renewability and sustainability of forest resources are
be considered to be nonrenewable. Because of this, it
determined by the combination of the frequency, scale,
seems most appropriate to consider soil in general as a pattern, and severity of ecosystem disturbance that
nonrenewable resource that therefore must be con- matches the rate of ecosystem recovery and the overall
served to maintain the renewability of the plant and ecology of the forest values that are to be sustained. Re-
animal crops that it produces. newability and sustainability in forestry cannot be evaluat-
ed by “snapshot” evaluations of ecosystem condition or by
oversimplified evaluations of rotation length or silvicul-
SUMMARY tural system. It also cannot be evaluated by looking at only
one value, such as timber, at a time. Only when evaluated
At the start of this book, we examined forests as func- analytically in terms of overall ecosystem response can we
tional ecosystems. This chapter returned to this systems determine whether a particular forest value will be sus-
perspective in the context of the renewability of forest re- tained under particular natural or management-induced
source values. disturbance regimes. Only when we learn to respect the
The wide variety of values that people want from full ecological, biological, and temporal diversity of nature
forests have traditionally been considered to be either re- and the cultural and social diversity of human society can
newable or nonrenewable. This is not a useful classifica- we expect that we will develop a sustainable relationship to
tion. The renewability of a resource value has more to do forests and other aspects of our environment. The com-
with the energy flow in the processes responsible for its plexity of this issue requires not only the accumulation of
renewal and the social time scales than whether the value ecological knowledge but also the development and use of
is a living or nonliving entity. Where there is a high appropriate forest ecosystem management simulation
energy flow, the time scale for renewal is short and the models with which to apply that knowledge in developing
value is considered renewable, and vice versa. If we can in- an appropriate set of relationships.
crease the energy flow in these processes, then we can in-
crease the resource renewability; if we damage these
processes and reduce energy flow, then we will increase
the time required for resource renewal and reduce re- STUDY QUESTIONS
newability. 1. What is an “ecological rotation”?
Using this concept, we can assess renewability and, 2. How can you evaluate the ER for a particular forest
hence, sustainability through the application of the idea value?
of ERs. By understanding rates of ecosystem recovery
from disturbance, we can identify combinations of rota- 3, What is a material resource?
tion length and ecosystem disturbance that will sustain 4. Can you evaluate the sustainability of a particular for-
particular ecosystem values at desired levels or patterns est management practice simply by looking at its im-
of change. Ecosystem recovery is closely related to the mediate effect on the ecosystem?
 Cc H i& E R

353
Ecology and Environmental
Ethics in Forestry
—

wr
Mi

qT evolutionary record suggests that only those values and environmental services provided by forest
species that adapt to their changing environment stands and landscapes. There are two ethical principles
and exist in harmony with that environment survive and responsibilities that are implicit in this definition:
the test of time. In the recent past, we, the human
species, have adopted the alternative strategy of adapt- 1. That forestry should change as the balance of val-
ing the environment to suit our own changing needs. ues desired from the world’s forests change; and
Rather than balancing our numbers and activities to 2. That foresters should resist both current practices
match the carrying capacity of our ecological niche, we and suggested changes in practices that are incon-
are expropriating the niches of many other species and sistent with the ecology and sociology of the new
altering the environment, thereby increasing, at least balance of desired values.
in the short to medium term, the carrying capacity of
the environment for humans. We are now discovering, These two responsibilities require that forestry
however, that there is an ultimate limit to how far we and foresters are sensitive to both the needs and de-
can go in this direction without undesirable, long-term sires of people, and to the ecology of the ecosystems
consequences. The time has come to start modifying they are managing to satisfy those needs and desires. It
our behavior to render it more compatible with the en- requires that foresters are both socially and environ-
vironment and to fulfill our ethical obligations to fu- mentally ethical in their activities.
ture generations. This requires a change in attitudes In these days of environmental awareness, the pro-
toward our environment and resources, a recognition fession of forestry is under increasing scrutiny from a
of the consequences of the current unbridled expropri- critical public. Areas of degraded forestland, which are
ation of the niches of other organisms, a clear identifi- our legacy from past logging and “site preparation
cation of the maximum level of harvesting or other treatments” that proved to be inappropriate for the
forms of utilization of ecosystem productivity that is particular site, stand as a mute testimony that in some
compauble with long-term maintenance of ecosystem areas Our expectations of resource sustainability have
function, and the acceptance and application of envi- been overly optimistic. In some parts of the world,
ronmentally sound behavior patterns and management foresters have lost a lot of their public credibility, and
practices. Among these changes is the need to learn if they are to retain their current role as decision mak-
how to preserve and improve the natural renewability ers in the forestry domain, then it is important that
of resources and how to classify forest resources ac- they do not claim to be practicing sustainable forestry
cording to their renewability. when they are not. Harvesting of nonrenewable
Forestry is the art, practice, science and business of forests almost certainly constitutes timber mining, and
managing forested landscapes to sustain the balance of this should be made explicit. Timber mining is not

608
 CHAPTER 23 Ecology and Environmental Ethics in Forestry 609

necessarily an inappropriate management goal, how- of energy and materials between individuals in ecolog-
ever. Its danger for the forestry profession is not in- ical systems.
herent in the practice itself but only when it occurs The human species is as much involved with ecol-
consciously or unconsciously under the guise of sus- ogy in its day-to-day existence as is any other species
tainable forestry. To allow such non-sustainable prac- of animal and, together with other species, is destined
tices to masquerade as sustainable forestry diminishes to adjust to the environment or ultimately succumb in
the credibility of foresters in the eyes of the public and the evolutionary struggle. If we wish to be continuing
the scientific community. actors in the ecological and evolutionary play
Much is said these days about the need for “humil- (Hutchinson 1965), we will have to act out a sustain-
ity” in the face of our incomplete understanding of able role. It seems to be high time that economics and
ecosystems and for the need to be “respectful” toward ecology walk the same road and that an ecological ap-
“nature.” This philosophy is interpreted by many to proach is adopted for all aspects of human activity. It is
mean that humans should protect nature and disturb it also high time that the science of forest ecology be-
as little as possible. Our knowledge is certainly incom- come linked more closely to the environmental move-
plete and “always” will be. However, our knowledge is ment to prevent the return of forestry to the earlier,
continuing to expand, and if we ensure that all rele- administrative stage of its evolution as it struggles to
vant existing knowledge is used in making manage- evolve through the ecologically based stage to the so-
ment decisions and adopt an adaptive management cial forestry stage (Table 2-1). Figure 23-1 revisits this
approach—learning about ecosystems by managing simple model of the evolution of forestry, incorporat-
them and learning from the results (Walters, 1986; ing the reversal of this developmental trend that can
Walters and Holling, 1990)—then continual improve- occur if well-intentioned individuals or organizations
ment in forest stewardship should be possible. impose a rigid, regulation-based approach to forestry
Where knowledge gained by science or experience that fails to respect the diversity of nature and of the
is incomplete, management must proceed with cau- values that society wants from the world’s forests.
tion. We must respect the ability of ecosystems to Such reversals serve only to repeat the disrespect for
maintain their functional processes in the face of nat- nature that accompanied the exploitative and adminis-
ural and human-caused disturbance, and we must de- trative stages of forestry in the past.
sign management-induced disturbance regimes to be It is the responsibility of forest ecologists and man-
within the limits to which a particular ecosystem is agers of the world’s forests to work with all sectors of
adapted. Being “respectful” of nature, however, does society to ensure that such reversals do not occur. In
not necessarily mean creating minimal disturbance. In closing, what better than to revisit the words of David
most forest ecosystems, maintaining ecosystem pro- Thoreau and Aldo Leopold.
ductivity, biodiversity, and other attributes within his- Henry David Thoreau, widely considered the
torical ranges requires some frequency, scale, and progenitor of the modern wilderness and nature con-
severity of disturbance. Respecting nature in such servation movement, was fascinated by nature’s wild-
ecosystems requires that we sustain appropriate dis- ness (Botkin, 2001). Although he was a great admirer
turbance regimes rather than adapt a “soft footprint” of civilization and its cultural, scientific, and techno-
or “preservation” approach (Attiwill, 1994a,b; Kim- logical achievements, he recognized the important
mins, 1996, 1999, 2000, 2002, 2004). spiritual counterbalance of unmanaged nature: “In
Much of traditional economics is concerned with wildness is the preservation of the world.”
expanding economies, with growth, and with maxi- However, he was also fascinated by the discrepancy
mizing short-term returns on investment. In this con- between the romantic, mystical, and unrealistic view
text, economics and ecology are dancing to different of nature held by the society of his time and the re-
tunes. However, economics in its broadest sense is ho- ality that he experienced when he penetrated into un-
mologous to ecology; it is a branch of human ecology managed and unroaded forest and mountain
in that it studies the interactions and exchange of en- landscapes. He asserted an abiding truth: that the best
ergy and materials between individuals in socio- way to understand nature is first-hand field experience.
economic systems. Conversely, ecology might be con- The environmental debate over forests and forestry
sidered to be the economics of the environment in has been conducted far too much by individuals who
that it is concerned with the interactions and exchange lack adequate field experience and knowledge of the
610 CHAPTER 23 Ecology and Environmental Ethics in Forestry

The Evolution of Forestry Revisited

Local people with > Sustainable Exploitation— Passive Management— Active Management
experience-based wisdom | Replaced by
Resource
Nonlocals without —————————>- Nonsustainable exploitation ———>- depletion
local knowledge
Leads t
| inet Variable
Sa a dministrative forestry ———> results; often
, single value
Pressure from Evolves into
ecologically Sustained
inappropriate Ecologically based forestry, ———> timber
belief systems often initially timber biased production
and incomplete
knowledge about ; Application of social and biophysical
nature sciences that respect the ecology
and sociology of desired values

Social forestry—ecologically based,

multivalue ecosystem management

Figure 23-1 The evolution of forestry revisited. From the early preforestry stage to the social
forestry stage, there is a continuing danger of reversal of the development sequence. The application
of biophysical and social sciences is constantly required to guide the evolution of forestry that will
sustain all the values desired by society in harmony with the functioning of the global ecosystem
(Kimmins, 2002).

ecosystems in question and at locations physically re- ecosystems contain certain levels of functional redun-
mote from these ecosystems. dancy (Perry and Amaranthus, 1997), that nature is
Aldo Leopold provided the key philosophical constantly undergoing stand-level changes in struc-
foundations for developing an ethical foundation for ture and species composition, and that this change is
conservation and forest management. A forester, part of and necessary for “healthy” ecosystem func-
farmer, hunter, and fisher but foremost a conserva- tioning (Attiwill, 1994a,b; Rogers, 1996).
tionist and philosopher, Leopold’s “The Land Ethic” Leopold’s quote, “A thing is right when it tends to
and other essays published in Round River and A Sand preserve the integrity, stability, and beauty of the biotic
County Almanac should be required reading for all community. It is wrong when it tends otherwise,” has
those who work in resource management and conser- been wrongly interpreted in terms of “balance of na-
vation (Leopold, 1953, in Leopold, 1966). One of his ture” and human aesthetic beauty. Elsewhere in his
widely quoted ideas is, “Jo keep every cog and wheel is the writing is the explicit reference to change as part of
first precaution of intelligent tinkering.” Frequently in- stability, integrity as being the maintenance of a nat-
terpreted in the environmental debate as implying ural range of variation in ecological processes, and
constant species lists and no-change forestry at a stand beauty as being the functional beauty—the interac-
level, Leopold was in fact referring to metapopula- tions and interdependence between the different
tions, a landscape-level concept not yet developed in components—of the ecosystem. He supported this
his time, rather than to stand-level species richness interpretation of the above sentence on the very next
and structure. Elsewhere in his writing (Leopold, page of Land Ethic where he says, “The evolution of a
1966), he rejects the fallacious “balance of nature” in- land ethic 1s an intellectual as well as emotional process.
terpretation ofVictorian times that implies that nature Conservation is paved with good intentions which prove to
is stable and, therefore, that change is antinature. be futile, or even dangerous, because they are devoid of
“This figure ofspeech fails to describe accurately what little critical understanding either of the land, or of economic
we know of the land mechanism.” He understood that land use.”
 CHAPTER 23 Ecology and Environmental Ethics in Forestry 611

Daniel Botkin, a contemporary thinker and writer Sustainable forest management must start with a
on the subject of conservation and a student of Tho- set of management objectives for a defined forest
reau'’s writings (Botkin, 2001), noted the ineffectiveness area. These objectives must reflect a balance between
of all paradigms of nature that do not directly respect what is ecologically possible and sustainable and
nature’s spatial and temporal diversity. The “perfect maintenance of ecosystems within the socially accept-
Swiss watch” analogy, in which it is assumed that the able portion of the historical natural range of varia-
system will fail and collapse if a single “cog” is re- tion. Humans are unique among the mammals in the
moved, is simply not an accurate reflection of the way degree to which we depend on our visual senses, the
that ecosystems function (Botkin, 1990). Echoing the degree to which we make decisions from our heart
ideas of Thoreau and Leopold, he noted that the only rather than our head (decisions based as much on our
way to understand nature is to study it directly in the emotions, values, and ethics as on our logical analysis
field, not theorize about it from some remote location. or instincts), and our dislike of environmental and
Successful management must be based on a re- aesthetic change (we do not really know whether
spect for nature as it is, not as we romantically might other species dislike change as much as we seem to,
wish it to be (Kimmins, 2000a). “Respect” has two other than when it denies them security, thermal pro-
groups of meanings: tection, and resource values). Because of these attrib-
utes, many in society equate aesthetically pleasing
¢ to honor, revere, and be deferential toward, and
forestry that produces little short-term visual change
* to notice with attention, to pay due regard to...
and satisfies our emotions with “good stewardship,”
the object of respect.
“ethical forestry,” and sustainability (Kimmins, 1999,
The first interpretation provides a basis for devel- 2000b).
oping the objectives of management—what are the Unfortunately, visual evidence is not always a reli-
values that are to be sustained in the ecosystem. ‘The able guide to the sustainability and ecological conse-
second asserts that to be successful in sustaining these quences of forestry. As a result, forestry should be
values, we must understand ecosystems and their spa- based on a balance between aesthetics and emotion-
tial and temporal diversity and develop site-specific based judgments and decisions based on an objective
and value-specific management strategies based on analysis of the ecology and sociology of the values that
this understanding. are to be sustained (Sheppard and Harshaw, 2000).
 _ —

”~

5 > * My awe AIA

~
THTTh wie * wav “Tuna
- = Lae uw — wily
in altri ett (iy :
pa!

rhivs.,
; hw
cvioy’
‘eer dei wernt
nt. mah
mpié

ey lett
Maen a
: ievtou le a Dae .
: “0 nus ® ae plies 3sie eye
; toh STOTT gagsueree' th sour
fk rhe
2’) cphel) once Gms
: Li seutet nah dsogaad bee ne AT Wwe
Aroril ti (ote aera: bane eine
aa it a rowen VOB iimwie searbeils Fea
“te
. .
| | <i boll ite snsneger
eee ltd
ar es Det y oO
¥
rr =I" ADS, ecnihrana 9
ae hee SOIT: Wt
eae
-

: ' aria eh ih

ee
wary Wait hs nel RaresTH ° . ’

y . (was seeing
— |
| ) ” ~ i~ Bit is:
— ~<a “win noice
: :
My argh on
’ “(es
;
| | ~ : a Ope Aloe
of
:
\vtewe ot os olTaege
, — n leg A te
ah
% ie ¥ ie corre? lah
a
rSyeaty ‘ouki
- : ;
ee '
o5y sgt
 Glossary of Technical Terms
Acclimation A change in the morphology, physiology, be- competitive abilities of other plant species. Allelochemicals
havior, and/or life cycle of an organism in response to may affect germination, growth, root development, mycor-
changes in the physical or biotic environment. A phenotypic rhizae, or the availability of nutrients by affecting decompo-
response to environmental pressures. sition and mineralization rates of soil organic matter.
Acid rain Rain that contains concentrations of hydrogen Allelopathy ‘Vhe production and release of chemicals by a
ions (H") greater than that of rain falling through “normal,” plant that have a more negative effect on other species than
unpolluted air. Acid rain usually contains significant concen- on the plant producing the chemical. A plant adaptation that
trations of oxides of sulfur and nitrogen, which, in combina- is thought to increase the competitive status of the species.
tion with H", form acids. Acid rain is often accompanied by
elevated concentrations of atmospheric ozone. The damage Allen rule In general, the extremities of homeothermic
caused to plants by acid rain is often the combined result of animals are smaller in cold than in hot climates.
the acidity, high loads of sulfur and nitrogen, ozone or other
Allogenic succession ‘(he changes in the living community
air pollutants, and the interaction of these with stresses and in the soil and microclimatic characteristics of an
caused by summer droughts, freezing damage to roots, in- ecosystem as a result of alterations in the physical environ-
sects, and diseases.
ment that are independent of changes in the living commu-
Acidity ‘The concentration of hydrogen ions in solution. It nity. Floods, wind, landslides, fire, drought, climate change,
is measured as pH, which is the log of the reciprocal of the and sediment deposition by a river are examples of factors
hydrogen ion concentration (i.e., pH = log 1/H* concentra- that can cause allogenic succession.
tion). The higher the concentration of H*, the greater the
Alluvial soil Soil developed in parent material transported
acidity but the lower the pH value.
and deposited by water. The material is well sorted into size
Actinomycetes A group of microorganisms that exhibit classes that vary from silts to sands to gravels, according to
characteristics of both fungi and bacteria. They consist of the speed of movement of the water. Characteristic of valley
nonseptate branching hyphae, which are often brightly col- bottoms and flood plains.
ored (e.g., the nitrogen-fixing Frankia).
Alpha diversity The diversity of species and the complex-
Adaptation A change in the genetic make-up of a popula- ity of community structure and function in a particular for-
tion as a result of genetic selection in response to changes in est stand or local ecosystem. Species diversity may simply be
the physical or biotic environment of that population. A species richness (number of species) or one of several indices
genotypic response to environmental pressures. that combine species richness with some measure of relative
commonness or rareness (species evenness), or some other
AEC Anion exchange capacity. This is the capacity of a soil measure of the relative abundance of species.
to absorb negatively charged ions such as nitrate (NO; ) or
phosphate (e.g., P;O,”) and hold them against the leaching Ammonification ‘The conversion of organic forms of ni-
action of water moving down through the soil. The AEC of trogen into inorganic ammonia gas (NH;) or ammonium
many forest soils is low. ions (NH,’). This is an important process in the mineraliza-
tion of organic matter that makes the nitrogen in unavailable
Aeolian soil Soil developed in parent material transported organic forms available for uptake by plants. Ammonifica-
by wind. The material is well sorted into particular particle tion is conducted by a variety of heterotrophic microorgan-
size classes according to the strength of the wind. Normally isms (fungi, bacteria, actinomycetes).
silt or fine sand-sized particles.
Anabatic winds Winds that result from the daytime warm-
Albedo ‘The percentage of solar radiation (including light) ing of a valley causing upslope winds.
that reaches the surface of the earth and is reflected back to
the atmosphere by that surface. Light-colored surfaces that Antagonism Relationship between two species in which at
reflect a lot of radiation have high albedo values (snow, sand). least one is adversely affected.
Dark-colored surfaces, such as peat soil or forests, have low Antecedent determinant A factor that contributes to the
values; they absorb most of the radiation that falls on them. determination of a condition or an event. It must exist or op-
Allelochemical A chemical that is produced by a plant and erate before the condition or event occurs and be determin-
released into its environment and in some way alters the istically related to it.

G-1
G-2 GLOSSARY

Antibiosis Antagonistic chemical interactions between dif- Biochemical or internal cycle Internal retranslocation of
ferent species. nutrients within a plant to satisfy immediate physiological
needs for nutrients that cannot be satisfied or satisfied so
Aridity he dryness of an environment, not simply in
rapidly or efficiently by uptake from the soil or atmos-
terms of the water content of or inputs to that environment phere.
but the ecological effectiveness and availability of water to
organisms in that environment. Bioclimatic rule Biological events in the spring are de-
layed 3 to 4 days for each higher degree of latitude and for
Assart effect he period of increased nutrient availability
each 100- to 130-m increase in altitude.
in the soil after natural or human-caused disturbance to a for-
est ecosystem. The increase results from one or more of the Biodiversity Variation in the biotic community. Used syn-
following: reduced nutrient uptake by plants, accelerated de- onymously with the term biological diversity. There are many
composition and mineralization of organic matter because of measures of biodiversity: genetic diversity, local species rich-
increased soil moisture and temperature, removal of mycor- ness and evenness, and local diversity in community structure
rhizal roots (the Gadgil effect), reduced competition for lim- and function (alpha diversity); variation in species richness
iting nutrients between microbes and plants, an increase in and community structure and function across the local land-
the mass of litter available for decomposition, an increase in scape (beta diversity); and changes over time in these
soil pH, and an increase in soil animal activity. Often associ- measures of biodiversity (temporal diversity). Landscape
ated with increased soil leaching and sometimes with denitri- (physical or ecological) diversity provides a framework for re-
fication. The assart effect results in increased site fertility and gional biodiversity (gamma diversity). Some people would
changes in the rate of growth and in species composition of prefer the term to be restricted to the biological diversity of
the plant community while it lasts. Nutrient-demanding ecosystems unaltered by human activity, but the term is now
species are favored, and these may grow for a period on sites so widely used that it is unlikely that this will happen.
where they are not found in the absence of disturbance.
Biogenic succession ‘The change in the living community
Assimilation efficency ‘The efficiency with which an or- and in the soil and microclimatic characteristics of an
ganism assimilates ingested food. ecosystem caused mainly by the animal and microbial com-
Association See Plant association. ponents of the living community. Insect outbreaks, mam-
malian herbivores, and epidemics of plant disease organisms
Autogenic succession ‘Uhe change in the living community are examples of biotic factors that can substantially alter the
and in the soil and microclimatic characteristics of an patterns of ecosystem development that would result from
ecosystem caused by the living community itself (auto = self; autogenic or allogenic processes acting alone.
genic = caused or generated); refers to effects caused by the
plant community. The major mechanisms of autogenic suc- Biogeochemical cycle ‘he circulation of nutrients between
cession are invasion and colonization, environmental alter- living organisms and the soil within a particular ecosystem.
ation, and species exclusion. Uptake, distribution within plants, losses from plants by lit-
terfall, defoliation, reproduction, foliar leaching, plant death,
Autotoxicity Antibiosis in which a species produces chemi- and decomposition and mineralization in the soil are the
cals that are inhibitory to its own seedlings.
major processes. This cycle receives inputs from the geo-
Autotrophs Organisms that can obtain the energy they need chemical cycle and loses nutrients to it. Nutrients are with-
for their metabolism from abiotic sources. Photoautotrophs drawn from it into the biochemical or internal cycle. Current
obtain energy from solar radiation. Chemoautotrophs obtain net primary production of an ecosystem is generally closely
energy by transforming inorganic chemicals. related to the quantity of growth-limiting nutrients circulat-
ing in this cycle, except where intentional cycling is able to
Base saturation (of soil) The percentage of the cation ex- supply much of the nutrient demand for new growth.
change capacity occupied by exchangeable cations other
Biogeoclimatic classification Classification of ecosystems
than hydrogen and aluminum (e.g., K*, Mg”*, Na*, Ca”).
on the basis of the potential climatic climax vegetation on
Bergman’s rule Homeotherms at colder, higher latitudes “mesic” sites with zonal soils. The classification is based on
are generally larger and have a lower surface area/weight ratio regional climatic classification (formations, regions, zones,
than species of the same genus at warmer, lower latitudes. and subzones) and the soil moisture and nutrient regimes
and the structure and species composition of the vegetation
Beta diversity Variation in species richness or some other
in local ecosystems (site associations, series, and types).
measure of alpha species diversity and of stand-level structur-
al and functional diversity across a local environmental gradi- Biogeoclimatic subzone A subdivision of a biogeoclimatic
ent such as elevation, or soil moisture and fertility gradients. zone according to differences in soils, humus forms, minor
The variation in measures of alpha diversity between the dif- vegetation, early seral species, and the secondary species that
ferent ecosystem types or seral stages in a local landscape. accompany the canopy dominants of mature or climax plant
 GLOSSARY G-3

communities. All subzones within a zone will have the same ing of the soil that resists change in pH when acid rain adds
dominant climax species on mesic sites with zonal soil. hydrogen ions).
Biogeoclimatic zone A climatically defined geographical Bulk density Mass of soil material per unit volume of soil.
area with a mosaic of soils and vegetation, the character of Bulk density has an important influence on soil aeration and
which broadly reflects the regional climate. The geographi- drainage and on root penetration.
cal limits of a zone are defined by the climatic climax or ma-
ture vegetation on mesic sites with zonal soil. Capillary forces Physical forces that hold water in small
spaces against the force of gravity or uptake by plants. Can
Biological diversity he diversity of life: diversity in cause low water availability to plants in fine-textured soils as
species, behavior, physiology, morphology, life cycles, ecolog- they begin to dry out.
ical characteristics, ecological function, habitat requirements,
Carbon allocation ‘The distribution of carbon and energy
etc.; often used interchangably with biodiversity. Different
(captured in the process of photosynthesis) to the various
from ecological diversity but influenced by it. Ecosystem di-
different organs of a plant. Often used to refer to the relative
versity includes biological diversity, but ecosystem diversity is
allocation of photosynthate between belowground and
not a subset of biological diversity as is often claimed.
aboveground organs or sometimes to the proportion of
Biomass ‘The mass of a living organism, population, or net primary production that is allocated to ephemeral fine
community. When applied to a tree, it normally refers to the roots. It affects root-shoot, height-diameter and other bio-
entire mass of the tree, even though much of this mass may mass ratios.
be heartwood, which is dead tissue. Sometimes used to refer
Cardinal temperatures A set of temperatures that define
to the total amount of organic matter in an ecosystem, both the overall temperature adaptations of a species: optimum
live and dead. temperatures, maximum and minimum effective tempera-
Biome A division of the continental ecosystem according tures, and maximum and minimum survival temperatures.
to the life form of the dominant vegetation of the climax or Cardinal temperatures vary for different organs and physio-
late seral stages. A biome consists of a plant formation to- logical processes and with the age of the organism, the time
gether with associated animal and microbial species and of a ofyear, and the organism’s condition.
life zone, which has a characteristic range of soil conditions Carnivore A meat-eating animal. A secondary consumer.
and regional climate (a subdivision of the continental cli- An animal that feeds (i.e., gets its energy and nutrients) by
mate). A group of ecosystems in which the primary produc- eating herbivores or other primary consumers.
ers have similar growth forms and consumers have broadly
similar feeding habits. Carrying capacity ‘The number or mass of a particular
species or group of species that can be sustained indefinitely
Biotic climax A climax community or seral stage, the char- in a given environment.
acter of which is determined by the impact of animals, such
as defoliators, browsers, or bark beetles, or other organisms Catabatic winds Winds that result from the nighttime
that arrest or alter the process of autogenic succession, cooling of ridges, causing downslope winds.
thereby resulting in repeated replacement by the same plant Causal determinism Situations in which an ecosystem
community. event, condition, or process is determined by a single an-
Boundary layer Thin layer of air around a leaf or other ob- tecedent determinant—the cause of the event, condition, or
ject whose temperature and/or humidity is significantly mod- process. Causally determined events or conditions are rela-
ified by that object in comparison with the surrounding air. tively easy to predict. Causal determinism is very uncom-
mon in forest ecosystems.
Brunisol A forest soil characteristic of coniferous or decid-
Cause A necessary and sufficient antecedent determinant.
uous forests in cool to warm, humid to dry climates that de-
A condition, process, or factor, the presence or operation of
velops in parent material of medium to high fertility.
which will always lead to the same event or condition and
Characterized by a modified B horizon (Bm) that fails to sat-
that is required for that event or condition to occur. Causal
isfy the criteria for a podzol or luvisol. Generally with a
determinants have a one-to-one relationship with the
moder or mull humus form.
event/condition that they cause. They are uncommon in
Bud bank The population of viable but dormant buds or ecological systems where most events or conditions are mul-
adventitious meristems beneath the stem bark or on the tiply determined.
roots of plants that will become active and grow if the CEC Cation exchange capacity. This is the capacity of a
branches or aboveground parts, respectively, are killed. soil to absorb positively charged ions such as potassium (K”),
Buffering capacity The ability of a system to resist change ammonium (NH,’), hydrogen (H’), or calctum (Ca?*). Most
in physical or chemical conditions (e.g., the chemical buffer- forest soils have a high CEC because of the abundance of
G-4 GLOSSARY

negatively charged soil organic matter and the negatively humid parts of the world, forests are the climatic climax veg-
charged clay—humus complex. etation. In some northern and high-elevation forests or in
very cool humid areas, closed forest is not the climatic cli-
Character species A plant species that is largely restricted
max condition. Sphagnum bog (muskeg) or open woodland
to and is indicative of a particular plant association or vege-
dominated by ericaceous shrubs is the climax in the very
tation or ecosystem unit under consideration. long-term absence of disturbance.
Chemoautotroph An organism that is able to obtain the Coarse fragment content The percentage of the soil vol-
energy that it needs for growth, reproduction, and life ume occupied by mineral particles larger than 2 mm.
processes from inorganic chemicals in its environment. It
does this by reducing these chemicals to a lower energy Coenocline The sequence of biotic communities along a
state, using the energy that is released in its metabolism. major environmental gradient; also called a community gra-
dient. The biotic part of an ecocline.
Chernozem A grassland soil characterized by a thick (>10
cm), dark, organic-rich Ah layer formed largely from the Colluvial soil Soil developed in parent materials trans-
turnover of grass roots. ported downslope under the influence of gravity.
Chilling requirement Part of the thermoperiodic adapta- Colonization ‘The first stage in autogenic succession after
tion of organisms in thermally variable environments with a disturbance has removed the previous biotic community or
stressfully cold winter or risk of spring frosts. Resumption of created openings where invading species can get established.
various biological activities, such as spring bud burst or Consists of invasion and survival.
awakening from winter dormancy, may require exposure to a
Commensalism A symbiotic relationship between two
minimum period below some critical temperature.
species in which one is benefited while the other is neither
Chinook (or foen) wind Warm and drying winds that often benefited nor harmed.
occur on the lee side of large mountain ranges.
Comminution ‘The breakup of dead organic matter (e.g.,
Clay-humus complex The negatively charged complex leaves, branches, roots) into smaller pieces by soil animals.
formed by stable humus molecules and clay particles in the This increases the surface area of the organic matter and
soil. This complex acts as an exchange site (CEC) for posi- mixes it with microbes (especially bacteria), thereby facilitat-
tively charged ions. It is a very important contributor to ing decomposition.
site fertility and acts to reduce leaching losses of soluble
Community ‘The assemblage of living organisms (plants,
cations.
animals, microbes) that interact with each other in energy
Clearcutting The removal of all trees in a single harvest flow and nutrient cycling processes in an ecosystem. The bi-
from a sufficiently large area that the “forest influence” is re- otic component of a particular ecosystem.
moved from at least 50% of the harvest area. The “clearcut”
Compensation point See Light compensation point.
condition persists until the regrowth of forest vegetation has
restored the forest influence on more than 50% of the site. Competition Occurs when two different species attempt
“Forest influence” refers to the microclimate; the soil cli- to utilize the same resource when that resource is in lim-
mate; and the effect of the trees on soil organisms, soil ited supply.
processes, and hydrological processes. It can also refer to
habitat conditions for aboveground wildlife species. Competitive exclusion No two species that compete for the
same set of limiting resources can coexist indefinitely. One of
Climatic climax A climax community and ecosystem con- the species will prove to be competitively superior and sur-
dition (i.e., a climax seral stage), the character of which is de- vive, whereas the other is excluded from the area in which
termined by the regional climate. Climatic climax ecosystem the competition is occurring. However, this will occur only
conditions are found only where the rates of autogenic suc- when the competition persists long enough for exclusion to
cession are high relative to the frequency and intensity of occur. In fluctuating environments, exclusion may not occur.
disturbances that cause successional retrogression or retard
Complex gradient ‘The sequence of physical environments
or prevent ecosystem development under the influence of
autogenic succession.
along a major environmental gradient. The physical part of
an ecocline.
Climax The end point of autogenic succession: the final
Computer model ‘The implementation on a computer of a
seral stage in a sere. A self-replacing community that is rela-
tively stable over several generations of the dominant plant conceptual model ofa system or a process.
species or very persistent in comparison with other seral Conditioning factor A factor determining the size and distri-
stages. The character of the climax community depends on bution of a population that operates in a density-independent
the frequency and intensity of ecosystem disturbance rela- manner, e.g., weather factors, availability of shelters. Most im-
tive to the rate of autogenic succession for the site. In many portant in physically extreme or variable environments.
 GLOSSARY G-5

Consistence (soil) Refers to the stickiness of soils and their orrhizal associations. This avoids the transfer of nutrients
ability to resist deformation under stress. from organic forms to mineral forms in soil solution by the
Constancy Lack of change over time, or in the face of process of decomposition and mineralization. Nutrients in
stress, of some biological or physical characteristics of the solution either may be leached or can be competed for by
the roots of other plants or by soil microbes. Direct nutrient
ecosystem. A measure of ecosystem stability.
cycling makes the biogeochemical cycle “tighter,” reducing
Contagion Clumping of individuals in a population. losses from this cycle. It enables plants that utilize this nutri-
ent cycling strategy to be more competitive by monopoliz-
Continuum Variation in species composition along an en- ing limited nutrient resources.
vironmental gradient in which there is no obvious grouping
or association of species. Each species is distributed along the Disclimax Occurs where communities are held in an
gradient solely in response to its genetically controlled toler- early successional condition by human activity or that of
ances to and requirements for factors of the environment. other animals.

Convergence See Ecological (successional) convergence. Disturbance Allogenic or biogenic processes that alter the
rate, pathway and/or the direction of successional change
Crown The horizontal layer of the vegetation where most that would be expected from autogenic succession in the
of the branches and foliage are found. ecosystem in question.
Crown fire Rapidly moving fire that burns through the Dormancy A period of arrested activity and altered metab-
crowns of woody vegetation, frequently leaving stems and olism undertaken to avoid a period of environmental stress
much of the forest floor relatively untouched. (e.g., drought, heat, low temperatures).
Crumb A porous soil ped or aggregate of mineral particles. Ecocline ‘The sequence of biotic communities and asso-
ciated physical environments along a major environmental
Cyclical stability Property of an ecosystem that changes gradient (the complex gradient), such as a major elevational
through a sequence of conditions or stages that brings it or latitudinal gradient.
back to the original condition.
Ecological actors A component of the metaphor of “Eco-
Decomposition ‘The loss of mass of dead organic matter as logical Theatre”. Ecological actors are the species and other
complex organic molecules are broken down into simpler measures of biological diversity that occupy the “Ecological
molecules or into their constituent chemical elements and Stage” at various times in the “Ecological Play”.
energy is released. Decomposition is mainly done by mi-
Ecological (successional) convergence Where the fre-
crobes (fungi and bacteria), but their activity often requires
quency and severity of disturbance are low relative to the
that the organic matter first be comminuted by soil animals.
rates of autogenic succession, the surface soil condition, the
Deductive reasoning A logical argument that goes from microclimate, and the biotic communities within a given cli-
the general to the specific (e.g., the development of a matic area will become increasingly similar and may eventu-
testable hypothesis from a scientific generalization). ally become essentially the same in a climatic climax seral
stage. Dry sites get moister. Wet sites get drier. Poor sites
Density-dependent factor A factor that determines the size get richer. Rich sites may get poorer. In most forest eco-
and distribution of a population, the effectiveness of which systems, one can observe a tendency for convergence to
increases as the density of the population increases. occur, but disturbance will normally prevent complete
Density-independent factor A factor that determines the convergence.
size and distribution of a population, the effectiveness of Ecological diversity Variation in the physical characteris-
which is not influenced by the density of the population on tics of ecosystems across a landscape caused by variations
which it is acting. in soil, slope, aspect, elevation, climate, and geology, and
Determinism A philosophical principle that holds that all the accompanying variation in biotic communities. Eco-
events or conditions are the result of identifiable antecedent logical diversity provides the ecological framework (the
in- “Ecological Stage”) within which biological diversity de-
determinants and that, ultimately, as our knowledge
velops. Much of the biological diversity across landscapes
creases, all events or conditions should be explainable (also
(beta and gamma diversity) is a direct reflection of ecolog-
know as Logical Positivism).
ical diversity.
Detrivore An organism that obtains its energy by consum- Ecological efficiencies “These determine the efficiency of
ing dead organic matter: a decomposer organism. Also energy transfer between trophic levels and thus influence
called a saprotroph. biomass storage in and the trophic structure of ecosystems.
Direct nutrient cycle The transfer of nutrients directly Utilization, assimilation, tissue growth, ecological growth,
from litterfall to plant roots via the fungal partners of myc- and trophic efficiencies.
G-6 GLOSSARY

Ecological group Group of species that share similar adap- Ecosystem diversity Diversity in the characteristics of O
—“
e
—

tations and tolerances to the conditions of their environment. ecosystems—their structure, function, complexity, the interac-
tion of their components and their change over time (ecologi-
Ecological landscape See Landscape. cal succession). Biological diversity is an important component
Ecological niche See Niche. of ecosystem diversity, but does not solely define it.
Ecological play A component of the metaphor of “Ecologi- Ecosystem health ‘There is no single definition of ecosys-
cal Theatre”. The ecological play is the sequence of biotic tem health. The concept can be defined only within the con-
communities, physical conditions and ecosystem processes text of the desired values that a particular seral stage of a
that successively occupy and are replaced on the ecological forest ecosystem or a particular forest landscape is supposed
stage over time as a result of ecosystem disturbance and re- to provide. One could say that if an ecosystem has its in-
covery. The ecological play is ecological succession. tegrity, then it is “healthy,” and vice versa.
Ecological pyramid A diagrammatic method of representing Ecosystem integrity The maintenance of an ecosystem
the biomass of, the numbers of organisms in, and the energy within the range of conditions or seral stages in which the
flow through the different trophic levels in an ecosystem. processes of autogenic succession operate normally to return
Ecological rotation ‘The time that is required between suc- the ecosystem to or toward its predisturbance condition.
cessive timber harvest—related disturbances for an ecosystem Ecosystem integrity is very different from the integrity of a
to recover to its predisturbance condition or to some desired particular seral stage or condition, such as the integrity of
new condition before the subsequent harvest. Ecological ro- the old-growth condition. An ecosystem that has been re-
tation is a function of both the degree of ecosystem alter- gressed from an old-growth condition to an earlier seral
ation by the disturbance and the rate of the autogenic stage may not have experienced any loss of ecosystem in-
process of recovery (the resilience of the ecosystem). tegrity, but there will have been a loss in the integrity of the
old-growth condition of that ecosystem.
Ecological succession ‘Vhe process by which a series of dif-
ferent plant communities and associated animals and mi- Ecosystem management Management of an ecosystem as
crobes successively occupy and replace each other over time an integrated system rather than managing individual
in a particular ecosystem or landscape location after a distur- ecosystem components and processes separately. Manage-
bance to that ecosystem. Includes the accompanying change ment to maintain ecosystem processes and the structures
in the nonliving environment (soil and microclimate). and functions of the ecosystem within historical or “natural”
(i.e. characteristic of that ecosystem) ranges of spatial and
Ecological theatre A metaphor in which ecological diversi- temporal variation, or a socially acceptable subset thereof
ty determines the “Ecological Stage”, which in turn defines
which is ecologically sustainable.
which “Ecological Plays” can be “acted” out, which in turn
determines which “Ecological Actors” (species, measures of Ecosystem site type A specific type of ecosystem, character-
biodiversity) can occupy the “Ecological Stage” at different ized by relatively homogeneous soil conditions, microcli-
times. matic conditions, a characteristic climax plant association,
and associated animals and microbes. Generally the smallest,
Ecology ‘The science that studies ecosystems and their
most homogeneous unit of ecosystem classification. A subdi-
components, processes, and change over time. The science
vision of an ecosystem (site) association according to varia-
that studies the abundance, dynamics, and distribution of or-
tion in soil conditions.
ganisms and their interactions with their living and nonliv-
ing environment. Ecotone ‘The interface or zone of transition between two
Ecosystem An ecological system composed of living organ- plant communities that differ in life form (physiognomy) or
in species composition, or both (e.g., the forest/grassland in-
isms (plants, animals, and microbes) and their nonliving en-
terface, or between two different types of forest). A zone of
vironment (climate and soil in the case of terrestrial
tension between two different communities. Characterized
ecosystems; aqueous environment and substrate in aquatic
by higher environmental, species, and structural diversity
ecosystems). To be an ecosystem, these components must be
than the adjacent communities. Sometimes referred to as the
spatially arranged and have the appropriate interactions that
boundary zone between different adjacent communities or
lead to the capture and storage of energy as biomass, a
community ages/conditions.
trophic structure, a circulation of nutrients, and change over
time (ecological succession). Ecosystems are characterized Ecotope ‘The physical components of an ecosystem: the soil
by five major attributes: structure, function, complexity, in- and atmospheric components, attributes, and processes of a
teraction of the components, and change over time. terrestrial ecosystem.
Ecosystem (site) association A\l ecosystems (sites) that have Ecotype A genetic subdivision of a species in response to
the same climatic climax vegetation potential. variations in the environment of that species over its range.
 GLOSSARY G-7

Most ecotypes that have been identified have arisen by nat- Exponential growth Growth in numbers or mass in which
ural selection in response to variations in the physical envi- the rate of change increases with time.
ronment (e.g., different seasonal day-length or temperature
Extrinsic population regulation Regulation of population
regimes, major differences in soils). Ecotypes can also repre-
size caused by factors external to the population, such as
sent genetic response to differences in the biotic environ-
predators, weather, disease, or fire.
ment (e.g., natural enemies).
Facilitation pathway A successional pathway in which
Edaphic climax A climax community or seral stage, the
species of later seral stages cannot invade an area and grow
character of which is determined by soil conditions; largely
until the earlier seral communities have moderated environ-
soil moisture, soil nutrient status, and soil aeration.
mental conditions (e.g., moisture, temperature, nutrient
Edaphic grid A two-dimensional graph with relative soil availability) that cannot be tolerated by the later seral species.
moisture status (soil moisture regime) on the y axis and soil
Fidelity A measure of the degree of restriction of a species
nutrient status (soil nutrient regime) on the v axis. This grid
to a particular plant association, ecosystem type, or environ-
can be used to show the distribution of plant associations or
mental condition. Species with high fidelity values have high
ecosystem types or the variation in growth of a species in re-
indicator value.
sponse to these soil characteristics. Used as the basis for site-
specific decisions about crop tree species, site treatments, Field capacity “Vhe amount of moisture that a soil can hold
and silvicultural systems. against the force of gravity. In coarse- and medium-textured
soils, it is the water content of a soil 2 or 3 days after soil sat-
Effective temperatures ‘The maximum and minimum tem-
uration, assuming no losses by evaporation and transpiration.
peratures within which an organism can successfully com-
plete its life cycle, irrespective of exposure time. Can also Forb An herb that is not a fern or a grass.
refer to the effective temperatures of an individual organ or Forest ecosystem An ecosystem dominated by trees, in which
physiological process. the microclimate, soils, hydrology, and nutrient cycling, the
Elasticity The speed with which an ecosystem returns to biomass creation, storage and turnover, the minor vegetation,
its original condition after disturbance. Also called re- animals and microbes, and the food chain processes reflect the
silience. A measure of ecosystem stability. dominance by large, long-lived, woody plants.
Endemic parasitism Parasitic interactions in which the Forest ecosystem management See Ecosystem management.
parasite may weaken or have little negative effect on its prey Forest floor A layer of dead organic matter and living or-
but does not kill it. Characteristic of long-standing and sta- ganisms on the surface of the mineral soil in a forest, the
ble parasite—host relationships. mass of which consists largely of dead plant parts in various
Energy flow diagram An elaboration of an energy ecolog- stages of decomposition. In some forest floors, dead micro-
ical pyramid to show energy storage in and transfers be- bial and soil animal matter may make a significant contribu-
tween the different trophic levels of both grazing and tion to the mass. In many forests, the annual death of fine
decomposer food webs. roots contributes more to the forest floor than the above-
ground litterfall of branches and leaves. Soil microbes and
Epicormic shoots Shoots formed from dormant buds or small soil animals are a vital component of the forest floor,
meristems beneath the bark after exposure of previously which plays a critically important role in many aspects of
shaded bark to full sunlight (e.g., by the removal or death of forest ecosystem function.
branches).
Forest influence The modification of the microclimate
Epidemic parasitism Parasitic interactions that generally (temperature, humidity, light, wind) and of soil chemistry,
lead to the death of the prey. Epidemic parasitism may be physical properties, biology, and microbiology caused by the
difficult to separate from predation. Characteristic of new or presence of trees. The forest influence is greatest in a closed
relatively recent parasite—-host relationships in which selec- forest but extends one to three tree heights into nontreed
tion has not yet had a chance to create a sustainable relation- areas, the distance depending on the latitude, slope, aspect,
ship between host and parasite. poleward or equator-ward side of a disturbance patch, time
Epiphyte A plant that grows on the aboveground organs of of day and year, and the variables being considered.
other plant species. Generally a commensal relationship in Fragmentation A term used in conservation biology to
which the epiphyte is benefited but there is little or only refer to the process that converts large areas of relatively
minor advantage or disadvantage to the host. Epiphytes can uniform vegetation into a mosaic of small patches of vegeta-
also have either symbiotic or antagonistic relationships with tion of different age class and wildlife habitat potential
their host but are not parasitic on the host. across the landscape. Where old-growth forest is frag-
Eutrophic Refers to rich nutrient conditions; fertile soils. mented, it may not sustain certain old-growth—dependent
G-8 GLOSSARY

species even if a significant portion of the landscape still has ing to the speed of movement of the melt water. Usually
old growth on it. Such animals or plants may require large coarse-textured (sand) or having a high content of well-
patches of mature or old-growth forest for their survival. rounded coarse fragments (gravel).
Frost heaving The lifting of the surface layer of mineral Gleysol A soil that develops under the influence of a fluc-
soil by the alternating formation and thawing of “needle” tuating water table, with much of the soil profile saturated
ice. The raised soil particles can return to their original posi- for part of the year. The zone of water fluctuation (the
tion when the ice melts, but plants lifted with the soil remain gleyed layer) is characterized by yellow or brown mottles in
elevated and may eventually be “heaved” right out of the soil. a gray matrix. The zone above the gleyed layer may be pod-
zolic, brunisolic, or luvisolic.
Frost pockets “Yopographic depressions that fill with cold
air and that may have below freezing air temperatures when Gradient analysis Study of the distribution of organisms
the surrounding air is above 0. along an environmental gradient.
Functional predator response ‘(he change in the number Gravitational water Water contained in the larger pores
of prey eaten per predator per day as prey density changes. of a saturated soil that can be removed by the force of grav-
ity over a 2- to 3-day period (or longer in the case of fine-
Fundamental niche he geographical range and habitat
textured soils). It is the difference between the water content
that a species can occupy and the ecological (functional) role
of a soil at saturation and at field capacity.
that it can fulfill in the ecosystem (i.e., its ecological niche)
as determined by its genetically determined tolerances and Greenhouse effect ‘The trapping of long wavelength radiant
requirements. The niche that a species can occupy in the ab- energy, emitted from the sun and the earth’s surface, by at-
sence of competition or other antagonistic interactions with mospheric “greenhouse gases” such as water vapor, CO), and
other species. methane. These gases have an effect similar to the glass in a
greenhouse, hence the name. Life on Earth depends on this
Gadgil effect he restriction on the abundance and activi-
effect, but an increased greenhouse effect caused by air pollu-
ties of bacterial decomposers in the soil caused by the pres-
tion could cause very undesirable global climate changes.
ence of mycorrhizal fungi that exert an antibacterial effect.
Ground fire Largely flameless fires that burn slowly
Gause’s hypothesis “That no two organisms that have the
through surface accumulations of organic matter (i.e., the
same ecological niche requirements can coexist indefinitely
forest floor).
in a stable but resource-limited environment. Also known as
the competitive exclusion principle. Ground frost Radiant cooling of the soil surface and the
air in contact with it to below 0°C when the general air body
Genome ‘The genetic constitution, or the entire gene pool,
is above the freezing point.
of a species.
Growth form ‘The gross morphology, or the overall struc-
Geochemical cycle ‘Vhe input and output of nutrients to
ture and stature, of a plant; plant physiognomy.
and from an ecosystem. Precipitation, CO, uptake by photo-
synthesis, slope seepage, mineral weathering, biological ni-
trogen fixation, and fertilization are examples of inputs. Habitat type A physical environment that will support a par-
Harvest removals, respiratory loss of carbon, losses in fire ticular climax plant association in the absence of disturbance.
and erosion, leaching loss, and gaseous losses of nitrogen Harvest index ‘The proportion of the total biomass of a
(e.g., denitrification) are examples of outputs. crop plant that is harvested.
Geometric growth Growth of an individual or population Heat sum A measure of the amount of heat to which an or-
in which the percentage increase remains constant, resulting ganism has been exposed. Often expressed as degree days.
in an increasing absolute growth rate with time. Same as ex- The annual or growing season sum of the daily difference
potential growth. between mean daily temperature and some threshold tem-
Geotropism ‘The orientation of plants and their organs in perature.
response to the pull of gravity. Herb A plant that lacks a permanent aboveground woody
Glacial till Unsorted soil parent material transported stem. It can be either an annual or a perennial.
by glaciers. Characterized by an unsorted mixture of size Herbivore A plant-eating animal. A primary consumer in a
classes, generally from clay to bolders, and a mixture of grazing food chain.
rounded and sharp-angled coarse fragments.
Heterotherm An organism that is normally a homeotherm
Glaciofluvial soil Soil developed in parent material trans- but that will periodically allow its internal temperature to
ported and deposited by fast-flowing glacial meltwaters. approach that of, or equilibrate with, the temperature of its
Characterized by being well sorted into size classes accord- immediate environment to conserve energy.
 GLOSSARY G-9

Heterotropbs Organisms that depend on biotic energy Interception loss Rain or snow intercepted by and held in
sources: the energy of complex organic molecules created by the vegetation canopy, which is lost back to the atmosphere
other organisms. They are unable to utilize the energy of by evaporation or sublimation before it can reach the soil.
sunlight or inorganic chemicals.
Intrinsic population regulation Regulation of population size
Homeotherm An animal that maintains a relatively con- caused by processes operating within the population itself, such
stant internal temperature by means of physiological as competition between individuals, behavior, and genetics.
processes; a “warm-blooded” animal. Homeotherms gener-
Island biogeography theory A theory that relates to habitat
ally employ an insulating outer layer such as fur, feathers, or
fat deposits in the skin. fragmentation. Based on the observation that species richness
on oceanic islands varies as a funciotn of island size and distance
Hydraulic conductivity Ability of a soil to conduct water; a to the mainland. The theory has application for patches of for-
function of the texture and structure of the soil and the re- est permanently isolated by “hard ecological edges” such as
sultant pore size distribution. agricultural or urban land. It frequently requires modification
Hydrophobicity Water repelancy. Can develop in the soil for patches of older forest temporarily surrounded by harvested
as a result of drying or fire. areas. As these areas regrow, the ecological edges soften and the
theory becomes less relevant.
Hydrophyte Plant adapted to saturated soils or to submer-
gence in water for part or all of the year. k-strategist A species whose reproductive strategy is to
produce fewer, larger seeds or offspring, each of which has a
Hydrosere ‘The sequence of seral stages that develop by relatively high probability of survival in comparison with an
primary successional process, both allogenic and autogenic, r-strategist.
in an initially hydric environment (e.g., a shallow lake, bog).
Krummbolz Stunted trees with a shrub-like growth form
Hygrotope ‘The relative moisture status of a site; soil mois- that grow at or near the altitudinal tree line. Caused by fac-
ture regime. tors such as wind pruning, desiccation, low temperatures,
snow damage, and very short growing seasons.
Indicator species A species that has a sufficiently consistent
association with some environmental condition or other Lacustrine soil Soil developed in parent material trans-
species that its presence can be used to indicate or predict ported by water and laid down in a lake. Characterized by a
that environmental condition or the potential for that other fine texture (silt, clay, or fine sand), well sorted, and a lack of
species. coarse fragments. Often there are distinct layers or varves in
lacustrine parent material, which may lack structure and be
Inductive reasoning ‘The logical process of developing
poorly drained.
generalizations from the observation of many specific exam-
ples of a phenomenon. Lamas shoots Shoots formed when summer-formed buds
on the leaders of trees with a determinate growth pattern fail
Infiltration Penetration of water into the surface of the soil.
to enter full dormancy and resume growth in the late sum-
Ingestad ratio The ratio of the concentration of a mer or early fall.
macronutrient to the concentration of nitrogen in plants
Landscape An area of sufficient size that it includes the full
that are under optimum nutrition. For seedlings of sev-
range of natural disturbance patches undergoing characteris-
eral tree species, this has been reported to be as follows:
tic sequences of autogenic successional recovery stages, so
nitrogen = 100, phosphorus = 13, potassium = 65, calcium =
that the area is a shifting mosaic of changing local ecosystems,
6, magnesium = 8.5.
the overall landscape character which is relatively constant.
Inhibition pathway A successional pathway in which an For many distrubance-driven forests, this requires a very large
early seral stage “captures” the site (i.e., controls the site re- area. Generally we deal with local landscapes, which are sub-
sources) and prevents invasion by species of the subsequent sets of true ecological landscapes, and consequently their
seral stage. Such arrested succession in which an early seral overall characteristics normally change over time.
stage becomes stable and self-replacing is called a ser-climax. Lapse rate The rate at which air temperature drops as you
Initial floristic composition pathway A successional path- go up in elevation.
way in which the pattern of seral stages is determined by the Leaf area index ‘The surface area of leaves in a plant com-
particular mixture of plant species propogules that arrive or munity per square meter of ground surface occupied by that
are already present in an ecosystem after disturbance. The community. May refer to the total leaf area, including all sides
later successional species do not require environmental al- of the leaf, or to the horizontally projected area of the leaf.
teration by the early successional species. The pattern of
seral stages can be variable and is often not easily pre- Life-form spectrum The relative abundance of the differ-
dictable. ent life forms in a biotic community.
G-10 GLOSSARY

Life form of vegetation The physiognomy of the vegeta- Mesophyte Plant adapted to moderate water supplies,
tion; its overall stature and gross morphology, e.g., herbs, without extremes of water availability.
grasses, shrubs, trees, climbers, mosses. Mesosere ‘The sequence of seral stages in a mesic environ-
Life zone The physical environment of a biome. The char- ment, from the pioneer to the climax stage, after a distur-
acteristic regional climate and the mosaic of soil types that bance that removes the existing biotic community and the
reflect the interaction of that climate with local topography modification of the physical and chemical environments that
and soil parent materials. it has caused.
Light compensation point ‘The intensity of photosyntheti- Mesotrophic Refers to medium nutrient conditions; mod-
cally active radiation (PAR) at which carbon and energy erate soil fertility levels.
losses as a result of respiration exactly balance carbon and
Microclimate The climate near the ground (e.g., within
energy gains by photosynthesis.
1 m), or the climate condition experienced by an individ-
Light saturation point ‘The intensity of PAR above which ual organism.
the rate of CO, fixation by photosynthesis per unit leaf area
does not increase as the intensity of PAR continues to increase. Micronutrient A chemical element that is essential, at
concentrations in the parts per million (ppm) range, for the
Little ice age ‘The period from the 1500s to the early normal growth and physiology of an organism.
1800s during which there was a distinct cooling of the
earth's climate. Glaciers advanced, snow fields expanded, Mixedwoods Mixed species forests in which the spatial ar-
and many temperature-dependent ecological processes ragement of groups of trees of the same species is such that
were altered more than 50% of the area is within ecotones between two
Local landscape See Landscape. or more tree species. Where the different species are aggre-
gated into single species groups such that more than 50% of
Loess Soil parent material transported to the site by wind. the area is outside the ecotones, one has a mosaic of mono-
Loess has a very uniform and fine texture, usually silty, loamy, cultures, not a mixedwood.
or fine sand. Soils developed in loess deposits are often fertile.
Model An abstract or a physical entity that represents in
Logistic population growth Population growth that fol-
some way the form and/or the function of real-world entities
lows an S, or sigmoid, pattern over time. There is an initial
and processes (e.g., a map, a book about forest ecosystems, a
period of exponential growth when resources are unlimited
computer model) of the forest ecosystem.
and the mortality rate is low, followed by a progressive slow-
ing of the growth rate as the population approaches K (the Moder A forest floor type in which there are distinct litter
carrying capacity) asymptotically. (L) and decomposing (F) layers and a humus (H) layer that
Long-day organism An organism that uses day lengths grades into and is partly mixed with the underlying mineral
longer than 14 to 16 hours to trigger some aspect oftheir life soil (i.e., there is an Ah layer). Unlike a mor forest floor, the
history. A photoperiodic adaptation. In plants, the response F layer is loose and friable, with little matting caused by fine
is actually based on the length of uninterrupted dark periods. roots and fungal mycelia, and it often has abundant soil ani-
mal activity. The Ah layer is less than 10 cm thick.
Luvisol A forest soil characteristic of coniferous forests in
cool to cold, humid climates and fine-textured (high clay Monoclimax theory ‘This theory of ecological succession
content) soil parent material. Characterized by a clay-en- says that, given sufficient time in the absence of disturbance
riched B horizon (Bt) that may be root restricting and has and irrespective of initial soil moisture and nutrient status,
reduced hydraulic conductivity (poorly drained). all sites in a climatic region will be altered by the processes
of autogenic succession to converge eventually to a single,
Macronutrient A chemical element that is essential, at
medium condition characterized by a self-replacing, rela-
concentrations in the parts per hundred (i.e., percent) range, tively stable climax community, the characteristics and
for the normal growth and physiology of an organism.
species composition of which reflect the regional climate.
Matric potential ‘The force with which soil holds water This successional convergence is the central idea of the
against the force of gravity and the uptake of water by monoclimax theory.
plant roots.
Mor A forest floor type in which there are distinct litter
Mesic site In the biogeoclimatic ecosystem classification (L), decomposing (F), and humus (H) layers and a sharp
system, a mesic site is a site on which the moisture experi- transition between the organic forest floor and the underly-
ences of the vegetation are determined mainly by the re- ing mineral soil. Dominated by fungi and having little or no
gional climate, rather than by soil and topographic factors. animal mixing, the F layer is densely matted with fine roots
In more general use, a mesic site is one with a moist or fresh and fungal mycelium. Generally acidic. Characteristic of
soil moisture regime. sites with slow decomposition of litter and low fertility.
 GLOSSARY G-11

Mortality Vhe population-reducing process of death, or Niche ‘The geographical range and habitat that a species can
the rate of death. The percentage of a population dying in a or does occupy and the ecological (functional) role that it can
particular period of time. or does fulfill in an ecosystem. The functional, adaptational,
and distributional characteristics of a species. A species has a
Mull A forest floor type in which there is a thin layer of
genetically controlled fundamental niche, but it generally oc-
fresh or recent litter (the L layer), virtually no F layer, and an
cupies only a subset of this (the realized niche) because of
Ah layer (a surface mineral soil horizon enriched with or-
competition and other antagonistic community interactions.
ganic matter) that grades slowly into the underlying mineral
soil. The Ah layer is more than 10 cm thick. Associated with Nitrification The soil process in which ammonium
productive, moist, and fertile sites and abundant soil animal (NH,*) ions are converted by microbes into nitrate ions
activity that mixes the organic forest floor material with the (NO,). This oxygen-requiring process is characteristic of
upper mineral soil (zoonenous Ah; rhizogenous Ah horizons moist, fertile soils and of the postdisturbance assart period.
can be formed by rapid turnover of fine roots in the upper The ammonium ions are produced from organic nitrogen
mineral soil). Also characteristic of grassland soils. compounds by the microbial process of ammonification.
Mutualism A relationship between two species in which Nitrogen fixation ‘The combining of the chemically inert
both are benefited. and unreactive atmospheric gas N) with atoms of hydrogen.
Biological nitrogen fixation is accomplished by certain mi-
Mycorrhiza A mutualistic relationship between a soil fun-
croorganisms that can reduce N, and combine it into or-
gus and a plant root in which the fungus obtains a supply of
ganic molecules such as amino acids and proteins.
high-energy organic molecules from the plant, and the plant’s
ability to get nutrients, especially phosphorus, and water from Nonsoil Unconsolidated materials at the surface of the
the soil are increased. The mycorrhizal relationship may also earth that do not satisfy the definition of soil.
confer on plant roots increased resistance to pathogens.
Numerical predator response “Vhe change in the number
Mycotrophy A nutritional adaptation in plants in which nu- of predators per unit area as prey density changes.
trient uptake from the soil is achieved largely by a mutualis-
Nutrient A chemical that is required for the growth and
tic association between plant roots and soil fungi (a
physiology of an organism.
mycorrhiza).
Old growth ‘This is one of the least well-defined terms in
Natality The population-increasing process of producing forest ecology! It may refer to a forest of very large trees, or
new individuals, or the rate of “birth.” The rate at which new of very old trees, or a forest that has reached its climax seral
offspring are produced per female or reproducing individual. stage, or simply one that satisfies criteria that are said to de-
fine the old-growth condition. Old-growth is also a phase of
Natural rate of increase (r) ‘The sum of additions to stand development that can occur towards the end of any
(births) and losses from (mortality) a population. It is the seral stage. Climax old-growth forest is not a type of forest
rate of population growth per capita. ecosystem but rather a condition that a forest ecosystem can
Needle ice Ice crystals formed in the surface of moist min- attain if sufficient time passes since the last disturbance.
eral soil when radiant cooling of that surface layer lowers Seral old-growth is a much more temporary condition than
the temperature below 0°C while the lower layers of soil are climax old-growth.
still above 0°C. Oligotrophic Refers to low nutrient conditions; low soil
fertility levels.
“New” forestry A philosophy of forest management devel-
oped in Oregon, in which a form of ecosystem management Ombrotrophic A nutrient-poor ecosystem condition in
replaces timber management. There is a switch from the which the source of nutrient input to the ecosystem is lim-
more traditional focus on soils as the major determinant of ited to precipitation and other atmospheric inputs. There
ecosystem sustainability to a more biocentric view; sustain- are no inputs from seepage or mineral weathering.
ability is related to biodiversity, coarse woody debris, and the
maintenance of “ecosystem structures” that provide habitat Omnivore An animal that acts as both an herbivore and a car-
for late-seral or old-growth—dependent. wildlife species. nivore, obtaining its food energy from both plants and animals.
Early seral conditions seem to be considered inferior to late Ordination A vegetation analysis technique that analyzes
seral or climax conditions (i.e., old growth). True ecosystem the distribution (the ordering) of plant species along envi-
management gives equal consideration to the physical and ronmental gradients.
biotic components of the ecosystem and their integration so
that the full natural range of variation in biotic and physical Osmotic potential ‘The suction pressure created by the
components and processes of the regional ecosystem is presence of dissolved chemicals in soil solution or the tissues
maintained across the landscape. of plants or animals.
G-12. GLOSSARY

PAR Photosynthetically active radiation, That portion of intercepts light and converts it into net production. Produc-
the visible wavelengths of the electromagnetic spectrum that tion/incident solar radiation x100.
is absorbed by photosynthetic pigments. Physiognomy The overall gross morphology of a plant.
Parasitism An antagonistic interaction between nwo species Phytochrome A light-absorbing pigment that is converted
in which one feeds on the other without necessarily killing it. into a biologically inactive form by exposure to far-red wave-
Parasites are normally much smaller than their prey. length radiation (i.e, to darkness) and to active form by ex-
Parent material See Soil parent material. posure to the red wavelengths of light. Important in
photoperiodism in plants.
P/E ratio The ratio of precipitation to the evaporative po-
tential of the air. A measure of the ecological effectiveness of Pioneer The seral stage at the beginning of a succession.
precipitation. Maps of P/E ratios often correlate well with Generally applies to a primary succession but may also refer
maps of broad vegetation types. to the first seral stage of a secondary succession.
Ped A soil aggregate; a group of mineral soil particles Plant association A plant community with a definite
bound together by colloidal clay, organic matter, or chemical species composition and characteristic physiognomy, grow-
bonding. Porous peds are called crumbs; nonporous peds are ing in particular, homogeneous habitat conditions. An asso-
called granules. ciation is a characteristic list of species of high fidelity. Plant
associations may be the result of species adapting to the
Percentage base saturation Sum of the exchangeable
presence of other members of the association or may repre-
cations, other than Al** and H”, held on the CEC, expressed
sent the distribution along an environmental gradient of a
as a percentage of the total CEC.
group of species that share common adaptations to that
Permafrost A soil condition in which most of the soil pro- physical gradient.
file is permanently frozen throughout most of its depth.
Plant formation A division of the vegetation of a conti-
Permanent wilting point The level of soil water availabil- nent based on the life form or physiognomy of the dominant
ity below which a plant is unable to recover its turgor after vegetation of the climax or late seral stage. The plant com-
the soil is rewetted. ponent of a biome,
pH Ameasure of acidity, or the activity of hydrogen ions, It is Podsol A forest soil characteristic of coniferous forests
the logarithm of the reciprocal of hydrogen ion concentration, growing in cool, humid climates on well-drained parent ma-
Phase diagram Used in population ecology to show the terial of medium to low nutrient status and medium to
relauonship between prey and predator/parasite popula- coarse texture, Characterized by an iron-enriched or iron-
tions, especially those that show regular oscillations or cy- and humus-enriched B horizon (Bf or Bhf) and a mor or
cles caused by the effect of the predator/parasite on the prey moder humus form. Acidic, commonly with a pale eluviated
population. Ae horizon.

Phenotype The behavioral, physiological, and morpholog- Poikilotherm An organism whose internal temperature is
ical characteristics of an organism determined by the influ- controlled by the temperature and radiation balance of its
ence of environmental factors on the genetically determined environment; “cold-blooded” animals, and plants and mi-
potential of that organism. crobes. Poikilotherm animals may maintain fairly constant
internal temperatures by moving between different temper-
Photoautotroph An organism that is able to obtain the ener- ature and radiation environments.
gy that it needs for growth, reproduction, and life processes
trom solar radiation that it captures with pigments. Most pho- Polyclimax theory A successional theory that says that ei-
toautotrophs are green plants. Autotroph = “self-feeding.” ther frequent ecosystem disturbance, by fire, wind, disease or
animals, or environmental conditions that cannot be modified
Photokinesis he relationship between the speed of move- by autogenic succession may prevent successional conver-
ment and general activity of animals to the intensity of light.
gence from occurring. Where the frequency and intensity of
Photon flux density A measure of the intensity of photo- disturbance are high relative to the rate of autogenic succes-
synthetically active radiation (PAR). sion, the landscape will be a mozaic of recently disturbed
stands and fire climax, edaphic climax, and biotic climax com-
Photoperiodism The sensitivity of plants and animals to
munities rather than converging to a common climatic climax.
variation in the relative lengths of day and night over the an-
nual cycle of seasons. Plants are mainly sensitive to the Population group of genetically related individuals.
length of the night.
Population cycle Changes in population size that have a
Photosynthetic efficiency he etticiency with which an in- relatively constant period of fluctuation and occur around a
dividual leaf, an entire plant, or an entire plant community relatively constant long-term mean density.
 GLOSSARY G-13

Porosity (soil) The proportion of the soil volume that is Realized niche The geographical area and habitat that a
occupied by air spaces or pores. species can occupy and the functional role that it can play in
Predation An antagonistic interaction between two species an ecosystem, in face of competition and other antagonistic
in which one species consumes much or all of another and interactions from other species.
normally kills it. Predators are often larger than their prey. Regulating factor A factor that determines the size and
Prey escape density ‘The prey density above which the per- distribution of a population that operates in a density-
centage of predation decreases. At prey densities above es- dependent manner, e.g., predation, disease, competition.
cape density, predation becomes less and less effective as a Most important in physically moderate environments.
population regulation mechanism. Predation is no longer a Relay floristics pathway A pathway of primary succession
density-dependent population control mechanism above es- in which the early seral communities alter the soil and mi-
cape density. croclimate in a way that facilitates the invasion and growth of
Primary succession Successional development of an subsequent seral communities. The early stages of this path-
ecosystem beginning after a disturbance that has removed all way are typically predictable and invariable. Midseral stages
of the modifications to microclimate and the geological sub- generally require the previous occupancy of the site by the
strate produced by the previous succession. Succession on pioneer seral stages before they can become established.
bare rock, in shallow lakes, or on parent soil materials. Renewable resource A resource that can be restored or re-
stores itself to the point at which it can be reused or rehar-
Production ecology of plants ‘The study of the system of
vested after a period of time that is within our current
ecological and physiological processes that determine leaf
economic or social planning time scale or that is renewed at
area, photosynthetic efficiency, total and net photosynthesis,
a rate that renders investment in its renewal economically or
allocation of photosynthate to different plant organs, the
socially attractive.
loss of biomass to litterfall and mortality, and the ratio be-
tween harvestable and nonharvestable biomass. Replacement fertility Level of fertility at which each
breeding unit produces sufficient offspring to replace itself.
Provenance ‘The geographical location and its physical
features where a particular genotype evolved. Generally Resilience See Elasticity.
used to describe the location of plant seed collection and de- Retrogression See Successional retrogression.
scribed by latitude, longitude, elevation, aspect, and climate.
Rhizome A perennial underground stem.
Pyral climax A climax community or seral stage, the charac-
ter of which is determined by periodic fire. A fire frequency of Rhizosphere ‘The zone of soil around a root in which the
once every 300 or 400 years may be sufficient to maintain a abundance, composition, and activity of the microbial com-
mid- or subclimax seral stage as a pyral climax. More frequent munity are influenced by the presence of the root.
fires can maintain earlier seral stages as a pyral climax. Rime A layer of ice crystals that grow on plant surfaces
Q,9 value ‘The increase in the rate of a biological process
when they cool to below 0°C by radiant cooling but are sur-
for a 10°C increase in temperature. Normally refers to the rounded by humid air above 0°C.
range of temperatures defined by the cardinal temperatures. Saprotroph An organism that obtains its energy by feeding
A value of the 2.0 suggests a doubling of the rate; 1:2 sug- on dead organic matter; a decomposer or detritus feeder.
gests a 20% increase. Seed bank The assemblage of viable but ungerminated seeds
r The natural rate of increase of apopulation; expressed as present in the soil or held in closed cones in the canopy that
the rate of increase in the population per unit time per mem- will germinate if the present vegetation is killed or removed.
ber of the population. Seed rain The annual “rain” ofseeds falling on an area as a
r-strategist A species whose reproductive strategy is to result of dispersal by wind or animals or simply released by
produce large numbers of small seeds, spores, or offspring, the vegetation on the site.
each of which has a very low probability of survival in com- Seedling bank ‘The population of suppressed seedlings pre-
parison with a k-strategist. sent in the understory. They will be released from suppression
Radiation frost Cooling of the soil surface by radiant if the overstory is thinned or removed by natural disturbance
emission on a clear night causing the temperature of the soil or forest management. Called advance regeneration by foresters.
surface and the air in contact with it to fall below 0°C. Selection harvest system A silvicultural system used in
Rain forest Forest that grows in climates that lack a major uneven-aged forests. Periodic (e.g., every 5 to 20 years) har-
growing season moisture deficit. Often defined as forests vests of trees of various sizes are done to yield timber and to
that receive >2000 mm annual precipitation evenly distrib- maintain a desired age and size class structure and species
uted over the year. mixture in the forest.
G-14 GLOSSARY

Selective harvest system Generally a system in which peri- Shelterwood system An even-age silvicultural system that
odic (e.g., every 30 to 50 years) harvests are made of all com- involves a series of heavy thinnings near the end of the rota-
mercial species down to some minimum diameter limit. tion to promote natural regeneration while maintaining a
Often called diameter-limit cutting. Frequently leads to modified microclimate and soil condition during the period
“high grading,” forest exploitation, and forest degradation. of regeneration establishment. The remaining overstory
Not considered to be an acceptable harvest system in sus- trees are removed when the regeneration is well established.
tainable forest management in many types of forest. Short-day organism An organism that uses short
Ser-climax A case of arrested succession whereby the days/long nights as an environmental signal to reproduce,
plant community of an early seral stage “captures” the site prepare for or enter dormancy, or complete some other as-
by developing sufficient root, rhizome and/or foliage mass pect of their life cycle (e.g., fall or spring flowering plants).
to monopolize soil moisture and nutrients and light. Early Characteristic of low and middle latitudes.
seral shrub or herb communities that develop unimpeded SMR _ Soil moisture regime. The moisture status of a site as
because of an initial lack of invasion by later seral species a result of topographic and soil factors, relative to the mois-
may remain stable and resist subsequent invasion by these ture availability expected on the basis of the local climate.
species for many decades or even centuries.
SNR_ Soil nutrient regime. The nutrient status of the site,
Seral stage ‘The identifiable stages in the development of a from lowest to highest fertility levels.
sere, from an early pioneer stage, through various early and
midseral stages, to late seral, subclimax, and climax stages. Soil The unconsolidated inorganic or organic surface of
The stages are identified by different plant associations (dif- the earth that is produced by physical, chemical, and/or bio-
ferent species composition and/or community structure); by logical processes that act on soil parent material over time
different ages of the dominant vegetation (usually related to and is capable of supporting plants. Soil provides anchorage,
differences in structure); and by different microclimatic, soil moisture, nutrients, and oxygen for roots and soil organisms.
and forest conditions. Soil parent material ‘Vhe unconsolidated inorganic mate-
Sere The sequence of biotic communities and associated rial from which mineral soil is formed by the action of phys-
nonliving environmental conditions that successively replace ical, chemical, and biological processes acting over time.
each other on a given area of land, from the time of a distur- Soil texture ‘he relative proportions of clay, silt, and sand
bance that has removed or disturbed the previous com- size particles in the fine fraction of the soil (i.e., the <2-mm-
munity and nonliving environmental conditions, to a final, diameter mineral soil particles).
relatively stable and self-replacing community (the climax).
The sere is divided into a series of seral stages. Species-area curve A graphical representation of the rate
of change in number of species in a sample plot as the size of
Serotiny Storage of seed in woody cones by trees or shrubs the plot is increased.
that remain closed until heated by a fire or by being exposed
on a hot ground surface. Serotiny ensures that abundant Species diversity One of the measures of biological diver-
seed is released when fire has reduced competition and cre- sity in forest ecosystems. The number of species in the
ated a suitable seedbed. ecosystem (species richness) or one of several indices that re-
flect the relative commonness of different species (species
Shade leaf A leaf produced by a shade-tolerant plant evenness).
growing under low light intensity. It has a low weight/sur-
face area ratio (a high specific leaf area), thin cuticle and Stability The tendency of a system to remain in its present
pallisade layer, and low compensation and saturation light condition, to persist in the face of disturbance, or to return
intensities. to its original condition after disturbance. There are many
different measures of stability.
Shade tolerance The ability of a plant to germinate, estab-
lish, survive, compete for resources, and grow in the shade of Stable age distribution ‘The distribution of the individuals
other plants. A complex characteristic of plants involving in a population between different age classes so that the age
seed size, physiological and morphologicat adaptations to class distribution remains constant over time. The distribu-
low light intensity, root-shoot ratios, height-diameter ratios, tion is different from the shape of the survivorship curve of
disease resistance, and the ability to compete for soil re- that species. Births do not equal deaths, and the population
sources at low light levels. Shade tolerance of trees often is either declining or expanding, but generally the latter.
varies with their age. Shade-tolerant tree species often grow Stand cycle ‘Vhe variation in biomass, species composition,
best in full sunlight or very light shade. Survival in the shade tree heights, stand density (stems per hectare), canopy lay-
may also reflect the ability to resist tipping or breakage by ers, and soil conditions from the time of timber harvest until
wind or snow when shaded plants have high height-diameter the subsequent timber crop is ready for harvest. Normally
and shoot-root ratios. applies to even-aged forests. The stand cycle is generally
 GLOSSARY G-15

shorter than the time scale of the four phases of stand dy- that the entire landscape will eventually support a single cli-
namics. matic climax biotic community. Only occurs where the rate
Stand dynamics The four stages in the development of a of autogenic succession is high relative to the frequency and
tree population: stand initiation, stem exclusion, understory severity of stand-replacing disturbance (same as ecological
reinitiation, old growth. [successional] convergence).

Stationary age distribution The distribution of the indi- Successional retrogression ‘The return of an ecosystem to
viduals in a population between different age classes so that an earlier seral stage or successional condition as a result of a
the proportion of the population in each age class remains disturbance that removes or modifies the existing com-
constant and in which births (natality) equal deaths (mortal- munity and creates soil and microclimate conditions charac-
ity) so that the population size remains constant. The shape teristic of that earlier seral stage.
of the age class distribution is the same as the shape of the Sunfleck A small patch of sunlight on the forest floor
“normal” survivorship curve of that species. where solar radiation has passed through a small gap in the
Stemflow ‘he concentration of precipitation by leaves and canopy unaltered in wavelength composition and largely
branches into flow pathways down the stem. Precipitation undiminished in intensity.
that reaches the soil via a plant stem. Sunscald Vhe killing of bark or other plant tissues (e.g.,
Structural diversity One of the measures of biological di- leaves) that had previously been shaded after their sudden
versity in forest ecosystems. It refers to the variation in tree exposure to full sunlight.
size and canopy layering, the variety of different life forms of Surface fire Fires with generally low intensity and severity
vegetation (trees, herbs, shrubs, mosses, climbers, epiphytes, and rapid rate of spread that burn litter and dead herbs and
etc.), and the relative size and abundance of standing dead shrubs. Surface fires generally do little damage to live trees
trees (snags) and decaying logs on the ground (coarse woody and soil where they occur frequently and there has been lit-
debris). Structural diversity refers to these features within a tle accumulation of fuels.
particular local ecosystem (alpha structural diversity) or to
variations in them between local ecosystems across the local Survival temperatures “Vhe maximum and minimum tem-
landscape (beta structural diversity). peratures that an organism or an organ can survive undam-
aged. Definition of survival temperatures includes both the
Structure of the plant community ‘The vertical arrange- temperature and the duration of exposure to that tempera-
ment of canopy layers and plants of different life form or the ture. Survival temperatures vary with age of the oganism and
horizontal variation in canopy closure and canopy layers, or time of year.
both. Community structure also includes standing dead
trees (snags) and decomposing logs on the forest floor Survivorship curves A graphical representation of the per-
(coarse woody debris). centage survival of an even-aged group of organisms (a co-
hort) over time from birth (animals) or germination (plants)
Structure of the soil Arrangement of the mineral and or- to the maximum physiological life span of the species.
ganic components of the soil to form a soil architecture con-
sisting of soil aggregates and pores of various sizes that Sustainable forestry Forest management that sustains a
facilitate soil water movement, gaseous exchanges with the continuous or periodic supply of a desired range of products,
atmosphere, root penetration, and the activities of soil or- services, or forest conditions from the area of forest in ques-
ganisms. tion. Based on the concept of nondeclining, long-term mean
values rather than constant values. Sustainable forestry on a
Subclimax ‘he seral stage preceding the climatic climax small area often implies uneven-age management. On a larger
seral stage. area, it can be either even-age or uneven-age management.
Succession See Ecological succession. Sustained yield A forest management term that refers to
Successional acceleration An increase in the rate of auto- the sustained supply of timber or some other forest product
genic processes and autogenic succession to above “normal” over time from a forest management unit. It does not imply
rates as a result of disturbance processes. constant yield or constant ecological conditions in any one
stand. It is based on the concept of fluctuations in yields and
Successional convergence ‘The alteration of physical and conditions around a nondeclining long-term mean. Sus-
chemical characteristics of the landscape by the process of tained yield from a small area of forest often implies uneven-
autogenic succession that reduces extremes and moves the age management. From a larger area, it can be either
initially different ecosystems toward a medium or average even-age or uneven-age management.
condition. The process by which wet sites become drier, dry
sites become wetter, poor sites become richer, rich sites be- Symbiosis A relationship between two species in which
come less rich, and microclimate extremes are moderated so neither species is harmed and at least one is benefited.
G-16 GLOSSARY

Temporal biodiversity ‘he change over time in measures Unstable age distribution An age class distribution in
of alpha species and structural diversity or in measures of which the percentage of the population in each age class is
landscape (beta) biodiversity as a result of disturbance and changing over time. The population is expanding or declin-
autogenic succession. ing but not consistently. The postwar “baby boom” has
given the human population of many countries an unstable
Territoriality An aspect of animal behavior in which an in- age distribution.
dividual will defend a territory to ensure adequate access to
resources. An important mechanism of population control in Utilization efficiency The efficiency with which organ-
some animal species. isms in one trophic level utilize the energy in the previous
trophic level. Equal to the ratio of ingestion to the net pro-
Thermal regulation Regulation of internal temperatures duction of the previous level.
by either physiological (homeotherms) or behavioral (poik-
Vapor pressure deficit The difference between saturation
ilotherms) mechanisms to keep them within the species’ car-
vapor pressure and the actual vapor pressure at any particu-
dinal temperature range.
lar temperature. The “drying power” of the air.
Thermoperiodism ‘The sensitivity of organisms to daily
Variable retention forestry Space-for-time substitution in
and seasonal variations in temperature. Heat sums, chilling
which components and processes of older forests are main-
requirements, and sensitivity to day/night temperature dif-
tained in younger, managed forests by retaining patches of
ferences are three aspects of thermoperiodism. In concert
unharvested forest within a matrix of harvested forest. This
with photoperiodism, it provides organisms with both a
sustains within the same local landscape ecosystem values
daily clock and a seasonal calendar.
that vary in their ecological rotation.
Throughfall Precipitation that either falls to the soil sur- View factor he proportion of the sky that an object or lo-
face directly through gaps in the canopy or drips from cation on the ground can “see.” The proportion of the radi-
branches and foliage. ation environment of an object that is accounted for by
Till soil Soil developed in parent material transported and unobstructed sky.
deposited by a glacier. Characterized by an unsorted mixture Water saver Plants that store water during times of abun-
of clay, silt, and sand particles; gravel, stones, and boulders; dance (e.g., succulents) or restrict the loss of water by mor-
and a mixture of rounded and sharp-edged stones and boul- phological or physiological adaptations.
ders.
Water spender Plants that maintain high rates of transpira-
Tissue growth efficiency ‘The ratio of net production to as- tion by ensuring an adequate supply of water to the leaves, usu-
similation in any consumer trophic level. ally by having high root-shoot ratios and deep root systems.
Tolerance pathway A successional pathway in which mid- Wilting point ‘The soil moisture potential at which plants
and late-successional species are reduced in growth by the close their stomata (temporary wilting point) or at which
presence ofearly seral species but are not prevented from in- plants will not recover turgor upon soil rewetting (perma-
vading and growing. In time, these later seral species will ex- nent wilting point).
clude the early seral species and succession will proceed.
The later seral species do not require environmental alter- Xerophyte Plant adapted to drought or low water availability.
ations by early seral species to enable them to invade. Xerosere ‘The sequence of seral stages in a xeric (dry) phys-
Tree line The altitudinal or high latitudinal limit of tree ical environment, from the pioneer to the climax stage, after
growth. a disturbance that removes the existing biotic community
and the modification of the physical and chemical environ-
Trophic efficiency The efficiency with which energy is
ments that it had caused.
transferred between trophic levels.
Zonal soil ‘The type of soil that, given sufficient time, will
Trophotope ‘The soil fertility status of a site; soil nutri-
develop under a particular regional climatic regime and the
ent regime. :
vegetation that reflects that climate. Zonal soils may not de-
Tropical forest Forest that grows between latitudes 23°N velop or will take a very long time to develop in soil parent
and 23°S. Includes tropical rain forest, tropical seasonal for- materials that are at the extremes of soil physical and chemi-
est, and tropical dry forest. cal conditions.
 Scientific and Common
Names of Plants

Scientific Name Common Name Life Form

Abies alba Mill. Silver fir Tree, evergreen conifer

Abies amabilis (Dougl.) Forbes Pacific silver fir Tree, evergreen conifer
Abies balsamea (L.) Mill. Balsam fir Tree, evergreen conifer
Abies concolor (Gord, & Glend.) Lindl White fir Tree, evergreen conifer
Abies fraseri (Pursh.) Poir Fraser fir Tree, evergreen conifer
Abies lasiocarpa (Hook.) Nutt. Subalpine fir Tree, evergreen conifer
Abies magnifica A. Murr. California red fir Tree, evergreen conifer
Acacia cornigera Willd. Bull’s horn acacia Tree, evergreen broadleaf
Acer circinatum Pursh. Vine maple Tree, deciduous broadleaf
Acer glabrum Torr. Rocky Mountain maple Tree, deciduous broadleaf
Acer grandidentatum Nutt. Bigtooth maple, canyon maple Tree, deciduous broadleaf
Acer japonicum Thunb. Japanese maple Tree, deciduous broadleaf
Acer macrophyllum Pursh. Bigleaf maple Tree, deciduous broadleaf
Acer pseudoplatanus L. Sycamore Tree, deciduous broadleaf
Acer rubrum L. Red maple Tree, deciduous broadleaf
Acer saccharum Marsh. Sugar maple Tree, deciduous broadleaf
Achillea lanulosa (Nutt.) Piper Yarrow Herb, forb
Adenostoma fasciculatum Hook & Arn. Chamise Shrub, evergreen
Aesculus mollissima Chinese chestnut Tree, deciduous broadleaf

Allium chamaemoly L. Wild onion Herb, forb

Alnus rubra Bong. Red alder Tree, deciduous broadleaf
Alnus sinuata (Regel) Rydb. Sitka alder Tree, deciduous broadleaf

Ammophila arenaria (L.) Link. Marram grass, beachgrass Herb, grass

Araucaria spp. Jussieu Southern pine Tree, evergreen conifer

Arbutus arizonica (Gray) Sarg, Arizona madrone Tree, evergreen broadleaf

Aster canadensis O. Ktze Horseweed Herb, forb

Bellis annua L. Daisy Herb, forb

Betula nigra L. River Birch Tree, deciduous broadleaf

Calamagrostis rubescens Buck. Pinegrass Herb, grass

Callitris spp. Ventenat Cypress pine Tree, evergreen conifer
Calluna vulgaris Salisb. Heather Shrub, evergreen

AP-1
AP-2 APPENDIX A __ Scientific and Common Names of Plants

Scientific Name Common Name Life Form

Ceanothus spp. L. Ceanothus Shrub, deciduous/evergreen

Clematis L. Clematis Climbing vine
Cornus spp. L. Dogwood Tree, deciduous broadleaf
Cryptomeria japonica (L.f) Don Sugi Tree, evergreen conifer
Cupressus macrocarpa Hartw. Monterey cypress Tree, evergreen conifer
Dicksonia spp. Tree ferns Tree, evergreen
Dryas drummondii Richards Yellow mountain avens Creeping shrub
Encelia californica Nutt. Encelia Shrub
Epilobium angustifolium L. Firewood Herb, forb
Eucalyptus spp. Benth. & Hook. Mallee, eucalyptus Tree, evergreen broadleaf
Eucalyptus marginata Sm. Jarrah Tree, evergreen broadleaf
Fagus grandifolia Ehrh. American beech Tree, deciduous broadleaf
Fagus sylvatica L. European beech Tree, deciduous broadleaf
Festuca ovina L. Sheep fescue Herb, grass
Fraxinus excellsior Bove ex DC. European ash Tree, deciduous broadleaf
Gaultheria shalion Pursh, Salal Shrub, evergreen
Hedaera helix L. Ivy Climbing shrub, evergreen
Hyoseris scabra Pourr. Herb, forb
Hypericum perforatum L. Klamath weed, St. John’s Wort Herb, forb
Juglans nigra L. Black walnut Tree, deciduous broadleaf
Juniper deppeana Steud. Alligator juniper Shrub, evergreen
Kalmia angustifolium L. Sheep laurel Shrub, evergreen
Kalmia polifolia Wang Swamp laurel Shrub, evergreen
Lantana camara L. Lantana Shrub, evergreen
Larix laricina (Du Roi) K. Koch Tamarack eastern larch Tree, deciduous conifer
Larix leptolepis Hort. ex Endl. Japanese larch Tree, deciduous conifer
Larix occidentalis Nutt. Western larch Tree, deciduous conifer
Ledum groenlandicum Oeder Labrador tea Shrub, evergreen
Liquidambar styraciflua L. Sweetgum Tree, deciduous broadleaf
Liriodenaron tulipifera L. Yellow-poplar Tree, deciduous broadleaf
Malus diversifolia (Bong.) Roem = Malus Western crab apple Tree, deciduous broadleaf
fusca (Raf.) Schneid Oregon crab apple
Malus pumila (L.) Mill. = Malus Common crab apple Tree, deciduous broadleaf
sylvestris (L.) Mill. Common apple
Metrosideros polymorpha Gaud. Ohia Tree, evergreen broadleaf
Myrica gale L. Sweet gale Shrub, deciduous
Nothofagus spp. Blume Southern beech Tree, deciduous broadleaf
Nothofagus fusca Oerst. Red beech Tree, deciduous broadleaf
Opuntia spp. Mill. Prickly pear cactus Succulent thorny
Oryza sativa L. Rice Herb, grass
Persea americana Mill. Avocado, alligator-pear Tree, evergreen broadleaf
Physiocarpus malvaceus (Greene) Kuntze Mallow ninebark Shrub, deciduous
Picea abies (L.) Karst. Norway spruce Tree, evergreen conifer
Picea engelmannii Parry Engelmann spruce Tree, evergreen conifer
Picea glauca (Moench) Voss White spruce Tree, evergreen conifer
Picea mariana (Mill.) B.S.P. Black spruce Tree, evergreen conifer
Picea sitchensis (Bong.) Carr. Sitka spruce Tree, evergreen conifer
Pinus aristata Engelm. Bristlecone pine, foxtail pine Tree, evergreen conifer
Pinus ayacahnita Ehreub. Mexican white pine Tree, evergreen conifer
 APPENDIX A Scientific and Common Names ofPlants

Scientific Name Common Name Life Form

Pinus banksiana Lamb. Jack pine Tree, evergreen conifer

Pinus cembra L. Swiss stone pine Tree, evergreen conifer
Pinus cembroides Zucc. Mexican pinyon pine Tree, evergreen conifer
Pinus contorta Dougl. Lodgepole pine Tree, evergreen conifer
Pinus densiflora Sieb. & Zucc. Japanese red pine Tree, evergreen conifer
Pinus edulis Engelm. Pinyon pine Tree, evergreen conifer
Pinus elliottii Engelm. Slash pine Tree, evergreen conifer
Pinus lecophylla Schiede & Deppe Chihuahua pine Tree, evergreen conifer
Pinus monticola Doug. Western white pine Tree, evergreen conifer
Pinus nigra Arnold Corsican pine Tree, evergreen conifer
Pinus palustris Mill. Long leaf pine Tree, evergreen conifer
Pinus ponderosa Doug. Ponderosa pine Tree, evergreen conifer
Pinus radiata D. Don Monterey pine Tree, evergreen conifer
Pinus resinosa Ait, Red pine Tree, evergreen conifer
Pinus rigida Mill. Pitch pine Tree, evergreen conifer
Pinus strobus L. Eastern white pine Tree, evergreen conifer
Pinus sylvestris L. Scotch (Scots) pine Tree, evergreen conifer
Pinus taeda L. Loblolly pine Tree, evergreen conifer
Plagiomnium insigne Mitt. Coastal leafy moss Moss
Polygonum distortoides Pursh. Snakeweed Herb, forb
Polypodium polypochoides Polypody fern, licorice fern Herb, fern
Polytrichum juniperinum Hedw. Hair-cap moss Moss
Populus deltoides Bartr. Eastern cottonwood Tree, deciduous broadleaf
Populus nigra var italica Muench. Lombardy poplar Tree, deciduous broadleaf
Populus tremuloides Michx. Quaking aspen Tree, deciduous broadleaf
Populus trichocarpa Torr. and Gray Black cottonwood Tree, deciduous broadleaf
Prosopis farata L. Mesquite Shrub, tree, deciduous broadleaf
Prunus spp. L. Cherry Tree, deciduous broadleaf
Prunus serotina Ehrh. Black cherry Tree, deciduous broadleaf
Pseudotsuga menziesii (Mirb.) Franco. Douglas-fir Tree, evergreen conifer
Pteridium aquilinum (L.) Kuhn Bracken fern Herb, fern
Purshia tridentata (Pursh.) DC Bitter brush, Antelope-brush Shrub, evergreen
Quercus spp. L. Oak Tree, deciduous broadleaf
Quercus arizonica Sarg. Arizona white oak Tree, deciduous broadleaf
Quercus borealis (Marsh.) = Quercus rubra L. Northern red oak Tree, deciduous broadleaf

Quercus emoryi Torr. Emory oak Tree, deciduous broadleaf

Quercus gambelii Nutt. Gambel oak Tree, deciduous broadleaf

Quercus hypoleucoides A. Camus Silverleaf oak Tree, deciduous broadleaf

Quercus tyrata Walt. Overcup oak Tree, deciduous broadleaf

Quercus oblongifolia Torr. Mexican blue oak Tree, deciduous broadleaf

Quercus petraea Sessile oak Tree, deciduous broadleaf

Quercus reticulata Hump & Bonpl

= Quercus rugosa Nee Live oak, netleaf oak Tree, evergreen broadleaf

Quercus rubra L. Northern red oak Tree, deciduous broadleaf

Ranunculus testiculatus Crantz Hornseed buttercup Herb, forb

Rhus spp. L. Sumac Shrub/tree, deciduous broadleaf

Robinia neomexicana Gray New Mexico locust Tree, deciduous broadleaf

Robinia pseudoacacia L. Black locust Tree, deciduous broadleaf

Rubus spectubilis Pursh. Salmonberry Shrub, deciduous

AP-4 = APPENDIX A _ Scientific and Common Names ofPlants

Salix spp. L. Willow Tree (or shrub) deciduous broadleaf

Secale cereale L. Winter rye Herb, grass
Sequoia sempervirens (D. Don) Endl. Redwood Tree, evergreen coniferous
Sheperdia canadensis (L.) Nutt. Soapberry Shrub, deciduous
Solidago spp. L. Goldenrod Herb, forb
Spirea douglasii Hook. Hardhack Shrub, deciduous
Taraxacum officinale Weber Dandelion Herb, forb
Taxus brevifolia Nutt. Pacific yew Tree, evergreen conifer
Taxus canadensis Marsh. Canada yew Tree, evergreen conifer
Thuja occidentalis L. Northern white cedar Tree, evergreen conifer
Thuja plicata Donn. Western redcedar Tree, evergreen conifer
Thuja standishii (Gord.) Carr. Japanese arbor-vitae Tree, evergreen conifer
Tsuga canadensis (L.) Carr. Eastern hemlock Tree, evergreen conifer
Tsuga heterophylla (Raf.) Sarg. Western hemlock Tree, evergreen conifer
Ulmus americana L. American elm Tree, deciduous broadleaf
Viburnum edule (Michx.) Raf. High-bush cranberry Shrub, deciduous
 Scientific and Common
Names of Animals

Common Name Scientific Name Type of Animal

Alaska black fish Dallia pectoralis (Bean) fish

Ant Pseudomyrex ferruginea (Smith) insect
Army worm Spodoptera exempta (Walk.) insect
Badger (e.g. American badger) Taxidea taxus (Schreber) mammal
Bears Ursidae mammal
Black bear Ursus americanus (Pallas) mammal
Grizzly bear Ursus arctos (L.) mammal
Beaver Castor canadensis (Kuhl) mammal
Budmoth, grey larch Zeiraphera griseana (Hubner) insect
Budworm, spruce Choristoneura fumiferana (Clemens) insect
Butterfly, monarch Danaus plexippus (L.) insect
Caribou, barren-ground Rangifer tarandus groenlandicus (L.) mammal
Caribou, mountain Rangifer tarandus montanus (Seton) mammal
Chalcid, golden (parasitic wasp Aphytis chrysomphali insect
(Aphelinid))
Cockroaches Blattidae insect
Cougar Felis concolor (L.) mammal
Cowbird, brown-headed Molothrus ater (Boddaert) bird
Cowbird, giant Scaphidura oryzivora bird
Coyotes Canis latrans (Say) mammal
Cuckoo Cuculus canorus bird
Deer, black-tailed Odocoileus hemionus colomblanus (Richardson) mammal
Deer, rocky mountain mule Odocoileus hemionus hemionus (Rafinesque) mammal
Deer, white-tailed Odocoileus virginianus (Zimmermann) mammal
Dove, rock Columba livia (Gmelin) bird
Elk Cervus elaphus canadensis (Erxleben) mammal
Fisher Martes pennanti (F. Erxleben) mammal

Fly, fruit Drosophila melanogaster (Meigen.) insect

Gophers, pocket mammal

Botta’s Thomomys bottae (Eydoux and Gervais)

Mountain Thomomys talpoides (Richardson)

AP-5
AP-6 | APPENDIXB Scientific and Common Names ofAnimals

Common Name Scientific Name Type of Animal

Plains Geomys bursarius (Shaw)

Yellow faced Pappogeomys castanops (Baird)
Goshawk Acclpiter gentilis (L.) bird
Grebe, western Aechmophorus occidentalis (Lawrence) bird
Grouse, red Lagopus lagopus scoticus (L.) bird
Grouse, ruffed Bonasa umbellus L. bird
Grouse, sharp-tailed Tympanuchus pnasianeitus (L.) bird
Grouse, spruce Canachizes canadensis (L.) bird
Grouse, blue Dendragopus obscurus (Say) bird
Grouse, willow Lagopus lagopus (L.) bird
Heath hen Tympanuchus cupido cupido (L.) bird
Hedgehogs Erinaceidae mammal
Hummingbirds Trochilidae bird
Lemmings (e.g. brown lemming) Lemmus sibiricus (Kerr) mammal
Little white heron, snowy egret Leucophoyx chula bird
Looper, pine Bupalis piniaria (L.) insect
Lynx Lyco lynx (L.) mammal
Marmot, arctic Marmota browerr (Hall and Gilmore) mammal
Marten Maries americana (Turton) mammal
Moose Aices aices (L.) mammal
Moth, peppered (pepper-and-salt) Biston cognataria betularia (Guenee) insect
Moth, green oak-roller Tortriz viridiane (L.) insect
Moth, gypsy Porthetria dispar (L.) insect
Mouse, field Microtus pennsylvanicus (Ord) mammal
Mouse, white-footed deer Peromyscus maniculatus (Wagner) mammal
Musk oxen Ovibos moschatus (Zimmermann) mammal
Oropendula, chestnut-headed Zarynchus wagleri bird
Owl, snowy Nyctea scandiaca (L.) bird
(Parasitic wasp (Aphelinid)) Aphytis tignanesis insect
(Parasitic wasp (Aphelinid)) Aphytis melinus insect
Pheasant, ring-necked Phasianus colchicus (L.) bird
Pigeon, passenger Enopistes migratorus (L.) bird
Pigeons Columbidae bird
Pillbug Armadillidium vulgare (Latreille) isopod
Psyllids Psyllidae insect
Ptarmigan, willow arctic Lagopus mutus (Montin) bird
Quail, bob-white Colinus virginianus (L.) bird
Rabbit, snowshoe (varying hare) Lepus americanus (Erxlehen) mammal
Racoons Procyon lotor (L.) mammal
Reindeer Rangifer tarandus tarandus (L.) mammal
Rook Corvus fragilegus bird
Salamander Urodela amphibian
Salmon, sockeye Oncorhynchus nerka (Walbaum) fish
Sawfly, European pine Diprion pini (L.) insect
Scale, California red Aonidiella auranat (Maskell) insect
Scale, cottony-cushion Icerya purchase (Maskell) insect
Shipworm Teredo navalis (L.) bivalve
Shrew, masked Sorex cunereus (Kerr) mammal
 APPENDIX B Scientific and Common Names ofAnimals — AP-7

Common Name Scientific Name Type of Animal

Shrew, short-tailed Blarina brevicauda (Say) mammal

Sowbug Porceilio scarber isopod
Squirrel, Monave ground Spermophilis monavensis (Merriam) mammal
Squirrel, red Tamiasciurus hudsonicus (Erxlehen) mammal
Starting (European) Sturnus vulgaris (L.) bird
Swallow Hirundinidae bird
Tern, arctic Sterna paradisaca (Brunnich) bird
Thar, Himalayan Hemitragus jemlaticus (H. Smith) mammal
Tit, great Parus major (L.) bird
Turkey, wild Meleagris gallopavo (Vieillot) bird
Tussock moth, Douglas-fir Orgyia pseudosugata McDunnough insect
Vedalia (beetle) Rodolia cardinalis (Muls.) insect
Vole, field Microtus pennsylvanicus (Ord) mammal
Vole, California Microtus californicus (Peale) mammal
Weevil, granary Sitophilus granarius (L.) insect
Wolf Canis tupus (L.) mammal
Wolverine Gulo gulo (L.) mammal
Woodland caribou Rangifer tarandus caribou (Gmelin) mammal
 Ch: 2 PTA

Z*
ore) iin? ~
~s a ale - sud)
~—

Le ay aya

ae ha = a
cas
owt yi

nat {Eee
>
4

y)

Wy ary ad ae
oe
 References Cited

Abee, A., and D. P. Lavender. 1972. Nutrient cycling in ,R. E. McMurtrie, W. J. Parton,J.Pastor, and H. H.
throughfall and litterfall in a 450-year-old Douglas-fir Shugart. 1991. State-of-the-art of models of produc-
stand. pp. 133-143.In J. F. Franklin et al. (eds.), tion—decomposition linkages in conifer and grassland
on Coniferous Forest Ecosystems.
Proceedings, Research ecosystems. Ecol. Appl. 1(2):118-138.
USDA For. Serv. PNW For. Range Exp. Sta., Portland, Ahlgren, C. E. 1970. Some effects of prescribed burning on
Ope. jack pine reproduction in northeastern Minnesota. Minn.
Aber,J.D., D. B. Botkin, andJ.M. Melillo. 1978. Predicting Agr. Exp. Stat. Misc. Rept. 94. 14 pp.
the effects of different harvesting regimes on forest floor Ahlgren, I. F. 1974. The effect of fire on soil organisms. pp.
dynamics in northern hardwoods. Can. 7. For. Res. 47-72. In 'T. T. Kozlowski and C. E. Ahlgren (eds.), Fire
8:306-315. and Ecosystems. Academic Press, New York.
, D. B. Botkin, and J. M. Melillo. 1979. Predicting ———, and C. E. Ahlgren. 1960. Ecological effects of forest
the effects of different harvesting regimes on productivi- fires. Bot. Rev. 26:483-533.
ty and yield in northern hardwoods. Can. 7. For. Res. , and C. E. Ahlgren. 1965. Ecology of prescribed
9:10-14. burning on soil microorganisms in a Minnesota jack pine
, andJ.M. Melillo. 1982a. FORTNITE: A computer forest. Ecol. 46:304-310.
model of organic matter and nitrogen dynamics in forest Aho, P. E. 1974. Decay. Pages Q1-Q17. In O.P. Cramer (ed)
ecosystems. Univ. Wisconsin, Research Bulletin, R3130. Environmental Effects of Forest Residues Management in the
49 pp. Pacific Northwest: A State-of-Knowledge Compendium.
, and J. M. Melillo. 1982b. Nitrogen immobilization USDA For. Serv. Gen. Tech. Rept. PNW-24.
in decaying hardwood leaf litter as a function ofinitial ni- Ajtay, G. L., P. Ketner, and P. Duvigneaud. 1979. Terrestrial
trogen and lignin content. Can. 7. Bot. 60:2263-2269. primary production and phytomass. pp. 129-181. Jn B.
, and J. P. Melillo. 1980. Litter decomposition: Mea- Bolin, E. T. Degens, S. Kempe, and P. Ketner (eds.), The
suring relative contributions of organic matter and nitro- Global Carbon Cycle. Scope Report 13. Wiley, New York.
gen to forest soils. Can. 7. Bot. 58:416-421. Albani, M. 2002. Spatial Analysis in a Successional Perspec-
, J. P. Melillo, and C. A. McClaugherty. 1990. Pre- tive: A Boreal Mixedwood Landscape in Northeastern
dicting long-term patterns of mass loss, nitrogen dynam- British Columbia. Doctoral thesis, University of British
ics, and soil organic matter formation from initial fine Columbia, Vancouver, BC.
litter chemistry in temperate forest ecosystems. Can. 7. Albrecht, F. O. 1967. Polymorphisme phasaire et biologie des
Bot. 68:2201-2208. acridiens migrateurs. Masson, Paris. 192 pp.
Aber, J. D., andJ.M. Melillo. 1982. FORTNITE: A Com- Albrecht, W. A. 1958. Soil Fertility and Animal Health. Fred
puter Model of Organic Matter and Nitrogen Dynamics Hahne Printing Co., Webster City, Iowa. 232 pp.
in Forest Ecosystems R3130. University of Wisconsin, Alcock, M. B. 1964. The physiological significance of defoli-
Madison. ation on the subsequent regrowth of grass-clover mix-
Adamowicz, W. L., W. White, and W. E. Phillips (eds.). tures and cereals. pp. 25-41. In A. J. Crisp (ed.), Grazing
1993. Forestry and the environment: Economic perspec- in Terrestrial and Marine Environments. Blackwell Scien-
tives. CAB International Wallingford, UK. 304 pp. tific Publications, Oxford.
Agee, J. K. 1974. Environmental impacts from fire manage- Alexander, M. E., and FE G. Hawksworth. 1976. Fire and
ment alternatives. Nat. Park Serv., West. Reg. Off. San dwarf mistletoes in North American coniferous forests. 7.
Francisco, Calif. 92 pp. For. 74:446-449.
. 1993. Fire Ecology ofPacific Northwest Forests. Islands Allee, W. C. 1931. Animal Agegregations: A Study in General
Press, Washington, D.C. 493 pp. Sociology. Univ. of Chicago Press. 431 pp.
Agren, G. I. 1983a. Nitrogen productivity of some conifers. . 1951. Cooperation Amongst Animals, With Human
Can. 7. For. Res. 13:494-500. Implications. Henry Schumann, New York. 233 pp.
. 1983b. The concept of nitrogen productivity in for- , A. E. Emerson, O. Park, T. Park, and K. P. Schmidt.
est growth modeling. Mitt. Forstl. Bundes-Versuchsanst. 1967. Principles of Animal Ecology. W. B. Saunders Co.,
147:199-210. Philadelphia. 837 pp.
. 1985. Theory of growth of plants derived from the Allen, S. E. 1964. Chemical aspects of heather burning. 7.
nitrogen productivity concept. Physiol. Plant. 64:17-28. Appl. Ecol. 1:347-367.

R-]
R-2 REFERENCES CITED

Almer, B. 1974. Effects of acidification on Swedish Lakes. in translation as Eyre, S. A. 1971. Regeneration patterns
Ambio 3(1):30-36. in the closed forest of Ivory Coast. World vegetation
Amadon, D. 1947. Ecology and evolution of some Hawaiian es).
birds. Evolution 1:63-68. Aulitzky, H. 1961. Standortsuntersuchungen in der sub-
Anderson,J. 1952. Fluctuations in the field hare populations alpinen Stufe. 4. Die Bodentemperaturen in der Kampf-
in Denmark compared with certain climatic factors. Pa- zone oberhalb der Waldgrenze und im subalpinen
pers on Game Research. Finnish Game Foundation. No. Zirben-Larchenwald. Mitt. Der forst. Bundes-Versuch.
8:41-43. Mariabrunn 59:153-208.
Andersson, T. 1991. Influence of stemflow and throughfall Axelsson, E., and B. Axelsson. 1986. Changes in carbon allo-
from common oak (Quercus robur) on soil chemistry and cation patterns in spruce and pine trees following irriga-
vegetation patterns. Can. 7. For. Res. 21:917-924. tion and fertilization. Tree Physiol. 2:189-204.
Andrewartha, H. G., and L. C. Birch. 1954. The Distribution B. C. Ministry of Forests. 1994. British Forest Practices Code.
and Abundance ofAnimals. Univ. of Chicago Press, Chica- Standards with Revised Rules and Field Guide References.
go. 782 pp. Ministry of Forests, Victoria, BC. 216 pp.
, and T. O. Browning. 1961. An analysis ofthe idea of Bailey, R. G. 1976. Ecoregions of the United States. USDA
“resources” in animal ecology. 7. Theoret. Biol. 1:83-97. For. Serv. Intermountain Reg., Ogden, Utah. Map, scale
Anonymous. 1991. 1989:2010—Wood and Wood Products. 1:7,500,000.
Food and Agriculture Organization, Rome. 39 pp. . 1978. Description of the ecoregions of the United
Armson, K. A. 1977. Forest Soils: Properties and Processes. States. USDA For. Serv. Intermountain Reg., Ogden,
Univ. Toronto Press, ‘loronto. 390 pp. Utah. 77 pp.
, and H. Struik. 1969. The effects of oxygen levels on Bajikov, A. D. 1949. Do fish fall from the sky? Science
the growth of black spruce and red pine seedlings. Univ. 109:402.
Toronto Faculty of Forestry, Glendon Hall Ann. Report Baker, F. S. 1950. The Principles of Silviculture. McGraw-Hill,
1967-68. New York, 414 pp.
, J. McG. Taylor, and E. Astley. 1973. The effect of Baker,J. B., and B. G. Blackmon. 1977. Biomass and nutri-
fire on organic layers of spodosols in the boreal forests of ent accumulation in a cottonwood plantation—The first
Ontario. Agron. Abstr. 1973:137. growing season. Soil Sci. Soc. Amer. Proc. 41:632-636.
Art, H. W., and P. L. Marks. 1971. A summary table of bio- , and W. M. Broadfoot. 1979. Site evaluation for
mass and net annual primary production in forest ecosys- commercially important southern hardwoods. USDA
tems of the world. pp. 3-32. In Forest Biomass Studies. For. Serv. Gen. Tech. Rept. SO-26. 51 pp.
IUFRO Univ. Maine, Orono. Baker, T. G., W.J.Dyck, P. G. Barton, G. R. Oliver, and G.
, FE. H. Bormann, G. K. Voigt, and G. M. Woodwell. Nicholson. 1990. Effect of irrigation with sewage efflu-
1974. Barrier Island forest ecosystem: Role of meteoro- ent on decomposition of litter in Pinus radiata forests.
logic nutrient inputs. Science 184:60-62. For. Ecol. Manage. 31:205-214.
Ash, M. J., P. C. Knoblock, and N. Peters. 1980. Energy Baldwin, I. T., andJ.C. Schultz. 1983. Rapid changes in tree
analysis of energy from forest options. ENFOR Project leaf chemistry induced by damage: Evidence for commu-
P-59. Can. For. Serv., Ottawa. 79 pp. nication between plants. Science 221:277-279.
Attiwill, P. J. 1994a. Ecological disturbance and the conserv- Baldwin, J. P. 1975. A quantitative analysis of the factors af-
ative management of eucalypt forests in Australia. For: fecting plant nutrient uptake from some soils. 7. Soil Sci.
Ecol. Manage. 63:303-348. 26:195-206.
. 1994b. The disturbance of forest ecosystems: The Ballard, R., and S. P. Gessel (eds.). 1983. IUFRO Symposium
ecological basis for conservation management. For: Ecol. on Forest Site and Continuous Productivity. USDA For.
Manage. 63:247-300. Serv. Gen. Tech. Rept. PNW-163, Portland, Ore. 406 pp.
Attiwill, P. M. 1966. The chemical composition of rainwater Ballard, T. M. 1972. Subalpine soil temperature regimes in
in relation to cycling of nutrients in mature Eucalyptus southwest British Columbia. Arctic Alpine Res. 4:139-146.
forests. Plant Sci. 24:390-406. , and D. W. Cole. 1974. Transport of nutrients to tree
. 1986. Nutrient cycling in a Excalyptus obliqua root systems. Can. F. For. Res. 4:563-564.
(LHerit) forest. [V. Nutrient uptake and nutrient return. , L.A. Black, and K. G. McNaughton. 1977. Summer
Aust. 7. Bot. 28:199-222. energy balance and temperatures in a forest clearcut in
. 1980. Interactions between carbon and nutrients in southwestern British Columbia. pp. 74-86. In Energy, Water
the forest ecosystem. Tiree Physiol. 2:389-399. and the Physical Environment ofthe Soil. 6th British Columbia
, and G. W. Leeper. 1987. Forest Soils and Nutrient Soil Sci. Workshop Report, Min. Agri., Victoria. B.C.
Cycles. Melbourne Univ. Press, Melbourne. 202 pp. Bangay, G. E., and C. Riordan. 1983. U.S.-Canada memo-
Aubréville, A. 1938. La forét coloniale: les foréts de l’Afrique randum of intent on transboundary air pollution. Impact
occidentale frangaise. Ann. Acad. Sci. Colon 9:1-245. (Also Assessment Work Group 1, Final Report. Jan. 1983.
 REFERENCES CITED R-3

Bange, G. G. J. 1953. On the quantitative explanation of Bell, M. A. M., J. M. Beckett, and W. F. Hubbard. 1974a.
stomatal transpiration. Acta Bot. Neerl. 2:255-297. Impact of harvesting on forest environments and re-
Banner, A.,J.Pojar, and G. E. Rouse. 1983. Postglacial paleo- sources. A review of the literature and evaluation of re-
ecology and successional relationships of a bog woodland search needs. Can. For. Serv., Dept. of Environ., Ottawa.
near Prince Rupert, British Columbia. Can. 7. For. Res. 141 pp.
13:938-947, , J. M. Beckett, and W. F. Hubbard. 1974b. Impact of
Bannikov, A. G. 1967. Some remarks concerning determina- harvesting on forest environments and resources. Anno-
tion of biomass of wild ungulates in natural geographical tated bibliography. Can. For. Serv., Dept. of Environ.,
zones of the USSR. pp. 255-260. In K. Petrusewicz (ed.), Ottawa. For. Tech. Rept. 3. 237 pp. + Suppl. No. 1.,
Secondary Productivity of Terrestrial Ecosystems (Principles 1976, 16 pp.
and Methods). IBP and Polish Academy of Science. Panst- Bell, P. R. 1982. Methane hydrate and the carbon dioxide
wowe Wydawnictwo Naukowe, Warsaw, Poland. question. pp. 401-406. In W. C. Clark (ed.), Carbon Diox-
Barclay-Estrup, P., and C. H. Gimingham. 1969. The de- ide Review: 1982. Clarendon Press, Oxford.
scription and interpretation of cyclical processes in a Bendell,J.F. 1974. Effects of fire on birds and mammals. pp.
heath community. 1. Vegetational change in relation to 73-138. In T. T. Kozlowski, and C. E. Ahlgren (eds.), Fire
the Calluna cycle.7.Ecol. 57:737-758. and Ecosystems. Academic Press, New York.
Barry, R. G., and R. J. Chorley. 1968. Atmosphere, Weather Benecke, U., and M. R. Davis. 1980. Mountain environ-
and Climate. Methuen, London. 319 pp. ments and subalpine tree growth. For. Res. Inst. N.Z.
Bates, C. G., and J. Roeser Jr. 1928. Light intensities re- For. Serv. Tech. Pap. No. 70. 288 pp.
quired for growth of coniferous seedlings. Amer. 7. Bot. Bennett, J. 1960. A comparison of selective methods and a
15:185-244. test of the pre-adaptation hypothesis. Heredity 15:65-77.
Batra, S. W. T., and L. R. Batra. 1967. The fungus gardens Bennett, W. D. 1960. Survey oflightning fire occurrences in
of insects. Scie. Ame. 217:112-120. Canada’s forests: 1950-1959. Pulp Pap. Res. Inst. Can.
Bazzaz, F. A. 1979. The physiological ecology of plant suc- Tech. Rept. 212.
cession. Ann. Rev. Ecol. Syst. 10:351-371. Berg, B., and C. A. McClaugherty. 1989. Nitrogen and
, and S. T. A. Pickett. 1980. Physiological ecology of phosphorus release from decomposing litter in relation
tropical succession: A comparative review. Amn. Rev. Ecol. to the disappearance oflignin. Can. 7. Bot. 67:1148-1156.
Syst. 11:287-310. Berger, U., and H. Hildenbrandt. 2000. A new approach to
Beadle, N. C. W. 1940. Soil temperatures during forest fires spatially explicit modelling of forest dynamics: Spacing,
and their effect on the survival of vegetation. 7. Ecol. ageing and neighborhood competition of mangrove
28:180-192. trees. Ecol. Model. 132:287-302.
. 1966. Soil phosphate and its role in moulding seg- Bernard, S. 1968. Thermoregulatory function of growing
ments of the Australian flora and vegetation, with special antlers (deer, caribou, goat). Nature 218:870-872.
reference to xeromorphylly and sclerophylly. Ecology Berndt, H. W. 1971. Early effects of forest fire on stream-
47:922-1007. flow characteristics. USDA For. Serv. PNW For. Range
Beamish, R. J. 1974. Loss of fish populations from unex- Exp. Sta. Res. Note PNW-148. 9 pp.
ploited remote lakes in Ontario, Canada, as a conse- , and W. B. Fowler. 1969. Rime and hoarfrost in
quence of atmospheric fallout of acid. Water Res. 8:85—95, upper-slope forests of eastern Washington. 7. For:
. 1975. Long-term acidification of a lake and result- 67:92-95.
ing effects on fishes. Ambio 4(2):98-102. Bessey, E. A. 1950. Morphology and Taxonomy of the Fungi.
Beard, J. S. 1973. The physiognomic approach. pp. Blakiston Co., Philadelphia. 791 pp.
355-386. In R. H. Whittaker (ed.), Ordination and Classi- Bilan, M. V. 1971. Some aspects of tree root distribution. pp.
fication of Communities. Part 5, Handbook of Vegetation Sci- 69-80. In Mycorrhizae. Proc. Ist N. American Conference on
ence. W. Junk b.v., Publishers, The Hague. Mycorrhizae. April 1969 USDA For. Serv., Misc. Pub. 1189.
Beaufait, W. R. 1960. Some effects of high temperatures on Billings, W. D. 1954. Temperature inversions in the pinyon-
the cones and seeds ofjack pine. For Sci. 6:194-199. juniper zone of a Nevada mountain range. Butler Univ.
Beck, S. 1956. The European corn borer. Pyrausta nubilalis Bot. Studies 11:112-118.
(Hubn.) and its principle host plant. II. The influence of Binkley, D. 1982. Nitrogen fixation and net primary produc-
nutritional factors on larval establishment and develop- tion in a young Sitka alder stand Can. 7. Bot. 60:281—284.
ment of the corn plant. Ann. Entomol. Soc. Amer. . 1983. Ecosystem production in Douglas-fir planta-
49:582-588. tions: Interaction of red alder and site fertility. For: Ecol.
Beggs, J. P. 1964. Growth inhibitor in soil. N.Z.7. Agr: Manage. 5:215-227.
108:529-535. . 1984. Does forest removal increase rates of decom-
Bélisle, J. 1980. Field Manual for Describing Soils. Ontario In- position and nitrogen release? For Ecol. Manage.
stitute of Pedology. Univ. of Guelph, Ontario. 8:229-233.
R-4 REFERENCES CITED

. 1995. The influence of tree species on forest soils: Bogert, C. M. 1959. Thermoregulation in reptiles, a factor
Processes and patterns. pp. 1-33. Jn D. J. Mead and L. S. in evolution. Evolution 3:195—211.
Cornforth (eds.), Proc., Trees and Soil Workshop. Lincoln Bolin, B. 1970. The carbon cycle. Sci. Amer. 223:125—132.
Univ. 1994. Agron. Soc. NZ Special Publ. No. 10. , E. T. Degens, S. Kempe, and P. Ketner (eds.). 1979.
Birdsey, R. A., A. J. Plantirga, and L. S. Heath. 1993. Past The Global Carbon Cycle. Scope Report 13. Wiley, New
and prospective carbon storage in United States forests. York. 491 pp.
For. Ecol. Manage. 58:33—40. Bolker, G. M., S. W. Pacala, F. A. Bazzaz, C. D. Canham,
Birk, E., and P. M. Vitousek. 1986. Nitrogen availability and and S. A. Levin. 1995. Species diversity and ecosystem
nutrient use efficiency in loblolly pine stands. Ecology response to carbon dioxide fertilization—Conclusions
67:69-79. from a temperate forest model. Global Change Biol.
Bishop, D. M., and M. E. Stevens. 1964. Landslides in 1:373-381.
logged areas in southeast Alaska. USDA For. Serv. Res. Bollen, W. B. 1974. Soil microbes. pp. B1-B41. In
Pap. NOR-1. 18 pp. Environmental Effects of Forest Residues Management in the
Biswell, J. K. 1960. Danger of wildfires reduced by pre- Pacific Northwest. USDA For. Serv. Gen. Tech. Rept.
scribed burning in ponderosa pine. Calif: Agr. 14(10):5—6. PNW-24.
Bjorkman, E. 1970. Mycorrhiza and tree nutrition on poor Bonan, G. B. 1989. Environmental factors and ecological
forest soils. Stud. for. suec. Skogs. #83, 24 pp. processes controlling vegetation patterns in boreal
Black, T. A. 1977. Micrometeorological studies of Douglas- forests. Landscape Ecol. 3:111-130.
fir. In J. J. Landsberg and C. V. Cutting (eds.), . 1992a. Processes in boreal forest. pp. 9-13. In H. H.
Environmental Effects on Crop Physiology. Academic Press, Shugart, R. Lemans, and G. B. Bonan (eds.), A Systems
New York. (Proc. 5th Symposium, Long Ashton Re- Analysis of the Global Boreal Forest. Cambridge Univ.
search Sta., Univ. of Bristol, London, 13-16 April 1975). Press, Cambridge, UK.
Blair,J.M., R. W. Parmelee, and M. H. Beare. 1990. Decay . 1992b. Soil temperature as an ecological factor in
rates, nitrogen fluxes, and decomposer communities of boreal forests. pp. 126-143. In H. H. Shugart, R. Le-
single or mixed-species foliar litter. Ecology 71:1976- mans, and G. B. Bonan (eds.), A Systems Analysis of the
1985. Global Boreal Forest. Cambridge Univ. Press, Cambridge,
Blair-West,J. R.,J.P. Coghlan, D. A. Denton,J. F. Nelson, UK.
E. Orchard, B.A. Scoggins, R.D. Wright, K. Myers, and Bonan, G. B., and H. H. Shugart. 1989. Environmental fac-
C.L. Jungueira. 1968. Physiological, morphological and tors and ecological processes in boreal forests. Annu. Rev.
behavioral adaptations to a sodium deficient environ- Ecol. Syst. 20:1-28.
ment by wild native Australian and introduced species of Bonan, G. B., and M. D. Korzuhin. 1989. Simulation of
animal. Nature 217:922-928. moss and tree dynamics in the boreal forests of interior
Blais, J. R. 1958. The vulnerability of balsam fir to spruce Alaska. Vegetatio 84:3 1-44.
budworm attack in northern Ontario with special refer- Bond, G. 1977. Some reflections on A/nus-type root nod-
ence to the physiological age of the tree. For. Chron. ules. pp. 531-537. In W. Newton,J. R. Postgate, and C.
34:405-422. Rodrigues-Barrneco (eds.), Recent Developments in Nitro-
Blinkenberg, G., H. Brix, S. de Muckadell, and H. Vedel. gen Fixation. Academic Press, New York.
1958. Controlled pollinations in Fagus. Silv. Genet. Bonnevie-Svendson, C., and O. Gjems. 1956. Amount and
7:116-122. chemical composition of the litter from larch, beech,
Bliss, L. C. 1962. Adaptations of arctic and alpine plants to Norway spruce and Scots pine stands and its effects on
environmental conditions. Arctic 15:117-144. the soil. Be + Norske Skogforsoksvesen 115-169.
. 1966. Plant productivity in alpine microenviron- Bormann, B. T. 1989. Podzolization and windthrow: Natur-
ments on Mount Washington, New Hampshire. Ecol. al fluctuations in long-term productivity and implications
Monogr. 36:125-155. for management. pp. 245. In S. P. Gessel, D. S. Lacate,
BockheimJ. G., T. M. Ballard, and R. P. Willington. 1975. Soil G. F. Weetman, and R. F. Powers (eds.), Sustained Produc-
disturbance associated with timber harvesting in south- tivity of Forest Soils. Proc. 7th N. Am. For. Soils Conf.,
western British Columbia. Can. 7. For: Res. 5:285-290. Univ. of B.C., Faculty of Forestry, Vancouver, B.C.
, and J.F. Leide. 1991. Foliar nutrient dynamics and , FH. Bormann, W. B. Bowden, R. S. Pierce, S. P.
nutrient-use efficiency of oak and pine on low fertility Hamburg, D. Wang, M. C. Snyder, C. Y. Li, and R. C.
soil in Wisconsin. Can. 7. For. Res. 21:925-934. Ingersall. 1993. Rapid N, fixation in pines, alder, and lo-
Bodenheimer, F. S. 1928. Welche Faktoren regulieren die cust: Evidence from the sandbox ecosystem study. Ecology
Individuenzahl einer Insektenart in der Natur? Biol. Zen- 74:583-598.
tralbl. 48:714-739. , F H. Bormann, and G. E. Likens. 1979. Pattern and
Boelter, D. H. 1964. Water storage characteristics of several Process in a Forested Ecosystem. Springer-Verlag, New
peats in situ. So#/ Sci. Soc. Amer. Proc. 28:433-435. York. 253 pp.
 REFERENCES CITED R-5

, and G. E. Likens. 1979a. Catastrophic disturbance ,and A. R. Ek. 1972. An evaluation of some effects of
and the steady state in northern hardwood forests. Amer. bole and branch pulpwood harvesting on site macronu-
Sci. 67:660-669. trients. Can. 7. For. Res, 2:407-412.
, and G. E. Likens. 1979b. Patterns and Processes in a , and A. R. Ek. 1973. Whole tree harvesting: nutrient
Forested Ecosystem. Springer-Verlag, New York. 253 pp. budget evaluation.7. For: 71:760-762.
Bornemisza, E., and R. Llanos. 1967. Sulphate movement, Boysen-Jensen, P. 1932. Die Stoffproduktion der Pflanzen.
adsorption, and resorption in three Costa Rican soils. Soil Gustav Fisher Verlag, Jena.
Set. Soc. Amer. Proc. 31:336-360. Bradshaw, A. D. 1952. Populations of Agrostis tenuis resistant
Bossel, H. 1991. Modelling forest dynamics: moving from to lead and zinc poisoning. Nature 169:1098.
description to explanation. For: Ecol. Manage. 42:129-142. Brady, M. A. 1996. Organic Matter Dynamics in the Surface
. 1996. TREEDYNE3 forest simulation model. Eco- Layers of Coastal Peat Deposits in Sumatra, Indonesia:
logical Modeling 90:187-227. The Effects of Vegetation and Edaphic Changes Associ-
, and H. Krieger. 1991. Simulation model of natural ated With Increasing Peat Depth. Doctoral thesis, Univ.
tropical forest dynamics. Ecol. Model. 59:37-71. of British Columbia, Vancouver.
, and H. Krieger. 1994. Simulation of multi-species Brady, N. C. 1984. The Nature and Properties of Soils. 9th ed.
tropical forest dynamics using a vertically and horizon- Macmillan, New York. 750 pp.
tally structured model. For: Ecol. Manage. 69:123-144. Braekke, F. H. (ed.). 1976. Impact ofAcid Precipitation on For-
Botkin, D. B. 1980. A grandfather clock down the staircase: est and Freshwater Ecosystems in Norway. SNSF-Project,
Stability and disturbance in natural ecosystems. pp. 1-10. Agric. Res. Council of Norway, Oslo. 111 pp.
In R. H. Waring (ed.), Forests: Fresh Perspectives from Braestrup, F. W. 1940. The periodic die-off in certain her-
Ecosystem Analysis. Proc. 40th Ann. Biol. Colloquium. bivorous mammals and birds. Science 92:354-355.
Oregon State Univ. Oregon State Univ. Press, Corvallis. Bramble, W. C., and W. R. Byrnes. 1972. A Long-Term Eco-
. 1981. Causality and succession. pp. 36-55. In D. C. logical Study of Game Food and Cover on a Sprayed Utility
West, H. H. Shugart, and D. B. Botkin (eds.), Forest Suc- Right-of-Way. Purdue University, Agric. Exper. Sta. Res.
cession. Springer-Verlag, New York. Bull. #885. 20 pp.
. 1990. Discordant Harmonies: A New Ecology for the Twen- Bray, J. R. 1961. An estimate of aminimum quantum yield of
ty-First Century. Oxford Univ. Press., New York. 241 pp. photosynthesis based on ecological data. Plant Physiol.
. 1993. Forest Dynamics: An Ecological Model. Oxford 36:371-373.
Univ. Press, Oxford. 309 pp. . 1962. Estimates of energy budget for a Typha (cat-
. 1994. Preface. pp. XXI-XXIV. In N. Sampson, tail) marsh. Science 136:1119-1120.
D. L. Adams, and M. J. Enger (eds.), Assessing Forest , and E. Gorham. 1964. Litter production in the
Ecosystem Health in the Inland West. Haworth Press, NY. forests of the world. Adv. Ecol. Res. 2:101-157.
. 1995. Our Natural History. The Lessons of Lewis and Brayshaw, ‘I. C. 1970. The dry forests of British Columbia.
Clark. Grosset/Putnam, New York. 300 pp. Syesis 3:17-43.
, J. FE Janak, andJ.R. Wallis. 1972. Some ecological Brink, V. C. 1959. A directional change in the subalpine for-
consequences of a computer model of forest growth. 7. est-heath ecotone in Garibaldi Park, British Columbia.
Ecol. 60:849-872. Ecology 40:10-16.
Botkin, D. B. 1993. Forest Dynamics: An Ecological Model. Ox- Brix, H. 1971. Effects of nitrogen fertilization on photosyn-
ford University Press, Oxford. 309 pp. thesis and respiration in Douglas-fir. For Sci.
, J. F Janak, and J. R. Wallis. 1972a. Rationale, limi- 17:407-414.
tations, and assumptions of a northeastern forest growth Broecker, W. S. 1973. Factors controlling CO, content in
simulator. IBM7. Res. Dev. 16:101-116. the oceans and atmosphere. pp. 32-50. In G. M. Wood-
, J. F Janak, and J. R. Wallis. 1972b. Some ecological well and E. V. Pecan, (eds.), Carbon and the Biosphere.
consequences of a computer model of forest growth. 7. AEC Symp. Series 30. Tech. Info. Center. Office of Info.
Ecol. 60:849-872. Services, USAEC.
Bowen, G. D. 1984. Tree roots and the use of soil nutrients. Brokaw, N. V. L. 1982. The definition of treefall gap and its
pp. 147-179. In G. D. Bowen and E. K. S. Nambiar (eds.), effect on measures of forest dynamics. Biotropica
Nutrition ofPlantation Forests. Academic Press, New York. 14(2):158-160.
,and E. K. S. Nambiar (eds.). 1984. Nutrition of Plan- , and S. M. Scheiner. 1989. Species composition in
tation Forests. Academic Press, New York. 516 pp. gaps and structure of a tropical forest. Ecology
Boyce, M. S., and A. Hanley (eds.) 1997. Ecosystem Manage- 70:538-541.
ment: Applications for Sustainable Forest and Wildlife Re- Brooke, R. C., E. B. Peterson, and V. J. Krajina. 1970. The
sources. Yale University Press, New Haven, CT. subalpine mountain hemlock zone. Ecol. West. N. Amer.
Boyle, J. R. 1975. Nutrients in relation to intensive culture 2:148-439.
of forest crops. lowa State 7. Res. 49:293-303. Brosset, C. 1973. Air-borne acid. Ambio 11(1-2):2-9.
R-6 REFERENCES CITED

Brouzes, R., J. Lasik, and R. Knowles. 1969. The effect of puters. Internal Rept., Systems Ecology Group, ETHZ
organic amendment, water content and oxygen on the in- Ziirich. 60 pp.
corporation of '°N, by some agricultural and forest soils. Bugmann, H. 1994. On the Ecology of Mountainous
Can. J. Microbiol. 15:899-905. Forests in a Changing Climate: A Simulation Study.
Brown,J.H.,Jr.1960. The role of fire in altering the species Doctoral Thesis, No. 10638. Swiss Federal Institute of
composition of forests in Rhode Island. Ecology Technology, Zurich, Switzerland. 258 pp.
41:310-316. Bugmann, H. 1991. Development of FORCLIM, a simpli-
Brown, S., G. Gertner, A. E. Lugo, andJ.Novak. 1981. Car- fied forest gap model for workstations and personal com-
bon dioxide dynamics of the biosphere. pp. 19-28. In puters. Internal Rept., Systems Ecology Group, ETHZ
W. J. Mitsch, R. W. Bossermann, and J. M. Klopatek Zurich. 60 pp.
(eds.), Energy and Ecological Modelling. Elsevier Sci. Publ., Bunge, M. 1959. Causality: The Place of the Causal Principle in
New York. Modern Science. Harvard Univ. Press, Cambridge, Mass.
Browning, T. O. 1962. The environments of animals and 380 pp.
plants. 7. Theoret. Biol. 2:63-68. Burger, D. 1972. Forest site classification in Canada. Mitt.
Broysen, de V. P., and N. M. Tainton (eds.). 1984. Ecological Ver. Forstl. Standortsk. Forstpflans. 21:20-36.
Effects of Fire in South African Ecosystems. Ecological Studies Burges, A. 1963. The microbiology of a podzol profile. pp.
48. Springer-Verlag, Berlin. 151-157. In J. Doekson and J. van der Drift (eds.), Sod
Brues, C. T. 1946. Insect Dietary. Harvard Univ. Press, Cam- Organisms. North Holland, Amsterdam.
bridge, Mass. 466 pp. , and F. Raw (eds.). 1967. Soil Biology. Academic
Brunig, E. F. 1977. The tropical forest—A wasted asset or an Press. London. 582 pp.
essential biospheric resource? Ambio 6:187-191. Burgess, R. L., and D. M. Sharpe (eds.). 1981. Forest Island
Bryant, J. In press. Role of vertebrate herbivores in deter- Dynamics in Man-Dominated Landscapes. Springer-Verlag,
mining community structure and diversity in the taiga. Jn New York. 311 pp.
K. Van Cleve et al. (eds.), Forest Ecosystems in the Alaskan Burns, G, P. 1923. Measurement of solar radiant energy in
Taiga. Springer-Verlag, New York. plant habitats. Ecology 4:189-195,
Bryant, J. P. 1981. Phytochemical deterrence of snowshoe Burton, P. J., and S. G. Cumming. 1991, Zelig.BC: User's
hare browsing by adventitious shoots of four Alaskan Guide to the Prototype of a Forest Succession Simulator
for
trees. Science 213:889-890. the Evaluation of Partial Cutting Options in British
. 1987. Felt leaf willow—Snowshoe hare interactions: Columbia. Univ. of British Columbia, Vancouver. 82
Plant carbon/nutrient balance and floodplain succession. pp.
Ecology 68:1319-1327. , and S. G. Cumming. 1995. Potential effects of cli-
, K S. Chapin, DI, and D. R. Klein: 1983. matic change on some western Canadian forests, based
Carbon/nutrient balance of boreal plants in relation to on phenological enhancements to a patch model of forest
vertebrate herbivory. Oikos 40:357-368. succession. Water Air Soil Pollut. 82:401-414.
, and P. J. Kuropat. 1980. Selection of winter forage , A. C. Balisky, L. P. Coward, S. G. Cumming, and
by subarctic browsing vertebrates: The role of plant D. D. Kneeshaw. 1992. The value of managing for biodi-
chemistry. Ann. Rev. Ecol. Syst. 11:261-285. versity. For. Chron. 68:225-237.
Bryson, R. A. 1974. A perspective on climatic change. Science Burton, P. J., and S. G. Cumming. 1991. Zelig.BC: Users
184:753-760. Guide to the Prototype of a Forest Succession Simulator
for the
Buchanan, B. A., K. T. Harper, and N. C. Frischknecht. Evaluation of Partial Cutting Options in British Columbia.
1978. Allelopathic effects of bur buttercup tissue on ger- University of British Columbia, Vancouver, BC.
mination and growth of various grasses and forbs in vitro Burton, P. J., and S. G. Cumming. 1995, Potential effects of
and in soil. Great Basin Nat. 38:19-96. climatic change on some western Canadian forests, based
Buck, C. C. (ed.). 1964. Winds over wildlands—A guide for for- on phenological enhancements to a patch model of forest
est management. USDA For. Serv. Agric. Hndbk No. 272. succession. Water Air Soil Pollut. 82:401-414.
33 pp. Busing, R. T. 1991. A spatial model of forest dynamics.
Budyko, M. I. 1955. Atlas of Heat Balance, 2nd ed. (1963) Vegetatio 92:167-169.
Leningrad. Publ. Meteorol. 19. 35 pp. - 1995. Disturbance and the population dynamics of
. 1974. Climate and Life. Academic Press, New York. Liriodendron tulipifera: Simulations with a spatial model
508 pp. of forest succession.7.Ecol. 83:45-53.
Buffington, L. C., and C. H. Herbel. 1965. Vegetational ,and E. E. C. Clebsch. 1987. Application of a sprace-
changes on a semidesert grassland range. Ecol. Monogr: fir forest canopy gap model. For: Ecol. Manage.
35:139-164. 20:151-169.
Bugmann, H. 1991. Development of FORCLIM, a simpli- Busing, R. T. 1991. A spatial model of forest dynamics.
fied forest gap model for workstations and personal com- Vegetatio 92:167-179,
 REFERENCES CITED R-7

Busing, R. ‘T. 1995. Disturbance and the population dynam- , A. H. F. Brown, and E. J. White. 1966. Litterfall,
ics of Liriodendron tulipifera: Simulations with a spatial leaf production and the effects of defoliation by Tortix
model of forest succession. 7. Ecol. 83:45-53. viridiana in a sessile oak (Quercus petraea) woodland. 7.
Busing, R. T., and E. E. C. Clebsch. 1987. Application of a Ecol. 54:65-85.
spruce-fir forest canopy gap model. For: Ecol. Manage. , A. H. F. Brown, and E. J. White. 1967. The nutrient
20:151-169. content of tree stem flow and ground flora litter and
Cain, S. A., and R. C. Friesner. 1929. Some ecological fac- leachates in a sessile oak (Quercus petraea) woodland. 7.
tors in secondary succession: Upland hardwood. I. Evap- Ecol. 55:615-627.
oration studies in the sycamore creek region. Butler Univ. Carmean, W. H. 1959. Litter weight not reduced following
Bot. Studies Pap. 1:1-15. clear-cutting of poor site oak stands in S.E. Ohio. 7. For:
Cairns, J., Jr. 1980. The Recovery Process in Damaged Ecosys- 57:207-209.
tems. Ann Arbor Science Publishers, Ann Arbor, MI. 167 . 1975. Forest site quality evaluation in the United
Pp: States. Adv. Agron. 27:207-269.
Cajander, A. K. 1926. The theory of forest types. Acta For. Carpenter, J. R. 1940. Insect outbreaks in Europe. 7. Anim.
Fenn. 29(3):1-108. Ecol. 9:108-147.
. 1949. Forest types and their significance. Acta For. Carroll, A., L. Safranyik, and D. Linton. 2002. Mountain
Fenn. 56(4):1-71. Pine Beetle. Canadian Forestry Service, Pacific Forest Re-
, and Y. Ilvessalo. 1921. Uber Waldtypen. II. Acta For: search Centre, Victoria, BC. Available at: www.pfe.forests
Fenn. 20(1):1-77. .ca/entomology/mpb/index_e.html
Caldwell, M. M. 1968. Solar ultra violet radiation as an eco- Carter, R. E., and K. Klinka. 1992. Variation in shade toler-
logical factor for alpine plants. Ecol. Monogr. 38:243-268. ance of Douglas-fir, western hemlock, and western red-
. 1990. Water paratism stemming from hydraulic lift: cedar in coastal British Columbia. For Ecol. Manage.
A quantitative test in the field. Israel 7. Bot. 39:395-402. 55:87-105.
Callaham, R. Z. 1962. Geographic variability in growth of Cattelino, P. J., I. R. Noble, R. O. Slatyer, and S. R. Kessell.
forest trees. pp. 311-325. In T. T. Kozlowski (ed.), Tree 1979. Predicting the multiple pathways of plant succes-
Growth. Ronald Press, New York. sion. Environ. Manage. 3:41-50.
Canham, C. D. 1989. Different responses to gaps among Caughley, G. 1970. Eruption of ungulate populations with
shade-tolerant tree species. Ecology 70(3):548-550. emphasis on Himalayan thar in New Zealand. Ecology
, J. S. Denslow, W. J. Platt,J.R. Runkle, T: A. Spies, §1:53-72.
and P. S. White. 1990. Light regimes beneath closed Cavers, P. B., andJ.L. Harper. 1966. Germination polymor-
canopies and tree fall gaps in temperate and tropical phism in Rumex crispus and R. obtusifolius. J. Ecol.
forests. Can. 7. For. Res. 20:620-631. 54:367-382.
Canham, C. D., K.D. Coates, P. Bartemucci, and S. Caza, C. L. 1993. Woody Debris in the Forests ofBritish Colum-
Quaglia. 1999. Measurement and modeling of spatially- bia: A Review of the Literature and Current Research. B.C.
explicit variation in light transmission through interior Min. For., Victoria, B.C., Land Management Rept., No.
cedar-hemlock forests of British Columbia. Can. 7. For. 78. 99 pp.
Res. 29:1775-1783. CDIAC. 1994. Carbon emissions update, pl, and ‘Trends
Cannell, M. G. R. 1982. World Forest Biomass and Primary 93: A compendium of data on global change. p. 8. In
Production Data. Academic Press, New York. 391 pp. FE M. O’Hara, Jr. (ed.), CDIAC Communication. Carbon
. 1989. Climate warming and spring phenology of Dioxide Information Analysis Centre, World Data Centre—A
trees. pp. 283-295. In D. P. Lavender (ed.), Woody Plant for Atmospheric Trace Gases. Oak Ridge Nat. Lab., Fall
Growth in a Changing Chemical and Physical Environment. 1994 Issue, No. 20.
Faculty of Forestry, Univ. of British Columbia, Vancou- Chabot, B. F, and D. J. Hicks. 1982. The ecology of leaf life
ver. spans. Ann. Rev. Ecol. Syst. 13:229-259.
, J. Grace, and A. Booth. 1989. Possible impacts of Chandler, R. F,, Jr. 1942. The time required for podzol pro-
climatic warming on trees and forests in the United file formation as evidenced by the Mendenhall glacial
Kingdom: A review. Forestry 62:337-364. deposit near Juneau, Alaska. Soi Sci. Soc. Amer. Proc.
, and EF. T. Last. 1976. Tree Physiology and Yield Im- 7:454-459,
provement. Academic Press, New York. 506 pp. Chapin, F. S., IH. 1980. The mineral nutrition of wild
, and R. I. Smith. 1986. Climate warming, spring plants. Ann. Rev. Ecol. Syst. 11:233-260.
budburst and frost damage on trees. 7. Appl. Ecol. , and L. Moilanen. 1991. Nutritional controls over
23:177-191. nitrogen and phosphorus resorption from Alaskan birch
Carlisle, A. 1976. The Utilization of Forest Biomass and Forest leaves. Ecology 72:709-715.
Industry Wastes for the Production and Conservation of Ener- , and K. van Cleve. 1981. Plant nutrient absorption
gy. Dept. Environ. Can. For. Serv., Ottawa. 54 pp. and retention under different fire regimes. pp. 301-321.
R-8 | REFERENCES CITED

In Fire Regimes and Ecosystem Processes. USDA For. Serv., Smith and T. M. Hinckley (eds.), Ecophysiology of Conifer-
Gen. Tech. Rept. WO-26. ous Forests. Academic Press, New York.
Chapman, R. W. 1928. The geile analysis of environ- Clark, D. E1994. Post-Fire Succession in the Sub-Boreal
mental factors. Ezology 9:1 11-122 Spruce Forests of the Nechako Plateau, Central British Co-
Chatarpaul, L., and A. Carlisle. 1983. Nitrogen fixation: A lumbia. Master’s thesis, Univ. of Victoria, Victoria, BC,
biotechnological opportunity for Canadian forestry. For 124 p .
Chron. 39:249-259, Clark, ER. 1976. The central hardwood forest. pp. 1-8. Jn
Cheng, T. C. 1970. Sywrbiosis. Organisms Living Together. Pe- J. S. Fralish, R.C. Schleisinger, and G.T. Weaver (eds.),
gasus, New York. 250 pp. Central Hardwood Conference. USDA For. Serv. NC For.
Chen, R., R. R. Twilley, and R. H. Chen. 1998. A gap dy- Exp. Sta. St. Paul. Minn.
namic model of mangrove forest development along gra- Clark, EF. E., and T. Rosswall (eds.). 1981. Terrestrial Nitrogen
dients of soil salinity and nutrient resources. 7 Fxvl. Cyeles. Processes, Ecosystem Strategies and Management Im-
86:37-31. pacts. Ecol. Bull. No. 33. Swedish Nat. Sci. Res. Council.
Chew, R. M., and A. E. Chew. 1970. Energy relationships of 714
the mammals of a desert shrub. Ecol. Monogr: 40:1-21. Clark, W. C. (ed.). 1982. Carbon Dioxide Review: 1982.
Chippendale, H. G., and W. E. J. Milton. 1934. On the wi- Clarendon Press, Oxford. 469 pp.
able seeds present in the soil beneath pastures. 7. Ecol. . K. H. Cook, and G. Marland et al. 1982. The car-
22:508-531. bon dioxide question: Perspectives for 1982. pp. 3-53. In
Chitty, D. 1960. Population processes in the vole and their W. C. Clark (ed.), Carbon Dioxide Review: 1982. Claren-
relevance to general theory. Can. 7. Zool. 38:99-113. don Press, Oxford.
/
. 1967. The natural selection of self-regulatory behav- Clarke, C. H. D. 1936. Fluctuations in Numbers of Ruffed ‘
iour in animal populations. Proc. Ecol. Soc. Aust. 2351-78 Grouse, Bonsa umbellus (Linne.) With Special Reference to .
Christanty, L. 1989. Analysis of the Sustainability and Man- Ontario. Univ. Toronto Studies. Biol. Ser. No. 41. 118 pp. .
agement of the Talwn-Kebun System of West Java, In- Clarke, G. L. 1967. Elements of Ecology. 3rd ed. Wiley, New
donesia. Doctoral thesis, Univ. of B.C., Vancouver, B.C. York. 560 pp.
276 pp. Clausen, J. 1965. Population studies of alpine and subalpine
. J. BR.Kimmins, and D. Mailly. 1997 “Without bam- races of conifers and willows in the California high Sierra
boo, the land dies”: The role of bamboo in an Indonesian Nevada. Evolution 19:56-68.
agroforestry system. For: Ecol. Manage. 91:83-91,. , D. D. Keck, and W. M. Hiesey. 1948. Experimental
. D. Mailly, J. P. Kimmins, 1996, “Without bamboo, studies on the nature of species. II. Environmental re-
the land dies”: II. Biomass, litterfall, and soil organic sponses of climatic races of Achillea. Carnegie Inst. Wash.
matter dynamics of a Javanese bamboo talunkebun sys- Publ. 581:1-129.
tem. Fer. Ecol. Manage. 87:75-88. Clayton, J. L. 1976. Nutrient gains to adjacent ecosys-
Christensen, N. L. 1973. Fireand the nitrogen cycle in Cal- tems during a forest fire: An evaluation. For. Sci.
ifornia chaparral. Science 181:66-67. 27:162-166.
Christensen, N. I, A. M. Bartuska, J. H. Brown, S. R. Car- Cleaves, E.T., D. W. Fisher, and O. P. Bricker. 1974. Chem-
penter, C. D’Antonio, R. Francis, J. F Franklin, J. A. ical weathering of serpentine in the eastern Piedmont of
MacMahon, R. F Noss, D. J. Parsons, J. H. Peterson, Maryland. Geol. Soc. Amer. Bull. 85:437-+44.
M. G. Turner, and R. G. Woodmansee. 1996, The scientif- , A. E. Godfrey, and O. P. Bricker. 1970. Geochemi-
ic basis far ecosystem management. Ea. Appl. 6:665-691, cal balance of a small watershed and its geomorphic im-
Christersson, L., and R. Sandstedt. 1978. Short-term tem- plications. Geol. Soc. Amer. Bull. 81:3015-3032.
perature variation in needles of Pins syvestris. L. Can. F. Clements, F. E. 1916. Plant succession. An analysis of the
Fer. Res. SAS0—-482. development of vegetation. Carnegie Inst. Wash. Publ.
Christian, J.J.1950, The adreno-pituitary system and popu- 242, 512 pp.
lation cycles in mammals. 7. Mamem. 31:247-259. Coale, A. J. 1974. The history of the human population. Sci.
. 1961. Phenomena associated with population densi- Amer. 231(3):41-51.
ty. Pree. Natl. daad. Sci, 47 428-448. # Cochran, P. H. 1969a. Lodgepole Pine Clearcut Size Affects
~. 1971, Population density and reproductive efficien- Minimum Temperatures Near the Soil Surface. USDA For.
cy. Biol, Reprad. 4:248-294. Serv. Pac. Northwest For. Range Exp. Sta. Res. Paper
. and D. E. Davis. 1964. Endocrines, behavior and PNW-26. 9 pp.
population. Same 146:1550-1560. - 1969b. Thermal Properties and Surface Temperatures of
Clancy, K. M.. M. R. Wagner, and P. B. Reich. 1995. Eco- Seedbeds. USDA For. Serv. Pac. Northwest For. Range
physiology and insect herbivory. pp. 125-180. Jn W. K. Exp. Sta. Misc. Publ. 12. 19 pp.
 REFERENCES CITED R-9

Cody, M. L. 1966. A general theory of clutch size. Evolution Coley, P. D. 1983. Herbivory and defense characteristics of
20:174-184. tree species in a lowland tropical forest. Ecol. Monogr.
Coffin, D. P., and W. K. Lauenroth. 1989. Disturbances and 53:209-233.
gap dynamics in a semiarid grassland: A landscape-level Collier, B. D., G. W. Cox, A. W. Johnson, and P. C. Miller.
approach. Landscape Ecol. 3:19-27. 1973. Dynamic Ecology. Prentice-Hall, Englewood Cliffs,
Coffin, D. P., and W. K. Lauenroth. 1989. Disturbances and NJ. 563 pp.
gap dynamics in a semiarid grassland: A landscape-level Comeau, P. G., andJ.P. Kimmins. 1986. The relationship be-
approach. Landscape Ecol. 3:19-27. tween net primary production and foliage nitrogen con-
Cole, D. W., W. J. B. Crane, and C. C. Grier. 1975. The ef- tent, and its application in modelling of forest ecosystems:
fect of forest management practices on water chemistry A study of lodgepole pine (Pinus contorta). pp. 202-223. In
in a second-growth Douglas-fir ecosystem. pp. 195-207. T. Fujimori and D. Whitehead (eds.), Crown and Canopy
In B. Bernier and C. H. Winget (eds.), Forest Soils and Structure in Relation to Productivity. Proc. [UFRO Symp.
Forest Land Management. Laval Univ. Press, Quebec. Workshop, For. Prod. Inst., Ibaraki, Japan.
, and S. P. Gessel, 1963. Movement of elements , andJ.P. Kimmins. 1989. Above- and below-ground
through a forest soil as influenced by tree removal and biomass and production of lodgepole pine on sites of dif-
fertilizer additions. pp. 95-104. In C. T. Youngberg (ed.), fering soil moisture regimes. Can. 7. For. Res. 19:447-454.
Forest Soil Relationships in North America. Oregon State Cumming, S. G., and P. J. Burton. 1994. Phenology-mediat-
Univ. Press, Corvallis. ed effects of climatic change on some British Columbia
, 5. P. Gessel, and S. F. Dice. 1967. Distribution and forests. Clim. Change 34:213-222.
cycling of nitrogen, phosphorus, potassium and calcium Connell, J. H. 1972. Community interactions on marine
in a second-growth Douglas-fir ecosystem. pp. 197-232. rocky intertidal shores. Ann. Rev. Ecol. Syst. 3:169-192.
In Symposium on Primary Production and Mineral Cycling in , and M. D. Lowman. 1989. Low diversity tropical
Natural Ecosystems. Univ. of Maine Press, Orono. rain forests: Some possible mechanisms for their exis-
, C. L. Henry, and W. L. Nutter (eds.). 1986. The For- tence. Amer. Nat. 134:88-119.
est Alternative of Treatment and Application ofMunicipal and , I. R. Noble, and R. O. Slatyer. 1987. On the mech-
Industrial Wastes. Univ. of Washington Press, Seattle, WA. anisms producing successional change. Oikos 50:136-137.
582 pp. , and R. O. Slatyer. 1977. Mechanisms of succession
, C. L. Henry, P. Schiess, and R. J. Zasoski. 1983. The in natural communities and their role in community sta-
role of forests in sludge and wastewater utilization pro- bility and organization. Amer. Nat. 111:1119-1144.
grams. pp. 125-143. In A. L. Page (ed.), Proc. 1983 Work- Cook, C., R. R. Askew, and J. A. Bishop. 1970. Increasing
shop on Utilization of Municipal Wastewater and Sludge on frequency of the typical form of the Peppered Moth in
Land. Univ. Calif., Riverside. Manchester. Nature 227:1155.
, and D. W.Johnson. 1980. Mineral cycling in tropi- Cooper, C. EF 1961. The ecology of fire. Sci. Amer.
cal forests. pp. 341-356. In C. T. Youngberg (ed.), Forest 204(4):150-160.
Soils and Land Use. Proc. 5th N. Amer. For. Soils Conf., Cooper, R. W. 1973. Trade-Offs Between Smoke From Wild
Dept. For. and Wood Sci., Colorado State Univ., Fort and Prescribed Fires. Proc. Int. Symp. Air Qual. and
Collins. Colo. Smoke from Urban and Forest Fires. Ft. Collins, Colo.
, and M. Rapp. 1981. Elemental cycling in forest Oct. 23-26, 1973.
ecosystems. pp. 341-409. In D. E. Reichle (ed.), Dynamic Cordes, L. D. 1972. An Ecological Study of the Sitka Spruce
Properties of Forest Ecosystems. International Biological Forest on the West Coast of Vancouver Island. Doctoral
Programme 23. Cambridge Univ. Press, Cambridge. thesis, Univ. ofB. C., Vancouver. 298 pp.
, and P. Schiess. 1978. Renovation of wastewater and Corke, C. T. 1958. Nitrogen transformations in Ontario for-
response of forest ecosystems. The Pack Forest Study. est podsols. pp. 116-121. In Proc. Ist N. American For. Soils
pp. 323-332. In H. L. McKim (ed.), State of Knowledge in Conf., Mich. State Univ. Agric. Exp. Stat., East Lansing.
Land Treatment of Wastewater. Vol. 1. U.S. Army Corps of Cornforth, I. S. 1970. Reafforestation and nutrient reserves
Engr. Cold Regions Res. and Engr. Lab., Hanover, NH. in the humid tropics.7.Appl. Ecol. 7:609-615.
, J. Turner, and C. Bledsoe. 1974. Requirement and Costanza, R., B. G. Norton, and B. D. Haskell (eds.). 1992.
uptake of mineral nutrients in coniferous ecosystems. pp. Ecosystem Health. Island Press, Washington, DC. 269 pp.
171-176. InJ. K. Marshall (ed.), The Below Ground Ecosys- Council on Environmental Quality. 1980. The Global 2000
tem. Dowden, Hutchinson and Ross, Stroudsberg, PA. Report to the President: Entering the Twenty-first Century.
Cole, L. C. 1957. Sketches of general and comparative de- Govt. Printing Office, Washington, DC. 766 pp.
mography. Cold Spring Harbour Symp. Quant. Biol. Cousens, J. 1974. An Introduction to Woodland Ecology. Oliver
22:1-15. and Boyd, Edinburgh. 151 pp.
R-10 REFERENCES CITED
«ee
te
ill
_—

Covington, W. W. 1976. Forest Floor Organic Matter and particularly in the Douglas-fir region. pp. 137-178. In
Nutrient Content and Leaf Fall During Secondary Suc- R. K. Hermann and D. P. Lavender (eds.), Even-Age
cession in Northern Hardwoods. Doctoral thesis, Yale Management, Paper 848. Oregon State Univ., Corvallis.
Univ., New Haven. 98 pp. Curtis, J. P. 1959. The Vegetation of Wisconsin. Univ. of Wis-
, and L. F. DeBano. (eds). 1994. Sustainable Ecological consin Press, Madison. 657 pp.
Systems: Implementing an Ecological Approach to Land Man- , and R. P. McIntosh. 1951. An upland forest contin-
agement. USDA For. Serv. Gen. Tech. Rept. RM-247. uum in the prairie-forest border region of Wisconsin.
Fort Collins, Co. Ecology 32:476-496.
, and M. M. Moore. 1994. Post-settlement changes Dale, V. H., T. W. Doyle, and H. H. Shugart. 1985. A com-
in natural fire regimes and forest structure: Ecological parison of tree growth models. Ecol. Model. 29:145—-169.
restoration of old-growth ponderosa pine forests. pp. , and M. Hemstrom. 1984. CLIMACS: A computer
153-181. In R. N. Sampson and D. L. Adams (eds.), model of forest stand development for western Oregon and
Assessing Forest Ecosystem Health in the Inland West. Ha- Washington. USDA For. Serv., Res. Pap. PNW-327.
worth Press, New York. , R. A. Houghton, and C. A. S. Hall. 1991. Estimat-
Cowan, R. L. 1962. Physiology of nutrition as. related to ing the effects of land-use change on global atmospheric
deer. pp. 1-8. In Proc. Ist National White-tailed Deer Dis- CO, concentrations. Can. 7. For. Res. 21:87-90.
ease Symposium. University of Georgia, Athens, Georgia. Dale, V. H., and M. Hemstrom. 1984. CLIMACS: A com-
Cowles, H. C. 1899. The ecological relations of the vegeta- puter model of forest stand development for western
tion of the sand dunes of Lake Michigan. Bot. Gaz. Oregon and Washington Res. Pap. PNW-327. USDA
27:95-117, 167-202, 281-308, 361-391. Forest Service, Washington, DC.
Cramer, O. P. 1974a. Air quality influences. In O. P. Cramer Dalton, D. C. 1963. Effect of dilution of the diet with an in-
(ed.), Environmental Effects of Forest Residues Management digestible filler on feed intake in the mouse. Nature
in the Pacific Northwest: A State-of-Knowledge Compendium. 197:909-910.
USDA For. Serv. Gen. Tech. Rept. PNW-24. Damman, A. W. H. 1971. Effect of vegetation changes on
. 1974b. Environmental Effects of Forest Residues Man- the fertility of a Newfoundland forest site. Ecol. Monogr:
agement in the Pacific Northwest. A State-of-Knowledge Com- 41:253-270.
pendium. USDA For. Serv. Gen. Tech. Rept. PNW-24. . 1977. The role of vegetation analysis in land classi-
Crane, W.J. B., andJ.C. G. Banks. 1992. Accumulation and fication. pp. 169-193. In Ecological Classification of Forest
retranslocation of foliar nitrogen in fertilized and irrigat- Land in Canada and Northwestern USA. Proc. Center for
ed Pinus radiata. For. Ecol. Manage. 52:201-223. Cont. Ed., Univ. of B. C., Vancouver.
Crawford, H. S. 1975. Soil-site and forest land management Daniel, T. W.,J.A. Helms, and F. S. Baker. 1979. Principles of
in relation to wildlife. pp. 571-581. In B. Bernier and Silviculture. 2nd ed. McGraw-Hill, New York. 500 pp.
C. H. Winget (eds.), Forest Soils and Forest Land Manage- Dansereau, P. 1951. Description and recording ofvegetation
ment. Laval Univ. Press, Quebec. upon a structural basis. Ecology 32:172-229.
Crisp, D. T. 1966. Input and output of minerals for an area of . 1957. Biogeography, an Ecological Perspective. Ronald,
Pennine moorland: The importance of precipitation, New York. 394 pp.
drainage, peat erosion and animals.7.Appl. Ecol. 3:327-348. Darling, F. F. 1938. Bird Flocks and the Breeding Cycle. A Con-
Crocker, R. L., andJ.Major. 1955. Soil development in rela- tribution to the Study of Avian Sociality. Cambridge Univ.
tion to vegetation and surface age of Glacier Bay, Alaska. Press, Cambridge. 124 pp.
J. Ecol. 43:427-448. Darlington, P. J. Jr., 1938. The origin of the fauna of the
Crutzen, P. J., and J. G. Goldammer (eds). 1993. Fire in the Greater Antilles, with discussion of dispersal of animals
Environment. The Ecological, Atmospheric, and Climatic Im- over water and through the air. Quart. Rev. Biol.
portance of Vegetation Fires. Wiley, New York. 400 pp. 13:274-300.
Cumming, S. G., and P. J. Burton. 1995. Phenology-mediat- Darwin, C. R. 1859. On the Origin ofSpecies by Means of Nat-
ed effects of climatic change on some British Columbia ural Selection, or the Preservation of Favoured Races in the
forests. Climatic Change 34:213-222. Struggle for Life. J. Murray, London. 502 pp.
Cummins, K. W. 1980. The multiple linkages of forests to Daubenmire, R. F 1952. Forest vegetation of northern
streams. pp. 191-198. In R. H. Waring (ed.), Forests: Idaho and adjacent Washington, and its bearing on con-
Fresh Perspectives from Ecosystem Analysis. Oregon State cepts of vegetation classification. Ecol. Monogr:
Univ. Press, Corvallis. 22:301-330.
Cunningham, P. M. 1963. The effect of clearing a tropical . 1954. Alpine timberlines in the Americas and their
soil.7.Soil Sci. 14:334-345. interpretation. Butler Univ. Bot. Stud. 11:119-136.
Curry, R. R. 1972. Geological and hydrological effects of - 1968. Plant Communities: A Textbook of Plant Synecol-
even-age management on productivity of forest soils, ogy. Harper & Row, New York. 300 pp.
 REFERENCES CITED R-11

. 1974. Plants and Environment: A Textbook of Plant DeByle, N. V. 1973. Broadcast burning of logging residues
Autecology. Wiley, New York. 422 pp. and the water repellency of soils. Northwest Sci. 47:77-87.
. 1976. The use of vegetation in assessing the produc- de Candolle, A. P. A. 1874. Constitution dans le Régne
tivity of forest lands. Bot. Rev. 42:115-143. Végétal des Groupes Physiologiques Applicables a la
, and J. B. Daubenmire. 1968. Forest vegetation of Géographie Ancienne et Moderne. Archives des Sciences
eastern Washington and northern Idaho. Wash. Agric. Physiques et Naturelles, Geneva.
Exp. Sta. Tech. Bull. 60. 104 pp. Deevey, E. S.,Jr.1947. Life tables for natural populations of
Davidson, E. A., D. D. Myrold, and P. M. Groffman. 1990. animals. Quart. Rev. Biol. 22:283-314.
Denitrification in temperate forest ecosystems. pp. 1960. The human population. Sci. Amer.
196-220. In S. P. Gessel, D. S. Lacate, G. F. Weetman, 203:195-204.
and R. F. Powers (eds.), Sustained Productivity of Forest . 1970. Mineral cycles. Sci. Amer. 223:149-158.
Soils. Proc. 7th N. Am. For. Soils Conf., Faculty of . 1973. Sulphur, nitrogen and carbon in the bios-
Forestry, Univ. of British Columbia, Vancouver. phere. pp. 182-190. In G. M. Woodwell and E. V. Pecan
Davis, G. K. 1968. Mineral elements in the nutrition of larg- (eds.), Carbon in the Biosphere. AEC. Symp. Series 30.
er mammals. Amer: Zool. 8:169-174. Tech. Info. Centre, Office of Info. Services. USAEC.
Davis, M. B. 1986. Climatic instability, time lags, and com- del Arco, J. M., A. Escudero, and M. V. Garrido. 1991. Ef-
munity disequilibrium. pp. 269-284. In J. M. Diamond fects of site characteristics on nitrogen retranslocation
and T. J. Case (eds.), Community Ecology. Harper & Row, from senescing leaves. Ecology 72:701-708.
New York. Dell, J. D., and F. R. Ward. 1971. Logging Residues on Doug-
. 1987. Invasions of forest communities during the las-Fir Region Clearcuts—Weights and Volumes. USDA For.
holocene: Beech and hemlock in the Great Lakes. pp. Serv. Res. Pap. PNW-115.
373-393. In A. J. Gray, M. J. Crawley, and P. J. Edwards DeLury, R. 1932. Astronomical periods in climate and life.
(eds.), Colonization, Succession and Stability. Blackwell Sci. pp. 213-221. In E. Huntington (ed.), Matamek Confer-
Publ., Oxford. ence on Biological Cycles. Matamek Factory, Canadian
. 1989. Insights from paleoecology on global change. Labrador.
Bull. Ecol. Soc. Am. 70:222-228. Delwiche, C. C. (ed.). 1981. Denitrification, Nitrification and
Davis, R. B. 1966. Spruce-fir forests of the coast of Maine. Atmospheric Nitrous Oxide. Wiley, New York. 286 pp.
Ecol. Monogr. 36:79-94. de Montigny, L. E., and G. F, Weetman. 1990. The effects
Davis, T. W. A., and P. W. Richards. 1933. The vegetation of of ericaceous plants on forest productivity. Jn B. D. Titus,
Moraballi Creek, British Guiana: An ecological study of a M. B. Lavigne, P. F Newton, and W. J. Meades (eds.),
limited area of tropical rain forest. 7. Ecol. 21:350-384. The Silvics and Ecology of Boreal Spruces. Proc. \UFRO
, and P. W. Richards. 1934. The vegetation of Mora- Working Party S.05-12 Ann. Meeting, Newfoundland.
balli Creek, British Guiana: An ecological study of a lim- Forestry Canada, Info. Rept. N-X-271.
ited area of tropical rain forest. 7. Eco/. 22:106-155. Denslow, J. S. 1985. Disturbance-mediated coexistence of
Dawson, T. E. 1993. Hydraulic lift and water use by plants: species. pp. 307-324. In S. T. A. Pickett and P. White
Implications for water balance, performance and plant- (eds.), The Ecology of Natural Disturbance and Patch Dy-
plant interactions. Oecologia 95:565—-574. namics. Academic Press, Orlando, Fla.
DeAngelis, D. L., R. H. Gardner, and H. H. Shugart, Jr. . 1987. Tropical rainforest gaps and tree species di-
1981. Productivity of forest ecosystems studied during versity. Ann. Rev. Ecol. Syst. 18:431-451.
IBP: The woodlands data set. pp. 567-672. In D. E. D’Eon, R. G. 2002. Landscape spatial Pattern and Forest Frag-
Reichle (ed.), Dynamic Properties of Forest Ecosystems. mentation in Managed Forests in Southeastern British Colum-
Cambridge Univ. Press, Cambridge. bia: Perceptions, measurements, and Scale. Doctoral thesis,
DeBach, P., and R. A. Sundby, 1963. Competitive displace- University of British Columbia, Vancouver, B.C. 107.
ment between ecological homologues. Hilgardia Desanker, P. V., and I. C. Prentice. 1994. MIOMBO—A
34:105-166. vegetation dynamics model for the miombo woodlands
DeBano, L. FE, J. F Osborn,J.S. Krammes, andJ. Letey, Jr. of Zambezian Africa. For. Ecol. Manage. 69:87-95.
1967. Soil Wettability and Wetting Agents: Our Current Desanker, P. V., and I. C. Prentice. 1994. MIOMBO: A veg-
Knowledge of the Problem. USDA For. Serv. Res. Pap. etation dynamics model for the miombo woodlands of
PSW-43. 13 pp. Zambezian Africa. For. Ecol. Manage. 69:87-95.
, and R. M. Rice. 1973. Water-repellent soils: their Desjardins, T., F Andreux, B. Volkoff, and C. C. Cerri.
implications in forestry. 7. For: 71:220-223. 1994. Organic Carbon and '°C contents in soils and soil
Debell, D. S., and C. W. Ralston. 1970. Release of nitrogen size-fractions, and their changes due to deforestation and
by burning light forest fuels. Soil Sci Soc. Amer. Proc. pasture installation in eastern Amazonia. Geoderma 61:
34:936-938. 103-118.
R-12 REFERENCES CITED

Devall, B., and G. Sessions, 1985. Deep Ecology. Peregrine Dominski, A. 1971. Accelerated nitrate production and loss
Smith, Salt Lake City, UT. in the northern hardwood forest ecosystem underlain
DeVelice, R. L. 1988. Test of a forest dynamics simulator in by podzol soils following clearcutting and addition of
New Zealand. N. Z. 7. Bot. 26:387-392. herbicides. Doctoral thesis, Yale Univ., New Haven.
DeVelice, R. L. 1988. Test of a forest dynamics simulator in 157 pp.
New Zealand. N. Z. 7. Bot. 26:387-392. Dorn, H. F. 1962. World population growth: an internation-
Development Forum. 1982. U.N. University and Div. Econ. al dilemma. Science 135:283-290. —

and Social Info. 5(10), p. 8. Douglas, A. E. 1919. Climatic Cycles and Tree Growth. Vol. 1. A
de Wit, C. T., and J. Goudriaan. 1974. Simulation of Ecologi- Study ofthe Annual Rings of Trees in Relation to Climate and
cal Processes. Centre for Agricultural Publishing and Doc- Solar Activity. Carnegie Inst. Wash. Publ. 289. 127 pp.
umentation, Wageningen, Netherlands. Downs, R. J. 1962. Photocontrol of growth and dormancy in
DeYoe, D. R., and K. Cromack. 1983. Mycorrhizae—A hid- woody plants. pp. 133-148. In T. T: Kozlowski (ed.), Tree s
den benefactor to forest trees. Oregon State University Growth. Ronald Press, New York.
Extension Service, EM 8247. 10 pp. Doyle, T. W. 1981. The role of disturbance in the gap dy-
Dickerson, R. E. 1978. Chemical evolution and the origin of namics of a montane rain forest: An application of a trop-
life. Set. Amer. 239(3):70-86. ical forest succession model. pp. 56-73. In D. C. West,
Dickinson, R. E. 1982. Modeling climatic changes due to car- H. H. Shugart, and D. B. Botkin (eds.), Forest Succession.
bon dioxide increases. pp. 103-142. Jn W. C. Clark (ed.), Springer-Verlag, Berlin.
Carbon Dioxide Review: 1982. Clarendon Press, Oxford. . 1982. The role of disturbance in the gap dynamics
Dickson, W. 1975. The acidification of Swedish Lakes. In- of a montane rain forest: An application of a tropical for-
stitute of Freshwater Res. Rept. No. 54. 20 pp. est succession model. pp. 56-73. In R. H. Waring (ed.),
Diebold, C. H. 1942. Effect of fire and logging upon the Forests: Fresh Perspectives from Ecosystem Analysis. Oregon
depth of the forest floor in the Adirondack region. Soi/ State Univ. Press, Corvallis.
Set. Soc. Amer. Proc. 6409-413. , R. S. Meldahl, G. A. Ruark, and W. G. Warren.
Dittmer, H. J. 1937. A quantitative study of the roots and 1990. Process Modeling of Forest Growth Responses to Envi-
root hairs of a winter rye plant (Secale cereale). Amer. F. ronmental Stress. Timber Press, Portland, Ore. 441 pp.
Bot. 24:417-420. Doyle, T. W. 1981. The role of disturbance in the gap dy-
Ditzer, T., A. Huth, H. Bossel, and J. Pfadenhauer. 1996. namics of a montane rain forest: An application of a trop-
Simulation des Wachstums von Dipterocarpaceen- ical forest succession model. pp. 56-73. In D. C. West,
Regenwald in Malaysia zur Bewertung von Nutzungsver- H. H. Shugart, and D. B. Botkin (eds.), Forest Succession.
fahren. lerhandlung. Gesellsch. Okol. 26:425-434. Springer-Verlag, New York.
Ditzer, T., R. Glauner, M. Forster, P. Kohler, and A. Huth. Drablos, D., and A. Tollan (eds.). 1980. Ecological Impact of
2000. The process-based stand growth model Formix Acid Precipitation. Proc. Internat. Conf. Sandford, Nor-
3-Q applied in a GIS environment for growth and yield way. SWSF Project Oslo, Norway.
analysis in a tropical rain forest. Jive Physiol. 20:367-381. Dressler, R. L. 1968. Pollination by euglossine bees.
Dix, R. L. 1960. The effects of burning on the mulch struc- Evolution 22:202-210.
ture and species composition of grasslands in western Drew, T. ]., andJ.W. Flewelling. 1977. Some recent Japan-
North Dakota. Eeolegy 1438-45. ese theories of yield-density relationships and their appli-
Dixon, R. K., S. Brown, R. A. Houghton, A. M. Solomon, cation to Monterey pine plantations. For Sci.
M. C. Trexler, and J. Wisniewski. 1994. Carbon pools 23:517-534.
and flux of global forest ecosystems. Science 263:185-190. , and J. W. Flewelling. 1979. Stand density manage-
. R. S. Meldahl, G. A. Ruark, and W. G. Warren. ment: An alternative approach and its application to
1990. Process Modeling of Forest Growth Responses to Envi- Douglas-fir plantations. For: Sci. 25:518-532.
ronmental Stress. Timber Press, Portland, Ore. +41 pp. Drobikovy, A. A. 1969. [Changes in the water regime of brown
Dobzansky, T. 1950. Evolution in the tropics. Amer: Sei. soils under the influence of fellings]. Pocvoved 6:54-62.
38:209-221, Drosdoff, M., R. B. Daniels, and J. J. Nicholaides. 1978.
Dochinger, L. S., and T. A. Seliga (eds.). 1976. Proceedings of Diversity ofSoils in the Tropics. Amer. Soc. Agronomy, Soil
the First International Symposium on Acid Precipitation and Sci. Soc. Amer. ASA Special Publ. No. 34. Madison, WI.
the Forest Ecosystem. USDA For. Serv. NE. For. Exp. Sta. 119 pp.
Gen. Tech. Rept. NE-23. 1072 pp. Drury, W. H. and I. C. T. Nisbet. 1973. Succession. 7.
Dodge, M. 1972. Forest fuel accumulation—A growing Arnold Arboretum, 54:331-368.
problem. Science 177:139-142. Duddridge, J. A., A. Malibari, and D. J. Read. 1980. Struc-
Dodson, C. H., R. L. Dressler, H. G. Hills, R. M. Adams, ture and function of mycorrhizal rhizomorphs with spe-
and N. H. Williams. 1969. Biologically active com- cial reference to their role in water transport. Nature
pounds in orchid fragrances. Science 164:1243-1249. 287:834-836.
 REFERENCES CITED R-13

Dunn, C. P., G. R. Guntenspergen, and J. R. Dorrey. 1983. Doeksen and J. van der Drift (eds.), Soil Organisms.
Catastrophic wind disturbance in an old-growth hemlock North-Holland Publ., Amsterdam.
hardwood forest, Wisconsin. Can. 7. Bot. 61:211-217. , D. E. Reichle, and D. A. Crossley, Jr. 1970. The
Durzan, D. J., and F. C. Steward. 1967. The nitrogen me- role of soil invertebrates in turnover of organic matter
tabolism of Picea glauca (Moench) Voss and Pinus and nutrients. pp. 145-172. In D. E. Reichle (ed.),
banksiana Lamb as influenced by mineral nutrition. Can. Analysis of Temperate Forest Ecosystems. Springer-Verlag,
f. Bot. 45:695-710. New York.
Dutch, J., and R. Walstenholme. 1994. The effects of sludge Edwards, M. V., J. Atterson, and R. S. Howell. 1963. Wind-
application to a heathland site prior to planting with Loosening of Young Trees on Upland Heaths. Forestry Com-
Sitka spruce. For. Ecol. Manage. 66:151-163. mission: Forest Record #50. London. 16 pp.
Duvall, V. L., and L. B. Whittaker. 1964. Rotation burning: Edwards, N. T., H. H. Shugart, Jr., S. B. McLauglin, W. F.
a forage management system for longleaf pine-bluestem Harris, and D. E. Reichle. 1981. Carbon metabolism in
ranges.7.Range Manage. 17:322-326. terrestrial ecosystems. pp. 499-536. In D. E. Reichle
Duvigneaud, P. 1971. Concepts sur la productivité primaire (ed.), Dynamic Properties of Forest Ecosystems. Cambridge
des écosystémes forestiers. pp. 111-140. In P. Duvigneaud Univ. Press, Cambridge.
(ed.), Productivity ofForest Ecosystems. UNESCO, Paris. Egerton, F. N. 1968. Studies of animal populations from
, and S. Denaeyer-De Smet. 1970. Biological cycling Lamarck to Darwin.7. Hist. Biol. 1:225-259.
of minerals in temperate deciduous forests. pp. 199-225. Egler, F. E. 1953. Vegetation management for rights-of-way
In D. E. Reichle (ed.), Analysis of Temperate Forest Ecosys- and roadsides. pp. 299-320. In Smithsonian Inst. Ann.
tems. Springer-Verlag, New York. Rept. Smithsonian, Washington, D.C.
Dyck, W. J., D. W. Cole, and N. B. Comerford (eds.) 1994. . 1954. Vegetation science concepts. I. Initial floristic
Impacts of Forest Harvesting on Long-term Site Productivity. composition—A factor in old-field vegetation develop-
Chapman & Hall, New York. 371 pp. ment. Vegetatio 4:412-418.
Dyck, W. J., and C. A. Mees. 1989. Research Strategies for Long- Ehhalt, D. H. 1973. Methane in the atmosphere. pp.
Term Site Productivity. IEA/BE A3 Rept. No. 8. For. Res. 144-158. In G. M. Woodwell and E. V. Pecan (eds.),
Inst., Rotorua, New Zealand. FRI Bull. No. 152.257 pp. Carbon and the Biosphere. AEC Symp. Series 30. Tech.
Dyck, W. J., and C. A. Mees. 1990. Impact of Intensive Har- Info. Centre, Office of Info. Serv. USAEC.
vesting on Forest Site Productivity. IEA/BE AE Rept. No. Ehrlich, P. R., and P. H. Raven. 1967. Butterflies and plants.
2. For. Res. Inst., Rotorua, New Zealand. FRI Bull. No. Sct. Amer. 216:104-113.
159. 175 pp. Eis, S. 1962. Statistical analysis of several methods for esti-
Dyck, W. J., and C. A. Mees. 1991. Long-term Field Trials to mation of forest habitats and tree growth near Vancou-
Assess Environmental Impacts of Harvesting. TEA/BE ver, B.C. For. Bull. No. 4, Fac. For., Univ. of B.C.,
T6/A6 Rept. No. 5. For. Res. Inst., Rotorua, New Vancouver.
Zealand. FRI Bull. No. 161. 215 pp. Ek, A. R., and R. A. Monserud. 1974. FOREST: A Comput-
Dyrness, C. T. 1965. Soil surface condition following tractor er Model for Simulating the Growth and Reproduction
and high-lead logging in the Oregon Cascade. 7. For. of Mixed Forest Stands. Res. Rept. A2635. Coll. Agric.
63:272-275. and Life Sciences. Univ. of Wisconsin, Madison. 14 pp.
. 1967. Mass Soil Movements in the H. 7. Andrews Ex- Ekern, P. C. 1964. Direct interception of cloud water at
perimental Forest. USDA For. Serv. Res. Pap. PNW-42. Lanaihole, Hawaii. Soil Sci. Soc. Amer. Proc. 28:419-421.
12 pp. E] Bayoumi, M. A., H. H. Shugart, and R. W. Wein. 1984.
Eaton, J. S., G. E. Likens, and F. H. Bormann. 1973. Modeling succession of eastern Canadian mixedwood
Throughfall and stemflow chemistry in a northern hard- forest. Ecol. Model. 21:175-198.
wood forest. 7. Ecol. 61:495—-508. E] Bayoumi, M. A., H. H. Shugart, and R. W. Wein. 1984.
Ebell, L. F. 1972. Cone-induction response of Douglas-fir to Modelling succession of eastern Canadian mixedwood
form of nitrogen fertilizer and time of treatment. Can. 7. forest. Ecol. Model. 21:175-198.
For. Res. 2:317-326. Elliot, K. J., and A. S. White. 1993. Effects of competition
, and E. E. McMullen. 1970. Nitrogenous substances from young northern hardwoods on red pine seedlings
associated with differential cone production responses of growth, nutrient use efficiency, and leaf morphology. For:
Douglas-fir to ammonium and nitrate fertilization. Can. Ecol. Manage. 57:232-255.
7. Bot. 48:2169-2177. Elton, C. 1924. Periodic fluctuations in the numbers of ani-
Ebermeyer, E. 1826. Die gesammte Lehre der Waldstreu mut mals: their causes and effects. Br. 7. Exp. Biol. 2:119-163.
Riicksicht auf die chemische Statik des Waldbaues. J. . 1927. Animal Ecology. Sedgwick and Jackson, Lon-
Springer-Verlag, Berlin. don. 207 pp.
Edwards, C. A., and G. W. Heath. 1963. The role of soil an- , and M. Nicholson. 1942. The ten-year cycle of
imals in breakdown of leaf material. pp. 76-84. In J. numbers of lynx in Canada.7.Anim. Ecol. 11:215—244.
R-14 REFERENCES CITED

Endler, J. A. 1993. The colour of light in forests and its im- . 1970. Seasonal changes in oak leaf tannins and nu-
plications. Ecol. Monogr. 63:1-27. trients as a cause of spring feeding by winter moth cater-
Epstein, E. 1972. Mineral Nutrition of Plants: Principles and pillars. Ecology 51:565-581.
Perspectives. Wiley, New York. 412 pp. , and H. Bostock. 1968. Seasonal changes in the tan-
Erby, 2002. The Earth Radiation Budget Experiment. nin content of oak leaves. Phytochemistry 7:87 1-880.
NASA Atmospheric Sciences. Available at: asd- Feller, M. C. 1974. Initial effects of clearcutting on the flow
www.larc.nasa.gov/erbe/ASDerbe.html of chemicals through a forest-watershed ecosystem in
Ericsson, T. 1994. Nutrient dynamics and requirements of southwestern British Columbia. Doctoral thesis, Univ. of
forest crops. N. Z. 7. For. Sci. 24:133-168. B.C., Vancouver.
Errington, P. L. 1963. Muskrat Populations. lowa State Univ. . 1982. The Ecological Effects of Slashburning
Press, Ames, Iowa. 665 pp. With Particular Reference to British Columbia: A Lit-
Esau, K. 1963. Plant Anatomy. Sth ed. Wiley, New York. 735 pp. erature Review. Land Management Report No. 23.
Escudero, A.,J.M. del Arco, L. C. Sanz, andJ.Ayala. 1992. ISSN 0702-9861. Ministry of Forests. Province of B.C.,
Effects of leaf longevity and retranslocation efficiency on Victoria.
the retention time of nutrients in the leaf biomass of dif- , and E. H. Hamilton. 1994. Effects of Slashburning
ferent woody species. Occologia 90:80-87. and Mechanical Site Preparation on Nutrient Status,
Eswaran, H., F. H. Beinroth,J.Kimble, and T: Cook. 1992. Plant Competition and Tree Growth in Engelmann
Soil diversity in the tropics: Implications for agricultural Spruce-Subalpine Fir Ecosystems. Res. Report to Re-
development. pp. 1-16. Jn R. Lal and P. A. Sanchez (eds.), search Branch, B.C. Min. Forests, Victoria, BC. Unpubl.
Myths and Science of Soils in the Tropics. Soil Sci. Soc. 186 pp.
Amer., Amer. Soc. Agronomy, SSSA Special Publ. No. , andJ. P. Kimmins. 1984. Effects of clearcutting and
29, Madison, WI. slashburning on streamwater chemistry and watershed
Evans, F. C. 1956. Ecosystem as the basic unit in ecology. nutrient budgets in southwestern British Columbia.
Science 123:1127-1128. Water Resource Res. 20:29-40.
Evans, G. C. 1956. An area survey method of investigating , J. P. Kimmins, and K. M. Tsze. 1983. Nutrient loss-
the distribution of light intensity in woodlands with par- es to the atmosphere during slashburns in southwestern
ticular reference to sunflecks, including an analysis of British Columbia. pp. 128-132. In Proc. 7th Conf. Fire and
data from rain forest in Southern Nigeria. 7. Ecol. Meteorology. Amer. Met. Soc. Boston, Mass.
44:391-428. Ferreira, M., and M. das Gragas. 1984. An analysis of the fu-
Evans, J. 1990. Long-term productivity of forest planta- ture productivity of Eucalyptus grandis plantations in the
tions—Status in 1990. IUFRO XIX World Congress, “Cerrado” region in Brazil: A nutrient cycling approach.
Montreal, Vol. 1:165—180. Doctoral thesis. Univ. of British Columbia, Vancouver.
. 1992. Plantation Forestry in the Tropics, 2nd ed. 230 pp.
Clarendon Press, Oxford, UK. Ferreira, M. G. M., J. P. Kimmins, and N. F Barros. 1984.
Everett, R. L. (comp.). 1994. Eastside Forest Ecosystem Impact of intensive management on phosphorus cycling
Health Assessment. Vol. IV. Restoration of Stressed in Eucalyptus grandis plantations in the savannah region,
Sites, and Processes. USDA For. Serv. Gen. Tech. Rept. Minas Gerais, Brazil. pp. 847-856. In Proc. Symp. on Site
PNW-GTR-330. Portland, Ore. 123 pp. Productivity of Fast Grown Plantations. YOFRO, Pretoria
Ewel,J.,C. Berish, B. Brown, N. Price, andJ. Raich. 1981. and Pieter Maritzburg, South Africa, Vol. 1.
Slash and burn impacts on a Costa Rican wet forest site. Fischer, A. G. 1960. Latitudinal variations in organic diver-
Ecology 62:816-829. sity. Evolution 14:64-68.
Eyre, F. H. (ed.). 1980. Forest Cover Types of the United States Fischlin, A., H. Bugmann, and D. Gyalistras. 1995. Sensitiv-
and Canada. Soc. Amer. For., Washington, DC. 168 pp. ity of a forest ecosystem model to climate parametriza-
Eyre, S. R. 1968. Vegetation and Soils: A World Picture. 2nd tion schemes. Environ. Pollut. 87:267-282.
ed. Edward Arnold, London. 328 pp. Fischlin, A., H. Bugmann, and D. Gyalistras. 1995. Sensitiv-
Farnworth, E. G., and F. B. Golley (eds.), 1973. Fragile ity of a forest ecosystem model to climate parametriza-
Ecosystems. Evaluation of research and applications in the tion schemes. Environ. Pollut. 87:267-282.
neotropics. Springer-Verlag, New York. 258 pp. Fisher, R. T. 1928. Soil changes and silviculture on the Har-
Fearnside, P. M. 1993. Deforestation in Brazilian Amazonia: vard forest. Ecology. 9:6-11.
The effect of population and land tenure. Ambio Fitton, E. M., and C. F. Brooks. 1931. Soil temperatures in
22:537-545. the United States. Mo. Weather Rev. 59:6-16.
Feeny, P. P. 1968. Effect of oak leaf tannins on larval growth Flinn, M. A., and R. W. Wein. 1977. Depth of underground
of the winter moth Operophtera brumata. 7. Insect. Physiol. plant organs and theoretical survival during fire. Can. rs
14:805-817. Bot. 55:2550-2554.
 REFERENCES CITED R-15

Florence, R. G., and P. H. Chuong. 1974. The influence of Fraser, L. 1993. The influence of salal on planted hemlock
soil type on foliar nutrients in plantations. Aust. 7. For: and cedar saplings on northern Vancouver Island. Mas-
Res. 6:1-8. ter’s thesis. Dept. of Botany, Univ. of British Columbia,
, and J. R. McWilliam. 1956. The influence of spac- Vancouver. 116 pp.
ing on seed production: Its application to forest tree im- Fredriksen, R. L. 1972. Nutrient budget of a Douglas-fir
provement. Si/lv. Genet. 5:97-102. forest on an experimental watershed in western Oregon.
Fogel, R., and K. Cromack, Jr. 1977. Effect of habitat and pp. 115-131. In J. F. Franklin and L.T. Dempster (eds.),
substrate quality on Douglas-fir (Pseudotsuga menziesii) Proceedings, Research on Coniferous Forest Ecosystems.
litter decomposition in western Oregon. Can. 7. Bot. USDA For. Serv., For. Range Exp. Sta., Portland. Ore.
55:1632-1644. , D. G. Moore, and L. A. Norris. 1975. The impact of
Foil, R. R. 1965. The effects of compaction on soil charac- timber harvest, fertilization and herbicide treatment on
teristics and seedling growth. Dissert. Abstr. 26(6):2955. streamwater quality in western Oregon and Washington. pp.
Folk, G. E., Jr. 1974. Textbook of Environmental Physiology. 283-313. In B. Bernier and C. H. Winget (eds.), Forest Soils
2nd ed., Lea and Febiger, Philadelphia. 465 pp. and Forest Land Management. Laval Univ. Press, Quebec.
Fonda, R. W. 1974. Forest succession in relation to river ter- Freedman, B. 1981. Intensive Forest Harvest: A Review ofNu-
race development in Olympic National Park, Washing- trient Budget Considerations. Info. Rept. M-X-121. Mar-
ton. Ecology 55:927-942. itimes For. Res. Centre, Can. For. Serv., Dept. Environ.,
Ford, E. B. 1931. Mendelism and Evolution. Methuen, Lon- Fredericton, N.B. 78 pp.
don. 116 pp. . 1995. Environmental Ecology. 2nd ed. Academic
. 1975. Ecological Genetics. 4th ed. Chapman & Hall, Press, New York. 606 pp.
London. 442 pp. , 5. Woodley, and J. Loo. 1994. Forestry practices
Ford-Robertson, F. C. (ed.). 1971. Terminology of Forest and biodiversity, with particular reference to the Mar-
Science, Technology, Practice, and Products. Vhe Multi- itime Provinces of eastern Canada. Environ. Rev. 2:33-77.
Lingual Forest ‘Terminology Series (English Language Frejka, T. 1974. World population projections: Alternative
Version) No. 1. Soc. Amer. For., Washington, DC. paths to zero growth. Pop. Bull. 29(5).
349 pp. Frelich, L. E. 2002. Forest Dynamics and Disturbance Regimes.
Forman, T. T- (ed.). 1979. Pine Barrens. Ecosystem and Land- Studies from Temperate Evergreen-Deciduous Forests. Cam-
scape. Academic Press, New York. 601 pp. bridge University Press, Cambridge. 266 pp.
Forman, R. T. T: 1995. Land Mosaics. Cambridge University Friend, A., A. Stevens, R. Knox, and M. Cannell. 1997. A
Press, Cambridge. process-based, terrestrial biosphere model of ecosystem
Fortin,J.A., R. Lavallée, and Y. Piché. 1979. Forest Utiliza- dynamics (Hybrid v3.0). Ecol. Model. 95:249-287.
tion for Energy and the Role ofNitrogen Fixation. A Litera- Friend, A. D., H. H. Schugart, and S. W. Running. 1993. A
ture Review. Contractors Report. DSS Contract physiology-based gap model of forest dynamics. Ecology
04SU.KLO011-8-0008. ENFOR Project P-9, Can. For. 74:792-797.
Serv., Chalk River, Ontario. 166 pp. Frey, T. E. A. 1973. The Finnish school and forest site-
Foster, N. W., and S. P. Gessel. 1972. The natural addition types. pp. 405-433. In R. H. Whittaker (ed.), Ordination
of nitrogen, potassium and calcium to a Pinus banksiana and Classification of Communities, Part 5, Handbook of Vege-
Lamb. forest floor. Can. 7. For. Res. 2:448-455. tation Science. W. Junk b.v., Publishers, The Hague.
, and I. K. Morrison. 1976. Distribution and cycling Fryer, J. H., and F. T. Ledig. 1972. Microevolution of the
of nutrients in a natural Pinus banksiana ecosystem. photosynthetic temperature optimum in relation to the
Ecology 57:110-120. elevational complex gradient. Can. 7. Bot. 50:1231-1235.
Fowells, H. A. 1965. Silvics of Forest Trees of the United States. Fujimori, T., S. Kawanabe, H. Saito, C. C. Grier, and T.
USDA For. Serv. Agric. Handbook #271. 762 pp. Shidei. 1976. Biomass and primary production in forests
Fox, W. 1990. Towards a Transpersonal Ecology. Shambhala, of three major vegetation zones of the northwestern
Boston. United States.7.Jap. For. Soc. 58:360-373.
Francke-Grosmann, H. 1967. Ecosymbiosis in wood- Fujiwara, K. 1970. [A study on the landslides by aerial pho-
inhibiting insects. pp. 141-204. In S. M. Henry (ed.), tographs.] Res. Bull. Exp. For., Hokkaido Univ., Japan
Symbiosis, Vol. Il. Academic Press, New York. 27:297-345.
Frank, F. 1957. The causality of microtine cycles in Ger- Fuller, W. H., S. Shannon, and P. S. Burgess. 1955. Effect of
many. 7. Wildl. Manage. 21:113-121. burning on certain forest soils of northern Arizona. For:
Franklin,J.E, W. H. Moir, G. W. Douglas, and C. Wieberg. Sci. 1:44—-50.
1971. Invasion of subalpine meadows by trees in the Cas- Fulton, M. R. 1991. A computationally efficient forest suc-
cade range. Washington and Oregon. Arctic Alpine Res. cession model: Design and initial tests. For. Ecol. Manage.
3:215-224. 42:23-34.
R-16 REFERENCES CITED

Fulton, M. R. 1991. A computationally efficient forest suc- phological and physical properties. Il. Chemical proper-
cession model: Design and initial tests. For: Ecol. Manage. ties. Soil Sci. Soc. Amer. Proc. 34:779—794.
42:23-34. Gessell, S. P. 1962. Progress and problems in mineral nutri-
Futuyma, D. J., and M. Slatkin (eds.). 1983. Coevolution. tion of forest trees. pp. 221-235. In T. T. Kozlowski (ed.),
Sinaner Associates, Sunderland, Mass. 555 pp. Tree Growth. Ronald, New York.
Gadgil, R. L., and P. D. Gadgil. 1971. Mycorrhiza and litter ,and A. N. Balci. 1965. Amount and composition of
decomposition. Nature 233:133. forest floors under Washington coniferous forests. pp.
, and P. D. Gadgil. 1975. Suppression of litter de- 11-23. In C. T. Youngberg (ed.), Forest Soil Relationships
composition by mycorrhizal roots of Pinus radiata N. Z. in N. America. Oregon State Univ. Press, Corvallis.
F. For, Sci. 5:33-41. Gessel, S. P., D. S. Lacate, G. F. Weetman, and R. F. Powers
, and P. D. Gadgil. 1978. Influence of clearfelling on (eds.) 1990. Sustained Productivity of Forest Soils. Pro-
decomposition of Pinus radiata litter. N. Z. F. For. Sci. ceedings of the 7th North American Forest Soils Confer-
8:213-224. ence, Faculty of Forestry, University of British
Gagnon,J.D. 1966. Free amino acids in needles in Abies bal- Columbia, Vancouver, BC. 525 pp.
samea and Picea mariana growing on different sites. Gholz, H. L. 1982. Environmental limits on aboveground
Nature 212:884. net primary production, leaf area, and biomass in vege-
Galloway, G., and J. Worrall. 1979. Cladoptosis: A repro- tation zones of the Pacific Northwest. Ecology
ductive strategy in black cottonwood? Can. 7. For. Res. 63:469-481.
9:122-125. , EK. Fitz, and R. H. Waring. 1976. Leaf area dif-
Gardner, C. A. 1957. The fire factor in relation to the vege- ferences associated with old-growth forest communities
tation of Western Australia. West. Austr. Nat. 5:166-173. in the western Oregon Cascades. Can. 7. For. Res.
Gardner, R. H., andJ. B. Mankin. 1981. Analysis of biomass 6:49-57.
allocation in forest ecosystems of IBP. pp. 451-497. In Gilbert, J. M. 1959. Forest succession in the Florentine
D. E. Reichle (ed.), Dynamic Properties of Forest Ecosystems. Valley, ‘Tasmania. Pap. Proc. Royal Soc. Tasmania
Cambridge Univ. Press, Cambridge. 93:129-151.
Garner, W. W., and H. A. Allard. 1920. Effect of the relative Gill, A. M. 1975. Fire and the Australian flora: A review.
length of the day and night and other factors of the envi- Aust. For. 38:1-25.
ronment on growth and reproduction in plants. 7. Agr. , R. M. Groves, and I. R. Noble (eds.). 1981. Fire and
Res. 18:553-606. the Australian Biota. Australian Academy of Science, Can-
Garrison, G. A., and R. S. Rummel. 1951. First-year effects berra. 582 pp.
of logging on ponderosa pine forest range lands of Ore- Gilpin, M. E. 1973. Do hares eat lynx? Amer. Nat.
gon and Washington.7. For: 49:708-713. 107:727-730.
Gates, D. M. 1962. Energy Exchange in the Biosphere. Harper Gimingham, C. H. 1972. Ecology of Heathlands. Chapman
& Row, New York. 151 pp. and Hall, London. 266 pp.
. 1965. Energy, plants and ecology. Ecology 46:1-24. Givnish, ‘T. J. 1988. Adaptation to sun and shade: A whole
. 1971. The flow of energy in the biosphere. Sci. plant perspective. Aust. 7. Plant Phys. 15:63-92.
Amer. 225:89-100. Glass, A. D. M. 1976. The allelopathic potential of phenolic
, and R. Janke. 1966. The energy environment of the acids associated with the rhizosphere of Pteridium aquil-
alpine tundra. Oecol. Plant. 1:39-62. inum. Can.F. Bot. 54:2440-2444.
Gause, G. F. 1934. The Struggle for Existence. Williams and Gleason, H. A. 1926. The individualistic concept of the
Wilkins Co., Baltimore. 163 pp. plant association. Bull. Torrey Bot. Club 53:7-26.
. 1935. Experimental demonstration of Volterra’s pe- . 1927. Further views on the succession concept.
riodic oscillations in the numbers of animals. 7. Exper: Ecology 8:299-326.
Biol. 12:44-48. . 1939. The individualistic concept of the plant asso-
Geiger, R. 1965. The Climate Near the Ground. Harvard ciation. Amer. Midl. Nat. 21:92-110.
Univ. Press, Cambridge, Mass. 611 pp. Glen-Lewin, D. C., R. K. Peet, and T. T. Veblen (eds.).
Geist, V. 1971. Mountain Sheep: A Study in Behavior and Evo- 1992. Plant Succession. Theory and Prediction. Chapman &
lution. Univ. of Chicago Press, Chicago. Hall, London. 352 pp.
Gergel, S E., and M. G. Turner (eds.) 2002. Learning Land- Godwin, G. E. 1968. The influence of wind on forest man-
scape Ecology. Springer, New York. agement and planning. pp. 60-66. In R. W. V. Palmer
Gerloff, G. C., D. G. Moore, andJ. T. Curtin. 1966. Selec- (ed.), Wind Effects on the Forest. Suppl. to For. 1968. Ox-
tive absorption of mineral elements by mature plants of ford Univ. Press, London.
Wisconsin. Plant Soil 25:393-405. Goldammer, J. G. (ed.). 1990. Fire in the Tropical Biota:
Gersper, P. L., and N. Holowaychuck. 1970. Effects of stem- Ecosystem Processes and Global Challenges. Ecological Studies
flow water on a Miami soil under a beech tree. I. Mor- 84. Springer-Verlag, New York. 497 pp.
 REFERENCES CITED R-17

. (ed.). 1992. Tropical Forests in Transition. Ecology of , P. B. Reich, and Y. Son. 1993. Canopy dynamics and
Natural and Anthropological Disturbance Processes. aboveground production of five tree species with differ-
Birkhauser Verlag, Basel. 269 pp. ent leaf longevities. Tree Physiol. 12:327-345.
, and M. J. Jenkins (ed.). 1990. Fire in Ecosystem Dy- , K. A. Vogt, and C. C. Grier. 1992. Carbon dynam-
namics. Mediterranean and Northern Perspectives. Springer- ics of Rocky Mountain Douglas-fir: Influence of water
Verlag, New York. 199 pp. and nutrient availability. Ecol. Monogr. 62:43-65.
Golding, D. L. 1970. The effects of forests on precipitation. Grace, J. B., and D. Tilman (eds.). 1990. Perspectives on Plant
For. Chron. 46(5):1-6. Competition. Academic Press, New York. 484 pp.
. 1978. Calculated snowpack evaporation during chi- Granat, L., R. O. Hallberg, and H. Rodhe. 1976. The glob-
nooks along the eastern slope of the Rocky Mountains in al sulphur cycle. pp. 89-134. In B. H. Svensson and R.
Alberta. 7. Appl. Meteorol. 17:1647-1651. Séderlund (eds.), Nitrogen, Phosphorus and Sulphur—
, and R. H. Swanson. 1978. Snow accumulation and Global Cycles. Scope Report 7. Ecol. Bull. (Stockholm)
melt in small forest openings in Alberta. Can. 7. For. Res. De
8:380-388. Grange, W. B. 1949. The Way to Game Abundance. Scribner’s,
Golley, F. B. 1960. Energy dynamics of a food chain in an New York. 365 pp.
old-field community. Ecol. Monogr. 30:187-206. Granhall, U., and H. Selander. 1973. Nitrogen fixation in a
. 1961. Energy values of ecological materials. Ecology subarctic mire. Oikos 24:8-15.
42:581-584. , and T. Lindberg. 1977. Nitrogen Fixation of Conif-
. 1977. Insects as regulators of forest nutrient cycling. erous Forest Sites Within the SWECON Project. Swed.
Trop. Ecol. 18:116-123. Conif. For. Project. Tech. Rept. 11. 39 pp. Uppsala, Swe-
, J. IT. McGinnis, R. G. Clements, G. I. Child, and den.
M. J. Duever. 1975. Mineral Cycling in a Tropical Moist Graumlich, L. J., and L. B. Brubaker. 1995. Long-term
Forest Ecosystem. Univ. Georgia Press, Athens. 248 pp. records of growth and distribution of conifers: Integration
, and E. Medina. 1975. Tropical Ecological Systems. of paleoecology and physiological ecology. pp. 37-62. In
Trends in Terrestrial and Aquatic Research. Ecological Studies. W. K. Smith and T. M. Hinckley (eds.), Ecophysiology of
Vol. II. Springer-Verlag, New York. 398 pp. Coniferous Forests. Academic Press, New York.
Golley, F. B. 1993. A History of the Ecosystem Concept in Ecolo- Green, G. R., and C. A. Evans. 1940. Studies on a popula-
gy. More Than the Sum of the Parts. Yale Univ. Press, New tion cycle of snowshoe hares on the Lake Alexander area.
Haven, CT. 254 pp. J. Wildl. Manage. 4:220-238, 267-278, 347-358.
Gordon, D. T. 1973. Damage From Wind and Other Causes in , and C. L. Larson. 1938. Shock disease and the
Mixed White Fir-Red Fir Stands Adjacent to Clearcuttings. snowshoe hare cycle. Science 87:298-299.
USDA For. Serv. Res. Pap. PSW-90, 22 pp. Greenland, D. J., and M. Kowal. 1960. Nutrient content of
Gordon, M. S., G. A. Bartholemew, and A. D. Grivell, et al. the moist tropical forest of Ghana. Plant Soil
1968. Animal Function: Principles and Adaptations. 12:154-174.
MacMillan, New York. 560 pp. Gregory, R. A. 1960. The Development of Forest Soil Or-
Gore, A. J. P. 1968. The supply of six elements by rain to an ganic Layers in Relation to Time in Southeastern Alaska.
upland peat area.7.Ecol. $6:483-495. USDA For. Serv. Alaska For. Res. Center Tech. Note
Gorham, E., P. Vitousek, and W. A. Reiners. 1979. The reg- INov4 745 pp:
ulation of chemical budgets over the course of terrestrial Grier, C. C. 1972. Effects of fire on the movement and dis-
ecosystem succession. Ann. Rev. Ecol. Syst. 10:53—-84. tribution of elements within a forest ecosystem. Doctor-
Gosz,J.R. 1984. Biological factors influencing nutrient sup- al. thesis, Univ. of Washington, Seattle. 167 pp.
ply in forest soils. pp. 119-146. In G. D. Bowen and E. K. . 1975. Wildfire effects on nutrient distribution and
S. Nambiar (eds.), Nutrition ofPlantation Forests. Academ- leaching in a coniferous ecosystem. Can. 7. For. Res.
ic Press, London. 5:599-607.
, G. E. Likens, and F. H. Bormann. 1972. Nutrient , D. W. Cole, C. T. Dyrness, and R. L. Fredriksen.
content of litter fall on the Hubbard Brook experimental 1974. Nutrient cycling in 37- and 450-year-old Douglas-
forest, New Hampshire. Ecology. 53:769-784. fir ecosystems. pp. 21-34. In R. H. Waring and R. L. Ed-
, G. E. Likens, and F. H. Bormann. 1976. Organic wards (eds.), Integrated Research in the Coniferous Forest
matter and nutrient dynamics of the forest and forest Biome. Conif. For. Biome Bull. No. 5., Ecosystem Analy-
floor in the Hubbard Brook forest. Oecologia 22:305—320. sis Studies, US/IBP. Univ. Washington, Seattle.
Gower, S. T.,J.G. Isebrands, and D. W. Sheriff. 1995. Car- , S. T. Gower, and K. A. Vogt. 1990. Douglas-fir pro-
bon allocation and accumulation in conifers. pp. ductivity: A conceptual model of its regulation by water
217-254. In W. K. Smith and T. M. Hinkley (eds.), and nutrient availability. pp. 257-273. In S. P. Gessel,
Resource Physiology of Conifers. Acquisition, Allocation, and D. S. Lacate, G. KFWeetman, and R. F. Powers (eds.),
Utilization. Academic Press, New York. Sustained Productivity of Forest Soils. Proc. 7th N. Amer. For.
R-18 REFERENCES CITED

Soils Conf. Faculty of Forestry, Univ. of British Columbia, Habeck, J. R., and R. W. Mutch. 1973. Fire-dependent
Vancouver. forests in the northern Rocky Mountains. 7. Quat. Res.
, and R. S. Logan. 1977. Old growth Pseudotsuga men- 3:408-424.
ziesii communities of a western Oregon watershed: Bio- Hacskaylo, E. (ed.). 1971. Mycorrhizae. Proc. Ist N. Amer:
mass distribution and production budgets. Ecol. Monogr. Conf. on Mycorrhizae. April 1969. USDA For. Serv. Misc. oe
ee
el
ee
Ge
Gt
i>
y

47:373—400. Publ. 1189.

, and S. W. Running. 1977. Leaf area of mature . 1972. Mycorrhiza: The ultimate in reciprocal para- —

northwestern coniferous forests: Relation to site water sitism. BioSci. 22:577-583.

balance. Ecology 58:893-899. Hagg, E. H., and V. J. Lieffers. 1991. The impact of
, K. A. Vogt, M. R. Keyes, and R. L. Edmonds. 1981. Calamagrostis canadensis on soil thermal regimes after log-
Biomass distribution and above- and below-ground pro- ging in northern Alberta. Can. 7. For: Res. 21:387-394.
duction in young and mature Abies amabilis zone ecosys- Haines, E. H. 1922. Influence of varying soil conditions on
tems of the Washington Cascades. Can. 7. For. Res. night air temperatures. Monthly Weather Rev.
11:155-167. 50:363-366.
, and R. H. Waring. 1974. Conifer foliage mass relat- Hairston, N. G., F. E. Smith, and L. B. Slobodkin. 1960.
ed to sapwood area. For. Sci. 20:205-206. Community structure, population control and competi-
Griffith, B. G., E. W. Hartwell, and T. E. Shaw. 1930. The tion. Amer. Nat. 64:421-425.
evolution of soils as affected by the old field white pine- Hall, C. A. S., andJ.W. Day, Jr. 1977. Ecosystem Modelling in
mixed hardwood succession in central New England. Theory and Practice: An Introduction With Case Histories.
Harv. For. Bull. 15. 82 pp. Wiley, New York. 684 pp.
Griggs, R. F. 1946. The timberlines of North America and Hall, J. E. (ed.). 1991. Alternative Uses for Sewage Sludge.
their interpretation. Ecology 15:80-96. Proc. WRC Internal Workshop. Pergamon Press, Oxford,
Grime, J. P. 1966. Shade avoidance and shade tolerance in UK.
flowering plants. pp. 187-207. In R. Bainbridge, C.G. Hall, R. B. 1974. Variations in DNA content per cell, and
Evans, and O. Rackham (eds.), Light as an Ecological Fac- repetitions in DNA of selected populations of Pinus
tor. Blackwell Sci. Publ., Oxford. banksiana Lamb. Doctoral thesis, University of Wiscon-
. 1974. Vegetation classification by reference to sin, Madison. 209 pp.
strategies. Nature (London) 250:26-31. Hall, G. M. J., and D. Y. Hollinger. 2000. Simulating New
. 1979. Plant Strategies and Vegetation Processes. John Zealand forest dynamics with a generalized temperate
Wiley, Chichester. 222 pp. forest gap model. Ecol. Appl. 10:115-130.
, and A. W. Jeffrey. 1965. Seedling establishment in Handley, W. R. C. 1954. Mull and mor formation in relation
vertical gradients of sunlight. 7. Ecol. 53:62 1-642. to forest soils. Bull. For. Commission, London 23.
Grinnell, J. 1917. The niche-relationship of the California . 1963. Mycorrhizal associations and Calluna heart-
Thrasher. Awk. 34:427-433. land afforestation. For. Comm. Bull. No. 36. Common-
Gross, A. O. 1928. The health hen. Mem. Boston. Soc Nat wealth For. Inst., Oxford. 70 pp.
Hist. 6:491-588. Hansen, H. P. (ed.). 1966. Arctic Biology. 2nd ed. Oregon
Grubb, P. J. 1977. The maintenance of species richness in State Univ. Press, Corvallis. 318 pp.
plant communities. The importance of the regeneration Hansen, J., A. Johnson, A. Lacas, S. Lebedeff, P. Lee, D.
niche. Biol. Rev. 52:107-145. Rind, and G. Russel. 1981. Climatic impact of increasing
Grumbine, R. E. 1994. What is ecosystem management? atmospheric carbon dioxide. Science 213:957-966.
Conserv. Biol. 8:27-38. Hansen, A. J, S. L. Garman, J. F. Weigand, D. L. Urban,
Gucinski, H., E. Vance, and W. A. Reiners. 1995. Potential W. C. McComb, and M. G. Raphael. 1995. Alternative
effects of global climate change. pp. 309-331. In W. K. silvicultural regimes in the Pacific Northwest: Simula-
Smith and T. M. Hinckley (eds.), Ecophysiology of Conifer- tions of ecological and economic effects. Ecol. Appl.
ous Forests. Academic Press, New York. 5:535-554.
Gullion, G. W. 1984. Managing Northern Forests for Hansen, J., R. Ruedy, M. Sato, M. Imhoff, W. Lawrence, D.
Wildlife. Minn. Agr. Exp. Sta. Misc.J.Ser. Publ. 13,442. Easterling, T. Peterson, and T. Karl. 2001. A closer look
77
42 .
at United States and global surface temperature change.
Gullison, J., and C. P-A. Bourque. 2001. Spatial prediction F. Geophys. Res. 106:239-247.
of tree and shrub succession in a small watershed in Hansen,J. E., R. A. Ruedy, and M. Sato. 2002. Global tem-
Northern Cape Breton Island, Nova Scotia, Canada. perature trends: 2002 summation. Goddard Institute for
Ecol. Model. 137:181-199. Space Studies, N.Y., N.Y. Available at: www. giss.nasa
Gunn, D. L. 1960. The biological background oflocust con- -gov/research/observe/surftemp. More references avail-
trol. Ann. Rev. Entomol. 5:279-300. able at www.giss.nasa.gov/gpol/authors/hauson.
 REFERENCES CITED R-19

Hanski, I., L. Hansson, and H. Henttoren. 1991. Specialist . 1969. The role of predation in vegetational diversi-
predators, generalist predators, and the microtine rodent ty. Brookhaven Symp. Biol. 22:48-62.
cycle. 7. Animal Ecol. 66:353-367. . 1977. Population Biology of Plants. Academic Press,
Harcomb, P. A. 1977a. The influence of fertilization on New York. 892 pp.
some aspects of succession in a humid tropical forest. , and I. H. McNaughton. 1962. The comparative
Ecology 58:1375-1383. biology of closely related species living in the same
. 1977b. Nutrient accumulation by vegetation during area. VII. Interference between individuals in pure and
the first year of recovery of a tropical forest ecosystem. mixed populations of Papaver species. New Phytol.
pp. 347-378. In J. Cairns, K. Dickson, and E. Herricks 61:175-188.
(eds.), Recovery and Restoration of Damaged Ecosystems. Harris, A. S., and W. A. Farr. 1974. The Forest Ecosystem
Univ. of Virginia Press, Charlottesville. of Southeast Alaska. 7. Forest Ecology and Timber man-
Hardin, G. 1960. The competitive exclusion principle. agement. USDA For. Serv. Gen. Tech. Rept. PNW-25.
Science 131:1292-1297. 109 pp.
Hardy, C. E., andJ.W. Franks. 1963. Forest Fires in Alaska. Harris, L. D. 1984. The Fragmented Forest. Island Biogeogra-
USDA For. Serv. Res. Pap. INT-5. 163 pp. phy Theory and the Preservation of Biotic Diversity. Univ. of
Harley, J. L. 1952. Associations between micro-organisms in Chicago Press, Chicago. 211 pp.
higher plants (mycorrhizae). Ann. Rev. Microbiol. Harris, W. F., D. Santantonio, and D. McGinty. 1980. The
6:367-386. dynamic below-ground ecosystems. pp. 119-129. In
. 1969. The Biology of Mycorrbiza. 2nd ed. Leonard R. H. Waring (ed.), Forests: Fresh Perspectives from Ecosys-
Hill, London. 334 pp. tem Analysis. Oregon State Univ. Press, Corvallis.
Harmer, R., and I. J. Alexander. 1985. Effects of root exclu- Harrison, E. A., and H. H. Shugart. 1990. Evaluating per-
sion on nitrogen transformation and decomposition formance of an Appalachian oak forest dynamics model.
processes in spruce humus. pp. 267-277. In A. H. Fitter Vegetatio 86:1-13.
(ed.), Ecological Interactions in Soil. Br. Ecol. Soc. Spec. Harrison, R., D. Xue, C. Henry, and D. W. Cole. 1994.
Publ. 4., Blackwell, Oxford. Long-term effects of heavy applications of biosolids on
Harmon, M. E., G. A. Baker, G. Spycher, and S. E. Greene. organic matter and nutrient content of a coarse-textured
1990a. Leaf-litter decomposition in the Pircea/Tsuga forest soil. For: Ecol. Manage. 66:165-177.
forests of Olympic National Park, Washington, U.S.A. Harrison, E. A., and H. H. Shugart. 1990. Evaluating per-
For. Ecol. Manage. 31:55-66. formance of an Appalachian oak forest dynamics model.
, W. K. Ferrell, and J. F. Franklin. 1990b. Effects on Vegetatio 86:1-13.
carbon storage of conversion of old growth forests to Hart, J. W. 1988. Light and Plant Growth. Unwin Hyman,
young forests. Science 247:699-702. Condon. 204 pp.
, and J. F. Franklin. 1989. Tree seedlings on logs in Hartman, G. F, and J. C. Scrivener. 1990. Impacts of
Picea-Tsuga forests of Oregon and Washington. Ecology forestry practices on a coastal stream ecosystem, Carna-
70:48-59. tion Creek, British Columbia. Canadian Dept. Fish and
, J. E. Franklin, F.J. Swanson, P. Sollins, S. V. Grego- Oceans, Ottawa. Ont. Bull. Fish. Aquatic Sci. 223.
ry, J.D. Lattin, N.W. Andrews, S.P. Clire, N.G. Aumen, Hartog, G. den, and H. L. Ferguson. 1978. Water Balance.
J.R. Sedell, E.W. Lienkaemper, K. Kromack, and J. W. Hydrological Atlas of Canada. Fisheries and Environment
Cummins. 1986. Ecology of coarse woody debris in tem- Canada, Ottawa.
perate ecosystems. Adv. Ecol. Res. 15:133-302. Hartzell, A. 1967. Insect ectosymbiosis. pp. 107-140. In
Harmon, M. E., J. M. Harmon, W. K. Ferrell, and D. S. M. Henry (ed.), Symbiosis. Vol. Ul. Academic Press,
Brooks. 1996a. Modeling the accumulation of forest New York.
products in the Pacific Northwest. Clim. Change 33: Harwood, C. E., and W. D. Jackson. 1975. Atmospheric
521-550. losses of four plant nutrients during a forest fire. Aust.
Harmon, M. E., S. L. Harmon, and W. K. Ferrell. 1996b. For. 38:92-99.
Modeling the historical patterns of tree utilization in the Hase, H., and H. Foelster. 1983. Impact of plantation
Pacific Northwest: Implications for carbon sequestra- forestry with teak (Tectona grandis) on the nutrient status
tion. Ecol. Appl. 6:641-652. of young alluvial soils in west Venezuela. For. Ecol. Man-
Harmon, M. E., and B. Marks. 2002. Effects of silvicultural age. 6:33-57.
practices on carbon stores in Douglas-fir-western hem- Haukioja, E. 1980. On the role of plant defences in the fluc-
lock forests in the Pacific Northwest, U.S.A.: Results tuation of herbivore populations. Oikos 35:202-213.
from a simulation model. Can. 7. For. Res. 32:863-877. Hauser, P. M. 1971. World Population: Retrospect and
Harper,J.L. 1961. Approaches to the study of plant compe- prospect. pp. 103-122. In Rapid Population Growth.
tition. Symp. Soc. Exper. Biol. 15:1-39. N.A.S. Johns Hopkins, Baltimore, 696 pp.
R-20 REFERENCES CITED

Havranek, W. M., and U. Benecke. 1978. The influence of 308-312. In C. S. Csallany, T.G. McLaughlin, and W.D.
soil moisture on water potential, transpiration and pho- Striffler (eds.), Watersheds in Transition. Amer. Water Re-
tosynthesis of conifer seedlings. Plant Soil 49:91-103. sources Assn. & Col. State Univ, Ft. Collins.
, and W. Tranquillini. 1995. Physiological processes , and J. H. Patric. 1965. Canopy and litter intercep-
during winter dormancy and their ecological signifi- tion of rainfall by hardwoods of eastern United States.
cance. pp. 95-124. In W. K. Smith and T. M. Hinckley Water Resources Res. 1:193-206.
(eds.), Ecophysiology of Coniferous Forests. Academic Press, Henderson, G. S., R. D. Hammer, and D. F. Gringal. 1990.
New York. Can measurable soil properties be integrated into a
Haydu, E. P., and R. N. Thut. 1971. Effects of forest chem- framework for characterizing forest productivity. pp.
icals on aquatic life in surface waters. pp. 159-171. In 137-154. In S. P. Gessel, D. S. Lacate, G. F. Weetman,
J. T. Krygier andJ.D. Hall (eds.), Proc. Symp. Forest Land and R. FE. Powers (eds.), Sustained Productivity of Forest
Uses and the Stream Environment. Oregon State Univ., Soil. Proc. 7th N. Amer: For. Soils Conf: Faculty of Forestry, rhL
e

Corvallis. Univ. of British Columbia, Vancouver.

Heal, O. W., M. J. Swift, and J. M. Anderson. 1982. Nitro- , and W. F. Harris. 1975. An ecosystem approach to
gen cycling in United Kingdom forests: The relevance of characterization of the nitrogen cycle in a deciduous for-
basic ecological research. Phil. Trans. R. Soc. Lond. B est watershed. pp. 179-193. In B. Bernier and C. H.
296:427-4+4. Winget (eds.), Forest Soils and Forest Land Management.
Heath, B., P. Sollins, D. A. Perry, and K. Cromak,
Jr. 1988. Laval Univ. Press, Quebec.
Asymbiotic nitrogen fixation in litter from Pacific North- Hendrickson, O. Q. 1988. Use of acetylene reduction for es-
west forests. Can. 7. For. Res. 18:68-74. timating nitrogen fixation in woody debris. Soil Sci. Soc.
Heginbottom,J. A. 1971. Some effects ofa forest fire on the Am. F. 52:840-844.
permafrost active layer at Inuvik. N.W.T. pp. 31-36. In . 1991. Abundance and activity of N,-fixing bacteria
R. J. E. Brown (ed.), Proc. Permafrost Active Layer Conf. in decaying wood. Can. 7. For. Res. 21:1299-1304.
C.N.R.C. Tech. Memo. No. 103. Henry, S. M. (ed.). 1967. Symbiosis. Vol. I. Associations ofMi-
Heilman, P. E. 1966. Change in distribution and availability croorganisms, Plants and Marine Organisms. Vol. IL.
of nitrogen with forest succession on north slopes in inte- Associations of Invertebrates, Birds, Ruminants and Other
rior Alaska. Ecology 47:825-831. Biota. Academic Press, New York.
. 1968. Relationship of availability of phosphorus and Henson, W. R. 1962. Convective transportation of
cations to forest succession and bog formation in interior Choristoneura fumiferana (Clem.). Proc. XI Internat. Kongr.
Alaska. Ecology 49:33 1-336. Fiir Entomol., Vienna 1960. I:4446.
Heinselman, M. L., and H. E. Wright (eds.). 1973. The eco- Hermann, R. K. 1977. Growth and production of tree roots:
logical role of fire in natural conifer forests of western A review. pp. 7-28. In J. K. Marshall (ed.), The Below-
and northern America. Quat. Res. 3:317-513. ground Ecosystem: A Synthesis of Plant-associated Processes.
Hellmers, H. 1962. Temperature effect on optimum tree Colorado State Univ., Fort Collins.
growth. pp. 275-287. In T. T. Kozlowski (ed.), Tree , and W. W. Chilcote. 1965. Effect of seedbeds on
Growth. Ronald, New York. germination and survival of Douglas-fir. Res. Pap. Ore.
. 1964. An evaluation of the photosynthetic efficiency For. Res. Lab. 4:1-28.
of forests. Quart. Rev. Biol. 39:249-257. Herrera, R., C. F Jordan, H. Klinge, and E. Medina. 1978.
. 1966a. Temperature action and interaction of tem- Amazon ecosystems. Their structure and functioning with
perature regimes in the growth of red fir seedlings. For: particular emphasis on nutrients. Interciencia 3:223-232.
Sci. 12:90-96. Heslop-Harrison, J. 1964. Forty years of genecology. Adv.
. 1966b. Growth response of redwood seedlings to Ecol. Res. 2:159-247.
thermoperiodism. For: Sci. 12:276-283. Hewlett, J. D., and W. L. Nutter. 1969. An Outline of Forest
, M. K. Genthe, and F. Ronco. 1970. Temperature af- Hydrology. Univ. of Georgia Press, Athens. 137 pp.
fects growth and development of Engelman spruce. For: Heyward, F. 1938. Soil temperatures during forest fires in
Sci. 16:447-452. the longleaf pine region.7. For: 36:478-491.
, and W. P. Sundahl. 1959. Response of Sequoia sem- Hibbs, D. E. 1979. The age structure of a striped maple
pervirens (D.Don) Endl. and Pseudotsuga menziesii (Mirb.) population. Can. 7. For. Res. 9:504—508.
Franco seedlings to thermoperiodism. For. Sci. Hiesey, W. M., and H. W. Milner. 1965. Physiology of eco-
12:276-283. logical races and species. Ann. Rev. Plant Phys.
Helmisaari, H.-S. 1992. Nutrient retranslocation in three 16:203-216.
Pinus sylvestris stands. For. Ecol. Manage. 51:347-367. Hik, D. S. 1994. Predation risk and the 10-year snowshoe
Helvey, J. D. 1972. First-year effects of wildfire on water yield hare cycle. Doctoral thesis. Univ. of British Columbia,
and stream temperature in north-central Washington. pp. Vancouver. 139 pp.
 REFERENCES CITED R-21

Hill, FE. B. 1973. Atmospheric sulphur and its links to the Horn, H. S. 1971. The Adaptive Geometry of Trees. Princeton
biota. pp. 159-181. In G. M. Woodwell and E. V. Pecan Univ. Press, Princeton. 144 pp.
(eds.), Carbon and the Biosphere. AEC Symposium Series . 1974. The ecology of secondary succession. Amn.
30. Tech. Info. Center, Office of Info. Services. USAEC. Rev. Ecol. Syst. 5:25-37.
Hillman, W. S. 1969. Photoperiodism and vernalization. In —. 1976. Succession. pp. 187-204. In R. M. May (ed.),
M. B. Wilkins (ed.), The Physiology of Growth and Develop- Theoretical Ecology: Principles and Applications. Blackwell
ment. McGraw-Hill, New York. Sci. Publ., Oxford.
Hills, G. A. 1976. An integrated, iterative holistic approach Hornbeck, J. W., G. E. Likens, R. S. Pierce, and F. H. Bor-
to ecosystem classification. pp. 73-97. In J. Thie and G. mann. 1975. Strip cutting as a means of protecting site
Ironside (eds.), Ecological (Biophysical) Land Classification in and streamflow quality when clearcutting northern hard-
Canada. Ecological Land Classification Series, No. 1. woods. pp. 209-225. In B. Bernier and C. H. Winget
Lands Directorate, Env. Canada, Ottawa. (eds.), Forest Soils and Forest Land Management. Laval
Hodges, J. D. 1967. Patterns of photosynthesis under natur- Univ. Press, Quebec.
al conditions. Ecology 48:234-242. Horton, K. W. 1956. The Ecology of Lodgepole Pine in Al-
Hogg, E. H., and V. J. Lieffers. 1991. The impact of berta and Its Role in Forest Succession. Can. Dept. N.
Calamagrostis canadensis on soil thermal regimes after log- Affairs and Nat. Resources, For. Br., For. Res. Div. Tech.
ging in northern Alberta. Can. 7. For. Res. 21:387-394. Note 45. 29 pp.
Holden, A. D. 1966. A chemical study of rain and stream Houghton, J. T., G. J. Jenkins, and J. J. Ephraums (eds.).
waters in the Scottish Highlands. Dept. Agric. Fish., 1990. Climate Change. The IPCC Scientific Assessment.
Scotland. WMO-UNEP. Cambridge Univ. Press, Cambridge, UK.
Holdridge, L. R. 1947. Determination of world plant forma- 365 pp.
tions from simple climatic data. Science 105:367-368. Houghton, R. A., J. E. Hobbie, J. M. Melillo, B. Moore, et
Holling, C. S. 1959. The components of predation as re- al. 1983. Changes in the carbon content of terrestrial
vealed by a study of small mammal predation of the Eu- biota and soils between 1860 and 1980: A net release of
ropean spruce sawfly. Can. Entomol. 9:293-320. CO, to the atmosphere. Ecol. Monogr. 53:236-262.
. 1965. The functional response of predators to prey Houghton, R. A., andJ.L. Hackler. 2002. Carbon flux to the
and its role in mimicry and population regulation. Mem. atmosphere from land-use changes. In Trends: A Com-
Entomol. Soc. Can. No. 45. 60 pp. pendium of Data on Global Change. Carbon Dioxide Infor-
. 1973. Resilience and stability of ecological systems. mation Analysis Centre, Oak Ridge National Laboratory,
Ann. Rev. Ecol. Syst. 4:1-23. U.S. Department of Energy, Oak Ridge, TN.
Holstener-Jorgensen, H. 1960. [Drift of soil into the west- House, H. L. 1965. Effects of low levels of the nutrient con-
ern edge of a plantation.] Forstt. Forsogsv. Danm. tent of a food and of nutrient imbalance on the feeding
26:389-397. and nutrition of a phytophagous larva, Celerio euphorbiae
Hom, J. L., and W. C. Oechel. 1983. The photosynthetic (L.) (Lepidoptera: Sphingidae). Can. Entomol. 97:62-68.
capacity, nutrient content, and nutrient use efficiency of . 1967. The role of nutritional factors in food selec-
different needle age-classes of black spruce (Picea mari- tion and preference as related to larval nutrition of an in-
ana) found in interior Alaska. Can. 7. For. Res. sect, Pseudoscarcophaga affimis (Diptera: Sarcophagidae) on
13:834-839. synthetic diets. Can. Entomol. 99:1310-1321.
Honer, T. G., W. R. Clark, and S. L. Gray. 1991. Determin- Howard, B. H. 1967. Intestinal micro-organisms of rumi-
ing Canada’s forest area and wood volume balance, nants and other vertebrates. pp. 317-385. In S. M. Henry
1977-1986. pp. 17-25. In D. G. Brand (ed.), Canada’s (ed.), Symbiosis. Vol. I. Academic Press, New York.
Timber Resources. Forestry Canada, Petawawa Nat. For. Howard, L. O., and W. F. Fiske. 1911. The importation into
Inst., Chalk River, Ont. Info. Rept. PI-X-101. the United States of the parasites of the gypsy moth and
Hoover, W. H., E. S. Johnston, and F. S. Brackett. 1933. the brown-tail moth: A report of progress with some
Carbon dioxide assimilation in a_ higher _ plant. consideration of previous and concurrent efforts of this
Smithsonian Inst. Misc. Coll. 87:1-19. kind. U.S. Dept. Agr, Bur. Ent. Bull. 91:105—109.
Hopkins, A. D. 1920. The bioclimatic law.7.Wash. Acad. Sci. Hoyle, M. C. 1973. Nature and Properties of Some Forest
10:3440. Soils in the White Mountains of New Hampshire.
Hopkins, H., and J. Eisen. 1959. Mineral elements in fresh USDA For. Serv. Res. Pap. NE-260. 18 pp.
vegetables from different geographic areas. Agric. Food Huffaker, C. B. 1958. Experimental studies on predation:
Chem. 7:633-638. dispersion factors and _ predator-prey oscillations.
Hori, T. (ed.). 1953. Studies on Fogs in Relation to Fog-Pre- Hilgardia 27:343-383.
venting Forest. Foreign Books Dept. Tanne ‘Trading Co., , and P. S. Messenger. 1964. The concept and signifi-
Sapporo. 399 pp. cance of natural control. pp. 74-117. In P. DeBach (ed.),
R-22 REFERENCES CITED

Biological Control of Insect Pests and Weeds. Chapman and . 1967b. Methods for uniform optimum fertilization
Hall, London. ; of forest tree plants. Proc. 14th IUFRO Congr. 3:265-269.
Humphreys, F. R., and M. Lambert. 1965. Soil temperature . 1971. A definition of optimum nutrient require-
profiles under slash and log fires of various intensities. ments in birch seedlings. II. Physiol. Plant. 24:118-125.
Aust. F. For. Res. 1(4):23-29. . 1979. Mineral nutrient requirements of Pinus
Hunt, C. B. 1972. Geology of Soils. Their Evolution, Classifica- sylvestris and Picea abies seedlings. Physiol. Plant.
tion and Uses. W. H. Freeman and Co., San Francisco. 45:373-380.
344 pp. . 1981a. Nutrition and growth of birch and grey alder e
Huntington, E. 1945. Mainsprings of Civilization. Wiley, seedlings in low conductivity solutions and at varied rela-
New York. 660 pp. tive rates of nutrient addition. Physiol. Plant. 52:454466.
Hurd, R. M. 1971. Annual tree-litter production by succes- 1981b. Nutrition and growth of coniferous
sional forest stands, Juneau, Alaska. Ecology 52:881-884. seedlings at varied relative nitrogen addition rates.
Hussain, S. B., C. M. Skan, S. M. Bashir, and R. O. Meeuwig. Physiol. Plant. 65:109-116.
1969. Infiltrometer studies of water-repellent soils on the . 1982. Relative addition rate and external concen-
east slope of the Sierra Nevada. pp. 127-131. In Symp. on tration: Driving variables used in plant nutrition re-
Water-Repellent Soils. Univ. of Calif., Riverside. search. Plant Cell Environ. 5:443-453.
Hustich, I. 1953. The boreal limits of conifers. Arctic , and G. I. Agren 1988. Nutrient uptake and alloca-
6:149-162. tion at steady-state nutrition. Physiol. Plant. 72:450-459.
Huston, M. A. 1979. A general hypothesis of species diversi- , and M. Kahr. 1985. Nutrition and growth of conif-
ty. Amer. Nat. 113:81-101. erous seedlings at varied relative nitrogen addition rates.
. 1992. Individual-based forest succession models and Physiol. Plant. 65:109-116.
the theory of plant competition. pp. 408-420. In D. L. , and A.-B. Lund. 1986. Theory and techniques for
De Angelis and L. J. Gross (eds.), Individual-Based Models steady state uinval nutrition and growth of plants. Scand.
and Approaches in Ecology: Populations, Communities and 7. For. Res. 1:439-453.
Ecosystems. Chapman & Hall, New York. Ingrens-Moller, H. 1957. Ecotypic response to temperature
, D. De Angelis, and W. Post. 1988. New computer and photoperiod in Douglas-fir. For. Sci. 3:78-83.
models _—_unify ecological theory. — BioScience Innes, J. L. 1993. Forest Health. Its Assessment and Status.
38(10):682-691. CAB International, Wallingford, UK. 677 pp.
, and T. Smith. 1987. Plant succession: Life history Irving, L. 1964. Terrestrial animals in cold: birds and mam-
and competition. Amer. Nat. 130:168-198. mals. pp. 361-377. In D. B. Dill et al. (eds.), Handbook of
Hutchinson, G. E. 1950. Survey of contemporary knowl- Physiology. Adaptation to the Environment. Amer. Physi-
edge of biogeochemistry. 3. The biogeochemistry of ver- ol. Soc., Washington DC. Waverly Press, Baltimore.
tebrate excretion. Bull. Amer. Mus. Nat. Hist. 96:1-554. Isaac, L. A. 1930. Seed flight in the Douglas-fir region. 7.
. 1957 Concluding remarks. Cold Spring Harbor Symp. For. 28:492-499.
Quant. Biol. 22:415-427. . 1938. Factors Affecting Douglas-Fir Seedlings.
. 1959. Hommage to Santa Rosalia, or why are there USDA Agric. Circ. 486. 45 pp.
so many kinds of animals? Ameri. Nat. 93:145-159. . 1946. Fog Drip and Rain Interception in Coastal
, and E.S. Deevey, Jr. 1949. Ecological studies on Forests. USDA For. Serv. PNW For. Range Exp. Stat.
populations. pp. 345-359. In G. S. Avery Jr. (ed.), Survey Paper 34. 16 pp.
of Biological Progress, Vol. 1. Academic Press, New York. Ivanov, M. V. 1981. The global biogeochemical sulphur
Hutchinson, G. L., R. I. Millington, and D. B. Peters. 1972. cycle. pp. 61-78. In G. E. Likens (ed.), Some Perspectives of
Atmospheric ammonia: adsorption by plant leaves. the Major Biogeochemical Cycles. SCOPE-17. Wiley, New
Science 175:771-772. York.
Hutchinson, G. E. 1965. The Ecological Theatre and the Eco- Ives, J. D. 1974. Permafrost. pp. 159-194. In J. D. Ives and
logical Play. Yale University Press, New Haven, CT. R. G. Barry (eds.), Arctic and Alpine Environments.
Hutt, F. B. 1949. Genetics of the Fowl. McGraw-Hill, New Methuen, London.
York. 590 pp. , and R. G. Barry (eds.). 1974. Arctic and Alpine Envi-
Ineson, P., J. Dutch, and K. S. Killham. 1991. Denitrifica- ronments, Methuen, London. 999 pp.
tion in a Sitka spruce plantation and the effect of Iwatsubo, G., and T. Tsutsumi. 1968. On the amount of
clearcutting. For. Ecol. Manage. 44:77-92. plant nutrients supplied to the ground by rain water in
Ingestad, T. 1967a. Nutrient needs of seedlings and young adjacent open plot and forests. II]. On the amount of
trees. pp. 139-141. In Proc. Colloq. For. Fertil. Vth Colloq. plant nutrients contained in run-off water. Bull. Kyoto
Internat., Potash Inst. Jyvaskyla, Finland. Univ. For. No. 40:140-156.
 REFERENCES CITED R-23

Jackson, R. M., and F. Rowe. 1966. Life in the Soil. Studies in Johnson, D. W. 1992. The effects of forest management on
Biology. No. 2, Edward Arnold, London. 60 pp. soil carbon storage. Water Air Soil Pollut. 64:83-120.
Jacob, F. H. 1958. Some modern problems in pest control. . 1994. Reasons for concern over impacts of harvest-
Sci. Prog. 46(181):30-45. ing. pp. 1-12. In W. J. Dyck, D. W. Cole, and N. B.
Janzen, D. H. 1966. Coevolution of mutualism between ants Comerford (eds.), Impacts of Forest Harvesting on Long-
and acacias in central America. Evolution 20:249-275. term Site Productivity. Chapman & Hall, London.
. 1967. Why mountain passes are higher in the trop- , D. W. Cole, and S. P. Gessel. 1979. Acid precipita-
ics. Amer. Nat. 101:233-249. tion and soil sulphate properties in a tropical and in a
. 1970. Herbivores and the number of tree species in temperate forest soil. Biotropica 11:38-42.
tropical forests. Amer. Nat. 104:501-528. , D. W. Cole, S. P. Gessel, M.J. Singer, and R.V. Min-
. 1973. Tropical agroecosystems. Science den. 1977. Carbonic acid leaching in a tropical, temper-
182:1212-1219. ate, subalpine, and northern forest soil. Arctic Alpine Res.
. 1974. Tropical blackwater rivers, animals, and mast 9:329-343.
fruiting by the Dipterocarpaceae. Biotropica 6:69-103. , J. Turner, andJ.M. Kelley. 1982. The effects of acid
. 1979. How to be a fig. Ann. Rev. Ecol. Syst. 10:13-51. rain on forest nutrient status. Water Resources Res.
Jarvis, P. G., and M. S. Jarvis. 1963. The water relations of 18:449-461.
tree seedlings. I. Growth and water use in relation to soil Johnson, E. A. 1992. Fire and Vegetation Dynamics: Studies
water potential. IT. Some aspects of the tissue water rela- from the North American Boreal Forest. Cambridge Uni-
tions and drought resistance. Physiol. Plant. 16:215-235, versity Press, Cambridge, UK. 125 pp.
501-516. , andJ.S. Rowe. 1975. Fire in the sub-arctic winter-
Jeffers,J. N. R. 1978. An Introduction to Systems Analysis, with ing ground of the Beverly Caribou Herd. Amer. Midl.
Ecological Applications. Edward Arnold, London. 198 pp. Nat. 94:1-14.
Jenkins, D., A. Watson, and G. R. Miller. 1963. Population Johnson, N. M. 1971. Mineral equilibria in ecosystem geo-
studies on red grouse, Lagopus lagopus scoticus (Lath.) in chemistry. Ecology. 52:529-531.
North-East Scotland.7.Anim. Ecol. 32:317-376. Johnson, N., and B. Cabarle. 1993. Surviving the Cut: Natur-
, A. Watson, G. R. Miller, and N. Picozzi. 1964. Unit al Forest Management in the Humid Tropics. World Re-
of grouse and moorland ecology. Rept. Nat. Conservancy sources Inst., Washington, DC. 71 pp.
1964:95-98. Johnson, P. L., and W. T. Swank. 1973. Studies of cation
Jennings, D. H., and D. L. Lee (eds.). 1975. Symbiosis. Sym- budgets in the southern Appalachians on four experimen-
posia of the Society for Experimental Biology No. XXIV. tal watersheds with contrasting vegetation. Ecology
Cambridge Univ. Press, Cambridge. 633 pp. 54:70-80.
Jenny, H. 1941. Factors ofSoil Formation. McGraw-Hill, New Johnson, D.W., and D.E. Todd. 1998. Harvesting effects on
York. 281 pp. long-term changes in nutrients of mixed oak forest. Soi/
. 1980. The Soil Resource. Origin and Behavior. Ecol. Sct. Soc. Amer. F. 62:1725-1735.
Stud. 37. Springer-Verlag, New York. 377 pp. Jones, J. B., and A. E. Luchsinger. 1979. Plant Systematics.
Jensen, J. 1962. [Investigations of the plant nutrient content McGraw-Hill, New York. 388 pp.
in rain]. Tidsskr. Planteavl. 65:894—906. Jones, J. R. 1969. Review and Comparison of Site Evalua-
Jensen, M. E., and P. S. Bourgeron (eds.). 1994. Ecosystem tion Methods. USDA For. Serv. Res. Paper RM-51. 27
management: Principles and applications, Vol. I. In Pp-
R. L. Everett (Team Leader), Eastside Forest Ecosystem Jones, K. 1976. Nitrogen fixing bacteria in the canopy of
Health Assessment. USDA For. Serv. Pacific N.W. Res. conifers in a temperate forest. pp. 451-463. In C. H.
Sta., Gen. Tech. Rept. PNW-GTR-318. Portland, Ore. Dickinson and T. F. Preece (eds.), Microbiology of Aerial
376 pp. Plant Surfaces. Academic Press, London.
Johnson, C. G. 1963. The aerial migration of insects. Sez. . 1970. Nitrogen fixation in the phyllosphere of the
Amer. 209:132-138. Douglas-fir, Pseudotsuga douglasii. Ann. Bot.
Johnson, C. G., Jr., R. R. Clausnitzer, P. J. Mehringer, and 34:239-244.
C. D. Oliver. 1994. Biotic and abiotic processes of east- . 1982. Nitrogen fixation in the canopy of temperate
side ecosystems: The effects of management on plant and forest trees: A re-examination. Ann. Bot. 50:329-334.
community ecology, and on stand and landscape vegeta- , E. King, and M. Eastlick. 1974. Nitrogen fixation
tion dynamics, Vol. III. In P. F. Hessberg (ed.), Assessment. by free-living bacteria in the soil and in the canopy of
Eastside Forest Ecosystem Health Assessment. R. L. Douglas-fir. Ann. Bot. 38:765-772.
Everett (comp.). USDA For. Serv. Gen. Tech. 322. Port- Jonsson, B., and R. Sundberg. 1972. Has the Acidification by
land, Ore. Atmospheric Pollution Caused a Growth Reduction in
R-24. REFERENCES CITED

Swedish Forests? Inst. for Skogproduktion. Rapport 20. Ray,J.J. 1993. On the nature of ecological integrity: Some
48 pp. closing comments. pp. 201-212. In S. Woodley, J. Kay,
Jordan, C. EF. 1971, A world pattern in plant energetics. and G, Francis (eds.), Ecological Integrity and the Manage-
Amer. Sci. 39325-4333. ment ofEcosystems. St. Lucie Press, Delray Beach, FL.
. 1985. Naarient Cycling in Tropical Forest one Kayll, A. J. 1968. The Role of Fire in Boreal Forest of Can-
Wiley, New York. 190 pp. ada. Petawawa For. Exp. Stat., Chalk River, Ontario.
. and R. Herrera. 1981. Tropical rain forests: rs nu- Info. Rept. PS-X-7.
trients really critical? Amer. Nat. 117:167-180. Keane, R. E., S. F Arno, andJ.K. Brown. 1989. Documen-
Joseph, P., T. Keith, L. Kline, J. Schanke, A. Kanaskie, and tation of the Ecological Process Model FIRESUM—A
D. Overhulser. 1991, Restoring forest health in the Blue FIRE sucession Model for Northern Rocky Mountain
Mountains: A 10 year strategic plan. For: Leg 61(2):3-12. Conifer Forests. USDA For. Serv., GTR INT-266.
Juday, C. 1940, The annual energy budget of an inland lake. Keane, R. E., S. FE Arno, andJ.K. Brown. 1989. FIRESUM: ~
Exalogy 21:438-450. An Ecological Process Model for Fire Succession in
Judson, O, P. 1994. The rise of the individual-based model Western Coniferous Forests. USDA Forest Service,
in ecology. Trends Eal, Evol. 91):9-14. . Washington, DC.
Jurgensen, M. F, R. T. Graham, M. J. Larsen, and A. F. Keane, R. E., K. C. Ryan, and S. W. Running. 1996. Simu-
Harvey. 1992, Clearcutting, woody residue removal, and lating effects of fire on northern Rocky Mountain land-
non-symbiotic nitrogen fixation in forest soils of the In- seapes with the ecological process model FIRE-BGC.
land Pacific Northwest. Can. 7. For: Res. 22:1172-1178 Tiree Physiol. 16:319-331.
. A. E. Harvey, R. T. Graham, M. L. Larsen,J.R. Ton- Keenan, R., and J. P. Kimmins, 1993. Ecological effects of
nand, and S. D. Page-Dumroese. 1990. Soil organic mat- cleareutting. Environ. Rev. 1:121-144.
ter, timber harvesting, and forest productivity in the inland Keenan, R. J. 1993. Structure and function of western red-
Northwest. pp. 392-415. In S. P. Gessel, D. S. Lacate, cedar and western hemlock forests on northern Vancou-
G. F Weetman, and R. FE Powers (eds.), Stustained Produc- ver Island. Doctoral thesis, Forest Science Dept.,
tivity ofForest Soils. Proc. Tth N. Am. For: Soils Conf. Faculty University of British Columbia, Vancouver. 167 pp.
of Forestry, Univ. of Briush Columbia, Vancouver. , C. E. Prescott, and J. P. Kimmins. 1995a. Litter
~ M. J. Larsen, R. T. Graham, and A. E. Harvey. production and nutrient resorption in western redcedar
1987. Nitrogen fixation in woody residue of northern and western hemlock forests on northern Vancouver Is-
Rocky Mountain conifer forests. Cen, 7. For Res. land, British Columbia. Can. 7. For: Res. 25:1850-1857.
17:1283-1288. , C. E. Prescott, J. P. Kimmins, J. Pastor, and B.
~ M. J. Larsen, S. D. Spano, A. FE. Harvey, and M. R. Dewey. 1995b. Changes in litter mass and chemistry dur-
Gale. 1984. Nitrogen fixation associated with increased ing decomposition in western redcedar and western hem-
wood decay in Douglas-fir residue. For Sa 30: lock forests on northern Vancouver Island, British
1038-1044. Columbia. Can. 7. Bot. (in press).
Kahn, H., and W. Brown. 1975, A world turning point. Keenan, R. G. 1993. Structure and function of western red-
Futurest. December. 284 pp. cedar and western hemlock forests on northern Vancou-
Kale, H. W. Il. 1965. Ecology and Bioenergetics of the ver Island. University of British Columbia, Vancouver,
Long-Billed Marsh Wren in Georgia Salt Marshes. Publ. BC. 177 pp.
Nuttal Ornith. Club, No. 5. Keenan, R. J., and J. P. Kimmins. 1995. Modeling carbon
Kalela, Q. 1962. On the fluctuations in the numbers of are- and nitrogen dynamics in western red cedar and western
uc and boreal small rodents as a problem of production hemlock forests. pp. 547-568. In W. W. McFee andJ.M.
biology. den. doad. Sa. Fenn, ATV 6621-38. Kelly (eds.), Carbon Forms and Functions in Forest Soils.
Katagin, S., and T. Tsutsumi. 1973. The relation between site Soil Science Society of America, Inc., Madison, WI.
condition and circulation of nutrients in forest ecosystems. Keever, C. 1950. Causes of succession on old fields of the
I. Littertall and nutrient contents.7.Fp. For: Sac. 35:83-90. Piedmont, North Carolina. Ecol. Monogr: 20:229-250.
Katznelson, H. 1965, Nature and importance of the rhizo- Keeves, A. 1966. Some evidence of loss of productivity with
sphere. pp. 187-209. In K. FE. Baker and W. C. Snyder successive rotations of Pinus radiata in the southeast of
(eds.), Exalogy ofSoil-Borne Plant Pathogens. Univ. of Calif. South Australia. Aust. For: 30:51-63.
Press, Berkeley. Keith, L. B. 1963. Wildlife’s Ten-Year Cycle. Univ. Wisconsin
Kaufmann, M. R., R. T. Graham, D. A. Boyee, Jr, W. H. Press, Madison. 201 pp.
Moir, L. Perry, R. T. Reynolds, R. L. Bassett, et al. 1904, - 1984. Demography and ecology of a declining
An Ecological Basis for Ecosystem Management. USDA snowshow hare population. Wildl. Monogr. 90.
For. Serv. Rocky Mt. For. Range Expt. Sta, Gen. Tech. » and L. A. Windberg. 1978. A demographic analysis
Rept. RM-246. Fr. Collins, CQ. 22 pp. of the snowshoe hare cycle. Wildl. Monogr. 58:1-50.
 REFERENCES CITED R-25

Keiter, R., and M. Boyce. 1991. The Greater Yellowstone Kettlewell, H. B. D. 1956. Further selection experiments on
Ecosystem. Yale University Press, New Haven, CT. industrial melanism in the Lepidoptera. Heredity
Keller, T: 1971. The effect of nutrition on gas exchange in 10:287-301.
Norway spruce. Allgem. Forst. U. Fagdztg. 142:89-93. . 1958. A survey of the frequencies of Biston betulavia
Kellman, M. C. 1970. The viable seed content of some for- (L.) (Lep.) and its melanic forms in Great Britain.
est soil in coastal British Columbia. Can. 7. Bot. Heredity 12:51-72.
48:1383-1385. Key, K. H. L. 1950. A critique of the phase theory of locusts.
Kellogg, W. W., R. D. Cadle, E. R. Allen, et al. 1972. The Q. Rev. Biol. 25:363-407.
sulphur cycle. Science 175:587-596. Keyes, M. R., and C. C. Grier. 1981. Above- and below-
Kellomaki, S., and H. Vaisanen. 1991. Application of a gap ground net reproduction in 40-year-old Douglas-fir
model for the simulation of forest ground vegetation in stands on low and high productivity sites. Can. 7. For. Res.
boreal conditions. For: Ecol. Manage. 421:35-47. 11:599-605.
, H. Vaisanen, H. Hanninen, T: Kolstrém, R. Lan- Khanna, P. K., and R. J. Raison. 1990. Mechanisms leading
hanen, U. Mattila, and B. Pajari. 1992. A simulation to a sustained growth response to N-fertilization in a
model for the succession of the boreal forest ecosystem. stand of Pinus radiata. pp. 221-233. In S. P. Gessel, D. S.
Silva Fenn. 26:1-18. Lacate, G. F.Weetman, and R. FE. Powers (eds.), Sustained
Kellomaki, S., and H. Vaisanen. 1991. Application of a gap Productivity of Forest Soils. Proc. 7th N. Am. For. Soils Conf.
model for the simulation of forest ground vegetation in Faculty of Forestry, University of British Columbia, Van-
boreal conditions. For. Ecol. Manage. 42:35-47. couver.
Kellomaki, S., H. Vaisanen, H. Hanninen, T. Kolstrém, R. Kienast, F. 1987. FORECE: A Forest Succession Model for
Lauhanen, U. Mattila, and B. Pajari. 1992. SIMA: A Southern Central Europe. Oak Ridge National Laboratory,
model for forest succession based on the carbon and ni- Oak Ridge, TN, ORNL/TM-2989.
trogen cycles with application to silvicultural manage- Kienast, F., and N. Kuhn. 1989. Simulating forest succession
ment of the forest ecosystem. Silva Carelica 22:85. along ecological gradients in southern Central Europe.
Kelsall, J. P. 1968. The Migratory Barren-ground Cariboo in Vegetatio 79:7-20.
Canada. Can. Wildl. Serv., Ottawa. 340 pp. , and N. Kuhn. 1989. Simulating forest succession
, E. S. Telfer, and T. D. Wright. 1977. The effects of along ecological gradients in southern Central Europe.
fire on the ecology of the boreal forest, with particular Vegetatio 79:7—20.
reference to the Canadian north: A review and selected Kilgore, B. M., and G. S. Briggs. 1972. Restoring fire to
bibliography. Can. Wildl. Serv., Occasional Paper No. high elevation forests in California.7.For. 70:266-271.
32. Ottawa, Canada. Kilian, C., and G. Lemée. 1956. Les xérophytes: Leur
Kendeigh, S. C. 1961. Animal Ecology. Prentice-Hall, Engle- économie d’eau. Handb. Pflanzen-physiol. 3:787-824.
wood Cliffs, NJ. 468 pp. Kimber, P. C. 1974. The root system of Jarrah (Eucalyptus
Kercher,J.R., and M. C. Axelrod. 1981. SILVA: A model for muarginata). For. Dept. W. A. Res. Pap. No. 10, 5 pp.
forecasting the effects of SO, pollution on growth and Kimmens, H. 1997. Balancing Act: Environmental Issues in
succession in a western coniferous forest. Ecol. Model. Forestry. 2nd ed. Univ. of British Columbia Press, Van-
23:165-184. couver. 305 pp.
, and M. C. Axelrod. 1984. A process model of fire Kimmins, J. P. 1970. Cyclic fluctuations in herbivore popu-
ecology and succession in a mixed-conifer forest. Ecology lations in northern ecosystems. A general hypothesis.
65:1725-1742. Doctoral thesis, Yale Univ., New Haven.
Kercher,J.R., and M. C. Axelrod. 1981. SILVA: A model for . 1971. Variations in the foliar amino acid composi-
forecasting the effects of SO, pollution on growth and tion of flowering and non-flowering balsam fir (Abies bal-
succession in a western coniferous forest. Ecol. Model. samea (L.) Mill.) and white spruce (Picea glauca (Moench)
23:165-184. Voss) in relation to outbreaks of the spruce budworm
Kercher,J.R., and M. C. Axelrod. 1984. A process model of (Choristoneura fumiferana (Clem.)). Can. 7. Zool.
fire ecology and succession in a mixed-conifer forest. 49:1005—1011.
Ecology 65:1725-1742. . 1972. The ecology of forestry: The ecological role
Kerfoot, O. 1968. Mist precipitation on vegetation. For. of man, the forester, in forest ecosystems. For. Chron.
Abstr. 29(1):8-20. 48:301-307.
Kershaw, K. A. 1973. Quantitative and Dynamic Ecology. Ed- . 1973. The renewability of natural resources: Impli-
ward Arnold, London. 183 pp. cations for forest management. 7. For: 71:290-292.
Kessell, S. R., and M. W. Potter. 1980. A quantitative suc- . 1974. Sustained yield, timber mining, and the con-
cession model for nine Montana forest communities. cept of ecological rotation: A British Columbian view.
Environ. Manage. 4:227-240. For. Chron. 50:27-31.
R-26 REFERENCES CITED

. 1977. Evaluation of the consequences for future tree , and K. A. Scoullar. 1994. Incorporation of nutrient
productivity of the loss of nutrients in whole-tree har- cycling in the design of sustainable, stand-level, forest
vesting. For: Ecol. Manage. 1:169-183. management systems using the ecosystem management
. 1985. Future shock in forest yield forecasting: The model FORECAST and its output format FORTOON.
need for a new approach. For. Chron. 61:503-512. InL.-O. Nilsson (ed.), Nutrient Uptake and Cycling in For-
. 1988. Community organization: Methods of study est Ecosystems. Ecosystem Res. Rept. No. 13. EUR 15465.
and prediction of the productivity and yield of forest , K. A. Scoullar, and M. J. Apps. 1991. FORCYTE-11. =
—

ecosystems. Can. F. Bot. 66:2654-2672. User’s Manual for the Benchmark Version. Forestry Canada,
. 1990. Modeling the sustainability of forest produc- ENFOR Project P-370. Northern Forestry Centre, Ed-
tion and yield for a changing and uncertain future. For: monton, Alberta. 294 pp.
Chron. 66:271-280. Kimmins, J. P. 1996a. The health and integrity of forest
. 1993a. Ecology, environmentalism and green reli- ecosystems: are they threatened by forestry? Ecosystem
gion. For. Chron. 69:285-289. Health. 2:5-18.
. 1993b. Scientific foundations for the simulation of Kimmins, J. P. 1996b. Importance of soil and role of ecosys-
ecosystem function and management in FORCYTE-11. tem disturbance for sustained productivity of cool tem-
Forestry Canada. Northern Forestry Centre, Edmonton. perate and boreal forests. Soil Sci. Soc. Amer. 7.
Info. Rept. NOR-X-328. 88 pp. 60:1643-1654.
1995a. Sustainable development in Canadian Kimmins, J. P. 1996. Biodiversity and its relationship to
forestry in the face of changing paradigms. For. Chron. ecosystem health and integrity. Policy Options. 17:3-7.
71(1):33-40. Kimmins, J. P. 1999. Biodiversity, beauty and the beast: Are
. 1995b. The health and integrity of forest ecosys- beautiful forests sustainable, are sustainable forests beau-
tems: Are they threatened by forestry? pp. 22-40. In tiful, and is “small” always ecologically desirable? For:
Managing Forests to Meet People’s Needs. Proc. Soc. Am. Chron. 75:955-960.
Foresters, Bethesda, MD. Kimmins, J. P. 2000. Respect for nature: An essential foun-
, D. Binkley, L. Chatarpaul, and J. de Catanzaro. dation for sustainable forest management. pp 3-24. In
1985. Biogeochemistry of Temperate Forest Ecosystems: R. G. D’Eon, J. Johnson, and E. A. Ferguson (eds.),
Literature on Inventories and Dynamics of Biomass and Ecosystem Management of Forested Landscapes: Directions
Nutrients. Can. For. Serv., Nat. For. Inst., Petawawa, and Implementation. UBC Press, Vancouver, BC.
Ont. Information Rept. PI-X-47 E/F 227 pp. Kimmins, J. P. 2001. Visible and non-visible indicators of
, P. G. Comeau, and W. A. Kurz. 1990. Modeling the forest sustainability: Beauty, beholders and belief sys-
interactions between moisture and nutrients in the con- tems. Pp 43-56. In S. R. J. Sheppard and H. W. Harshaw
trol of forest growth. For. Ecol. Manage. 30:361-379. (eds.), Forests and Landscapes Linking Ecology, Sustainability
, and M. C. Feller. 1976. Effect of clearcutting and and Aesthetics. CABI Publishing, Wallingford, UK.
broadcast slashburning on nutrient budgets, streamwater Kimmins, J. P. 2002. Future shock in forestry. Where have
chemistry and productivity in western Canada. pp. we come from; where are we going; is there a right way to
186-197. In Proc. XVI IUFRO World Congress, Div. I. manage forests? Lessons from Thoreau, Leopold, Tof-
, and B. Hawkes. 1978. Distribution and chemical fler, Botkin and Nature. For, Chron. 78:263-271.
content of fine roots in an old growth white-subalpine fir Kimmins, J. P. 2003. “Old-growth forest: An ancient and
stand in central B.C. Can. 7. For. Res. 8:265-279. stable sylvan equilibrium, or a relatively transitory
, and K. A. Scoullar, 1979. FORCYTE: A computer ecosystem condition that offers people a visual and
simulation approach to evaluating the effect of whole tree emotional feast? Answer-it depends,” For. Chron. 79. In
harvesting on the nutrient budgets of northwest forests. press.
pp. 266-273. In S. P. Gessel et al. (eds.), Forest Fertiliza- Kimmins, J. P. 2004. Emulation of natural forest distur-
tion Conference. Proc. Inst. For. Resources, Univ. of bance. What does this mean? In A.H. Perera, L. J. Buse
Wash., Seattle, Contrib. No. 40. and M. Weber (eds) Emulating Natural Forest Landscape
, and K. A. Scoullar, 1981. FORCYTE—A computer Disturbances: Concepts and Applications, Columbia Univ.
simulation approach to evaluating the effect of whole tree Press, N.Y., N.Y. In press.
harvesting on nutrient budgets and future forest produc- Kimmins,J.P., D. Mailly, and B. Seely. 1999. Modelling for-
tivity. Mitt. D. Forstl. Bundes-versuchsanstalt, Wien est ecosystem net primary production: The hybrid simu-
140:189-205. lation approach used in FORECAST. Ecol. Model.
, and K. A. Scoullar, 1983. FORCYTE-10: A User’s 122:195-224.
Manual. Faculty of Forestry. Univ. of B.C. Vancouver. Kimura, M. 1960. Primary production of the warm-temper-
112 pp. ate laurel forest in the southern part of Osumi Peninsula,
 REFERENCES CITED R-27

Kyushu, Japan. Misc. Rpt. Res. Inst. Nat. Resources. Publ. No. 45, Amer. Soc. Agron.; Soil Sci. Soc. Amer.,
Tokyo. 52/53:36-47. Madison, Wis.
King, H. G. C., and C. W. Heath. 1967. The chemical Klomp, H. 1970. The determination of clutch-size in birds.
analysis of small samples of leaf material and the relation- A review. Ardea 58:1:124.
ship between disappearance and composition of leaves. Kluijver, H. N. 1951. The population ecology of the Great
Pedobiologia 7:192-197. Tit, Parus m.major. Ardea 39:1-135.
Kira, T., and T. Shidae. 1967. Primary production and Knabe, W. 1976. Effects of sulfur dioxide on terrestrial veg-
turnover of organic matter in different forest ecosystems etation. Ambio 5:213-218.
of the western Pacific.7.#pn. Ecol. 17:70-87. Knapp, R. 1974. Cyclic successions and ecosystem ap-
, K. Shinozaki, and K. Hozumi. 1969. Structure of proaches in vegetation dynamics. pp. 91-100. In R.
forest canopies as related to their primary productivity. Knapp (ed.), Vegetation Dynamics, Part 8, Handbook of Veg-
Plant Cell Physiol. 10:129-142. etation Science. W. Junk b.v., Publishers, The Hague.
Kitchell, J. FE, R. V. O’Neill, D. Webb, G. W. Galleppe, S. Knight, H. 1966. Loss of nitrogen from the forest floor by
M. Bartell, J. F. Konce, and B. J. Ausmus. 1979. Con- burning. For: Chron. 42:149-152.
sumer regulation of nutrient cycling. BioScience 29:28-34. Knowles, P., and M. C. Grant. 1983. Age and size structure
Kittredge, J. 1940. A comparison of forest floors from plan- analyses of Engelmann spruce, ponderosa pine, lodge-
tations of the same age and environment. 7. For: pole pine and limber pine in Colorado. Ecology 64:1-9.
38:729-732. Knowles, R. 1965. The significance of non-symbiotic nitro-
. 1948. Forest Influences. The Effects of Woody Vegetation on gen fixation. Soil Sci. Soc. Amer. Proc. 29:223.
Climate, Water and Soil. McGraw-Hill, New York. 394 pp. . 1975. Interpretation of recent '°N studies of nitro-
Kleiber, M. 1961. The Fire of Life: An Introduction to Animal gen in forest ecosystems. pp. 53-65. Jn B. Bernier and
Energetics. Wiley, New York, 454 pp. C. H. Winget (eds.), Forest Soils and Forest Land Manage-
Klein, D. R. 1968. The introduction, increase, and crash of ment. Laval Univ. Press, Quebec.
reindeer on St. Matthew Island. 7 Wildl. Manage. Kobe, R. K., and K. D. Coates. 1997. Models of sapling mor-
32:350-367. tality as a function of growth to characterize interspecific
Klemmedson, J. O. 1975. Nitrogen and carbon regimes in variation in shade tolerance of eight tree species of north-
an ecosystem of young dense Ponderosa pine in Arizona. western British Columbia. Can. 7. For. Res. 27:227-236.
For. Sci. 21:163-168. Koch, P. 1978. Five new machines and six products can
. 1992. Decomposition and nutrient release from triple commodity recovery from southern forests. 7. For:
mixtures of Gambel oak and ponderosa pine leaf litter. 76:767-772.
For. Ecol. Manage. 47:249-261. . 1980. Concept for southern pine plantation opera-
Klinka, K. 1976. Ecosystem Units, Classification, Interpre- tion in the year 2020.7. For. 78:78-82.
tation and Mapping in the UBC Research Forest. Doc- . 1992. Wood versus non-wood materials in U.S.A.
toral thesis, Univ. of British Columbia, Vancouver. residential construction. Some energy-related global im-
, R. N. Green, R. L. Trowbridge, and L. E. Lowe. plications. For: Prod. 7. 42(5):3 1-42.
1981. Taxonomic Classification of Humus Forms in Kohler, P., and A. Huth. 1998. The effects of tree species
Ecosystems of British Columbia. First Approximation. grouping in tropical rainforest modelling: Simulations
Land Memnt. Rep. No. 8. Ministry of Forests, Victoria, with the individual-based model FORMIND. Ecol.
BGs54 pp. Model. 109:301-321.
, W. D. van der Horst, F. C. Nuszdorfer, and R. G. Kojima, S. 1979. Biogeoclimatic zones of Hokkaido Island,
Harding. 1980a. An ecosystematic approach to a subunit Japan.7. Coll. Liberal Arts 12:97-141.
plan. Koprino River watershed study. BCFS Land Mgmt. , 1981. Biogeoclimatic ecosystem classification and
Rept. No. 5. Ministry of Forests, Victoria, BC. 118 pp. its practical use in forestry. 7. Coll. Liberal Arts 14:41-75.
, W. D. van der Horst, F. C. Nuszdorfer, and R. G. Kolb, T. E., W. R. Wagner, and W. W. Covington. 1994.
Harding, 1980b. An ecosystematic approach to forest Concepts of forest health: Utilitarian and ecosystem per-
planning. For, Chron. 56:97-103. spectives.7.For, 92:10-15.
. Wang, G. J. Kayahara, R. E. Carter, and B. A. Komarek, E. V. 1971. Principles of fire ecology and fire
Blackwell. 1992. Light-growth response relationships in management in relation to the Alaskan environment. pp.
Pacific silver fir (Abies amabilis) and subalpine fir (Abies la- 1-22. In C. W. Slaughter et al. (eds.), Fire in the Northern
siocarpa). Can. F. Bot. 70:1919-1930. Environment—A Symposium. USDA For. Serv., PNW
Klock, G. O. 1982. Some soil erosion effects on forest soil For. Range Exp. Sta., Portland, Ore.
productivity. pp. 53-66. In Determinants of Soil Loss Toler- Képpen, W. 1923. Die Klimate der Erde: Grundriss der Khi-
ance. Proc. Symp., Ft. Collins, Colorado. ASA Special makunde. DeGruyter, Berlin. 369 pp.
R-28 REFERENCES CITED

Kormanik, P. P., W. C. Bryan, and R. C. Schultz. 1977. The and J. P. Decker. 1944. Relation between light in-
role of mycorrhizae in plant growth and development. tensity and rate of photosynthesis of loblolly pine and
pp. 1-10. In H. M. Vines (ed.), Physiology ofRoot-Microor- certain hardwoods. Plant Physiol. 19:350-358. i
lg
ti
at
en
Ta
a

ganisms Associations. Proc. Symp., S. Sect. Amer. Soc. , and T. T. Kozlowski. 1979. Physiology of Woody
Plant Physiol., Atlanta. Plants. Academic Press, New York. 811 pp.
Kormondy, E. J. (ed.). 1965. Readings in Ecology. Prentice- Krauchi, N. 1994. Modelling forest succession as influenced
Hall, Englewood Cliffs, NJ. 219 pp. by a changing environment. Mitteilung. Eidgenossisch.
. 1969. Concepts of Ecology. Prentice-Hall, Englewood Forschungsanst. Wald Schnee Landsch. 69:145-271.
Cliffs, NJ. 209 pp. Kriuchi, N., and F. Kienast. 1993. Modelling subalpine for-
Korstian, C. F, and N. J. Fetherolf. 1921. Control of stem est dynamics as influenced by a changing environment.
girdle of spruce transplants caused by excessive heat. Water Air Soil Pollut. 68:185-197.
Phytopathology 11:485-490. Kriuchi, N., and H. J. M. van Grinsven. 1995. Application
, and T. S. Coile. 1938. Plant Competition in Forest of the model FORSUM to the Solling spruce site. Ecol.
Stands. Duke Univ. Sch. For. Bull. 3. Durham, N.C. 125 pp. Model. 83:219-228.
Korzukhin, M. D., M. T. Ter-Mikaelian, and R. G. Wagner. Krammes, J. S. 1965. Seasonal debris movement from steep
1996. Process versus empirical models: Which approach mountainside slopes in southern California. pp. 85-89. In
for forest ecosystem management? Can. 7. For: Res. U.S. Dept. Agric. Misc. Publ. No. 970.
26:879-887. Krasny, M. E., and M. C. Whitmore. 1992. Gradual and
Koskimmes,J. 1955. Ultimate Causes of Cyclic Fluctuations sudden forest canopy gaps in Alleghany northern hard-
in Numbers in Animal Populations. Papers on Game Re- wood forests. Can. J. For. Res. 22:139-143.
search. Finnish Game Foundation. No. 15. 29 pp. Kraus,J.F, and A. E. Squillace. 1964. Inheritance of yellow
Kowal, N. E. 1969. Effect of leaching on pine litter decom- oleoresin and virescent foliage in slash pine. Si/vae Genet.
position rate. Ecology 50:739-740. 13:114-116.
Kozlovsky, D. G. 1968. A critical evaluation of the trophic Krebs, C. J. 1964. The Lemming Cycle at Baker Lake,
level concept. 1. Ecological efficiencies. Ecology 49:48-60. Northwest Territories, During 1959-62. Arctic Inst. N.
Kozlowski, T. T. 1962. Photosynthesis, climate and tree Amer. Tech. Pap. No. 15.
growth. pp. 149-164. In T. T. Kozlowski (ed.), Tree . 1978. Ecology: The Experimental Analysis of Distribu-
Growth. Ronald, New York. tion and Abundance. 2nd ed. Harper & Row, New York.
. (ed.). 1968-1978. Water deficits and plant growth. I. 678 pp.
1968. Development, control and measurement. Ul. 1968. , R. Boonstra, S. Boutin, M. Dale, S. Hannon, K.
Plant water consumption and response. Il. 1972. Plant re- Martin, A. R. E. Sinclair, J. N. M. Smith, and R. Turk-
sponses and responses and control of water balance. TV. 1977. ington. 1992. What drives the showshoe hare cycle in
Soil water measurement, plant responses and breeding for Canada’s Yukon? pp. 886-896. In D. R. McCullough and
drought resistance. V. 1978. Water and plant disease. Acade- R. H. Barrett (eds.), Wildlife 2001: Populations. Elsevier,
mic Press, New York. London.
, and C. E. Ahlgren (eds.). 1974. Fire and Ecosystems. , 9. Boutin, R. Boonstra, A. R. E. Sinclair, J. N. M.
Academic Press, New York. 542 pp. Smith, M. R. T. Dale, K. Martin, and R. Tarkington.
Krajina, V. J. 1965. Biogeoclimatic zones in British Colum- 1995. Impact of food and predation on the snowshoe
bia. Ecol. West. N. Amer. (Dept. Bot., Univ. of British hare cycle. Science 269:1112-1115.
Columbia, Vancouver.) 1:1—17. , M. S. Gaines, B. L. Keller, J. H. Myers, and R.H.
. 1966. Biogeoclimatic Zones of the Hawaiian Islands Tamarin. 1973. Population cycles in small rodents.
and Their Variation by Volcanic Activities. Paper Pre- Science 179:35-41.
sented to 11th Pac. Sci. Congr., Tokyo. , and J. Myers. 1974. Population cycles in small
. 1969. Ecology of forest trees in British Columbia. mammals. Adv. Ecol. Res. 8:267-399.
Ecol. West. N. Amer. (Dept. Bot., Univ. of British Co- Krefting, L. W., and H. L. Hansen. 1969. Increasing browse
lumbia, Vancouver.) 2(1):1-147. for deer by aerial applications of 2,4-D.7.Wildl. Manage.
. 1972. Ecosystem Perspectives in Forestry. H. R. 33:784-790.
MacMillan Lectureship in Forestry. Univ. of British Co- Kruckeberg, A. R. 1954. The ecology of serpentine soils. II.
lumbia, Vancouver. 31 pp. Plant species in relation to serpentine soils. Ecology
Kramer, P. J. 1957. Some effects of various combinations of 35:267-274.
day and night temperatures and photoperiod on the Krug, E. C., and C. R. Frink. 1983. Acid rain on acid soil. A
growth of loblolly pine seedlings. For: Sci. 3:-45-55. new perspective. Science 221:520-525.
. 1983. Water Relations ofPlants. Academic Press, New Krumlik, G. J. 1979. Comparative study of nutrient cycling
York. 489 pp. in the subalpine Mountain Hemlock Zone of British Co-
 REFERENCES CITED R-29

lumbia, Doctoral thesis, Univ. of B.C., Vancouver. 195 Landsburg, H. E. 1958. Trends in climatology. Science
Pp: 128:749-758.
Kucera, C. L., and J. H. Ehrenreich. 1962. Some effects of Landsberg, J. 2001. Modelling forest ecosystems: State-of-
annual burning on central Missouri prairie. Ecology the-art, challenges and future directions. Can. 7. For. Res.
43:334-336. 33:385-397
Kiichler, A. W. 1947. A geographical system of vegetation. Landsberg, J. J.,and S. T. Gower. 1997. Applications of Physi-
Geogr. Rev. (N.Y.) 37:233-240. ological Ecology to Forest Management. Academic Press, San
. 1964. Potential Natural Vegetation of the Conter- Diego, CA. 354 pp.
minous United States. Manual to accompany map. Amer. , and R. H. Waring. 1997. A geneeralized model of
Geogr. Soc. Spec. Publ. No. 36. 116 pp. + Rev. Map, 1965. forest productivity using simplifies concepts of radiation-
. 1967. Vegetation Mapping. Ronald, New York. 472 pp. use efficiency, carbon balance and partitioning. Forest
Kurz, W. A. 1989a. Net primary production, production al- Ecology and Management 95: 209-228.
location and foliage efficiency in second growth Dou- , and N. C. Coop. 1999. Modelling forest productiv-
glas-fir stands with differing site quality. Doctoral thesis. ity across large areas and long periods. Natural Resource
Univ. of B.C., Vancouver. Modeling 12:383-411.
. 1989b. The significance of shifts in carbon alloca- Langmaid, K. K. 1964. Some effects of earthworm invasion
tion patterns for long-term site productivity research. pp. in virgin podzols. Can. 7. Soil Sci. 44:34-37.
149-164. In W. J. Dyck and A. C. Mees (eds.), Research Langner, W. 1954. Beitrag zur Lésung des Problems der Be-
Strategies for Long-Term Site Productivity. Proc. TEA/BE fruchtungsoverhaltnisse im Wald mittels einer Mendelspal-
Workshop. For. Res. Instit., Rotorua, New Zealand. tung. Compte rendus, 11e Congrés, Union internationale
IEA/BE Rept. 8. des instituts de recherches forestiéres, 1953:459-467.
, and M. J. Apps. 1992. Atmospheric carbon and Pa- Larsen, J. A. 1974. Ecology of the northern forest border.
cific Northwest forests. pp. 69-80. In G. Wall (ed.), pp. 341-369. In J. D. Ives and R. G. Barry (eds.), Arctic
Implications of Climate Change for Pacific Northwest Forest and Alpine Environments. Methuen, London.
Management. Dept. Geog., Publ. Ser., Occasional Paper Larsen, P. R. 1963. Stem form development of forest trees.
No. 15. Univ. of Waterloo, Ontario. For. Sci. Monogr. 5:1-42.
, and M. J. Apps. 1993. Contribution of northern Lasch, P., M. Lindner, B. Ebert, M. Flechsig, F-W. Gers-
forests to the global C cycle: Canada as a case study. tengarbe, F. Suckow, and P. C. Werner. 1999. Regional
Water Air Soil Pollut. 70:163-176. impact analysis of climate change on natural and man-
, M. J. Apps, S. J. Beukema, and T. Lekstrum. 1995a. aged forests in the Federal State of Brandenburg, Ger-
20th century carbon budget of Canadian forests. Tellus many. Environ. Model. Assess. 4:273-286.
47B:170-177. Laukhart, J. B. 1957. Animal cycles and food.7. Wildl. Man-
, M. J. Apps, B. J. Stocks, and W. J. A. Volney. 1995b. age. 21:230-234.
Global climate change: Disturbance regimes and bios- . 1962. Wildlife population fundamentals. 27th N.
pheric feedbacks of temperate and boreal forests. pp. Amer. Wildl. Conf. 27:233-242.
119-133. In G. M. Woodwell and F. T. Mackenzie (eds.), Laurance, W. F., L. V. Ferreira, J. M. Rankin-de Merona,
Biotic Feedbacks in the Global Climatic System. Will the and S. G. Laurance. 1998. Rain forest fragmentation and
Warming Feed the Warming? Oxford Univ. Press, Oxford, the dynamics of Amazonian tree communities. Ecology
UK. 79:2032-2040
Kiirpick, P., U. Kiirpick, and A. Huth. 1997. The influence Lavender, D. P. 1987. Woody Plant Growth in a Changing
of logging on a Malaysian Dipterocarp rain forest: A Chemical and Physical Environment. Proc. (UFRO Symp.,
study using a forest gap model. 7. Theor: Biol. 185:47—-54. Dept. of Forest Sciences, Univ. of British Columbia,
Lack, D. 1947. Darwin’s Finches. Cambridge Univ. Press, Vancouver. 314 pp.
Cambridge. 208 pp. Lawrence, D. B. 1939. Some features of the vegetation of
. 1954. The Natural Regulation of Animal Numbers. the Columbia River gorge with special reference to
Oxford Univ. Press, New York. 343 pp. asymmetry of forest trees. Ecol. Monogr. 19:217-257.
Lahti, D. C. 2001. The “edge effect on nest predation” hy- . 1950. Glacier fluctuation for six centuries in south-
pothesis after twenty years. Biol. Conserv. 99:365-374. eastern Alaska and its relation to solar activity. Geogr: Rev.
Lal, R., and P. A. Sanchez. 1992. Myths and Science ofSoils in 40:191-223.
the Tropics. Soil Sci. Soc. Amer., Amer. Soc. Agronomy. . 1958. Glaciers and vegetation in southeastern Alas-
SSSA Special Publ. No. 29, Madison, Wis. 185 pp. ka. Amer. Sci. 46:89-122.
Lamprecht, H. 1966. The two exmutants /aciniata and , R. E. Sehonihi, A. Quispel, and G. Bond. 1967.
asplenifolia of Fagus sylvatica and their genetical basis. Role of Dryas drummondi in vegetation development. 7.
Phyton. Horn. 11:188-198. Ecol. 55:793-813.
R-30 REFERENCES CITED

Layne, J. N., and M. A. V. Raymond. 1994. Communal nest- Leverenz, I.W., and D. J. Lev. 1987. Effects of carbon-diox- _
ing of southern flying squirrels in Florida. 7. Manrmal. ide-induced climate change on the natural ranges of six
75:110-120. major commercial tree ies in the western United
Leaf, A. L. (ed.). 1979. Imepact ofIntensive Harvesting on Forest States. pp. 123-155. In W. E. Shands andJ. S. Hoffman
Nutrient Cyeling. Proc. Symp., Coll. Environ. inte and (eds.), The Greenhouse Effect, Climate Change, and U.S.
For. State Univ. Syracuse, NY. 421 pp. Forests. Conservation Foundation, Washington, DC.
Ledig, F. T., andJ. H. Fryer. 1971, The Serotinous iad Levitt, J. 1958. Frost, drought and heat resistance.
Habit in Pinus rigida as Related to Selection in Introgres- Proto; 6:1-87.
sion, Contrib. Pap. Work. Group on Quant. Genetics. . 1972. Responses ofPlants to Environmental Stress. Aca-
TUFRO Congr. Gainesville, Fla, demic Press, New York. 697 pp.
. and D. R. Korbobo. 1983. Adaptation of sugar Lewis, W. M.,Jr. 1974. Effects of fire on nutrient movement
maple populations along altitudinal gradients: Photosyn- in a South Carolina pine forest. Ecology 55:1120-1127.
thesis, respiration, and specific leaf weight. Amer. 7. Bot. Leyton, L., and L. Z. Rousseau. 1958. Root growth of
70:256-265. seedlings in relation to aeration. pp. 467-475. In K. V.
Lee, J. J., and D. L. Inman. 1975, The ecological role of Thimann (ed.), Physiology of Forest Trees. Ronald Press,
consumers—An integrated ecosystem view. Ecology New York.
S6:1455-1458. Li, C, J. Cui, and C. Trettin. 2003. Modelling impacts of man-
Leemans, R., and I. C. Prentice. 1987. Description and sim- agement on carbon sequestration and trace gas emissions in
ulation of tree-layer composition and size distributions in forested wetland ecosystems. Institute for the Study of
a primaeval Picve-Pinus forest. egetatio 69:147-156. Earth, Oceans and Space, Univ. of New Hampshire,
, and I. C. Prentice. 1989, FORSKA, a General For- Durham, NH 03824. Submitted to J. Environ. Manage.
est Succession Model. Medd. Vaxt-biol. Inst., Uppsala. Lieberman, M., and D. Lieberman. 1989. Forests are not
60 pp. just Swiss cheese: Canopy stereogeometry of non-gaps in
Leemans, R. 1992. The biological component of the simula- tropical forests. Ecology 70:550-552.
tion model for boreal forest dynamics. pp. 428-445 In Likens, G. E. 1981. Sore Perspectives ofthe Major Biogeochem-
H. H. Shugart, R. Leemans, and G. B. Bonan (eds.), 4 ical Cycles. Scope 17.Wiley, New York. 175 pp.
Systems Analysis of the Global Boreal Forest. Cambridge . FH. Bormann, and N. M. Johnson. 1969. Nitrifica-
University Press, Cambridge, UK.. tion: Importance to nutrient losses from a cutover for- —

Leemans, R., and I. C. Prentice. 1987. Description and sim- ested ecosystem. Science 163:1205-1206.
ulation of tree-layer composition and size distributions in . FH. Bormann, N. M. Johnson, D. W. Fisher, and
a primaeval Picea-Pinus forest. | agenetio 69:147-156. R. S. Pierce. 1970. Effects of forest cutting and herbicide
Leemans, R., and I. C. Prentice. 1989. FORSKA, a General For- treatment on nutrient budgets in the Hubbard Brook
et Sucesion Medel. Medd. Vaxtbiol. Inst. Uppsala, Sweden. watershed-ecosystem. Ecol. Monogr: 40:23-47.
LeMay, V., and P. Marshall. 2003. Forest modeling for . F H. Bormann, R. S. Pierce,J.S. Eaton, and N. M.
ecosystem management, forest certification, and sustain- Johnson. 1977. Biogeochemistry of a Forested Ecosystem.
able management. Can, 7. For: Res. 33(Special Edition): Springer-Verlag, New York. 146 pp
386-541. Lillesand, T. M., and R. W. Kiefer. 1980. Remote Sensing and
Leonard, R. E. 1961. Interception of Precipitation by Image Interpretation. Wiley, New York. 612 pp.
Northern Hardwoods. USDA For. Serv., NE For. Exp. Lindeman, R. L. 1942. The trophic-dynamic aspect of ecol-
Sta., Upper Darby. Station Paper No. 159. 16 pp. ogy. Ecology 23:399-418.
Leopold, A. 1966. 4 Sand County Almanac. With Exsays on Linder, S. (ed.). 1981. Understanding and Predicting Tree
Conservation from round Rivers. Ballentine Books, N-Y. Growth. Studia Forestalia Suecica No. 160. 87 pp.
295pp. . 1987, Response to water and nutrients in coniferous
Lertzman, K. P. 1992. Patterns of gap-phase replacement in ecosystems. pp. 180-202. In E. D. Schulze and H.
a subalpine, old-growth forest. Ecology 73:657-669. Zwolfer (eds.), Potentials and Limitations of Ecosystem
Lett, C., C. Silber, and N. Barret. 1999. Comparison of a Analysis. Analysis and Synthesis. Ecological Studies 61.
cellular automata network and an individtal-based model Springer-Verlag, New York.
for the simulation of forest dynamics. Exo Model. Lindner, M. 1999. Waldbaustrategien im Kontext méglicher
121:277-293. Nlimainderungen. Forstwissenschaftliches Centralblatt 118:
Lett, C. and J. Walter. 1999. Validation of an individual- 1-13.
based model of forest dynamics using self-thinning rela- Lindner, M. 2000. Developing adaptive forest management
tionships. Comptes Rendus de TAoadémie des Sciences strategies to cope with climate change. Tree Physiol.
Serie-lll-Sciences de la Vie 322:879-8835, 20:299-307.
 REFERENCES CITED R-31

Lindner, M., P. Lasch, and M. Erhard. 2000. Alternative for- Sharpe (eds.), Forest Iskand Dynamics in Man-Dominated
est Management strategies under climatic change— Landscapes. Springer-Verlag, New York. 311 pp.
prospects for gap model applications in risk analyses. Loveless, A. R. 1961. A nutritional interpretation of sclero-
Silva Fennica 34:101-111. phylly based on differences in the chemical composition
Lindner, M., R. Sievanen, and H. Pretzsch. 1997. Improving of sclerophyllous and mesophytic leaves. Ann. Bot.
the simulation of stand structure in a forest gap model. 25:168-184.
For. Ecol. Manage. 95:183-195. . 1962. Further evidence to support a nutritional in-
Lines, R., and R. S. Howell. 1963. The Use of Flags to Es- terpretation of sclerophylly. Ann. Bot. 26:551-561.
timate the Relative Exposure of Trial Plantations. Lowe, V. P. W. 1969. Population dynamics of the red deer
Forestry Record #51. Forestry Commission, London. (Cervus elaphus L.) on Rhum. 7. Anim. Ecol.
Sepp. 38:425-457.
Liski, J, A. Pussinen, K. Pingoud, R. Makipaa, and T. Kar- Ludwig, D., D. D. Jones, and C. S. Holling. 1978. Qualita-
alainan 2001. Which rotation length is favorable to car- tive analysis of insect outbreak systems: The spruce bud-
bon sequestration? Can. 7. For. Res. 31:2004-2013. worm and the forest. 7. Anim. Ecol. 44:315-332.
Liu, S., R. Munson, D. W. Johnson, S. Gherini, K. Sum- Luken, J. O. 1990. Directing Ecological Succession. Chapman
mers, R. Hudson, K. Wilkinson, and L. F. Pitelka. 1992. Hall, London. 251 pp.
The nutrient cycling model (NuCM): Overview and ap- Lull, H. W. 1951. Forest fire smoke of September 1950. 7.
plication. pp. 583-690. In D. W. Johnson and S. E. Lind- For. 49:286.
berg (eds.), Atmospheric Deposition and Forest Nutrient Lulla, K. 1981. Remote sensing in ecological studies. Can. 7.
Cycling. A Synthesis of the Integrated Forest Study. Ecological Remote Sens. 7:97-107.
Studies 91. Springer-Verlag, New York. Lundmark, J. E. 1983. [Growth effects of whole-tree har-
Liu, J., and W. W. Taylor (eds.) 2002. Integrating Landscape vesting on different site types—forecasts based on basics
Ecology into Natural Resource Management. Studies in Land- of biology]. Swedish Univ. of Agr. Sciences. Skogsfakta.
scape Ecology. Cambridge University Press, Cambridge, Suppl. No. 1, pp. 24-31.
UK. 480 pp. Lundmark,J. E. 1986. [Short-term and long-term effects on
Livingston, B. E., and F. Shreve. 1921. The Distribution of site productivity by soil treatments in forestry]. Swedish
Vegetation in the United States as Related to Climatic Univ. of Agr. Sciences. Skogsfakta Konferens No. 9. pp.
Conditions. Carnegie Inst. Wash. Publ. 284. 590 pp. 126-132.
Lloyd, P. S. 1971. Effect of fire on the chemical status of Lutz, H. J. 1956. Ecological effects of forest fires in the inte-
herbaceous communities of the Derbyshire Dales.7.Ecol. rior of Alaska. USDA Tech. Bull. No. 1133. 121 pp.
59:261-273. , and R. F. Chandler, Jr. 1946. Forest Soils. Wiley,
Logan, K. T: 1966. Growth ofTree Seedlings as Affected by New York. 514 pp.
Light Intensity: II. Red Pine, White Pine, Jack Pine and Lutz, W., C. Prinz, and J. Langgassner. 1994. The ITASA
Eastern Larch. Dept. For. Can., Publ. 1160. 19 pp. world population scenarios to 2030. pp. 391-421. In W.
Longman, K. A., and J. Jenik. 1974. Tropical Forest and its En- Lutz (ed.), The Future Population of the World. What Can
vironment. Longman, London. 196 pp. We Assume Today. UASA, Earthscan Publications Ltd,
Lonsdale, W. M. 1990. The self-thinning rule: Dead or London.
alive. Ecology 71:1373-1388. Luxmoore, R. J.,R. Oren, D. W. Sheriff, and R. B. Thomas.
Lorenz, R. W. 1939. High Temperature Tolerance of Forest 1995. Source-sink-storage relationship of conifers. pp.
Trees. Univ. Minn. Agr. Exp. Stat. Tech. Bull. 141.25 pp. 179-216. In W. K. Smith and T: M. Hinkley (eds.),
Lotka, A. J. 1925. Elements ofPhysical Biology. Williams and Resource Physiology of Conifers. Acquisition, Allocation, and
Wilkins, Baltimore. Utilization. Academic Press, New York.
Loucks, O. L., A. R. Ek, W. C. Johnson, and R. A. Mon- MacArthur, R. H. 1958. Population ecology of some warblers
serud. 1981. Growth, aging and succession. pp. 37-85. In of north eastern coniferous forests. Ecology 39:599-619.
D. E. Reichle (ed.), Dynamics Properties of Forest Ecosys- ———,, andJ. W. MacArthur. 1961. On bird species diversi-
tems. International Biological Programme 23. Cambridge ty. Ecology 42:594-598.
Univ. Press, Cambridge. -, and E. O. Wilson. 1967. The Theory of Island Bio-
Lovat, L. 1911. Moor management, pp. 372-391. In L. geography. Princeton Univ. Press, Princeton, NJ.
Lovat (ed.), The Grouse in Health and Disease. London. MacDougal, D. T., and E. B. Working, 1921. Another high
Love, D. 1970. Subarctic and subalpine: Where and what? temperature record for growth and endurance. Science
Arctic Alpine Res. 2:63-71. 54:152=153.
Lovejoy, T. E., and D. C. Oren. 1981. The minimal critical MacGregor, H. G. 1994. Literature on the effects of envi-
size of ecosystems. pp. 7-12. In R. L. Burgess and D. M. ronment and forests on water quality and yield. Canadian
R-32 REFERENCES CITED

Forest Service, Maritimes Region, Natural Resources , L. Christanty, and J. P. Kimmins. 1996. “Without
Canada. R&D Report No. 10. Fredericton, NB. 119 pp. bamboo, the land dies”: Nutrient cycling and biogeo-
MacHattie, L. B. 1963. Winter injury of lodgepole pine fo- chemistry of a Javanese bamboo talun-kebun system. For.
liage. Weather (London) 18:301-307. Ecol. Manage.
MacKenzie, D. 1994. Where has all the carbon gone? New , and H. A. Margolis. 1992. Forest floor and mineral
Sci. Jan. 8:30-33. soil development in Casuarina equisetifolia plantations on
MacLean, D. A. 1978. Fire and the nutrient cycle of New the coastal sand dunes of Senegal. For: Ecol. Manage.
Brunswick pine and hardwood stands: Field studies and 55:259-278.
computer simulation studies. Doctoral thesis, Univ. of Mailly, D. 1998. Influence of the disturbance regime on for-
New Brunswick, Fredericton. 324 pp. est succession in the Coastal Western Hemlock dry sub-
, and R. W. Wein. 1980. Simulation ofwildfire effects zone, B.C. Doctoral Thesis. Faculty of Forestry,
on the nitrogen cycle of a Pinus banksiana ecosystem in University of British Columbia, Vancouver, BC.
New Brunswick, Canada. Ecol, Model. 10:167-192. Mailly, D., J. P. Kimmins, and R. T. Busing. 2000. Distur-
, S.J. Woodley, M. G. Weber, and R. W. Wein. 1981. bance and succession in a coniferous forest of northwest-
Fire and nutrient cycling. In R. W. Wein and D. A. ern North America: Simulations with Dryades, a spatial
MacLean (eds.), The Role of Fire in Northern Circumpolar gap model. Ecol. Model. 127:183-205.
Ecosystems. Scope 18. Wiley, New York. Majawa, A. O. 1977. Phytoecological impacts and manage-
MacLulich, D. A. 1937. Fluctuations in the numbers of ment implications of the Douglas-fir tussock moth near
British tetraonids.7.Anim. Ecol. 21:128-153. Kamloops, British Columbia. Master’s thesis, Univ. of
. 1937. Fluctuations in the Numbers of Varying Hare B.C., Vancouver.
(Lepus americanus). Univ. of Toronto Press, ‘Toronto. Mikela, A. 2001. Modelling tree and stand growth: Towards
MacMahon, J. A. 1980. Ecosystems over time succession a hierarchical treatment of multi-scale processes. pp.
and other types of change. pp. 27-58. In R. H. Waring Can. J. For. Res. 33:398-409.
(ed.), Forests: Fresh Perspectives From Ecosystem Analysis. Major, J. 1951. A functional, factorial approach to plant
Oregon State Univ. Press, Corvallis, Ore. ecology. Ecology 32:392-412.
Mace, A. C.,Jr. 1970. Soil Compaction Due to Tree Length 1974, Accumulation of ash elements and pH
and Full Tree Skidding With Rubber Tired Skidders. changes. pp. 215-218. In R. Knapp (ed.), Vegetation Dy-
Minn. For. Res. Note. No. 214. 4 pp. namics, Vol. 8, Handbook of Vegetation Science. Junk b.v.,
Machta, L. 1973. Prediction of CO) in the atmosphere. pp. Publishers, The Hague.
21-31. In G. M. Woodwell and E. V. Pecan (eds.), Carbon Malanson, G. P. 1996. Effects of dispersal and mortality on
in the Biosphere. AEC Symp. Series 30, Tech. Info. Center, diversity in a forest stand model. Ecol. Model. 87:103-110.
Off. of Info. Services. USAEC. Malanson, G. P., and M. P. Armstrong. 1996. Dispersal
Mackay, J. R. 1970. Disturbances to the tundra and forest probability and forest diversity in a fragmented land-
tundra environment of the western Arctic. Can. Geotech. scape. Ecol. Model. 87:91-102.
F. 7:420-432. Malcolm, D. C., 1975. The influence of heather on silvicul-
Mackenzie, J. M. D. 1952. Fluctuations in the numbers of ture practice—An appraisal. Scottish For. 29(1):14-24.
British tetraonids. 7. Anim. Ecol. 21:128-153. Malkonen, E. 1974. Annual primary production and nutri-
Mader, D. L., and H. W. Lull. 1968. Depth, Weight and ent cycle in some scots pine stands. Comm., Inst. For.
Water Storage of the Forest Floor in White Pine Stands Fenn., 84.5. 87 pp.
in Massachusetts. USDA For. Serv., NE For. Exp. Sta. Malthus, T. R. 1978. An Essay on the Principle of Population as
Res. Pap. No. NE-109. 35 pp. it Affects the Future Improvement of Society. Johnson, Lon-
Madge, D. S. 1965. Leaf fall and disappearance in a tropical don.
forest. Pedobiologia 5:273-288. Marchand, D. E. 1971. Rates and modes of denudation,
Madgewick, H. A. I., andJ.D. Ovington. 1959. The chemi- White Mountains, eastern California. Amer- 'F-Sa,
cal composition of precipitation in adjacent forest and 270:109-135.
open plots. Forestry. 32:14-22. Margalef, R. 1975. Diversity, stability and maturity in natur-
. (ed.). 1990. Boron, copper and zinc in plantation al ecosystems. pp. 151-160. In W. H. van Dobben and
forestry. For. Ecol. Manage. Spec. Issue 37:1-221. R. H. Lowe-McConnell (eds.), Unifying Concepts in Ecolo-
Mahendrappa, M. K. 1974. Chemical composition of stem- gy. W. Junk, b.v., Publishers, The Hague.
flow from some eastern Canadian tree species. Can. 7. Margolis, H., R. Oren, D. Whitehead, and M. R. Kauf-
For. Res. 4:1-7. mann. 1995. Leaf area dynamics of conifer forests. pp.
Mailly, D. 1996. The influence of disturbance regime on 181-223. In W. K. Smith and T. M. Hinckley (eds.),
forest succession in the CWHdm subzone, British Co- Ecophysiology of Coniferous Forests. Academic Press, New
lumbia. Doctoral thesis, Univ. of B.C., Vancouver. York.
 REFERENCES CITED R-33

Marion, G. M. 1979. Biomass and nutrient removal in long- Maser, C., R. F. Tarrant, J. M. Trappe, and J. F. Franklin
rotation stands. pp. 98-110. In Proc. Impact of Intensive (eds.). 1988. From the Forest to the Sea: A Story of Fall-
Harvesting on Forest Nutrient Cycling. Coll. Environ. Sci. en Trees. USDA, For. Serv., Portland, Oreg. Gen. Tech.
& For., SUNY, Syracuse. Rept. PNW-GTR-229. 153 pp.
Marks, G. C., and T. T. Kozlowski, 1973. Ectomycorrhizae: Mason, B. 1966. Principles of Geochemistry. 3rd ed. Wiley,
Their Ecology and Physiology. Academic Press, New York. New York. 329 pp.
444 pp. Mason, D. I. 1915. The life history of lodgepole pine in the
Marks, P. 1974. The role of pin cherry (Prunus pensylvanica Rocky mountains. USDA, For. Serv., Bull. 154. 33 pp.
L.) in the maintenance of stability in northern hardwood Materna, J. 1979. The influence of air pollution on the sta-
ecosystems. Ecol. Monogr. 44:73-88. bility of spruce ecosystems. pp. 343-354. In E. Klimo
Marland, G., T. A. Boden, and R. J. Andres. 2002. Global, (ed.), Stability of Spruce Forest Ecosystems. Proc. Symp.
regional and national CO, emissions. In Trends: A Com- MAB/TUFRO, Univ. of Agri., Brno, Czechoslovakia.
pendium of Data on Global Change. Carbon Dioxide Infor- Matthews, J. D. 1963. Factors affecting the production of
mation Analysis Centre, Oak Ridge National Laboratory, seed by forest trees. For. Abs. 24:i-xiii.
U.S. Department of Energy, Oak Ridge, TN. Mattson, W. J., Jr. 1980. Herbivory in relation to plant ni-
Marriage, P. B., and S. I. Warwick. 1980. The selection of trogen content. Ann. Rev. Ecol. Syst. 11:119-161.
herbicide resistant biotypes of Chenopodium album L. , and N. D. Addy. 1975. Phytophagous insects as reg-
Weed Res. 20:9-15. ulators of forest primary production. Science
Marshall, A. J., and H. S. deS. Disney. 1957. Experimental 190:515-522.
induction of the breeding seasons in a xerophilous bird. May, R. H. 1975. Stability in ecosystems: some comments.
Nature 180:647. pp. 161-168. In W. H. van Dobben and R. H. Lowe-
Marshall, W. H. 1954. Ruffed grouse and snowshore hare McConnell (eds.), Unifying Concepts in Ecology. W. Junk,
populations on the Cloquet Experimental Forest, Min- b.v., Publishers, The Hague.
nesota.7.Wild. Manage. 18:109-112. . 1978. The evolution of ecological systems. Sci.
Martin, C. W., and R. S. Pierce. 1979. Clearcutting patterns Amer. 239(3):160-175.
affect nitrate and calcium in streams in New Hampshire. Mayewski, P. A., and F. White. 2002. The Ice Chronicles. The
J. For. 78:268-272. Quest to Understand Global Climate Change. University
Martin, P. S. 1966. Africa and Pleistocene overkill. Nature Press of New England, University of New Hampshire,
212:339-342. Hanover, NH. 233 pp.
. 1967. Africa and Pleistocene overkill. pp. 75-120. In McArdle, R. E., W. H. Meyer, and D. Bruce. 1961. The
P. S. Martin and H. E. Wright (eds.), Pleistocene Extinc- yield of Douglas-fir in the Pacific Northwest. USDA For.
tions. Yale Univ. Press, New Haven. Serv., Wash. DC., Tech. Bull. 201. Revised. 74 pp.
Martin, W. L. 1985. Post-clearcutting forest floor nitrogen McBean, G. A., and G. Thomas. 1991. Regional climate
dynamics and regeneration response in the Coastal West- change for the Pacific Northwest. pp. 19-30. In G. Wall
ern Hemlock wet subzone. Doctoral thesis, University of (ed.), Implications of Climate Change for Pacific Northwest
B.C., Vancouver. 350 pp. Forest Management. Dept. Geog. Publ., Ser. Occ. Pap.
Martin, P. 1992. EXE: A climatically sensitive model to No. 15. Univ. of Waterloo, Ontario.
study climate change and CO, enhancement effects on McBeath, J. H., G. P. Juday, G. Weller, and M. Murray
forests. Aust. 7. Bot. 40:717-735. (eds.). 1984. The potential effects of carbon dioxide-
Martin, P. H. 1990. Forest succession and climate change: induced climatic change in Alaska. Conf. Proc. Sch. Agr.
Coupling land-surface processes and ecological dynam- and Land Res. Memt., Univ. of Alaska, Fairbanks. Misc.
ics. Cooperative Thesis between the University of Cali- Publ. 83-1. 208 pp.
fornia at Berkeley and the National Center for McClaugherty, C. A.,J.Pastor,J.D. Aber, andJ.M. Melillo.
Atmospheric Research (NCAR), Boulder, CO. Universi- 1985. Forest litter decomposition in relation to soil ni-
ty Microfilms International (UMI) Catalog Number trogen dynamics and litter quality. Ecology 66:266-275.
9126692, 342 pp. McCook, L. J. 1994. Understanding ecological community
Marx, D. H. 1991. The practical significance of ectomycor- succession. Causal models and theories, a review.
rhizae in forest establishment. pp. 54-90. In Ecophystology Vegetatio 110:115-147.
of Ectomycorrhizae ofForest Trees. The Marcus Wallenberg McDermott, R. E. 1954. Seedling tolerance as a factor in
Foundation, Symposia Proc. 7, Falun, Sweden. bottom-land timber succession. Mo. Agr. Exp. Stat. Res.
, and W. C. Brian. 1976. The significance of mycor- Bull. 557. 11 pp.
rhizae to forest trees, pp. 107-117. In B. Bernier and McFee, W. W., and E. L. Stone. 1965. Quantity, distribution
C. H. Winget (eds.), Forest Soils and Forest Land Manage- and variability of organic matter and nutrients in a forest
ment. Laval Univ. Press, Quebec. podzol in New York. Soil Sci. Soc. Amer. Proc. 29:432-436.
R-34 REFERENCES CITED

McFarlane, P. 2003. Products from sustainably managed Mekaru, T., and G. Uehara. 1972. Anion absorption in fur-
forests. Unpublished public lecture presentation. De- ruginous tropical soils. Soil Sci. Soc. Amer. Proc.
partment of Wood Sciences, Faculty of Forestry, Univer- 36:296-300.
sity of British Columbia, Vancouver, BC. Melillo,J.M.,J.D. Aber, andJ.F. Muratore. 1982. Nitrogen
McIntosh, R. P. 1980. The relationship between succession and lignin control of hardwood leaf litter decomposition
and the recovery process in ecosystems. pp. 11-62. In J. dynamics. Ecology 63:621-626.
Cairns, Jr. (ed.), The Recovery Process in Damaged Ecosys- Menaut,J. C., J. Gignoux, C. Prado, and J. Clobert. 1990.
tems. Ann Arbor Science Publishers, Ann Arbor, MI. Tree community dynamics in a humid savanna of the
McLachlin, R. 1983. Winter wren dispersion in coastal Céte d’Ivoire: Modelling the effects of fire and compe-
western hemlock forest (dry subzone) in southwestern tition with grass and neighbours. 7. Biogeogr. 17:471-
British Columbia. Doctoral thesis, Univ. of B.C., Van- 481.
couver. Messier, C. 1993. Factors limiting early growth of western
McLean, D. L., and M. R. Lockman. 1967. Forest Fire red cedar, western hemlock and Sitka spruce seedlings on
Losses in Canada, 1966. Can. Dept. Forest and Rural ericaceous-dominated clearcut sites in coastal British Co-
Devel., For. Fire Res. Inst. Rept. lumbia. For: Ecol. Manage. 60:181—206.
McMurthie, R. E., and J. J.Landsberg. 1992. Using a simu- , and J. P. Kimmins. 1990. Nutritional stress in Picea
lation model to evaluate the effects of water and nutrients sitchensis plantations in coastal British Columbia: The ef-
on the growth and carbon partitioning of Pinus radiata. fects of Gaultheria shallon and declining soil fertility.
For. Ecol. Manage. 52:243-260. Water Air Soil Pollut. 54:257-267.
McNaughton, F. G., and T. A. Black. 1973. A study of evapo- Messier, C., M. J. Fortin, F. Schmiegelow, F. Doyon, S. G.
transpiration from a Douglas-fir forest using the energy Cumming, J. P. Kimmins, B. Seely, C. Welham, and J.
balance approach. Water Resources Res. 9:1579-1590. Nelson. 2003. Modelling tools to assess the sustainability
McNaughton, S. J., and L. L. Wolf. 1973. General Ecology. of forest management scenarios. In P. J. Burton, C.
Holt, Rinehart and Winston, New York. 710 pp. Messier, D. W. Smith, and W. L. Adamowicz (eds.),
MeNeill, W. M. 1955. Studies of surface vegetation and nat- Towards Sustainable Management of the Boreal Forest. NRC
ural regeneration in a mature Scots pine wood in Ab- Research Press, Ottawa, Ontario.
erdeenshire. Forestry 28:49-65. Messina, M. G., W. J. Dyck, and I. R. Hunter. 1985. The
McPhee, W. W., and E. L. Stone. 1965. Quantity, distribu- nutritional consequences of forest harvesting, with spe-
tion and variability of organic matter and nutrients in a cial reference to the exotic forests of New Zealand,
forest podzol in New York. Soil Sci. Soc. Amer. Proc. IEA/BE Project CPC-10. Report #1. Forest Research In-
29:432-436. stitute, Rotorua, New Zealand. 57 pp.
McPherson, J. K., and C. H. Muller. 1969. Allelopathic ef- Metz, L. J., and M. H. Farrier. 1971. Prescribed burning and
fects of Adenostoma fasciculatum. “Chamise” in the Cali- soil mesofauna on the Santee Experimental Forest. pp.
fornia chaparral. Ecol. Monogr. 39:177-198. 100-106. In Prescribed Burning Symposium. USDA For.
McWhirter, N., and R. McWhirter. 1975. Guiness Book of Serv. SE For. Exp. Sta., Ashville, SC.
Records. Guiness Superlatives, Enfield, UK. Mielke, D. L., H. H. Shugart, and D. C. West. 1978. A
Meadows, D. H., D. L. Meadows, J. Randers, and W. W. Stand Model for Upland Forests of Southern Arkansas. Oak
Behrens, III. 1972. The Limits to Growth. A Report for the Ridge National Lab, Oak Ridge, TN, ORNL/TM-6225.
Club of Rome’s Project on the Predicament of Mankind. Pan Mielke, D. L., H. H. Shugart, and D. C. West. 1978. A stand
Books, London. 205 pp. model for upland forests of southern Arkansas. Oak
Means, J. E. (ed.). 1982. Forest Successional Stand Development Ridge National Laboratory, Oak Ridge, TN,
Research in the Northwest. Forest Res. Lab., Oregon State ORNL/TM-6225.
Univ., Corvallis. 170 pp. Mikola, P. 1962. Temperature and tree growth near the
Means, T. H. 1927. Fog precipitated by trees. 7 For. northern timber line. pp. 265-274. In T. T. Kozlowski
25:1015-1016. (ed.), Tree Growth. Ronald, New York.
Meentemeyer, V. 1978. Macroclimate and lignin control of . 1970. Mycorrhizal innoculation in afforestation. Int.
litter decomposition rates. Ecology 59:465—472. Rev. For. Res. 3:123-197.
, E. O. Box, and R. Thomson. 1982. World patterns Millard, P., and M. F. Proe. 1993. Nitrogen uptake, parti-
and amounts of terrestrial plant litter production. tioning and internal cycling in Picea sitchensis (Bong.)
BioScience 32:125-128. Carr. as influenced by nitrogen supply. New Phytol.
Meiklejohn, J. 1953. The effect of bush burning on the mi- 125:113-119.
croflora of some Kenya soils. Proc. 6th Int. Conf. Microbi- Miller, F R., and K. L. Adams (eds.). 1992. Wise Manage-
ol. 10:317-319. ment of Tropical Forests: Proceedings of the Oxford Conference
 REFERENCES CITED R-35

on Tropical Forests 1992. Oxford Forestry Institute, Ox- , A. F. M. van Hees, and H. H. Bartelink. 1991. Suc-
ford, UK. cession models as an aid for forest management in mixed
Miller, G. R., D. Jenkins, and A. Watson. 1966. Heather stands in The Netherlands. For Ecol. Manage. 42:
performance and red grouse populations. 1. Visual esti- 111-127.
mates of heather performance. 7. Appl. Ecol. 3:313-326. Mohren, G. M. J., A. F. M. van Hees, and H. H. Bartelink.
Miller, H. G. 1979. The nutrient budgets of even-aged 1991. Succession models as an aid for forest management
forests. pp. 221-256. In E. D. Ford, J. Atterson, and in mixed stands in The Netherlands. For: Ecol. Manage.
D. C. Malcolm (eds.), The Ecology of Even-aged Forest 42:111-127.
Plantations. Inst. Terrestrial Ecol., Cambridge. Moller, C. M., D. Muller, and J. Nielson. 1954. Graphic
, J. M. Cooper, andJ. D. Miller, 1976. Effect of nitro- presentation of dry matter production of European
gen supply on nutrients in litter fall and crown leaching beech. Forst. Fors. Danm. 21:327-335.
in a stand of Corsican pine.7.Appl. Ecol. 13:233-248. Moncrief, L. W. 1970. The cultural basis for our environ-
, J. M. Cooper, J. D. Miller, and O. J. Pauline, 1979. mental crisis. Science 170:508-512.
Nutrient cycles in pine and their adaptation to poor soils. Money, D. C. 1972. Climate, Soils and Vegetation. 2nd ed.
Can. fF. For. Res. 9:19-26. Univ. Tutorial Press, London. 272 pp.
, andJ.D. Miller. 1976. Effects of nitrogen supply on Monk, C. D. 1966. An ecological significance of evergreen-
net primary production in Corsican pine. 7. Appl. Ecol. ness. Ecology 47:504—-505.
13:249-256. Monning, E., and J. Byler. 1992. Forest Health and Ecolog-
, and B. L. Williams. 1968. Research on forest soils ical Integrity in the Northern Rockies. USDA For. Serv.
and tree nutrition. pp. 143-147. In Rep. For. Res. London. FPM Rep. 92-7. 19 pp.
Miller, R. S. 1964. Ecology and distribution of pocket go- Montes, R., and E. Medina. 1977. Seasonal changes in nutri-
phers (Geomyidae) in Colorado. Ecology 45:256-272. ent content of leaves of savanna trees with different eco-
Miller, W. F. 1966. Annual changes in foliar nitrogen, phos- logical behaviour. Geo. Eco. Trop. 1:295-307.
phorus and potassium levels of loblolly pine (Pinus taeda Mooney, H. A., and M. Godran (eds.). 1983. Disturbance and
L.) with site and weather factors. Plant Soil 24:369-378. Ecosystems. Ecological Studies 44. Springer-Verlag, New
Miller, C., and D. L. Urban. 1999. Forest pattern, fire, and York. 297 pp.
climatic change in the Sierra Nevada. Ecosystems 2:76-87. , and F. Shropshire. 1967. Population variability in
Milne, A. 1957. The natural control of insect populations. temperature related to photosynthetic acclimation. Oecol.
Can. Entomol. 89:193-213. Plant. 2:1-13.
. 1962. On a theory of natural control of insect popu- Moore, J. M., and R. W. Wein. 1977. Viable seed popula-
lation.7.Theoret. Biol. 3:19-26, 32-42, 48-50. tions by soil depth and potential site recolonization after
Mitchell, J. E., J. B. Waide, and R. L. Todd. 1975. A pre- disturbance. Can. 7. Bot. 55:2408-2412.
liminary compartment model of the nitrogen cycle in a Moreau, R. E. 1966. The Bird Faunas ofAfrica and its Islands.
deciduous forest ecosystem. pp. 41-57. In F. G. Howell Academic Press, New York. 424 pp.
et al. (eds.), Mineral Cycling in Southeastern Ecosystems. Morgenstern, E. K. 1969. Genetic variation in seedlings of Picea
Tech. Info. Centre, Oak Ridge, TN. 898 pp. CONF- mariana (Mill.) B.S.P. I. Correlation with ecological factors.
740513. IL. Variation patterns. Si/v. Genet. 18:151—160, 161-167.
Mitsch, W. J., R. W. Bosserman, andJ.M. Klopatek. 1981. Morris, R. F. 1963. The dynamics of epidemic spruce bud-
Energy and Ecological Modelling. Elsevier Sci. Publ., New worm populations. Mem. Entomol. Soc. Can. 31:1-332.
York. 839 pp. Morrison, I. K. 1974. Mineral Nutrition of Conifers With
Mitton, J. B. 1995. Genetics and the physiological ecology Special Reference to Nutrient Status Interpretation: A
of conifers. pp. 1-36. In W. K. Smith and T. M. Hinckley Review of Literature. Can. For. Serv. No. 1343. 74 pp.
(eds.), Ecophysiology of Coniferous Forests. Academic Press, . 1984. Acid rain. A review ofliterature on acid depo-
New York. sition effects in forest ecosystems. For. Abstr. +5:483-506.
Mladenoff, D. J., and W. L. Baker. 1999. Spatial modelling Moseholm, L., B. Andersen, and J. Johnson. 1988. Acid Depo-
of forest landscape change: Approaches and applications. sition and Novel Forest Decline in Central and Northern Eu-
Cambridge University Press, Cambridge, UK. 352 pp. rope. Nordic Council of Ministers, Copenhagen. 91 pp.
Mohren, G. M. J. 1987. Simulation of forest growth applied Moser, M. 1959. Die kiinstliche Mykorrhizaimpfung an
to Douglas-fir stands in the Netherlands. Doctoral the- Forstpflanzen. Ill. Die Impfmethodic im Forstgarten.
sis. Landbouwuniversiteit, Wageningen, The Nether- Forstwiss. Cbl. 78:193-202.
lands. 184 pp. Moss, R. 1967a. Food selection and nutrition in ptarmigan
, and F, Kienast (eds.). 1991. Modeling forest succes- (Lagopus mutus). Symp. Zool. Soc. Lond. No.
sion in Europe. For. Ecol. Manage. 42:1-142. 21:207-216.
R-36 REFERENCES CITED

. 1967b. Probable limiting nutrients in the main food , R. O. Squire, R. Sands, and G. H. M. Will. 1984.
of red grouse (Lagopus lagopus scoticus). pp. 369-379. In K. Manipulation of water and nutrients in plantations of fast
Petrusewicz (ed.), Secondary Productivity of Terrestrial growing species. pp. 489-506. In IUFRO Symposium on
Ecosystems. IBP, Krakow. Fast Growing Plantations. So. Africa For. Res. Inst., Preto-
. 1972. Food selection by red grouse (Lagopus /. scoti- ria, South Africa.
cus) in relation to chemical compostion. 7. Anim. Ecol. National Academy of Sciences. 1975. Productivity of World
44:233-244. Ecosystems. Proc. Symp. Nat. Acad. Sci., Washington,
Mount, A. B. 1969. Eucalypt ecology as related to fire. Proc. DC. 166 pp.
Tall Timbers Fire Ecol. Conf. 9:75-108. NCSENC. 1995. Fundamental research in forest produc-
Mueller-Dombois, D., and H. Ellenberg. 1974. Aims and tion. pp 8-15. In North Carolina State Forest Nutrition
Methods of Vegetation Ecology. Wiley, New York. 547 pp. Coop. 24th Annual Report. Dept. of Forestry, N.C. State
Mukammal, E. I. 1965. Ozone as a cause of tobacco injury. Univ., Raleigh, NC.
Agric. Meteorol. 2:145—-165. Neary, D. C., G. J. Jokela, N. B. Comerford, S. R. Colbert,
Muller, C. H. 1969. Allelopathy as a factor in ecological and T. E. Cooksey. 1990. Understanding competition for
processes. Vegetatio 18:348-357. soil-nutrientation the key to site productivity on south-
. 1970. Phototoxins as plant habitat variables. Recent eastern coastal plain forests. pp. 432-450. In S. P. Gessel,
Adv. Phytochem. 3:105-121. D. S. Lacate, G. F Weetman, and R. F. Powers (eds.),
, R. B. Hanawalt, andJ.K. McPherson. 1968. Allelo- Sustained Productivity of Forest Soils. Proc. 7th N. Am. For-
pathic control of herb growth in the fire cycle of Califor- est Soils Conf: Faculty of Forestry, Univ. of British Colum-
nia chaparral. Bull. Torrey Bot. Club 95:225-231. bia, Vancouver.
Muller-Stoll, W. R. 1947. Der Einfluss der Ernahrung auf die Neiland, B. J. 1971. The forest-bog complex of southeast
xeromorphie der Hochmoor pflanzen. Planta 35:225-251. Alaska. Vegetatio 22:1-64.
Murcia, C. 1995. Edge effects in fragmented forests: Impli- Nelson, B. W., V. Kapos, J. B. Adams, W. J. Oliveira,
cations for conservation. Tiree 10:58-62. O. P. G. Braun, and I. L. Do-Amaral. 1994. Forest dis-
Murray, K. F. 1965. Population changes during the 1957-58 turbance by large blowdown in the Brazilian Amazon.
vole (Microtus) outbreak in California. Ecology Ecology 75:853-858.
46:163-171. Newman, K. 1979. Sapwood basal area as an estimator ofin-
Mustanoja, K. J., and A. L. Leaf. 1965. Forest fertilization dividual tree growth. Master’s thesis, Sch. For., Ore.
research, 1957-1964. Bot. Rev. 31:151-246. State Univ., Corvallis. 65 pp.
Mutch, R. W. 1970. Wildland fires and ecosystems—A hy- Newton, M., B. A. El Hassan, andJ. Zavitovski. 1968. Role
pothesis. Ecology 51:1046-105 1. of red alder in western Oregon forest succession. pp.
Myers,J.H. 1988. Can a general hypothesis explain popula- 73-84. In J. M. Trappe et al. (eds.), Biology of Alder.
tion cycles of forest Lepidoptera? Adv. Ecol. Res. USDA For. Serv., PNW For. and Range Exp. Sta., Port-
18:179-242. land, Ore.
. 1990. Population cycles of western tent caterpillars: Nicholson, A. J. 1933. The balance of populations. 7. Anim.
Experimental introductions and synchrony of fluctua- Ecol. 2:132-148, 178.
tions. Ecology 71:989-995. Niciporovic, A. A. 1968. Evaluation of productivity by study
Myers, K. 1967. Morphological changes in the adrenal of photosynthesis as a function of illumination. pp.
glands of wild rabbits. Nature 213:147-150. 261-270. In Functioning of Terrestrial Ecosystems at the Pri-
Myers, N. 1980. Conversion of Tropical Moist Forest. Nat. mary Production Level. UNESCO, Paris.
Acad. Sci., Wash. D. C. 205 pp. Niering, W. A., and F. E. Egler, 1955. A shrub community of
. 1982. Depletion of tropical moist forests: A com- Viburnum lentago stable for twenty-five years. Ecology
parative review of rates and causes in the three main re- 36:356-360.
gions. Acta Amaz. 12:745-758. Nihlgard, B. 1972. Plant biomass, primary production
Nageli, W. 1946. Weitere Untersuchungen iiber die Wind- and distribution of chemical elements in a beech and a
verhaltnisse im Bereich von Windschutzstreifen. Mitteil. planted spruce forest in South Sweden. Oikos
schweiz. Anstalt forstl. Versuch. 24:659-737. 23:69-81.
Nambiar, E. K. S., and D. N. Fife. 1991. Nutrient retranslo- Nilsson, L. O., R. F. Hiittl, and U. T. Johansson. 1995.
cation in temperate conifers. Tree Physiol. 9:185—-207. Nutrient Uptake and Cycling in Forest Ecosystems. Kluwer
, and R. Sands. 1993. Competition for water and nu- Academic Publ., Dordrecht. 685 pp.
trients in forests. Can. 7. For. Res. 23:1955-1968. Niu, C-H., K. Cromak, Jr., and C. Y. Li. 1989. Association
, R. Squire, R. Cromer, J. Turner, and R. Boardman. of nitrogen-fixing bacteria with eclomycorrhizae in
1990. Management of water and nutrient relations to in- Douglas-fir. pp. 253-254. In S. P. Gessel, D. S. Lacate,
crease forest growth. For: Ecol. Manage. 30:1-486. G. F. Weetman, and R. F. Powers (eds.), Sustained Produc-
 REFERENCES CITED R-37

tivity ofForest Soils. Proc. 7th N. Am. For. Soils Conf. Facul- . 1971. Fundamentals of Ecology. 3rd ed. W. B. Saun-
ty of Forestry, Univ. of B.C., Vancouver. ders Co., Toronto. 574 pp.
Noble, I. R., and R. O. Slatyer. 1977. Post-fire succession of Odum, H. T. 1956. Efficiencies, size of organisms, and com-
plants in mediterranean ecosystems. pp. 27-63. In munity structure. Ecology 37:592-597.
Mooney, H. A. and C. E. Conrad (eds.), Proc.—Symp. on . 1957. Trophic structure and productivity of Silver
the Environmental Consequences of Fire and Fuel Manage- Springs, Florida. Ecol. Monogr. 27:55-112.
ment in Mediterranean Ecosystem. USDA For. Serv. Gen. , and R. F. Pigeon (eds.). 1970. A Tropical Rain For-
Tech. Rept. WO-3. est. A Study of Irradiation and Ecology at El Verde,
, and R. O. Slatyer, 1980. The effect of disturbance Puerto Rico. Nat. Tech. Info. Serv., Springfield, Vt. 1678
on plant succession. Proc. Ecol. Soc. Aust. 10:135-145. PP:
, and R. O. Slatyer. 1980b. The use of vital attributes Oke, T. R. 1978. Boundary Layer Climates. Methuen, Lon-
to predict successional changes in plant communities don. 372 pp.
subject to recurrent disturbance. Vegetatio 43:5-21. O’Laughlin, J., R. L. Livingston, R. Thier, andJ.T. Powers
Nodvin, S. C., and T. A. Waldop. (ed) 1991. Fire and the En- (eds.). 1989. Maintaining the Long-Term Productivity ofPa-
vironment: Ecological and Cultural Perspectives. Proc. Int. cific Northwest Forest Ecosystems. Vimber Press, Portland,
Symp., USDA For. Serv., S. E. For. Expt. Sta., Asheville, Ore. 256 pp.
NC. Gen-Tech, Rpt. SE-69. 426 pp. Old, S. M. 1969. Microclimate, fire and plant production in
Nohrstedt, H-O. 1988. Nitrogen fixation (C,H, reduction) an Illinois prairie. Ecol. Monogr. 39:355-384.
in birch litter. Scand. 7. For. Res. 3:17-23. Oliver, C. D. 1981. Forest development in North America
Nortcliff, S., and J. B. Thornes. 1977. Water and cation following major disturbances. For Ecol. Manage. 3:
movement in a tropical rain forest environment. 1. Objec- 153-168.
tives, experimental design and preliminary results. Dis- , and B. C. Larson. 1990. Forest Stand Dynamics.
cussion Paper No. 62. London School of Economics, McGraw-Hill, New York. 467 pp.
London. Olmsted, N. W., andJ. D. Curtis. 1947. Seeds of the forest
Noss, R. F. 1992. The wildlands project: Land conservation floor. Ecology 28:49-52.
strategy. Wild Earth 1(Special Issue):10-25. O’Loughlin, C. L. 1974. A study of tree root strength dete-
Noss, R. F., and A. Cooperrider. 1994. Saving Nature's Lega- rioration following clearcutting. Can. 7. For. Res.
cy: Protecting and Restoring Biodiversity. Island Press, 4:107-113.
Washington, DC. Olson J. S. 1958. Rates of succession and soil changes on
Nuszdorfer, F.C. 1982. Vertical distribution and biomass of southern Lake Michigan sand dunes. Bot. Gaz.
fine roots in three subalpine forest plant associations in 119(3):125-170.
southwestern British Columbia. Master’s thesis, Univer- . 1963. Energy storage and the balance of producers
sity of B.C., Vancouver. 172 pp. and decomposers in ecological systems. Ecology
Nye, P. H. 1954. Some soil forming processes in the humid 44:322-331.
tropics. I. A field study of a catena in a West African For- . 1982. Earth’s vegetation and atmospheric carbon
est. f. Soil Sci. 5:7-21. dioxide. pp. 388-398. In W. C. Clark (ed.), Carbon Diox-
. 1961. Organic matter and nutrient cycles under ide Review: 1982. Clarendon Press, Oxford.
moist tropical forest. Plant Soil 13:333-346. O’Neill, R. V. 1976. Ecosystem persistence and _het-
, and D.J.Greenland. 1960. The Soil Under Shifting erotrophic regulation. Ecology 57:1244-1253.
Cultivation. Commonw. Bur. Soil Sci., Harpenden, Tech. ,and D. L. DeAngelis. 1981. Comparative productiv-
Commun. No. 51, 156 pp. ity and biomass relations of forest ecosystems. pp.
, and D. J. Greenland. 1964. Changes in the soil after 411-449. In D. E. Reichle (ed.), Dynamic Properties of For-
clearcutting tropical forest. Plant Soil 21:101-112. est Ecosystems. International Biological Programme 23.
Oberlander, G. T. 1956. Summer fog precipitation of the Cambridge Univ. Press, Cambridge.
San Francisco peninsula. Ecology 37:85 1-852. , W. F. Harris, B. S. Ausmus, and D. E. Reichle. 1975.
Oberle, M. 1969. Forest fires: suppression policy has its eco- A theoretical basis for ecosystem analysis with particular
logical drawbacks. Science 165:568-571. reference to element cycling. pp. 28-40. In F. W. Howell,
O’Connor, F. B. 1957. An ecological study of the enchy- J. B. Gentry, and M. H. Smith (eds.), Mineral Cycling in
traeid worm population of a coniferous forest soil. Orkos Southeastern Ecosystems. ERDA Symp. Ser., U.S. Energy
8:161-199. Res. and Devel. Admin. CONF-7405 13.
Odum, E. P. 1962. Relationships between structure and , and D. E. Reichle. 1980. Dimensions of ecosystem
function in the ecosystem. Jpn. 7. Ecol. 12:108-118. theory. pp. 11-26. In R. H. Waring (ed.), Forests: Fresh
. 1969. The strategy of ecosystem development. Perspectives from Ecosystem Analysis. Oregon State Univ.
Science 164:262-270. Press, Corvallis.
R-38 REFERENCES CITED

Oosting, H. J. 1956. The Study of Plant Communities: An In- . 1963. Cold resistance in woody plants. Bot. Rev.
troduction to Plant Ecology. 2nd ed. W. H. Freeman and 29:123-201.
Co., San Francisco. 440 pp. Parkinson, J. A. 1984. Nitrogen and phosphorus retranslo-
Oppenheimer, H. R. 1951. Summer drought and water bal- cation from needles of Douglas-fir growing on three site-
ance of plants growing in the near east. 7. Ecol. types. Master’s thesis, Univ. of B.C., Vancouver. 116 pp.
39:357-362. Pastor,J.,R. H. Gardner, V. H. Dale, and W. M. Post. 1987.
Oreskes, N., K. Shrader-Frechette, and K. Belitz. 1994. Ver- Successional changes in nitrogen availability as a poten-
ification, validation, and confirmation of numerical mod- tial factor contributing to spruce declines in boreal
els in the earth sciences. Science 263:641-646. North America. Can. 7. For. Res. 17: 1394-1400.
Orians, G. H. 1975. Diversity, stability and maturity in nat- .and W. M. Post. 1985. Development ofa Linked Forest
ural ecosystems. pp. 139-150. In W. H. van Dobben and Productivity-Soil Process Model. Oak Ridge National Labo-
R. H. Lowe-McConnell (eds.), Unifying Concepts in Eco- ratory, Env. Sci. Div. Pub. 2455. 162 pp.
logy. W. Junk, b.v., Publishers, The Hague. , and W. M. Post. 1986. Influence of climate, soil
, and F. A. Pitelka. 1960. Range management for the miosture, and succession on forest carbon and nitrogen
animal ecologist. Ecology +1:406. cycles. Biogeochemistry 2:3—27.
Ostman, N. L., and G. T. Weaver. 1982. Autumnal nutrient , and W. M. Post. 1988. Response of northern forests
transfers by retranslocation, leaching and litter fall in a to CO)-induced climatic change. Nature 334:55—58.
chestnut oak forest in southern Illinois. Can. 7. For. Res. Pastor, J., and W. M. Post. 1985. Development of a linked
12:49-51. forest productivity-soil process model. Oak Ridge Na-
Ovington, J. D. 1957. Dry matter production by Pinus tional Laboratory, Oak Ridge, TN. 162 pp.
sylvestris L. Ann. Bot. 21:287-314. Pastor, J., and W. M. Post. 1986. Influence of climate, soil
. 1959. Mineral content of plantations of Pinus moisture, and succession on forest carbon and nitrogen
sylvestris L. Ann. Bot. 23:75-88. cycles. Biogeochemistry 2:3-27.
. 1961. Some aspects of energy flow in populations of Pauley, S. S. 1958. Photoperiodism in relation to tree im-
Pinus sylvestris. Ann. Bot. 25:12-20. provement. pp. 557-571. In K. V. Thimann (ed.), The
. 1962. Quantitative ecology and the woodland Physiology ofForest Trees. Ronald, New York.
ecosystem concept. Adv. Ecol. Res. 1:103-192. , and T. O. Perry. 1954. Ecotypic variation of the pho-
Pacala, S$. W., C. D. Canham, J. Saponara, J. A. Silander, toperiodic response in Populus. 7. Arnold Arbor. 35:167-188.
R. K. Kobe, and E. Ribbens 1996. Forest models defined Pausas, J. G. 1999. Response of plant functional types to
by field measurements: Estimation, error analysis and dy- changes in the fire regime in Mediterranean ecosystems:
namics. Ecol. Monogr. 66(1):1-43. A simulation approach.7. Veg. Sci. 10:717-722.
, C. D. Canham, and J. A. Silander,
Jr. 1993. Forest Pausas, J. G., M. P. Austin, and I. R. Noble. 1997. A forest
models defined by field measurements: I. The design of a simulation model for predicting eucalypt dynamics and
northeastern forest simulator. Can. 7. For. Res. habitat quality for arboreal marsupials. Ecol. Appl.
23:1980-1988. 7:921-933.
Pacala, S. W., C. D. Canham, J. Saponara, J. A. Silander, Pearlstine, L., H. McKellar, and W. Kitchens. 1985. Model-
R. K. Kobe, and E. Ribbens. 1996. Forest models defined ling the impacts of a river diversion on bottomland forest
by field measurements: Estimation, error analysis and dy- communities in the Santee River floodplain, South Car-
namics. Ecol. Monogr. 66:1-43. olina. Ecol. Model. 29:283-302.
Pacala, S$. W., C. D. Canham, and J. A. J. Silander. 1993. Pearman, G. I. (ed.). 1980. Carbon dioxide and climate: Aus-
Forest models defined by field measurements: I. The de- tralian Research. Aust. Acad. Sci., Canberra. 217 pp.
sign of a northeastern forest simulator. Can. 7. For. Res. Pears, N. V. 1968. The natural altitudinal limit of forest in
23:1980-1988. the Scottish Grampians. Oikos 19:71-80.
Paine, R. T. 1966. Food web complexity and species diver- . 1972. Interpretation problems in the study of tree-
sity. Amer. Nat. 100:65-75. line fluctuations. In J. A. Taylor (ed.), Forest Meteorology:
Palmer, R. W. V. (ed.). 1968. Wind Effects on the Forest. Ox- An Aberystinyth Symposium. Univ. College of Wales,
ford Univ. Press, London. 93 pp. Aberystwyth.
Park, T. 1962. Beetles: Competition and populations. Science Peng, C. H.,J.X. Liu, Q. L. Dang, M. J. Apps and H.Jiang.
138:1369-1375. 2002. TRIPLEX: A generic hybrid model for predicting
, PH. Leslie, and D. B. Mertz. 1961. Genetic strains forest growth and carbon and nitrogen dynamics. Eco-
of Tribolium: Their primary characteristics. Physiol. Zool. logical Modelling 153:109-130.
34:62-80. Perera, A. H., L. J. Buse, and M. Weber (eds.) In press.
Parker, J. 1953. Photosynthesis of Picea excelsa (P. abies) in Emulating Natural Forest Landscape Disturbances: Concepts
winter. Ecology 34:605-609. and Applications. Columbia University Press, New York.
 REFERENCES CITED R-39

Perrin, C. J. 1981. On the summer regulation of nitrogen Pielou, E. C. 1981. The usefulness of ecological models: A
and phosphorus transport in a small stream of southwest- stock taking. Q. Rev. Biol. 56:17-31.
ern British Columbia. Master’s thesis, Univ. of B.C., Van- . 1991. After the Ice Age. The Return ofLife to Glaciated
couver. 144 pp. North America. Univ. of Chicago Press, Chicago. 366 pp.
Perry, T. O. 1962. Racial variation in the day and night tem- Pierce, R. S., C. W. Martin, C. C. Reeves, G. E. Likens, and
perature requirements of red maple and loblolly pine. F. H. Bormann. 1972. Nutrient loss from clearcutting in
For. Sci. 8:336-344. New Hampshire. pp. 285-295. In S. C. Csallany, T. G.
1971. Dormancy of trees in winter. Science McLaughlin, and W. D. Striffler (eds). Watersheds in
171:29-36. Transition. Proc. Symposium, Ft. Collins, Colo.
Peters, R. L., and T. E. Lovejoy (eds.). 1992. Global Warming Pignatti, S., and A. U. Savoia. 1989. Early use of the succes-
and Biological Diversity. World Wildlife Fund. Conference on sion concept by G. M. Lancisi in 1714. Vegetatio
Consequences ofthe Greenhouse Effect for Biological Diversity. 84:113-115.
Yale University Press, New Haven. 386 pp. Pimental, D., W. Dritschilo, J. Krummel, and J. Kutzman.
Peterson, C. J., W. P. Carson, B. C. McCarthy, and S. T. A. 1975. Energy and land constraints in food protein pro-
Pickett. 1990. Microsite variation and soil dynamics duction. Science 190:754-761.
within newly created treefall pits and mounds. Oikos , L. E. Hurd, A. C. Bellotti et al. 1973. Food produc-
58:39-46. tion and the energy crisis. Science. 182:443-449.
Pfister, R. D. 1976. Forest Habitat Type Classification, , W. D. Nagel, and J. L. Madden. 1963. Space-time
Mapping, and Ecoregions in the Northern Rocky Moun- structure of the environment and the survival of parasite-
tains. Discussion paper, XVI ['UFRO Congress, Oslo. host systems. Amer. Nat. 97:141-167.
, and S. F. Arno. 1980. Classifying forest habitat , E. C. Terhune, R. Dyson-Hudson et al. 1976. Land
types based on potential climax vegetation. For. Sci. degradation: Effects on food and energy resources.
26:52-70. Science 194:149-155.
, B. L. Kovalchik, S. F. Arno, and R. C. Presby. 1977. Pisek, A., W. Larcher, W. Moser, and I. Pack. 1969. Kardi-
Forest Habitat Types of Montana. USDA For. Serv. Gen. nale ‘Iemperaturbereiche der Photosynthese und Grenz-
Tech. Rept. INT-34. Intermountain For. and Range Exp. temperaturen des Lebens der Blatter verschiedener
Sta. 174 pp. Spermatophyten. HI Temperaturabhangigkeit und opti-
Phillips,J.1931. The biotic community. 7. Ecol. 19:1-24. maler ‘Iemperaturbereich der Netto-Photosynthese.
. 1934. Succession, development, the climax and the Flora (Abt. B) 158:608-630.
complex organism: An analysis of concepts. Part I. 7. Ecol. Pitelka, F. A. 1957. Some aspects of population structure in
22:554-571. the short-term cycle of the brown lemming in northern
. 1935a. Succession, development, the climax and the Alaska. Cold Spr. Hbr. Symp. Quart. Biology 22:237-251.
complex organism: An analysis of concepts. Part IL. 7. . 1964. The nutrient recovery hypothesis for arctic
Ecol. 23:210-246. microtive cycles. I. Introduction. pp. 55-56. In A. J. Crisp
. 1935b. Succession, development, the climax and the (ed.), Grazing in Terrestrial and Marine Environments.
complex organism: An analysis of concepts. Part II. 7. Blackwell Sci. Publ., Oxford.
Ecol. 22:488-508. Pogrebniak, P. S. 1930, tiber die Methodik von Standort-
Phillipson, J. 1966. Ecological Energetics. Edward Arnold, suntersuchungen in Verbindung mit Waldtypen.
London. 57 pp. Verh.II. Int. Congr. Forstl. Versuchsanstalten, 1929.
Phipps, R. L. 1979. Simulation of wetlands forest dynamics. Stockholm.
Ecol. Model. 7:257-288. Pojar, J., K. Klinka, and D. V. Meidinger. 1987. Biogeocli-
Pianka, E. R. 1970. On r- and K-selection. Amer. Nat. matic ecosystem classification in British Columbia. For:
104:592-597. Ecol. Manage. 22:119-154.
Pickett, S. T. A., and P. S. White (eds.). 1985. The Ecology of Pollard, W. G. 1979. The prevalence of earthlike planets.
Natural Disturbance and Patch Dynamics. Academic Press, Amer. Sct. 67:653-659.
New York. 472 pp. Polster, H. 1950. Die physiologischen Grundlagen der Stof-
, 8. L. Collins, and J.J. Armesti. 1987a. Models, mech- ferzeugung im Walde. Bayerischer Landwirtschaftsverlag,
anisms and pathways of succession. Bot. Rev. 53:335-371. Miinchen.
, §. L. Collins, andJ. J.Armesti. 1987b. A hierarchi- Poole, T. B. 1961. An ecological study of the Collembola in
cal consideration of causes and mechanisms of succes- a coniferous forest soil. Pedobiologia 1:113-137.
sion. Vegetatio 69:109-114. Population Reference Bureau. 1983. World Population Data
Pickett, S. T. A., and M. L. Cadenasso. 1995. Landscape Sheet. Pop. Ref. Bureau, Inc., Washington, DC.
ecology: spatial heterogeneity in ecological systems. Precht, H. et al. 1973. Temperature and Life. Springer-Verlag,
Science 269:33 1-334. New York. 779 pp.
R-40 REFERENCES CITED

Prentice, I. C., M. T. Sykes, and W. Cramer. 1993. A simu- Pugnaire, F. I., and F. S. Chapin III. 1993. Controls over nu-
lation model for the transient effects of climate change trient resorption from leaves of evergreen mediterranean
on forest landscapes. Ecol. Model. 65:5 1-70. species. Ecology 74:124-129.
Prentice, I. C., M. T. Sykes, and W. Cramer. 1993. A simu- P’Yavchenko, N. I. 1960. The biological cycle of nitrogen
lation model for the transient effects of climate change and mineral substances in bog forests. Soviet Soil Sci.
on forest landscapes. Ecol. Model. 65:51-70. 6:593-602.
Prescott, C. E. 1995. Does nitrogen availability control rates of Pyronzinski, K. A., and D. W. Mallock. 1975. The origin of
litter decomposition in forests? Plant Soil 168-169:83-88. land plants: A matter of mycotrophy. Biosystems
, M. A. McDonald, and G. F. Weetman. 1993a. Avail- 6:153-164.
ability of N and P in the forest floors of adjacent stands of Rafes, P. M. 1970. Estimation of the effects of phytophagous
western redcedar—western hemlock and western hem- insects on forest production. pp. 100-106. In D. E.
lock—amabilis fir on northern Vancouver Island. Can. 7. Reichle (ed.), Analysis of Temperate Forest Ecosystems.
For. Res. 23:605-610. Springer-Verlag, New York.
, B. R. Taylor, W. F. J. Parsons, D. M. Durall, and D. . 1971. Pests and the damage they cause to forests.
Parkinson. 1993b. Nutrient release from decomposing pp. 357-367. In P. Duvigneaud (ed.), Productivity of Forest
litter in Rocky Mountain coniferous forests: Influence of Ecosystems. UNESCO, Paris.
nutrient availability. Can. 7. For. Res. 23:1576-1586. Rainey, R. C. 1973. Airborne pests in the atmospheric envi-
, and G. F Weetman (eds.). 1994. Salal Cedar Hemlock ronment. Weather 28:224-239.
Integrated Research Program: A Synthesis. Faculty of Ralston, C. W., and G. E. Hatchell. 1971. Effects of
Forestry, Univ. of British Columbia, Vancouver. 85 pp. prescribed burning on physical properties of soil.
Preston, F. W. 1962. The canonical distribution of common- pp. 68-85. In Prescribed Burning Symposium Proceed-
ness and rarity. Ecology 43:185-215, 410-432. ings. USDA For. Serv., SE For. Exp. Stat., Ashville,
Price, P. W., C. E. Bouton, P. Gross, B. A.McPherson,J. N. SC.
Thompson, and A. E. Weis. 1980. Interactions among Ramanthan, V., B. R. Backstrom, and E. F. Harrison. 1989.
three trophic levels: Influence of plants on interactions Climate and the earth’s radiation budget. Physics Today
between insect herbivores and natural enemies. Amn. Rev. 42:22-34.
Ecol. Syst. 11:41-65. Ramensky, L. G. 1924. [The main regularities of the vegeta-
Price, D. T., D. H. Halliwell, M. J. Apps, W. A. Kurz, and tion cover] (In Russian). Véstnik opy tnogo déla Sredne-
S. R. Curry. 1997. Comprehensive assessment of carbon Chernoz. Obl., Voronezh. pp. 37-73.
stocks and fluxes in a Boreal-Cordilleran forest manage- Ramsey, G. S., and D. G. Higgins. 1982. Canadian Forest
ment unit. Can. 7. For. Res. 27:2005-2016. Fire Statistics 1980. Can. For. Serv., Info. Rept. PI-X-17.
Pritchett, W. L. 1979. Properties and Management of Forest Ranjitsinh, M. K. 1979. Forest destruction in Asia and the
Soils. Wiley, New York. 500 pp. South Pacific. Ambio 8:192-201.
Puritch, G. S. 1973. Effects of water stress on photosynthe- Raper, S. M., K. Steinbeck, I. S. Moss, and D. Whitehead.
sis, respiration and transpiration of four Abies species. 1992. Water use efficiency and transpiration of Robinia,
Can, 7. For. Res. 3:293-298. Liquidambar, and Platanus sprouts in the southeastern
. 1981. Nonvisual Remote Sensing of Trees Affected USA. For. Ecol. Manage. 51:259-268.
by Stress. A Review. Can. For. Serv., Pac. For. Res. Cent. Rapp, M., M. C. LeClerc, and P. Loissant. 1979. The nitro-
For. Tech. Rept. 30. 38 pp. gen economy in a Pinus pinea L. stand. For. Ecol. Manage.
Proctor, J. (ed.). 1989. Mineral Nutrients in Tropical Forest and 22:221-231.
Savanna Ecosystems. Spec. Publ. No. 9, British Ecol. Soc., Raunkiaer, C. 1934. The Life Forms of Plants and Statistical
Blackwell Sci. Publ., Oxford. 473 pp. Plant Geography. Clarendon Press, Oxford. 632 pp.
. 1995. Rainforests and Their Soils. Unpubl. Mss. Read, D. J. 1984. Interactions between ericaceous plants and
Univ. of Stirling, Scotland. their competitions with special reference to soil toxicity.
Proe, M. FE, andJ.Dutch. 1994. Impact of whole-tree har- Asp. Appl. Biol, 5:195-209.
vesting on second rotation growth of Sitka spruce: The . 1991. Mycorrhizal fungi in natural and semi-natural
first 10 years. For: Ecol. Manage. 66:39-54. plant communities. pp. 27-53. In Ecophysiology of Ectomye-
, and P. Millard. 1994a. Relationships between nutri- orrbizae of Forest Trees. The Marcus Wallenberg Founda-
ent supply, nitrogen partitioning and growth in young tion, Symposia Proc. 7, Falun, Sweden.
Sitka spruce (Picea sitchensis). Tree Physiol. 14:75-88. Read, R. A. 1964. Tree Windbreaks for the Central Great
, and P. Millard. 1994b. Effect of P supply upon sea- Plains. USDA For. Serv., Rky. Mt. For. Range Exp. Sta.
sonal growth and internal cycling of P in Sitka spruce Agr. Hdbk. No. 250. 68 pp.
(Picea sitchensis (Bong) Carr.) seedlings. Plant Soil Redmond, D. R. 1959. Mortality of rootlets of balsam fir de-
168-169:3 13-317. foliated by the spruce budworm. For; Sci. 5:64-69.
 REFERENCES CITED R-41

Reed, K. L., and S. G. Clark. 1976. Succession Simulator: A . 1984. Allelopathy. 2nd ed. Academic Press, New
Coniferous Forest Simulator. Coniferous Forest Biome York. 422 pp.
Bull. 11. Univ. of Washington, Seattle. , and S. K. Pancholy. 1972. Inhibition of nitrification
Rehfeldt, G. E., C. C. Ying, D. L. Spittlehouse, and D. A. by climax ecosystems. Amer: 7. Bot. 59:1033-1040.
Hamilton. 1999. Genetic responses to climate in Pinus Rice, R. M., and G. T. Foggin. 1971. Effects of high inten-
contorta. Niche breadth, climate change, and reforesta- sity storms on soil slippage on mountainous watersheds
tion. Ecol. Monogr. 69:375—407. in Southern California. Water Resources Res.
Reich, P. B., T. Koike, S. T. Gower, and A. W. Schoette. 7:1485-1496.
1995. Causes and consequences of variation in conifer Richards, B. N. 1968. Effect of soil fertility on the distribu-
leaf life-span. pp. 225-254. In W. K. Smith and T. M. tion of plant communities as shown by pot culture and
Hinckley (eds.), Ecophysiology of Coniferous Forests. Acade- field trials. Comm. For. Rev. 47:200-210.
mic Press, New York. Richards, P. W. 1952. The Tropical Rain Forest: An Ecological
, M. B. Walters, and D. S. Ellsworth. 1992. Leaf life- Study. Cambridge Univ. Press, London. 450 pp.
span in relation to leaf, plant and stand characteristics . 1969. Speciation in the tropical rain forest and the
among diverse ecosystems. Ecol. Monogr. 62:365-392. concept of the niche.7.Linn. Soc. (Biol.) 1:149-153.
Reichle, D. E. (ed.). 1981. Dynamic Properties of Forest Ecosys- . 1973. The tropical rain forest. Sci. Amer. 229:58-67.
tems. International Biological Programme 23. Cambridge . 1977. Tropical forest and woodlands: An overview.
Univ. Press, Cambridge. 683 pp. Agro-ecosystems 3:225-238.
, R. V. O'Neill, and W. F. Harris. 1975. Principles of Ricklefs, R. E. 1973. Ecology. Chiron. Press, Portland, Ore.
energy and materials exchange in ecosystems. pp. 27-43. In 861 pp.
W. H. van Dobben and R. H. Lowe-Connell (eds.), Roberts, S. D., andJ.N. Long. 1992. Production efficiency
Unifying Concepts in Ecology. W. Junk b.v. Publishers, The ofAbies lasiocarpa: Influence of vertical distribution of leaf
Hague. area. Can. 7. For. Res. 22:1230-1234.
Reifsnyder, W. E., and H. W. Lull. 1965. Radiant Energy in Robertson, G. P., andJ.M. Tiedje. 1984. Denitrification and
Relation to Forests. USDA For. Serv. ‘Tech. Bull. No. nitrous oxide production in successional and old-growth
3445141 pp. Michigan forests. Soi Sci. Soc. Am. F. 48:383-389.
Remezoy, N. P., and P. S. Pogrebnyak. 1969. Forest Soil Sci- Robertson, R. A., and G. E. Davis. 1965. Quantities of
ence. Israel Program of Scientific Translations, Jerusalem. plant nutrients in heather ecosystems. 7. Appl. Ecol.
261 pp. 2:211-219.
Remmert, H., and K. Wunderling. 1970. Temperature dif- Robinson, R. K. 1972. The production by roots of Ca/luna
ferences between arctic and alpine meadows and their of a factor inhibitory to the growth of some mycorrhizal
ecological significance. Oecologia 4:208-210. fungi.7.Ecol. 60:219-224.
Rennie, P. J. 1955. Uptake of nutrients by mature forest Robinson, D. C. E. 1996. GIZELA: Model description,
growth. Plant Soil 7:49-95. user’s guide and tutorial. ESSA Technologies Ltd., Van-
. 1957. The uptake of nutrients by timber forest and couver, BC.
its importance to timber production in Britain. Q. 7. For. Rodhe, W. 1955. Can plankton production proceed during
§1:101-115. winter darkness in subarctic lakes? Proc. Int. Assoc. Theor:
Repetto, R., and M. Gillis. 1988. Public Policies and Misuse of Appl. Limnol. 12:117-122.
Forest Resources. Cambridge Univ. Press, New York. 432 pp. Rodin, L. E., and N. I. Bazilevich. 1967. Production and Min-
Revelle, R. 1974. Food and population. Sc. Amer. eral Cycling in Terrestrial Vegetation. Oliver & Boyd, Lon-
231:161-170. don. 288 pp.
Reynolds, E. R. C., and F. B. Thompson (eds.). 1988. Forests, Roe, A. L. 1967. Seed Dispersal in a Bumper Spruce Seed
Climate and Hydrology. Regional Impacts. United Nations Year. USDA For. Serv. Res. Pap. INT-39, 10 pp.
University, Tokyo. 217 pp. Rollo, D. 1978. The behavioral ecology of terrestrial slugs.
Rice, E. L. 1964. Inhibition of nitrogen-fixing and nitrifying Doctoral thesis, Univ. of B.C., Vancouver. 444 pp.
bacteria by seed plants. Ecology 45:824-837. Romell, L. G. 1935. Ecological Problems of the Humus Layer of
. 1965. Inhibition of nitrogen-fixing and nitrifying the Forest. Cornell Univ., Ithaca, N.Y.
bacteria by seed plants. II. Characterization and identifi- Rosen, K., and T. Lindberg. 1980. Biological nitrogen fixa-
cation of inhibitors. Physiol. Plant. 18:255-268. tion in coniferous forest watershed areas in central Swe-
1967. Chemical warfare between plants. Bros. den. Holarct. Ecol. 3:137-140.
38:67-74. Ross, H. H. 1957. Principles of natural coexistence indicated
1968. Inhibition of nodulation of innoculated by leafhopper populations. Evolution 11:113-129.
legumes by pioneer plant species from abandoned fields. Roussel, L. 1948. Convert et photométrie. Bull. Soc. For
Bull. Torrey Bot. Club 95:346-358. Franche-Comte 125:313-326.
R-42 REFERENCES CITED

Rowe, J. S. 1956. Uses of undergrowth plant species in pine beetle, Dendroctonus — ponderosae —_Hopk.
forestry. Ecology 37:461-473. (Coleoptera:Scolytidae). Canadian Forest Service Info. Re-
. 1961a. The level-of-integration concept and ecolo- port BC-X386. 35 pp.
gy. Ecology 42:420-427. Salisbury, E. J. 1970. The pioneer vegetation of exposed
. 1961b. Critique of some vegetational concepts as muds and its biological features. Phil. Trans. R. Soe. B.
applied to forests of northwestern Alberta. Can. 7. Bot. Biol. Sei. 259:207-255.
39:1007-1017. Salisbury, EK B., and C. Ross. 1969. Plant Physiology.
. 1972. Forest Regions of Canada. Dept. of Envi- Wadsworth, Belmont, Calif. 747 pp.
ronment, Can. For. Serv., Publ. No. 1300. 172 pp. + Sampson, R. N., and D. L. Adams (eds.). 1994, Assessing For-
maps. est Ecosystem Health in the Inland West. Haworth Press,
Runkle, J. R. 1981. Gap regeneration in some old-growth New York. 461 pp.
forests of the eastern United States. Ecology Sanderson, J., and L. D. Harris (eds.) 2000. Landscape Ecol-
62:1041-1051. ogy. Lewis Publishers.
. 1982. Patterns of disturbance in some old-growth Sands, R., P. E. Kriedemann, and P. P. Cotterill. 1984. Water
mesic forests of eastern North America: Ecology relations and photosynthesis in three families of radiata
63:1533-1546. pine known to differ in their response to weed control.
. 1992. Guidelines and Sample Protocol for Sampling For. Ecol. Manage. 9:173-184.
Forest Gaps. USDA For. Serv. Gen. Tech. Rept. PNW- , and D. R. Mulligan. 1990, Water and nutrient dy- e
GTR-283. Portland, Ore. +4 pp. namics and tree growth. For: Ecol. Manage. 30:91-111,.
Running, S. W., and S. T. Gower. 1991. FOREST-BGC, a Santantonio, D., and J. C. Grace. 1987. Estimating fine-root
general model of forest ecosystem processes for regional production and turnover from biomass and decomposi-
applications: II. Dynamic carbon allocation and nitrogen tion data: A compartment-flow model. Can. 7. For Res.
budgets. Tree Physiol. 9:147-160. 17:900-908.
Russell, C. E. 1983. Nutrient cycling and productivity of na- Satali, H., and V. Huhta. 1991. Soil fauna increase Betula
tive and plantation forests at Jari Forestal, Para, Brazil. pendula growth: Laboratory experiments with coniferous
Doctoral thesis, Univ. of Georgia, Athens. 133 pp. forest floor. Ecology 72:665-67 1.
Russell, W. M. S. 1969. Man, Nature and History: Controlling Satchell, J. E. 1967. Lumbricidae. pp. 259-322. In A.
the Environment. Natural History Press, Garden City, Burgess and F. Raw (eds.), Sof Biology. Academic Press,
NY. 252 pp. London.
Ruth, R. H. 1967. Silvicultural effects of skyline crane and Satoo, T. 1956. Drought resistance of some conifers at the
high-lead yarding. 7. For: 65:251-255. first summer after their emergence. Bull. Tokyo Univ. For:
Rutter, A. J. 1963. Studies in the water relations of Pinus 51:1-108.
sylvestris in plantation conditions. I. Measurement of , and H. A. I. Madgwick. 1982. Forest Biomass. W.
rainfall and interception.7. Ecol. 51:191-203. Junk, The Hague. 152 pp.
. 1968. Water consumption by forests. pp. 23-84. In Satterlund, D. R. 1972. Wildland Watershed Management.
T. T. Kozlowski (ed.), Water Deficits and Plant Growth. Wiley, New York. 370 pp.
Vol. I. Academic Press, New York. Sauer, C. O. 1969. Agricultural Origins and Dispersals. 2nd
Ryan, D. F, and F H. Bormann. 1982. Nutrient resorption ed. MIT Press. Cambridge, Mass. 110 pp.
in a northern hardwood forest. BioScience 32:29-32. Savage, S. M. 1974. Mechanism of fire-induced water repel-
Ryan, P. J. 1991. Productivity in Perspective. 3rd Australian lency in soil. Soil Sci. Soc. Amer. Proc. 38:652-657.
Forest Soils and Nutrition Conference. Forestry Commis- Sawyer, C. N. 1966. Basic concepts of eutrophication. 7.
sion of NSW, Sydney. 252 pp. Water Pollut. Con. Fed. 38:737-744.
Ryan, M. G., D. Binkley, and J. H. Fownes. 1997. Age- Schaetzl, R. J., S. F Burns, D. L. Johnson, and T. W. Small.
related decline in forest productivity: Pattern and 1989. ‘Tree uprooting: Review of impacts on forest ecol-
process. Adv. Ecol. Res. 27:213-262. ogy. legetatio 79:165-176.
Sabins, F. F. 1978. Remote Sensing Principles and Interpreta- Scheffer, V. B. 1951. The rise and fall of a reindeer herd. Sei.
tion. W. H. Freeman and Co., San Francisco. 426 pp. Monthly 73:356-362.
Sachs, D., and J. A. Trofymow. 1991. Testing the Perfor- Schlesinger, W. H. 1977. Carbon balance in terrestrial detri-
mance of FORCYTE-11 Against Results From Shawni- tus. dnn. Rev. Ecol. Syst. 8:51-81.
gan Lake Thinning and Fertilization Trials on , and W. A. Reiners. 1974. Deposition of water and
Douglas-Fir. Forestry Canada, Pacific Forestry Centre, cations on artificial foliar collectors in fir krummholz of
Victoria. Info. Rept. BC-X-324. 58 pp. New England mountains. Ecology 55:378-386.
Safranyik, L., H. Barclay, A. Thomson, and W. G. Riel. Schneider, A. (ed.). 1989, Deforestation and “Development” in
1999. A population dynamics model for the mountain Canada and the Tropics: The Impact on People and the Envi-
 REFERENCES CITED R-43

ronment. Centre for International Studies, Univ. Coll. of pine forest stand: a gap model approach. Clim. Change
Cape Breton, Sydney, NS. 251 pp. 34:263-268.
Schneider, S. H. 1974. The population explosion: can it Shao, G., S. Zhao, and G. Zhao. 1991. Application of GIS in
shake the climate. Ambio 3(3-4):150-155. dynamic simulation of forested landscapes. Chinese 7.
. 1975. On the carbon dioxide-climate confusion. 7. Appl. Ecol. 2:103-107.
Atmos. Sci. 32:2060-2066. Sharma, N. P. (ed.). 1992. Managing the World’s Forests. Look-
,and R. D. Dennett. 1975. Climatic barriers to long- ing for Balance Between Conservation and Development.
term energy growth. Ambio 4:66-74. Kendall/Hunt Publ., Dubuque, IA. 605 pp.
Schonland, B. F. J. 1950. The Flight of the Thunderbolt. Ox- , R. Rowe, K. Openshaw, and M. Jacobsen. 1992.
ford Univ. Press, Oxford. 152 pp. World forests in perspective. pp. 17-31. In N. P. Sharma
Schowalter, T. D., T. E. Sabin, S. G. Stafford, andJ.M. Sex- (ed.), Managing the Worla’s Forests. Looking for Balance Be-
ton. 1991. Phytophage effects on primary production, tween Conservation and Development. Kendall/Hunt Publ.,
nutrient turnover, and litter decomposition of young Dubuque, IA. 605 pp.
Douglas-fir in western Oregon. For. Ecol. Manage. Shea, S. R. 1979. An ecological approach to the control of
42:229-243. jarrah dieback. For. Focus 21:7-19.
Schroter, C. 1926. Das Pflanzenleben der Alpen. Raustein, Shelford, V. E. 1951a. Fluctuations of non-forest animal
Zurich. 1288 pp. populations in the upper Mississippi Basin. Ecol. Monogr.
Schultz, A. M. 1964. The nutrient recovery hypothesis for 21:149-181.
arctic microtine cycles. I. Ecosystem variables in rela- . 1951b. Fluctuations of forest animal populations in
tion to arctic microtine cycles. pp. 57-68. In A. J. Crisp east central Illinois. Ecol. Monogr. 21:183-214.
(ed.), Grazing in Terrestrial and Marine Environments. Sheppard, S. R. J., and H. W. Harshaw (eds.) 2000. Forests
Blackwell Sci. Publ., Oxford. and Landscapes. Linking Ecology, Sustainability and Aesthet-
. 1969. A study of an ecosystem: the arctic tundra. pp. ics. CABI Publishing, Wallingford, UK
77-93. In G. van Dyne (ed.), The Ecosystem Concept in Sheriff, D. W., H. A. Margolis, M. R. Kaufmann, and P. O.
Natural Resource Management. Academic Press, New B. Reich. 1995. Resource use efficiency. pp. 143-178. In
York. W. K. Smith and T. M. Hinkley (eds.), Resource Physiology
Schurhoff, P. N. 1924. Die Plastiden. Jn K. Linshaner (ed.), of Conifers, Acquisition. Allocation, and Utilization. Acade-
Handbuch der Pflanzen-anatomie. Vol. I. p. 10. mic Press, New York.
Schweers, W., and F. H. Meyer. 1970. Einfluss der Mykor- , E. K. S. Nambiar, and D. N. Fife. 1986. Relation-
rhiza auf den Transport von Assimilaten in die Wurzel. ship between nutrient status, carbon assimilation and
Der dtsch. Bot. Gesell. 83:109-119. water use efficiency in Pinus radiata (D.Don) needles.
Scott, D. R. M. 1955. Amount and Chemical Composition Tree Physiol. 2:73-88.
of the Organic Matter Contributed by Overstorey and Shibata, N., T. Ibaragi, and M. Ishii. 1951. [Studies on the
Understorey Vegetation to Forest Soil. Yale Univ. School influence of variation of forest conditions on the soil. II.
For. Bull. 62. 73 pp. Effects of clear felling on soil in Hinoki forest]. pp.
Seely, B.,J.P. Kimmins, C. Welham, and K. Scoullar. 1999. 133-135. Trans. 59th Meeting Jap. For. Soc.
Defining stand-level sustainability, exploring stand-level Shimwell, D. W. 1971. The Description and Classification of
stewardship.7. For. 97:4-11. Vegetation. Univ. Washington Press, Seattle. 322 pp.
Seely, B., C. Welham, andJ. P. Kimmins. 2002. Carbon se- Shugart, H. H. 1998. Terrestrial Ecosystems in Changing En-
questration in a boreal forest ecosystem: Results from the vironments. Cambridge University Press, Cambridge,
ecosystem simulation model, FORECAST. For. Ecol. UK.
Manage. 169:123-135. Shugart, H. H., M. S. Hopkins, I. P. Burgess, and A. T.
Segaard, B. 1969. Resistance studies in Thuja. Forstl. Forsol- Mortlock. 1980. The development of a succession model
gsv. Damm. 31:279-398. for subtropical rain forest and its application to assess the
Seidel, S., and D. Keyes. 1983. Can We Delay a Greenhouse effects of timber harvest at Wiangaree State Forest, New
Warming? USEPA, Washington, DC. 215 pp. South Wales. 7. Environ. Manage. 11:243-265.
Selye, H. 1946. The general adaptation syndrome and the Shugart, H. H., and I. R. Noble. 1981. A computer model of
disease of adaptation. 7. Clin. Endocrinol. 6:117-230. succession and fire response to the high altitude Eucalyp-
Shao, G. FE, P. Schall,J.F. Weishampel, and G. Shao. 1994. tus forest of the Brindabella Range, Australian Capital
Dynamic simulations of mixed broadleaved-Pinus ko- Territory. Aust. 7. Ecol. 6:149-164.
raiensis forests in the Changbaishan Biosphere Reserve Shugart, H. H., and D. C. West. 1977. Development of an
of China. For: Ecol. Manage. 70:1-3. Appalachian deciduous forest succession model and its
Shao, G., G. F. Shao, and T. M. Smith. 1996. Potential im- application to assessment of the impact of the chestnut
pacts of climate change on a mixed broadleaved-Korean blight. 7. Environ. Manage. 5:161-170.
R-44 REFERENCES CITED

Shugart, H. H. 1984. A Theory ofForest Dynamics. The Ecolog- , J. M. Goshine, G. Holdsworth, C. J. Krebs, S.
ical Implications of Forest Succession Models. Springer- Boutin,J.N. M. Smith, R. Boonstra, and M. R. T. Dale.
Verlag, New York. 278 pp. 1993. Can the solar cycle and climate synchronize the e
S
S
S

, M. S. Hopkins, I. P. Burgess, and A. T. Mortlock. snowshoe hare cycle in Canada? Evidence from tree rings
1980. The development of a succession model for sub- and ice cores. Amer: Nat. 141:173-198. —

tropical rain forest and its application to assess the effects Siren, G. 1995. The development of spruce forest on raw
of timber harvest at Wiangaree State Forest, New South humus sites in northern Finland. Acta Forest. Fenn. 62.
Wales.7.Environ. Manage. 11:243-265. Sirois, L. 1992. The transition between boreal forest and
, P. J. Michaels, T. M. Smith, D. A. Weinstein, E. B. tundra. pp. 196-215. In H. H. Shugart, R. Lemans, and
Rastetter, R. G. Woodmansee, and P. G. Risser. 1988. G. B. Bonan (eds.), A Systems Analysis of the Global Boreal
Simulation models of forest succession. pp. 125-151. In Forest. Cambridge Univ. Press, Cambridge, UK.
T. Rosswall, R. G. Woodmansee, and P. G. Risser (eds.), , G. B. Bonan, and H. H. Shugart. 1994. Develop-
Scales and Global Change: Spatial and Temporal Variability ment of a simulation model of the forest-tundra transi-
in Biospheric and Geospheric Processes. SCOPE (Scientific tion zone of northeastern Canada. Can. 7. For. Res.
Committee on Problems ofthe Environment) 35. John Wiley, 24:697-706.
New York. Sirois, L., G. B. Bonan, and H. H. Shugart. 1994. Develop-
, and I. R. Noble. 1981. A computer model of succes- ment of a simulation model of the forest-tundra transi-
sion and fire response to the high altitude Eucalyptus for- tion zone of northeastern Canada. Can. 7. For. Res.
est of the Brindabella Range, Australian Capital 24:697-706.
Territory. Aust. 7. Ecol. 6:149-164. Skole, D., and C. Tucker. 1993. Tropical deforestation and
, and D. C. West. 1977. Development of an Ap- habitat fragmentation in the Amazon: Satellite data from
palachian deciduous forest succession model and its ap- 1978 to 1988. Science 260:1905-1910.
plication to assessment of the impact of the chestnut Skutch, A. F. 1967. Adaptive limitation of the reproductive
blight.7.Environ. Manage. 5:161-170. rate of birds. [bis 109:579-599,
, and D. C. West. 1980. Forest succession models. Slankis, V. 1958. The role of auxin and other exudates in myc-
BioScience 30:308-313. orrhizal symbionts of forest trees. pp. 427-443. In K. V.
, D. C. West, and W. R. Emanuel. 1981. Patterns and ‘Thimann (ed.), Physiology ofForest Trees. Ronald, New York.
dynamics of forests: An application of simulation models. . 1971. Formation of ectomycorrhizae of forest trees
pp. 74-94. In D. C. West , H. Shugart, and D. B. Botlain in relation to light, carbohydrates and auxins. pp.
(eds.), Forest Succession: Concepts and Application. Springer- 151-167. In Mycorrhizae. Proc. First N. Amer. Conf. on
Verlag, New York. Mycorrhizae. Misc. Publ. 1189. USDA For. Serv.
Siivonen, L., and J. Koskimmes. 1955. Population Fluctua- Slatyer, R. O. 1967. Plant-Water Relationships. Academic
tions and the Lunar Cycle. Papers on Game Research. Press, London. 366 pp.
Finnish Game Foundation, No. 14. 22 pp. . (ed.). 1977. Dynamic Changes in Terrestrial Ecosys-
Silverman, M. P., and E. F Munoz. 1970. Fungal attack on tems: Patterns of Change and Techniques for Study and
rock: solubilization and altered infrared spectra. Science Applications to Management. MAB Technical Notes No.
169:985—987. 4. 30 pp. UNESCO.
Simard, S. W., M. D. Jones, and D. M. Durall. 2002. Carbon Slaughter, C. W., R. J. Barney, and G. M. Hansen (eds.).
and nutrient fluxes within and between mycorrhizal 1971. Fire in the Northern Environment—A Symposium.
plants. pp. 33-74. In M. G. A. van Heijden and I. Saun- USDA For. Serv. PNW For. Range Exp. Sta., Portland,
ders (eds.), Mycorrhizal Ecology. Ecological Studies 157. Ore. 275 pp
Springer-Verlag, Berlin. Slobodkin, L. B. 1961. Growth and Regulation ofAnimal Pop-
Simberlof, D. 1998. Measuring diversity of communities ulations. Holt, New York.
and ecosystems with special reference to forests. Pp. Small, E. 1972. Photosynthetic rates in relation to nitrogen
66-81. In F. L. Bunnell and J. FE Johnson (eds.), recycling as an adaptation to nutrient deficiency in peat
Measuring Biodiversity for Forest Policy and Management. bog plants. Can. 7. Bot. 50:2227-2233.
Univ. of British Columbia Press, Vancouver. Smith, C. T., P. N. Beets, W. J. Dyck, and L. A. Morris
Simons, E. L. 1972. Primate Evolution: An Introduction to (eds.). 1994. Understanding plant nutrient uptake and
Man’; Place in Nature. Macmillan, New York. 322 pp. supply—Opportunities for managing site productivity.
Simpson, G. G. 1964. Species density of North American Proc. IEA/BA Workshop, Omapere, Northland, NZ. N.
recent mammals. Syst. Zool. 13:57-73. Z. F. For. Service 24:129-428.
Sinclair, A. R. E. 1995. Mammal Populations: Fluctuation, Smith, D. W., and G. C. Bowes. 1974. Loss of some ele-
Regulation and Conservation. Unpubl. mss., Dept. of Zool- ments in flyash during old-field burns in southern On-
ogy, University of British Columbia, Vancouver. tario. Can.F. Soil Sci. 54:215-224.
 REFERENCES CITED R-45

Smith, FE. 1970. Analysis of ecosystems. pp. 2-18. In D. E. Soil Survey Staff. 1975. Soil Taxonomy. A Basic System of Soil
Reichle (ed.), Analysis of Temperate Forest Ecosystems. Classification for Making and Interpreting Soil Surveys.
Springer-Verlag, Berlin. USDA Agric. Handbook 436. Washington, DC. 754 pp.
Smith, H. S. 1935. The role of biotic factors in the determi- Sollins, P., R. A. Goldstein,J.B. Mankin, C. E. Murphy, and
nation of population densities. 7 Econ. Entomol. G. L. Swartzman. 1981. Analysis of forest growth and
28:873-898. water balance using complex ecosystem models. pp.
Smith, N. G. 1968. The advantage of being parasitized. 537-565. In D. E. Reichle (ed.), Dynamic Properties ofFor-
Nature 219:690-694. est Ecosystems. International Biological Programme 23.
Smith, R. B., and E. F. Wass. 1976. Soil Disturbance, Vege- Cambridge Univ. Press, Cambridge.
tative Cover and Regeneration on Clearcuts in the Nel- Solomon, A. M. 1986. Transient response of forests to CO3-
son Forest District, British Columbia. Pacific For. Res. induced climate change: Simulation modeling experi-
Centre, Victoria, Canada. Report No. BC-X-151. 37 pp. ments in eastern North America. Oecologia 68:567-579.
Smith, R. L. 1966. Ecology and Field Biology. Harper & Row, ,M. L. Tharp, D. C. West, G. E. Taylor,J.M. Webb,
New York. 686 pp. andJ.M. Trimble. 1984. Responses of Unmanaged Forests to
Smith, T. M., and M. Huston. 1989. A theory of the spatial CO,-Induced Climatic Change: Available Data, Initial Tests,
and temporal dynamics of plant communities. Vegetatio and Data Requirements. DOE/NDD-0053-TRO09. U.S.
83:49-69. Department of Energy, Washington, DC.
, and D. L. Urban. 1988. Scale and resolution of for- , D. C. West, and A. J. Solomon. 1981. Simulating
est structural pattern. Vegetatio 74:143-150. the role of climate change and species immigration in
Smith, W. H. 1985. Forest quality and air quality. 7. For forest succession. pp. 36-55. In D. C. West, H. H.
83:82-92. Shugart, and D. B. Botkin (eds.), Forest Succession.
, D. M. Post, and F. W. Adrian. 1979. Waste manage- Springer-Verlag, New York.
ment to maintain or enhance productivity. pp. 304-320. Solomon, A. M. 1986. Transient response of forests to CO-
In Impact ofIntensive Harvesting on Forest Nutrient Cycling. induced climate change: Simulation modelling experi-
Proc. Symp., Coll. Environ. Sci. and For., SUNY, Syra- ments in eastern North America. Oecologia 68:567-579.
cuse. Solomon, A. M., M. L. Tharp, D. C. West, G. E. Taylor,J.M.
Smith, W. K., and T. M. Hinckley (eds.). 1995a. Resource Webb, andJ.M. Trimble. 1984. Responses of unmanaged
Physiology of Conifers. Acquisition, Allocation and Utilization. forests to CQO,-induced climatic change: Available data,
Academic Press, New York. 396 pp. initial tests, and data requirements DOE/NDD-0053-
, and T. M. Hinckley (eds.). 1995b. Ecophysiology of TROO9. U.S. Department of Energy, Washington, DC.
Coniferous Forests. Academic Press, New York. 338 pp. Solomon, A. M., D. C. West, and A. J. Solomon. 1981. Sim-
Smith, T. M., and D. L. Urban. 1988. Scale and resolution ulating the role of climate change and species immigra-
of forest structural pattern. Vegetatio 74:143-150. tion in forest succession. pp. 154-177. In D. C. West,
Smithers, L. A. 1957. Thinning in Lodgepole Pine Stands in H. H. Shugart, and D. B. Botkin (eds.), Forest Succession.
Alberta. Can. Dept. N. Affairs and Nat. Resources, For- Springer-Verlag, New York.
est Br., For. Res. Div. Tech. Note 52. 26 pp. Sopper, W. E. 1975. Wastewater recycling on forest lands. pp.
Snell, G. L., S. C. Sterns-Smith, and E. W. Mogren. 1979. 227-243. In B. Bernier and C. H. Winget (eds.), Forest Soils
Macronutrient Transport in Relation to Harvesting in and Forest Land Management. Laval Univ. Press, Quebec.
Colorado Spruce-Fir Forests. Paper distrib. at Impact of , and S. N. Kerr (eds.). 1979. Utilization of Municipal
Intensive Harvesting on Forest Nutrient Cycling. Sewage Effluent and Sludge on Forest and Disturbed Land.
SUNY, Syracuse. Pennsylvania State Univ. Press, University Park, Pa. 547 pp.
Society of American Foresters. 1993. Task Force Report on Sousa, W. P. 1984. The role of disturbance in natural com-
Sustaining Long-term Forest Health and Productivity. Soci- munities. Aun. Rev. Ecol. Syst. 15:353-391.
ety of American Foresters, Bethesda, MD. 83 pp. Southwood, T. R. E. 1961. The number of species of insect
Séderlund, R., and B. H. Svensson. 1976. The global nitro- associated with various trees. 7. Anim. Ecol. 30:1-8.
gen cycle. pp. 23-73. In B. H. Svensson and R. Soderlund Spears, J. S. 1979. Can the wet tropical forest survive?
(eds.), Nitrogen, Phosphorus and Sulphur—Global Cy- Comm. For. Rev. 58:165-180.
cles. Scope Report 7. Ecol. Bull. (Stockholm) 22. Specht, R. L., P. Rayson, and M. E.Jackman. 1958. Dark Is-
Séderstrém, B. 1991. The fungal partner in the mycorrhizal land Heath (Ninety-Mile Plain, South Australia). VI.
symbiosis. pp. 5-26. In Ecophysiology of Ectomycorrhizae of Pyric succession: changes in composition, coverage, dry
Forest Trees. The Marcus Wallenberg Foundation, Sym- weight and mineral status. Aust. 7. Bot. 6:59-88.
posia Proc. 7, Falun, Sweden. Spies, T. A., andJ.F. Franklin. 1989. Gap characteristics and
Soil Science Society of America. 1973. Glossary of Soil Science vegetation response in coniferous forests of the Pacific
Terms. Soil Sci. Soc. Amer., Madison, Wis. 34 pp. Northwest. Ecology 70:543-545.
R-46 REFERENCES CITED

, J. FE. Franklin, and M. Klopsch. 1990. Canopy gaps and C. FE Jordan. 1978, Nutrient retention by the
in Douglas-fir forests of the Cascade Mountains. Can. 7. root mat of an Amazonian rain forest. Ecology
For. Res. 20:649-658. 59:434-437.
Spilsbury, R. H., and D. S. Smith. 1947. Forest Site Types of Stark, R. W. 1959. Climate in relation to winter mortality of
the Pacific Northwest. Dept. Lands and Forests, B.C. the lodgepole needle miner, Recurvaria starki Free., in —_
For. Serv., Victoria, B.C. 46 pp. Canadian Rocky Mountain Parks. Can. 7. Zool.
Sprugel, D. G. 1975. Dynamic structure of wave-regener- 37:753-761.
ated Abies balsamea forests in the northeastern United Statistics Canada. 1976. Profile Studies. The Age-Sex Struc-
States.7.Ecol. 64:889-911. ture of the Canadian Population. 1971 Census of Canada.
. 1984. Density, biomasss, productivity, and nutrient Catalogue 99-703 Vol: V—Part:1 (Bull. 5.1-3). Ottawa.
cycling changes during stand development in wave- Steinberg, R. A. 1951. Correlations between protein-carbo-
regenerated balsam fir forests. Ecol. Monogr. 54:165-186. hydrate metabolism and mineral deficiencies in plants.
, M. G. Ryan, J. Brooks, K. A. Vogt, and T. A. Mar- pp. 359-386. In E. Truog (ed.), Mineral Nutrition of
tin. 1995. Respiration from the organ level to the stand. Plants. Univ. of Wisconsin Press, Madison.
pp. 255-299. In W. K. Smith and T. M. Hinkley, (eds.), Steinbrenner, E. C. 1955. The effects of repeated tractor
Resource Physiology of Conifers. Acquisition, Allocation, and trips on the physical properties of forest soils. Northwest
Utilization. Academic Press, New York. Sci. 29:155-159.
Spurr, S. H. 1957. Local climate in the Harvard Forest. . 1976. Factors that influence the productivity of
Ecology 38:37—46. western hemlock. XVI [UFRO World Congress, Oslo.
, and B. V. Barnes. 1973. Forest Ecology. 2nd ed. Proc. Div. 1:167-185.
Ronald, New York. 571 pp. Steinhart,J.S., and C. E. Steinhart. 1974. Energy use in the
Squire, R. O. 1983. Review of second rotation silviculture of U.S. food system. Science 184:307-316.
P. radiata plantations in southern Australia: Establish- Steinlin, H. J. 1982. Monitoring the world’s tropical forests.
ment, practice and expectations. pp. 130-137. In R. Bal- Unasylva 34:2-8.
land and S. P. Gessel (eds.), J[UFRO Symposium on Forest Stenberg, P., E. H. DeLucia, A. W. Schoettle, and H.
Site and Continuous Productivity. USDA For. Serv. Gen. Smolander. 1995. Photosynthetic light capture and pro-
Tech. Rept. PNW-163. Portland, Ore. cessing from cell to canopy. pp. 3-38. In W. K. Smith and
Squire, R. O., P. W. Farrell, D. W. Flinn, and B. C. Aeberli. T. M. Hinkley (eds.), Resource Physiology of Conifers. Acqui-
1985. Productivity of first and second rotation stands of sition, Allocation, and Utilization. Academic Press, New
radiata pine on sandy soils. II. Height and volume growth York.
at five years. Aust. For. 48:127-137. Stenlid, G. 1958. Salt losses and redistribution of salts in
Squire, R. O., D. W. Flinn, and P. W. Farrell. 1979. Produc- higher plants. Encyc Plant Physiol. 4:615—637.
tivity of first and second rotation stands of radiata pine Stern, K., and L. Roche. 1974. Genetics ofForest Ecosystems.
on sandy soils. I. Site factors affecting early growth. Aust. Springer-Verlag, New York. 330 pp.
For. 42:226-235. Stewart, O. C. 1956. Fire as the first great force employed
Staaf, H. 1982. Plant nutrient changes in beech leaves dur- by man. pp. 115-133. In W. L. Thomas, Jr. (ed.), Man’
ing senescence as influenced by site characteristics. Oecol. Role in Changing the Face of the Earth. Univ. of Chicago
Plant. 3:161-170. Press, Chicago.
Stachurska-Hagen, T: 1980. Acidification Experiments in Stewart, W. D. P. 1966. Nitrogen Fixation in Plants. Athlone
Conifer Forest. 8. Effects of Acidification and Liming on Press, London. 168 pp.
Some Soil Animals: Protozoa, Rotifera and Nematoda. Stoessl, A. 1970. Antifungal compounds produced by higher
Internal Report 74/80. SNSF-project, Aas—NLH, Nor- plants. Rec. Adv. Phytochem. (3):143-180.
way. Stone, E. 1975. Effects of species on nutrient cycles and soil
Stachurski, A., and J. R. Zimka. 1975. Methods of studying change. Philosoph Trans. R. Soc. London (B) 271:149-162.
forest ecosystems: Leaf area, leaf production and with- Stone, E. C..1957. Dew as an ecological factor. Ecology
drawal of nutrients from leaves of trees. Ekol. Pol. 38:407-422.
23:637-648. . 1958. Dew absorption by conifers. pp. 125-153. In
Stage, A. R. 1973. Prognosis Model for Stand Development. K. V. Thimann (ed.), The Physiology of Forest Trees.
USDA For. Serv., Intermountain For. Range Exp. Stn., Ronald, New York.
Ogden, Utah. Res. Pap. INT-137. , and H. A. Fowells. 1955. Survival value of dew
Stage, A. R. 2001. How models are connected to reality— under laboratory conditions with Pinus ponderosa. For: Sci.
Evaluation criteria for their use in decision support. Can. 1:183-188.
7. For. Res. 400. Strand, L. 1980. Acid Precipitation and Regional Tree Ring
Stark, N. 1972. Nutrient cycling pathways and litter fungi. Analyses. Internal Report 73/80. SNSF-project, Aas—
BioScience 22:355-360. NLH, Norway.
 REFERENCES CITED R-47

Strang, R. M. 1972. Ecology and land use of the barrens of Western United States. USDA For. Serv., Gen. Tech.
western Nova Scotia. Can. 7. For. Res. 2:276-290. Rept. PNW-21. 14 pp.
. 1973. The rate of silt accumulation in the Lower Swartzman, G. L., and S. P. Kaluzny. 1987. Ecological Simu-
Peel River, Northwest Territories. Can. 7. For. Res. lation Primer. Macmillan, New York. 370 pp.
3:457-458. Switzer, G. L., and L. E. Nelson. 1972. Nutrient accumula-
Struwe, S., and A. Kjoller (eds.). 1990. Denitrification in tion and cycling in loblolly pine (Pinus taeda L.) planta-
forest ecosystems. For. Ecol. Manage. Spec. Issue 44:1-92. tion ecosystems: The first 20 years. Soil Sci. Soc. Amer.
Stuart, T. S. 1968. Revival of respiration and photosynthesis Proc. 36:143-147.
in dried leaves of Polypodium polypodioides. Planta , L. E. Nelson, and W. H. Smith. 1968. The mineral
83:185-206. cycle in forest stands. pp. 1-9. In Forest Fertilization. The-
Sturges, F. W., R. T. Holmes, and G. E. Likens. 1974. The ory and Practice.’TVA, Muscle Shoals, Tenn.
role of birds in nutrient cycling in a northern hardwood Syers, J. K., J. A.Adams, and T. W. Walker. 1970. Accumu-
ecosystem. Ecology 55:149-155. lation of organic matter in a chronosequence of soils de-
Stutzbach, S. J., A. L. Leaf, and R. E. Leonard. 1972. Varia- veloped on wind-blown sand in New Zealand. 7. Soil Sci.
tion in forest floor under a red pine plantation. Soil Sci. 21:146-153.
114:24-28. Tamarin, R. H. 1978. Population Regulation. Benchmark Pa-
Suchting, H., and A. Christmann. 1935. [On humus layer pers in Ecology. 7. Dowden, Hutchinson and Ross,
destruction in the forest floor. III. Humus destruction Stroudsburg, Pa. 391 pp.
during clearcutting.| Mitteil. Fortswirt. u. Forstwiss. Tamm, C. O. 1964. Determination of nutrient requirements
6:425-446. of forest stands. Int. Rev. For: Res. 1:115-170.
Sukachey, V., and N. Dylis. 1964. Fundamentals of Forest Bio- 1976. Acid precipitation and forest soils. pp.
geocoenology. Oliver and Boyd, Edinburgh. 672 pp. 681-684. In L. S. Dochinger and T. A. Seliga (eds.), Proc.
Sullivan, T. P. 1978. Conifer seed predation by the deer Ist Int. Symp. Acid Precip. USDA For. Serv., Gen. Tech.
mouse: a problem in reforestation. Doctoral. thesis, Rept. NE-23.
Univ. of B.C., Vancouver. 181 pp. , and E. B. Cowling. 1976. Acidic precipitation and
. 1979. The use of alternative foods to reduce conifer forest vegetation. pp. 848-855. In L. S. Dochinger and
seed predation by the deer mouse (Peromyscus manicula- T. A. Seliga (eds.), Proc. Ist Int. Symp. Acid Precip. USDA
tus).F.Appl. Ecol. 16:475-495. For. Serv., Gen. Tech. Rept. NE-23.
Sutton, R. F. 1969. Form and Development of Conifer Root ‘Tanner, E. V. J. 1980. Litterfall in montane rain forest of Ja-
Systems. Comm. For. Bureau. Comm. Agric. Bureau, maica and its relation to climate. 7. Ecol. 68:833-848.
Oxford Tech. Communication No. 7, 131 pp. ‘Tanner, J. T. 1966. Effects of population density on growth
Sutton, W. R. J. 1993a. The world’s need for wood. Paper rates of animal populations. Ecology 47:733-745.
presented at “The Globalization of Wood: Supply ‘Tansley, A. G. 1920. The classification of vegetation and the
Processes, Products and Markets.” Forest Products Soci- concept of development. 7. Eco/. 8:118-149.
ety and Oregon Forest Resources Institute, Portland, . 1935. The use and abuse of vegetational concepts
Ore: and terms. Ecology 16:284-307.
. 1993b. For Environmental Reasons, Should We or . 1939. The British Islands and Their Vegetation. Cam-
Should We Not Harvest British Columbia (B.C.) bridge Univ. Press, Cambridge. 930 pp.
Forests? Unpublished mss. Forestry Canada, Victoria, . 1941. Note on the status of salt-marsh vegetation
BC, and the concept of “formation.” 7. Ecol. 29:212-214.
Swaine, M. D., and T. C. Whitmore. 1988. On the defini- Tarrant, R. F, and J. M. Trappe. 1971. The role of A/nus in
tion of ecological species groups in tropical rain forests. improving the forest environment. Plant Soil. Spec.
Vegetatio 75:81—86. vol.:335-348.
Swank, W. T., andJ.W. Elwood. 1971. The seasonal and an- Taylor, B. R., D. Parkinson, and W.J. F. Parsons. 1989. Ni-
nual flux of cations for forested ecosystems in the Ap- trogen and lignin content as predictors of litter decay
palachian Highlands. Paper. 2nd Nat. Biol. Congr., rates: A microcosm test. Ecology 70:97-104.
Miami Beach, Fla. October 23-26. 34 pp. , C. E. Prescott, W. J. F. Parsons, and D. Parkinson.
Swanson, F.J., and C. T. Dyrness. 1975. Impact of clearcut- 1991. Substrate control of litter decomposition in four
ting and road construction on soil erosion by landslides Rocky Mountain coniferous forests. Can. jf. Bot.
in the western Cascade Range, Oregon. Geology 69:2242-2250.
3:393-396. Teskey, R. O., R. W. Sheriff, D. Y. Hollinger, and R. B.
Swanston, D. N. 1969. Mass Wasting in Coastal Alaska. Thomas. 1995. External and internal factors regulating
USDA For. Serv., Res. Paper PNW-83, 15 pp. photosynthesis. pp. 105-140. In W. K. Smith, and T. M.
. 1974. Slope Stability Problems Associated With Hinkley, (eds.), Resource Physiology of Conifers. Acquisition,
Timber Harvesting in Mountainous Regions of the Allocation, and Utilization. Academic Press, New York.
R-48 REFERENCES CITED

Tharp, M. L. 1978. Modeling Major Perturbations on a Timmis, R., and J. Worrall. 1974. Translocation of dehard-
Forest Ecosystem, Master’s thesis, Univ. of Tennessee, ening and bud-break promoters in climatically “split”
Knoxville. Douglas-fir. Can. 7. For. Res. 4:229-237. C—
e
t—~
L

Tharp, M. L. 1978. Modeling Major Perturbations on a Forest Tinbergen, L. 1960. The natural control of insects in
Ecosystem. M. Sc. University of Tennessee, Knoxville. pinewoods. 1. Factors influencing the intensity of preda- ~

Theodorou, C., and G. D. Bowen. 1970. Mycorrhizal re- tion by songbirds. Arch. Neerl. Zool. 13:265-343.
sponses of radiata pine in experiments with different Ting, I. P., and W. M. Duggar, Jr. 1968. Non-autotrophic
fungi. Aust. For. 34:183-191. carbon dioxide metabolism in cacti. Bot. Gazette
, and G. D. Bower. 1990. Effects of fertilizer on lit- 129:9-15.
terfall and N and P release from decomposing litter in a Tisdale, S. L., and W. L. Nelson. 1966. Soil Fertility and Fer-
Pinus radiata plantation. For. Ecol. Manage. 32:87-102. tilizers. Macmillan, New York. 694 pp.
Thie, J., and G. Ironside (eds.). 1976. Ecological (Biophysical) Titus, B. D., S. S. Sidhu, and A. U. Mallik. 1995. A Sum-
Land Classification in Canada. Ecol. Land. Classif. Ser. mary of Some Past Studies on Kalmia angustifolium L.: A
No. 1. Lands Directorate, Environ. Canada. Problem Species in Newfoundland Forestry. Can. For.
Thielges, B. A. 1968. Altered polyphenol metabolism in the Serv., Newfoundland and Labrador Region, St. John’s
foliage of Pinus sylvestris associated with European pine Info. Rep. N-X-296. 68 pp.
sawfly attack. Can. 7. Bot. 46:724-725. Tollan, A. (ed.). 1978. Annotated bibliography 1974-1977.
Thirgood,J.V. 1981. Man and the Mediterranean Forest. Aca- SNSF project. Acid Precipitation—Effects on Forest and
demic Press, New York. 180 pp. Fish. Aaa—NLH, Norway. 39 pp.
Thomas, B. R., and E. R. Burroughs,
Jr. 1975. Significance of Tomlinson, G. H. 1981. Tree Die-Back in Quebec and the N.E.
soil classification in forest land management. pp. 599-616. United States—A Hypothesis Suggesting the Cause and a
In B. Bernier and C. H. Winget (eds.), Forest Soils and For- Proposed Experimental Investigation. Res. Memo., Rept.
est Land Management. Laval Univ. Press, Quebec. #3. Proj. #74-7124-13. Res. Centre, Senneville, Quebec.
Thomas, W. A. 1969. Accumulation and cycling of calcium Tonu, O. 1983. Metsa suktessiooni ja tasandilise strktuuri
by dogwood trees. Ecol. Monogr. 39:101-120. imiteerimisest. Yearbook of the Estonian Naturalist Society
Thompson,J.N. 1982. Interaction and Coevolution. Wiley In- 69:110-117.
terscience, New York. 180 pp. ‘Torrenueva, A. L. 1975. Variation in Mineral Flux to the
Thornthwaite, C. W. 1948. An approach towards a rational Forest Floors of a Pine and a Hardwood Stand in the
classification of climate. Geogr: Rev. 38:55-94. Georgia Piedmont. Doctoral. thesis, Univ. Georgia,
, and F. K. Hare. 1955. Climatic classification in Athens. 110 pp.
forestry. Mnasylva 9:50-59. ‘Toumey, J. W., and R. Kienholz. 1931. Trenched Plots
Tiedje, J. M. 1978. Denitrification in soil. pp. 362-366. In Under Forest Canopies. Yale Univ. School For. Bull. No.
Schlessinger (ed.), Microbiology. Am. Soc. Microbiol., 30. 31 pp.
Washington, DC. Tranquillini, W. 1954. Uber den Einfluss von Ubertempera-
Tilman, D. 1982. Resource Competition and Community Struc- turen der Blatter bei Dauereinschluss in Kiivetten auf die
ture. Princeton Univ. Press, Princeton, NJ. 296 pp. dkologische CO,-Assimilationsmessung. Ber. Deut. Bot.
. 1985. The resource ratio hypothesis of plant succes- Ges. 67:191-204.
sion. Am. Naturalist 125:827-852. . 1959. Die Stoffproduktion der Zirbe (Pinus cembra
. 1988. Plant Strategies and the Dynamics and Structure L.) an der Waldgrenze wahrend eines Jahres: I Standert-
of Plant Communities. Princeton Univ. Press, Princeton, sklima und CO,-Assimilation. II. Zuwachs und CO,-Bi-
NJ. 360 pp. lanz. Planta 54:107-151.
, D. 1990a. Mechanisms of plant competition for nu- , and W. Schutz. 1971. Uber den Einfluss von
trients: The elements ofa predictive theory of plant com- Warme auf das Photosynthesevermogen der Zirbe (Pinus
petition. pp. 117-141. In J. B. Grace and D. Tilman cembra L..) und der Alpen rose (Rhododendron ferrugineum
(eds.), Perspectives on Plant Competition. Academic Press, L.) in winter. Rep. Kevo. Subarctic Res. Stat. 8:158-166.
San Diego, CA. ‘Transeau, E. N. 1926. The accumulation of energy by
. 1990b. Constraints and trade-offs: Towards a pre- plants. Ohio7.Sci. 26:1-10.
dictive theory of competition and succession. Oikos Treshow, M. 1970. Environment and Plant Response. Mc-
58:3-15. Graw-Hill, New York. 422 pp.
, and J. A. Downing. 1994. Biodiversity and stability ‘Trewartha, G. T. 1954. An Introduction to Climate. McGraw-
in grasslands. Nature (London) 367:363-365. Hill, New York. 402 pp.
Timmer, V. R., and E. L. Stone. 1978. Comparative foliar Trexler, M. C., and C. A. Haugen. 1993. Keeping it Green.
analysis of young balsam fir fertilized with NPK. and Using Tropical Forestry to Mitigate Global Warming. World
lime. Soil Sci. Soc. Amer. F. 42:125-130. Resources Institute, Washington, DC.
 REFERENCES CITED R-49

Triska, F.J., and K. Cromack,Jr. 1980. The role of wood de- late trajectories in restoration ecology. Marine Pollut.
bris in forests and streams. pp. 171-190. In R. H. Waring Bull. 37:404-419.
(ed.), Forests: Fresh Perspectives from Ecosystem Analysis. Ugolini, F. C. 1966. Soils. pp. 29-72. In Soil Development and
Oregon State Univ. Press, Corvallis. Ecological Succession in a Deglaciated Area of Muir Inlet,
Troll, C. 1939. Luftbildplan und okologische Boden- Southeastern Alaska. Institute for Polar Studies, Report
forschung. pp 241-298. Z. Ges. Erdkunde, Berlin. 20, Part 3. Ohio State Univ., Columbus.
‘Truog, E. 1947. Soil reaction influence on availability of , and D. H. Mann. 1979. Biopedological origin of
plant nutrients. Soil Sci. Soc. Amer. Proc. 11:305-308. peatlands in south east Alaska. Nature 281:366-368.
Trust, T. J., and R. W. Coombs. 1973. Antibacterial activity Ulrich, A., and F. J. Hill. 1967. Principles and practices of
of S{—thujaplicin. Can. 7. Microbiol. 19:1341-1346. plant analysis. pp. 11-24. In Soil Testing and Plant Analysis.
Tsutsumi, T. 1971. Accumulation and circulation of material Part II. Plant Analysis. Spec. Publ. Ser. No. 2. Soil Sci.
elements in forest ecosystems. pp. 543-552. In P. Du- Soc. Amer., Madison, Wis.
vigneaud (ed.), Productivity of Forest Ecosystems. UN- Ulrich, B., R. Mayer, and P. K. Khanna. 1980. Chemical
ESCO, Paris. changes due to acid precipitation in a loess-derived soil in
Tubbs, C. H. 1973. Allelopathic relationships between yel- central Europe. Soil Sci. 130:193-199.
low birch and sugar maple seedlings. For Sci. 19: , P. Benecke, W. F. Harris, P. K. Khama, and R.
139-145. Meyer. 1981. Soil processes. pp. 265-338. In D. E. Re-
Tucker, G. F, and W. H. Emmingham. 1977. Morphologi- ichle (ed.), Dynamic Properties of Forest Ecosystems. Inter-
cal changes in leaves of residual western hemlock after national Biological Programme 23. Cambridge Univ.
clear and shelterwood cutting. For. Sci. 23:195-203. Press, Cambridge.
Tukey, H. B.,Jr.1969. Implications of allelopathy in agricul- UN Univ. & Div. Economic and Social Info./DPI. 1982.
tural plant science. Bot. Rev. 35:1-16. Dev. Forum 10(5):8.
. 1970. The leaching of substances from plants. Ann. UNESCO. 1978. Tropical Forest Ecosystems. A State-of-
Rev. Plant Physiol. 21:305-324. Knowledge Report. UNESCO/UNEP/FAO. Nat. Resour.
, and J. V. Morgan. 1963. Injury to foliage and its ef- Res. UNESCO. Paris. 683 pp.
fects upon the leaching of nutrients from aboveground Unger, L. 1992. Dwarf Mistletoes. Forestry Canada. Forest
plant parts. Physiol. Plant. 16:557-564. Pest Leaflet No. 44. Pacific Forestry Centre, Victoria,
Turmanina, V. I. 1965. The strength of tree roots. Bjull. BC. 7 pp.
Mosk. Obsc. Ispry. Prir. (Otd. Biol.) 70:36-45. Urban, D. L. 1990. A Versatile Model to Simulate Forest Pat-
Turner, J. 1975. Nutrient Cycling in a Douglas-fir Ecosys- tern—A _User’s Guide to Zelig Version 1.0. Environ. Sci.
tem With Respect to Age and Nutrient Status. Doctoral. Dept., Univ. Virginia, Charlottesville. 39 pp.
thesis, Univ. of Wash., Seattle. . 1993. A User’s Guide to ZELIG Version 2 with Notes on
. 1977. Effect of nitrogen availability on nitrogen cy- Upgrades From Version 1. Department of Forest Sciences,
cling in a Douglas-fir stand. For: Sci. 23:307-316. Colorado State University, Fort Collins, Colo. 39 pp.
, and M. J. Singer. 1976. Nutrient distribution and , and H. H. Shugart. 1992. Individual-based models
cycling in a subalpine coniferous forest ecosystem. 7. of forest succession. pp. 249-292. In D. G. Glen-Lewin,
Appl. Ecol. 13:295-301. R. K. Peet, and T. T. Veblen (eds.), Plant Succession. Theo-
Turner, N. 1963. The gypsy moth problem. Conn. Agric. ry and Prediction. Chapman & Hall, London.
Exp. Sta. Bull. 655. Urban, D. L. 1990. A Versatile Model to Simulate Forest Pattern:
Turner, M. G. 1989. Landscape ecology: The effect of pat- A User’s Guide to Zelig Version 1.0. Environmental Science
tern on process. Annu. Rev. Ecol. System. 20:171-197. Department University of Virginia, Charlottesville.
Turner, M. G., R. H. Gardner, and R. V. O’Neill. 2001. Urban, D. L. 1993. A user’s Guide to ZELIG Version 2 with
Landscape ecology in theory and practice: Pattern and Notes on Upgrades from Version 1. Department of Forest
process. Springer-Verlag, New York. Sciences, Colorado State University, Fort Collins,
Turner M. G., W. H. Romme, R. H. Gardner, R. V. O’Neill, CO.
and T. K. Kratz. 1993. A revised concept of landscape U.S. Corps of Engineers. 1956. Snow Hydrology. Summary
equilibrium: Disturbance and stability on scaled land- Report of the Snow Investigations. North Pac. Dw.
scapes. Landscape Ecol. 8:213-227. USCE. Portland, Ore. 437 pp.
Turrell, F M. 1934. Leaf surface of a twenty-one-year-old U.S. Department of Agriculture. 1958. Alpine Snowfields
catalpa tree. Iowa Acad. Sci. Proc. 41:79-84. Gain Moisture From Air. USDA For. Serv., Rocky Mt.
. 1936. The area of the internal exposed surface ofdi- For. Range Exp. Sta. Ann. Rept. 1957:3-4.
cotyledon leaves. Amer. 7. Bot. 23:255-264. . 1964. Winds Over Wild Lands—A Guide for Forest
Twilley, R., V. Rivera-Monroy, R. Chen, and L. Botero. Management. USDA For. Serv. Agric. Handbook No.
1998. Adapting an ecological mangrove model to simu- 272. Washington, DC. 33 pp.
R-50 REFERENCES CITED

. 1971. Prescribed Burning Symposium. USDA For. . 1984. Nutrient storage, retranslocation and rela-
Serv., SE For. Exp. Sta., Ashville, SC. 160 pp. tionship of stress to nutrition. pp. 181-209. In G. D.
. 1979. Effects of Fire on Soil. USDA For. Serv., Bowen and E. K. $. Nambiar (eds.), Nutrition of Planta-
Gen. Tech. Rept. WO-7, 34 pp. tion Forests. Academic Press, London.
. 1980. Environmental Consequences of Timber van der Kamp, B. J. 1986. Nitrogen fixation in cottonwood
Harvesting. Symp. Proc. USDA For. Serv., Gen..Tech. wetwood. Can. 7. For. Res. 16:1118-1120.
Rept. INT-90, Ogden, Utah. 526 pp. , andJ.Worrall. 1990. An unusual case of winter bud
USDA Forest Service. 1993. Healthy Forests for America’s damage in British Columbia interior conifers. Can. 7. For:
Future. A strategic plan. USDA For. Serv. MP-1513. 58 pp. Res. 20:1640-1647.
Uvarov, B. P. 1931. Insects and climate. Trans. Entomol. Soc. van der Maarel, E. (ed.). 1980. Succession, W. Junk b.v., The
London 79:\-247. Hague. 151 pp.
. 1961. Quantity and quality in insect populations. . 1993. Some remarks on disturbance and its relations
Proc. R. Entomol. Soc. London (C). 25:52-59. to diversity and stability. 7. Veg. Sci. 4:733-736.
Vaartaja, O. 1952. Forest humus quality and light conditions Vandermeer,J.H. 1969. The competitive structure of com-
as factors influencing damping-off. Phytopathology munities: An experimental approach with protozoa.
42:501-506. Ecology 50:362—371.
. 1959. Evidence of photoperiodic ecotypes in trees. Van der Tak, J., C. Haub, and E. Murphy. 1979. Our popu-
Ecol. Monogr. 29:91-111. lation predicament: a new look. Popul. Bull. 34(5):1-48.
, and P. J. Salisbury. 1965. Mutual effects im vitro of van Dobben, W. H., and R. H. Lowe-McConnell (eds.).
micro-organisms isolated from tree seedlings, nursery 1975. Unifying Concepts in Ecology. W. Junk b.v., The
soil and forests. For: Sci. 11:160-168. Hague. 302 pp.
Van Arsdel, E. P. 1965. Micrometeorology and plant disease Van Dyne, G. M. 1969. The Ecosystem Concept in Natural Re-
epidemiology. Phytopathology 55:945—950. source Management. Academic Press, New York. 383 pp.
van Breemen, N. 1990. Deterioration of forest land as a re- Van Eimern, J., R. Karschon, L. A. Razumova, and G. W.
sult of atmospheric deposition in Europe: A review. pp. Robertson. 1964. Windbreaks and Shelterbelts. Tech.
40-48. In S. P. Gessel, D. S. Lacate, G. F. Weetman, and Note No. 59. World Meteorological Organization.
R. F. Powers (eds.), Sustained Productivity of Forest Soils. WMO. No. 147. TP. 70. Geneva.
Proc. 7th N. Am. For. Soils Conf. Univ. of B.C., Faculty of Van Kooten, G. C., W. A. Thompson, and I. Vertinsky.
Forestry, Vancouver. 1993. Economics of reforestation in British Columbia
van Cleve, K., FE. S. Chapin, II, P. W. Flanagan, L. A. when benefits of CO, reduction are taken into account.
Viereck, and C. T. Dyrmess (eds.). 1986. Forest Ecosystems pp. 227-247. In W. L. Adamomicz, W. White, and W. E.
in the Alaskan Taiga. A Synthesis of Structure and Function. Phillips (eds.), Forestry and the Environment: Economic Per-
Ecol. Studies 57. Springer-Verlag, New York. 230 pp. spectives. CAB International, Wallingford, UK.
, and L. L. Noonan. 1971. Physical and chemical van Miegroet, H., D. W. Cole, and P. S. Homann. 1990.
properties of the forest floor in birch and aspen stands in The effect of alder forest cover and alder forest conver-
interior Alaska. Proc. Soil Sci. Soc. Amer. 35:356-360. sion on soil fertility and productivity. pp. 333-354. In S.
, and L. A. Viereck. 1981. Forest succession in rela- P. Gessel, D. S. Lacate, G. F. Weetman, and R. F. Powers
tion to nutrient cycling in the boreal forest of Alaska. pp. (eds.), Sustained Productivity of Forest Soils. Proc. 7th N.
185-211. In D. C. West, H. H. Shugart, and D. B. Botkin Am. For. Soils Conf. Faculty of Forestry, Univ. of B.C.,
(eds.), Forest Succession. Concepts and Application. Springer- Vancouver.
Verlag, New York. , D. Zabowski, C. T. Smith, and H. Lundkvist. 1994.
van Daalen, J. C., and H. H. Shugart. 1989. OUTENI- Review of measurement techniques in site productivity
QUA—A computer model to simulate succession in the studies. pp. 287-362. In W. J. Dyck, D. W. Cole, and
mixed evergreen forests of the southern Cape, South N. B. Comerford (eds.), Impacts of Forest Harvesting on
Africa. Landscape Ecol. 2:255-267. Long-term Site Productivity. Chapman & Hall, London.
van Daalen,J.C., and H. H. Shugart. 1989. OUTENIQUA: Van Vechten, G. W. 1960. The ecology of the timberline
A computer model to simulate succession in the mixed and alpine vegetation of the Three Sisters, Oregon. Doc-
evergreen forests of the southern Cape, South Africa. toral. thesis, Oregon State Univ., Corvallis. 111 pp.
Landscape Ecol. 2:255-267. Van Wagner, C. E. 1972. Duff consumption by fire in east-
Van den Burg, J. 1976. Problems related to analysis of soil ern pine stands. Can. 7. For: Res. 2:34-39.
fertility. XVI IUFRO World Congress, Oslo. Proc. Div. Varley, G. C. 1970. The concept of energy flow applied to a
1:148-163. woodland community. pp. 389-405. In A. Watson (ed.),
van den Driessche, R. 1974. Prediction of mineral nutrient Animal Populations in Relation to Their Food Resources.
status of trees by foliar analyses. Bot. Rev. 40:347-394. Blackwell, Oxford.
 REFERENCES CITED R-51

, G. R. Gradwell, and M. P. Hassel. 1973. Insect Popu- , J. R. Gosz, C. C. Grier, J. M. Melillo, W. A. Rein-
lation Ecology: An Analytical Approach. Blackwell Sci. ers, and R. L. Todd. 1979. Nitrate losses from disturbed
Publ., Oxford. 212 pp. ecosystems. Science 204:469-473.
Vasil’eva, I. N. 1968. Physical properties and water require- , and W. A. Reiners. 1975. Ecosystem succession and
ments of soils in the Sereboyanyi Bor experimental for- nutrient retention: A hypothesis. BioScience 25:376-381.
est. pp. 10-50. In L. P. Rysin and M. V. Nadezhdina , and R. L. Sanford. 1987. Nutrient cycling in moist
(eds.), Long-Term Biogeocoenotic Investigations in the South- tropical forest. Annu. Rev. Ecol. Syst. 17:137-167.
ern Taiga Zone. Transl. 1964 Symposium, USDA and Vogt, K. A., C. C. Grier, and D. J. Vogt. 1986. Production,
Natural Science Foundation, Washington, DC. turnover and nutrient dynamics of above- and below-
Veblen, T. T. 1989. Tree regeneration response to gaps ground detritus of world forests. Adv. Ecol. Res.
along a transandean gradient. Ecology 70:541-543. 15:303-377.
. 1992. Regeneration dynamics. pp. 152-187. In , R. L. Edmonds, C. C. Grier, and S. R. Piper. 1980.
D. C. Glenn-Lewin, R. K. Peet, and T. T. Veblen (eds.), Seasonal changes in mycorrhizal and fibrous-textured
Plant Succession. Theory and Prediction. Chapman & Hall, root biomass in 23- and 180-year-old Pacific silver fir
London. stands in western Washington. Can. 7. For. Res.
Veihmeyer, F. J. 1956. Soil Moisture. Handb. Pflanzen Physi- 10:523-529.
ol. 3:64-123. , R. L. Edwards, C. C. Grier, and S. R. Piper. 1981.
Vézina, P. E., and D. W. K. Boulter. 1966. The spectral Seasonal changes in biomass and vertical distribution of
composition of near ultraviolet and visible radiation be- mycorrhizal and fibrous-textured conifer fine roots in
neath forest canopies. Can. 7. Bot. 44:1267-1284. 23- and 180-year-old subalpine Abies amabilis stands. Can.
Viereck, L. A. 1965. Relationship of white spruce to lenses f. For. Res. 11:223-229.
of perennially frozen ground, Mount McKinley National , C. C. Grier, C. E. Meier, and R. L. Edmonds. 1982.
Park, Alaska. Arctic 18:262-267. Mycorrhizal role in net primary production and nutrient
. 1970. Forest succession and soil development adja- cycling in Abies amabilis ecosystems in western Washing-
cent to the Chena River in interior Alaska. Arctic Alpine ton. Ecology 63:370-380.
Res. 2(1):1-26. 7 @. Go Grier, Cal Meier, andsMi Rakevesn 1983).
. 1973. Ecological effects of river flooding and forest Organic matter and nutrient dynamics in forest floors of
fires on permafrost in the taiga of Alaska. pp. 60-67. In young and mature Abies amabilis stands in western Wash-
Permafrost. The N. Amer. Contrib. ‘To the Second Inter- ington, as affected by fine-root input. Ecol. Monogr.
nat. Conf. Nat. Acad. Sci. Washington, DC. 53:139-157.
Vince-Prue, D. 1975. Photoperiodism in Plants. McGraw- Voigt, G. K. 1960. Distribution of rainfall under forest
Hill, New York. 444 pp. stands. For. Sci. 6:2-9.
Viro, P. J. 1953. Loss of nutrients and the natural nutrient . 1968. Variation in nutrient uptake by trees. pp.
balance of the soil in Finland. Com. Inst. For. Fenn. 20-27. In Forest Fertilization, Theory and Practice. Proc.
42:1-50. Symp., Gainesville, Fla. TVA, Muscle Shoals, Ala.
. 1974. Effects of forest fire on soil. pp. 7-45. In T. T. . 1971. Mycorrhizae and nutrient immobilization. pp.
Kozlowski and C. E. Ahlgren (eds.), Fire and Ecosystems. 122-131. In Mycorrhizae. Proc. 1st N. Amer. Conf. on
Academic Press, New York. Mycorrhizae. USDA For. Serv., Misc. Publ. 1189.
Vitousek, P. M. 1977. The regulation of element concentra- , B. N. Richards, and E. C. Mannion. 1964. Nitrogen
tions in mountain streams in the northeastern United utilization by young pitch pine. Soil Sci. Soc. Amer. Proc.
States. Ecol. Monogr. 47:65-87. 28:707-709.
. 1981. Clearcutting and the nitrogen cycle. pp. Volterra, V. 1926. Variazioni e gluttnazioni del numero d’in-
631-642. In F. E. Clark and T. H. Rosswall (eds.), dividui in specie animali conviventi. Mem. Acad. Lincei
Nitrogen Cycling in Terrestrial Ecosystems: Processes, Roma. 2:31-113.
Ecosystem Strategies, and Management Implications. Ecol. Von Frisch, K. 1967. The Dance Language and Orientation of
Bull. 33, Swedish Nat. Sci. Research Council, Stock- Bees. Harvard Univ. Press, Cambridge. 566 pp.
holm. Wagener, W. W. 1961. Past fire incidence in Sierra Nevada
, andJ. S. Denslow. 1986. Nitrogen and phosphorus forest. 7. For. 59:739-748.
availability in treefall gaps of a lowland tropical rain for- Wagner, M. R. 1994. The healthy multiple-use forest
est. f. Ecol. 74:1167-1178. ecosystem: An impossible dream. pp. 185-188. In W. W.
, J. R. Gosz, C. C. Grier, J. M. Melillo, and W. A. Covington and L. F. DeBano (eds.), Sustainable Ecological
Reiners. 1982. A comparative analysis of potential nitrifi- Systems: Implementing an Ecological Approach to Land Man-
cation and nitrate mobility in forest ecosystems. Ecol. agement. USDA For. Serv. Gen. Tech. Rept. RM-247.
Monogr. 52:155-177. Fort Collins, Colo.
R-52 REFERENCES CITED

Wahl, E. W. 1968. A comparison of the climate of the east- Wang, Y.-P,, and P. G. Jarvis. 1990. Description and valida-
ern United States during the 1830’s with the current nor- tion of an array model—MAESTRO. Agr: For. Meteorol.
mals. Monthly Weather Rev. 96:73-82. 51:257-280.
, and T. L. Lawson. 1970. The climate of the mid Wardle, P. 1968. Engelmann spruce (Picea engelmannii
nineteenth century United States compared to current Engel.) at its upper limits on the Front Range, Colorado.
normals. Monthly Weather Rev. 98:259-265. Ecology 49:483—495.
Waide,J.B., and W. T. Swank. 1975. Nutrient recycling and . 1974. Alpine timberlines. In J. D. Ives and R. G.
the stability of ecosystems: Implications for forest man- Barry (eds.), Arctic and Alpine Environments. Methuen,
agement in the southeastern U.S. pp. 404-424. In Proc. London.
1975 SAF Annual Meeting. S.A.F., Washington, DC. Wareing, P. F. 1969. Germination and dormancy. In M. B.
, and W. T. Swank. 1977. Simulation of potential ef- Wilkins (ed.), The Physiology of Plant Growth and Develop-
fects of forest utilization on the nitrogen cycle in different ment. McGraw-Hill, New York.
southeastern ecosystems. pp. 767-789. In D. L. Correll Waring, R. H. 1980. Vital signs of forest ecosystems. pp.
(ed.), Watershed Research in Eastern North America. A 131-136. In R. H. Waring (ed.), Forests: Fresh Perspectives
Workshop to Compare Results. Vol. I. Chesapeake Bay Cntr. from Ecosystem Analysis. Oregon State Univ. Press, Cor-
Environ. Studies. Smithsonian Inst. Edgewater, Md. vallis.
Waksman, S. A. 1932. Principles ofSoil Microbiology. 2nd ed. , W. H. Emmingham, H. L. Gholz, and C. C. Grier.
Wilkins and Wilkins Co., Baltimore. 894 pp. 1978. Variation in maximum leaf area of coniferous
Waldrop, T. A., E. R. Buckner, H. H. Shugart, and C. E. forests in Oregon and its ecological significance. For. Sci.
McGee. 1986. FORCAT: A single-tree model of stand 24:13 1-140.
development on the Cumberland plateau. For: Sci. , and J. EF Franklin. 1979. Evergreen coniferous
32:297-317. forests of the Pacific Northwest. Science 204:1380-1386.
Waldrop, T. A., E. R. Buckner, H. H. Shugart, and C. E. , R. H., K. Newman, and J. Bell. 1981. Efficiency of
McGee. 1986. FORCAT: A single-tree model of stand tree crowns and stemwood production at different
development on the Cumberland plateau. For Sci. canopy densities. Forestry 54:129-137.
32:297-317. , J.J. Rogers, and W. 'T. Swank. 1981. Water relations
Wali, M. K., and V. J. Krajina. 1973. Vegetation environ- and hydrologic cycles. pp. 205-264. In D. E. Reichle
ment relationships of some sub-boreal spruce zone (ed.), Dynamic Properties of Forest Ecosystems. Internat.
ecosystems in British Columbia. Vegetatio 26:237-381. Biol. Programme 23. Cambridge Univ. Press, Cam-
Wall, G. 1991. Implications ofClimate Change for Pacific North- bridge.
west Forest Management. Dept. Geography, Univ. of Water- , W. G. Thies, and D. Muscato. 1980. Stem growth
loo, Occasional Paper No. 15. Waterloo, Ontario. 221 pp. per unit of leaf area: A measure of tree vigour. For. Sci.
Wallwork, J. A. 1970. Ecology ofSoil Animals. McGraw-Hill, 26:112-117.
New York. Waring, R. H., and S. W. Running. 1998. Forest Ecosys-
Waloff, Z. 1966. The upsurges and recessions of the desert tems—Analysis at Multiple Scales. 2nd Ed. Academic
locust plague: A historical survey. Anti-locust Memoirs Press, San Diego, CA. 370 pp.
8:1-111. Warming, E. 1909. Oecology of Plants: An Introduction to the
Walstad, J. D., and P. J. Kuch (eds.). 1987. Forest Vegetation Study ofPlant Communities. Oxford Press, Oxford. 422 pp.
Management for Conifer Production. John Wiley, New Wassink, E. C. 1959. Efficiency of light energy conversion
York. 523 pp. in plant growth. Plant Physiol. 34:356-361.
, J. D., S. R. Radosevich, and D. V. Sandberg. 1990. Watson, R. T., H. Rohde, H. Oeschger, and U. Siegent-
Natural and Prescribed Fire in Pacific Northwest Forests. holer. 1990. Greenhouse gases and aerosols. pp. 1-40. In
Oregon State University Press, Corvallis, Ore. J. T. Houghton, G. J. Jenkins, andJ.J. Ephraums (eds.),
Walter, H., E. Harnickell, and D. Mueller-Dombois. 1975. Climate Change. The IPCC Scientific Assessment. Cam-
Climate-Diagram Maps ofthe Individual Continents and the bridge Univ. Press, Cambridge, UK.
Ecological Climatic Regions of the Earth. Springer-Verlag, Watson, R. E., D. C. Albritton and the CORE Writing
Berlin. 36 pp. and 9 maps. Team. (eds.) 2001. Climate Change 2001: Synthesis Report.
Walters, C. J. 1986. Adaptive Management of Renewable Re- Intergovernmental Panel on Climate Change. Cambridge
sources. Macmillan, New York. 374 pp. University Press, Cambridge, UK. 397 pp.
, and C. S. Holling. 1990. Large-scale management Watt, A. S. 1919. On the causes of failure of natural regener-
experiments and learning by doing. Ecology ation in British oakwoods.7. Ecol. 7:173-203.
71:2060-2068. . 1934. The vegetation of Chiltern with special refer-
Wang, J. Y. 1960. A critique of the heat unit approach to ence to the beechwoods and their seral relationships. Part
plant response studies. Ecology 41:785-790. I. 7. Ecol. 22:230-270.
 REFERENCES CITED R-53

- 1947. Pattern and process in the plant community. land Res. Index No. 183. Pulp Pap. Res. Instit. Can. 38
4. Ecol, 35:1-22. PP:
- 1955. Bracken versus heather, a study in plant soci- , and B. Webber. 1972. The influence of wood har-
ology. 7. Ecol. 43:490-506. vesting on the nutrient status of two spruce stands. Can.
Watt, K. FE. 1968. Large-scale biological wave phenomena: F. For. Res. 2:351-369.
Fur bearer predator-prey cycles in the boreal forest and Wei, X., J. P. Kimmens and G. Zhou, 2003. Disturbances
outbreaks of insect pests. pp. 143-156. In Ecology and Re- and the sustainability of long-term site productivity in
source Management. McGraw-Hill, New York. lodgepole pine forests in the central interior of British
Watt, A. S. 1947. Pattern and process in the plant com- Columbia—an ecosystem modelling approach. Ecological
munity. 7. Ecol. 35:1-22. modelling 164:239-256.
Watts, J. A. 1982. The carbon dioxide question: data sam- Weidemann, E. 1951. Ertragskundliche und waldbauliche
pler. pp. 431-469. In W. C. Clark (ed.), Carbon Dioxide Grundlagen der Forstwirtschaft. Das Hauptergebnis zer
Review: 1982. Clarendon Press, Oxford. Zehnjaehrigen Arbeiten. der (ehem.) Presussischen
Weatherell, J. 1953. The checking of forest trees by heather. Forstlichen Versuchsanstalt.3.Auflage. J.D. Sauerlander’s
Forestry 26(1):37-40. Verlag. Frankfurt an Main. 346 pp.
Weaver, H. 1951. Fire as an ecological factor in the south- Wein, R. W. 1978. The role of fire in the degradation of
western ponderosa pine forests. 7. For. 49:93-98. ecosystems. pp. 193-209. In M. W. Holdgate and M. J.
Weaver, J. E. 1968. Prairie Plants and Their Environment. Woodman (eds.), The Breakdown and Restoration of Ecosys-
University of Nebraska Press, Lincoln. 276 pp. tems. Plenum, New York. 322 pp.
, and F. E. Clements, 1938. Plant Ecology. 2nd ed. , and D. A. MacLean (eds.). 1981. The Role of Fire in
McGraw-Hill, New York. 601 pp. Northern Circumpolar Ecosystems. Scope; Wiley and Co.,
, and N. W. Roland. 1952. Effects of excessive natur- ‘Toronto. 322 pp.
al mulch on development, yield and structure of native , andJ.M. Moore. 1977. Fire history and rotations in the
grassland. Bot. Gazette 114:1-19. New Brunswick Acadian forest. Can. 7. For. Res. 7:285-294.
Webb, T., II,J.E. Kutzbach, and EF A. Street-Perrot. 1985. Weinstein, D. A., H. H. Shugart, and D. C. West. 1982. The
20,000 years of global climate change: Paleoclimatic re- Long-Term Nutrient Retention Properties of Forest Ecosys-
search plan. pp. 182-218. In T. F Malone and J. G. tems: A Simulation Investigation. Oak Ridge National
Roederer (eds.), Global Change. Cambridge Univ. Press, Lab., Oak Ridge, Tenn., ORNL/TM-8472.
Cambridge, UK. Weinstein, D. A., H. H. Shugart, and D. C. West. 1982. The
Webber, B. D. 1977. Biomass and nutrient distribution pat- long-term nutrient retention properties of forest ecosys-
terns in a young Pseudotsuga menziesii ecosystem. Can. Ff. tems: A simulation investigation. Oak Ridge National
For. Res. 7:326-334. Laboratory, Oak Ridge, TN, ORNL/TM-8472.
Weber, A., B. Gilbert, J. P. Kimmins, and C. E. Prescott. Weir, J. 1969. Importation of nutrients into woodlands by
2003. Factors limiting the early survivorship of western rooks. Nature 221:487-488.
redcedar (Thuja plicata) on northern Vancouver Island. Weiser, C. J. 1970. Cold resistance and injury in woody
Can. 7. For. Res. 33:854-861. plants. Science 169:1269-1278.
Webster,J. R.,J.B. Waide, and B. C. Patten. 1975. Nutrient Weller, D. A. 1987. A re-evaluation of the -3/2 power rule
recycling and the stability of ecosystems. pp. 1-27. In of plant self thinning. Ecol. Monogr. 57:23-43.
Howell, F. G. et al. (eds.), Mineral Cycling in Southeastern Wellington, W. G. 1952. Air mass climatology of Ontario
Ecosystems. Tech. Info. Centre, U.S. Energy Res. & north of Lake Huron and Lake Superior before out-
Devel. Admin. CONF-7405 13. breaks of the spruce budworm and the forest tent cater-
Wedin, D. A., and D. Tilman. 1990. Species effects on nitro- pillar. Can. 7. Zool. 30:114-127.
gen cycling: A test with perennial grasses. Oecologia . 1954. Weather and climate in forest entomology.
84:43 1-433. Meteorol. Monogr. 2:11-18.
Weetman, G. F. 1967. Nitrogen deficient black spruce on . 1957. Individual differences as a factor in popula-
raw humus soils in northern Quebec—Response to thin- tion dynamics: The development of a problem. Can. 7.
ning and urea treatments. Proc. XIVth IUFRO Con- Zool. 35:293-323.
gress, Miinchen. IV:608-611. IUFRO Secretariat, . 1960. Qualitative changes in animal populations
Federal Forest Research Inst., Vienna. during changes in abundance. Can. 7. Zool. 38:289-3 14.
, and R. Fournier. 1983. Graphical diagnoses of . 1964. Qualitative changes in populations in unstable
lodgepole pine responses to fertilization. Soi! Sci. Soc. environments. Can. Entomol. 96:436-451.
Amer. 7. 46:1280-1289. . 1965. Some maternal influences on progeny quality
, and V. Timmer. 1967. Feather Moss Growth and in the western tent caterpillar Malacosoma pluviale
Nutrient Content Under Upland Black Spruce. Wood- (Dyar.). Can. Entomol. 97:1-14.
 RS RROD
Wek © &_ ae§_ 2 Ws, RE SE ee ee
Ne
RRR Se Ree, RR RB ORE
——~nt 2 ee RRS:

<= BWA i RR aaa

~

WOR RA VA Ve. ha. eR a See

WR, BR RA Rae HRA RRR SA
a waatentaies d
en, SR. Tee ee
wake Se
ae
Week FR aa \ Sek PER Bike, Rete
RDS SAR
Wess,RR &_al
F Geek, SN eS
Fe
WW WAY Bh Reseant§,
SR Be Ne Bie fine Me,
Bey Geeta
Wee DP CK HR Dheeex cui PD 5 ete, OR Ree
SS eee eee Li
SR Se
Wena, CF OOK Te pemtinee. 2 oR, ee,
SRE, ee SN RR
Ne ee
Se Geet. Ge EN Weierete- cn,
Ce ee ee
eek ofAQemeeeee Fe SP
“Thc Rae
Were MSG the ekg TR
aT

———. am. TR Rome SEE ite

Nee s,
 REFERENCES CITED R-55

Wickman, B. E. 1992. Forest Health in the Blue Mountains: ley (eds.), Ecophysiology of Coniferous Forests. Academic
The Influence ofInsects and Disease. USDA For. Serv. Gen. Press, New York.
‘Tech. Rept. PNW-GTR-295. Portland, Ore. 15 pp. Woodwell, G. M. 1970. The energy cycle of the biosphere.
Wiebes, J. T. 1979. Co-evolution of figs and their insect pol- Sci. Amer. 223:64-74.
linators. Ann. Rev. Ecol. Syst. 10:1-12. . 1982. Earth’s vegetation and the carbon dioxide
Wiedemann, E. 1935. Uber die schaeden der Streunutzung question. pp. 399-400. In W. C. Clark (ed.), Carbon Diox-
im Deutschen Osten. Forstarchiv. 11:386-390. ide Review: 1982. Clarendon Press, Oxford.
Wiedmann, E. 1935. [Damage due to litter utilization in , and D. B. Botkin. 1970. Metabolism of terrestrial
eastern Germany]. Forstarchiv. 11:386-390. ecosystems by gas exchange techniques: The Brookhaven
Wiegert, R. G., and F. C. Evans. 1967. Investigations of sec- approach. pp. 73-85. In D. E. Reichle (ed.), Analysis of
ondary productivity in grasslands. pp. 499-518. In K. Temperate Forest Ecosystems. Springer-Verlag, New York.
Petrusewicz (ed.), Secondary Productivity of Terrestrial , and R. H. Whittaker. 1967. Primary production and
Ecosystems. Polish Acad. Sci., Warsaw. the cation budget of the Brookhaven forest. pp. 151-166.
, and D. F. Owen. 1971. Trophic structure, available In Proc. Symp. Primary Production and Mineral Cycling in
resources and population density in terrestrial vs. aquatic Natural Ecosystems. Univ. of Maine Press, Orono.
ecosystems.7. Theoret. Biol. 30:69-81. , R. H. Whittaker, W. A. Reiners, G. E. Likens, C. C.
Wiens, J. A. 1989. Spatial scaling in ecology. Func. Ecol. Delwiche, and D. B. Botkin. 1978. The biota and the
3:385-397. world carbon budget. Science 199:141-146.
Wilczynski, C. J., and S. T. A. Pickett. 1993. Fine root bio- Wooldridge, D. D. 1960. Watershed disturbance from trac-
mass within experimental canopy gaps: Evidence for a tor and skyline crane logging. 7. For: 58:369-372.
below-ground gap.7. Veg. Sci. 4:571-574. World Commission on Environment and Development
Williams, G. R. 1954. Population fluctuations in some (WCED). 1987. Our Common Future. Oxford Univ.
northern hemisphere game birds (Tetraonidae). 7. Anim. Press, Oxford, UK. 400 pp.
Ecol. 23:1-34. Worrall, J. 1983. Temperature-bud burst relationships in an
Wilson, B. F. 1970. The Growing Tree. Univ. Massachusetts amabilis and subalpine fir provenance test replicated at
Press, Amherst, Mass. 152 pp. different elevations. Silvae Genet. 32:203-209.
Wilson, J. W. 1959. Notes on wind and its effects in arctic- Wright, H. E., Jr. 1974. Landscape development, forest
alpine vegetation.7. Ecol. 47:415—427. fires, and wilderness management. Science 186:487—495.
Winters, R. K. 1974. The Forest and Man. Vantage Press, Wuebbles, D. J., and J. Edmonds. 1988. A Primer on Green-
New York. 393 pp. house Gases. U.S. Dept. of Forestry, Office of Energy Re-
Witkamp, M. 1966. Decomposition of leaf litter in relation search, Washington, DC. DOE/NBB0083. 100 pp.
to environment, microflora, and microbial respiration. Wyne-Edwards, V. C. 1962. Animal Dispersion in Relation to
Ecology 47:194-200. Social Behavior. Oliver and Boyd, Edinburgh. 653 pp.
Wittich, W. H. L. 1962. Classification, mapping and inter- —-—. 1964. Population control in animals. Sci. Amer.
pretation of soils for forestry purposes. Proc. Sth World 211:68-74.
For. Congr., Seattle, Washington 1960. 5:502-507. . 1965. Self-regulating systems in populations of ani-
Wollum, A. G., I, and G. B. Davey. 1975. Nitrogen accu- mals. Science 147:1543-1548.
mulation, transformation and transport in forest soils. pp. Xiao, G. 1994. The role of root-associated fungi in the dom-
67-106. In B. Bernier and C. H. Winget (eds.), Forest inance of Gaultheria shallon. Doctoral. thesis. Dept. of
Soils and Forest Land Management. Laval Univ. Press, Botany, Univ. of British Columbia, Vancouver. 137 pp.
Quebec. Yanai, R. D., T. J. Fahey, and S. L. Miller. 1995. Efficiency
Wong, S. C. 1980. Effects of elevated partial pressure of of nutrient acquisition by fine roots and mycorrhize. pp.
CO, on rate of CO, assimilation and water use efficiency. 75-103. In W. K. Smith and T. M. Hinkley (eds.),
pp. 159-166. In G. I. Pearman (ed.), Carbon Dioxide and Resource Physiology of Conifers. Acquisition, Allocation, and
Climate. Australian Research. Aust. Acad. Sci., Canberra. Utilization. Academic Press, New York.
Woodley, S.,J.Kay, and G. Francis. 1993. Ecological Integrity Yarie, J. 1980. The role of understory vegetation in the nu-
and the Management of Ecosystems. St. Lucie Press, Delray trient cycle of forested ecosystems in the Mountain
Beach, Fla. 220 pp. Hemlock Biogeoclimatic Zone. Ecology 61:1498—15 14.
Woods, P. V., E. K. S. Nambiar, and P. Smethurst. 1992. Ef- . 1981. Forest fire cycles and life tables: a case study
fect of annual weeds on water and nutrient availability to from interior Alaska. Can. 7. For. Res. 11:554-564.
Pinus radiata trees in a young plantation. For. Ecol. Man- Yarie, J. A. 1986a. A preliminary comparison of two ecosys-
age. 48:145—-163. tem models, FORCYTE-10 and LINKAGES for in-
Woodward, F. I. 1995. Ecophysiological controls of conifer terior Alaska which spruce. pp. 95-103. In G. I. Agren
distribution. pp. 79-94. In W. K. Smith and T. M. Hinck- (ed.), Predicting Consequences ofIntensive Forest Harvesting
R-56 REFERENCES CITED

on Long-term Productivity. Proc. IEA/FE Project CPC-10 Zavitovski, J., and M, Newton, 1968a, Ecological impor-
Workshop, Jadraas. Swedish Univ., Agric. Sciences Re- tance of snowbrush Ceanothus velatinas in the Oregon
port 26. Uppsala. Cascades. Ecology 49:1 134-L145,
. 1986b. FORCYTE—Enxtension of a stand level , and M. Newton. 1968b, Effect of organic matter
growth and yield model utilizing nitrogen dynamicsto and combined nitrogen on nodulation and nitrogen fixa-
taiga white spruce forests. pp. 190-204. In K. van Cleve, tion in red alder. pp. 200-223. In J. M. Trappe et al.
EF. S. Chapin, Il, P. W. Flanagan, L. A. Dyrness, and (eds.), Biology of Alder. USDA For. Serv, PNW For.
C. T. Viereck (eds.), Forest Ecosystems of the Alaskan Taiga. Range Exp. Sta., Portland, Ore.
A Synthesis of Structure and Function. Springer-Verlag, Zeide, B. 1987, Analysis of the 3/2 power law of self-
New York. thinning. For: Sed, 33:517-537,
Yin, X., J. A. Perry, and R. K. Dixon. 1989. Influence of Zeman, L. J. 1973. Chemistry of trophospheric fallout and
canopy removal on oak forest floor decomposition. Can. streamflow in a small mountainous watershed near Van-
J. For. Res. 19:204-214. couver, British Columbia. Doctoral. thesis, Univ. of B.C.
Yoda, K., T. Kira, H. Ogawa, and K. Hozumi. 1963. Self- Vancouver.
thinning in overcrowded pure stands under cultivated 1975. Hydrochemical balance of a British
and natural conditions. 7 Biol. Osaka City Univ. Columbian mountainous watershed, Catena 2:81-94,
14:107-129. Zimmerman, R. R.,J.P. Greenberg, S. QO. Wandiga, and P.J.
Young, H. E. 1968. Challenge of complete tree utilization. Crutzen. 1982. ‘Termites: A potentially large source of at-
For. Prod.7.18:83-86. mospheric methane, carbon dioxide and molecular hy-
. 1972. Woody Fibre Farming: An Ecologically Sound and drogen. Science 218:563-S65.
Productive Use of Right-of-Ways. Univ. Maine Sch. For. Zlotin, R. I., and K. S. Khodashova, 1980, The Role ofAn-
Res. Pap, Orono. 19 pp. imals in Biological Cycling of Forest-Steppe Ecosystems.
Youngberg, C. T. 1966. Forest floors in Douglas-fir forests: Dowden, Hutchinson and Ross, Stroudsburg, Pa.
dry weight and chemical properties. Soil Sci. Soc. Amer. 221 p.
Proc. 30:406—409. Zobel, D. B., W. A. McKee, G. M. Hawk, and C. T. Dyr-
Yu-Zhen L., Z.-S. Dong, $. Garman, Z. L. Yu, and S. D. ness. 1974. Correlation of forest communities with envi-
Zhao. 1998. A succession and silviculture model for the ronment and phenology on the H. J. Andrews
broad-leaved Korean pine forests of Changbai Mountain Experimental Forest, Oregon. pp, 48-36. Im RH. War-
area.7.For Res. 9:176-118. ing and R. L. Edmonds (eds.), Integrated Research in the
Zabowski, D., M. F. Skinner, and P. T. Rygiewicz. 1994. Coniferous Forest Biome. Bull. No. 5. Conit. For. Biome,
Timber harvesting and long-term productivity: Weather- US/IBP, Univ. Washington, Seattle.
ing processes and soil disturbance. For: Ecol. Manage. , A. McKee, G. M- Hawk, and C. 'T. Dyrness. 1976,
66:55-68. Relationship of environment to composition, structure
Zasada, J. C., T. L. Sharik, and M. Nygren. 1991. The re- and diversity of forest communities of the central western
productive process in boreal forest trees. pp. 85-134. In Cascades of Oregon. Ecol. Monogr: 46:135-156,
H. H. Shugart, R. Lemans, and G. B. Bonan (eds.), 4 Sys- Zwolinski, M. J. 1971. Effects of fire on water infiltration
tems Analysis of the Global Boreal Forest. Cambridge Univ. rate in a ponderosa pine stand. Hydrol. Water Resour, Aris.
Press, Cambridge, UK. Southwest 1:107-112.
 Index

A Ant(s), in soil, 308-309 methods used in, 165, 166

Abiotic theory of population regulation, 385, 386 Antagonistic interactions between species in overall structure of system for, 165, 167
Aboveground litterfall, energy loss due to, communities, 416-424 topographic sequences of sites types within
49-50 allelopathy and, 420-422 subzones and, 163—164
Acclimation, 453 antibiosis and, 419-420 zones and subzones in, 161-164
Acidity. See pH competition and, 422-424 Biogeoclimatic regions, 159
Acid rain, 122-124 physical exploitation and, 416-419 Biogeoclimatic zones, 159, 161-162
Actinomycetes, in soil, 301 Anthocyanin pigment, 174-175 topographic sequences of site types within,
Adaptation, forestry and, 456-457 Antibiosis, 419-420 163-164
Adaptive management (AM), 518 species displacement by, 485-486 Biological diversity, 406, 429-438
Adhesive forces, 290 Apparent photosynthesis, definition of, 43 ecological, 429
Adsorption, 291 Arbuscules, of mycorrhizae, 302 genetic variation as measure of, 442-448
Aeolian soil, 315 Assart effect, 127-128 global patterns of, 432-436
Aeration Assimilation efficiency, 53 island biogeography theory and forest
energy flow increases in detritus food webs Autecology, 27 fragmentation and, 436-437
after clearcutting and, 68 Autogenic succession, 31, 33, 465 measures of, 429-431
of soil, 289 community structure as result of, 410-411 spatial scales for evaluating, 431-432
Age-distribution diagrams, 377 Autotoxicity, species displacement by, 486 stability of communities and ecosystems and,
Aggregation, biotic factors and, 368 Autotrophs, 36 437-438
Agriculture in soil, 299 Biological evolution, 441-442
development of, 4-5, 12, 13-14 Biological integration, levels of, 29-31, 32
slash-and-burn (swidden), 264 B Biological mechanisms, in sedimentary cycles,
Air, evaporative power of, 270 Bacteria, in soil, 299-301 84-85
Albedo, 172 Bark, fire-resistant, 338 Biological organization, levels of, 27-28
radiation budgets and, 204 Belowground litterfall, energy loss due to, 50 Biological spectrum, 148
Algae, in soil, 304 Bergman’s rule, 220-221 Biomass
Alkalinity. See pH Beta diversity, 431 definition of, 43
Allelochemicals, 420, 421 Bioassay models, historical, 577 distribution between parts of plants, 51
Allelopathy, 420-422 Biochemical cycles, 74, 105-110 ecological pyramid of, 38-39
Allen rule, 220-221 succession and, 507 of forest, redistribution of, 66-67
Allogenic mechanisms of ecological succession, Bioclimatic rule, 228 net productivity and, 50-51
sil Biodiversity of ecosystems, stability and Biomes, 228, 410
Allogenic succession, 465-466 successional stage related to, 507-510 Biophysical classification, 165-166
Alluvial soil, 315 Biogenic succession, 31, 466 Biota, rate of environmental alteration by, 489
Alpha diversity, 431 Biogeochemical cycles, 7+ Biotic climaxes, 470
Alpha species diversity, 441 within ecosystems, 86-105 Biotic communities, 405
Ammonification, 113 direct nutrient cycling and, 105 patterns of. See Forest ecotones; Species
Anabatic winds, 237 litter decomposition and, 98-102 distribution
Anemochory, 241 nutrient distribution in plants and, 91 Biotic environment, coevolution and, 454-456
Animal(s) nutrient losses from plants and, 91-97 Biotic potential, 30, 374
behavior of nutrient uptake by plants and, 86-91 Biotic theory of population regulation, 385
temperature adaptation of, 225-226 understory vegetation in, 103-105 Black water rivers, 118
water excess or deficit and, 275-276 within plants, 105-110 Blocky soil, 287
wind effects on, 246-247 succession and, 506-507 Boron, deficiency of, plant visual symptoms
effects of fire on, 341-343 Biogeochemistry, 72-138 accompanying, 314
hibernation in, 222-223 of animal nutrition, 79-81 Boundary layer, 268
nutrition of, 79-81 biogeochemical cycle and. See Branchioles, 275
photoperiodism in, 193-194, 196-197 Biogeochemical cycles Braun-Blanquet floristic classification system,
soil and, 323 efficiency of forest ecosystems and, 113-115 11593}, IS, UBKe Th)7/
solar radiation intensity and, 189 forest management and, 124-134 Broods, size of, 381-382
temperature and. See Temperature, animals of forests, fire and, 344-345 Bud bank, 396-397
and geochemical cycles and, 73-74, 81-86 Budding, initiation of, temperature and, 214
tree line and, 361 human activity and, 120-124 Buffering
water availability to, 269-270 long-term declines in forest productivity chemical, of ecosystems, 122-123
wind effects on, 246-248 aul, 13355 1337) of soil pH, 294
wind transport of, 247-248 nitrogen cycling and, 110-113 Bulk density, ofsoil, 288-289
Animal communities, succession in, 510-511, nutrient cycling in tropical forest ecosystems Burrowing, water excess or deficit and, 275
alls and, 115-120 Buttress formation, in plants, 244
Animal habitats, coarse wood debris as, 62 over stand cycle and succession, 134-135,
Animal population, human population growth 136 C
and, 5, 7 of plant nutrition, 74-79 Calcium
Animation, suspended, water excess or deficit succession and, 505—507 deficiency of, 78
Biogeoclimatic classification, 159-165 plant visual symptoms accompanying, 314
and, 275
Anion exchange capacity (AEC), 294 hygrotypes in, 160-161 in soil, availability of, 296

I-1
I-2 INDEX

Capillarity, 290
Capillary forces, 290
Capillary fringe, 265
Capillary rise, 265
Carbon allocation, 60 Contagion, 367
Carbonation, of soil, 293 Continuum hypothesis of species distribution,
Carbon balance, tree line and, 359
Carbon dioxide
atmospheric, role of forests and forestry m re ma 380
controlling, 553-559 Cohost survivorship curves, 377, 380
concentrations of, shade tolerance and, Cold. See Freezing; Temperature Copper, deficiency of, plant visual symptoms
188-189 accompanying, 314
us cycle, 81-82 Crop, definition of, 42-43
veel
Shenk ga and,
OD 232-234 animal, succession in, 510-511, 512 Crown classes, of tree layers, 407
iologicaldiners and,
20-43 Crown fires, 330
Crown tattering, by wind, 244 2
global cates aes 432-436 Crumbs, in soil, 287
island y theory and forest Cryptophytes, 219
nutrition, 80 tation and, 436-437 Crystallization, latent heat of, 201
production ecology at level of, 54 measures of, 429-431 Cyclical succession, linear succession versus,
Carrying capacity, 375 spatial scales for evaluating, 431-432 495-498
lack of response to, 376 stability of communities and ecosystems Cyclic population fluctuations, 389-390
reduction by high population density. and, 437-438 Cyclonic precipitation, 258
376-377 competitive exclusion principle and,
Catabatic winds, 237 424-426 D
Cation exchange capacity (CEC), 293-294 degree to which communities dominate sites Daily degree hours, 214
Causal determinism, 18 and resist invasion and, 490 Darcy’s law, 264, 292
Ccellular— freezing point of, lowering of, ecological niche concept and, 427-429 Daubenmire approach to floristic classification,
forest management and, 438-439 155-158
Centralhuts floristic classification of plants, 407-411 Day-length indifferent (day-neutral)
system, 153, 155, 156, 157 community structure versus ecosystem organisms, 192
Chamaephytes, 219 function and, 411 Day/night solar variation, 190. See also
Character species, 155 snags and large decomposing logs as Photoperiodism
Chelation, 293 components of structure, 411 Deciduousness, nutrients and, 109
Chemical buffering, of ecosystems, 122-123 structure as result of disturbance and Decomposer organisms. See Saprotrophs
Chemoautotrophs, 36 successional development, 410-411 Decomposer (detritus) trophic chains, 37
Chernozems, 316, 318 structure as result of plant life form, Decomposition, of litterfall. See Detritus
Chinook winds, 239 407-410 entries; Litterfall
Chi-square, 42n self-replacing (climax), 406 Deductive reasoning, 388
Chromosomes, +43 species interaction in, 411-424 Defoliation, by herbivores, nutrient loss from
Cc yea a 241 antagonistic, 416-424 plants by, 95-96
Clay,287 symbiotic, 412-416 Deforestation, carbon release due to, 62
Clay-humus complex, 293 stability of, biological diversity and, 437-438 Degree days, 215
Clay loam, 287 Community ecology, 28 Demographers, 377
Clearcutting. See also Forest management Community gradients, 349. See also Forest Denitrification, 113
biogeochemical effects of, 124-131 ecotones; Species distribution bacteria and, 299
assart effect and, 127-128 Community models, spatially explicit, 585-587 gaseous losses of nitrogen due to, 131
gaseous losses of nitrogen, 131 Competition Density-dependent theory of population
leaching losses, 128-131 interspecific, 422-424 regulation, 385
losses in harvested materials, 124-127 species displacement by, 486-488 Density-independent theory of population
effects on energy in forest ecosystem, 66-69 Competition coefficients, 423 regulation, 385, 386
succession and, 511-313 Competitive exclusion principle, 423-426 Desertification, 236
temperature effects of, 230 Competitor adaptation, 472 Determinate tree species, 192-193
Climate Complete-tree harvesting, biogeochemical Determinism, principle of, 18-19
biotic potential and, 30 effects of, 124 Detritus feeders. See Saprotrophs
change in, carbon storage and, 345 Complex gradients, 349 Detritus trophic chains, 54-57
litter decomposition and, 102 Complexity, in ecosystem concept, 29 Detritus trophic web
rate of successional change and, 488 Comprehensive theory of population forest management effects on, 67-69
soil development and, 316 regulation, 385, 386-388 grazing trophic web versus, 39-40
temperature and, 208 Compression wood, 243-244 Detrivores. See Saprotrophs
Climatic classification, of forest ecosystems, Compromise theory of population regulation, Dew, 258
143-146 385, 386-388 Dew point, 256
Climatic climaxes, 469 Computer models, 571 Diagnostic species, 153, 155
Climatic diagrams, 145 application to site nutrient depletion Diapause, 222, 223
Climatic formations, 159 through harvesting of forest biomass, Differential species, 155
Climax 573-574, 576-577, 578 Diffusion transfer, 87
arguments for and against concept of, Computer models of succession, 475 Diploid number, 443
498-500 Computer simulation models, 569 Dipoles, 256
closed forest and, 498 development of, 571-572 Direct nutrient cycling pathway, 105
Climax communities, 406, 466 Conceptual models, 568-569 Disclimaxes, 468-469
Climax forests, 500-501 application to site nutrient depletion Disease control, in forest management,
Climax pattern hypothesis, 470-471 through harvesting of forest biomass, 402-403
Clines, 454 372-573 Disease organisms, shade tolerance and,
Clutches, size of, 381-382 Conditioning mechanisms, 387 187-188
 INDEX I-3

Disease transmission, in populations, 371-372 linear and cyclical, climax concept and, biological diversity and, 437-438
Dispersal, reproduction and, 368 495-501 water input into
Dispersion, 87-88, 367 management strategies with respect to, forest influences on, 259-260
Dissolved organic carbon (DOC), in 521-523, 524, 525 precipitation as, 256, 258-259
sedimentary cycles, 82 mechanisms of, 466 wind effects on, 248-249
Disturbance(s). See also Forest disturbance mesotrophic, 464 Ecosystem change, 31, 33-34
community structure as result of, 410-411 models of, 466 Ecosystem classification, 141-168
frequency and severity of, succession and, need for complexity in, 591-594 climatic, 143-146
490-491 oligotrophic, 464 ecosystematic, 158-166
gap, 528-530 rates of successional change and, 488-491 biogeoclimatic, 159-165
Disturbance patches, sizes of, frequency climate and soil condition and, 488 biophysical, 165-166
distribution of, 539-540 degree of change and, 489 physiographic, 146-147
Division of labor, membership in populations degree to which communities dominate vegetative, 147-158
and, 370-371 sites and resist invasion and, 490 dominance type as basis of, 150
Dominance, vegetative classification based on, frequency and severity of disturbance and, floristic composition as basis of, 150-158
150 490-491 physiognomic, 148-150
Dormancy, of plants, 219-220 longevity of organisms dominating sites Ecosystem concept, 28-29
Drizzle, 258 and, 489-490 components of, 28-29
Drought-resistant stages, of animals, water rate at which biota alter environment and, importance of, 29, 30-31
excess or deficit and, 276 489 Ecosystem ecology, 28
Drought stress, tolerance of, 274 recent theories and models of, 471-476 Ecosystem management (EM), 517, 533-542
Dry excretion, water excess or deficit and, 275 computer models, 475-476 fragmentation, connectivity, and edges and,
Duff mull, 298 Connell and Slatyer’s three-pathway 537-539
Dynamic survivorship curves, 377, 380 model, 473-475 key themes in, 534-536
hierarchy of successional causes, 475, +76 landscape ecology and, 536-537
E importance of life history characteristics tools in, 541
Earthworms, 308 in, 472-473 patch size and event size frequency
Ecoclines, 349 recognition of importance of individual distributions and, 539-540
Ecological actors, 429 species in, +72 sustainable, sustained yield management and
Ecological classification, 165-166 Tilman’s ideas, 473 timber mining versus, 601-602
Ecological convergence, 469 sere types and, 491-495 Ecosystem management models
Ecological diversity, 431. See also Biological primary hydrarch succession and for soil fertility assessment, 314-315
diversity hydrosere and, 494-495 stand-level, 582, 584-585
Ecological energetics, 502, 504-505 primary mesarch succession and mesosere Ecosystem models, spatially explicit, 585-587
Ecological groups, 351-352 and, 493-494 Ecotherms, 210
Ecological growth efficiency, 53 primary xerarch succession and xerosere Ecotypes, 441
Ecological models, linkage of harvest schedules and, 491-493 Ectendomycorrhizae, 302, 412
models to, 588 successional change mechanisms and, 476-488 Ectomycorrhizae, 302, 412
Ecological niche concept, 427-429 alteration of ecosystem physical Ectotrophic mycorrhizae, 88, 302
Ecological play, 429 characteristics, 478-485 Edaphic climaxes, 470
Ecological pyramids, 37-39 colonization, 477-478, 479 Edaphic grid, 147, 165
variations in, 39-41 displacement of species by antibiosis, Effective temperature range, 209
Ecological rotation, 509 autotoxicity, and competition, 485-488 Eluvial horizon, 316
as framework for emulation of natural forest terminology related to, 464-466 Eluviation, 316
disturbance, 523-524, 526-528 types of change and, 463-464 Emigration, population size and, 383-385
sustainability and, 602-605 “vital attributes” model of, 501-502, 503 Emulation of natural forest disturbance (ENFD)
design of sustainable forest management wind and, 248 definition of, 521
and, 605 Ecology ecological rotation concept as framework for,
ecological succession and, 603-604 development of, 25-26 523-524, 526-528
site nutrient capital and, 604-605 green religion and, 16-17 Endomycorrhiza, 412
Ecological stage, +29 production. See Production ecology Endotherms, 210
Ecological succession, 31, 33-34, 463-515 subdivisions of field, 27-28, 32 Endotrophic mycorrhyziae, 302
allogenic, 465-466 use and misuse of term, 26-27 Energy. See also Production ecology
in animal communities, 510-511, 512 Ecoregions, 157 flow of, fire and, 343-344
autogenic, 465 Ecosystem(s) sources for living organisms, 36-37
biogenic, 466 alteration of physical characteristics of, of water, 290
classical concepts and models of, 467-471 succession and, 478-485 Energy-balance approach, to climatic
climax pattern hypothesis, 470-471 as basic unit of ecology, 30 classification, 143, 145
contribution to present understanding of effects of fire on, 343-346 Energy flow, ecological pyramid of, 39
succession, 471 biogeochemical, 344-345 Energy-flow diagrams, definition of
monoclimax theory, 468-469, 470 carbon storage in forests and, 345-346 Energy imports/exports, ecological pyramid
polyclimax theory, 469-470 energy flow and, 343-344 and, 41
concepts of, 466-467 function of Environmental alteration, populations and, 371
ecological rotation and, 603-604 changes during succession, 502, 504-507 Environmental ethics, 608-611
ecosystem function changes during, 502, community structure versus, +11 Environmental filters, 397
504-507 genetic constitution of, 605-606 Environmental heterogeneity hypothesis of
biogeochemistry and, 505-507 health of, 547-551 biological diversity, 435-436
ecological energetics and, 502, 504-505 integrity of, 551-553 Environmental issues, 545-563
ecosystem stability, biodiversity, and definition of, 551-551 ecosystem integrity, 551-553
successional stage and, 507-510 of forest landscape, 551-552 forest and ecosystem health, 547-551
eutrophic, 464 seral stage integrity versus, 551 green religion and, 16-17
forest management and, 511-514 stability of role of forests and forestry in global carbon
clearcutting effects and, 511-513 biodiversity and successional stage and, cycles and climate change, 553-563
slashburning effects and, 513-514 507-510 control of atmospheric CO2, 553-559
 genetic variation and, 456-457
human population growth and, 8-9
water and, 281-282 al
Forest sections, 153
Forest site-type classification, 150-151, 152
renner gisicamny 363-364

Freezi
paar igs freezing point of
cellular fluids, 22
damage caused
of soil, = and, 206-207
Frontal precipitation, 258
Frost cracks, 215
Frostdamage, to plants,216

Function, in ecosystem concept, 28-29

Functional diversity, 430

sensssen
and STI-STS, SR SIF
Tea
iemmoonof S2H-521 Gamma dncrks, 431-432
<AMaOOR OE xv Emulator of namral Gap disturbance, 528-530
Terese disturbance (ENFD) Gap dynamics, 497
swalbscse (gap) SE 330 Gap models, 580-582, 583
~anadie commen aod, S19 Gaseous cycles, 81-82
~eaaoue-ad, SSS Gas exchange
“OeeX cuniogy, development
of,1721 between plants and atmosphere, wind effects
Steet costes, SR 50-563 on, 241
im soil, 289
Gause’s hypothesis, 423-426
Gene flow, 450
Gene pools, 450
Genetic constitution of ecosystem, 605-606
Genetic diversity
forestry and, 456-457
as measure of biological diversity, 442-448
in populations and, 370
Qw-gexster wee lines and, 34 3356 natural, sources of, 443-444
Salar wee linesand, 361-363 qualitative, +44-+46
Roremauiges, SEN
Sosse does, 87 2e
Tex panty of 76+
mamane, ATL Geochemical cycles, 81-86
mf os, 10E
“Oses ayggentinen, $37—F33, $32
“Qeex growin, lear area relared no, 57-60
Foeesc cme, 33>
“Oe Gamagement. See aée Clearcumng; a e distribution of organisms,
perature limitations on
seageucnemment effecss
of 174-134 saetatrendinal or altitudinal, nities
 INDEX [-5

polar and upper altitudinal, 226 defoliation by, nutrient loss from plants by, Internally represented models, 568-569
Geographical information systems (GISs), 541 95-96 Intraspecific competition
Geological cycles. See Geochemical cycles nutrition, 79-80 membership in populations and, 370
Geological mechanisms, in sedimentary cycles ) production ecology at level of, 51-54 in populations, 371
85-86 Heterotherms, 210, 222-223 Intrinsic theory of population regulation, 385,
Geometric growth, 373-374 Heterotrophs, 36-37 386
Germination in soil, 300-301 Iron
effect of fire on, 340 Hibernation, 222-223 deficiency of, 78
temperature and, 214 Hidden heat of crystallization, 201 plant visual symptoms accompanying, 314
Glacial till soil, 315 High-elevation tree lines. See Forest ecotones in soil, bacteria and, 299
Glacio-fluvial soil, 315 Higher-order consumers. See Carnivores Island biogeography theory, 436—437
Gleysols, 293 Historical bioassay models, 577
Gleyzation, 293 Historical range of variation (HRV), 519 J
Global warming, 232-234 Homeotherms, 210, 221 JABOWA model, 580
possible effects of, 214 physiological temperature adaptations of, Juglone, 420
Golger’s rule, 220-221 222
Governing mechanisms, 387 wind effects on, 247 Kk
Granules, in soil, 287 Human activities Krummholz tree growth, 243, 356
Graphs, 570 biogeochemical effects of, 120-124 k-strategists, 383
Grasses, 407 global temperatures and, 232-234
Gravitational potential, 290 selection pressures caused by, 448-450 1p
Gravitational water, 291 world radiation budget and, 232-234 Lacustrine soil, 315
Grazing trophic chains, 37 Humidity Lakes, acid rain effects on, 122-123
Grazing trophic web control by animals, water excess or deficit Laminar flow, 237-238
detritus trophic web versus, 39-40 and, 276 Landform, physiographic classifications and,
forest management effects on, 69 solar radiation intensity and, 171 146-147
Greenhouse gases, 202 Humus, raw, 298 LANDIS model, 587
carbon dioxide as, 60 Hybrid vigor, 454 Landscape ecology, 536-537
global temperature and, 232-234 Hydrarch succession, 464 tools in, 541
Green religion, 16-17 Hydration, of soil, 292 Landscape metrics, 541
Gregarious phase, in phase theory of locusts, Hydraulic lift, 259, 273 Landscape models, 541, 587-589
384 Hydric sites, 160, 161 Landslides, wind swaying of trees and, 248
Gross production, 43 Hydrogen bonds, in water, 256 Latent heat ofcrystallization, 201
Ground fires, 330, 331 Hydrological mechanisms, in sedimentary Latent heat of vaporization, 201
Ground vegetation, floristic classification cycles, 85-86 Laterites, 293
emphasizing, 150-151, 152 Hydrolysis, of soil, 292 Laterization, 293
Growth Hydrophobicity, 263-264 Layers, of plant communities, 407
of animals, insufficient summer warmth for, Hydrophobic soils, 322 Leaching
226 Hydrophytes, 272 from clearcutting, 128-131
new, allocation of, shading and, 186-187 Hydroseres, 464, 494-495 of soil, 296-297
of plants, temperature and, 211-216 Hygric sites, 160, 161 Leaf area index, 46
Growth efficiency, 53 Hygrotypes, classification of, 160-161 Leaves
Growth form, vegetative classification by, 148 Hypervolumes, n-dimensional, 428 age of, photosynthetic rate and, 180
Guano islands, 85 area of, forest growth and, 57-60
Guilds, 430 I morphology of, light intensity and, 187
Gullar flutter, 222 Immigration, population size and, 383-385 orientation of, morphology of, light intensity
Indeterminate tree species, 192-193 and, 187
H Individualistic hypothesis of species solar radiation intensity and, 184-186
Habitat-type approach to floristic distribution, 353 Lianas, as plant community layer, +07
classification, 155-158 Individualistic view of succession, 472 Lichenometry, 489
Hail, 258 Inductive reasoning, 388 Lichens, 413
Haloseres, 464 Industrial Revolution, 13 Life
Hand texturing, 288 Infiltration, of water into soil, 263-264 functional view of, 35-36
Haploid number, 443 Infrared radiation, 170-171 origins of, 35
Hartig net, 302 Ingested ratios, 77 Life form(s), of plants, 218
Harvesting Initial floristic composition, 472, 478 Life-form classification, of vegetation, 148
by clearcutting. See Clearcutting Innate capacity for increase, 374 Life history, importance to succession,
long-term productivity related to, 137 Insects 472-473
of northern forests, temperature and, 232 control of, in forest management, 402-403 Life history diversity, 430
Harvest schedules models, linkage to diapause in, 222, 223 Life tables, 377
ecological models, 588 pollination by, as mutualism, 414-415 Life zones, 228, 410
Heat. See a/so Temperature population regulation of, temperature in, Light. See also Solar radiation
conduction of, 205-206 2 characteristics of, 169-170
of fusion, 255 wind transport of, 247-248 definition of, 169
loss of, by animals in summer, 224-225 Instantaneous life tables, 377, 380 importance in forestry, 196-198
of vaporization, 255 Integument, of animals, water excess or deficit infrared, 170-171
latent, 201 and, 275 reaching ground, 171
Heat sums, 214, 215-216 Interaction, in ecosystem concept, 29 ultraviolet, 170-171
Heliophytes, 183 Interception loss, 260-261 ecological effects of, 174-175
Hemicryptophytes, 219 Interception storage capacity, 260 Light demanding plants, 186
Herb(s), as plant community layer, 407 Interdependence, in ecosystem concept, 29 Lignotubers, 339
Herbivores, 37 Internal cycles, 74 Linear succession, cyclical succession versus,
consumption by, energy loss due to, 47-49 succession and, 507 495-498
Lithoponics, 322

Nazality, population size and, 381-383, 384

Natural range of variation (NRV), 519 ~
Natural rate of increase, 374 -
Natural selection, 26, 442
Needle ice, 216-217

rai: a 432

572-577
of computer models, 573-574, 576-377,

of conceptual models, 572-573

of pictorial models, 573, $74, 575 gaseous losses of, denitrification causing, 131
of word models, 573 plant content of, 80
bioassay, historical, 577 in soil, availability of, 295
complexity of, 589-594 Nitrogen cycling, 110-113
Logistic growth, 374-381 degree of, 589-591 Nitrogen fixation, 110, 112
actual populations and, 375-381 in successional models, need for, 391-394 Nitrogen productivity, 57
Long-day organisms, 192 computer simulation, stages in development Nocturnal behavior, water excess or deficit
Low-elevation tree lines, 354, 356 of, 571-572 and, 275
Lungs, internal, water excess or deficit and, externally represented, 569-571 Nomadism, water excess or deficit and, 276
275 hybrid, 579-587 Nonsoil, 285
gap models, 580-582, 583 Nucleotides, +4
M spatially explicit individual tree, Numbers, ecological pyramid of, 38
Macronutrients, 75 community, and ecosystem models, Nutrients
MAESTRO, 579 585-587 acquisition of, 76
Magnesium stand-level ecosystem management competition for, shade tolerance and, 186
deficiency of, 78 models, 582, 384-585 cycling of, biogeochemical cycle and. See
plant visual symptoms accompanying, 314 internally represented (conceptual), 568-569 Biogeochemical cycles
in soil, availability of, 296 landscape, 587-589 depletion through harvesting of forest
Manganese, deficiency of, plant visual with dynamic links to FORECAST, biomass, models applied to, 372-577
symptoms accompanying, 314 developing, 588-589 of computer models, 373-574, 576-577,
Massenerhebung effect, 358 FEENIX, 588 578
Massive soil, 287 LANDIS, 587 of conceptual models, 572-573
Mass transfer, 87 linkage of harvest schedules models to of pictorial models, 573, 574, 575
Mass wasting, 316, 324-325 ecological models and, 588 of word models, 573
Material resources SELES, 587-588 essential, 75
as concentration phenomena, 599-600 Tardis, 588 loss of, fire and, 345
nonrenewable, 597, 598 limitations of, 566-567 oligotrophic condition and, 114
nonrenewable aspects of forest ecosystem process simulation, 577, 579 in plants
and, 605-607 reasons for using, 567-568 cycling of, 105-110 |
rate of formation as function of energy flux of succession, 466 distribution of, 91
through resource-forming system, classical, 467-471 losses of, 91-97
600-401 recent, 471-476 measurement of, 311-312
renewability of, 597-598 Moder forest floors forest floors, 298 plant uptake of nutrients and, 86-91
effect of use on, 601 Moisture. See Humidity; Water in soil
sustainability and ecological rotation and, Mollusks, in soil, 308 availability of, 295-296
602-605 Molybdenum, deficiency of, plant visual fire and, 337
timber mining versus sustained yield symptoms accompanying, 314 measurement of, 310-311
management versus sustainable Monoclimax theory of succession, 468-469, plant utilization of, 322-323
ecosystem management and, 601-602 470 water interaction of, 278-280, 281
Mathematical models, 569, 570-571 Mor forest floors, 298 Nutrient use efficiency, 57
Matric potenual, 290 Mortality, population size and, 383 Nutrition
Matric suction, 290 Mountain winds, 237 of animals, 79-81
Mature forest floors, 297-298 Mull forest floors, 298-299 of forests, 78-79
Mechanisms of succession, 466, 476488 duff, 298 Nutritional deficiencies, 76-78
alteration of ecosystem physical Multiple determinism, 18
characteristics, 478-485 Mutualism, 412-415 ie)
colonization, 477-478, 479 with continuous contact, 412-413 Obligate hibernation, 223
species displacement by antibiosis, without continuous contact, 414415 Oligotrophic condition, 114
autotoxicity, and competition, 485-488 Mycangium, 414 Oligotrophic succession, 464
Mesarch succession, 464 Mycorrhizae Omnivores, 37
Mesic sites, 160-161 ectotrophic, 88 Organic matter, of soil, 297-299
Mesophytes, 272 net productivity related to, 51 loss due to fire, 331-332
 INDEX I-7

Organism size, ecological pyramid and, 40-41 Picture models, 569, 570 Plant morphology, solar radiation intensity
Orographic precipitation, 258 Pioneers, 464 and, 183-186
Osmotic potential, 290 Pioneer species, +77 Plant potential, 292
Other-feeding organisms. See Heterotrophs Plant(s) Plant-site relationships, assessment of, fertility
Overstory, floristic classification and, 151, 153, adaptations of, to water excess or deficit, assessment by, 312-313
154-155 270-274 Plinthite, 293
Oxidation, of soil, 293 bending of, by wind, 242 Pneumatophores, 272
burial by wind deposits, adaptation to, 248 Poikilotherms, 210, 221
P buttress formation in, 244 Polar tree lines, 361-363
Palisade cells, 184-185 competition with trees for water, 280 Pollen, dissemination by wind, 241
Paramo, 228 dormancy of, 219-220 Pollution, wind transport of, 248
Parasitism, 418 effects on wind, 249-251 Polyclimax theory of succession, 469-470
Parent material of soil, 285 fire effects on. See Fire, effects on plants Polymorphic behavior hypothesis, 389
soil development in different types of, 320, growth form of, 407 Polyphagous predators, 393
321 growth of Population(s), 367-404
talus, 315-316 fertility assessment by, 312 advantages of membership in, 368-371
Particle size, in soil, 285-287 fire and, 340 age structure of, 377-381
Pathogens, tree line and, 361 inorganic chemistry of, 75 disadvantages of membership in, 371-372
Pedalfers, 316 interception of precipitation by, 260-261 origin of, 367-368
Pedocals, 316, 318 life forms of, 218 Population cycles, 389
Pedogenesis, 320. See also Soil, development of moisture balance of, wind effects on, 241 Population ecology, 27-28
Percent base saturation, 294 morphology of, temperature adaptations of, Population growth, 3-10, 372-381
Perched water table, 265 217-219 eruptions and, 377
Percolation, 263 Mycorrhizal relationship of. See Mycotrophy geometric (exponential), 373-374
Permafrost, 206 nutrients in. See Nutrients, in plants historical background of, 4-5, 6, 7
Permanent wilting percentage (PW%), 271 photoperiodism in, 192-193, 194-197, 197f implications for global resources, 5, 7-9
Permanent wilting point, 271, 292 physiognomy of, 407 logistic, 374-381
Permissive hibernation, 223 physiology of, temperature adaptations of, Malthus’s position on, 26
pH 219-220 Population size
energy flow increases in detritus food webs population ecology of, 394-402 cyclical fluctuations in, 389-390
after clearcutting and, 68 change in distribution of trees over time determinants of, 381-385
of forest floor, 102 and, 401-402 immigration and emigration, 383-385
of litter, 102 density-dependent regulation of mortality, 383
of soil, 294, 480-482 populations and, 398, 400-401 natality, 381-383, 384
fire and, 336-337 density-independent regulation of forest management and, 402-403
Phanerophytes, 219 populations and, 398 natural regulation of, 385-388
Phase theory oflocusts, 384 recruitment of new plants into population abiotic (density-independent) school of,
Phosphorus and, 395-397 386
deficiency of, 78 stand development phase as result of biotic (density-dependent) school of, 385
plant visual symptoms accompanying, 314 competition and self-thinning and, 401 comprehensive (compromise) school of,
in soil, availability of, 295 survival and, 397-398, 399, 400 386-388
Photoautotrophs, 36 radiation budgets and, 204 self-regulation (intrinsic) school of, 386
Photokinesis, 189 redistribution of water by, 261-263 of plants, 394-402
Photoperiod, 174 snow interception and redistribution by, 263 change in size distribution of trees over
Photoperiodism, 190-192 soil and, 320-323 time and, 401-402
in. animals, 193-194, 196-197 anchorage and, 320-321 density-dependent, 398, 400-401
definition of, 190 moisture supply and, 321-322 density-independent, 398
in plants, 192-193, 194-197, 197f nutrient supply and, 322-323 recruitment of new plants into, 395-397
Photosynthesis solar radiation intensity variations and, stand development phases as result of
apparent, definition of, 43 175-189 competition and self-thinning and, 401
intensity of solar radiation and, 176-183 morphology and, 183-186 survival and, 397-398, 399, 400
real, definition of, 43 photosynthesis and, 176-183 regulation of
temperature effects on, 180-181 shade tolerance and, 186-189 predation and, 390-394
Photosynthetic efficiency, 45-46 stem shape of, wind modification of, theories about, 385-388
Photosynthetic light saturation curve, 176 243-244 Porosity, of soil, 288
Phototaxis, 189 temperature effects on, 211-220 Potassium
Phototropism, 189 adaptations and, 217-220 deficiency of, plant visual symptoms
Physical environment growth response and, 211-216 accompanying, 314
modification by organisms, 368 injuries and, 216-217 in soil, availability of, 295-296
variations in, populations and, 368 temperature of, wind effects on, 242 Potential evapotranspiration (PE), 143, 145
Physical exploitation, 416-419 tree line and, 361 Potworms, in soil, 309-310
consumptive, 418-419 visual assessment of, fertility assessment by, Precipitation. See also specific forms of
nonconsumptive, 416-418 314 precipitation
Physical interference, in populations, 372 water availability to, 269-270 climatic classification based on, 143
Physical models, 569, 570 water balance of, 268-269 distribution by wind, effects on plants, 246
Physiognomic classification, of vegetation, wind effects on, 251 frontal (cyclonic), 258
148-150 water effects on distribution and production as input ofwater into ecosystem, 256,
Physiographic classifications, landform, 146-147 of, 276-278, 279 258-259
Physiological drought, 217 wind effects on. See Wind, effects on intensity of, water availability to plants and
Physiological ecology, 488 vegetation animals and, 270
Phytochrome, 190-191 wind training of, 243 interception by vegetation, 260-261
Pictorial models, application to nutrient Plant associations, 405 orographic, 258
depletion through harvesting of forest Plant formations, 410 Predation, 418-419
biomass, 573, 574, 575 Plant habitats, coarse wood debris as, 62—63 population regulation and, 390-394
I-8 INDEX

experimental investigation of, 390-393 ae

functional and numerical tor
and, 393-39, 395 i of, effect of fire on, 339-340
Predation hypothesis of biological diversity, ination of, SeeGermination
434-435 size of, shade tolerance and, 187
Predator pits, 419 easoning,
Predormancy, 219 Recruitment, of new plants into population,
395-397
Red belt damage, 207 pressures for, 48-450
energy
and nutrients at levelof,74-79 —— of sail, 293 a res for, 450, 452-454 A
ion ecology
atlevel of, 43-51 egeneration complexes, 495 rye en reer x
te landscape diversity 1-432 Sel ing organisms, utotrophs
Relay floristics, 472, 478 Selfregulation theory of population
Remote sensing, 172, 541 regulation, 383, 386
Renewability ofresources, 597-598 Selt-replacing communities, 406
effect of use on, 601 Self&chinning rule, 398, 400
nonrenewable aspects of forest eoasystem Seral species, 464
and, 605-607 Seral stage, 31, 33-34, 411, 464
biogeochemical cycles affecting, 134-135,
Process simulation model, 577, 579 136
Producer organisms. See Autotrophs of animals, insufficient summer warmth for, integrity of, ecosystem integrity versus, F31
Production, definition of, 43 226 length of, longevity of organisms dominating
Production ecology, 35-71 dispersal and, 368 sites and, 489490
carbon allocation and storage and, 60-64 membership in populations and, 370 Serclimaxes, 468
carnivores and, 54 nutrient lass from plants due to, 96 Seres, 31
at consumer trophic levels, 51-54 of plants, adaptations to fire and, 339-340 types of, 491495
ecological pyramids and, 37-41 Reserve acidity, of soil, 24 Shade leaves, T8S4+-186
energy benefit/cost relationships of forest Residence time, of organic matter, 56 Shade tolerance, 186-189
production and, 64, 65, 66 Residual soil, 315 Shade tolerant species, 178
energy-flow diagrams and, +1—42 Resource-ratio hypothesis of succession, 473 Shelterbelts, wind and, 250-
energy requirements of wood and other Resource tradeotts, 473 Shifting steady-state mosaic, 536
materials compared and, 64, 66 Respiration Short-day organisms, 191-192
energy sources for living organisms and, energy loss due to, 46-47 Shrubs, as plant community layer, 407
36-37 temperature effects an, 180-181 Silt loam, 287
forest management effects and, 66-69 Retranslocation, 109 Single-grained soils, 287
leaf area and forest growth and, 57-60 Rhizomes, 339 Site, 363-364
at primary producer level, 43-51 Rhizosphere, 204, 305 Site nutrient capital, ecological rotation and,
terminology used in, 42-43 Rime, 258-259 sustainability and, 604-605
trophic chains and, 37, 54-57 Ring shake, 253 Site series, 165
trophic webs and, 37 Roots, 304-306, Sve alse Mycorrhizae; Skylight, 171
Productivity Mycorrhizal relationships Slash-and-burn agriculture, 264
definition of, 43 competition among species and, 306 Slashburning
of forests, long-term declines in, 135, 137 plant nutrient uptake though, 87-88 biogeochemical effects of, 131-134
Profile diagrams, 148-149 soil utilization by, 320-321 succession and, 513-514
Protection, membership in populations and, Root system sorption gone, 310 Slash harvesting, 67
368-370 rselection, 383 Snags, as community structure component,
Provenances, 453 strategists, 383 4ul
Providential ecology, 26 Ruderal adaptation, 472 Snow, 258
Psammoseres, +64 distribution by wind, effects on plants, 246
Puddled soil, 289 S duration of, tree line and, 360-361
Punnett square, 445 Sate sites, 397, 477 plant interception and redistribution of, 263
Pyral climaxes, 470 Salt, carnage by wind, effects on plants, 42-243 Soil(s), 284-328
Salt licks, 323 biology of, 299-310, See ade Soil fauna; Soil
Sand, 287 flora
Qualitative genetic variation, 4+—-H6 carriage by wind, effects on plants, 242-243 chemical properties of, 202-297
Quanutative genetic variation, H6—H8 Sandy loam, 287 cation and anion exchange properties and,
Quasi-organisms, +06 Santa Anna winds, 23° 293-294
Saprotrophs, 37 leaching and, 296-297
R Saturation point (SP), 176 nutrient availability and, 295-206
Radiation balance, clearcutting and, 230 Scientific reductionism, 365 pH and, 204
Radiation budget, 202-205, 206t Sciophytes, 183 root distribution and, 306
albedo and, 204 Seasonal hibernation, 223 weathering and, 292-293
human effects on, 232-234 Seasonal variation condition of, rate of successional change and,
vegetation and, 204 in animals, 193-104 488
view factor and, 203-204 solar, 190 definition of, 284
Radiative index of dryness, 145 Sea spray, effects on plants, 243 development of, 315-320
Rain Secondary consumers, See Carnivores major factors determining, 315-320
acid, 122-124 Secondary production, definition of, 43 — in different types of parent material,
drizzle, 258 Secondary suceession, 464 320
heavy, 258 Sedimentary cycles, 82-86 wind and, 248
Rainfall, convectional, 256 biological mechanisms of, 84-85 woody debris decay in, 63
Rainwater, nutrient leaching from plants by, geological/hydrological mechanisms of, disturbance of, by forest management,
92-95 85-86 323-324
 INDEX [-9

ecological significance of, 320-323 mesofauna, 306, 308-310 water loss to, 269
for animals, 323 microfauna, 306, 310 Stress, in populations, 371
for plants, 320-323 Soil flora, 299-306 Stress tolerator adaptation, 472
fertility assessment of, 310-315 macroflora, 304-306 Structural diversity, 430
assessment by plant production, 312 microflora, 299-304 Structure, in ecosystem concept, 28
ecosystem management models for, Soil moisture regime (SMR) classes, 160-161 Structureless soil, 287
314-315 Soil reaction, 294 Structure management silviculture, 519
field fertilizer trials and assessment by Soil resources, 606-607 Subalpine zone, 356
visual symptoms and, 313-314 competition for, shade tolerance and, 186 Subassociations, 155
foliar analysis to measure, 311-312 Solar constant, 171 Subclimaxes, 468
nutrient level measurement and, 310-311 Solar radiation, 169-199 Succession. See Ecological succession
plant-site relationships and, 312-313 importance in forestry, 196-198 Successional convergence, 469
fertility of intensity variations in, 175-189 Successional pathways, 466
litter decomposition and, 102 effects on animals, 189 Successional regression, 466
root distribution in soil and, 306 humidity and, 171 Sulfur
fire effects on, 331-338 photosynthesis and, 176-183 deficiency of, plant visual symptoms
biological, 337-338 plant morphology and, 183-186 accompanying, 314
chemical, 336-337 shade tolerance of plants and, 186-189 in soil
physical, 331-336 physical nature of, 169-174 availability of, 296
frozen, radiation budget and, 206-207 spectral quality variations in, 174-175 bacteria and, 299
hydrophobic, 322 temporal variations in, 189-196 Sun leaves, 184
importance for forest management, 323-327 animals and, 193-194 Sun scald, 198, 217
physical property alterations and, 326-327 plants and, 190-193, 194-196, 197 Sunspots, change in, carbon storage and,
soil disturbance and, 323-324 Solifluction, 217 345-346
soil stability and, 324-325, 326f Solitary phase, in phase theory of locusts, 384 Superorganisms, 29
soil temperature alterations and, 325-326 Solonchak soil, 296 Supra-organisms, 406
moisture storage by, 269 Solonetz soil, 296 Surface fires, 330-331
nutrient acquisition from, 76 Solum, 285 Surface runoff, regulation by decaying logs, 63-64
nutrients in, fire and, 337 Solution, of soil, 292 Survival, plant population ecology and,
organic matter of, 297-299 Solvents, water as, 255 397-398, 399, 400
loss due to fire, 331-332 SORTIE model, 586 Survival temperature range, 209-210
pH of, 480-482 SPACE model, 586 Survivorship curves, 377
fire and, 336-337 Species Suspended animation, water excess or deficit
physical properties, root distribution in soil displacement of, by antibiosis, autotoxicity, and, 275
and, 306 and competition, 485-488 Sustainable ecosystem management, sustained
physical properties of, 285-290 diversity of, 429-430 yield management and timber mining
aeration, 289 evenness of, 430 versus, 601-602
alteration by forest management, 326-327 genetic diversity within, 429 Sustained yield management, timber mining
bulk density, 288-289 individual, recognition of importance of, +72 and sustainable ecosystem
consistence, 288 Species distribution, 349-364 management versus, 601—602
porosity, 288 forest ecotones and. See Forest ecotones Swidden agriculture, 264
structure, 287-288 possible patterns of, 349-351 Symbiosis, commensal, 415-416
temperature, 289-290 schools of thought concerning, 351-354, 355 Symbiotic bacteria, in soil, 301
texture, 285-287 site and, 363-364 Symbiotic interactions, between species in
plant uptake of nutrients from, 97-98 spatial, 367 communities, 412-416
porosity of, fire and, 332-333 Species interaction, in communities, +1 1-424 commensalist, 415-416
residual, 315 antagonistic, 416-424 mutualistic, 412-415
scarification of, 512 symbiotic, 412-416 Synecology, 28
single-grained, 287 Species richness, 430 Synusiae, of plants, 407
stability of, forest management and, Specific heat, 255
324-325, 326 Spiracle, 275 T
structure of, fire and, 332-333 Springtails, in soil, 309 ‘Talus parent materials, 315-316
temperature of, 208 Stability, of ecosystems, biodiversity and TARDIS model, 588
fire and, 333-336 successional stage related to, 507-510 ‘Taxonomic diversity, 429
root distribution in soil and, 306 Stable age distribution, 380 ‘Temperature, 201-235
transported, 315-316 Stand cycles, 495 animals and, 220-226
water in, 264-265, 290-292 Stand dynamics, 466 adaptations of, 221-226
fire and, 333, 334, 335 Standing crops, definition of, 42-43 air temperature and, 223-225
root distribution in soil and, 306 Stand initiation, 411 poikilotherms and homeotherms and, 221
water infiltration into, 263-264 Stand-level diversity, +31 climatic classification based on, 143
water percolation in, 263 Stand-level ecosystem management models, distribution of organisms related to, 226-228
wind erosion of, 248 582, 584-585 lower latitudinal or altitudinal limits of,
zonal, 160 Static life tables, 377, 380 226-228
Soil biota, 299. See also Soil fauna; Soil flora Stationary age distribution, 377, 380 polar and upper altitudinal limits of, 226
soil development and, 317-320 Statistical determinism, 18 energy flow increases in detritus food webs
Soil churning, 217 Stembreak, 244 after clearcutting and, 68
Soil classification, of forest ecosystems, 147 Stem exclusion, 411 excessively high, 210-211
Soil creep, 316 Stemflow, 261, 263 excessively low, 211
Soil fauna, 306-310 Stem girdle, 217 geographical and temporal variations in,
fire and, 337-338 Stenohaline organisms, 270 202-208
functions of, 54-55 Stochastic view of succession, 472 climate and microclimate and, 208
litter decomposition by, 100 Streams heat conduction and convection and,
macrofauna, 306, 308 log input to, 63 205-207
I-10 INDEX

radiation budget and, 202-205, 206t Trophotopes, 163 plant adaptations to excess or deficit of,
of soil temperatures, 208 Tropical forest ecosystems, nutrient cycling in, 270-274
topography and, 207 115-120 excess moisture and, 272
global, human effects on, 232-234 True dormancy, 219 moisture deficits and, 272-274
human effects on, 232-234 Turbulent flow, 237, 238-239 plant distribution and production and,
importance in forestry, 228-232 276-278, 279
photosynthesis and, 180-181 U ; in plants, wind effects on, 241
of plants, wind effects on, 242 Ultraviolet radiation, 170-171 redistribution by vegetation, 261-263
plants and, 211-220 ecological effects of, 174-175 - in soil, 264-265, 290-292
adaptations of, 217-220 Understory reinitiation, 411 fire and, 333, 334, 335
growth response of, 211-216 Understory vegetation, nutrient cycling and, plant utilization of, 321-322
temperature-related injuries and, 216-217 103-105 root distribution in soil and, 306
respiration and, 180-181 Unions, 155 as solvent, 255
of soil, 289-290 Utilization efficiency, 52 water balance of plants and, 268-269
fire and, 333-336 Water balance, of plants, ‘wind effects on, 251
forest management and, 325-326 v Water chemistry profile, 114-115
root distribution in soil and, 306 Valley winds, 237 Water cycle, 256-269
terminology related to, 209-210 Vaporization Water lifting, 290
tree line and, 358-359 heat of, 255 Water molecules, Dipoles, 255, 256
‘Temperature inversions, 207 latent heat of, 201 Water potential, 290
Temporal change, in ecosystem concept, 29 Vapor pressure deficit (VPD), 260 Water savers, 273
Temporal diversity, 430-431 Variable retention (VR) forestry, 319, 588-589 Water spenders, 273
Temporary wilting point, 271 Vegetation. See Plant(s); Tree(s) Water table, perched, 265
Termites, in soil, 308 Vegetation dynamics, 466 Water use efficiency (WUE), 57, 277-278
Tertiary consumers. See Carnivores Vegetation provinces, 157 Weathering, of soil, 292-293
Texture, of soil, 285-287 Vegetation regions, 157 Whole-tree harvesting, 67
Thallophytes, 407, +10 Vegetation zones, 157 biogeochemical affectsof, 124-127
as plant community layer, +07 Vegetative classification, 147-158 Wildlife management, 402
Thermoperiodism, 212-216 dominance type as basis of, 150 Wilting point
Therophytes, 219 floristic composition as basis of, 150-158 permanent, 271, 292 —
Three-pathway model of succession, of physiognomic, 148—150 temporary, 271
Connell and Slatyer, 473-475 Velvet, 224 Wind, 236-254
Throughfall, 261 Verbal models, 569 effects on animals, 246-248
Timberline, definition of, 356 Vernalization, 214 effects on ecosystems, 248-249
Timber management, +02 Vertebrates, in soil, 308 effects on vegetation, 239, 241-246
Timber mining, sustained yield management Vesicles, of mycorrhizae, 302 by distribution of precipitation, 246
and sustainable ecosystem management Vesicular-arbuscular mycorrhizae (VAM), 302 morphologic, 242-246
versus, 601-602 View factor, radiation budgets and, 203-204 physiological, 241-242
Time, soil development and, 320 Vital attributes model of succession, 501-502, reproductive, 241
Time hypothesis of biological diversity, 432, 434 503 extreme events related to, 248-249
Time lags, population growth and, 376 mountain, 237
Time scales, evolutionary, +56 Ww significance for forestry, 251-253
Tissue growth efficiency, 53 Warfare, impact on forests, 14 temporal and spatial variations in, 236-239,
Tools, development of, 12 Water, 255-283. See also Humidity 240
‘Topography animal adaptations to excess or deficit of, tree line and, 360
radiation budget and, 207 275-276 valley, 237
soil development and, 316 availability to plants and animals, 269-270 vegetation effects on, 249-251
Trachea, 275 chemical composition of, availability to Wind tatter, 244
Trade winds, 237 plants and animals and, 270 Windthrow, 244-246, 251-253
Transpiration, water loss to, 266-268 competition for, between trees and minor Wood, energy requirements of, comparison
Transpiration stream, 271 vegetation, 280 ‘with other materials, 64, 66
Transported soils, 315-316 energy flow increases in detritus food webs Wood supply, Industrial Revolution and, 13
Tree(s) after clearcutting and, 68 Word models, 569, 570
change in distribution over time, 401-402 from food, 276 application to site nutrient depletion
competition with minor vegetation for water, forestry and, 281-282 through harvesting of forest biomass,
infiltration into soil, 263 573
fire-resistant bark of, 338 inputs into ecosystems, 256, 258-260
krummholz growth of, 243 forest system influences on, 259-260 xX
as plant community layer, +07 interception and redistribution of snow and, Xerarch succession, 464
stand dev elopment phases and, 401 263 Xeric sites, 160, 161
wind swaying of, 248 interception of precipitation by vegetation Xerophytes, 272-274
Tree farming, 598 and, 260-261 Xeroseres, 464, 491-493
Tree lines (limit) loss of
definition of, 356-357 by drainage, 269 ¥
high-elevation. See Forest ecotones to evaporation, 265-266 Yarding, 324
low-elevation, 354, 356 to streams, 269 Yield, definition of, 42-43
polar, 361-363 to transpiration, 266-268
Tree models, spatially explicit, 385-587 metabolic, animal use of, 276 Z
TRIPLEX model, 584-585 nutrient acquisition from, 76 ZELIG model, 586
Trophic chains, 37 nutrient interaction with, 278-280, 281 Zine, deficiency of, plant visual symptoms
Trophic efficiency, 53 percolation of, 263 accompanying, 314
Trophic levels, 37 physical state of, availability to plants and Zonal soils, 160
Trophic webs, 37 animals and, 270 Zonation, 518-519
 .ed.gov (MI
suet

PEARSON
ice

Saddle River
Prent ™
Sada c 8
va45
all L PDer N]
in
L

ON
LUNA
(AN
.WIWW_ bren
a