National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 06/21/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home
Preface FIRE EFFECTS GUIDE
Objectives
Fire Behavior
Fuels Sponsored by:
Air Quality
Soils & Water
Plants
National Wildlife Coordinating Group
Wildlife
Cultural Res.
Grazing
Fire Use Working Team
Mgmt.
Evaluation Copies of the guide (NFES 2394) can be ordered form:
Data Analysis
Computer National Interagency Fire Center
Soft. Great Basin Area Cache
Glossary 3833 S. Development Ave.
Bibliography Boise ID 83702
Contributions
 National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 06/20/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home PREFACE
Preface
Objectives by Dr. Bob Clark and Melanie Miller
Fire Behavior
Fuels
A. Purpose
Air Quality
Soils & Water
Plants The Federal government manages a variety of ecosystems across the
Wildlife United States, including deserts, grasslands, tundra, shrublands,
Cultural Res. forestlands, estuaries, and riparian zones. These ecosystems range
Grazing from arid to humid, warm to cold, and sea level to over 10,000 feet
Mgmt. elevation. Fires naturally occur in almost all of these ecosystems, with
Evaluation fire characteristics determined by climate, vegetation, and terrain.
Data Analysis
Computer The purposes of this Guide are to summarize available information on
Soft. fire effects principles and processes, provide references for additional
Glossary information, and provide guidelines for the collection, analysis, and
Bibliography evaluation of wild and prescribed fire effects data. Basic mechanisms of
Contributions fire effects are described so that the reader will be able to understand
and interpret fire effects literature, and evaluate observed results that
conflict with those presented in published reports. The goal is to improve
fire management by improving our ability to manage fire effects.

The Guide was written as an aid for resource managers and fire
managers. It can be used for managing and evaluating wildfires;
developing and implementing emergency fire rehabilitation plans;
planning, monitoring, and evaluating prescribed fires; developing activity
plans such as timber management plans, allotment management plans,
and threatened and endangered species recovery plans; and providing
fire management input for land use plans.

B. Assumptions
Ecosystems have evolved with, and adapted to, specific fire regimes. In
a particular ecosystem, natural fires occurred with fairly specific, albeit
irregular, frequency and typical season of occurrence; with characteristic
fireline intensity and severity; and characteristically did or did not involve
the crowns of trees or shrubs. Gross differences occurred among
ecosystems. For example, frequent, low intensity, surface fires were
common in ponderosa pine ecosystems, whereas fires in big sagebrush
were probably less frequent, of higher intensity, and killed much of the
sagebrush overstory. High intensity, stand replacement fires at long
intervals were characteristic of some forest types, while annual fires may
have been common on some Great Plains grasslands. Despite this
variability in fire regimes, universal principles and processes govern
response of ecosystem components to fire. Recognition and
understanding of the principles and processes can help our
understanding of the variability in postfire effects that is often reported in
the literature, and differences between reported results and local
observations on burned areas. This knowledge will enable resource and
fire managers to predict and evaluate fire effects, regardless of
ecosystem or fire regime.

Fire effects are the result of an interaction between the heat regime
created by the fire and the properties of ecosystem components present
on the site. For example, plant species in vegetation types that have
evolved with frequent fire tend to be much more resistant to fire than
species from plant communities that rarely burned. The effects of a fire
burning under the same conditions may be very different on soils of
different textures or chemical properties. Variation in fire effects may
also occur within ecosystems because of differences in site
characteristics, fuel conditions, and weather prior to, during, and after
the fire. A fire may have different effects upon the same site if it occurs
in different seasons or within the same season but with different fuel,
duff, and soil moisture. For these reasons, it is important to document
conditions under which the fire occurred, and the characteristics of the
fire, as part of any effort to monitor postfire effects.

The words fire intensity, severity, fireline intensity, and burn severity are
often used interchangeably in the literature. The following terminology is
used throughout this Handbook to describe the properties of fire. All
definitions that describe the behavior of a flaming fire are those used in
the Fire Behavior Prediction System (Rothermel 1983), including fireline
intensity, the rate of heat release per linear foot of the flaming front. Burn
severity is a qualitative assessment of the heat pulse toward the ground,
and relates to subsurface heating, large fuel and duff consumption, and
consumption of litter and organic layers beneath isolated trees and
shrubs. The terms fire intensity and intensity are used by some authors
to describe the overall heat regime of a fire. They are generic terms that
are often confused with fireline intensity, and are not used as a synonym
for fireline intensity in this Handbook.

The Guide recognizes that a natural fire regime cannot be perpetuated

in unnatural communities. Timber harvest practices, grazing patterns
and degree of use, the accidental or deliberate introduction of exotic
plants and animals, other cultural activities that alter fuel continuity and
loading, and the modification of historic fire patterns through active
suppression have changed many plant communities. Interruption of fuel
continuity by livestock grazing, road construction, and other
developments has resulted in fires that are less frequent, smaller, and of
lower fireline intensity, in some ecosystems. The introduction of exotic
plants such as cheatgrass, coupled with anthropogenic ignition sources,
has greatly increased fire size and frequency in other ecosystems.

Active suppression has resulted in large areas, especially in shrublands

and forests, that are extremely susceptible to fire. The exclusion of fire
has resulted in a larger proportion of vegetation in older age classes,
except in regulated forests, which are more susceptible to insect and
disease infestations. The amount of dead plant material has increased,
either accumulated on the ground or retained on plants. In plant
communities with historically short fire cycles, the absence of fire has
allowed the development of fuel ladders between the surface and the
overstory. Fires which do occur are often carried into the tree crowns by
large accumulations of down dead woody fuels or understory trees,
causing a stand replacement fire in forest types where historically,
overstory trees were rarely killed. In vegetation communities with long
natural fire cycles, younger, intermixed, less flammable age classes of
vegetation are not as prevalent as they would have been under a natural
fire regime. Coupled with the increased incidence of insect and disease,
the continuity of highly flammable stands has increased, resulting in
greater potential for extremely large fires. Plans for the

C. Handbook Organization

The chapters of this Guide discuss different elements that relate to our
management of fire effects and specific responses of different
ecosystem components to fire. This Handbook recognizes that separate
discussions of fire effects on fuels, soils, watershed, plants, and wildlife
are artificial, because fire effects are an integration of the responses of
all of these components to fire. Despite the fact that fire effects occur
holistically, ecosystem components are discussed individually as a
means of organizing the information. Chapters describe basic principles
and processes that regulate fire effects, including fire behavior and
characteristics, fuels, air quality, soils and watershed, plants, wildlife,
and cultural values. Considerations for management of fire effects on
these resources, and a discussion of appropriate techniques for
monitoring fire effects, are contained in each of these chapters.
Monitoring is included in this Handbook because techniques that
accurately describe long-term trends in plant community condition, for
example, are not adequate to detect significant and sudden changes
caused by burning. Because an understanding of prefire and postfire
grazing management, data analysis, and documentation and evaluation
procedures is critical to sound management and monitoring of fire
effects, chapters on each of these topics are also included. Resource
management is goal oriented. The first chapter in this Guide is a
discussion of goals and objectives and how they fit into planning for the
use and management of fire.

This page was last modified 06/20/01

|Disclaimer| | Privacy| | Copyright| |USFWS Main Page| |Webmaster|

 National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 05/31/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home CHAPTER I - DEVELOPMENT OF OBJECTIVES

Preface
Objectives By Dr. Tom Zimmerman
Fire Behavior
Fuels
A. Introduction
Air Quality
Soils & Water
Plants Management is the process of anticipating the future, setting objectives,
Wildlife implementing an action, achieving an output, and performing an
Cultural Res. evaluation comparing the output to the objective. Management is not
Grazing possible without setting objectives. Clear and easily communicated
Mgmt. objectives facilitate the management process.
Evaluation
Data Analysis In land management programs, the desired outcome of management
Computer actions is expressed as management objectives. Objectives represent
Soft. an important component of all land management programs and are the
Glossary single most important factor driving all management actions.
Bibliography
Contributions B. Definitions and Qualities of Good Objectives

1. Goals and Objectives - Definition. In land management, both goals

and objectives are important. Goals are primary and basic products of
the long range management plans. These goals are commonly referred
to as land use decisions. Goals are relatively short statements that
discuss what the public lands are to be used for and where the uses will
occur. Each statement addresses a land use, but is not limited to the
principal or major use.

Objectives are a necessary component of the planning process; they

provide a bridge between goals and the implementation phase.
Objectives describe what procedures will be used and when actions will
be completed.
During the planning of fire management projects, objectives are
formulated and used as the basis for development of an action plan.
Interdisciplinary (I.D.) teams coordinate various concerns and develop
objectives for a project. The I.D. teams are composed of resource
specialists from different disciplines who address concerns of the
affected resources and resolve conflicts among resource disciplines that
arise from specific management actions.

2. Qualities of Good Objectives. Fire management objectives must be

made up of certain attributes or they will not convey the necessary
guidance. Good objectives must be informative and SMART. Objectives
that are SMART are:

S - Specific - what will be accomplished, using limiting factors, and

identifying the range of acceptable change from the present to the
proposed condition.

M - Measurable - the present and proposed condition must be

quantifiable and measurable.

A - Achievable - can be achieved within a designated time period.

R - Related/Relevant - related in all instances to the land use plan goals

and relevant to current fire management practices.

T - Trackable - objectives must be trackable over time and must include

a definite timeframe for achievement, monitoring, and evaluation.

3. Kinds of Objectives.

a. Land use decisions (goals). These are broad statements, usually

specified in land management plans, that deal with large areas over long
time periods (e.g., 10 years). Land use decisions establish resource
condition objectives; the allowable, limited, or excluded uses for an area
(land use allocations) and the terms and conditions for such use; and
management actions that will be taken to accomplish multiple use goals.

b. Resource management objectives. Resource management

objectives identify the changes in water, soil, air, or vegetation from the
present to proposed conditions. Resource objectives can also describe
an existing resource condition that should be maintained.
c. Treatment objectives. These are very well-defined statements that
describe what a treatment must accomplish in order to meet a stated
resource management objective. This type of objective is site-specific
and must utilize the SMART concept.

Any statement that is an objective must identify the change from

present conditions to the proposed conditions (the changes that are
planned) and the limiting factors.

C. How Objectives Relate to Project Inventory, Development,

Implementation, Monitoring, and Evaluation

Objectives are an important part of management actions and are

prerequisite to sound land and resource management. Objectives not
only drive the planning system, they also drive the full spectrum of
project implementation, monitoring, and evaluation.

During the fire planning process, for example, the planner uses resource
management objectives (standards) as guidance to determine what fire
management responses and activities are necessary. These standards
then provide guidance in determining what and how much information
should be collected prior to and during project implementation. At this
point, knowledge of fire effects becomes a necessary part of the
planning process. Fire effects information helps to determine what will
be done, how many resources are needed, how much funding the fire
program will need, and what should be evaluated to ensure efficient
accomplishment of the workload.

D. Relationships of Different Tiers (Levels) of Planning to

Objectives

Generally, objectives start as issues when the land use planning

process is initiated. (Issues are usually conflicts between two or more
resource uses or demands that must be resolved in the plan.) Issues are
generally defined in terms of the desired state of achievement for
environmental values and socioeconomic conditions affected by
management activities and resource decisions. The next step is
development of alternatives that include a range of ways to resolve the
issues. After the preferred alternative is selected, local guidance for
resource functions is developed that contains resource management
objectives.
Land use planning systems used by most federal agencies are divided
into five distinct tiers: national, geographically defined management
areas, individual resource functions, and strategic and tactical site
specific implementation (Table I-1). National policy is established in
public laws, federal regulations, Executive Orders, and other
Presidential, Secretarial, and Director approved documents. Policy
guidance for planning is developed, as needed, through interpretation of
national policy, public participation activities, and from coordination and
consultation with other federal agencies.

Table I-1: Relationship of Planning Tiers to Fire Management

Objectives, Products and Fire Effects Applications

Planning Tier Type of Product Fire Effects

Objectives Applications
National Policy and National policies
Regulations and guidance
regarding fire
presence and
exclusion in
wildland
ecosystems
Geographically Land use Resource Integration of fire
defined decisions Management and resource
management Plan (BLM, NPS) management
area Comprehensive within a
Conservation geographically
Plan (FWS) defined
Integrated management
Resource area
Management
Plan (BIA) Forest
Land
Management
Plan (FS)
 Habitat The role of fire in
Local guidance Management resource
Resource
for individual Plan management
management
resource Compartment within a specific
objectives
components Plan administrative
Allotment Plan unit.
Fire Identification of
Management appropriate
Activity Plans allocation of fire
Fire suppression, fire
Management use and fuels
Strategic site Strategic
Plan management
specific program
Fire activities
implementation objectives
Management necessary to
Action Plan achieve
Wilderness Fire resource
Management management
Plan objectives
Tactical site Treatment Prevention Plan Interpretation
specific objectives Presuppression and analysis of
implementation Plan site specific fire
Escaped Fire effects to guide
Situation development
Analysis and
Postfire implementation
Rehabilitation of a program of
Plan action to
Prescribed Fire accomplish
Plan treatment
Other objectives

Resource management plans developed for geographically defined

management areas establish the combinations of land and resource
uses; related levels of investment, production, and/or protection to be
maintained; and general management practices and constraints for
various public land resources. These are set forth as the terms,
conditions, and decisions that apply to management activities and
operations and are presented in the form of multiple-use prescriptions
and plan elements.

The third planning tier, developed at a local level, provide guidance for
individual resource functions. At this level the role of fire is discussed,
and how fire can be used or is detrimental in achieving the individual
resource objectives.

Site specific stratigic and tactical implementation plans are the final step
in the fire planning process. The primary role of these plans is to identify
operational guidance to accomplish site specific treatment objectives. To
continue with the example of the fire management component, Fire
Management Activity Plans delineate areas to receive different levels of
fire suppression, fire use, and fuels treatment. Resource management
objectives developed at this level are derived directly from land use
decisions. Prescribed fire plans refer to resource management
objectives developed in activity plans and identify treatment objectives.
Resource management objectives referenced in prescribed fire plans
describe second order fire effects, the indirect effects of fire treatment
that occur over the longer term, such as increased plant productivity,
changes in species composition, or increased off-site water yield. Fire
treatment objectives are developed from the resource management
objectives and state exactly what immediate effects the fire must create
in order to achieve the resource objectives. Fire treatment objectives
describe first order fire effects such as plant mortality, fuel reduction, or
duff consumption. An example of a fire treatment objective is: to remove
90 percent of existing sagebrush crown cover, using fireline intensities
that consume sagebrush crowns, leaving residual stems that are six
inches or less in height.

E. Summary

Land management programs are objective driven. Objectives must be

based on an amount of information sufficient to determine if a change
from the present condition to the proposed condition can be achieved.
Establishing objectives is a task of major importance and deserves an
allotment of sufficient attention and time. Both objectives and fire effects
information become more precise as site specificity increases.

This page was last modified 05/31/01

|Disclaimer| | Privacy| | Copyright| |USFWS Main Page| |Webmaster|

 National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 06/20/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home CHAPTER II - FIRE BEHAVIOR AND CHARACTERISTICS

Preface
Objectives by Melanie Miller
Fire Behavior
Fuels
A. Introduction
Air Quality
Soils & Water
Plants Frequently a fire is described as hot or cool, high or low intensity, or flaming or
Wildlife smoldering. Too often fire behavior and characteristics are not described at all.
Cultural Res. Standard terminology exists for describing the behavior, characteristics, and heat
Grazing Mgmt. regime of wildland fires. Monitoring and documentation of fire behavior and
Evaluation characteristics according to these standard terms can increase understanding of
Data Analysis the relationship between the effects created by a specific fire and that fire's heat
Computer regime, and make comparisons among different fires possible.
Soft.
Glossary The behavior of the flaming front of a surface fire can be predicted with fire
Bibliography behavior technology. Other characteristics of the surface fire, such as duration of
Contributions all phases of combustion and penetration of heat into duff and soil layers cannot
be predicted with existing models. The rate of heat release and growth of crown
fires can be estimated. It is important to understand some of the different
properties a wildland fire can have, how they can be described, which can be
predicted, and how the various aspects of a fire's heat regime can be related to
the fire treatment.

This chapter contains a brief overview of principles of fire behavior and

characteristics. More detailed information can be obtained from formal courses in
fire behavior and in courses that are prerequisites for certain prescribed fire
positions. The chemistry and phases of the combustion process are described in
the Air Quality chapter of this Guide, Chapter IV.B.1. The effect of fuel moisture
on fire behavior is described in this chapter, but factors affecting fuel moisture
content and fuel consumption are discussed in Chapter III.B, this Guide.

B. Principles and Processes

1. The Fire Environment. Wildland fire is influenced by three interacting classes

of variables: fuels, topography, and air mass (Countryman 1972).

a. Fuels. Wildland fuels provide the energy source for fire. Fuels consist of both
living and dead vegetation, the latter in various stages of decay. Fuels occur in
three fairly distinct strata: ground, surface, and aerial. A fire can burn in one, two,
or all three strata at once, or change the layer in which it is burning as fuels and
environmental conditions change throughout an area. Fuels are discussed in
greater detail in Chapter III. B.1., this Guide.

(1) Ground fuels. Ground fuels are all combustible materials below the surface
litter layer. These fuels may be partially decomposed, such as forest soil organic
layers (duff), dead moss and lichen layers, punky wood, and deep organic layers
(peat), or may be living plant material, such as tree and shrub roots.

(2) Surface fuels. Surface fuels are those on the surface of the ground,
consisting of leaf and needle litter, dead branch material, downed logs, bark, tree
cones, and low stature living plants.

(3) Aerial fuels. Aerial fuels are the strata that is above the surface fuels and
include all parts of tree and tall shrub crowns. The aerial fuel layer consists of
needles, leaves, twigs, branches, stems, and bark, and living and dead plants that
occur in the crowns such as vines, moss, and lichens.

(4) Ladder fuels. Ladder fuels bridge the gap between surface and aerial fuels.
Fuels such as tall conifer reproduction can carry a fire from the surface fuel layer
into tree crowns.

b. Topography. Topography includes slope, aspect, elevation, and how these

elements are configured. Topography can change suddenly, particularly in
mountainous terrain, and its influence on fire behavior can rapidly change as well.

(1) Direct effect.

(a) Slope is an extremely important factor in fire behavior because the flames of a
fire burning upslope are positioned closer to the fuels ahead of the fire. This dries
and preheats the fuels at a greater rate than if they were on flat terrain.

(b) Topography channels wind and can create turbulence and eddies that affect
fire behavior. Topography also affects diurnal air movement, influencing the
velocity of day time upslope and night time downslope winds.

(2) Indirect effect.

(a) The combined effects of aspect and elevation create different microclimates
that affect vegetation distribution and hence fuel type.

(b) Fuel moisture can vary with aspect, elevation, and vegetation type. This is
discussed further in Chapter III.B.4, this Handbook.
c. Air mass. Weather components such as temperature, relative humidity,
windspeed and direction, cloud cover, precipitation amount and duration, and
atmospheric stability are all elements of the air mass. These values can change
quickly over time, and significantly with differences in aspect and elevation. The
air mass affects fire both by regulating the moisture content of fuel (discussed in
Chapter III.B.4.), and by its direct effect on the rate of combustion. The following
is a brief discussion of the effect of air mass factors on fire behavior and
characteristics.

(1) Temperature. Atmospheric temperature affects fuel temperature. The ease of

ignition, the amount of heating required to raise fuel to ignition temperature (320
C.; 608 F.) (Burgan and Rothermel 1984), depends on initial fuel temperature.
The most important effect of temperature, however, is its effect on relative
humidity and hence on dead fuel moisture content. (See Chapter III.B.4.).

(2) Windspeed. Wind has a significant effect on fire spread. It provides oxygen to
the fuel and, combined with slope, determines which way the fire moves. Wind
tips the flame forward and causes direct flame contact with fuel ahead of the fire
(Burgan and Rothermel 1984). These fuels are preheated and dried by this
increased transfer of radiant and convective heat. Windspeed has the most
influence on fire behavior in fuel types with a lot of fine fuels, such as grasslands.

2. Combustion Process.

a. Two stage process. Within a wildland fire, the processes of pyrolysis and
combustion occur simultaneously (Ryan and McMahon 1976 in Sandberg et al.
1978).

(1) Pyrolysis. When first heated, fuels produce water vapor and mostly
noncombustible gases (Countryman 1976). Further heating initiates pyrolysis, the
process by which heat causes chemical decomposition of fuel materials, yielding
organic vapors and charcoal (ibid.). At about 400F. (204 C)., significant amounts
of combustible gases are generated. Also at this temperature, chemical reactions
start to produce heat, causing pyrolysis to be self-sustaining if heat loss from the
fuel is small. Peak production of combustible products occurs at when the fuels
are about 600 F. (316 C.) (ibid.).

(2) Combustion. Combustion is the process during which combustible gases and
charcoal combine with oxygen and release energy that was stored in the fuel
(Countryman 1976) as heat and light.

b. Phases of combustion. The following summary is derived from Ryan and

McMahon (1976 in Sandberg et al. 1978), except where noted. For a more
complete discussion of the phases of combustion, see Sandberg et al. (1978).

(1) Pre-ignition phase. In this phase, heat from an ignition source or the flaming
front heats adjacent fuel elements. Water evaporates from fuels and the process
of pyrolysis occurs, the heat-induced decomposition of organic compounds in
fuels.

(2) Flaming phase. Combustible gases and vapors resulting from pyrolysis rise
above the fuels and mix with oxygen. Flaming occurs if they are heated to the
ignition point of 800 to 900F. (427 to 482 C.), or if they come into contact with
something hot enough to ignite them, such as flames from the fire front
(Countryman 1976). The heat from the flaming reaction accelerates the rate of
pyrolysis. This causes the release of greater quantities of combustible gases,
which also oxidize, causing increased amounts of flaming (Ryan and McMahon
1976 in Sandberg et al. 1978).

(3) Glowing phase. When a fire reaches the glowing phase, most of the volatile
gases have been driven off. Oxygen comes into direct contact with the surface of
the charred fuel. As the fuel oxidizes, it burns with a characteristic glow. This
process continues until the temperature drops so low that combustion can no
longer occur, or until all combustible materials are gone.

(4) Smoldering phase. Smoldering is a very smoky process occurring after the
active flaming front has passed. Combustible gases are still being released by the
process of pyrolysis, but the rate of release and the temperatures maintained are
not high enough to maintain flaming combustion. Smoldering generally occurs in
fuel beds with fine packed fuels and limited oxygen flow such as duff and punky
wood. An ash layer on these fuel beds and on woody fuels can promote
smoldering by separating the reaction zone from atmospheric oxygen (Hartford
1993).

3. Fire Behavior Prediction. The Fire Behavior Prediction System is a collection

of mathematical models that were primarily developed to predict the behavior of
wildland fires (Rothermel 1983). The models include those used to forecast
behavior, area and perimeter growth of a surface fire; models that estimate spot
fire potential, crowning potential, and crown fire behavior; and fire effects models
that predict tree crown scorch height and tree mortality.

Solutions for most of these models can be obtained from nomograms (Albini
1976) and the BEHAVE system. BEHAVE is a set of programs for use on
personal computers (Andrews 1986; Andrews and Chase 1989; Burgan and
Rothermel 1984). More information about the BEHAVE system is contained in
Chapter XII.C.1, this Guide.

4. Fire Spread Model. A fire spread model was developed by Rothermel in 1972
that allows managers trained in the use of the model to make quantitative
estimates of fire behavior. The model is a mathematical representation of fire
behavior in uniform wildland fuels. The fire spread model describes the processes
that control the combustion rate: moisture evaporation, heat transfer into the fuel,
and combustible gas evolution (Rothermel 1972).

a. Assumptions. Basic assumptions of the fire spread model are (Rothermel

1983):

(1) The fire is burning in a steady state in homogeneous surface fuels, not in
crown or ground fuels.

(2) The percent slope and aspect are uniform.

(3) The wind is constant in both velocity and direction.

(4) The model describes fire behavior within the flaming front. The model does not
describe behavior after the fire front has passed, such as during fuel burnout.

(5) The behavior of the fire is no longer influenced by the source of ignition or by
suppression activities.

These assumptions are often violated when prescribed burning because ignition
is often used to manipulate the fire. A common objective for burning is to
consume fine fuels before the fire reaches a steady state. The predicted values
do provide an estimate of fire behavior if a prescribed fire escapes.

b. Inputs to the Fire Spread Model. Required inputs to the fire spread model
include fuel model, fuel moisture content, slope, and wind.

(1) Fuel model. A fuel model is a mathematical representation of the amount and
kind of fuels present. The Fire Behavior Prediction System provides 13 standard
fuel models that describe the characteristics of the portions of the fuel complex
carrying the fire. Custom fuel models more closely describing a specific fuel
situation also can be developed using the BEHAVE program (Burgan and
Rothermel 1984). (See XII. C.1.a., this Guide.)

(a) Categories. In most situations, the flaming front of a fire advances through
fine fuels such as grass, shrub foliage, litter, and small diameter down dead
woody fuels. Wildland fuels can be grouped into four categories, according to the
nature of the carrier fuels.

i. Grass or grass dominated: the primary carrier of the fire is grass.

ii. Shrub dominated: the primary carrier of the fire is either shrubs or litter beneath
shrubs.

iii. Timber litter dominated: the primary carrier of the fire is litter beneath a timber
(tree) stand.
iv. Logging slash: the primary carrier of the fire is residual material left from
logging operations.

(b) Fuel properties. Fuel particles within a fuel complex have physical properties
that influence the way they burn. The 13 standard fire behavior fuel models have
specified physical properties (Anderson 1982). Properties can be changed to
create a custom fuel model that may better describe a particular fuel complex.
(See XII.C.1.a., this Guide.) Fuel properties that are the most important for
determining the way a fire will behave include the following.

i. Fuel loading. The amount of live and dead fuel is expressed in weight per unit
area. Loadings are grouped by particle size class and are usually expressed in
tons per acre (kilograms per square meter). Total fuel is all plant material both
living and dead present on a site. Available fuel is the amount of fuel that will burn
under a specific set of fire conditions.

ii. Fuel size class. Dead fuels are divided into size classes based on diameter:
less than 1/4-inch, 1/4 to 1-inch, 1 to 3 inches, and greater than 3 inches. (Metric
equivalents of these size classes are: less than 0.6 centimeters, 0.6 to 2.5
centimeters, 2.5 to 7.6 centimeters, and greater than 7.6 centimeters.) Fuel size
class is related to the rate at which particles wet and dry. This is discussed further
in Chapter III.B.4., this Guide.

iii. Size class distribution. Fires usually start and spread in fine fuels, that is, those
less than 1/4 inch in diameter. These fuels ignite increasingly larger size classes
of fuels. If fine fuels or an intermediate size class are missing, a fire may not ignite
or may not spread.

iv. Surface area to volume ratio. The surface area to volume ratio is a function of
the particle size: the more finely divided the fuel material, the larger the ratio.
Because small fuel particles have a large surface area compared to their volume,
they dry out and ignite more rapidly than larger particles. Therefore, fine fuels
usually have the most influence on fire behavior.

v. Fuel bed depth. Fuel bed depth is the depth of the surface fuel layer, i.e., the
average height of surface fuels contained in the combustion zone of a spreading
fire front.

vi. Packing ratio. The packing ratio is a measure of the compactness of the fuel
bed. Expressed as a percentage, the packing ratio is the percentage of the fuel
bed that is composed of fuel, the remainder being air space between the
individual fuel particles (Burgan and Rothermel 1984). A fuel bed with no fuel has
a packing ratio of zero, while a solid block of wood has a packing ratio of one
(ibid.). A very open or porous fuel bed burns slowly because individual fuel
particles are located so far apart that little heat is transferred among particles. A
very compact fuel bed also burns slowly because airflow among the fuel particles
is impeded, and there are large numbers of fuel particles that must be heated to
ignition temperature. For every size of fuel particle, there is an optimum packing
ratio at which heat transfer and oxygen produce the most efficient combustion
(Burgan and Rothermel 1984). Compactness also influences the drying rate of
fuel.

vii. Bulk density. Bulk density is the actual fuel weight per unit. It is calculated by
dividing the weight per unit area by the fuel bed depth. It is a measure of the oven
dry weight of fuel per cubic foot of the fuel bed, usually expressed as pounds per
cubic foot. The higher the bulk density of the fuel, the slower the spread rate,
because more fuel must be preheated to ignition temperature in order for the fire
to spread.
viii. Fuel continuity. Fuel continuity is a description of the distribution of fuels. Fire
spread is most likely in continuously distributed fuels. The greater the fuel
discontinuity, the higher the fireline intensity required for fire spread. Fuel
continuity is described in terms of both horizontal and vertical continuity.
Horizontal continuity relates to the horizontal distances between fuel particles and
relates to percent cover. The proximity of tree or shrub crowns affects the ease
with which fire can spread in a live fuel strata. Vertical continuity describes the
proximity of surface fuels to aerial fuels and affects the likelihood that a fire can
move into the vegetative canopy.

ix. Heat content. The most important aspect of fuel chemistry influencing fire
behavior is heat content. This value expresses the net amount of heat that would
be given off if the material burns completely (or at 100 percent efficiency), rated
as Btu per pound of fuel. The heat content for all species of dead woody fuel is
essentially the same (Albini 1976). The presence of pitch in wood, and of volatile
compounds such as oils and waxes in some live fuels, increases heat content,
and thus flammability.

x. Live fuels. Some fuel types contain a significant component of live fuels in the
surface fuel layer, including shrubs, grasses, and forbs. The importance of live
fuels to fire behavior can change throughout the year. Their volume can increase
significantly during greenup and the early part of the growing season. They can
lose their foliage at the end of the growing season or during a drought. Seasonal
fluctuations in moisture content occur that significantly affect flammability. The
moisture cycles within live fuels are discussed in more detail in Chapter III.B.5.,
this Guide.

While technically live fuels, mosses and lichens do not have physiologically
controlled seasonal moisture cycles. Their moisture content is very sensitive to
changes in temperature and relative humidity and can become as low as that of
surface litter layers. A dry surface layer of mosses and lichens can readily carry a
fire in black spruce forests in Alaska (Dyrness and Norum 1983).

The volatile compounds in some species of live fuels allow them to burn at a
higher moisture content than if there are few or no volatiles (Norum 1992).
Sagebrush (Artemisia spp.) is considered to be a moderately volatile fuel, while
chaparral shrubs, conifers, and dead juniper are highly volatile fuels (Wright and
Bailey 1982).

Fire behavior in stands of shrubs containing volatile compounds can be extreme.

This is attributed not only to their chemical content, but also to the high
percentage of dead material that some of these stands of shrubs contain, and the
ideal mixture of fuel to air within the shrub canopy (Burgan 1993).

(2) Fuel moisture. Fuel moisture content describes how wet or dry the fuels are.
Moisture content is the single most important factor that determines how much of
the total fuel is available for burning, and ultimately, how much is consumed. Fuel
moisture determines if certain fuels will burn, how quickly and completely they will
burn, and what phases of combustion the fuels will support. Fuels with a higher
moisture content reduce the rate of energy release of a fire because moisture
absorbs heat released during combustion, making less heat available to preheat
fuel particles to ignition temperature (Burgan and Rothermel 1984). Ignition will
not occur if the heat required to evaporate the moisture in the fuels is more than
the amount available in the firebrand (Simard 1968). Environmental factors
regulating dead fuel moisture content, and the relationship between fuel moisture
content and fuel consumption, are discussed in III.B.4., this Guide.

(a) Fuel moisture formula. Fuel moisture content is the percent of the fuel
weight represented by water, based on the dry weight of the fuel. In a word
equation, it is:

Percent Moisture Content = Weight of Water / Oven-dry Weight

of Fuel x 100

Moisture content can be greater than 100 percent because the water in a fuel
particle may weigh considerably more than the dry fuel itself. For example, a
green leaf may contain three times as much water as there is dry material,
leading to a moisture content of 300 percent. Moisture content of duff and organic
soil can be over 100 percent. Methods to measure and calculate fuel moisture
content are described in Chapter III.D., this Guide.

(b) Moisture of extinction. The extinction moisture content is the level of fuel
moisture at which a fire will not spread. It is a function of the fuel type and fuel
bed geometry (Byram et al. 1966 in Albini 1976). The moisture of extinction is
much lower for light, airy fuels such as fine grass, about 12 to 15 percent
(Sneeuwjagt 1974 in Albini 1976), than it is for dense fuel beds such as pine
needles, in which it has been measured at 25 to 30 percent (Rothermel and
Anderson 1966 in Albini 1976). Under favorable burning conditions, the moisture
of extinction has little effect on fire behavior, but when "conditions for burning are
poor, it can cause significant changes in predicted fire behavior" (Rothermel
1983).

(3) Slope. The steepness of slope is measured as the rise of the ground in feet
for every horizontal foot traversed, commonly referred to as "rise over run."

Percent Slope = Rise / Run x 100

Percent slope can be measured directly with instruments or calculated from

topographic maps.

(4) Wind. Both windspeed and direction are used as inputs to the Fire Behavior
Prediction System.

(a) Midflame windspeed. The speed of the wind is measured at the midpoint of
the height of the flames because this best represents the wind that blows directly
on the fire. Most weather forecasts, and most weather measurement stations,
give the windspeed at 20 feet (6 meters) above the ground or above local
obstructions. For fire behavior calculations, the 20-foot windspeed is reduced to
the speed occurring at the midflame height. This compensates for the friction
effect of vegetation and land surface that slows the speed of the wind. The
adjustment factor varies with vegetation type, amount of canopy closure, and
position on slope.

20 Foot Windspeed x Wind Adjustment Factor = Midflame

Windspeed

(b) Effective windspeed. As an intermediate step in obtaining solutions to the

fire spread model, effective windspeed is determined. This value integrates the
additive effects of slope steepness with a wind that is moving across or up a
slope.

c. Outputs of the Fire Spread Model. The accuracy of predictions depends on

how representative the fuel model chosen is of the fuels on the site, how
accurately inputs are measured or estimated, and to what degree the situation
meets the spread model assumptions. For predictions to be within a factor of two
of actual fire behavior (from one-half to two times) is considered to be an
acceptably accurate estimate (Norum 1993). The model is flexible enough that an
experienced practitioner can make fairly good projections of fire behavior by
carefully estimating or measuring the input values and tempering the results with
judgment. Personal experience in a particular fuel type is necessary for refining
output from this model.

(1) Forward rate of spread. One of the most important measures of fire behavior
is the speed at which the fire moves across the landscape. The spread model
calculates the rate of spread at the head of fire when the fire reaches its full,
steady state speed. It predicts the speed of a fire burning in surface fuels,
spreading on a single, unified front, that is not influenced by other ignitions. Rate
of spread is generally stated in chains per hour, feet per minute, or meters per
minute.

(2) Flame length. A second spread model output is the length of the flames when
the fire has reached its full, forward rate of spread. Flame length is the distance
along the slant of the flame from the midpoint of its base to its tip Flame height is
the perpendicular distance from the ground to the flame tip and is not predicted by
the fire spread model.

(3) Fireline intensity. Fireline intensity describes the nature of a fire in terms of
its rate of energy release. Fireline intensity is the amount of heat given off by a
fire along each foot of the leading edge of the fire each second, usually expressed
as Btu per lineal foot of fireline per second.
(4) Heat per unit area. Another measure of the energy released from a fire is
heat per unit area. It is the total amount of heat released in each square foot of
the flaming fire front, usually expressed as Btu per square foot. All of the heat
given off in the flaming front is included in this value, regardless of the length of
time that the flaming front persists. For a given area with a specific amount and
distribution of fuel, heat per unit area is inversely related to fuel moisture content.
Heat released in flaming combustion that occurs as fuels burn out after the
flaming front has passed is not included in the heat per unit area value.

(5) Reaction intensity. Reaction intensity is a rate of heat release per unit area
of flaming fuels, usually expressed in Btu per square foot per minute. This is the
amount of energy released each minute by a square foot of flaming front,
compared to heat per unit area which measures the total amount of energy given
off per square foot. For a given fuel complex, reaction intensity can vary
significantly with differences in moisture content.

d. Other predictable aspects of fire behavior.

(1) Probability of ignition. The probability of ignition, expressed as a

percentage, is an estimate of the probability that a spark or firebrand landing on
representative fuels will start a fire (Rothermel 1983). It is based on the amount of
heat required to bring fine fuel to ignition temperature. Model inputs are fine fuel
moisture, ambient air temperature, and the amount of shade.

(2) Maximum spotting distance. For many fuels situations, it is possible to make
reasonably accurate estimates of the maximum distance to which fire may spot
ahead by airborne embers (Rothermel 1983). The inputs required include the
source of the embers, i.e. whether it is burning piles or trees; the species of tree,
and their size and shape; the topography at and downwind from the fire; and the
20 foot windspeed. The model calculates the farthest distance a live ember is
likely to be carried. It does not estimate how many burning embers will be lofted,
or if the ember will ignite a spot fire. However, a combination of maximum spotting
distance with the probability of ignition provides a workable idea of how far a fire
may spot and the probability that it will cause a new fire.

(3) Crown fires.

(a) Classes. Van Wagner (1977) grouped crown fires into three classes based
upon their dependence on the behavior of the surface fire.

i. Passive crown fires are those in which trees torch as individuals, ignited by the
surface fire. These fires spread at essentially the same rate as surface fires.
Trees torch within a few seconds with the entire crown enveloped in flames from
its base to the top.
ii. Active crown fires are those in which a solid flame develops in the crowns. The
surface and crown fires advance as a single unit dependent upon each other.

iii. Independent crown fires advance in the crowns alone, independently of the
behavior of the surface fire.

(b) Crowning potential. The conditions necessary to cause the ignition of the
crowns of trees or tall shrubs can be estimated. A probability of crown fires can be
calculated, given the foliar moisture content and the height of the lowest part of
the crowns. From these, an estimate can be derived of the fireline intensity
needed to ignite the crowns (Rothermel 1983).

(c) Wind-driven vs. plume-dominated crown fires. The following discussion is

taken from Rothermel (1991).

i. Wind-driven crown fire. A running crown fire can develop when winds blow
flames from torching trees into adjacent tree crowns, or slope effectively
accomplishes the same thing. Strong winds are the major force pushing the fire,
and its spread rate can be greatly accelerated by slope. A strong convection
column rapidly develops that is tipped over by the wind.

ii. Plume-dominated crown fire. A plume-dominated crown fire behaves quite

differently from one driven by wind. Plume-dominated crown fires occur when
windspeeds are fairly low. A strong convection column develops that rises above
the fire, rather than leaning over before the wind. Air movement within the
convection column generates the winds that cause significant rates of crown fire
spread.

(d) Predicting size and intensity of crown fires. Rothermel (1991) presents
methods for estimating and displaying the important elements of the behavior of a
wind-driven crown fire. The model is applicable to coniferous forests of the
northern Rocky Mountains, or forests with similar structure and fuels. Using these
methods, an experienced fire behavior analyst can predict the rate of spread of a
wind-driven crown fire, the length of flames, the time period when a particular
crown fire will run, the probable area and perimeter of the crown fire, and the
maximum rate of crown fire spread.

A method for calculating and comparing the power of a fire with the power of the
wind is provided. The power of the fire is the heat energy released by combustion
that drives the convection column, expressed as foot pounds per second per
square foot. If the power generated by the fire is close to or exceeds that of the
wind, a plume-dominated crown fire may develop. The onset of a plume-
dominated fire may cause a sudden acceleration of the fire and faster spread
rates than predicted. The model can thus predict the potential for onset of a
plume-dominated fire, but not its behavior.
5. Relationships between Fire Behavior and Fire Effects. Few fire effects are
known to be directly related to the behavior of the surface fire, that is to its spread
rate, flame length, or rate of heat release. The following effects can be estimated
from outputs of the Fire Behavior Prediction System.

a. Crown scorch height. There is a direct relationship between crown scorch

height and flame length, ambient air temperature, and wind. All of these variables
are used to measure the height above the surface of the ground that lethal
temperatures occur. The height of tree crown scorch can be predicted from these
values, using a model (SCORCH) in the Fire Behavior Prediction System. (See
XII.D.1.a, this Guide.)

b. Tree mortality. A model (MORTALITY) estimates the percentage of tree

mortality from scorch height, tree height, tree diameter, and crown ratio for eight
species of conifers that occur in the northern Rocky Mountains. (See XII.D.1.b,
this Guide.)

c. Total heat pulse to the site. Heat per unit area is a good estimate of the total
heat pulse to the site when all of the fuel that is burned is consumed by the
passing flame front. However, because this value does not account for long-term
burnout of heavy fuels or organic soil layers, heat per unit area is not a very good
estimate of the heat regime of the fire when much of the fuel, litter, or duff
consumption occurs after the flaming front has passed.

6. Aspects of a Fire's Heat Regime that Cannot Presently Be Predicted.

Many aspects of the heat regime of a fire cannot presently be predicted with any
known model. Many of the most important and influential effects of fire on a site
and its biological components are related to aspects of the heat regime of a fire
that are not described by the Fire Behavior Prediction System.

Fireline intensity, as described in II.B.3.b.(3), is a rate of heat release that is

related to flame length. However, the release of energy in flames has little
relationship to the amount of subsurface heating (Hungerford 1989). Peak
subsurface temperatures and the amount and duration of soil heating are not
related to any value predicted by the Fire Behavior Prediction System. The effects
of fire on fuels, soils, watershed, understory vegetation, and wildlife habitat cannot
be estimated from measures of fireline intensity or flame length alone.

a. Fuel burnout time. This is the length of time that fuels continue to burn after
the flaming front has passed, including all phases of combustion. The length of
the fuel burnout period is related to fuel properties and fuel moisture but cannot
be estimated by any known method.

b. Duration of smoldering and glowing combustion. Smoldering and glowing

combustion are related to the amount of fuel, its size class distribution, thickness
of duff and organic layers, and the moisture content of heavier fuels and duff. We
cannot presently predict the duration of time during which these combustion
phases will occur.

c. Total heat pulse to the site. Total heat pulse considers not only the heat
released in flames but also that released by smoldering and glowing combustion.
Heat per unit area only includes the amount of heat that is released in the flaming
front. Extensive studies in physics modelling is currently underway at the
Intermountain Fire Sciences Lab which may provide means to calculate the total
heat pulse to the site.

d. Soil heating. Most heat produced by the flaming front moves upward.
Downward movement of heat from flames cannot presently be predicted, but it is
not believed to be a significant source of subsurface heat. Most soil heating
results from long term fuel, duff, and organic layer burnout. Neither this heat, nor
its penetration into soil layers, has been modelled.

e. Burn severity. Burn severity is a term that qualitatively describes classes of

surface fuel and duff consumption. Large diameter down, dead woody fuels and
organic soil horizons are consumed during long-term, smoldering and glowing
combustion. The amount of duff or organic layer reduction is also called depth of
burn, or ground char (Ryan and Noste 1985). Because the amount and duration
of subsurface heating can be inferred from burn severity, this variable can be
related to fire effects on plants and soils. Factors regulating fuel and duff
consumption, and thus burn severity, are discussed in Chapter III.B.2. and 3. The
relationship between burn severity and its effects on plants is described in
Chapter VI.B.1.c. and VI.B.2.c.

(1) Descriptive classes. An example of a set of burn severity classes is given

below. Agency specific guidelines for assessing burn severity are described in
USDI-NPS (1992).

(a) Unburned.

(b) Scorched. Foliage is yellow; litter and surface vegetation are barely burned or
singed.

(c) Low severity. Small diameter woody debris is consumed; some small twigs
may remain. Leaf litter may be charred or consumed, and the surface of the duff
may be charred. Original forms of surface materials, such as needle litter or
lichens may be visible; essentially no soil heating occurs.

(d) Moderate severity. Foliage, twigs, and the litter layer are consumed. The duff
layer, rotten wood, and larger diameter woody debris is partially consumed; logs
may be deeply charred; shallow ash layer and burned roots and rhizomes are
present. Some heating of mineral soil may occur if the soil organic layer was thin.
(e) High severity. Deep ash layer is present; all or most organic matter is
removed; essentially all plant parts in the duff layer are consumed. Soil heating
may be significant where large diameter fuels or duff layers were consumed. The
top layer of mineral soil may be changed in color; the layer below may be
blackened from charring of organic matter in the soil.

(2) Relationship between fireline intensity and burn severity. There can be
many combinations of fireline intensity and burn severity on any site, depending
on fuel loading and distribution, and site weather and moisture conditions at the
time of the fire. For example, given a site with good, continuous surface fuels, and
a deep litter/organic layer, any of the following combinations of fireline intensity
and burn severity can occur (as well as a lot of intermediate combinations).

(a) High fireline intensity/high burn severity. Both the carrier fuels and organic
layer are dry. The result is a fire with high fireline intensity that exhibits vigorous
fire behavior, that is also a deep burning, high severity fire. Flames are long, large
fuels are removed, soil organic layers are consumed, and the long duration fire
causes a significant amount of subsurface heating.

(b) High fireline intensity/low burn severity. The carrier fuels are dry, but the
litter/duff layer is wet. The result is a fire with high fireline intensity, that exhibits
vigorous fire behavior, but which is a very low severity fire because the organic
layer is too wet to burn. Flames are long but little subsurface heating occurs.

(c) Low fireline intensity/high burn severity. The carrier fuels and surface litter are
moist, and litter/duff layers are dry. The result is a fire of low fireline intensity that
may barely cover the area, but of high burn severity wherever the litter/duff layer
ignites because it is dry enough to burn. Even though the surface fire was of little
apparent consequence, a significant amount of soil heating can occur, caused by
the consumption of dry duff layers, peat, and/or large diameter downed woody
fuels.

(d) Low fireline intensity/low burn severity. The carrier fuels are moist and the
litter/duff layer is wet. The result is a fire with low fireline intensity that also has
low severity.

(3) Application to shrub dominated communities. Burn severity concepts can

also apply to litter and duff layers beneath isolated trees and shrubs.

(1) Low severity fire may just scorch the litter beneath the shrub or tree crown.

(2) A moderate severity fire may consume some basal litter and organic matter,
but residual material remains. Some heating of deeper organic layers and soil
may occur.
(3) A high severity fire removes all litter and duff, leaving only an ash layer.
Significant amounts of soil heating can only occur where there is a high degree of
consumption of thick, basal organic layers. Isolated patches of severely burned
ground may occur where shrubs used to be, surrounded by extensive areas
where little soil heating occurred.

f. Burn pattern. The pattern of a fire is the mosaic of burned and unburned
vegetation and fuels. It can be further defined in terms of the degree of heating
and consumption of fuels and vegetation, such as scorched compared to severely
burned areas. A pattern can occur in the tree canopy, shrub canopy, in surface
fuels, or in litter, duff, and organic layers. The size of the mosaic can vary from
acres of scorched, consumed, and unburned patches in the canopy, to mosaic
patterns of burned and unburned fuels and litter layers of only a few feet, or even
inches. The effects of fire are closely related to the pattern of the fire, both on a
large and small scale. Fire effects vary considerably with burn pattern because it
reflects the variation in the fire's heat regime, above, at, and below the surface.

Significant variations in burn pattern are the result of differences in fuel continuity,
fuel loading, fuel moisture, aspect, wind, and ignition methods and techniques.
Whether a fire will become a surface or crown fire, and its effects on fuel
consumption and soil heating, can be estimated by a person skilled in fire
behavior or prescribed fire. However, there are presently no computational tools
with which to predict the exact burn pattern that will occur.

C. Resource Management Considerations

1. Levels of Fireline Intensity. Different levels of fireline intensity, along with

corresponding flame lengths, have special meaning both for the design of
prescriptions for prescribed fire, and in wildfire management activities. From
widely held and commonly agreed upon experience, the following are reliable
rules (Rothermel 1983).

a. When fireline intensity is below 100 Btu per foot of fireline per second, flame
lengths are less than 4 feet (1.2 meters).

(1) Fires can generally be attacked at the head or flanks of the fire by persons
using hand tools.

(2) Handlines should be adequate to hold the fire.

b. Fireline intensity 100 to 500 Btu per foot of fireline per second; flame lengths
are between 4 and 8 feet (1.2 to 2.4 meters).

(1) Fires are too intense for direct attack at the head of the fire by persons using
hand tools.
(2) Handline cannot be relied upon to hold the fire.

(3) Equipment such as bulldozers, pumpers, and retardant aircraft may still be
effective.

(4) Fires are potentially dangerous to personnel and equipment.

c. Fireline intensity 500 to 1,000 Btu per foot of fireline per second; flame lengths
are between 8 and 11 feet (2.4 to 3.4 meters).

(1) Fires may present serious control problems, such as torching out, crowning,
and spotting ahead.

(2) Control efforts at the head of the fire probably will be ineffective. Indirect attack
is probably the only means of suppression.

(3) Fires are definitely dangerous to personnel and equipment.

d. Fireline intensity above 1,000 Btu per foot of fireline per second; flame lengths
are greater than 11 feet (3.4 meters).

(1) Crowning, spotting, and major fire runs are probable.

(2) Control efforts at the head of the fire are ineffective by any known means of
suppression. Indirect attack and tactical counterfiring may be the only means to
slow the spread of the fire in certain directions.

(3) Fires are extremely dangerous to personnel and equipment in the immediate
vicinity of the fire.

These values have obvious implications for holding actions on prescribed fires
and suppression actions on wildfires. If only hand crews are available to hold a
prescribed fire, and handlines are the only lines of control, then prescription
variables (inputs to the spread model) should be set so that surface fires do not
exceed 100 Btu per second per foot, nor flame lengths exceed 4 feet (1.2
meters).

2. Relationship between Moisture Content of Big Sagebrush and Fire

Behavior. In vegetation types dominated by shrubs, moisture content of foliage
can be a dominant factor in the behavior of wildland fires. Within a given
geographical area, it is possible to determine threshold levels of foliar moisture
content that relate to degrees of fire behavior activity and difficulty of control. In
order to obtain such a database, foliar moisture content levels must be
documented in areas where fires are occurring and fire behavior observations
must be recorded. Sampling at established intervals over a period of several
years and relating moisture levels to easily identifiable growth stages of the plants
would provide the most useful information.

Threshold levels of moisture that relate to fire behavior in sagebrush have been
determined for Nevada and eastern Oregon.

a. Nevada. When Greg Zschaechner worked for the Bureau of Land Management
in Nevada on the Great Basin Live Fuel Moisture Project, he established
guidelines that relate the moisture content of big sagebrush (Artemisia tridentata)
to fire behavior and effective suppression tactics. Suppression tactics are
included in the following descriptions because they provide additional description
of the behavior of the fire. These levels are most accurate within Nevada but may
serve as general guidelines elsewhere.

(1) 181 percent and above. Fires will exhibit VERY LOW FIRE BEHAVIOR with
difficulty in continued burning. Residual fine fuels from the previous year may
carry the fire. Foliage will remain on the stems following a burn. Fires can
generally be attacked at head or flanks by persons using handtools. Handlines
should hold the fire without any problems. Fires will normally go out when the
wind dies down.

(2) 151 percent to 180 percent. Fires will exhibit LOW FIRE BEHAVIOR with fire
beginning to be carried in the live fuels. Both foliage and stem material up to 1/4
inch (0.6 centimeters) in diameter will be consumed by the fire. Burns will be
generally patchy with many unburned islands. Engines may be necessary to
catch fires at the head. Handline will be more difficult to construct but should hold
at the head and flanks of the fire.

(3) 126 percent to 150 percent. Fires will exhibit MODERATE FIRE BEHAVIOR
with a fast continuous rate of spread that will consume stem material up to 2
inches (5.1 centimeters) in diameter. These fires may be attacked at the head
with engines but may require support of dozers and retardant aircraft. Handline
will become ineffective at the fire head but should still hold the flanks. Under high
winds and low humidities, indirect line should be given consideration.

(4) 101 percent to 125 percent. Fires will exhibit HIGH FIRE BEHAVIOR leaving
no material unburned. Head attack with engines and dozers will be nearly
impossible on large fires, but may still be possible on smaller, developing fires.
Flanking attack by engines and indirect attack ahead of the fire must be used.
Spotting should be anticipated. Fires will begin to burn through the night, calming
down several hours before sunrise.

(5) 75 percent to 100 percent. Fires will exhibit EXTREME FIRE BEHAVIOR.
Extreme spread rates and moderate to long range spotting will occur. Engines
and dozers may be best used to back up firing operations and to protect
structures. Indirect attack must be used to control these fires. Fires will burn
actively through the night.

(6) 74 percent and below. Fires will have ADVANCED FIRE BEHAVIOR with
high potential to control their environment. Large acreage will be consumed in
very short time periods. Backfiring from indirect line such as roads must be
considered. Aircraft will need to be cautious of hazardous turbulence around the
fire.

b. Eastern Oregon. Fire behavior and its relationship to moisture content in

sagebrush has been monitored on Oregon rangelands east of the Cascades
(Clark 1989). The following moisture levels indicate how readily a fire can
propagate, given that adequate fine fuels are present between sagebrush plants
to carry the fire, or that sagebrush density is high enough for flames to reach
between plants where herbaceous fuels are sparse.

(1) Above 90 percent. Fire behavior is docile. The fire may or may not spread
and is easy to control.

(2) 60 to 90 percent. Fire is much more difficult to control. Fire is likely to burn
actively throughout the night, especially if wind is present.

(3) Less than 60 percent. The fire displays extreme fire behavior and rates of
spread, and is essentially uncontrollable by normal suppression methods.

3. Effect of Fuel Type Changes on Fire Behavior. The dominating factor

regulating fire behavior is wind in some fuel types and moisture content in others.
The behavior of fires in fuel types with a large component of fine materials, such
as the grass models, is most influenced by wind. Fuel moisture is much more
important than wind in regulating the activity of fires in fuel types with a lot of
larger diameter, dead woody fuels. Wind, while influential, is not so dramatically
important in heavy fuels as it is in grass or shrub type fuels.

Fire behavior can drastically change when a fire moves into a different fuel type. If
a fire moves from an area of logging residue to one dominated by cured grassy
fuels, flame length and fireline intensity will probably decrease, but the rate of
spread is likely to increase significantly. An optimal prescription for burning the
logged unit to reduce hazard fuels would include low moisture content in smaller
size classes of fuels and low windspeeds. Under these conditions, the desired
amount of consumption in the harvested area would be achieved, and any escape
into the grass fuels outside of the unit could easily be caught.

4. Effect of Long-Term Drying on Heat Release. Long periods of limited

precipitation result in deep drying of the surface organic layers. Deeper drying of
the entire fuel complex leads to an increase in fire behavior because of greater
involvement of larger fuels and surface organic fuels in the fire front. Because
more fuels burn in the initial stages of flaming combustion, fireline intensities and
flame lengths can be greater. More heat can also be generated during smoldering
and glowing phases of combustion, as deeper organic layers may burn. Fire
effects may be much more notable than if the site had burned under less droughty
conditions.

5. Burn Pattern. When igniting a prescribed fire, the pattern of burn that is being
obtained should be noted throughout the ignition period. If the desired mosaic is
not being obtained, alteration of ignition pattern may change the percent of the
prescribed fire area that is actually being covered by fire.

6. Firewhirls. Firewhirls are tight, spinning vortices filled with flame and hot gas
that have the appearance of a small tornado of fire. They can cause severe
difficulty in controlling a wildfire or prescribed fire by spreading pieces of flaming
material great distances beyond the project area.

When igniting a prescribed fire using strip headfires, it is important to let one strip
of fire burn down in intensity before igniting the next strip. This avoids
concentrated mutual convection, competition for incoming air, and a high
probability of initiating firewhirls at the ends of the strips. Also, by skillfully
designing the ignition pattern and sequence, the risk of firewhirls developing on
lee slopes, and where two fires merge, can be minimized.

D. Methods to Monitor Fire Behavior and Characteristics

Fire prescriptions contain elements that define ranges of acceptable weather and
moisture conditions that produce the desired fire behavior and characteristics.
Monitoring for a prescribed fire can include monitoring of weather and fuel
moisture before the fire to determine the daily weather patterns in a particular
area and to determine how close moisture conditions are to the prescribed range.
Some factors vary diurnally, such as temperature, relative humidity, and the
associated moisture content of small diameter fuels. Other prescription elements,
such as moisture content of soil organic layers or live fuel moisture content,
decrease slowly, and weekly monitoring is often adequate to detect change.

It is important to monitor all elements of the prescription during a fire to determine

that the fire remains within prescription, that the fire behavior predictions were
adequate, and to correlate with the subsequent effects of the fire treatment.
Whenever possible, information on fuel moisture, fire behavior, and fire
characteristics should be obtained in the same location(s) as fire effects data
collection occurs. In order to most effectively monitor rates of fire spread, flame
length, and burn severity, some equipment may need to be installed before a fire.

If site characteristics vary on the burned area, specific site attributes should be
documented as observations about fire behavior are made. Fuel type, vegetation
type, slope, and aspect should be recorded, as well as a notation about the
location where the observation is made. Whether the fire is a heading, backing, or
flanking fire should be noted at the same time as observations are made.

1. Burning Conditions.

a. Fuel moisture. Fuel moisture is a critical determinant of fire behavior and

characteristics. Techniques to monitor fuel moisture are described in Chapter
III.D.4, this Guide.

b. Weather. An important part of monitoring fire behavior and characteristics is to

have a good record of weather that occurred during the time that a fire occurs. A
standard set of weather observations should be taken at regular intervals during
the fire: temperature, relative humidity, windspeed and direction, clouds or other
indicators of instability, and the presence of thunderstorms. Standard procedures
for monitoring weather are detailed in Finklin and Fischer (1990). Agency specific
guidance on weather data collection is available in USDI-NPS (1992).

2. Fire Behavior and Characteristics. Agency specific guidelines for monitoring

fire behavior and characteristics are described for forests and for grassland and
brush types in USDI-NPS (1992). The following description provides additional
methods.

a. Rate of spread. Observations of rate of fire spread should only be taken after
the fire has reached a steady state, because this is what the fire behavior system
predicts. Rate of spread measurements are difficult to document on prescribed
fires with center or perimeter firing patterns, or narrow strip headfires. In these
situations the fire often has not reached a steady state, or its behavior is
influenced by the ignitions that have occurred in adjacent areas. Whether the fire
is heading, backing, or flanking at the point of observation should be noted. The
following discussion is taken largely from Zimmerman (1988).

(1) Visual observation. Visual observation and pacing of distances can be used
to take rate of spread measurements, particularly on a slowly moving fire. A
stopwatch is used to determine how long it takes a fire to cover a specific
distance. Rate of spread is calculated from the time/distance relationship.

(2) Metal tags. Numbered metal tags can be thrown at or near the flaming front.
A stopwatch is started when the front crosses a tag and a second tag is thrown
ahead of the fire. When the flaming front crosses the second tag, the stopwatch
records the elapsed time. The distance between the two tags is measured by
pacing or steel tape, and the spread rate calculated.

(3) Grid marking system. When high spread rates are expected, and/or when it
is not safe to be immediately adjacent to the fire, fire behavior can be measured
using a grid system installed before the fire. Spacing of markers should be related
to the expected rate of forward spread of the fire. Reference point markers can
consist of materials such as flagging tape tied to branches or poles painted with
bright paint. Times are recorded with a stopwatch or wrist watch as the fire burns
past each marker, and rate of spread is determined.

(4) Sketch map. Sketch maps of the fire perimeter can be made at different times
during the period of fire growth, a useful technique if reference points are plentiful
or the fire will cover a large area. This method requires a good vantage point or
the use of an aircraft. Rate of spread can later be calculated by dividing the
distances between different landmarks by the time periods it took to cover these
distances.

(5) Photography. Pictures of the fire can be taken at specific intervals and time
noted when each photo is taken. A 35 mm camera with split lens can be used
(Britton et al. 1977), in which one side of the image is focused on a watch, and
the other on the flame. Cameras are now commercially available that record date
and time on each image. The use of black and white infrared film greatly
increases the value of this technique because it increases the quality of an image
recorded through visually obscuring smoke.

(6) Video camera. Video cameras can be very successfully used when
monitoring fire behavior. Time and other observations can be recorded on an
audio track while recording the visual image. The advantages of video cameras
include the potential for making a complete record of fire as it burns in specific
areas, the fact that image quality can be immediately assessed, and that cameras
are relatively inexpensive and very portable.

A computerized image analysis system has been used to study video tapes. A
grid representing a known size or distance is set on the first frame, and
subsequent measurements can be made from the screen image (McMahon et al.
1987).

b. Flaming residence time. Residence time is the amount of time that it takes
the flaming front of the fire to pass a particular point. Residence time can be
difficult to measure because of the indefinite trailing edge of the fire as
concentrations of fuel continue to flame. It can be estimated from observations,
still photography, or a video camera. The video position analyzer system (ibid.)
also can be used to obtain more accurate residence time estimates. Use of
infrared sensors or film are extremely useful when smoke obscures flames.

c. Fuel burnout time. Residence time discussed above is only a measure for
flaming combustion. For monitoring that will later be related to fire effects, an
estimate of the total duration of smoldering combustion of large diameter fuels
and duff layers is important. Observation, repeated photography of the same
points particularly with black and white infrared film, repeated video camera
images, or a Probe-eye® can record fuel burnout time. Infrared images can sense
higher temperatures caused by continued smoldering or glowing combustion
when no visible signs of combustion are present. While it is not important to
document the duration of long-term combustion to the exact minute, it is important
to note whether smoldering combustion lasts for only a few minutes, or a few
hours, or several days.

d. Flame length. Flame length is measured along the slant of the flame. The
accuracy of estimation of flame length can be increased by installing reference
points that provide scale. Steel posts with 1-foot sections alternately painted red
and white or metal flags attached every 3 feet (the choice depends on the
expected scale of the flames) set in the burn area work very well (Rothermel and
Deeming 1980). These markers can be the same as those used to measure rate
of fire spread.

(1) Observations. Flame length data are usually obtained from visual
observations of average flame length at set intervals. Flame length is usually
recorded at the same time as rate of spread observations are made.

(2) Photography. Flame length can be documented with cameras and time and
location of observation of each exposure recorded. Accuracy is enhanced by use
of infrared film.

(3) Video camera. Not only are video cameras an excellent way of documenting
fire behavior, the passive image analyzer mentioned above (McMahon et al.
1987) allows a very accurate measurement of flame length. After a grid of known
size is established on the first frame, the tape is advanced until a representative
flame is seen on the screen. The image is frozen on the screen, and the flame is
outlined on the screen with a cursor. Computer software then calculates flame
length.

e. Burn pattern. A map of the burned area can be made at both a gross and
detailed scale. For general monitoring purposes, a map of the burned area can
show areas where the tree or shrub canopy was removed, areas where the fire
was an underburn, and areas the fire did not burn at all. Information on burn
pattern can be obtained by a walk through the burned area, by long transects, or
with photography. A low elevation aerial photo, or an oblique photo taken from a
high vantage point such as a hill or a tree, can be measured with a dot grid to
determine burn pattern. For large wildfires, satellite imagery can be used to obtain
information on the pattern of burned and unburned areas, and where the fire was
a surface fire or a crown fire. When choosing imagery for analysis, it must be
remembered that up to about 2 weeks may pass before scorch damage to
overstory tree foliage is apparent.

f. Burn severity/Depth of burn. The pattern of burn severity on the surface of

the ground can be quite complex, because it varies with the distribution of prefire
fuel loading and arrangement, thickness of litter and duff layers, and moisture
content of surface and ground fuels. While mapping the pattern of burn in the
surface fuels and vegetation for an entire burned area may be too large a task,
burn severity, and the degree of canopy removal, should be noted in the areas
where fire effects monitoring sites are located.

Surveyor pins or bridge spikes can be used in easy and practical way to monitor
depth of burn. The pins or spikes are pounded into the ground before the fire, with
a cross piece or top of the spike level with the top of the litter layer. After the fire,
the amount of pin exposed is a measure of the depth of organic material
removed. The amount of residual organic layer at each pin site can be measured
to obtain an estimate of duff removal. Use of an inexpensive metal detector can
make it much easier to relocate metal pins after the fire.

3. Potential Control Problems. The occurrence of any of the following during a

prescribed fire should be noted and recorded and the Burn Boss or Fire Behavior
Analyst notified.

a. Spotting. If spotting is occurring outside the burn perimeter, record the time of
occurrence, distance from the fire front, and location on a map.

b. Torching or crowning. Torching or crowning trees may produce spots and

may indicate that the situation requires extra caution. Note the time and location
of occurrence and any relationship observed with surface fire behavior.

c. Firewhirls. The location and time of any firewhirls should be recorded.

Observations about the fuels in the area of the fire whirls, or any relation to
ignition method or technique, should be noted.
d. Fire behavior exceeding prescription limits. Any observation of rate of
spread or flame length that exceeds specified limits could provide a potential
control problem. Fire behavior less than that predicted is not necessarily a control
problem, but can lead to an improper site treatment, and should be reported.

E. Summary

Knowledge of the behavior and characteristics of wildland fire are important both
for managing fire and for understanding and interpreting the effects of fire. The
heat regime created by a fire varies with the amount, arrangement, and moisture
content of flammable materials on a site. Trained and experienced people can
predict (within a factor of two) some aspects of the behavior and heat release of a
flaming front of a fire, and some associated fire effects such as crown scorch.
However, many fire effects are related to characteristics of fire that are not related
to the behavior of the flaming front and cannot presently be forecast.

This page was last modified 06/20/01

|Disclaimer| | Privacy| | Copyright| |USFWS Main Page| |Webmaster|

 National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 05/31/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home CHAPTER III - FUELS

Preface
Objectives By Melanie Miller
Fire Behavior
Fuels
A. Introduction
Air Quality
Soils & Water
Plants Fuel characteristics strongly influence how a given site will burn under
Wildlife specific environmental conditions. Fire also has an effect on fuels,
Cultural Res. because fire requires the consumption of fuel. Besides removing fuel,
Grazing Mgmt. fire can result in the creation of additional surface fuel, as vegetation
Evaluation drops fire-killed leaves, needles, and twigs, and dead shrubs and trees
Data Analysis fall and become part of the surface fuel layer.
Computer
Soft. All effects of fire on resources result from its effect on fuels, because
Glossary the way that fuels burn determines the heat regime of a fire. Each fire
Bibliography varies in the amount of fuels that burn, the size class distribution of
Contributions fuels that burn, the rate at which fuels burn, how much soil organic
matter burns, and whether living plants become fuel. The nature of fuel
consumption determines the peak temperatures reached, the duration
of heat, and the stratification of heat above and below the surface.

Fuels such as litter, snags, and downed trees, have important effects on
a site. Freshly deposited litter protects the soil surface from erosion by
raindrops. Unburned logs and fallen fire-killed trees provide locations for
mycorrhizae, nitrogen fixation, and habitat for birds, mammals, and
insects. Standing snags provide habitat for many animal species that
utilize this specific habitat.

This chapter will discuss the factors that regulate the effect of fire on
fuels. Different types of fuels are defined, and the factors that control
the amount of fuel and organic layer consumption are described. Dead
and live fuel moisture content are discussed in great detail because
moisture content is the most important determinant of the combustion
process and the heat regime of the fire. Basic principles of fuel
succession, the changes in the fuel complex over time, are
summarized. Those properties of fuels that affect the behavior of a
flaming fire are described in Chapter II of this Guide, Fire Behavior and
Characteristics. Those aspects of combustion that affect smoke
production are discussed in Chapter IV, Air Quality, this Guide. The role
of downed logs and organic matter in regulating soil nutrients and their
relationship to fire are discussed in the Soils chapter of this Guide,
Chapter V. Wildlife use of dead woody material as habitat is described
in Chapter VII, Wildlife Habitat.

B. Principles and Processes

1. Fuel Classification. Fuel is all vegetative biomass, living or dead,

that can be ignited by lightning or an approaching fire front. Wildland
fuels have been grouped into various classes.

a. Natural fuels vs. activity fuels. Natural fuels result from plant
growth and death, loss of foliage, branch breakage, and tree blowdown.
Activity fuels are similar to natural fuels but they are distributed
differently in time and space due to human activity such as logging,
thinning, chaining, and herbicide use.

b. Down, dead woody fuels. This class of fuels includes dead twigs,
branches, stems, and boles of trees and shrubs that have fallen and lie
on or above the ground (Brown et al. 1982). Wood can be either sound
or rotten. Sound wood is essentially intact. It may have checks or
cracks, but it still retains its structure. Rotten wood is partially
decomposed. Material is punky or can be easily kicked apart. It can be
important to distinguish between sound and rotten large diameter
woody fuels because their moisture retention and combustion
characteristics are very different. (See B.2.a.(3) and B.4.d.(3), this
Chapter.)

c. Soil litter and organic layers.

(1) Litter. Litter is the top layer of the forest floor, typically composed of
loose debris such as fine twigs, and recently fallen leaves or needles,
little altered in structure by decomposition. Litter can also include loose
accumulations of debris fallen from rangeland shrubs, and dead parts of
grass plants lying on or near the surface of the ground. Some surface
feather moss and lichen layers are also considered to be litter because
their moisture response is similar to dead fine fuel.

(2) Organic layers.

(a) Duff. Duff is the partially decomposed organic material of the forest
floor that lies beneath the freshly fallen twigs, needles and leaves. It is
equivalent to the fermentation and humus layers of soil.

(b) Organic soils. Soils that are essentially composed of deep layers of
organic matter form wherever production of organic matter exceeds
rates of decomposition. They frequently develop in poorly drained areas
where plant material partially decomposes in water or in saturated
environments. Organic soils can be extensive in wetlands and in cool,
moist climates (Buol et al. 1973). The amount of incorporated mineral
material can vary significantly. The organic content of these soils can
burn if soil moisture content is low enough.

d. Live fuels. Live fuels are living vascular plants that may burn in a
wildland fire. Live fuels include trees, shrubs, grasses and grass-like
plants, forbs, and ferns. Because of seasonal variation in their moisture
content, the flammability of their foliage can vary significantly.
Herbaceous plants, i.e., grasses and forbs, can cure, changing from a
live fuel to a dead fuel. The leaves and older needles of trees and
shrubs dry and fall from the plants, adding to the surface litter layer at
the end of the growing season. Living plants may contain a large
component of dead material, such as dead branchwood in older shrubs.
This increases their flammability and the likelihood that the entire plant
will be consumed by fire. Live fuel moisture cycles are described in
greater detail in section B.5. of this chapter.

e. Fuel strata. Fuels have been classed as surface, ground, aerial, or

ladder fuels. These terms are described in Chapter II.B.1.a., this Guide.

f. Total fuel vs. available fuel. Total fuel is the total amount of fuel
present on a site. Available fuel is the amount that can burn, under a
given set of conditions. The amount of available fuel depends on fuel
size, arrangement, and moisture content (Brown and See 1981). Fuel
size can affect availability if there are inadequate amounts of smaller
sized fuels to burn and transfer enough heat to larger size fuels to raise
them to ignition temperature. Standing tree boles are not considered to
be available fuel because they are extremely unlikely to burn in wildland
fires, except for smoldering in punky snags.

There can be long-term changes in fuel availability. The total amount of

fuel within a stand increases as plants grow, while the distribution of
fuel within a stand changes when snags or branches fall or foliage
drops. There can also be short term changes in availability due to fuel
moisture content. At any given point in time, the most important factor
affecting fuel availability is its moisture content, because this
determines whether fuel can ignite, and whether it can sustain
combustion.

2. Factors Regulating Dead Fuel Consumption.

a. Relationship to physical and chemical properties of fuel. The key

fuel properties that affect fire behavior were described in Chapter
II.B.3.b. The following discussion explains how some of these factors
influence fuel consumption.

(1) Fuel size. Fuels less than 1/4-inch diameter are almost completely
consumed by fire over a wide range of burning conditions. Most
branchwood between 1/4-inch and 3 inches is also consumed (Martin et
al. 1979). A fair prediction of the consumption of large diameter dead
woody fuels is possible if the average preburn diameter is known
(Reinhardt et al. 1991).

(2) Fuel continuity. Fuel continuity relates to the proximity of individual

pieces of fuel and also of different fuel strata. It affects fire spread, how
much of an area ignites, and how much fuel is consumed. Breaks in fuel
continuity contribute to patchy fire spread and may result in patchy fuel
consumption.

(a) Forests. Large diameter fuels in local accumulations are more likely
to be consumed than if these fuels are scattered. Anderson (1983)
found that large downed woody fuels need to be within a distance of
about 1.5 diameters of each other for interactive burning to occur.

(b) Rangelands/grasslands. Fuel consumption in range and grass types

can be closely related to fuel continuity. If fuels are sparse, windspeed
may not be adequate to spread the fire, limiting the amount of fuels that
are ignited and consumed.
(3) Quality. Wood may be sound, rotten, or partially rotten. In north
Idaho, consumption of rotten wood was greater than that of sound
wood, even though moisture content of rotten wood was higher.
Consumption of completely rotten pieces was higher than that of
partially rotten pieces (Reinhardt et al. 1991).

(4) Heat content. The heat content of wood is about 8,000 Btu per
pound (Albini 1976). Pitch adds to the flammability of wood because its
heat content is about 15,000 Btu per pound (Carmen (1950 in Byram
1959). Pitchy fuels can burn at a much higher moisture content than
those without pitch. A damp pitchy stump that is ignited is often
completely consumed by fire.

(5) Fuel moisture content. The major effect of moisture on small fuel
consumption is simply whether fuels are dry enough to ignite. Eighty to
90 percent of fine woody forest fuels are consumed wherever fire
spreads (Brown et al. 1985). This is also true for fine rangeland fuels.
The proportion of large diameter woody fuels consumed is more
strongly influenced by their moisture content than by any other factor
(Reinhardt et al. 1991).

b. Relationship to phase of combustion. The phase of combustion

during which dead woody fuels are consumed is related to their size.

(1) Small diameter fuels. Fine fuels tend to be consumed during flaming
combustion. However, the arrangement of small woody fuels sometimes
does not provide enough mutual heating during the flaming state for
complete consumption to occur. Blackened branches may burn off or
fall into the ash and generate enough mutual heat for more flaming
combustion to occur. Eventually the amount of heat that is generated
decreases and can no longer support flaming, and the remaining
consumption of these small pieces occurs by smoldering and glowing
combustion (Norum 1992).

(2) Large diameter fuels. While the surface layer of woody fuels may
initially support flames, most of the consumption of large woody fuels,
both sound and rotten, occurs in the smoldering and glowing phases of
combustion. Glowing combustion in large woody fuels commonly lasts
for 10 to 20 times longer than the flaming phase (Anderson 1983).

3. Factors Regulating Consumption of Duff and Organic Soils.

a. Moisture content. As is true for large diameter woody fuels,
moisture content is the most important variable influencing consumption
of duff and soil organic layers. Duff and soil temperatures remain below
the boiling point of water (100 C.) until all moisture is evaporated
(Hartford and Frandsen 1992). Heating of organic layers to the high
temperatures required for ignition cannot occur while moisture is
present.

Specific relationships have been observed between duff moisture and

duff consumption. At less than 30 percent duff moisture content, duff
layers burn on their own once ignited, a threshold level observed in the
southwest U. S., Pacific northwest, and the northern Rockies (Brown et
al. 1985). At 30 to 120 percent duff moisture content, the amount of
consumption of duff depends on the amount of consumption of
associated fuel. When duff moisture content is greater than 120
percent, duff essentially will not be consumed.

Similar relationships have been found for organic soils. Peat is a

deposit of slightly or non-decayed organic matter, while the organic
content of muck is markedly decomposed (Buckman and Brady 1966).
Peat burns well when its moisture content is below about 30 percent
(Craighead 1974 in Hermann et al. 1991). Blocks of organic soil from
south central Florida sustained smoldering up to 135 percent moisture
content (McMahon et al. 1980 in Frandsen 1987).

Wet pocosin muck does not burn, but once the water table has lowered,
these soils can ignite and sustain combustion. However, the moisture
limits for ignition are not known (Frandsen 1993). Research is presently
being conducted to determine factors affecting consumption in pocosin
organic muck (Frandsen et al. 1993).

b. Surface fuel consumption. Heat generated by consumption of

surface fuels can dry, preheat, and then ignite the duff layer. The
amount of consumption of large diameter fuels was related to duff
reduction and mineral soil exposure in north Idaho (Brown et al. (1991),
and western Oregon and Washington (Little et al. 1986; Ottmar et al.
1990, Sandberg 1980).

c. Preburn duff depth. Duff consumption was strongly related to

preburn duff depth in the northern Rocky Mountains (Brown et al.
(1991); in jackpine (Stocks 1989 in ibid.); Alaska black spruce (Dyrness
and Norum 1983); white spruce/subalpine fir (Blackhall and Auclair
1982 in ibid.); and southwestern ponderosa pine (Harrington 1987), but
not in deeper duff layers of the Pacific Northwest (Sandberg 1980; Little
et al. 1986). d. Inorganic content. Mineral soil becomes mixed with soil
organic layers by freeze-thaw cycles, insect and small animal activity,
overland flow, windthrow, and management actions, particularly
skidding of logs (Hartford 1989). Mineral material affects combustion of
organic layers because it absorbs some of the heat that would
otherwise preheat combustible materials (Frandsen 1987). The greater
the amount of mineral material in the organic layer, the lower the
moisture content of the organic layer had to be before it would burn
(ibid.). If the ratio of mineral particles to organic matter (mass ratio) was
greater than about 4 to 1, smoldering did not occur (ibid.).

e. Phase of combustion. Almost all consumption of duff and organic

soils occurs during the smoldering and glowing phases of combustion.
Combustion can continue for hours, days, and in the case of pocosin
soils, for weeks after ignition, if organic layers are dry (Frandsen 1993).

4. Dead Fuel Moisture. Fuel moisture has a significant effect on fuel

availability and fuel consumption because it suppresses combustion.
Part of the heat produced by the combustion of wood is used to drive off
moisture in adjacent woody fuel. If the moisture content is high, the heat
generated may be insufficient to dry these fuels and heat them to
ignition temperature, and the fire will not continue to burn.

Fuel moisture is the ratio of the weight of moisture in the fuel to that of
the dry weight of the fuel. The formula for fuel moisture and its effect on
fire behavior is described in Chapter II.B.4.b.(2), this Guide. Moisture
effects on woody fuel consumption were discussed in B.2 a. (5), this
Chapter, and on duff and organic consumption in B.3.a. The following
discussion describes the most important factors affecting moisture
content of dead woody fuels, litter layers, and duff and organic layers.
Live fuels and their moisture cycles are discussed in B.5., this Chapter.

a. Wetting and drying process. Water in fuels can be present in liquid

or vapor form.

(1) Liquid water. Liquid water comes from rainfall, snowmelt, or

condensation. It can be present both on the surface, and within cell
cavities (Schroeder and Buck 1970). At the fiber saturation point (about
30 to 35 percent of the fuel dry weight), the cell wall holds as much
water as it physically can, but no liquid water is present within cell
cavities (McCammon 1976).

Liquid water is readily absorbed by fuels through their surface, filling cell
cavities and intracellular spaces (Schroeder and Buck 1970). In liquid
water, molecules travel with different speeds and directions. A water
molecule at or near the surface of a layer of water can attain a high
enough speed after colliding with another molecule to escape from the
liquid water into the air. By this process, called evaporation, a liquid
water molecule becomes a water vapor molecule (ibid.). Evaporation is
the primary drying process when fuels are saturated. It decreases in
importance as fuels dry below the fiber saturation point.

(2) Water vapor. The following discussion is derived from Schroeder

and Buck (1970), except where noted. Water present in the atmosphere
in the form of a gas is called water vapor. That part of the atmospheric
pressure due to the presence of water vapor is called vapor pressure.
The maximum amount of vapor that the atmosphere can hold when it is
saturated depends on the air temperature. Water vapor molecules
move from an area of higher concentration to one with lower
concentration until vapor pressure is equal.

Water molecules in fuels can be bound to cellulose molecules or held

by capillary action in tiny openings in the cell wall (Simard 1968).
Molecules closest to the cell walls are held the most tightly. Successive
layers of water molecules are held with progressively weaker bonds
until the cell walls become saturated. At less than saturation, water
vapor moves between a fuel particle and the atmosphere if the vapor
pressure of the layer of water in the fuel does not equal the vapor
pressure of the atmosphere. If the vapor pressure within the outer layer
of fuel is greater than that of the atmosphere, moisture escapes to the
atmosphere, and fuel moisture content decreases. If the vapor pressure
of the atmosphere is greater than the vapor pressure within the outer
layer of fuel, the fuel takes water vapor from the atmosphere, increasing
the fuel moisture content.

b. Equilibrium moisture content. Equilibrium moisture content (EMC)

is the "value that the actual moisture content approaches if the fuel is
exposed to constant atmospheric conditions of temperature and relative
humidity for an infinite length of time" (Schroeder and Buck 1970). The
EMC determines the amount of water vapor that a specific piece of
wood can hold (Simard 1968). A unique EMC exists for each
combination of atmospheric temperature and relative humidity, with the
associated vapor pressure (Schroeder and Buck 1970). If fuel moisture
content is greater than EMC, vapor diffuses out of the fuel, and the fuel
becomes drier. If fuel moisture content is less than the EMC, water
vapor transfers into the fuel particle and the fuel becomes wetter.

Atmospheric temperature and relative humidity are never constant and

tend to vary diurnally. Equilibrium moisture content also varies. Because
fuels usually take up and release moisture at a slower rate than the
temperature and humidity changes, the actual fuel moisture content
lags behind the equilibrium moisture content. The greater the difference
between the equilibrium moisture content and the fuel moisture content,
the more rapidly vapor diffusion occurs, and the more rapidly the fuel
particle exchanges moisture with the atmosphere. As a particle
approaches equilibrium moisture content, the exchange occurs more
slowly. Fuel moisture content never reaches equilibrium moisture
content, because other physical processes prevent a complete
exchange of vapor (Schroeder and Buck 1970). A fuel that is gaining
moisture stabilizes at a lower moisture content than a fuel that is drying
(Simard 1968). Van Wagner (1987 in Viney 1991) noted a 2 percent
lower EMC for wetting fuels compared to drying fuels.

c. Timelag theory.

(1) Timelag principle. Drying and wetting of unsaturated dead woody

fuel has been described by the timelag principle. A timelag has been
defined as the length of time required for a fuel particle to reach
approximately 63 percent of the difference between the initial moisture
content and the equilibrium moisture content.(1)

(2) Timelag period. Under standard conditions, defined as 80 F. and 20

percent relative humidity, the length of time that it takes a fuel particle to
reach 63 percent of EMC is a property of the fuel and is referred to as
the timelag period (Schroeder and Buck 1970).

(3) Timelag classification. The proportion of a fuel particle exposed to

weather elements is mathematically related to its size. Small diameter
fuel particles have large surface area to volume ratios. Moisture levels
in these fine fuels can change rapidly with changes in temperature and
relative humidity. Large diameter fuel particles have small surface area
to volume ratios, and their moisture content changes very slowly in
response to changes in temperature and relative humidity. Time lag
thus increases with increasing fuel diameter.

(a) Dead woody fuel timelag classes. Downed dead woody fuels have
been grouped into size classes that reflect the rate at which they can
respond to changes in atmospheric conditions (Lancaster 1970). The
classes relate to an idealized surface area to volume ratio and an
average timelag that represents each fuel class. Classes relate to the
theoretical length of time required to reach 63 percent of EMC.

i. 1-hour timelag fuels - less than 1/4-inch diameter (less than 0.6 cm).

ii. 10-hour timelag fuels - between 1/4-inch and 1-inch diameter (0.6 to

2.5 cm).

iii. 100-hour timelag fuels - between 1-inch and 3 inches diameter (2.5
to 7.6 cm).

iv. 1000-hour timelag fuels - between 3 and 8 inches (7.6 to 20.3 cm)
diameter.

(b) Forest floor timelag classes. There is a loose correspondence

between these timelag classes and forest floor litter and duff, although
the deeper the duff layer, the more approximate is the relationship. The
corresponding classes assigned for fire danger rating purposes were
(Deeming et al. 1977):

i. 1-hour timelag fuels - dead herbaceous plants and uppermost layer of

forest floor litter.

ii. 10-hour timelag fuels - layer of litter extending from just below the
surface to 3/4 of an inch below the surface.

iii. 100-hour timelag fuels - forest floor from 3/4 inch to 4 inches below
the surface.

iv. 1000-hour timelag fuels - forest floor layer deeper than 4 inches
below the surface.

(4) Timelag of other fine fuels. Weathered aspen leaves, tree lichen
(Alectoria jubata) and some cheatgrass fuel beds were shown to act as
1-hour timelag fuels (Anderson 1990). The surface layer of lichens and
mosses that carries fire in Alaska responds as a 1-hour fuel to
temperature/relative humidity changes (Mutch and Gastineau 1970).
However, conifer needle litter of some species belongs in the 10-hour
timelag category (Anderson 1990), despite its high surface area to
volume ratio. Other factors such as surface covering influence the rate
at which fuel moisture changes in response to environmental
conditions.

d. Fuel properties that affect dead fuel moisture content.

(1) Surface covering. The presence of a surface coating of organic

material can limit movement of water, whether liquid or vapor (Simard
and Main 1982). Dead woody fuel with bark gained and lost moisture at
two-thirds the rate of the same diameter fuels without bark (Simard et
al. 1984).

Moisture exchange in recently cast conifer needle litter is inhibited by a

coating of fat, waxes, and cutin deposits (Anderson 1990). Anderson
(ibid.) noted timelags of 5 to 34 hours for recently cast conifer needle
litter, rather than the expected timelag of 10 hours. Weathering causes
the breakdown and removal of needle coatings that slow vapor transfer.
Timelags of 2 to 14 hours were measured in weathered conifer litter,
which is still slower than the timelag of less than two hours expected for
that diameter of particle (ibid.).

(2) Composition. The material of which a fuel is composed affects its

structure, porosity, ability to gain or lose atmospheric moisture, and the
movement of vapor within the particle. Composition and fuel moisture
response properties vary significantly among dead woody fuels,
deciduous leaf litter, grass litter, and coniferous needle litter (Simard
and Main 1982).

(3) Amount of decomposition. Woody fuels that have been affected by

weathering and decomposition often develop deep cracks that increase
their surface area to volume ratio. Both liquid water and vapor can enter
or leave the fuel through these splits in the wood, increasing the rate of
moisture exchange. There may be few naturally occurring forest fuels
that are actually 1000 hour timelag fuels, because almost all large
pieces of wood have cracks that effectively increase their wetting and
drying rates (Miller 1988, personal observation).
The cell structure in highly decomposed wood, such as rotten logs, has
broken down, and moisture can travel more easily through this material
than through solid wood. The moisture content of rotten wood can be
very different from that of sound wood of the same diameter.

(4) Thickness and density of litter or duff layer. Because litter and duff
layers have porosities of 70 to 90 percent, air can diffuse through them
at 60 to 80 percent of the diffusion rate in free air (Fosberg 1975). The
particles of organic matter in these layers exchange moisture with the
atmosphere in the void space in the litter and duff layer (ibid). The
moisture within the voids seeks equilibrium with the external
atmosphere (ibid.). Van Wagner (1979) observed that the drying
environment within a 3 centimeter (1.2 inches) deep needle litter layer
was less favorable at the bottom than at the top because the lower part
of the layer was farther from the drying surface. Anderson (1990) noted
that moisture diffusion rates were slower if litter fuel beds were deeper
or more densely packed. The lower part of a litter and/or duff layer can
become matted and tightly bound by fungal strands (Harrington and
Sackett 1990). The wetting and drying response of this layer is likely to
be slower than that of the more loosely packed material nearer the
surface because of slower rates of water vapor diffusion.

e. Soil moisture effects on fuel moisture content. Afternoon moisture

content of eucalyptus (Eucalyptus spp.) leaf litter placed on wet soils
was much higher than litter placed on dry soils (Hatton et al. (1988).
Moisture appeared to diffuse upwards from the wet soil, increasing the
relative humidity environment of the leaf litter, causing higher litter
moisture content (ibid.). The biggest effect of the wet soil was noted in
early mornings, probably because wet soils made more moisture
available to condense on dead leaves. Litter on the dry soils was dry
enough for the litter to burn throughout the night, while litter on wet soil
plots was too wet to burn.

Active surface fire behavior occurred throughout many night time

burning periods during the 1988 fire season in Yellowstone National
Park. Night time moisture recovery of lodgepole pine litter was much
slower and reached lower maximum levels than expected (Hartford and
Rothermel 1991). The limited amount of surface litter moisture recovery
was partially attributed to the lack of moisture in the air and in the soil
that could contribute to an increase in night time litter moisture content
(ibid.).(2)
f. Effect of weather factors on fuel moisture content.

(1) Precipitation duration. Wood absorbs water as long as the surface is

wet, so precipitation duration is usually more important than
precipitation amount in determining moisture content of dead woody
fuels (Fosberg 1971 in Simard and Main 1982). The rate of diffusion of
liquid water into wood is usually less than the rate at which precipitation
occurs, so much of the rain water drips off before it can soak into the
wood.

(2) Precipitation amount. The amount of precipitation is more important

than the duration of precipitation in determining the moisture content of
duff, organic soils, and accumulations of organic materials that occur
beneath isolated trees and shrubs. Duff layers and organic soils retain
much of the precipitation that falls, allowing it to slowly soak into the fuel
particles (Simard 1968 in Simard and Main 1982).

(3) Temperature. Temperature affects both the humidity of the air and
its vapor pressure, and thus the equilibrium moisture content. Higher
fuel temperatures decrease relative humidity in the microclimate near
the ground (Rothermel et al. 1986), which also decreases the EMC.
Higher fuel temperatures increase the tendency of bound water vapor to
diffuse away from the fuel, thus drying it further (Schroeder and Buck
1970). Fuel temperature is affected by slope, aspect, time of day, cloud
cover, canopy cover, sun angle, and the albedo of the fuel.

(4) Relative humidity. Relative humidity has a significant effect on

moisture content of small diameter fuels because water vapor can
readily penetrate into or escape from the center of small fuel particles.
Diurnal changes in relative humidity have little effect on the moisture
content of large diameter fuels because their large volume prevents
rapid movement of moisture molecules between the surface of the fuel
and its center. Relative humidity can have a major effect on large fuel
moisture content if there is a long period of time without precipitation.

(5) Wind. Wind has its most important drying effect on woody fuels
when liquid water is evaporating because it removes any layer of water
vapor that may be adjacent to the fuel. Wind has a greater effect on wet
fuel particles that are above the surface, causing them to dry more
rapidly than material on the ground (Simard and Main 1982). When fuel
is below the fiber saturation point and most vapor loss is by diffusion,
the effect of wind becomes less important as the fuel becomes drier
(Schroeder and Buck 1970). Wind has a more significant drying effect
on small diameter fuels than on large diameter fuels, duff, or organic
layers.

g. Relationship of topography to fuel moisture content.

(1) Fuel moisture tends to vary with topographic position. Fuels are less
directly exposed to sun on north slopes than south slopes so their
moisture content tends to be higher. Temperatures are generally cooler
and humidities higher at upper elevations, so fuel moistures are usually
higher than at lower elevations.

(2) Topography partially determines the strength of any night time

inversion layer that forms. If a steep inversion and temperature gradient
forms, fuel moisture recovery can be fairly high because of low
temperatures and high relative humidities.

If an inversion forms in a valley, a thermal belt may form at the top of

the inversion layer. In this belt, temperatures are warmer than at lower
elevations within the inversion, and warmer than at higher elevations
because temperature decreases with altitude. Higher night time
temperatures, lower relative humidities, and lower fuel moistures occur
within the thermal belt than at other locations along the slope. Fires can
remain active throughout the night within the thermal belt, while activity
is limited below the inversion layer (Schroeder and Buck 1970).

5. Live Fuel Moisture.

a. Effect of live fuel moisture on fire. Live fuels can either be a heat
sink or a heat source in a wildland fire, depending on their moisture
content. If live fuel moisture levels are high enough, they absorb some
of the heat produced by associated burning fuels without themselves
igniting, and thus do not contribute to the progress of the fire. If live fuel
moisture is low, the combustion of dead fuels readily produces enough
heat to desiccate and ignite the live fuels, which then add to the total
amount of heat released by the fire (Burgan and Rothermel 1984). Live
fuels can thus retard, stop, or contribute to fire spread.

b. Factors regulating live fuel moisture.

(1) Internal factors. Moisture content of living plants is controlled largely

by species morphology and physiology. The amount of water in plant
tissue, and thus its moisture content, relates closely to events during a
plant's seasonal growth cycle (plant phenology). For a given species,
the maximum and minimum moisture content values and the average
values during different parts of the growing season are controlled more
closely by the plant structure and its adaptations to the general climate
of the area, than by daily weather. Seasonal timing of drying for specific
deciduous shrub, forb, and grass species were found to be similar
between wet and dry growing seasons, although moisture levels were
generally higher in the wet season (Brown et al. 1989).

(2) Site factors. Site conditions can cause differences in moisture

content within the same species, possibly because of physiological
conditioning or even a genetic adaptation to the site (Reifsnyder 1961).
Differences in foliar moisture content within a single species were
related both to differences in substrate and the amount of shading
provided by a forest canopy (Blackmarr and Flanner 1968).

(3) Climatic variation. Climate affects such factors as the timing and
length of the growing season, the length of the green-up period, and the
existence of seasonal periods of cold- induced dormancy or drought or
heat-induced quiescence.

c. Differences among species groups. There are characteristic

differences in seasonal moisture patterns for groups of species.
Deciduous leaved woody plants tend to have higher moisture content
values than evergreen leaved plants, and the seasonal pattern of
moisture changes tends to vary more. Coniferous trees have entirely
different foliar moisture patterns than deciduous trees. Herbaceous
species moisture levels can be higher or lower than that of associated
shrubs, depending on the species present and the time of year. There
are differences in average maximum and minimum moisture values
among species within any group, depending upon the morphology of
the species, and the relative amount of new and old growth on the
plant.

d. Deciduous leaved shrubs. The general pattern for deciduous

leaved shrubs is for moisture to rapidly increase to a peak level soon
after bud break and begin to decrease after all new seasonal growth
has occurred. Moisture then slowly declines for the remaining part of
the growing season until leaves cure.
Data from Alaskan aspen stands illustrate variation in moisture content
levels among deciduous species, as well as variability due to site
differences (Norum and Miller 1981). Maximum spring moisture content
of leaves and small twigs of highbush cranberry (Viburnum edule)
(Illustration III-1, page III-24) was about 325 percent, but its moisture
content dropped to about 225 percent by midsummer where it remained
for most of the growing season. Rose (Rosa acicularis) on that same
site in that same season had a spring maximum value near 375
percent, but its moisture content decreased to about 175 percent where
it remained until fall curing. Maximum moisture levels for that same
species of rose on a drier aspen site were less than 250 percent and
persisted at about 165% for much of the growing season. Blueberry
(Vaccinium uliginosum) (Illustration III-2, page III-25), a smaller stature
deciduous species on a black spruce site nearby, had spring maximum
moisture value of less than 250 percent and spent most of the summer
at about 125 percent moisture content (Norum and Miller 1981).

For all of these species, moisture content did not significantly decrease
as fall coloration appeared on the leaves. Moisture content began to
drop markedly as the abscission layer formed at the bases of the
petioles and cut off water transport to leaves, when obvious drying and
browning of the leaves occurred (Miller 1981).

e. Evergreen leaved shrubs. The general pattern for broad-leaved

evergreen shrubs is more complex than for deciduous species because
evergreen shrubs sometimes retain old leaves for several years. They
tend to have lower spring maximum values and much lower growing
season average values than deciduous species. Values increase from
an overwintering minima as new growth is added in the spring, or at
other times of the year when precipitation triggers growth after a dry
season. Values decrease significantly after new growth ceases. Some
evergreen leaved species develop ephemeral leaves in late winter and
early spring. Average foliar moisture content drops significantly as these
seasonal leaves cure and drop from the shrub.

A typical profile for sagebrush (Artemisia tridentata) moisture content

would be a rise from early to late spring from about 150 percent to
about 250 percent, with a subsequent decline to 60 percent or less in
mid to late summer (Schmidt 1992). Riedel and Petersburg (1989)
found that the lowest summer levels for sagebrush moisture (Illustration
III-3, page III-26) were reached one month earlier in one year than the
previous summer. Sagebrush flammability has been related to threshold
levels of moisture content. (See II.C.2., this Guide.)

In the Alaskan interior, maximum foliar moisture content levels for

Labrador tea (Ledum decumbens) were only 145 percent, and that peak
value occurred about a month after the maximum moisture values were
reached in associated deciduous shrub species (Norum and Miller
1981). Moisture content of new leaves of chamise (Adenostoma
fasciculatum) in California were at 125 percent in late May, dropped to
about 60 percent in early September, and rose to about 90 percent
when the plants again became physiologically active in early December
(Dell and Philpot 1965). Maximum moisture levels for galberry foliage
(Ilex glabra) averaged about 140 percent in North Carolina, while
minimum values were about 100 percent (Wendel and Story 1962).
Maximum values for redbay (Persea borbonia) foliage were about 120
percent, while fall and winter minima were around 60 percent (ibid.).

f. Herbaceous plants. Herbaceous moisture content can also vary

significantly among species. Moisture levels can be much higher at the
beginning of the growing season than for other species groups because
all of the plant is new tissue. Also, because there is no residual
material, all of the plant can become cured, sometimes before the end
of the growing season. This is especially notable for grasses and other
species in areas with hot, dry summer weather. In north Idaho, moisture
content of cheatgrass (Bromus tectorum), an annual grass, for example,
was measured to be 150 percent on June 20, but was only 9 percent on
July 20 (Richards 1940). All of the plant material, once cured, responds
to atmospheric conditions as a dead fine fuel, as reflected by the 9
percent moisture level just cited.

Some species of grasses and forbs in some regions can produce new
growth in the fall, after a summer of quiescence, thus causing fall green-
up and associated increase in moisture content. Green-up is caused by
renewed growth of perennial species and germination of seeds.

Some herbaceous species do not cure and dry out during the summer,
rather only begin a significant amount of curing as frost occurs in the
fall. In north Idaho, moisture content of fireweed (Epilobium
angustifolium) plants was 426 percent on June 20 and 241 percent on
September 10 (ibid.). In interior Alaska, bluejoint reedgrass
(Calamagrostis canadensis) was first measured at about 400 percent
moisture content on May 27 when the plants had about 1 to 1-1/2 feet
of leaf growth. Moisture content of plants declined to about 260 percent
by June 30 and was about 200 percent on August 28, just before the
first frost (Norum and Miller 1981). In north central Michigan, large
leaved aster (Aster macrophyllus) was measured at about 420 percent
moisture content at the beginning of June, and the lowest moisture level
observed for the rest of the summer fluctuated around 300 percent
(Loomis and Blank 1981).

Data for most herbaceous species show only a slow decrease in

moisture levels after early growing season maxima. However, in
western Wyoming, grasses and forbs had some increase in moisture
content in response to mid-summer rain. By September, however, the
drying trend was not altered by rainfall (Brown et al. 1989).

g. Coniferous trees. Moisture content of coniferous foliage also varies

significantly with season but the pattern is quite different than that
shown by deciduous and herbaceous species. Coniferous species
retain their needles for several years; the number of years is a species
characteristic.(3) For most species, the lowest level of moisture content
of needles formed in previous

years occurs in late spring, during about the same time period in which
buds expand, and new needles and twigs are formed. Moisture content
of old needles increases during most of the growing season to a
maxima during late summer and/or early fall (depending on species and
region).

Moisture content of new needles is very high as buds break, needles

grow, and stems elongate, but starts dropping significantly about the
same time as the new terminal bud on the end of the current year's
growth forms. The moisture content of new foliage drops to about the
same level as that of old foliage late in the growing season. In the
southeastern U.S., conifers may flush more than once during the
growing season. The moisture cycle in older foliage may be different
from that of conifers that grow in climates with winter cold and/or shorter
growing seasons.

The difference between low and high moisture values for 1-year-old
black spruce (Picea mariana) foliage in Alaska varied from 28 to 40
percentage points on different collection sites (Norum and Miller
unpublished). Seasonal lows occurred in June and seasonal high
values in August (Illustration III-4, page III-27). Seasonal low values for
Douglas-fir foliage occurred in mid to late June in Montana, with a peak
value reached by early September (Philpot and Mutch 1971; Rothermel
1980). The range between high and low values varied from about 20 to
40 percentage points.

Crown fires were much more prevalent in Douglas fir trees burning in
late spring and early summer experimental fires than on those sites
burned in late summer and early fall (Norum 1975). Low springtime
foliar moisture values may explain the observed difference. The peak of
the fire season in boreal latitudes is usually shortly after summer
solstice, and the low foliar moisture content of black spruce may be a
factor in tree crown ignition. However, the fire season in the western
United States generally peaks in August, at a time when moisture
content of conifers is increasing to a seasonal high. The readiness of
western species of conifers to crown during extreme fire weather is not
due to low foliar moisture values, although it may be enhanced by early
drying of the oldest needles.

Measured moisture content of live foliage in northeast Oregon,

southwest Idaho (Miller 1988), and northwest Wyoming (Hartford and
Rothermel 1991) in the extreme fire season of 1988 was not different
from normal moisture values for that time of the year. Experimental
evidence suggests that conifers can rapidly transport water into their
foliage when heated to temperatures that occur in a wildland fire
(Cohen et al. 1990). This can delay branch and foliage drying and could
inhibit crown heating and ignition. During drought conditions, trees may
not be able to transport water into their crowns, increasing their
flammability and hence their crowning potential (ibid.).

6. Fuel Succession. Vegetative biomass tends to accumulate over

time. However, not all biomass is available fuel. Biomass is all of the
vegetation on a site, while available fuel is what can burn. Fuel
succession is the change in the fuel complex over the long term,
including changes in loading, size distribution, availability, and live to
dead ratios. These changes are the net result of the counteracting
processes of accumulation and depletion (Brown 1987).

a. Accumulation.

(1) Litter layer. The amount of foliage that is produced each year affects
the amount of new litter that accumulates. Because coniferous trees
retain their needles for several years, there may be no relationship
between the productivity in a particular year and the amount of needle
litter added to the surface fuel layer.

(2) Dead woody fuels. Insects, disease, suppression of individual trees

in young stands, and death of lower branches of trees can provide a
source of dead woody fuels. These events, and the timing of the
addition of the fuels, occur irregularly, as branch material is broken or
entire trees fall because of wind, and heavy snowfalls (Brown 1975).
Fire can kill trees and shrubs, and whatever woody material is not
consumed can become surface dead woody fuels.

(3) Duff and organic layers. Material is added to these organic layers as
the lower part of the litter layer, and moss and lichen layer, decompose.
Rotten woody material is gradually incorporated into the forest floor,
and becomes part of the duff layer.

(4) Live fuels. Shrubs, herbaceous plants, and young conifers can
establish and/or increase in volume. Branches die, and increase the
flammability of trees and shrubs.

b. Depletion.

(1) Litter. Dead plant material can oxidize and essentially disappear
during one growing season. It can remain into the next growing season,
be compacted beneath additional litter, and decompose enough to
become part of the duff layer.

(2) Dead woody fuels. Dead woody fuels physically deteriorate and
settle over time, and compactness increases as supporting branches
decay (Brown 1975). A more compact fuel bed is less well aerated, and
may dry more slowly, have a higher moisture content, and be a more
favorable environment for additional decomposition.

(3) Live fuels. Productivity of an understory layer of shrubs and

herbaceous plants can decrease significantly as the canopy closes,
resulting in a much lower annual addition of litter. A coniferous fuel
ladder can grow tall enough that its lower branches no longer provide a
bridge between surface and crown fuels.

c. Patterns of fuel succession.

(1) Forests. The generality that downed woody fuels accumulate over
time is, in many cases, not true (Brown and See 1981). The amount of
forest fuel depends on stand history, whether the stand was visited by
insects, disease, wind, and fire, and at what intervals. The size and
pattern of disturbance, and amount of fuel that results, can vary with the
event, and tree and branch mortality can be compounded by drought.
Agee (1993) also relates the amount of forest fuel to stand disturbance.
Changes in the amount of fine and coarse woody fuels over time relate
to the amount of biomass present before a disturbance, the severity of
the disturbance, and successional patterns after the stand is disturbed
(ibid.).

(a) Relationship to stand disturbance. When a wildland fire occurs in a

forested stand, the severity of the impact on the stand, and resulting
amount of surface fuels and rate of their accumulation, can vary
(Muraro 1971 in Brown 1975). For example, if a fire occurs in a
lodgepole pine stand that burns only in surface litter layers, it can kill or
weaken many of the trees but not consume much of the foliage. Surface
fuels increase moderately as trees die and fall. A fire in a lodgepole
pine stand that burns into the duff layer can consume many structural
tree roots. This makes trees susceptible to rapid blowdown, and fine
fuels are added to the stand at a much higher rate than after a lower
severity fire. A high intensity crownfire in a lodgepole pine stand can
burn off many of the fine branches in the tree crowns. If it also burns
deeply into the duff layer, most of the trees will fall fairly quickly. Most of
the fuel added would be large diameter material. Because downed trees
are not supported by small diameter branchwood, they would come into
contact with forest floor sooner and decompose more readily.

Whether the young stand of lodgepole pine that establishes after fire
has a low or high dead fuel loading also depends on the frequency of
fire. The stand that develops after a fire that caused rapid blowdown of
trees with a lot of branchwood would have a high loading of dead fuel in
all size classes. If a fire occurs in this young stand of trees, much of the
crown-stored seed could be destroyed and most of the fuel consumed.
A sparse stand of lodgepole pine could subsequently establish that has
a much lower loading of dead woody fuel than the previous stand
(Muraro 1971 in Brown 1975).

(b) Varying patterns among live and dead fuels. Fuel succession is
more complicated if live and dead fuels are involved (Brown and See
1981). There may be an increase in one class of fuel while another is
decreasing or becoming unavailable. Dead woody fuel may decompose
while an understory of trees establishes. The loading of dead woody
fuel may increase while some trees become tall enough to be much
less available to surface fire (Brown and See 1981). Early successional
herbs, such as bracken fern in western Oregon, can cause a high
loading of fine fuels before the canopy closes and shades out these
plants (Isaac 1940 in Agee 1993).

(c) Relationship to stand age. There is no clear relationship in the

northern Rockies (Brown and See 1981) between stand age and
amount of dead woody material. The amount of fuel in young and
mature forests cannot be related to age because too many other factors
are involved. The only generalities are that downed woody fuel loadings
tend to become predictably high as stands acquire old growth
characteristics, but loading is unpredictable from age alone in young,
immature, and mature stands (ibid.).

In western Oregon and Washington, stand replacing fires generally

occur at much longer intervals than they do in the northern Rocky
Mountains. Fuels in the 0 to 3-inch (to 7.6 cm) range are usually at their
highest levels in early stages of postfire succession (Agee 1993). 1000-
hour fuel biomass is highest in mid-successional stages when some
stems die because of naturally occurring self-thinning. Biomass of larger
logs is greatest in the oldest stands.

(d) Relationship to site productivity. Fuel loading and site productivity

are not well correlated (Brown and See 1981). In warm, moist forest
types, productivity is fairly high, but fuel may not accumulate because
the decomposition rate keeps up with fuel production (ibid.). In cool, dry
forest types, productivity tends to be low, but a relatively higher
proportion of biomass may accumulate as fuel because decomposition
is limited (ibid.).

(e) Relationship to fire exclusion. In many areas of the western U.S.,

naturally occurring fires used to occur at a fairly high frequency. With
the onset of organized fire suppression activities, and the removal of
fine fuels by livestock grazing, wildland fires became an infrequent
event in many forest types. If fire exclusion has removed several fire
cycles from a forested stand, the ecological effect is much more
profound than if fire has only been effectively suppressed for one-third
of the length of a stand's natural fire rotation. In parts of the
southwestern U.S., for example, the exclusion of fire from ponderosa
pine stands that previously burned at intervals ranging from 2 to 10
years has resulted in higher loadings of litter, forest floor duff, and in
some cases, down dead woody fuels (Harrington and Sackett 1990).
While the amount of fuel may not be predictable from age, it is logical to
conclude that there is more fuel in stands without understory fire for 80
to 100 years than if underburns had continued to occur at frequent
intervals.

(2) Shrublands.

(a) Sagebrush. The percentage of dead stemwood in sagebrush

(Artemisia tridentata) plants increases with age. However, when
modelling the effect of higher proportions of dead branchwood on fire
behavior, only a small increase was found (Brown 1982). The total
amount of fuel correlates to the height of the stand, but stand height
does not correlate well with age (ibid.). The amount of fuel in older
stands of sagebrush is greater if the volume and density of shrubs has
increased.

(b) Chaparral. Old stands of chaparral have been observed to be more

flammable than young stands, and this difference has been attributed to
an increasing proportion of dead branch material in older age classes of
shrubs (Paysen and Cohen 1990). No correlation was found between
the percentage of dead branch material and age of chamise in southern
California (ibid.).

Because all leaves and fine branch material in the chaparral canopy
tends to be consumed by fire when foliar moisture content is low, a
stand with more leaves and twigs has more fuel. For any given site, the
amount of biomass tends to increase with age of the stand, and it may
be this increase in total biomass that causes the higher flammability
observed in old stands. However, because of variability in site
productivity and species composition, it cannot be said that every stand
of chaparral of a certain age is more flammable than a stand that is
younger.

C. Resource Management Considerations

The primary ways to manipulate fire effects on fuels are to modify fuel
availability and to change the way an area is ignited and burned.

1. Fuel Availability.
a. Fuel moisture.

(1) Change the prescribed fuel moisture. When planning a prescribed

fire, the moisture contents specified in the prescription can be chosen to
achieve selected effects on fuels.

(a) Fine fuel moisture. Fine fuel moisture indirectly affects overall fuel
consumption by determining which fuels ignite. Fine fuel moisture is
defined by specifying different ranges of temperature and relative
humidity in the prescription.

(b) Large fuel moisture. In forested areas, the moisture content of large
diameter woody fuels is the chief factor affecting the amount of total
consumption. Remember that rotten woody material can burn at a much
higher moisture content than sound material of an equivalent size.

(c) Duff and organic layers. Consumption of soil organic material is also
directly related to its moisture content (Brown et al. 1985).

i. At moisture content greater than about 120 percent, duff will not burn.

ii. At moisture content less than about 30 percent, duff will sustain
combustion on its own once ignited.

iii. The amount of consumption of duff between 30 and 120 percent

moisture content depends on the amount of consumption of associated
fuels. (d) Live fuels. By prescribing the moisture content of live fuels in
the surface fuel layer, the amount of their flammability and consumption
is regulated. The amount of scorching of a conifer canopy may be
greater early in the growing season when trees are just becoming
physiologically active and foliar moisture content is lower than it is later
in the year. Live fuels may be consumed, regardless of their moisture
content, if a large loading of dry, dead woody material burns.

(2) Alter the fuel moisture. Use of water or foam changes the moisture
and burning characteristics of fuel. These techniques are commonly
used to build fireline and protect specific features, such as wildlife trees.

b. Fuel loading and distribution.

(1) Remove the fuels. Less fuel is available, and there is less potential
for heat release, if fuels are removed from a site. Fuels can be removed
by:

(a) Grazing.

(b) Commercial thinning of forests.

(c) Firewood sales.

(d) Yarding unmerchantable material to a central location during forest

harvesting operations.

(2) Change the fuels. If fuel distribution or arrangement is changed,

flammability, and the potential for heating, changes.

(a) Crushing. Crushing fuels increases fuel bulk density and can make
the rate of burning slower. However, if crushing compacts fuels to a
more ideal arrangement, it may enhance combustion.

(b) Lopping and scattering. Cutting and scattering of branches during a

logging or thinning operation makes fuel continuity more uniform, but
also decreases the potential for concentrated heating where piles of
branches would have been located.

(c) Piling or windrowing. Piling or windrowing fuels breaks up the

continuity and decreases the likelihood that fire can spread. A fire that
starts (or is started) in a pile or windrow has a greater potential for
subsurface heating if low moisture content of larger pieces and low
amounts of intermixed mineral soil permit a high degree of
consumption.

(d) Chaining. Chaining woodlands or shrub dominated areas alters the

distribution and continuity. If removal of the trees or shrubs allows more
grasses and forbs to grow (or if they are seeded), the flammability of the
site will be significantly higher because of the presence of downed
woody fuels.

(e) Herbicide. The use of herbicide to kill shrubs and woodland trees
results in a large amount of standing dead vegetation. Intermixture of
newly established grasses and forbs will result in a highly flammable
site.
c. Fuel chemistry. Application of long term fire retardants inhibits fuel
ignition and hence fuel consumption.

2. Ignition.

a. Backing vs. heading fires. Backing fires usually result in more

complete fuel consumption, particularly of litter and duff layers, than
heading fires.

b. Mass firing. Use of mass firing techniques, such as center firing or

concentric firing, may result in more complete consumption of fuels, for
a given moisture regime, than if a backing or heading fire were used.

c. Ignition devices. Use of ignition devices such as a heli-torch that

can apply a lot of fire in a short period of time can result in a fire that
causes more woody fuel consumption than if surface ignition were
used.

D. Methods To Monitor Fire Effects

Fuels inventory data are collected to facilitate accurate prescription

development, to determine if fuel consumption objectives are met, and
to relate fuel reduction to fire effects on other resources. Fuel moisture
data can determine whether prescribed conditions are met, and
document the conditions that correlate with specific amounts of fuel
consumption and related aspects of the heat regime of the fire. If smoke
emissions are a critical factor in a prescribed fire program, both fuel
moisture and fuel consumption data can be used to predict emissions,
refine prescriptions, and obtain an accurate estimate of the amount of
emissions produced by a particular prescribed fire. While mineral soil is
not a fuel, soil moisture data can provide important information for
documentation and interpretation of fire effects that are related to sub-
surface heating.

1. Fuel Loading. The type and amount of fuels inventory should match
the objective for

doing the inventory, because fuels data can be expensive and time
consuming to collect. Specific techniques have been developed for
inventorying or estimating living and dead biomass in forest and
rangeland vegetative types, many of which were developed specifically
for assessing fuels. The time of year when sampling is performed can
be critical if any component of live vegetation is being measured,
particularly grasses and forbs. Sampling performed before the full
amount of seasonal growth has occurred can produce serious
underestimates in fuel loading. Sampling during the normal fire season,
or during the specific time of year when a prescribed fire is planned to
occur, is recommended. Agency specific guidance for fuels
measurement in forests and in grassland and brush is provided in USDI-
NPS (1992).

a. Destructive sampling. Destructive sampling is the clipping, sorting

by size category, and weighing of all fuel in a representative area. This
is an extremely accurate way to collect fuels data but is also time
consuming and expensive. All of the other procedures detailed here
derive estimates of fuel loading from specific sets of measurements.

b. Estimating weight of herbaceous fuels. There are many

techniques for estimation of weight and production of herbaceous
rangeland vegetation because of its use as livestock forage. Most
techniques for weight estimation can be placed into one of three
categories:

1) clipping and weighing, 2) estimation, and 3) a combination of

weighing and estimation (Brown et al. 1982). Details on use of these
and related methods can be found in Hutchings and Schmautz (1969),
and Chambers and Brown (1983).

c. Estimating shrub weight.

1) Rangeland shrubs. Average height of an entire stand of big

sagebrush can be estimated by multiplying 0.8 times the average height
of the tallest plants in that stand (Brown 1982). Average sagebrush
height and percent can be converted to tons/hectare (tons/acre) (ibid.).
Martin et al. (1981) developed estimates for average loading by percent
of crown cover for big sagebrush, antelope bitterbrush (Purshia
tridentata), snowbrush ceanothus (Ceanothus velutinus), and greenleaf
manzanita (Arctostaphylos patula).

2) Forest shrubs. Shrub biomass can be estimated from basal stem

diameters for 25 species common in the northern Rocky Mountains
(Brown et al. 1982).
d. Live/dead ratio. The live/dead ratio within plants can be obtained by
ocular estimation or through more time consuming destructive sampling
techniques.

e. Inventory of dead woody fuels and duff.

1) Direct measurement. Brown et al. (1982) provides comprehensive

procedures for inventorying downed woody material, forest floor litter
and duff, herbaceous vegetation, shrubs, and small conifers. Field
sampling methods include counting and measuring diameters of
downed woody pieces that intersect vertical sampling planes,
comparing quantities of litter and herbaceous vegetation against
standard plots that are clipped and weighed, tallying shrub stems by
basal diameter classes, tallying conifers by height classes, and
measuring duff depth (ibid.). All of these procedures can be completed
at one sample point in about 15 minutes. The authors recommend that
at least 15 to 20 sample points be located in an area where fuel
estimates are desired. Although these procedures apply most
accurately in the Interior West, techniques for estimating biomass of
herbaceous vegetation, litter, and downed woody material apply
elsewhere (ibid.).

Formulas for calculating fuel loading from field measurements are found
in Brown (1974). Anderson (1978) provides graphs from which loading
can be estimated. The calculation procedures are converted into a
computer program listed in Brown et al. (1982). Agency fire
management staff may have software that can be used to calculate fuel
weights from these inventory data.

2) Photo series. A photo series developed for a specific fuel type in a

defined geographic area can be used to obtain an estimate of fuels. The
stand of interest is compared to pictures of similar stands in which fuel
inventories have been conducted. Precision is intermediate when
compared to other methods for obtaining fuels information. Photo series
are more accurate for assessing fire potential than for estimating fuel
loads (Fischer 1981a). Photo guides are available for natural and
activity fuels for coastal and interior forest types in the Pacific Northwest
(Maxwell and Ward 1976a, 1976b, 1980); for forest residues in two
Sierra Nevada conifer types (Maxwell and Ward 1979); for natural fuels
in Montana (Fischer 1981b, c, and d); for thinning slash in north Idaho
(Koski and Fischer 1979); and for natural forest residues in the southern
Cascades and northern Sierra Nevada (Blonski and Schramel 1981).
Supplementary information on fire behavior and resistance to control
were compiled for existing photo guides for Pacific Northwest coastal
forest (Sandberg and Ward 1981); for two Sierra conifer types (Ward
and Sandberg 1981a) and for Northwest ponderosa and lodgepole pine
types (Ward and Sandberg 1981b). There are presently no photo series
for the Great Basin or southwest U.S. Fischer (1981a) explains how a
photo guide is constructed with enough detail for a field office to
prepare a series on specific fuel types.

2. Woody Fuel Consumption. Fuel consumption is measured by

comparing prefire fuel loading data with data collected after a wildfire or
prescribed fire is extinguished. If a quantitative fuel reduction objective
was set, and a related fuel inventory technique selected and performed
before the fire, that same inventory must be conducted again. Changes
in fuel loading can be less precisely estimated by comparing photo
series pairs that match the appearance of the site before and after
burning.

In some cases, total downed woody fuel increases after a fire because
of the addition of branchwood and boles of trees that fell as a result of
the fire. If this has occurred, the observation should be recorded with
field data, as it will help interpret fuels data when the project is being
evaluated.

3. Litter/Duff Reduction. Techniques for measuring litter and duff

reduction are described in Chapter II.D.8., this Guide, "Burn
Severity/Depth of Burn."

4. Fuel Moisture.

a. What should be sampled. Categories of fuel moisture that can be

related to the heat regime of a fire and to fire effects include the
following:

- less than 1/4-inch diameter down dead woody fuels (1 hour fuels)

- 1/4 to 1-inch diameter down dead woody fuels (10 hour fuels)

- 1 to 3-inch diameter down dead woody fuels (100 hour fuels)

- 3 to 8-inch diameter sound down dead woody fuel (1000 hour fuels)
- large diameter rotten down dead woody fuel

- surface litter

- thin duff layer

- upper part of a deep duff layer

- lower part of a deep duff layer

- organic soils

- organic layers beneath isolated trees and shrubs

- mineral soil

- tree foliage

- shrub foliage

- herbaceous plants

Moisture data required varies with vegetation type, expected fire

behavior and fire characteristics, the fuel situation on the site, and the
objectives for conducting the fire. While not a fuel, mineral soil is
included in this list because of its role in regulating heat transfer into
soil. (See Chapter V.B.1.a, this Guide.)

b. Where fuel moisture should be sampled. The following discussion

is derived from Norum and Miller (1984), and Sackett (1981). Fuel
moisture samples should be collected within the proposed burn unit and
be representative of the area. Samples should span the range of
vegetative conditions, fuel conditions, elevation, aspect, and slope on a
prescribed fire site, because these variables can lead to notably
different fire characteristics and fire effects. Notably wet and dry
microsites should be sampled, along with the areas between them. This
also applies to shaded and exposed spots, greater and lesser
concentrations of fuel, older and younger stands, and any other within-
plot variations that might influence fuel moisture content.
If fuels inside and outside of the prescribed fire unit are notably
different, as in the case of a clearcut, fuel moisture outside the burn unit
should also be monitored and documented. Differences in anticipated
fire behavior within and outside of the intended fire area help to
determine the probability of a spot starting a fire outside of the unit and
the needed contingency suppression forces in case the fire escapes.

c. The number of samples to collect. Prefire variability in moisture

content of fuels can be fairly high. Prefire samples can be used to
determine how many samples must be collected to guarantee the
needed sampling precision. See Chapter XI.B.1., this Handbook, for an
example of how to determine how large a sample size is required.

d. Direct sampling. Detailed discussions of fuel moisture sampling

methods and drying procedures are given in Norum and Miller (1984)
and Countryman and Dean (1979). While these two publications were
designed for specific geographic locations, the general principles
involved can be applied to other parts of the country.

(1) Containers. Commonly used containers are aluminum soil sample

cans, paint cans, nalgene bottles, and wide-mouth glass jars. Plastic
bags, even if they have a tight seal, are not suitable for sample
collection. Moisture can escape through small pores in the plastic,
especially if the sample sits for a while before processing. Moisture from
the sample can condense on the bag, and be lost when the sample is
transferred to another container for drying. Use of plastic bags for
sample collection can result in underestimation of sample moisture
content.

(2) General field procedures. Detailed procedures for collecting fuel

moisture samples in Alaska were developed by Norum and Miller
(1984). The publication contains many general procedures which can
be followed in any part of the country. Some general guidelines include:

(a) If recent rain, frost or dew have left obvious moisture on the surface
of the plants, sample moisture content may be overestimated.

(b) Material collected from living plants, leaf litter, and upper duff layers
becomes fairly stiff as it dries, and may expand, causing it to spring
from the sample containers during the drying process. Material must be
loosely packed into sample containers. Stems and leaves of live fuels
can be cut into small pieces as they are placed in the sample can.

(c) Samples must be kept cool, dry, and out of direct sunlight until they
are processed. Countryman and Dean (1979) recommend placing
samples within an ice chest until they can be brought back to the lab for
processing. Lunch coolers with a container of ice can also be used. If
samples cannot be processed immediately, refrigerate them, still
sealed, until they can be weighed.

(e) Live fuels. Guidelines for collecting specific species of plants are
given in Norum and Miller (1984), and can be adapted to other species.
Plant material sampled should consist only of living foliar material, not
dead branches, dead leaves, flowers, or fruits. A consistent manner of
sampling is most important, both for each species of plant and
throughout the growing season. Plant growth stage at the time of
sampling should be noted.

(3) Processing samples. A basic requirement for processing of fuel

moisture samples is a top-loading beam or torsion balance scale,
capable of measuring to 0.1 gram. If many samples must be processed
over the course of a field season, or several seasons, the cost of an
electronic balance may be justified because of the time saved and
accuracy that is gained.

Several different means exist for determining fuel moisture content in

the office or lab once samples have been collected.

(a) Xylene distillation. The xylene distillation method is a laboratory

procedure which produces extremely precise estimates of moisture
content for both live and dead fuels. However, this method is
comparatively expensive and takes a significant time to perform. It will
not be discussed further here.

(b) Microwave oven. Microwave ovens have been used successfully to

dry dead woody fuels (Norum and Fischer, 1980). McCreight (1981) did
not recommend use of a microwave oven for drying live fuels.

(c) Computrac®. The Computrac, Model FS-2A is a moisture analyzer

that weighs and dries a small sample and provides a moisture content
on a dry-weight basis. Material is dried in an automatically controlled
oven chamber, continuously weighed, and moisture content calculated.
Results are obtained within about 10 to 20 minutes for dead woody
fuels, or about one hour for live fuels. Fuels can be dried at 95 C. (203
F.), minimizing any loss of volatiles.

While quite accurate, a major disadvantage of the Computrac is the

very small size of the sample which can be processed, approximately 3
to 10 grams of material. In order to obtain a representative sample,
many samples must be subsequently processed. A second major
disadvantage of the Computrac is its high purchase price.

(d) Drying ovens. Detailed procedures for use of a scale and drying
oven can be found in Countryman and Dean (1979) and Norum and
Miller (1984). Processing of fuel moisture samples in a drying oven has
long been the standard for measurement of fuel moisture content.
Samples are weighed on a scale to the nearest 0.1 gram, dried in the
oven, and then weighed again to determine the amount of water lost.
Ovens are customarily set to 100 C. (212 F.) for dead woody fuels, and
80 C. (176 F.) for live fuels. The standard drying time is 24 hours. Major
advantages of a drying oven are that many samples can be processed
simultaneously, and accurate values are obtained if proper procedures
are followed. The disadvantage is the 24 hour delay in arriving at the
values for moisture content.

e. Fuel moisture meters. Several brands of fuel moisture meters are

presently available that provide a direct measurement of fuel moisture.
Most meters work by measuring the electrical resistance between two
probes which are inserted into a piece of wood. Most of these meters
were developed for testing the moisture content of kiln dried lumber and
are most accurate at lower moisture values. Some of these probes are
calibrated on a wet weight basis, not a dry weight basis, and will not
give answers that can be used as input to fire behavior prescriptions.
Most of the probes are less than an inch in length and cannot penetrate
deeply enough into large diameter wood to measure its moisture
content. However, a fairly accurate measurement of large fuel moisture
content can be made by cutting across the diameter of a large piece of
woody fuel and inserting the probe into the freshly exposed surface.
Because meters were developed to measure moisture content of wood,
a fairly dense substance, they cannot give an accurate reading of
moisture content within litter or duff layers, or of soil. These meters are
not suitable for live fuel moisture estimation because the probes cannot
be adequately inserted into the live fuels, and most meters do not
operate at high moisture contents.
f. Ways to estimate dead fuel moisture content.

(1) Calculation.

(a) Fine fuels. The moisture content of fine dead woody fuels can be
estimated with several different computation models. All models use
inputs which describe the environment in which the fuel is located,
temperature, relative humidity, slope, and time of year. The most simple
but marginally accurate calculation method is available in tabular form in
the course materials for S-390, Intermediate Fire Behavior, and S-590,
Fire Behavior Analyst. A more accurate estimate can be made using the
fine fuel moisture model (MOISTURE) in the BEHAVE system. (See
XII.C.1.B.)

(b) Large diameter downed fuels. There is a regionally specific model

that accurately predicts the moisture content for large diameter dead
woody fuels, the ADJ-Th (Adjusted Thousand Hour) model developed
by Ottmar and Sandberg (1985). This model applies to 3 to 9-inch
diameter Douglas-fir and western hemlock logging slash in western
Washington and Oregon.

(2) Fuel sticks. A standard set of fuel moisture indicator sticks consists
of four, 1/2 inch diameter ponderosa pine sapwood dowels spaced one-
quarter inch apart on two 3/16-inch- diameter hardwood pins. They do
not measure any specific fuel but rather "measure the net effect of
climatic factors affecting flammability" (Davis 1959). When completely
dry, the sticks weigh 100 grams. Their moisture content can be
obtained by weighing them, using any of several types of commonly
available scales. Procedures for use of fuel sticks are described in
detail in Finklin and Fischer (1990).

Fuel sticks have important limitations. The differing density of the wood
of which the sticks are made can cause dowels made from the same
board to give different fuel moisture values when exposed to the same
environment. Response characteristics of the sticks can change
significantly with continued exposure and wood aging. A fuel stick
should be discarded after one season's use, and more often if rapid
weathering or checking has occurred. A fuel stick must be exposed at
least five days before moisture readings will be accurate. Because of
the variation in fine fuel moisture content caused by microsite
differences, use of only one set of fuel sticks to represent moisture
conditions for a prescribed fire may give a very inaccurate estimate.

E. Summary

Fuels are an integral part of most wildlands. At some time after death,
or while still alive, all vegetation becomes potential fuel. The single most
important factor controlling the flammability and consumption of fuels is
their moisture content. The moisture content of dead wildland fuels is
regulated by environmental factors, while that of living plants is largely
controlled by physiological processes. Other fuel properties can also
affect the degree of consumption. All direct effects of fire result from the
characteristics of the heat regime of the fire, which is controlled by the
manner in which fuels burn. Management of fuels is important because
by doing so, the heat regime of a fire is also regulated.

1. 0.63 approximates the value 1 minus 1/e, where e is the base for
natural logarithms (Schroeder and Buck 1970). This value is used to
describe fuel moisture relationships because the shape of the drying
and wetting curve as a function of time is approximately logarithmic.

2. The presence of unweathered organic coatings that limited vapor

movement in and out of the most recently cast needle litter was another
likely cause of the slow moisture response (Hartford and Rothermel
1991).

3. Conifers of the Larix genera (larches and tamarack) have deciduous

needles, and their moisture content will not be discussed here.

This page was last modified 05/31/01

|Disclaimer| | Privacy| | Copyright| |USFWS Main Page| |Webmaster|

 National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 05/31/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home CHAPTER IV - AIR QUALITY

Preface
Objectives By Larry Mahaffey and Melanie Miller
Fire Behavior
Fuels
A. Introduction
Air Quality
Soils & Water
Plants Wildland fires produce smoke, an air pollutant. Smoke that is a result of
Wildlife human activities is subject to legal restrictions imposed by state,
Cultural Res. Federal, and local governments. Prescribed fire is a planned event, and
Grazing Mgmt. Federal land managers have a mandate to manage prescribed fire
Evaluation smoke. Land managers must have a clear understanding of the
Data Analysis regulations and processes that must be complied with to manage
Computer smoke. The liability for downwind effects is the responsibility of the
Soft. prescribed fire managers who produced the smoke.
Glossary
Bibliography The National Environmental Policy Act (NEPA) is the law that
Contributions establishes fundamental environmental policy for the U.S. and provides
the process for considering the full range of impacts in planning land
use activities. Agencies have the responsibility to disclose possible air
pollution impacts from land management projects. The Federal land
manager is required to conduct NEPA analysis if the project includes a
"significant" amount of burning, may have impacts on sensitive vistas or
visibility, or is located near a public roadway.

The 1977 Clean Air Act (CAA) mandates the protection of human health
and the prevention of significant deterioration of air quality, and
establishes acceptable levels of emissions. States are charged with the
responsibility for protecting air quality. States write State
Implementation Plans (SIPs) to interpret and enforce the Clean Air Act,
including the identification of Designated Areas (DA), principal
population centers or other areas requiring protection of air quality.
Designated Class I Areas include specific National Parks, Wilderness
Areas, and certain Indian reservations. A goal for Federal Class I Areas
is to prevent any future impairment of visibility and remedy any existing
impairment of visibility that results from human-caused air pollution. The
1990 Amendments to the Clean Air Act specify that individual States
must consider smoke from wildland fires in their SIPs. Requirements for
prescribed fires can be established by States in the SIP that are more
stringent than those in the CAA.

Anyone using prescribed fire must consider smoke management.

Smoke management requirements and procedures vary because of the
different amounts of fuel burned, fuel type, topography, meteorology,
and presence of smoke sensitive areas (Mathews et al. 1985). The
following questions can help a manager determine the level of smoke
management that is needed, and whether an increased emphasis on
smoke management is required.

1. Is the public informed of agency resource objectives for using

prescribed fire?

2. Is the amount of acreage treated with prescribed fire predicted to

change significantly?

3. Does topography or meteorology cause poor smoke dispersion?

4. Will prescribed fires cause or contribute to increased levels of air

pollutants?

5. Will smoke from prescribed fires result in public health and safety
problems or complaints?

6. Are Best Available Control Measures (BACM) used to manage

prescribed fire smoke?

The effects of smoke on the airshed and the public and the
opportunities to reduce these impacts will be discussed in this chapter.
Managers have the responsibility to do the best job possible to control
and mitigate the impacts of smoke that result from their actions or
treatments.

B. Principles and Processes of Fire Effects

1. Combustion Process.

a. Chemistry of combustion. The following is a summary of

information provided by Byram (1959). Wood is a chemically complex
substance, composed primarily of cellulose and lignin, of which carbon,
hydrogen, and oxygen are the primary constituents. When wood burns
in a completely efficient manner, it combines with atmospheric oxygen,
and produces water, carbon dioxide, and energy. Some of the water
that results from combustion is evaporated from the fuel, but a larger
proportion is a product of the chemical reaction.

b. Phases of combustion. The four phases of the combustion process

are described in Chapter II.B.2, this Guide. Some strategies for smoke
management rely on manipulation of the amount of fuel consumed in
each combustion phase. The types of emissions and factors regulating
their production will be discussed with respect to the phases of
combustion. For a more complete discussion of the phases of
combustion, see Sandberg et al. (1978).

(1) Pre-ignition phase. During the pre-ignition phase, gases, vapors,

tars and charcoal are produced. The proportions and amounts vary
widely according to the conditions under which pyrolysis occurs. If rapid
heating occurs during pyrolysis, less charcoal, a lot of tar, and highly
flammable gases are produced. Slow heating during the pyrolysis
process results in the production of more charcoal, little tar, and lower
amounts of flammable gases (Sandberg et al. 1978).

(2) Flaming phase. The following is from Ryan and MacMahon (1976 in
Sandberg et al. 1978). The principal chemical by-products of flaming
combustion are carbon dioxide and water. However, some pyrolyzed
substances cool and condense without passing through the flaming
zone. Other substances are only partially oxidized as they pass through
the flames, and many combustion by products are produced. Low
molecular weight organic compounds may remain as gases and are
dispersed by wind. Tar droplets and particles of soot result from the
cooling and condensation of compounds with higher molecular weights.
Visible smoke consists of mostly tar, soot, and water vapor.

(3) Smoldering phase. The lower temperatures of the smoldering

phase allow some gases to condense as visible smoke. Smoldering
fires produce at least twice the emissions of flaming fires. Heat release
is inadequate to loft the smoke as a convection column, so smoke stays
near the ground and may persist in relatively high concentrations. Most
of the smoke produced consists of tar droplets less than one micron in
size (Johansen et al. 1985).

(4) Glowing phase. During the glowing phase, combustion is fairly

efficient. Carbon monoxide and carbon dioxide are released, but no
visible smoke is formed (ibid.).

c. Combustion efficiency. If combustion of fuels in wildland fires was

100 percent efficient, the burning of one ton of wood would release
3,670 pounds (1,665 kilograms) of carbon dioxide and 1,080 pounds
(490 kilograms) of water (Sandberg and Dost 1990). All of the carbon in
the fuel would oxidize to carbon dioxide. However, the combustion of
fuels in wildland fires is not a completely efficient process. The most
important reason for incomplete combustion is that wind cannot deliver
enough oxygen to the combustion zone to mix efficiently with all of the
flammable gases produced (Ryan and McMahon 1976 in Sandberg et
al. 1978). There are differences between heading fires and backing fires
in the proportion of time spent in the different combustion phases listed
above (ibid.).

(1) Heading fires. A heading fire is one in which the flaming front
moves ahead rapidly. The fire may be pushed by the wind, move
upslope, or be influenced by both factors. These fires burn with
relatively high fireline intensity, moving quickly from one fuel element to
another. The main combustion zone moves before most fuel elements
are completely consumed by fire. The flames continue ahead, leaving
behind a large area of smoldering fuel (ibid.).

(2) Backing fires. A backing fire burns into the wind or downslope.
Because the flames move more slowly, a higher proportion of fuel is
consumed in the flaming zone of the fire, leaving less fuel to smolder
after the flaming front has passed.

(3) Smoke production. For a given fuel bed and set of burning
conditions, a heading fire causes more total smoke production than a
backing fire. A heading fire generally results in more fuel consumed in
the smoldering phase of combustion than does a backing fire, and
smoldering fuels produce more smoke than fuels burned in flames. A
backing fire is a more efficient fire because more fuel is consumed in
flaming combustion, and less smoke production results.
d. Fuel properties that affect smoke production. Fuel properties that
affect smoke production are those that influence the phase of
combustion in which fuel consumption occurs, and the total amount of
fuel consumed. These factors are discussed more completely in
Chapter III. Fuels.

(1) Fuel particle size, arrangement, and continuity. The smaller the
size of the fuel particle, the more quickly it can ignite and be consumed.
The arrangement of fuel particles affects the amount of oxygen that
reaches them. More tightly packed fuel, such as a bed of juniper or
spruce needles, burns less efficiently, and produces more smoke than a
loosely packed fuel bed, such as one of ponderosa pine needles. Fuel
continuity is a factor because if fuel particles are too widely spaced,
sustained ignitions cannot occur; flames are unable to ignite adjacent
fuels.

(2) Fuel loading. A site with large amounts of fuel can generate more
smoke than a site with little fuel. The size class distribution of the fuel is
also important, because the proportion of fuel in each size class affects
the proportion that may be consumed in flaming versus smoldering
combustion. Smaller diameter fuels, such as loosely packed grass litter,
fine branchwood, and live moss and lichens burn almost entirely in
flames with little residual smoldering. In contrast, large diameter
downed woody fuels such as those found in logging slash are rarely
consumed in flaming combustion, and thus have higher potential to emit
large amounts of residual smoke.

(3) Fuel moisture. The moisture content of the different size classes of
fuel affects smoke production because it influences fuel availability and
combustion temperatures. Extremely dry fuels burn rapidly and
completely, while wet fuels burn slowly or not at all. Any moisture
released from the fuels absorbs some heat energy from the fire, limiting
combustion temperatures (Ryan and McMahon in Sandberg 1978). If
larger size classes of fuels have a high moisture content, most or all of
the heat released by flames will be expended evaporating water, and
little consumption of large diameter fuels occurs. Fuel moisture, its role
in combustion, and its relationship to past and present atmospheric
conditions, is discussed more completely in Chapter III, Fuels.

2. Emissions. Emission products from fires vary greatly, depending

upon the type of fuel, fireline intensity, fuel moisture, wind, and
temperature of the fire.

a. Combustion products. Hundreds of different compounds are

emitted in the smoke from wildland fires. More than 90 percent of the
mass of smoke emitted from wildland fires consists of carbon dioxide
and water. Carbon in the fuel is also converted to particulate matter,
carbon monoxide, aldehydes, and hydrocarbons, as well as complex
organic materials (Johansen et al. 1985). Nitrogen oxides and
hydrocarbons produced by the fire can react together in the presence of
sunlight and produce ozone and organic oxidants. Ozone production
occurs in the top of a smoke plumes where there is more light, and in
downwind areas where smoke is less dense (Sandberg and Dost 1990).

Because most of the adverse effects of smoke are related to the

amount of smoke produced, fire managers need to know how much
smoke is generated. The answer can be estimated from two numerical
expressions: emission factor and emission rate.

b. Emission factors. An emission factor is the mass of contaminant

emitted to the atmosphere by the burning of a specific mass of fuel, and
is expressed in pounds per ton in the English system or grams per
kilogram as the metric equivalent (Johansen et al. 1985). An emission
factor can be calculated for a single fire, or a single combustion stage of
one fire, or it can be a statistical average for a geographical area or a
set of similar fires (Sandberg and Dost 1990).

(1) Carbon dioxide. The carbon dioxide emission factor for prescribed
fires ranges from 2,200 to 3,500 pounds per ton of fuel consumed (1098
to 1747 g/kg) (Sandberg and Dost 1990). The combination of carbon in
the fuel with atmospheric oxygen during combustion results in the
production of a greater weight of carbon dioxide than the original weight
of the fuel. Carbon dioxide is a "greenhouse gas", i.e., it may have an
effect on the global radiation budget and may be a factor in potential
global climate change. However, carbon dioxide is also released when
wood and other organic matter decays. Logging removes forest fuels
from sites and can reduce the amount of carbon dioxide that would be
released if the site burned. Fire suppression is not an effective way to
mitigate this carbon dioxide release from many wildland fuels because
in most cases, suppression only postpones burning. Decomposition by
fire has occurred for millions of years in most of the vegetation types in
western and northern North America. In the absence of fire, fuels tend
to accumulate, ignition eventually occurs, and more carbon dioxide may
be released than would have occurred under a natural fire regime. More
fuel may be present, and fuel consumption may be more complete.

(2) Particulate matter. Particulate matter is the most important

category of pollutants from wildland fire, because it reduces visibility
and can absorb and transmit harmful gases. Particles vary in size and
chemical composition, depending upon fireline intensity and the
character of the fuels. Proportionately larger particles are produced by
fires of higher fireline intensity (longer flames) than are found in low
intensity and smoldering combustion fires (Ward and Hardy 1986 in
USDA Forest Service and Johns Hopkins University 1989). Particulate
matter emission factors for forest fuel types range from 4 to 180 pounds
per ton (2 to 90 g/kg). For prescribed burning of logging slash,
particulate production ranges from 18 to 50 pounds per ton (9 to 25
g/kg) for broadcast burning and 14 to 30 pounds per ton (7 to 15 g/kg)
for piled slash. The amount of particulate released when burning
sagebrush/grass fuel types averages 45 pounds per ton (22.5 g/kg),
mixed chaparral ranges from 24 to 30 pounds per ton (12 to 15 g/kg),
and emission factors for pinyon-juniper (slashed) range from 22 to 35
pounds per ton (11 to 17.5 g/kg) (Hardy 1990). The exact amount
depends on the fuel type, the fuel arrangement, and the manner of
combustion.

Emission factors for particulate matter less than 2.5 microns in diameter
(PM2.5) range from 9 to 32 pounds per ton (4.5 to 16 g/kg) for
prescribed fires in the Pacific Northwest, averaging about 22 pounds
per ton (11 g/kg). Emission factors are highest during the inefficient
smoldering combustion stage and lowest during flaming combustion.

The amount of smoke produced depends on the total amount of fuel

consumed. For example, even though the emission factor for sagebrush
is higher than that for chaparral or pinyon-juniper, total smoke
production from burning sagebrush is often lower because the total
amount of fuel on a sagebrush site is generally less than on a chaparral
or pinyon-juniper dominated site.

(3) Other emissions. Emission factors are available for other products
of combustion such as the invisible gases. Emission factors for carbon
monoxide range from 70 pounds per ton (35 g/kg) during flaming
combustion to 800 pounds per ton (399 g/kg) for some smoldering fires.

Volatile organic compounds are a diverse class of substances

containing hydrogen, carbon, and other elements such as oxygen. They
include methane, polynuclear aromatic hydrocarbons (PAH's), and
aldehydes and related substances. PAH's are not free in the
environment as vapor, but are incorporated in fine particulates that are
respirable. Methane and aldehydes are emitted as gases. Emissions for
volatile organics vary from 4 to 50 pounds per ton (2 to 25 g/kg) of fuel
burned, about half of which is commonly methane.

c. Emission rate. An emission rate is the amount of smoke produced

per unit of time (pounds/minute or grams/second). The portion of the
total amount of combustible fuel that a fire will consume for a given set
of conditions is called available fuel, and is usually measured in tons
per acre (kg/sq m) for forest fuels, and pounds per acre (kg/ha) for
rangeland fuels. The land manager can make better estimates of
emission rates from a prescribed burn if the amount of fuel consumed in
each combustion phase is known. (See B.1.b.)

Fuel consumption rates are expressed as area burned per unit of time:
acres per minute. Combustion rates can be calculated whether line-type
ignition is used for backing or heading fires or area-type ignition is used
in natural or activity fuels.

In order to estimate the emission rate, the following variables are

required: available fuel (tons/acre), the combustion rate (acres/minute),
and the emission factor (pounds/ton). The emission rate
(pounds/minute) can be calculated by the following equation:

Emission Rate = Available Fuel x Combustion Rate x Emission

Factor

The emission rate is used as an input to models that predict air pollutant
concentrations. Such models can be used to assess the impact of
smoke on visibility sensitive areas such as highways, cities, airports,
and parks (Johansen et al. 1985).

3. Human Health Risk from Smoke. There is a growing awareness

that smoke from wildland fires can expose individuals and populations
to hazardous air pollutants. Concern is increasing over the risk to
firefighters and the general public from exposure to toxins, irritants, and
known carcinogens in smoke. A rigorous risk assessment is needed to
address this increasingly sensitive issue. Although there is a low
probability that public health is at risk, fireline workers are more likely to
be harmed. Firefighters can be exposed to high levels of lung toxins
such as aldehydes, acids, and particulates; to carcinogens such as
polycyclic aromatic hydrocarbons, formaldehyde and benzene; and to
carbon monoxide. These exposures may be at high levels for short
periods, or at low levels for weeks at a time. The amount of some
hazardous components of smoke, such as formaldehyde and respirable
particulate matter, is well correlated to the amount of carbon monoxide
(Reinhardt 1989). Relatively inexpensive devices for measurement of
carbon monoxide (CO dosimeters) may provide a practical means to
help recognize and prevent exposure of firefighters to dangerous levels
of smoke.

The most likely effects of smoke on health are the aggravation of

existing diseases or increased susceptibility to infection. Those most
susceptible to exposure to air toxins include very young children and
individuals with chronic lung disease or coronary heart disease. The
effects of smoke on human health are discussed in detail in Sandberg
and Dost (1990), Dost (1991), and the comprehensive study plan
prepared by the USDA Forest Service and Johns Hopkins University
(1989). The following discussion is summarized from these sources.

a. Criteria pollutants. The National Ambient Air Quality Standards are

a set of goals established by the Environmental Protection Agency for
acceptable levels of six air pollutants that are potentially harmful to
public health. These criteria pollutants are particulate matter, carbon
monoxide, sulfur dioxide, nitrogen dioxide, ozone, and lead.

(1) Particulate matter. The size class distribution of particles produced

by forest fires is bimodal. Most of the particles have an average
diameter of either 0.3 micrometers or greater than 10 micrometers
(USDA Forest Service and Johns Hopkins University 1989). The
proportion of particles in wood smoke less than 2.5 micrometers in
diameter ranges from 50 to 90 percent (Sandberg and Dost 1990).
Particles less than about 10 micrometers in diameter are able to
traverse the upper airways (nose and mouth) and enter the lower
airways starting with the trachea (Raabe 1984). As the particle size
decreases below 10 micrometers, increasing proportions of particles are
able to enter the trachea, and penetrate to the deeper parts of the
airways prior to deposition. It should also be noted that once such
particles reach the lower airways, it is likely that they will deposit on
surfaces in the deepest parts of the lungs, the "pulmonary" zone--that
part of the respiratory tract most sensitive to chemical injury (Morgan
1989 in Sandberg and Dost 1990).

National air quality standards assume that the components of

particulate matter are essentially the same, regardless of location and
source. However, the constituents of particulate matter vary widely. The
particulate emitted by burning of vegetation has a much different
composition and effect than that present in urban areas (Dost 1990).
Particulate matter from vegetation fires consists mainly of condensed
organic compounds. In urban areas, particulates contain compounds
that rarely occur in rural vegetation smoke, such as masonry dust, fly
ash, and asbestos. Also, other compounds present in urban air lead to
a variety of chemical reaction products not likely to be found in
association with wildland smoke. Smoke from wildland fuels is not as
environmentally damaging as urban smoke.

(2) Carbon monoxide. Carbon monoxide is a product of combustion

that is rapidly diluted at short distances from a fire and therefore poses
little or no risk to community health (Sandberg and Dost 1990).
However, carbon monoxide can be present at high enough levels near
a fire to pose a hazard to firefighters, depending upon the
concentration, duration, and level of activity of the firefighters at the time
of exposure. Carbon monoxide is a chemical asphyxiant that interferes
with oxygen transport in blood. Pilots exposed to carbon monoxide have
developed headaches, fatigue, decreased concentration and impaired
judgement. Data also suggest that long-term exposure to low levels of
carbon monoxide produce accelerated arteriosclerosis, increasing the
risk of cardiovascular diseases such as heart attack and stroke (USDA
Forest Service and Johns Hopkins University 1989).

(3) Oxides of sulfur and nitrogen. Because forest fuels contain minute
amounts of sulfur and somewhat higher levels of nitrogen, it is expected
that these criteria pollutants are formed when wildland fuel is burned.
Increased levels of oxides of sulfur have never been measured near
wildland fires. Some oxides of nitrogen form, but the amount produced
by forest burning is not significant enough to be of concern (Sandberg
and Dost 1990).

(4) Lead. While a serious problem in urban pollution, lead is not a

natural constituent of smoke from wildland fuels. It is assumed that lead
may be only a minor component of wildland fire smoke when it has
been deposited onto fuels from atmospheric sources, such as
contaminated urban air that has moved into wildland areas (Ward and
Hardy 1986 in Sandberg and Dost 1990).

(5) Ozone. Much of the following discussion is summarized from Dost

(1990). At lower altitudes, ozone is a common component of air, but at
low enough levels to have an insignificant impact on human health. At
high concentrations, ozone is a respiratory irritant, and can have a
significant impact on individuals who already have serious respiratory
impairment. The atmospheric chemistry of ozone formation is quite
complex. It can form in the presence of atmospheric hydrocarbons
generated in large quantities by the combustion of vegetation. The
photochemical reactions that create ozone occur in areas of a smoke
column that are penetrated by ultraviolet wavelengths of sunlight,
particularly in the upper part of the plume. Ozone is therefore not likely
to be a pollutant of concern to people in the immediate vicinity of a fire,
although fire crews working at high elevations may find increased levels
of this substance (USDA Forest Service and Johns Hopkins University
1989). Ozone may pose a problem in downwind areas affected by
smoke, particularly in urban areas where ozone concentrations from
other sources may already be high.

b. Non-criteria pollutants. Ambient standards have not been

specifically identified by the Environmental Protection Agency for all
components of smoke, although some of these substances can have
negative effects on human health. The following discussion, taken
largely from Dost (1990), describes the two groups of volatile organic
compounds most likely to impact human health.

(1) Aldehydes. Aldehydes are classed as irritants, and some are

potentially carcinogenic. The two aldehyde compounds found in smoke
that are most likely to pose health problems are formaldehyde and
acrolein.

(a) Formaldehyde. Formaldehyde is a very common atmospheric

contaminant, found in association with building materials, textiles,
preservatives, and medical activities. It has been measured as a by-
product of burning wood, although little is known about its production in
wildland fires. Formaldehyde is probably the most abundantly produced
aldehyde, and is likely responsible for nose, throat and eye irritation in
firefighters exposed to smoke. At higher concentrations, it may cause a
reflexive decrease in breathing rates. Formaldehyde may not only thus
contribute to mucosal irritation commonly experienced by firefighters, it
also may interfere with their ability to obtain adequate oxygen at times
when energy is most needed. Formaldehyde is rapidly transformed in
the body to formic acid, a known toxin with a very slow removal rate.
Formaldehyde also may be present in decreasing amounts in the
smoke plume downwind, being slowly removed by chemical reactions.

(b) Acrolein. Acrolein is formed by few natural processes other than

combustion. It has been measured in emissions from fireplace smoke,
and studies suggest that greater amounts are produced in inefficient
fires. Acrolein is a more potent irritant than formaldehyde. It has similar
effects as formaldehyde to the respiratory system, but may also have
severe toxic effects on cells. Individuals exposed to acrolein may have
a decreased ability to repel respiratory infections. Acrolein is degraded
by sunlight, and it is assumed that it slowly dilutes downwind with other
components of smoke in the plume. If initial concentration levels are
high, acrolein could be a significant irritant at a considerable distance
from its source.

(2) Polynuclear aromatic hydrocarbons (PAH). Polynuclear aromatic

hydrocarbons are a class of products that have been detected in wood
smoke. These benzene containing compounds are incorporated as fine
particles that are respirable. Some PAH compounds have carcinogenic
properties. The likely risk to the general public of developing cancer
because of exposure to these chemicals from prescribed fire is very
small because of the rapid dilution of these products in the smoke
plume (Sandberg and Dost 1990). However, very little is known about
potential effects on firefighters from these compounds due to a lack of
research on their production in wildland fires.

C. Resource Management Considerations

1. Control Strategies. When wildland fires occur, managers must

consider the impacts on air quality and mitigate adverse effects
whenever possible. Wildfire is evaluated through the Escaped Fire
Situation Analysis, and prescribed fire through the Environmental
Assessment and Prescribed Fire Planning process. There are strategic
and tactical measures that can limit the amounts and mitigate the
impacts of smoke from fires. The mitigation of adverse impacts can be
accomplished through the selection and implementation of an
appropriate control strategy. Managers can use these strategies to
allow the burn to take place and yet reduce the risk of adverse effects of
smoke. Clear resource management objectives and careful monitoring
and evaluation of smoke impacts are keys to successful smoke control.
Managing smoke from wild or prescribed fires requires a daily prediction
of smoke accumulations and whether they will reach unacceptable
levels. Choice of suppression strategies and tactics must include a
consideration of smoke effects on safety and visibility.

a. Avoidance. Avoidance is a strategy that considers meteorological

conditions when scheduling burns to avoid incursions of smoke into
sensitive areas. Burning should occur on days when weather conditions
allow the transport of smoke away from populated areas. Smoke may
not be such a limiting factor in sparsely populated areas, but any
downwind effects should be considered when burning. The wind
direction during both the active burning period (flaming stage) and the
smoldering period must be considered. At night, downslope winds can
carry smoke toward smoke sensitive areas, or allow valley bottoms to fill
with smoke. Residual smoke emitted during the smoldering stage is
especially critical.

b. Dilution. The dilution strategy controls the amount of emissions or

schedules the rate of burning to limit the concentration of smoke in
sensitive areas. The concentration of smoke can be reduced by diluting
it through a greater volume of air, either by scheduling during good
dispersion periods or burning at slower rates (burning narrow strips or
smaller areas). However, burning at a slower rate may mean that
burning continues into the late afternoon or evening when atmospheric
conditions may become more stable. Burn when weather systems are
unstable, but not at extremes of instability. The time of day at which
ignition occurs is also important. Consider early morning ignitions to
take advantage of weather conditions where improved mixing will occur
as atmospheric heating takes place. Avoid days with low morning
transport wind speed, less than 4 miles per hour (6.5 km/hr). Use firing
methods to rapidly build a smoke column to vent smoke up to the
transport wind and larger volumes of air. Using mass-ignition or rapid
ignition will loft the column up and away from the unit, allowing for better
dispersion and reduced emissions during the smoldering phase.
Generally, a burn early in the day encounters improving ventilation; an
evening burn encounters deteriorating ventilation.

c. Emission reduction. Emission reduction is an effective control

strategy for decreasing the amount of regional haze and avoiding
smoke intrusions into Designated Areas (DA's) (Sandberg et al. 1985).
It reduces the smoke output per unit area, and is a concept applicable
in both forest and rangeland areas. Most emission reduction techniques
are based on limiting the consumption of larger fuels and soil organic
layers. Large fuel consumption can be reduced by physically removing
or scattering the larger fuels or burning when the larger fuels and duff
are too wet to carry fire. Burning when the larger fuels or duff are wet
will produce fewer emissions and allow rapid extinguishment of the fire.
When windrowing and piling debris, allow fuels to dry before piling, and
avoid mixing dirt into the pile. Emissions can also be reduced by use of
a backing fire that results in more fuel consumption in the flaming stage,
producing less smoke (Sandberg and Peterson 1987).

2. Techniques to Minimize Smoke Production and Impacts. Some

smoke management techniques have application to both wildfire and
prescribed fire situations, while others apply specifically to prescribed
fire management.

a. All fire situations.

(1) Be sure that each burning operation has clear and concise
management objectives that consider the impacts of smoke.

(2) Ensure that burn prescriptions and ignition plans provide for optimal
smoke dispersion for the specific circumstances of the fire.

(3) Use the best weather data available to ensure adequate smoke
dispersal. This includes obtaining spot weather and transport wind
forecasts from the National Weather Service, taking weather at the burn
site for several days prior to ignition, and validating the fire prescription
and spot forecast with onsite weather observations. Wind speed and
direction over the area can be checked by release of a helium balloon,
or by observing the smoke from a test fire.

(4) Burn when conditions allow rapid dispersion. The atmosphere

should be unstable so smoke will rise and dissipate, but not so unstable
that control problems result.

(5) Burn when fuel moistures are higher and consume only those fuels
that are specified in the treatment objectives. Higher duff moisture
shortens the smoldering phase, thereby reducing residual smoke and
particulate production.

(6) Mass ignition allows burning to occur with higher fuel moistures.
Higher temperatures generated by mass fire cause smoke to rise to a
greater height above terrain than if a line ignition is used.

(7) Use a backing fire. The slow rate of spread and long residence time
result in a higher fraction of fuel consumption in the flaming stage of
combustion rather than in the smoldering stage. Since total smoke
production per unit of fuel burned is considerably less during flaming
combustion, backing fires favor lower total smoke production.

(8) The volume of smoke in a geographic area must be considered

when making management decisions about prescribed burns,
prescribed natural fires, or wildfires.

b. Prescribed fire.

(1) Burn other than in the "traditional" late summer and fall season. The
impact on the air resource can be spread over a longer period, thereby
reducing the possibility of a heavy smoke load on a particular day. Be
careful of night burns because predicting smoke drift is more difficult,
although night burning can be successful if properly planned and
implemented.

(2) Burn fuel concentrations, piles, landings, and jackpots outside of the
prescribed burning season. This increases the number of units that can
be burned without overloading the airshed on days with good dispersal
conditions.

(3) Public criticism of a burn program can be decreased by limiting its

impact on recreational users. Avoid burning on days when smoke may
affect Class I Areas and heavily visited recreational areas, or on
holidays when many visitors may be using public lands.

(4) Using prescribed natural fire requires close monitoring of fuel

loadings, fuel moistures, normal weather patterns, and down wind
receptors in the area that may be affected by smoke drift.

(5) For prescribed natural fires, daily certification that the fire remains in
prescription must include an assessment of smoke dispersal.

3. Participation in State and Local Smoke Management Programs.

State and some local air quality agencies have mandatory smoke
management programs. Programs are tailored to the needs of local and
regional prescribed fire managers, while working to minimize adverse
impacts of smoke.

a. Comply with air pollution and smoke management regulations. Know

the regulations for your State and local area when developing the
prescription. Details on State and local laws and regulations can be
obtained from agency fire management or air quality staff.

b. Be pro-active in protecting air quality. Take part in the development

(or update) of the State Implementation Plan that contains rules that
govern prescribed burning. Working with State and local air quality
agencies provides an opportunity for field input and some control over
the future of prescribed fire.

D. Methods to Monitor Fire Effects

This section suggests methods for monitoring smoke effects that are
practical for management purposes. Although there may be few State
regulations that require monitoring of prescribed fire smoke, there are
stewardship principles and ethical reasons that make monitoring a
compelling aspect of a smoke management policy. As the first step,
managers must develop and maintain an awareness of air quality
monitoring techniques. Monitoring allows the evaluation of program
adequacy and the effectiveness of communication with local air quality
personnel. Implementation of air quality monitoring does not require
having an elaborate array of monitoring instruments or hiring a
monitoring contractor to evaluate fires.

While no readily available operational smoke monitoring techniques

accurately predict the effect of a specific fire on air quality,
understanding principles of air quality monitoring can result in better
smoke management decisions. Some states are currently charging fees
for burning, such as one fee to register each acre and an additional fee
to burn the acre. This money is used to support the smoke
management program and provide monitoring services for agencies
doing prescribed burning. Local fire management officers should
determine the proper level of monitoring and incorporate it into the burn
plan. They should develop an objective method to monitor and evaluate
the effectiveness of their smoke management efforts. Monitoring
practices can range from simple to very complex programs as
determined by managers or by the states. Agency specific guidance
that identifies smoke management monitoring techniques and
frequencies is provided in USDI-NPS (1992). The following are some
practical procedures for monitoring and modelling smoke.

1. Visual Techniques.

a. Visual estimation. Visual estimation is the most common smoke

monitoring method in use. Although most visual methods are subjective
and limited, they are still very useful. When burning near smoke
sensitive areas, a spotter on a hill away from the fire can watch where
the smoke goes and relay information to the Burn Boss.

b. Aircraft tracking. Aircraft tracking of smoke plumes can be used to

verify the source and trajectory of the smoke. It is used by some
regulatory agencies to detect violations of air quality/smoke
management regulations. This procedure provides a means to observe
the loading of the airshed and to determine if additional burning should
be limited.

2. Instrumentation.

a. Nephelometer. A nephelometer is an electronic device that

measures the amount of particulate in a sample of air. This optical
device measures the amount of light reflected from particles in the
enclosed sample space. A nephelometer can be useful for safety
monitoring, such as by measuring the amount of smoke on a highway.
The machine could be programmed to flash lights as a warning when
visibility is poor.

b. Filter sampler. Filter samplers draw a known volume of air through a

filter. The filter is weighed before and after the sampling period, and the
weight of particulate per volume of air can be calculated.

3. Computer Models.

a. SASEM. The Simple Approach Smoke Estimation Model (SASEM) is

a screening model developed by the Bureau of Land Management and
approved by the States of Wyoming and Arizona for estimating smoke
impacts from prescribed fires. This model calculates emissions, and
uses the emission figure to calculate down-wind concentrations of
particulates. Estimated particulate loadings are compared quantitatively
against ambient air quality standards to see if standards may be
exceeded. (See Chapter XII.E.1., this Guide.)

b. TAPAS. The Topographic Air Pollution Analysis System (TAPAS) is

a system of models for predicting the dispersion of air pollution over flat
or mountainous terrain. Data on topography, wind speed, and direction
are used to model plume direction and speed of smoke. Documentation
and more information is available from the air quality staff, National
Biological Survey, Environmental Science and Technology Center, Fort
Collins, Colorado.

c. TSARS. The Tiered Smoke/Air Resources System (TSARS) is a

group of computer programs that allows smoke management to be
performed in a series of increasingly advanced levels of proficiency
(Riebau et al. 1991). Tools with varying degrees of sophistication are
available to model smoke production from wildland fires, producing
results with different degrees of resolution. The level of analysis
conducted can be matched to the complexity of the problem, or the
expertise of the person using the model. Simple to use tools which
produce easily interpreted results can be used at field levels, while
more central offices in an agency would have access to more elaborate
techniques. The components of the TSARS system were all derived
from existing models, but have been modified for fires as the emission
source and to have a consistent appearance to the user. See Chapter
XII.E.2. for a more complete discussion of TSARS.

d. PUFF. PUFF is a plume trajectory model for multiple fires being

developed for the Pacific Northwest. PUFF uses input on emission
production, and models smoke dispersion for a specific grid of
atmospheric temperature and pressure. Atmospheric conditions are
derived from a National Weather Service model. These data can be
input automatically from a computer that is operated by a private
contractor. For more information, contact the U. S. Forest Service,
Global Environmental Protection Project, Forestry Sciences Laboratory,
Seattle, Washington.

E. Summary

The effects of smoke on health, air quality, and regional haze is very
important to all land managers. They must recognize the need to
manage smoke from wildland fires using the Best Available Control
Measures. Every manager must determine the level of smoke
management necessary to provide the least impact on the public, both
in terms of health and visibility. The effects of smoke on firefighters also
must be considered when managing wildland fires. If federal agencies
do not take a rational, voluntary approach to smoke management, a
mandatory approach may be provided that makes it more difficult to
meet resource management goals and objectives.

This page was last modified 05/31/01

|Disclaimer| | Privacy| | Copyright| |USFWS Main Page| |Webmaster|

 National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 06/20/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home CHAPTER V - SOILS, WATER, AND WATERSHEDS

Preface
Objectives By Dr. Bob Clark
Fire Behavior
Fuels
A. Introduction
Air Quality
Soils & Water
Plants Fire, either wild or prescribed, may have a wide range of effects on the soils,
Wildlife water, and watershed resources of forestlands, shrublands, grasslands, and
Cultural Res. wetlands. The wide range of effects is due to the inherent preburn variability in
Grazing Mgmt. these resources, and to fire behavior characteristics, season of burning, and
Evaluation prefire and postfire environmental conditions such as timing, amount, and duration
Data Analysis of rainfall. Further, effects of fire on some soils and watersheds are poorly
Computer documented and poorly understood. Thus, generalization about effects of fire on
Soft. these resources is likely to be more risky and overstated than with any other
Glossary managed resource.
Bibliography
Contributions This chapter, in keeping with the theme of this Guide, will discuss principles and
processes that contribute to the effects of fire on soils, water, and watersheds.
Where applicable, the opportunity to manage or influence these processes will be
discussed. Also, methods for monitoring the effects of fire on these resources will
be described, recognizing fire management investigation and monitoring needs for
management purposes are not normally required to the degree and precision that
are required at the research level. Finally, other factors are critical to "correct"
prediction and evaluation of fire effects on soils and water; Chapter III.D. in this
Guide is particularly relevant.

Excellent texts that provide basic information on soil taxonomy, properties,

physics, and hydrology include those written by Soil Survey Staff (1975), Pritchett
(1979), Hanks and Ashcroft (1980), and Branson et al. (1981). The ecology of the
soil nitrogen cycle is well described in (Sprent 1987). Soil descriptions used in this
chapter follow the nomenclature of the Soil Survey Staff (1975) unless otherwise
noted. References to litter layers are described in Youngberg (1981). Important
terms are identified in the Glossary.

B. Principles and Processes of Fire Effects

1. Soils. Most information about the effects of fire on soils is from forested land
and chaparral; also, much information is predicated on the effects of wildfire, not
prescribed fire. By extrapolation, this situation has frequently led to the conclusion
that fire is always detrimental to soils, including shrubland and grassland soils.
However, in a history of fire research, Schiff (1962) indicated that researchers
began documenting about five decades ago that in addition to negative effects, fire
occasionally had beneficial effects on soil, and often had no measurable effect;
further, the negative effects often were short-lived. These data are not meant to
imply that the effects of fire are unimportant, because any negative effect, however
small, can have substantial postfire consequences. The effects of fire on the soil
resource are induced by soil heating, by removal of the protective cover of
vegetation, litter and duff, or by the concentration of plant material substances in
the soil. These effects are described in detail in Chandler et al. (1983), Wells et al.
(1979), Wright and Bailey (1982), and other references listed in the Bibliography.

a. During the fire.

(1) Soil heating. The magnitude of the heat pulse into the soil depends on fuel
loading, fuel moisture content, fuel distribution, rate of combustion, soil texture, soil
moisture content, and other factors. The movement of heat into the soil is not only
dependent upon the peak temperature reached, but even more so upon the length
of time that the heat source is present. Because fuels are not evenly distributed
around a site, a mosaic of soil heating occurs. The highest soil temperatures are
associated with areas of greatest fuel consumption and the areas that have the
longest duration of burning. In forested areas, high subsurface soil temperatures
usually occur beneath fuel accumulations, with the highest temperatures most
likely to be found in association with consumption of large piles of dry harvest
residue or windthrow, or very thick duff layers. Because the pattern of soil heating
varies significantly around a site, with differences in both the amount and duration
of soil heating, a range of fire effects on soils can occur on one burned area.

Duff and soil moisture contents are critical regulators of subsurface heating. In a
controlled soil heating experiment, the heat load into wet duff and mineral soil
averaged 20 percent of the heat load that penetrated dry duff and mineral soil
(Frandsen and Ryan 1986). Peak temperatures were more than 1000 F (538 C)
greater where duff and soil were dry. DeBano (1977) estimated that about 8
percent of the heat generated in California chaparral fires is absorbed and
transmitted downward into the soil. In general, "lightly burned" forests will cause
maximum soil surface temperatures between 212 and 482 F (100 and 250 C) and
the temperature 0.4 to 0.8 inches (1.0 and 2.0 centimeters) below the surface will
not exceed 212 F (100 C) (Chandler et al. 1983). In "moderately burned" areas,
surface temperatures are typically in the 572 to 752 F (300 to 400 C) range and
may be between 392 and 572 F (200 and 300 C) at the 0.4 inch (1.0 centimeter)
depth (Chandler et al. 1983). A "severely burned" area may result in surface
temperatures approaching 1400 F (760 C) (Chandler et al. 1983).

In contrast, rangelands support considerably lighter fuel loadings and frequently

result in fires of shorter duration that produce less subsurface heating. Rangeland
fires typically result in soil surface temperatures from 212 to 730 F (100 to 388 C)
although extremes to 1260 F (682 C) have been reported. The highest surface
temperatures are probably associated with local accumulations of loosely arranged
litter and intense winds created by the fire (Wright and Bailey 1982). The greatest
subsurface heating likely occurs where thick, dry litter layers are consumed
beneath shrubs and isolated trees. The soil heat pulse, including both amount and
duration (DeBano 1979), is instrumental in eventual effects of fire on plants (see
Chapter VI.B.1.c., and VI.B.2.c., this Guide) and in physical, chemical, and
biological effects on soils.

Less is known about heat effects on wetland soils. Due to the high water content of
wetland soils, penetration of heat generated by a surface fire can be significantly
less than in mineral soils. Since many wetland soils are composed of significant
amounts of organic materials, and organic matter has a lower thermal diffusivity
than mineral soil, penetration of heat can be further reduced. However, organic soil
layers can become dry enough to burn. Significant amounts of heat can be
generated when organic soils burn, particularly in drought situations when the fire
burns deeply into organic layers.

(2) Postfire temperature increases. Soil temperature may increase after a fire
because of the removal of vegetative cover, consumption of fuels, thinning or
removal of the litter and/or duff layer, and the enhanced "black body" thermal
characteristics of charred material on the surface. This is of great significance in
Alaska where permafrost (permanently frozen soil) is present. The soil layer above
the permafrost thaws each summer, and is called the "active layer." Soil
temperatures usually increase after a fire because fire removes the overstory
vegetation, blackens the surface, and consumes some of the layer of moss,
lichens, and semi-decomposed organic matter that insulated the soil from summer
warmth. Soil temperatures were 9 to 11 F (5 to 6 C) greater at depths of 4 to 20
inches (10 to 51 centimeters) after fire in a black spruce/feathermoss stand in
interior Alaska (Viereck and Foote 1979). Eight years after this fire, the depth of
the active layer had increased from about 18 inches to 72 inches (46 to 183
centimeters) (Viereck and Schandelmeier 1980). The depth of the active layer
eventually stabilizes, and then decreases to its original thickness. The length of
time before this occurs depends upon the rate at which new vegetation grows and
shades the soil surface, and how long it takes for a soil organic layer to develop
that has the same insulating properties as the organic layer that was removed by
the fire.

Under similar moisture regimes, warmer soils increase the rate of decomposition,
and nutrient availability to postfire vegetation. Within physiological limits, higher
soil temperatures also improve growing conditions for plants. In Alaska, deeper
annual soil thawing increases the depth of soil available for rooting. This makes
additional nutrients, especially nitrogen, available to plants, simply because they
are not in frozen soils (Heilman 1966; 1968). Postfire vegetation productivity
generally increases significantly after fire on permafrost sites (Viereck and
Schandelmeier 1980), although the duration of this effect is undocumented.
b. Physical effects. Heating may cause changes in some physical properties of
soils, including the loss or reduction of structure, reduction of porosity, and
alteration of color. Most frequently, however, the important consequences include
the reduction of organic matter, exhibition of increased hydrophobicity
(nonwettability), and increased erosion due to the loss of protective plant and litter
cover. Organic matter and hydrophobicity are discussed below under the heading
"c. Direct chemical effects."

Erosion by wind (aeolian), water, or gravity often, but not always, increases
following fire. The severity and duration of the accelerated erosion depend on
several factors, including soil texture, slope, recovery time of protective cover, the
amount of residual litter and duff, and postburn precipitation intensity. Raindrop
splash, sheet and rill erosion, dry ravel, soil creep, and even mass wasting can
occur. In extreme cases, such as steep, chaparral sites in the San Gabriel
Mountains of southern California, erosion rates of more than 150 tons of debris per
acre have been measured after wildfires (Krammes and Osborn 1969). It is
reasonable to assume that hydrophobicity contributed to the extreme erosion rates
reported from these areas (DeBano 1979). Extreme rates, up to 165 tons per acre,
have also been reported following severe wildfires on timbered and chaparral sites
with 40 to 80 percent slopes in Arizona (Wright and Bailey 1982). More commonly,
erosion rates, even on steep slopes, range from about 23 to 52 tons per acre on
granitic, sandstone, and shale-derived soils, and 7 to 10 tons per acre on
limestone-derived soils (Wright and Bailey 1982). It is unclear from the literature
how much of the soil movement is attributable to fire, because preburn soil
movement or soil movement from unburned "control" areas is seldom reported.

Excessive erosion may not occur for several years after burning (Wright and Bailey
1982) because root systems of top-killed shrubs can maintain soil stability. Mass
wasting apparently occurs when root systems begin to decay. If this occurs, it is
reasonable to assume that rapid reestablishment of soil-stabilizing, deep-rooted
shrubs (rather than shallow rooted grasses) is critical, especially on steep slopes.
It has also been reported that coarse-textured soils are more erodible than fine-
textured soils (Wright and Bailey 1982). This may explain why little soil movement
occurs, even on steep slopes, following prescribed fires on sites with fine-textured
Mollisol soils in Wyoming and elsewhere. In Alaska, however, fine textured
permafrost soils tend to be much more erosive than coarse textured permafrost
soils. Coarse textured soils usually have a low water content, while fine textured
soils may contain as much as 50 percent ice. Postfire erosion on ice-rich
permafrost soils occurs much more frequently where firelines have been
constructed than on sites that have burned, because fires are seldom severe
enough to completely remove the organic layer (Viereck and Schandelmeier
1980).

c. Direct chemical effects. Several chemical changes in soils may occur as a

direct result of fire, including an increase in pH on some sites; the formation of
water repellant soil layers, hydrophobicity, on some sites; and reduction in organic
matter.
(1) Organic matter. The reduction of incorporated organic matter is critical if it
occurs, on arid, semi-arid, and forested sites, because organic matter is a basic
reservoir of the site nutrient (especially nitrogen) budget (Sprent 1987; Harvey et
al. 1987). Organic matter helps regulate the hydrologic cycle and the
carbon/nitrogen ratio, provides a site for nitrogen fixation by N-fixing bacteria, and
maintains soil structure porosity and the cation exchange capacity. The amount of
soil organic matter consumed by fire depends on soil moisture content, amount
and duration of heating, and amount of organic matter available for combustion or
distillation. For example, peat soils may burn extensively, whereas fire rarely
affects most rangeland soils. Similarly, saturated soils rarely loose any organic
matter whereas substantial losses may occur in dry soils.

(2) Hydrophobicity. The hydrophobicity phenomena is most common in the

chaparral soils of southern California. Although not completely understood, the
process by which hydrophobic soil layers form has been described in some detail
by DeBano (1981). Organic compounds in litter, probably aliphatic hydrocarbons,
are distilled during combustion, migrate into the soil profile, and condense on soil
particles, forming a water repellant layer. The phenomena is most severe in dry,
coarse textured (sandy) soils that are heated to 349 to 399 F (176 to 204 C). It is
least severe in wet, fine textured soils where temperatures remain below 349 F
(176 C). It also appears that high temperatures, above 550 F (288 C), destroy the
compounds. These data suggest that fires that heat soils to an intermediate range
of temperature are more likely to cause the formation of a non-wettable layer than
fires that only heat the surface of the soil, or those that cause deep penetration of
high temperatures; and that certain plant communities, such as those containing
chaparral species, are more likely to be affected.

It is important to recognize that hydrophobicity occurs naturally, in the absence of

fire, on forestlands, shrublands, grasslands, agricultural lands, and even golf
greens around the world (DeBano 1969a, 1969b, 1981). In addition to the
potentially severe problem in southern California, it has also been reported,
although less severe and of shorter duration, in every western State except
Alaska, New Mexico, and Wyoming (Branson et al. 1981, DeBano 1969a). There
are several reported "benefits" of hydrophobicity, including evaporation control and
water harvesting (DeBano 1981). One additional, unreported benefit occurred in
central Oregon where precipitation limited reestablishment of lodgepole pine
(Pinus contorta) following a severe wildfire. The presence of hydrophobic layers
beneath large burned logs channeled water to inter-log areas, providing adequate
soil moisture for seedling establishment.

(3) Acidity/alkalinity. pH is a standard measure of acidity or alkalinity, with 7.0 (i.e.,

the concentration of H+ ions is 10 7 equivalents per liter) being neutral on the pH
scale of 1 to 14. The scale is logarithmic, so that water or soil with a pH of 5 is ten
times more acidic than water or soil with a pH of 6. "Pure" water is neutral,
although "pure" rainfall may have pH values between 5.4 and 5.6 due to
absorption of CO2 that reacts to form one or more weak acids. Understanding the
pH concept allows understanding of the mechanisms by which fire alters soil pH
and thus the soil nutrient regime.

The combustion process releases bound nutrients, many in elemental or radical

form. Certain positive ions, collectively called cations, are stable at typical
combustion temperatures and remain onsite after burning. They are subsequently
washed into the soil where they exchange with H+ ions; the resulting increase in
H+ ions in solution increases the pH. Nutrient availability is related to soil acidity
(c.f., Tisdale and Nelson, 1975). Elements critical for plant growth, such as
nitrogen and phosphorus, become more available to plants after a fire that raises
the pH of an acidic soil. Fire can significantly enhance site fertility when it raises
the pH on cold, wet, acidic sites.

Fire-induced increases in soil pH are widely reported (Chandler et al. 1983, Wright
and Bailey 1982). Most cases of increased pH occurred on forest soils where the
initial pH was acidic, and a large amount of organic material burned. Increases in

nutrient availability may be highly significant. Rarely do arid or semi-arid soils,

which are typically alkaline, exhibit increased pH after burning. Those that do are
near neutral initially, may increase a few tenths of a pH unit, then return to preburn
pH levels within a year or two after burning. Little effect on the soil nutrient regime
occurs.

d. Soil biota. Soil fauna are variously affected by fire (Ahlgren 1974, Chandler et
al. 1983, Daubenmire 1968a, DeBano 1979, Mueggler 1976, Wright and Bailey
1982). Aboveground, soil-related herbivores and carnivores usually suffer drastic,
but temporary declines, and may be eliminated by "clean" fires (Wright and Bailey
1982). Sub-surface animals respond differently, depending on both amount and
degree of soil heating, the size of preburn populations, and the specific organism
in question. One study of Douglas-fir (Pseudotsuga menziesii) residue reduction
burning found that the bacteria Streptomyces were not affected by burning but
mold populations were significantly reduced. In contrast, prescribed burning in jack
pine (Pinus banksiana) resulted in greatly increased Streptomyces populations
that were still increased into the third postburn growing season. Even where
bacterial populations immediately decrease after burning, they typically increase
dramatically following the first significant postburn rainfall (Chandler et al. 1983).

Fire induced changes in the soil environment may favor one soil microorganism to
the detriment of another. Reaves (et al. 1990) reported that growth of populations
of species of Trichoderma, a soil fungus, was encouraged in soils sampled from a
ponderosa pine (Pinus ponderosa) site that had been burned by prescription. In a
laboratory study, these fungi inhibited growth of Armillaria ostoyae, one of several
species of Armillaria responsible for serious root diseases in coniferous forests
and plantations.

(1) Soil moisture content. Fire-caused mortality of soil microorganisms can be

related to the amount of moisture in the soil when a fire occurs. Nitrosomonas and
Nitrobacter, two bacteria groups critical to nitrification (Huber et al. 1977, Sprent
1987), are killed at 284 F (140 C) in "dry" soil but at 167 and 122 F (75 and 50 C),
respectively, in "wet" soil (Chandler et al. 1983). This suggests that this sensitivity
to heat may be critical to the recovery of low-nitrogen ecosystems.

Water in soil increases the rate of conductance so that elevated temperatures are
reached more quickly in surface layers, especially in coarse soils. Therefore, the
premise that soil temperature cannot exceed 212 F (100 C) until all moisture is
evaporated is academic with respect to certain organisms that have lethal
thresholds below the boiling temperature of water. It is important to note that
susceptibility to the heat pulse is usually dependent on time-temperature
interaction rather than peak temperature alone. If nitrogen fixing bacteria are a
concern, the best treatment may be to burn when soils are wet or moist because
they restrict the heat pulse to deeper soil layers.

(2) Mycorrhizae. Fire can have a significant, although indirect, effect on soil
mycorrhizae. Mycorrhizal fungi form a symbiotic relationship with roots of most
higher plant species of both forests and rangelands. The fungal strands absorb
water and nutrients (particularly phosphorus) from the soil and translocate them to
the roots of the host plant. The host plant provides photosynthetic products to the
fungi. The presence of mycorrhizae can lengthen root life and protect them against
pathogens (Harley and Smith 1983 in Perry et al. 1987), and can be critical for the
establishment of some species of trees. Most mycorrhizal roots occur in surface
soil horizons, particularly the organic soil layer, and decaying wood, especially
large diameter decomposing logs. If fire removes most of the organic matter on a
forested site, productivity may be significantly reduced for many years (Harvey et
al. 1986). If fire kills all species of plants that sustain mycorrhizal associations,
spores of these fungi may die after several years. It may then be difficult for
desired species of plants to reestablish, either by natural regeneration, planting, or
direct seeding.

An important mechanism for reintroduction of mycorrhizal fungi on burned forested

areas is dispersal by chipmunks (Tamias spp.). These animals eat fruiting bodies
of mycorrhizal fungi in adjacent unburned areas, and spread spores in burned
areas when they defecate (Maser 1978b and McIntyre 1980 in Bartels et al. 1985).
Downed logs provide important travel lanes and home sites for chipmunks.
Therefore, the presence of residual logs after a wildland fire enhances the
reestablishment of mycorrhizal fungi, both by enhancing habitat for chipmunks,
and providing suitable microsites for mycorrhizal infection and growth.

Little is known about the ecology of mycorrhizae in rangelands (Trappe 1981).

Most plants of arid and semi-arid rangelands are mycorrhizal, and many of these
same relationships may be true, particularly the association of mycorrhizae with
organic matter.

e. Soil nutrients. Nutrient changes occur during combustion. Two distinctions

germane to the discussion of nutrients include total site nutrient budgets vs. soil-
borne nutrients, and total nutrients vs. available nutrients. For example, sites with
large volumes of woody material have considerable portions of site budgets bound
in organic matter, in forms unavailable to plants. When this material burns, a large
amount of nutrients may remain on the site in ash, may leave the site in fly ash or
via overland flow, or may volatilize and leave the site in gaseous form. In any
event, if bound nutrients leave the site, the site budget decreases but the soil
reservoir may remain unchanged (Owensby and Wyrill 1973). Second, even
though part of a nutrient's budget may be removed, the remaining portion may be
converted into a different, more available form. This latter case is common with
nitrogen, which volatilizes at low temperatures; when volatilization occurs, the site
budget decreases, but usually the ammonium (NH4+) form increases. The
ammonium form is directly usable by plants. It is also converted to nitrite by
Nitrosomonas, then to nitrate, which is also directly usable by plants, by the
Nitrobacter group. The net result is that while the total amount of nitrogen on a site
decreases, the amount of available nitrogen frequently increases.

Postfire nitrogen accretion occurs by such means as fixation by heterotrophic

bacteria and symbiotic fixation by nodulated plant roots. On forested sites, many
"nitrogen fixers" such as alder (Alnus spp.) and ceanothus (Ceanothus spp.),
provide rapid recovery of nitrogen (Farnsworth et al. 1978, Raison 1979,
Rodriguez-Barrueco and Bond 1968). Bacterial fixation in decomposing wood can
also provide an important postfire nitrogen source. In contrast, on chronically
nitrogen-deficient sites, including many semi-arid rangelands (Whitford 1986),
nitrogen may be depleted by burning too frequently. Burning at intervals of less
than 5 to 8 years depleted nitrogen on tobosagrass (Hilaria mutica) sites (Sharrow
and Wright 1977).

Nitrogen is often the growth-limiting factor on many sites, and is therefore of major
interest. Sulfur also volatilizes at low temperatures and its loss also may be
important (Tiedemann 1987). Most of the remaining nutrients typically increase or
remain unchanged after burning (Chandler et al. 1983, Mueggler 1976, Wright and
Bailey 1982).

The cations released to the soil during combustion may be substantial where fire
consumes heavy fuel loads on forest sites. However, this so-called "ash effect" is
probably minimal on most rangelands. A rangeland site supporting 1,000 pounds
(454 kilograms) of completely consumed vegetation per acre that contained 1
percent calcium, would only add about 10 pounds (4.5 kilograms) of calcium
"fertilizer" per acre. Most vegetation contains about 3 to 6 percent cations.

2. Water. Wildland fire may affect both water quality and water quantity. The
effects are summarized in Chandler et al. (1983), Tiedemann et al. (1979), Wright
(1981), and Wright and Bailey (1982).

a. Water quantity. Plants, especially phreatophytes, transpire enormous

quantities of water. It follows that breaking the soil-plant-atmosphere continuum
should result in a net reduction in water loss. This concept has been applied, with
mixed results, in attempts to increase water yield from watersheds (Branson et al.
1981, Davis 1984, Sturges 1983). For example, conversion of shrublands to
grasslands has been thought to increase off-site water yield. More recently,
Hibbert (1983) suggested that such practices in areas with annual precipitation
less than about 15 to 20 inches (38 to 51 centimeters) will probably not result in
increased off-site flow. It may be difficult to increase off-site water yield by any
practical means in areas where evapotranspiration greatly exceeds precipitation.

There are no conclusive studies that clearly demonstrate that fire causes long-term
increased water yield (Settergren 1969). Temporary (for a few years) increases
may occur following large, "clean" fires because although direct evaporation may
increase, water detention by litter and debris, and transpiration, both decrease.
However, the effect is quickly reduced as vegetation and litter return.
Demonstration of the "increased yield" is difficult because the effect is often
temporally shorter than natural variation in climatic events, and because increased
evaporation from the soil surface may compensate for reduced transpiration.
There is good circumstantial evidence that greater accumulations of snow may
occur following fires that remove some tree cover because of decreased
interception of snow by the canopy. However, if the burned area exceeds about
four times the height of surrounding cover, snow accumulation may decrease due
to wind scour (Haupt 1979). In contrast, water quality may be dramatically affected
by fire.

b. Water quality. The literature is replete with evidence of fire-induced changes in

water quality, including increased sedimentation and turbidity, increased stream
temperatures, and increased concentrations of nutrients resulting from surface
runoff (Buckhouse and Gifford 1976, DeByle and Packer 1972, Feller and Kimmins
1984, Helvey et al. 1985, Nissley et al. 1980, Richter and Ralston 1982, Striffler
and Mogren 1971, Tiedemann et al. 1979, Wright et al. 1976). The implication is
clear: wild and prescribed fires, on forestlands, shrublands, and grasslands, have
the potential to decrease on and off-site water quality, and should be mitigated.
Effects may be short or long-lived. In a study on 26 to 28-inch (66 to 71
centimeters) annual precipitation rangeland with Mollisol and Inceptisol soils,
Wright et al. (1976) found that level areas were unaffected, but adverse effects
lasted for 9 to 15 months on moderate (8-20 percent ) slopes and 15 to 30 months
on steep (37 to 61 percent) slopes. Wright et al. (1976) further found that the
average sediment yield was less than 0.01 tons per acre during the first six months
after burning from the level sites but was about 10-fold greater on moderate slopes
and 100-fold greater on steep slopes.

Mesic, forested sites revegetate much more quickly, but also may be exposed to
greater, and often more intense, rainfall. In a study following a wildfire on a
ponderosa pine site in central Washington, where annual precipitation is about 23
inches (58 centimeters), Helvey et al. (1985) found that annual sediment yields
increased as much as 180-fold above prefire levels. The yields were still 12-fold
greater after seven years. A carefully controlled study on a larch (Larix
occidentalis), Douglas-fir, and Engelmann spruce (Picea engelmannii) site in
western Montana (DeByle and Packer 1972) found that sediment returned to
preburn levels after about four years. Erosion rates in the Montana study remained
below 0.01 tons per acre per year throughout the study.

Methods for mitigating accelerated sedimentation due to fire have not been fully
developed. Sedimentation may be reduced by the protection of steep slopes,
retention of wide buffers along water courses, rapid revegetation, the presence of
residual fuel and duff, and the exclusion of use until recovery.

Fire may induce sudden changes in water chemistry. Such changes probably
result from nutrients that are carried into water courses from burned areas.
Typically, several forms of nitrogen, phosphorus, and most cations show increases
in stream water after burning (Tiedemann et al. 1979). Chemistry is most often
altered during the first few storms following fire. Changes include increases in
bicarbonates, nitrates, ammonium, and organic nitrogen (Chandler et al. 1983).
These nutrients usually are not hazardous to humans but may contribute to
eutrophication or algal blooms. Water quality typically returns to preburn levels
within one to two years. Some fire retardant chemicals used during fire
suppression may be toxic to aquatic animals; the addition of these chemicals near
or in water courses should be avoided until specific consequences are clarified.

Stream temperatures also often increase after fire occurs. Usually the temperature
increase is due to the removal of overhead protective vegetation rather than direct
heat flux from the fire. Elevated stream temperatures are detrimental to most cold
water fish species. Therefore, protection of streamside vegetation, and quick
revegetation of burned areas, are critical to stream rehabilitation.

C. Resource Management Considerations

1. Expertise on Interdisciplinary (I. D.) Teams. Expertise in soils and hydrology

is required on interdisciplinary teams.

a. A soil scientist, knowledgeable about fire effects, should be assigned to

interdisciplinary teams involved with fire prescription development, site selection,
emergency fire rehabilitation projects, and wildland fire suppression activities.

b. A hydrologist should be assigned to the I.D. team, or at least consulted, if wild

or prescribed fire might affect water quality, on or off-site.

2. Statistical Analysis. Statistical analysis is necessary to assess the effects of

fire on soils and hydrology.

a. Physical and chemical characteristics of soils typically are extremely variable.

Fire effects can vary significantly around a site because of differences in the
amount of soil heating. A biometrist or statistician should be consulted before any
sampling is undertaken.
b. Adjacent, unburned "control" sites should be used for comparison with burned
sites whenever possible to evaluate the effects of fire on soils or water. A
biometrist or statistician should be consulted for appropriate sampling and
comparison methods. 3. Limited Ability to Extrapolate to Other Sites. Much of
the fire literature describes the effects on soils and hydrology of intense wildfires.
Such information should be extrapolated to different regions, soils, environmental
conditions, types of fire behavior and characteristics, and to prescribed fires with
caution.

4. Variability of Effects. Because fire effects on soils and water are highly
variable, consideration should be given to locally documenting effects and relating
the effects to fireline intensity, burn severity, fuel, duff, and soil moisture content at
the time of the fire, and other appropriate factors.

5. Factors Related to Postfire Erosion. Potential for wind, water, or gravity

(especially dry ravel) erosion should be given strong consideration in the timing
(i.e., fall vs. spring) of prescribed fires, and in the methods, timing, and species
proposed for emergency fire rehabilitation.

a. Delayed recovery of vegetation and slope steepness appear to be important

factors in accelerated erosion.

b. The presence of large woody debris and duff after a fire helps to protect the soil
from erosion.

6. Management of Soil Heating. The amount of soil heating caused by prescribed

fires in forest or woodland areas can be managed.

a. The distribution of soil heating is affected by the choice to broadcast burn, pile
burn, or burn windrows. Also, the piling method may be important because
machine piles tend to be "dirtier," and hold heat longer, than hand piles.

b. Small diameter, unmerchantable trees (whips) can be slashed just before fire,
when they are still green and will not burn well, and thus can contribute little to soil
heating.

c. Higher levels of utilization or yarding some unmerchantable material in areas

with heavy dead and down fuel loads can decrease the amount of potential soil
heating.

d. Burning an area while moisture content of large diameter fuels, lower duff, and
soil is high will limit the duration of the fire and the amount of heat penetration into
lower soil layers.

e. Rapid ignition techniques (e.g., aerial drip torch) can sometimes be used to
shorten the duration of the burn, and thus the amount of soil heating.
7. Leaving Woody Material. When prescribed burning, it is important to leave
some coarse, woody debris on the site for nutrient cycling and mycorrhizal
function. Agencies may have specific requirements for retention of downed, woody
material.

8. Riparian Areas.

a. Buffer strips. When prescribed burning, leave unburned strips of vegetation

along riparian areas to serve as slope stability buffers, and decrease the potential
for stream sedimentation. Width of buffer strips should be in accordance with
applicable agency policy.

b. Season of fire. Riparian areas should be burned, if necessary, in spring when

conditions are favorable for rapid recovery of adjacent vegetation.

c. Use of fire retardant. Use of fire retardant chemicals in or near waterways

should be avoided. Fire retardant has the greatest impact on small or slow moving
bodies of water.

d. Firelines. On erosive soils and/or steep slopes, restrict the location of firelines
that lead directly into water courses. Rehabilitate any firelines that were
constructed as soon as possible. Replacement of soil and plant material removed
during construction is an effective method of fireline rehabilitation.

9. Salvage Logging. Know the potential for soil erosion when considering or
planning salvage logging operations after wildfire.

a. Road construction may increase the amount of soil erosion and mass
movement. Also, some areas (e.g., Western Oregon) have restrictions to limit "off
road" use to minimize compaction. These restrictions may dictate the appropriate
logging method.

b. A choice may be made to helicopter log or to not log at all.

10. Need for Closures. It may be necessary to close burned areas to all types of
vehicular use, and other uses, for several years because of increased erosion
potential.

D. Methods to Monitor Fire Effects

The effects of fire on soils and water are usually extremely variable over time and
space due to variations in soil characteristics and plant communities; in the
intensity, duration, and timing of postfire precipitation; and in the heat regime of the
fire. Many methods used to monitor changes in soils or water quality are time
consuming, expensive, and often require elaborate laboratory facilities. Therefore,
methods used to monitor fire effects for day-to-day management purposes are
usually less extensive and intensive than methods used for research. This section
suggests methods for monitoring fire effects that are practical for management
purposes. A more complete understanding of soil monitoring techniques can be
gained from Black (1965) and Golterman and Clymo (1969).

1. Soil Temperature. Soil temperature, by itself, may not be particularly revealing.

However, it may add valuable insight when used in conjunction with other
information.

a. Heat sensitive paint. Temperatures can be monitored by placing chips covered

with heat sensitive paints at the duff/soil surface and at various depths below the
surface. These paints turn color at a temperature specific to each kind of paint, but
do not indicate duration of heat. However, it is likely that the presence of high
temperatures may increase the certainty of impacts. Temperature sensitive
devices and paints are available from many forestry suppliers.

b. Electronic equipment. Extremely accurate data can be obtained through use

of sophisticated electronic equipment such as thermistors and thermocouples. A
thermocouple can be wired to a strip chart recorder which provides a continuous
record of temperature. These are expensive, and can be unreliable unless properly
used. Unless research is being conducted on a prescribed fire, their use is
probably not warranted. If they are used, advice should be obtained from the
research community on appropriate composition, size, number, and placement of
thermocouples.

2. Soil Physical Properties. The primary physical effect of fire on soil is the
removal of protective cover, which allows accelerated erosion. Erosion can be
estimated using predictive models, qualitative guides, or quantitative methods. To
isolate fire effects from other effects, burned areas should be compared with
adjacent, unburned control areas. The following practical, quantitative methods are
suggested for obtaining direct estimates.

a. Erosion.

(1) An appropriate number of "depth-of-burn" pins can be randomly placed onsite

before burning to measure duff removal and/or subsequent soil movement. The
pins may be of the v-notched survey pin type (with the pin inserted such that the
notch is at duff or soil surface level), or the t-bar survey pin type (with the pin
inserted such that the cross member rests on the duff or soil surface) (McRae et
al. 1979). Bridge spikes also work well. Remeasurements over time will provide an
indication of soil movement. Frost heaving and pin disruption by animals may
produce erroneous values. Depth of soil removed can be expressed as volume or
mass by using appropriate mathematical conversions. This method of estimating
soil movement appears to work well for relatively uniform erosion of minor depth,
such as wind erosion, but does not work well for irregular soil movement, such as
gully erosion, or for mass wasting.
(2) Soil erosion bridges (Blaney and Warrington 1983, Ranger and Frank 1978)
are excellent devices for estimating the amount of soil leaving the site, but may
require special or additional equipment. Staff soil scientists, or soil scientists from
the USDA Soil Conservation Service, universities, or experiment stations should
be consulted.

(3) Soil catchments or erosion troughs (Ryan 1982, Wright et al. 1976) are used to
collect material after it leaves the site. Commonly, paired watersheds (one
member of the pair receives the treatment and the other serves as an untreated
control) are used to estimate off-site movement of soil. This method requires the
construction of catchments and may not be practical on wildfires or on most
prescribed fires.

(4) Other methods are available for estimating accelerated erosion, including
models such as the Revised Universal Soil Loss Equation (Renard et al. 1991),
radioactive markers (Lance et al. 1986), and photogrammetry (Lyon et al. 1986).
These methods are especially useful for research and special management needs,
but may not be practical in most fire situations.

b. Hydrophobicity. Hydrophobicity is most easily estimated by the water drop

penetration time method (DeBano 1981). A water drop is placed on the sample
surface and the length of time to be absorbed is monitored. DeBano suggests that
water droplets remaining longer than 5 seconds indicate water repellency. A
further refinement of this method uses the surface tension of various ethanol
solutions; a water repellency index is then obtained by dividing the critical surface
tension into the time (up to 600 seconds) required for water drop absorption. An
index value greater than 10 indicates extreme repellency, 1 to 10 indicates
moderate repellency, 0.1 to 1 indicates slight repellency, and less than 0.1
indicates wettability.

c. Other physical properties. Fire-induced changes in other physical properties

of soils, including soil structure, porosity, and rate of infiltration can be measured
using standard methods (Black 1965) for soil analysis. A soil scientist familiar with
the methods should perform these analyses.

3. Soil Chemical Properties.

a. Soil water content. Soil moisture content partially regulates the heat pulse into
the soil. Because of this importance it is discussed separately here. Several
methods of soil water determination are readily available (Roundy et al. 1983) but
the gravimetric method (Gardner 1965) is probably the most reliable and
commonly used method. An appropriate number of 1 to 100 gram samples are
collected in soil cans, weighed (wet weight), oven dried to constant weight
(normally at 212 F [100 C]), and reweighed (dry weight). Soil water content is then
calculated according to:

soil water content (percent) = (wet weight - dry weight x 100) / dry weight

Drying at temperatures greater than 212 F (100 C) can cause volatilization of soil
organic matter, resulting in a loss of materials other than water from the sample,
and an overestimation of soil moisture content.

b. Soil pH. The acidity of soil is readily determined using soil paste or aqueous soil
suspension and glass electrode pH meters (Peech 1965). Although "standards"
are used for meter calibration, it is important to concurrently analyze soil samples
from adjacent, untreated soils for comparison, because variations occur among
meters and investigators.

c. Soil conductivity. The electrical conductivity of soil caused by the presence of

soluble salts is readily determined by using a solu-bridge (Bower and Wilcox
1965). Extracted soil solution is read on the bridge, corrected for temperature, and
reported in millimho (unit of conductance). High soil salt content is an indicator that
salt-tolerant species should be planted.

d. Other chemical properties. Measurement of other chemical properties of soils

that are likely to be affected by the fire require special laboratory equipment and
procedures. These are probably beyond practicality for routine fire effects
analyses. Bureau soil scientists or soil scientists of other agencies, universities, or
experiment stations should be contacted if such analyses are necessary.
4. Water Quantity. Increases in off-site water yield due to burning are most easily
determined by measuring changes in streamflow volume before and after burning.
If available, paired watersheds should be used. Other agencies, such as the USDA
Soil Conservation Service or DOD Army Corps of Engineers, often have gaging
stations or use other methods to determine flow volumes on many streams and
rivers. In addition, these agencies and many universities and experiment stations
often have portable devices that can be used to assess changes in water yield.

5. Water Quality. Several descriptors of water quality can be estimated with

minimal investment of time, equipment, training, and personnel, and include
turbidity, conductivity, dissolved oxygen content, and temperature. Kits and
relatively inexpensive instruments are commercially available for sampling these
properties. Experiment stations, universities, and Federal and State water quality
agencies can provide assistance.

a. Sedimentation and turbidity. Sedimentation and turbidity reduce the quality of

spawning areas and reduce photosynthetic activity. These effects can be amplified
by fires that occur near water and may persist for several years or longer. Platts et
al. (1983) described general methods for sampling and evaluating stream
conditions. Specific procedures for the nephelometric turbidity estimation method
are contained in the EPA (1979) publication on water quality evaluation methods.

b. Conductivity. Changes in the specific conductance due to increased ionic

composition of the water may provide a useful estimate of the addition of nutrients
or contaminants to water. Specific procedures for the use of conductivity meters
are found in EPA (1979).

c. Dissolved oxygen. A dissolved oxygen content of about 5 milligrams or more

per liter may be necessary to maintain aerobic conditions and support cold water
fisheries in Western streams (EPA 1976, Thurston et al. 1979). Because the
addition of sediment and nutrients following fire may reduce the oxygen content
below acceptable levels, oxygen contents can be monitored to estimate potential
impacts on fisheries. Inexpensive kits and meters for measuring dissolved oxygen
contents are readily available from chemical and forestry equipment suppliers. The
EPA (1979) described specific procedures for using the membrane electrode and
modified Winkler methods of dissolved oxygen analysis. A hydrologist, fisheries
biologist, or chemist should be contacted for recommendations of preferred
methods and equipment in specific locations.

d. Temperature. Reasonably correct estimates of temperature may be important

because water temperature, outside of some fairly narrow ranges, can dramatically
influence algal blooms, fish survival and reproduction, and a host of other
biological activities. However, precise estimates of temperatures in streams can be
difficult to obtain because such factors as diurnal variation, angle of the sun,
shading, and flow can contribute to error. These factors, as well as thermal
layering, can cause equally bad estimates in lakes and ponds. Therefore,
depending on the significance of temperature in a particular situation, a
temperature sampling scheme should be carefully designed. It should also be
noted that data obtained from monitoring and recording devices are usually more
reliable than "grab samples" obtained with hand-held mercurial thermometers.
Specific procedures for estimating the temperature of water are found in EPA
(1979).

e. Other chemical properties. Measurement of other chemical properties of

water, such as the concentrations of specific chemicals or nutrients, are probably
beyond the practical reach of most land management agencies. Such analyses
require special and expensive laboratory equipment and training. If such analyses
are necessary, a hydrologist, fisheries biologist, or chemist should be contacted,
and appropriate procedures (EPA 1979, Golterman and Clymo 1969, Hem 1970)
applied. Analyses can sometimes be completed using inexpensive soil or water
testing kits. Results from such testing are not definitive and should remain suspect
until confirmed by standard laboratory procedures.

E. Summary

The effects of fire on soils, water, and watersheds are extremely variable. In some
cases, such as accelerated erosion, the outcome is reasonably predictable and
mitigating measures such as rapid revegetation are necessary. In other cases,
such as change in off-site water yield after burning, the outcome is much less
predictable because it appears to depend on site-specific characteristics and on
unpredictable climatic events. The application of mitigating measures must be
based on local experience and local research. In almost all cases, the
establishment of a local data base would provide useful information for future
events.

This page was last modified 06/20/01

|Disclaimer| | Privacy| | Copyright| |USFWS Main Page| |Webmaster|

 National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 05/31/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home CHAPTER VI - PLANTS

Preface
Objectives By Melanie Miller and Jean Findley
Fire Behavior
Fuels
A. Introduction
Air Quality
Soils & Water
Plants This chapter discusses the interaction of fire with plants. It explains the basic principles
Wildlife and processes that determine how plants are affected by fire, and the factors that
Cultural Res. control plant response after the fire. Documentation of burning conditions and fire
Grazing characteristics provides important information for understanding postfire vegetative
Mgmt. response. Use of appropriate techniques when monitoring specific effects of fire on
Evaluation vegetation is necessary to detect changes that occur in the postfire plant community.
Data Analysis The goal is to enable managers to predict fire effects on plants based upon knowledge
Computer of burning conditions and prefire species and community characteristics, and to
Soft. interpret the causes for observed variability in postfire vegetative response.
Glossary
Bibliography The response of plants to fire can vary significantly among fires and on different areas
Contributions of the same fire. Both variability in the heat regime of the fire and differences in plant
species' abilities to respond affect the postfire outcome. Fireline intensity, burn severity,
total duration of combustion, soil heating, time of the year, and time since the last fire
all influence mortality or survival of the plants, and thus subsequent recovery. Postfire
effects also depend upon the characteristics of the plant species on the site, their ability
to resist the heat of a fire, and the mechanisms by which they recover after fire. Plant
recovery can be affected by factors that vary with growing season, or age of the plant.
Whether the plants that first appear after a fire successfully establish on the site can be
influenced by external factors such as postfire weather, postfire animal use, and plant
competition.

The inherent abilities of plants to respond to fire depend partially on the fire regime to
which the plant community has adapted. For example, a community may
characteristically have been subject to frequent, low intensity, low severity understory
fires, or the site may have experienced infrequent high intensity fires that killed all
standing vegetation. Knowing the "natural" role of fire on a site gives an indication of
the type of plant adaptations to fire that may be present.

The most significant sources of heat from most fires are downed dead surface fuels,
litter, and duff layers. However, dead branches, leaves, or needles within a plant itself
can produce a considerable amount of heat. Old decadent stands of shrubs may
produce a more intense fire than a young shrub stand, which may have little dead or
dry material and cannot be ignited. The amount of dead woody fuel, thickness of litter
and duff layers, and amount of dead material within or around a living plant may be
greater than "natural" if fire has been excluded from an environment in which fires used
to occur at a moderate to high frequency. In this situation, the impact of fire on the
vegetation may be different than it would have been under natural conditions because
of the potentially higher temperatures and longer duration of fire that can occur.

B. Principles and Processes of Fire Effects

1. Plant Mortality. Fire-related plant tissue mortality is dependent upon both the
temperature reached and the duration of time it is exposed to that temperature. The
lowest temperature at which plant cells die is between about 50 to 55 C (122 to 131 F)
(Baker 1929 in Wright and Bailey 1982). Some plant tissue may be able to withstand an
exposure to 60 C (140 F) for a few seconds, but dies if exposed for about 1 minute.
Plant tissue can sustain higher temperatures for greatly decreasing periods of time.
Douglas-fir (Pseudotsuga menziesii) needles can tolerate temperatures of 70 C (158 F)
for only about 0.01 second (Silen 1960 in Martin 1963). Additionally, some plant
tissues, particularly growing points (meristems or buds) tend to be much more sensitive
to fire heat when they are actively growing and their tissue moisture is high, than when
tissue moisture content is low (Wright and Bailey 1982). Thus, plant tissues more
readily die after exposure to a specific temperature for a certain length of time when
actively growing than when they are physiologically dormant or quiescent, or have
finished active growth for the year. Susceptible plant tissue may not be directly
exposed to fire heat, because it is protected by other tissues such as bark or bud
scales, or is buried in duff or soil. Plant mortality depends on percentage of tissue
killed, location of dead tissue, reproductive mechanisms, and species ability to recover
from injury.

a. Crown mortality. Both structural and physical characteristics affect the likelihood
that the aboveground part of a woody plant will be killed by a given fire. Important
crown characteristics include branch density, ratio of live to dead crown material,
location of the base of the crown with respect to surface fuels, and total crown size
(Brown and Davis 1973). Small size buds are more likely to be lethally heated because
of their small mass. Large buds, such as on some of the pines, are more heat resistant.
For conifers, long needles provide more initial protection to buds than short needles
that leave the bud directly exposed to fire heat (Wagener 1961 in Ryan 1982).

The moisture content of new needles, leaves, and small twigs, the foliar moisture
content, fluctuates throughout the growing season. It is highest during the period of
active leaf formation and shoot elongation (greenup), subsequently declines to a lower
level during the remainder of the growing season, and drops again when foliage cures
(Norum and Miller 1984). For conifers, the moisture content of new foliage follows the
above pattern, while moisture levels in older needles drop in the spring, and rise again
in late spring and early summer (Gary 1971; Chrosciewicz 1986). Moisture content
influences foliar flammability because leaves and twigs containing more water require a
greater amount of heat to raise them to ignition temperature. Coniferous tree crowns
seem to be more susceptible to crown damage in the spring than they are in the fall
because tissue moisture of new growth is highest at about the same time the moisture
content of old foliage is near its seasonal low and more flammable. The foliage of some
shrubs, particularly those with evergreen leaves, contains flammable compounds that
allow foliage to burn more readily than if these compounds were not present
(Countryman and Philpot 1970; Shafizadeh et al. 1977).

The scorching of a tree crown is primarily caused by peak temperatures associated

with the passage of the flaming fire front (Wade 1986). The height above the surface to
which crowns are scorched, (crown scorch height), can be estimated from flame length,
an output of the Fire Behavior Prediction System, ambient air temperature and
windspeed (Albini 1976). (See XII.D.1.a., this Guide, for a more complete discussion of
how this model works.) Long-term heating caused by burnout of fuel concentrations
after the flaming front has passed can also scorch crowns. The percent of crown
volume with scorched foliage is a better indicator of fire impact than crown scorch
height because it considers the proportion of live foliage remaining (Peterson 1985).
For conifers with short needles, and trees and shrubs with small buds, crown scorch is
often equivalent to crown death because of lack of protection afforded the buds (Wade
1986), and the low heat resistance of small buds and twigs.

Crown consumption is the result of the ignition of needles, leaves, and twigs. Needle
ignition occurs at about 400F (220C) (Wade and Johansen 1986). For fire resistant
conifers with long needles, such as ponderosa pine (Pinus ponderosa), and/or large or
well protected buds that are buried in wood such as western larch (Larix occidentalis),
crown consumption is often a better indicator of crown mortality than crown scorch. For
these species, bud and twig death generally only occurs where foliage is consumed by
fire (ibid.).

b. Stem mortality. Trees and shrubs can be killed by lethally heating the cambium, the
active growth layer that lies beneath the bark. Bark surface texture can affect its
likelihood of ignition, whether stringy and flammable, or smooth. Fire resistance of tree
stems is most closely related to bark thickness, which varies by species and with age.
The cambium layer of thin barked trees such as lodgepole pine (Pinus contorta),
subalpine fir (Abies lasiocarpa), aspen (Populus tremuloides), madrone (Arbutus
menziesii), or most of the spruces (Picea spp.) is usually dead beneath any charred
bark. Heat released in the flaming front, and hence flame length, can be a good
indicator of the amount of injury sustained, and even the mortality of thin barked
species. External char is not a good indicator of cambium damage on thick barked
trees such as ponderosa and Jeffrey pine (Pinus jeffreyi), Douglas-fir, or western larch
(Ryan 1982). The cambium beneath thick bark layers is usually only killed by heat
released over a long duration, such as from burnout of logs, and deep litter and duff
layers, which cannot be predicted by the Fire Behavior Prediction System. The amount
of bole damage was a better indicator of postfire survival of Douglas-fir after a series of
spring and fall underburns than either scorch height or the percentage of crown volume
scorched (Ryan et al. 1988). Once tree cambium is wounded by fire or mechanical
damage, it is often more susceptible to additional injury by fire, both because the bark
is thinner near the scar, and because of pitch that is often found in association with
wounds. A model that estimates tree mortality based on species, and the amount of
crown and bole damage is described in XII.D.1.b., this Guide.

Bark thickness, texture, and the presence of wounds or pitch also can affect the
likelihood of mortality of shrub stems. However, because of the relatively small
diameter of most shrub stems, most stems are girdled by any fire that reaches into their
canopy, unless heat is present for only a very short period of time.

c. Root mortality. As with tree and shrub crowns and stems, there are physical and
structural characteristics that affect root damage. Structural support roots growing
laterally near the surface are more susceptible to fire damage and consumption than
those growing downward. Roots found in organic layers are more likely to be lethally
heated or consumed than those located in mineral soil layers. This makes shallow-
rooted trees more subject to postfire windthrow. Most plants have feeder roots. Tree
feeder roots collect most of its water and nutrients, are very small in diameter and are
usually distributed near the surface. If most of the feeder roots are located in soil
organic layers rather than in mineral soil, they are much more subject to lethal heating
and consumption. While this may not always kill the tree, it can place the tree under
significant stress. If fire has been excluded for a long time from areas that formerly had
a high fire frequency, increased amounts of root (and bole) damage may result from
fires smoldering in accumulations of litter beneath trees (Herman 1954 and Wagener
1955 in Wade and Johansen 1986).

Damage to roots and other subsurface plant parts cannot be predicted by the general
behavior of the surface fire, nor by any specific descriptors of surface fire activity, such
as fireline intensity, flame length, or rate of spread. Temperatures reached in the
flaming front may be extremely high, but most of that heat is directed upwards. The
mortality of buried plant parts is much more dependent on the total residence time of
the fire, the length of time a heat source is present (Wade 1986), not just the length of
time flaming combustion occurs. The subsurface heat regime of a fire is influenced by
the amount of surface dead fuel, the amount and compactness of litter and duff, and
the moisture content of those materials.

Burn severity (see II.B.6.e.), a qualitative measure of the amount of consumption of

surface fuels and duff layers, is an indicator of subsurface heating. Soil moisture also
retards penetration of heat into soil layers (Shearer 1975; Frandsen and Ryan 1986),
thus protecting subsurface plant structures. There is some relationship between heat
per unit area, the total amount of heat released in flaming combustion, and root
damage, particularly if only a thin layer of combustible fuel is present. However, if
moderate to heavy accumulations of surface dead fuel or organic layers exist, their
consumption in smoldering and glowing combustion is the best indicator of significant
amounts of subsurface heating. (Factors that regulate fuel and duff consumption, and
thus burn severity, are discussed more completely in III.B.2. and III.B.3., this Guide.)
There may be considerable damage or consumption of roots when little or no damage
is apparent in tree or shrub crowns (Geiszler et al. 1984 in Wade and Johansen 1986).
The amount of subsurface plant mortality can be indirectly estimated by knowing the
location of roots and other buried reproductive structures, and relating this to classes of
burn severity.

Plant mortality is often the result of injury to several different parts of the plant, such as
crown damage coupled with a high percentage of cambial mortality. Mortality may not
occur for several years. Death is often the result of secondary infection by disease,
fungus, or insects, because the resistance of plants to these agents is often lowered by
injury, and wound sites provide an entry point for pathogens (Littke and Gara 1986). A
plant weakened by drought, either before a fire or after wounding, is also more likely to
die.

2. Vegetative Regeneration. Sprouting is a means by which many plants recover after

fire. Sprouts can originate from plant parts above the ground surface or from various
levels within the litter, duff, and soil layers. Where sprouts originate and the depth
below the surface at which buried plant parts are found can be species specific
characteristics (Flinn and Wein 1977). Heat released in the flaming front of the fire can
have a direct impact on mortality of sprouting sites that are above the ground surface.
The same factors that control mortality of roots affect mortality of buried reproductive
structures of woody plants, grasses, and forbs.

a. Location of dormant buds. Dormant buds are often located on laterally growing
stems. Stolons are stems that run at or near the surface of the ground, producing
plants with roots at intervals, such as a series of strawberry plants. Rhizomes are
laterally growing underground stems located at varying depths in litter, organic, and
mineral soil layers. They have a regular network of dormant buds that can produce new
shoots and roots. Rhizomes are a structure common to many plants, including blue
huckleberry (Vaccinium globulare), thimbleberry (Rubus parviflorus), Oregon grape
(Mahonia spp.), common snowberry (Symphoricarpos albus), shiny-leaf spiraea
(Spiraea betulifolia), heartleaf arnica (Arnica cordifolia), gambel oak (Quercus
gambelii), bittercherry (Prunus emarginata) and chokecherry (Prunus virginiana).

Many plants have buds located in the tissue of upright stems, above or below the
surface of the ground, such as bitterbrush (Purshia tridentata), bigleaf maple (Acer
macrophyllum), rabbitbrush (Chrysothamnus spp.), winterfat (Ceratoides lanata), and
mountain-mahogany (Cercocarpus spp.). Bud masses may also be present in branch
axils. Paper birch and madrone are species that have root collar buds located in stem
tissue at the point where roots spread out from the base of the stem. Lignotubers,
burls, and root crowns are names for masses of woody tissue from which roots and
stems originate, and that are often covered with dormant buds (James 1984). Dormant
buds may be deeplyburied in wood, and may be located far below the surface if the
tissue mass is large. Chamise (Adenostoma fasciculatum), serviceberry (Amelanchier
alnifolia), scrub oak (Quercus dumosa), tanoak (Lithocarpus densiflora), alder (Alnus
spp.), and mallow ninebark (Physocarpus malvaceus) all produce sprouts from these
buried woody structures.

A caudex is an upright underground stem common in many forb species, such as

arrowleaf balsamroot (Balsamorhiza sagittata) and lupine (Lupinus spp.), that develops
new leaves and a flowering stem each year. Other stemlike reproductive structures are
bulbs and corms, which are essentially buried, short thickened stems with a bud or
buds covered with fleshy leaves. Some species have dormant buds or bud primordia
located along the surface of roots from which new shoots can originate, such as aspen,
fireweed (Epilobium angustifolium), and horsebrush (Tetradymia spp.).

b. Postfire sprouting process. Postfire sprouting occurs by a very orderly process.

The following discussion describes a likely model of the interactions that control postfire
shoot development in woody plants. Similar processes probably regulate sprouting in
grasses and forbs. The growth of most dormant buds or bud primordia of woody plants
is controlled by a phenomenon called apical dominance. Growth hormones, particularly
auxin, a plant hormone manufactured in actively growing stem tips and adjacent young
leaves, are translocated to dormant buds, which prevent them from developing into
new shoots (Schier et al. 1985). If these plant parts are removed, the source of growth
hormones is eliminated. The balance of plant hormones within the buds changes (ibid.).
Growth substances in roots, particularly cytokinins, that are translocated to the buds
can cause dormant buds to sprout, or stimulate bud primordia to differentiate into
shoots (ibid.). Cytokinins may already be present in buds, and a decrease in the ratio of
auxins to cytokinins provides the stimulus for bud outgrowth (ibid.). The buds that
become shoots are usually those nearest to the part of the plant killed by the fire. If
dormant buds are destroyed, new buds may form in wound tissue, called callus, and
subsequently produce shoots (Blaisdell and Mueggler 1956). Once new shoots are
actively growing, they produce growth hormones that are translocated to other dormant
buds that are farther away from the point of damage, suppressing their growth (Schier
1972; Bilderback 1974).

If the organic layer is thinned or disturbed, additional light may reach the tips of
rhizomes and stimulate them to grow towards the surface and produce shoots (Barker
and Collins 1963; Trevett 1956; Miller 1977). It has also been observed that
decapitating a rhizomatous plant causes laterally growing rhizomes to turn upwards
and become shoots (Schier 1983). Additional rhizomes often form in response to
vigorous aerial plant growth (Kender 1967), which may subsequently produce
aboveground shoots. Sprouts from new rhizomes or lateral roots may recolonize areas
where all reproductive plant parts were killed by a fire. Plants may sprout soon after a
fire, or not until the following spring if the fire occurs after the plants have become
dormant for the winter (Miller 1978). Warmer soil temperatures after burning may
enhance the amount of sprouting that occurs (Zasada and Schier 1973). The initial
energy required to support growth until the sprout is photosynthetically self-sufficient
comes from carbohydrates and nutrients stored in the reproductive structures or in
adjacent roots (James 1984).

Postfire sprouting ability can vary with plant age. Young plants that have developed
from seed may not be able to sprout until they reach a certain age, which varies by
species (Smith et al. 1975; Tappeiner et al. 1984). For a given species such as
bitterbrush, an older plant may be able to produce few, if any, sprouts that survive
(Ferguson 1988). Other factors also may lead to decreased amounts of postfire
sprouting in older bitterbrush. These include the higher amount of dead material within
old crowns, and the deep organic layer found beneath some old plants that cause
increased potential for lethal heating during a fire (Clark 1989; Miller 1988).

Aspen produces sprouts from healthy roots. Decreased amounts of postfire sprouting
observed from older aspen stands (Schier 1975) may be because the condition of
many of the roots has deteriorated to the point where they cannot sprout (Zasada, pers.
conv. 1989). Aspen stands in Alaska can resprout vigorously after fire when they are
150 to 200 years old, perhaps because the incidence of pathogens in Alaskan aspen
stands is relatively low (ibid.). In areas such as the Lake States, aspen stands are often
killed by cankers by the age of 50 to 70 years (ibid.) An aspen stand that is producing a
few understory suckers still has the capacity to sprout after a fire (DeByle 1988a).
Some plants therefore replace themselves, forming a new aboveground stem, but use
essentially the same root system as before -- vegetative regeneration. Some plants
may spread and develop new individuals from different locations along their roots or
rhizomes. Shoots may form their own root system and become separate individuals.
This is called vegetative reproduction by some, but these plants are genetically the
same as the parent plant, and represent growth of the clone (Zasada 1989). True
reproduction only occurs when a new genetic individual is formed, by establishment
and growth of a new seedling (ibid.). Sexual reproduction of new individuals of gambel
oak occurs when plants establish from seeds. Gambel oak can also regenerate
vegetatively, replacing itself by sprouting from a lignotuber, and extending the clone by
developing new plants from buds on rhizomes (Tiedemann et al. 1987).

c. Relationship of sprouting to burn severity. A strong relationship exists between

burn severity, a measure of the amount of heating at and below the ground surface,
and postfire sprouting in forested areas (Miller 1977; Dyrness and Norum 1983; Ryan
and Noste 1985; Morgan and Neuenschwander 1988).

(1) A light severity fire occurs under moist fuel conditions, or where little fuel is present.
Woody debris is partially consumed but some small twigs and much of the larger
branchwood remain. Leaf litter may be charred or consumed, and the surface of the
duff layer also may be charred. A light severity fire may kill reproductive parts at or very
near the surface such as stolons, or stem buds that are not well protected by bark
layers. It has little effect on most buried plant parts and significant amounts of postfire
sprouting can occur.

(2) A moderate severity fire occurs when fine and smaller diameter dead fuels and
surface litter and organic layers are dry, but large dead fuels and lower organic layers
are still moist. Foliage, twigs and the litter layer are consumed. Duff, rotten wood, and
much of the woody debris are removed. Logs are deeply charred. This type of fire kills
or consumes plant structures in litter and in the top part of the duff layer, such as
stolons and shallow rhizomes, and may kill buds on portions of upright stems that are
beneath the surface, and buds on the upper part of root crowns. Sprouting occurs from
buds in deeper duff or soil layers. Moderate severity fires frequently cause the greatest
increases in stem numbers of rhizomatous shrubs (Miller 1976), and of root sprouters,
such as aspen (Brown and Simmerman 1986).

(3) A high severity fire occurs when large dead fuels and organic layers are dry. It
consumes all litter, twigs, and small branches, most or all of the duff layer, and some
large diameter dead, down woody fuels, particularly rotten material. Significant
amounts of soil heating can occur, especially near fuel concentrations. This kind of fire
can eliminate plants with reproductive structures in the duff layer, or at the duff-mineral
soil interface, and may lethally heat some plant parts in upper soil layers. Sprouting can
only occur from deeply buried plants parts, which may still be a significant amount for
species with deep roots such as aspen, or deep rhizomes such as gambel oak. Killing
all belowground reproductive structures usually occurs only where there is a long
duration surface heat source, such as beneath a large pile of woody debris that
sustains almost complete combustion. Observations show that the concept of burn
severity also can be related to fire effects on sprouting rangeland shrubs. The severity
of burn relates to the depth of the litter layer beneath a shrub (Zschaechner 1985), and
its moisture content when the fire occurs. A light severity fire may scorch litter beneath
the shrub crown, but causes little or no damage to reproductive buds buried in
stemwood or soil, although it can kill buds at the surface of the soil or those not
protected by wood. Most sprouting plants will likely regenerate after this type of fire. A
moderate severity fire consumes some basal litter and organic matter, and can kill
some reproductive buds. Buds located in deeper litter layers may be lethally heated
even if the litter is not consumed, and sprouting of some species may be reduced or
eliminated. A high severity fire can consume all litter and organic matter beneath a
shrub, and kills all buds and roots in or near the organic layer. This kind of fire favors
shrubs with buds and roots buried so deeply in the soil beneath the plant that they do
not receive a lethal dose of heat. Fires which occur where there are deep
accumulations of litter beneath shrubs and isolated trees, or significant amounts of
dead lower branches that burn off and smolder beneath a shrub crown, are more likely
to lethally heat roots and reproductive structures than a fire that occurs where there is
sparse litter and few dead branches. Reproductive buds of rangeland shrubs that are
located on roots or rhizomes at some distance away from the parent plant are not likely
to be killed by fire because fuels are often sparse in these locations.

d. Postfire sprouting of grasses. Grass meristems, growing points, are the point
where new leaf tissue is formed during the active growing period, and resumes after
summer quiescence or winter dormancy. New growth also may occur by "tillering,"
branching from dormant axillary buds in the plant crown or on rhizomes. Burning of all
live leaves may stress the plant, and cause subsequent death. However, a more
common cause of death of grasses is the lethal heating of meristems and buds.
Sensitivity of meristems and dormant buds to heating relates to their location with
respect to the soil surface and the fuel provided by dead grass and shrub litter, and
other associated fuels, including shrub canopies. Meristems of some grasses form on
long shoots which are elevated above the surface and are readily exposed to fire heat.
Their postfire recovery depends upon growth form and whether any basal meristems
and buds survive. A detailed description of the vegetative regeneration process in
grasses can be found in Dahl and Hyder (1977).

Physiologically active meristems are more susceptible to heat than when they are
dormant or quiescent (Wright 1970). Mortality of cool season grasses which green up
early in the growing season can be caused by the burning of the litter of associated
warm season grasses that are still dormant and hence more heat resistant. A high
mortality of perennial grasses also may occur if fire burns in cured litter of annual
grasses while perennials are still actively growing.

(1) Response of stoloniferous and rhizomatous grasses. Stoloniferous grasses, those

which spread by stolons, such as black grama, are sensitive to heating. Many species
of grass are rhizomatous, with meristems and buds buried beneath the litter, duff, or
soil surface. Whether rhizomatous grasses are stimulated or killed by fire depends on
rhizome location with respect to the surface, whether rhizomes are located in mineral
soil or in organic layers, the moisture content of the litter, organic, and soil layers, and
the amount and duration of heat generated by the surface fire. Rhizomatous grasses
often respond positively to rangeland fires because meristems and buds are usually
protected by soil, and a long-term surface heat source over large, contiguous areas is
rarely present (Wright and Bailey 1982). In forested areas, grass rhizomes are more
likely to be located in litter or duff layers or in association with dead woody fuels, and
the potential for lethal heating is higher than it is in rangeland situations.

Litter and decomposed organic matter derived from rhizomatous graminoids such as
cattails (Typha latifolia), reeds (Phragmites communis), and rushes (Juncus spp.) can
accumulate to such thick levels that they completely fill areas of open water in
wetlands. Occasional severe fire can have a criticalrole in maintenance of these
wetlands. Fires which occur after long dry periods when water levels are low can
consume much of this organic accumulation, restoring areas of open water. Prescribed
fire is recognized as a management tool for this purpose in the Delta Marshes of
Manitoba (Ward 1968). This same role in wetland maintenance for wildland fire has
been noted in Alaska (Kelleyhouse 1980), and in the coastal plain of the southeast U.S.
(Hermann et al. 1991).

(2) Response of bunchgrasses. The location of meristems and dormant buds of

bunchgrasses can be near the surface of the bunch above the level of the soil, or at
various depths below the soil surface within or below the bunchgrass litter. Buds and
meristems can be readily exposed to lethal temperatures, or be fairly well protected if
deeply buried in unburned organic materials or in soil. Fuel and moisture characteristics
affect the amount of heat generated. Wright (1971) discusses the relationship between
stem coarseness and the rate at which a bunchgrass clump burns. Fine stemmed
grasses with a dense clumping of basal stems can burn slowly and generate a fair
amount of heat that can be transferred to meristems and buds. Fire tends to pass fairly
quickly through coarse stemmed bunchgrasses, which usually have little material
concentrated at their base near reproductive structures. Fires tend to burn more rapidly
through small diameter bunches in comparison to large diameter bunches, with larger
bunches more likely to have enough fuel to release significant enough amounts of heat
to affect growing points (Wright and Klemmedson 1965). The amount of surface litter,
i.e., the amount of fine fuel, depends on the amount of use by livestock and wildlife,
production in this and previous seasons, and the time since the last fire in areas
receiving little utilization.

(3) Relationship of moisture conditions and fire behavior to mortality. The moisture
content of fine aerial litter, accumulations of basal litter, dead bunchgrass centers,
adjacent shrubs and shrub litter layers, and dead woody fuels all affect the amount and
duration of heat that the meristems will receive. Mineral soil moisture can control how
much heat is received by plant parts located in soil layers. While it is true that moist soil
conducts heat better than dry soil, the moisture in surface soil layers must first be
evaporated before heating of deeper layers occurs (Albini 1975 in Miller 1977). Moist
heat, i.e., steam, may more effectively heat meristems than dry heat, and may be a
cause of higher mortality when fires occur where greenup has begun in some plants.
However, wet fuel doesn't burn, so the likelihood of a long duration fire under damp fuel
and soil conditions that will kill all active bunchgrass meristems and dormant buds is
very low, and heat penetration into organic and soil layers is minimal under these
conditions (Frandsen and Ryan 1986). If flammable shrubs ignite, dry and preheat
adjacent bunchgrass clumps, bunchgrass mortality may be higher than on a similar site
with few shrubs that burned under the same conditions (Zschaechner 1985).

For a given litter moisture content, windspeed controls how quickly a fire passes over a
plant and the rate at which the litter burns. Fires have been observed in northwest
Colorado burning at windspeeds of 10 to 14 miles per hour (16 to 22.5 kilometers per
hour) with rates of spread greater than 88 feet/minute (27 meters/ minute) during dry
summer conditions. These fires charred only the tops of the crowns of bluebunch
wheatgrass and Indian ricegrass plants that were 4 to 5 inches (10 to 13 centimeters)
in diameter. Fire may have moved through grass litter too quickly to have a long
enough residence time to ignite grass crowns, and little grass mortality occurred
(Petersburg 1989).

(4) Relationship of damage to postfire sprouting. A fire may move quickly through a
bunchgrass stand with little residual burning. At the other extreme, the dead center of a
bunchgrass plant may ignite, smolder, and burn for hours. Conrad and Poulton (1966)
developed damage classes for bunchgrasses: 1) unburned, although foliage may be
scorched; 2) plants partially burned, but not within 2 inches (5 centimeters) of the
crown; 3) plants severely burned, but with some unburned stubble less than 2 inches;
4) plants extremely burned, all unburned stubble less than 2 inches and mostly
confined to an outer ring; 5) plants completely burned, no unburned material above the
root crown.

Postfire response of a bunchgrass plant can be related to these damage classes,

particularly for those species with meristems above the mineral soil surface (ibid.). The
highest postfire sprouting potential usually is found in those plants with only some
surface litter removed. The amount of sprouting tends to decrease as the amount of
basal litter consumption increases, with new shoots most likely to appear from the
outside edge of the bunch when little unburned stubble remains. Plant mortality is most
likely if all plant material above the root crowns is consumed. Survival and recovery
after a specific amount of fire caused damage must also be considered with respect to
phenology and other seasonal factors that affect plant response. (See Sections B.4.a.;
B.4.c.)

3. Seedling Establishment.

a. Seedbed. Requirements for successful germination and establishment vary for

different species. Organic seedbeds, even rotting logs, may be able to successfully
support seedling establishment and survival if water is not limiting during the growing
season (Zasada 1971). However, moss, litter, and duff are poor seedbeds in many
climates because they frequently dry out in the summer, killing the seedling if the root
has not yet reached mineral soil. Other attributes of organic seedbeds may also inhibit
seedlings (Zasada et al. 1983). For many species, the best seedbed is exposed
mineral soil, and microsites where most or all of the organic layer has been removed by
fire provide the greatest chance for seedling survival. Soil does not dry out as readily as
organic material, and nutrients may be more readily available in ash. Competition from
sprouting plants may be reduced.

On hot, dry, exposed sites, seedling germination and establishment may occur more
readily if some organic material remains as mulch, especially if the seeds are covered
(Clark 1986). Ponderosa pine seedlings are more likely to establish if seeds land on
bare mineral soil, and the ungerminated seeds are subsequently covered by litter
(McMurray 1988). Allelopathic chemicals, those that inhibit the germination and/or
establishment of seeds of plants of other species, are commonly found in the litter
beneath certain plant species, including chamise (McPherson and Muller 1969), Utah
juniper (Juniperus osteosperma), and singleleaf pinyon (Pinus monophylla) (Everett
1987a). Fire can volatilize these chemicals and allow additional seed germination.
Some species that must establish from seed may be temporarily eliminated from a
burned area because their establishment is not favored by conditions created by a fire.
They may require shade, have slow growth of their primary root, or a high water
requirement. Some tree species such as pinyon and juniper, may establish a few years
after a fire in the shade of plants that established first, but subsequently can grow in full
sun (Everett 1987b).

b. Seedbank. The seedbank, the supply of seeds present on a site, is composed of

buried seeds, those stored in the tree canopy, and those that are deposited annually.
Seeds of some species, such as willow, are very short-lived, and are part of the seed
bank for only a short time. Other species have extremely long-lived seeds, and become
a fixed member of the soil seedbank, once their seeds are dispersed.

Seed dispersal mechanisms vary. Light seeds may be windblown while heavier seeds
may skid across the surface of the snow. Some seeds have wing-like structures which
enhance their movement through the air. Seeds with barbs or hooks may be carried by
animals. Hard-coated seeds ingested along with their fruit may pass through the bird or
animal, with an enhanced likelihood of germination. Seed dispersal from unburned
areas depends on the amount of available seed, the distance of the seed source from
the burned area, the prevailing wind direction, and the type of seed.

The supply of seeds of a specific species can be greatly influenced by the amount of
annual seed production, which can vary significantly (Zasada et al. 1978).
Regeneration of conifers may be limited because cone crops are poor during the period
of time when exposed mineral soil seedbed is present.

Surviving plants on or near the burned area may not be old enough to produce seed
(Zasada 1971; Barney and Frischknecht 1974), or may be too old to produce much
viable seed.

The dispersal of seeds from plants occurs at a time that is characteristic for that plant,
and can last for different durations of time (Zasada 1986). The time of fire occurrence
with respect to seed dispersal can determine whether a species can regenerate
promptly. Heat from the fire may kill seeds in the canopy and seeds that have recently
been distributed onto the site. Seeds of a certain species may require a period of cold
before they can germinate, so seedlings of that species will not appear until the
following spring.

Seeds in immature cones in tree canopies may have survived the fire, and may
continue to ripen even though the foliage was killed (ponderosa pine) (Rietveld 1976),
or the bole was completely girdled by fire (white spruce) (Zasada 1985). Serotinous
lodgepole pine cones retain their seeds because of the presence of a resin bond
between the cone scales. These cones do not open and release their seeds unless
heated to 45 to 50C (113 to 122F), a temperature that melts the bond (Lotan 1976).
Numerous lodgepole pine seeds are often released after heating of the canopy during a
fire. However, there is considerable variation in the amount of cone serotiny, both on
individual trees, and geographically (ibid.). Black spruce (Picea mariana) cones are
"semi-serotinous", i.e., they open and release their seeds over a period of years
(Zasada 1986). Because cones are usually bunched near the top of the tree, some
cones are often shielded from fire heat and provide a postfire seed source.

c. Stimulation of buried seed. There may be an enormous reserve of seed stored in

litter, duff, and soil. Seed may accumulate on the surface and be gradually buried by
litter, or may be cached by rodents and birds (West 1968; Tomback 1986). Seed of
some species may remain viable for many years, with dormancy imposed by an
impermeable seed coat (Stone and Juhren 1953). Plants such as snowbrush
ceanothus (Ceanothus velutinus), raspberry (Rubus idaeus), geranium (Geranium
bicknellii), and corydalis (Corydalis sempervirens), as well as many annuals of
California chaparral may appear on a site after a fire where they were not apparent
before the fire. Seeds of snowbrush ceanothus remain viable for 200 to 300 years
(Gratkowski 1962 in Noste and Bushey 1987). Germination of some species is
enhanced by fire that can melt or crack the seed cuticle or otherwise scarify the
impermeable seed coat (Keeley 1987; Rasmussen and Wright 1987).

Requirements for optimum germination may be very specific, such as redstem

ceanothus that has the highest amount of germination after exposure to moist heat at
80 C (176 F) (Gratkowski 1973), explaining its higher germination after high severity
fires than low severity fires (Leege 1968). Chemicals leached from charred materials
stimulate germination of many species of California chaparral and coastal sage scrub
(Keeley 1987). Increased light levels caused by removal of vegetation can induce or
enhance seed germination (Keeley 1987). Some species, such as chamise and
hoaryleaf ceanothus (Ceanothus crassifolius), produce a certain proportion of seeds
that will only germinate after fire treatment, while other seeds from the same plant will
germinate under any suitable moisture and temperature conditions (Christensen and
Muller 1975a). Annual species may appear from stored seed after a fire, but may
disappear in a few years as site conditions change, a common phenomenon in
California chaparral (Sweeney 1956; Muller et al. 1968; Christensen and Muller 1975b).

Germination of seeds of chaparral communities is adapted to wildfires that normally

occurred during fairly hot, dry, late summer or fall conditions. Seeds of some species of
chaparral communities require dry heat to induce germination, but are killed by lower
temperatures if they have imbibed moisture. Other seeds require higher temperatures
for a longer duration to induce germination than generally occur under spring burning
conditions (Parker 1989). If chaparral sites are burned under moist spring conditions,
germination of both of these types of seeds is often very much reduced. This is a
particular concern for maintenance of a seed bank of chaparral "fire-following annuals"
as well as shrub species that can only reproduce from seed (Parker 1987).

d. Dual response plants. Some plants will recover after a fire both by resprouting and
by germination of duff stored seeds, while obligate seeders reproduce by seed only.
Obligate seeders often have seedlings with better potential for establishment than
seedlings from species that sprout (Parker 1984). Other plants have a two-stage
response to fire. They sprout from surviving reproductive structures, then produce
seed, right onsite, that can readily utilize available seedbed. Fireweed and rabbitbrush,
for example, can sometimes gain temporary dominance over a site for these reasons.

4. Factors Influencing Postfire Plant Recovery and Growth.

a. Climate and weather. Different parts of the country have characteristic seasonal
distribution of temperature and precipitation. The overall pattern of seasonal plant
growth (phenology) relates to climate, such as the time of the year when most growth
occurs; occurrence of late summer quiescence, and the onset of winter dormancy.
However, the timing and rate of plant development and total amount of growth can vary
greatly with seasonal weather (Mueggler 1983). The date when plants begin growth,
flowering, and cease growing all relate to seasonal weather (Sauer and Uresk 1976).
The average annual occurrence of the wildfire season in various parts of the country is
closely related to climate, while the actual timing and severity is related to fuel amount
and conditions, and the weather that occurs that year. Generalizations can be made
about weather trends and patterns for a particular region, but there are always
exceptions. Long-term averages do not reflect the wide range of conditions possible.

(1) Prefire weather. Prefire weather can affect the plant growth stage at the time of
burning. The amount and availability of fuel is influenced by weather. Fires in the
cheatgrass region tend to be much larger in years with high winter precipitation and
spring rain, resulting in high production of fine fuel. Burned acreage in the Sonoran
Desert tends to be higher after two winters of above average precipitation that
promotes growth of winter annuals, which subsequently dry and provide fuel (Rogers
and Vint 1987). The moisture content of heavy fuels and deep litter and duff layers is
closely related to temperature and precipitation in previous months, and thus the
likelihood that these fuels are available to burn and provide a long-term heat source is
weather dependent (Brown et al. 1985). Fire size, and its degree of impact on
vegetation, is influenced by fuel availability, and burning conditions at the time of the
fire. Drought, anomalous high winds and low humidity, or high summer precipitation all
affect the immediate impact of a fire in a particular area in a certain year.

(2) Postfire weather. Postfire weather can affect plant survival. Sprouting plants must
produce enough growth to restore food reserves before the next period of high use,
and this growth can be enhanced or limited by weather. Without restoration of
carbohydrate reserves, the plant may die. Plants which sprout late in the season also
can die because they have too little time or energy to harden off for the winter. The
amount of autumn rain can determine whether germination of seed of some species
occurs in the fall or the following spring. Late summer rains (Thill, Beck, and Callihan
1984) followed by a dry period can cause germination, and subsequent death, of many
seedlings. Weather in the following years affects the rate of recovery from burning by
influencing productivity. Drought can place additional stress on injured plants, and
increase the likelihood that they will die. Postfire weather is a primary factor in
determining range readiness for postfire grazing use. The weather cannot be
controlled, but it is important to document it. Fires burned on similar sites in different
years with the same burning weather may have widely varying results because of
differences n prefire and postfire weather. Analysis of these records may explain the
reasons for significant variations in postfire response, and the "success" or "failure" of a
specific prescribed fire project.
b. Carbohydrates.

(1) Carbohydrate cycle. Carbohydrates are starches and sugars manufactured by

plants and used to provide energy for metabolism, and structural compounds for growth
(Trlica 1977). Carbohydrates which are manufactured in excess of those used are
stored in various parts of the plant, such as roots, rhizomes, root crowns, stem or leaf
bases (Cook 1966a), or evergreen leaves. There is a seasonal cycle of depletion and
restoration of total available carbohydrates (TAC) that relates to events in the growth
cycle of the plant. The most rapid depletion usually occurs during greenup, to support
initial development and growth of leaves and shoots. Stored carbohydrates levels also
may be lowered during the period of flower and fruit development. Carbohydrates are
required to prepare plants for winter, the "hardening off" process, as well as for
respiration and cellular maintenance during winter dormancy (McCarty and Price 1942
in Trlica 1977), and quiescence during late summer drought conditions (Hanson and
Stoddart 1940 in Trlica 1977). Roots must be maintained even while the aboveground
part of the plant is not actively growing. A major depletion also can be caused by heavy
or repeated grazing or browsing, as the plant may need to use reserves to support
subsequent new growth if inadequate leaf area remains on the plant to provide energy
for growth and new tissue formation. "In general, too heavy, too early, or too frequent
grazing or defoliation result(s) in declining vigor of vegetation" (Hedrick 1958 in Trlica
1977).

Restoration occurs when production by photosynthesis exceeds demands for growth

and respiration. The beginning and length of the restoration period varies. Cook (1966a
and b) discusses the seasonal carbohydrate cycle. Some plants rapidly deplete, but
then quickly restore carbohydrate reserves, all within about the first month of growth.
Squirreltail (Elymus elymoides) is a plant that exhibits this V-shaped carbohydrate
cycle. U-cycles are shown by plants that deplete carbohydrate reserves over a much
longer period of time, and don't begin to make a significant restoration of reserve
carbohydrates until later in the growing season. Bitterbrush, a species with a U-cycle,
does not show a major increase in its level of reserves until August or September, only
accumulating carbohydrates from the period of seed formation until leaf fall (McConnell
and Garrison 1966). The timing of highs and lows in the carbohydrate cycle thus
corresponds with the growth states of the plant. The cycle can differ from year to year
because the timing of phenological events can vary significantly among years (Sauer
and Uresk 1976; Schmidt and Lotan 1980; Turner and Randall 1987). The amplitude of
the cycle will vary with growing conditions, the amount of growth produced, and the
amount of carbohydrate used for other physiological processes, all of which affect the
amount left for storage.

(2) Relationship to fire. Energy and material for initial plant regrowth after a fire depend
on the availability of reserve carbohydrates. The biggest negative impacts from burning
may occur during the lowest point in a plant's annual carbohydrate cycle, usually during
the early seasonal growth period. The low survival of chamise sprouts after spring
prescribed fires has been attributed to low winter and spring carbohydrate reserves
because of high spring demand for growth, flowering, and fruiting (Parker 1987). For
other species, the effects are most negative if the plant is burned late in the growing
season when reserves are being rapidly replenished because the plant uses a
considerable amount of stored carbohydrates to sprout, but does not have enough time
to restore reserves before winter dormancy (Trlica 1977; Mueggler 1983).As a result of
burning during an unfavorable growth period with respect to stored carbohydrates, a
plant or any sprouts that it produces may die during the next long period of
carbohydrate demand, such as summer quiescence or winter dormancy. If the plant
survives, its productivity in the next few years may be greatly reduced. An additional
consideration when burning old stands of woody plants is that energy reserve levels of
these plants may be low, because annual production is low, and/or much of the plant's
carbohydrate production is used to maintain old plant tissue. Meager amounts of
sprouting observed from old bitterbrush plants may be partially due to low levels of
stored root carbohydrates.

The degree of dependency for regrowth that a plant has on carbohydrate reserves after
fire depends on whether any photosynthetically capable material, such as sheath
leaves on stubble, survived. If some plant tissue that can photosynthesize remains or
rapidly regenerates, newly grown leaf material soon manufactures all of the
carbohydrates that the plant needs for growth and respiration. However, the initial spurt
of growth after a fire likely requires use of some stored carbohydrates, even if only for a
day or two. Evidence from clipping and grazing studies has shown that the recovery of
grass plants is more related to the removal of growing points than to the carbohydrate
level at the time of defoliation (Caldwell et al. 1981; Richards and Caldwell 1985).
However, fire may have a greater impact on grass plants than severe defoliation
because a majority of the carbohydrates used to initiate regrowth are derived from the
basal portion of the older tillers, and these may not survive a fire.

c. Postfire plant competition. Plant competition occurs when growth and reproduction
of one plant is hampered by the presence of another, or, when the resources of a site
required by one plant are reduced by another (Harris 1977). Plants compete the most
for whatever is in shortest supply - particularly water, nutrients, and light. Whether
competition occurs and the degree to which it occurs depends on the species present
on the site, the number of plants present, and the site conditions (Samuel and Depuit
1987; Brand 1986). Simultaneous requirements for limited resources such as water and
light can place individuals in competition with each other. Whether certain species
compete depends on the timing of germination and growth, germination and
establishment requirements, rate of growth, and requirements for water and nutrients.
Some species have an innately high ability when in a seedling state to compete with
seedlings of other species (Samuel and DePuit 1987). The ability of a plant to respond
to changes in the supply of nutrients or water varies by species (ibid.). Some species
can take better advantage of changes in the postfire environment than other species
can, which may give them a competitive advantage.

Fire affects plant competition by changing the numbers and species of existing plants,
altering site conditions, and inducing a situation where many plants must reestablish on
a site. In a postfire situation established perennial plants that are recovering
vegetatively usually have an advantage over plants developing from seed because they
can take up water and nutrients from an existing root system while seedlings must
develop a new root system (ibid.). Natural regeneration of shrubs may severely limit
growth of naturally occurring or planted conifers because of competition for light or
moisture (Stein 1986; Haeussler and Coates 1986). If perennial plants are few, or their
postfire survival is low, and a seed source is present, seedlings may establish and
dominate the community for varying periods of time. Certain species may be favored,
such as ceanothus (Parker 1984), because of the sheer volume of seeds on a site.
Cheatgrass (Bromus tectorum) has such a great postfire advantage over seedlings of
most native grasses because roots of cheatgrass seedlings can grow at much cooler
soil temperatures than those of most native perennial grasses, and can proliferate
much more rapidly at warmer soil temperatures than can roots of natives. Cheatgrass
seedlings can deplete soil moisture in the spring before other species get their roots
down into the soil profile (Thill, Beck, and Callihan 1984).

Grass seeded for postfire erosion control in forested areas may easily overtop conifer
seedlings. In chaparral areas, seeded grasses compete with sprouts and seedlings of
native plants (Barro and Conard 1987). Litter from seeded grasses may increase the
flammability of these sites to higher levels than would occur if only native vegetation
recovered on the site (Cohen 1986 in Barro and Conard 1987). A second fire after a
short time interval might kill all seedlings of native species, often before they have
produced much seed, decreasing the number of seeds in the soil seed bank.
Conversely, if seeded crested wheatgrass establishes on a cheatgrass site after it
burns, the amount of litter, and fire frequency, can decrease.

A lack of fire can also increase plant competition. One hundred year old stands of
juniper usually have very low cover of shrubs and grass (Barney and Frischknecht
1974), probably because of juniper's superior ability to extract soil water, as well as the
inhibitory effect of juniper litter on germination and establishment of seedlings of shrub
and herbaceous species (Everett 1987b). Herbaceous production in the vicinity of
sagebrush plants decreases as sagebrush cover increases, because of root
competition (Frischknecht 1978). Young stands of conifers that develop in the absence
of fire beneath mature overstories of ponderosa pine compete for moisture and
nutrients with the mature trees (Wyant et al. 1983), weakening them and making them
susceptible to insects and disease.

d. Animal use. If burning occurs in close association with heavy use of the plant
community by livestock or wildlife, either before or after the burn, plant recovery may be
delayed or prevented. Heavy postfire use of perennial plants in the first growing season
after a fire is likely to cause the most harm, particularly in arid and semiarid range
communities (Trlica 1977). Depending upon the plant community and its production
capabilities, some use after the first full growing season may not have a negative
impact, and may even be desirable, as in tobosagrass communities. Two full growing
seasons of postfire rest are necessary before plants can sustain much utilization in the
Intermountain west after wildfire (Wright and Bailey 1982). A longer recovery period is
necessary if weather has been unfavorable for growth, or if establishment of plants
from seeds is required to completely revegetate the site. Desert plants required more
than seven years of recovery after moderate defoliation (Cook and Child 1971 in Trlica
1977), and some shrubland sites may require this long a period of postfire rest if
recovery of browse species is desired. See Chapter IX.B.2 and B.3 for additional
discussion on this topic.

5. Plant Productivity. Fire can affect postfire plant productivity. Short-term decreases
can be caused by plant mortality, reduction in basal area of grasses, forbs, and shrubs,
changes in species composition to less productive plants, and reduced availability of
soil nutrients. Increases are caused by fire induced vegetative reproduction and
regeneration, fire enhanced seedling germination and establishment, improvements in
the soil nutrient regime, and increases in soil temperature. Warmer soil temperatures
often result in earlier greenup on burned areas, particularly in grassland and rangeland
environments.

Removal of thick layers of litter and organic matter in tall grass, wetland, and boreal
environments increase soil temperature and nutrient availability, enhancing plant
growth (Vogl 1973 and Hulbert 1969 in Young 1986). An occasional fire is very
important for rejuvenating cold, nonproductive forest sites in interior Alaska (Yarie
1983), and this is likely also true for many tundra sites. Where permafrost is present,
many nutrients are tied up in frozen organic layers, and are unavailable to plants
(Heilman 1966; 1968). Fire's removal of insulating organic matter and the blackened
surface cause deeper annual soil thawing, and a greater depth and higher temperature
of the rooting zone. Soil acidity decreases and rates of nutrient cycling increase.
Vegetatively regenerating plants and seedlings use these nutrients, significantly
enhancing growth. Eventually, organic matter accumulates and becomes an effective
soil insulator, causing a decline in both growing season soil temperatures and
associated plant productivity.

There may be a significant decrease in productivity during the initial postfire recovery
period, then an increase in production after one or several years. Some conifers have
reduced growth the first growing season after the fire, but show increased growth rates
in subsequent years caused by the removal of competing trees (Reinhardt and Ryan
1988a). Total productivity may not change, but can shift among classes of plants on the
site, such as from conifers that are killed by a fire to shrubs, grasses, and forbs
(Volland and Dell 1981). Total site productivity may actually decrease, but production of
shrubs, grasses, and forbs often increases over prefire levels (Harniss and Murray
1973; Dyrness and Norum 1983). On sagebrush sites, total prefire productivity may not
be reached until sagebrush again dominates the site (Bunting 1985), because its deep
root system can allow it to utilize site resources that are physically unavailable to other
plants.

The length of productivity changes depends on the ecosystem, the degree of change
caused by burning, and the resulting amount of change in species composition in the
postfire plant community. A low intensity, low severity fire may have little effect, while a
shift from an old coniferous forest to a shrubfield may result in long-term changes in
plant production. Site productivity in the first few years after fire will likely be higher if a
significant amount of vegetative regeneration occurs, than if plants on the site must
reestablish from seed. Sprouts can obtain nutrients and carbohydrates for initial growth
from the parent plant while a seedling often has access to only a small nutrient reserve
in seed, and may initially grow fairly slowly. Seedling establishment and growth are
much more dependent on site conditions and postfire weather. Snowbrush seedlings
grow slowly until age 4 or 5, but then grow rapidly until about age 10, while sprouts of
snowbrush may grow from 1 to 2 feet (0.3 to 0.6 meters) per year from the time growth
is induced (Peterson 1989). Exceptions to this general rule do occur. Obligate seeders,
plants that must regenerate from seed, can be adapted for making rapid growth on
burned or disturbed sites (Parker 1984).
Greatly increased amounts of flowering and fruiting may occur, including a significantly
enhanced output of grass seed and berries (Daubenmire 1975; Young 1986;
Christensen and Mueller 1975b). Changes in production are caused by the same
factors that increase vegetative productivity: warmer soil temperatures, improved
nutrient availability, and removal of senescent, woody material that requires a lot of
energy to maintain. For a given species, flower and fruiting generally occur sooner on
sprouts than on plants that develop from seed. For some species, flower buds are
formed on the previous year's growth, so it takes two growing seasons for flowers and
fruits to appear. Increased levels of fruit or seed production may only persist for a few
years of burning. Improvements in forage amount and availability, and increases in
flowering and fruiting are key reasons for wildlife and livestock attraction to newly
burned areas.

C. Resource Management Considerations

Fire effects on plants cannot be understood unless their survival and reproductive
strategies with respect to fire are understood. Some plants resist fire by characteristics
such as thick bark or buds that can withstand scorching temperatures. A site can be
repeatedly burned, and many of these plants survive. Plants may have their surface
parts completely consumed, but endure the fire because belowground reproductive
structures typically survive. Some plants are almost always killed by fire, and their
seedlings cannot tolerate immediate postfire conditions. It can be said that these
species avoid fire, because they are only found on sites that are fire-free for long
periods of time (Rowe 1983).

Plants can be divided into four basic groups with respect to postfire revegetation of a
site (Stickney 1986), as defined by their source and time of establishment. Survivors
are species with established plants on the site that can regenerate after a fire.
Colonizers are species that establish on the site from seed. Residual or onsite
colonizers originate from seed that is present on the site at the time of the fire. Off-site
colonizers develop from seed that is carried from off the site. Secondary off-site
colonizers develop from off-site seed, but not until site conditions are mitigated by the
plants that established first. Initial establishment of a plant is only the first step,
because its long-term survival and productivity is affected by competition with other
plants and by weather.

The following management considerations summarize key elements to consider with

respect to predicting, observing, and interpreting the effects of fire on plants. They are
derived from information explained in greater detail in the text of this chapter, as well as
in Chapter II. Fire Behavior and Characteristics, and Chapter III. Fire Effects on Fuels.

1. Plant Mortality.

a. Relationship to fire behavior, fire characteristics, and fuels.

(1) Flame length relates to the amount of crown scorch and canopy consumption.

(2) Dry concentrations of down, dead woody fuels can ignite and provide a long-term
heat source that can damage a tree crown, tree stem, roots, or buried reproductive
structures.

(3) The amount of heating that results from combustion in the flaming front of a
prescribed fire can be regulated. Ignition methods and techniques must be selected
with consideration for fuel conditions, weather, and slope steepness and concavity.

(a) The width of the flaming zone can be manipulated by controlling the number of lines
of strip headfires that are ignited at once (Norum 1987), and the spacing between
them.

(b) Regulating the interval between lines of strip headfires controls flame length,
because the shorter the interval between lines, the shorter the flames (ibid.)

(c) Use of rapid ignition techniques can greatly increase the rate of heat release and
decrease the duration.

b. Crown scorch height.

(1) The height to which tree crowns are being scorched is often not obvious during
ignition of a prescribed fire.

(2) Scorch height can be estimated from current weather, and observed flame lengths,
using the graphs in Albini (1976, pages 63 to 66).

(3) If scorch height is too high, then ignition can be altered to lower flame lengths, or
the fire may be curtailed until more moderate burning conditions occur.

(4) Too high a scorch height can indicate that the fire prescription may require
modification to reduce scorch heights, such as by prescribing increased fuel moistures,
or lower air temperatures when the fire is ignited.

c. Mortality of crowns.

(1) Dormant buds have varying degrees of sensitivity to fire heat. Sensitivity relates to
size, the presence of protective bud scales or needles, and whether they are
physiologically active or dormant.

(2) Foliage flammability and sensitivity to scorching temperatures varies seasonally,

especially because of changes in foliar moisture content.

(3) Foliage flammability varies by species according to branch density, the presence of
lichens, presence of flammable compounds, retention of ephemeral or evergreen
leaves or needles, and the proximity of the crown base to the surface of the ground.

d. Mortality of tree stems and cambium.

(1) Thick barked species are more resistant to fire heat than thin barked species.
(2) Duration of heating is generally more important than peak temperature in
determining damage to thick barked trees and shrubs.

e. Mortality of roots and other buried reproductive plant parts.

(1) Potential for heating to lethal temperatures relates to the plant part and its location.

(a) Depth of roots or reproductive structures below the surface.

(b) Whether plant parts are located in litter, soil organic layers, or mineral soil.

(2) The potential for heating relates most closely to the duration of heat released during
the consumption of accumulations of dead woody fuels or deep litter and duff layers.
Duff reduction relates to its moisture content (Norum 1977; Brown et al. 1985). See
Chapter III.B.3.a. for moisture content guidelines for consumption of organic soil layers.

(3) Moist soil retards the penetration of heat and protects buried plant parts.

2. Postfire Sprouting.

a. Process. The physiological processes that control postfire sprouting are essentially
the same for trees, shrubs, forbs, and grasses.

b. Species specific characteristics. The type of plant part on which dormant buds are
located, the subsurface distribution of reproductive structures, and the depth below the
surface from which new shoots can develop are species specific characteristics.

c. Relationship to burn severity. Sprouting is closely related to burn severity because

the number of postfire sprouts relates to the number of reproductive buds or bud
primordia that survived the fire. A species may be enhanced or harmed depending on
how deeply lethal temperatures penetrated below the surface, and the characteristic
depth of its reproductive structures.

d. Spread from adjacent areas. On sites where all reproductive structures were killed,
sprouts may develop from rhizomes or roots that colonize the area from adjacent, less
severely burned areas.

e. Bunchgrasses. Bunchgrass species also have reproductive buds located at

characteristic depths below the surface, and with respect to accumulations of dead
basal material.

(1) Moisture contents of basal litter, dead centers of plants, and soil are critical for
determining the amount of consumption of a bunchgrass plant.

(2) There is a potential for additional heating of bunchgrasses from burning of adjacent
shrubs, with the amount of heat related to shrub species, density, and flammability.
(3) The amount of consumption of a bunchgrass plant of a particular species can be
related to its potential for postfire sprouting, because it relates to the amount of physical
damage to growing points and dormant buds.

3. Postfire Reproduction by Seed.

a. Seed ecology. The likelihood that a species will reestablish from seed depends
upon its seed ecology.

(1) Germination and establishment requirements.

(2) Whether its seed is sensitive to heating or is stimulated by heat or chemicals

leached from charred materials.

(3) Length of period of seed viability.

b. Seed source.

(1) Distance from living, seed producing plants.

(2) How much seed in organic and soil layers survived the fire.

c. Timing of fire.

(1) Production of current year's crop of seeds.

(2) Age of plants on or near the site, whether they were old enough to produce seed.

d. Soil seedbank. Some species of plants may establish from duff or soil stored seed
and produce a significant amount of biomass.

(1) The length of time that a species persists depends on its habitat requirements and
how the site conditions change as plant succession proceeds.

(2) Different plant communities have characteristically different species and numbers of
seeds in their seed bank, that also vary in longevity.

e. Relationship to burn severity.

(1) The amount of bare mineral seedbed created.

(2) The amount of heat stimulation or mortality of specific species.

4. Carbohydrates.

a. Plant phenology. Plant growth stage is related to the level of stored carbohydrates
that provide energy for initial postfire vegetative regrowth.
(1) The amount and timing of high and low levels, and rates of recovery, of stored
carbohydrates varies by species, and with conditions in a particular growing season.

(2) Recovery of a plant may be most affected if a plant is burned during a low point in
its carbohydrate cycle, or when there is not enough time for the plant to rebuild stored
carbohydrate levels before the next period of high demand.

b. Animal use. Prefire and postfire use of a site by livestock and/or wildlife must be
evaluated and managed, particularly important if heavy utilization has occurred.

5. Postfire Plant Productivity.

a. Tree growth. Postfire productivity of surviving trees relates to the amount of injury to
crowns, stems, and roots.

b. Sprouting woody plants. Rapid recovery of perennials may occur by postfire

sprouting if reproductive structures were not killed.

c. Seed and fruit production. Seed and fruit production generally increase much more
quickly from plants that regenerate vegetatively, than from plants that must establish
from seed.

6. Direct Seeding and Planting.

a. Postfire rehabilitation considerations.

(1) The requirement for seeding is determined by the specific situation on the burned
area and the management objectives for the area. Factors such as erosion control,
native species restoration, limiting establishment of annual exotics, and meeting wildlife
habitat requirements are major considerations in the decision whether or not to do
postfire rehabilitation.

(2) The likelihood of survival of native species should be assessed before artificial
reseeding is planned. The percentage mortality of individual plants should be
estimated, and likely methods of recovery determined, such as vegetative regeneration
or plant establishment from stored seeds. (See B., this chapter.) Reseeding is not
necessary where recovery of native plants will occur.

(3) Prefire species composition may be determined by inspection of adjacent unburned

areas. A seed mixture of species adapted to the site results in the highest likelihood of
establishment, as well as the greatest long-term diversity and productivity. Grass, forb,
and shrub mixes have been successfully seeded on some Federally managed
rangelands.

(4) Seeded grasses may compete with other desirable species.

i. Seeded grasses can interfere with the establishment of native plants, and limit the
future seed bank of those species.

ii. Seeded grasses can provide significant competition to planted trees and shrubs.

iii. Where postfire erosion is a significant threat, seeding annual or short-lived perennial
grasses may allow greater recruitment of native plants than seeding long-lived
perennial species.

b. Need for rapid replanting. It is often necessary to plant tree seedlings on

productive sites as soon as possible after fire because of potential competition from
naturally regenerating shrubs, grasses, and forbs. Rapid reseeding of rangeland sites
is required if an objective is to establish perennial species on a site dominated by
annual exotics.

c. Prescribed fire considerations.

(1) Residual logs and duff can enhance site productivity by providing sites for
mycorrhizal infection, and nitrogen fixation, both of which are beneficial to
establishment and growth of tree seedlings. (See V.B.c.(4) and (5), this Guide.)

(2) Residual downed logs and shade from standing dead trees provide shade that can
aid establishment of planted seedlings or natural regeneration on dry forest habitat
types.

7. Effect of Postfire Weather. Postfire weather has a significant effect on the rate and
amount of postfire vegetative recovery.

8. Need for More than One Treatment.

a. Dual treatment. A site may require burning after mechanical, chemical, or manual
treatment to kill residual target species or seedlings developed from residual seed.

b. Maintenance burning. Repeated burning at regular intervals may be necessary to

prevent reinvasion of the site by seedlings of undesired species. The desired burning
interval is related to the natural regime that fire used to play in the vegetation
community on the site.

D. Methods to Monitor Fire Effects

A variety of monitoring methods have been employed to study vegetative attributes and
their changes over time. Those methods most appropriate for postburn studies will be
reviewed here. Monitoring schemes chosen to evaluate the effects of fire on both
individual plants and plant communities must be sensitive to the responses observed
from the perturbation of burning. If the fire has been planned, methods selected by the
observer to evaluate changes in the vegetative component must necessarily follow the
objectives of the fire so that the vegetative responses can be properly evaluated.
Preburn measurements are critical, and thoughtful establishment of almost any preburn
study will provide valuable information.
Specific attributes of vegetation or plant communities affected by fire may be expressed
in the generally accepted terms of cover, density, frequency of occurrence, weight,
species composition, number, height, vigor, growth stages, age classes, and
phenology. Plant mortality, injury to trees, and burn severity, all a direct function of
burning, also merit consideration because they directly or indirectly relate to postfire
effects. Definitions of each of these attributes will be given as monitoring methods
appropriate for each are discussed.

In addition to selection of the most appropriate methods of study, two other

considerations are vital for a successful monitoring program: control plots must be
established outside the area of the fire so that universal factors that may be influencing
results, such as climate or insect infestations, can be separated from effects of the fire
itself; and, timing of studies must be planned so that plants have reached maximum
growth, and repeat studies should be taken as near to the same time as possible as the
initial studies were conducted. For planned burns, control plots must be in similar
vegetative communities as the area to be burned in order to make valid conclusions
regarding fire effects.

Standard references discussing both the philosophy and methodology of vegetation

sampling include Cook and Stubbendieck (1986), Greig-Smith (1983), Mueller-Dombois
and Ellenberg (1974), Pieper (1973), and Brown (1954). A rangeland monitoring guide
describes in detail the more commonly used techniques for monitoring range trend,
some of which may be valuable in monitoring fire effects (USDI-BLM 1985a). Additional
guidance for prefire and postfire vegetation monitoring is found in USDI-NPS (1992).

The matrix in Table VI-1 relates the specific effects of fire on vegetation to measurable
vegetation and site attributes, so that appropriate methods of study can be most
efficiently chosen for the effects to be measured. In designing any sampling scheme,
the community type being sampled must be considered in determining which
methodology to be employed, as well as size and shape of plots to be used. Chambers
and Brown (1983) outline appropriate quadrat sizes and shapes for specific
methodologies in a variety of vegetation types. It is important to work with qualified
personnel to design valid sampling schemes and methods of analysis. (See Chapter XI,
this Guide, for a discussion of sampling and statistical analysis.)

1. Cover. Cover refers to the area on the ground covered by the combined aerial parts
of plants expressed as a percent of the total area. Specifically measured are either
basal cover, which is the vertical projection of the root crown on to the ground, or foliar
cover, which includes the projection of all plant parts vertically on to the ground. Cover
of litter, rocks, or any other physical parameter on the ground may be determined using
cover measurements.

Points and point frames are also used to measure cover (Chambers and Brown 1983;
Floyd and Anderson 1983), and are particularly suited to dense or rhizomatous
vegetation where intensive sampling is desired. A disadvantage of using points is that
an extremely large number of points must be collected to obtain a representative
sample of the population. Specific methods include vertical and inclined point frames,
points along line transects, and pace transects. First hit only or all hits through the
various canopies may be recorded. Usually, aerial canopy is used for trees, shrubs and
broadleaf perennial forbs, and basal crown is used for grasses and single-stemmed
forbs. Where vegetation is identified by layer, cover may exceed 100 percent. Cover
using quadrat frames has been employed by Daubenmire (1959). The method uses
canopy coverage classes for each species within a given frame.

Table VI-1: Vegetation and Site Attributes Useful for Evaluating Selected Fire Effects.

FIRE EFFECTS

Reproduction
ATTRIBUTES Mortality Productivity Structure
Sprouts Seedlings
Cover I* I D
Density D** D D
Frequency I I I
Weight I D
Species Composition D D I
Number D D
Height I D
Crown scorch D/I*** I*** I
Crown consumption D*** I*** I D/I
Stem char D/I*** I***
Burn severity D/I*** D/I*** I I D/I

* I = Indirect Relationship
** D = Direct Relationship
*** = Depends upon species

2. Density. Density is the number of plants or parts of plants per unit area, although
older literature may use the term to refer to the attribute of cover. Density is highly
useful and frequently used for evaluating effects of fire on mortality and reproduction. It
is generally straightforward, easily measured and readily understood. Density is not
particularly useful in describing community structure and the relationship of species
importance to one another, but can be highly valuable in tracking response of individual
species to fire. For example, the methodology may measure the number of seedlings,
shrubs, or trees per unit area. Response of rhizomatous or suckering plants, such as
Western wheatgrass or aspen, may be measured in terms of stems or ramets per unit
area with this methodology also. Sprouting shrubs are often tracked using density
measurements before and after controlled burning.

Quadrats used to sample density may be small frames or large plots, depending on
size and abundance of the species studied. Species to be monitored must initially be
defined; size of plots will follow so that the physical sampling does not become
cumbersome. Belt transects to measure density of shrubs before and after burning are
frequently used in rangeland situations.
3. Frequency of Occurrence. A quantitative expression of the presence or absence of
individuals of a species in a population is termed frequency of occurrence, or simply
frequency. It is the ratio between the number of sample units that contain a species and
the total number of sample units. Its sensitivity in accurately reflecting the population
parameters is a direct function of the size of the quadrat used for sampling. Because
this method does not measure any plant or plant community attribute directly, its
usefulness in fire management has been somewhat limited. Frequency is an integrator
that encompasses plant size and shape, density, distributional patterns, number, and a
host of other physical attributes. Because of the often severe nature of fire's effects on
a plant community, wide swings in frequency values may present problems in both
sampling and interpretation of results.

Quadrat sizes used to sample frequency will vary based on vegetative characteristics
and the size and distribution of the species being sampled. Frequency values between
20 and 80 percent are considered necessary to both describe the plant's occurrence
and detect change over time. Smith and others (1986) describe a nested frequency
configuration to sample more than one species in a specific series of transects. If a
single frame is used, it is critical that the quadrat size used to collect the initial set of
frequency data be used on all subsequent data collections so frequency values are
comparable. Statistical analysis cannot be conducted if quadrat sizes have been
changed during the course of monitoring.

4. Weight. A measure of the mass of some aspect of an ecosystem may be defined as

weight. As used in both ecological and fire literature, the term needs considerable
redefining to be understood and thus useful. The term biomass refers to the total
weight of living plants and animals above and below ground in a given area at a given
time. Because total biomass is obviously beyond normal capabilities to measure,
aboveground plant biomass becomes a standard reference in describing a weight
aspect of plant communities. It is the total amount of living plants above the ground in a
given area at a given time. A virtually synonymous term, standing crop, may be used as
both a time and weight indicator of biomass, and refers to the total amount of living
plant material in aboveground parts per unit of space at a given time, with particular
emphasis on the specified date. Vegetation samplers often seek to measure peak
standing crop, that being the maximum amount of living tissue when accumulation is
greatest. Aboveground phytomass, which includes dead attached parts, is a standard
expression for all organic plant parts in a specified area.

Aboveground phytomass data are useful for fire managers particularly for writing
prescriptions and understanding fire behavior. The fuel load, or aboveground
phytomass, carries the fire; knowing the precise fuel load prior to burning not only
contributes to designing a successful fire, but permits evaluation of observed fire
behavior and results of the prescription. In addition, an objective of many prescribed
burns is to increase the yield of specific species, or groups of species such as grasses,
on a site. Weight data must be collected in order to evaluate success in meeting the
objective.

Methods employed to sample weight include clipping of quadrats, estimates, double

sampling, use of height-weight curves, and use of capacitance meters. Dense, uniform
vegetation requires fewer quadrats for clipping studies. The vegetation is clipped to a
specific height in a specified size quadrat, air or oven-dried, and weighed. For total
aboveground plant phytomass, everything but unattached litter is clipped. Litter also
may be added if total fuel loading is desired. For aboveground plant biomass, also
referred to as current year's production or current annual growth, all dead material must
be removed from the material clipped. Weights may be estimated in the quadrats, or a
double sampling scheme may be employed whereby some plots are clipped and some
are estimated, and the actual weights from the clipped plots used to adjust estimated
weights in the nonclipped plots. For shrubs, regressions of crown volume, stem
lengths, or stem diameters have been used to estimate current annual production.
Height-weight curves have been developed by researchers, particularly for individual
grass species, based on the relationship of plant height to weight in the various
segmented portions. Capacitance meters, which require recalibration for each site and
sampling date, rest over vegetation to be sampled and, using the difference between
the dielectric constant of herbage and air, estimate weight of underlying vegetation.

Actual clipping or sampling yields good information that can provide more than weight
data alone. Fuel moisture content can be calculated; species can be sampled
individually or lumped into categories; and botanical composition by weight can be
obtained. The attribute is useful not only in determining changes in productivity on an
area, but may provide information on kinds and amounts of wild and domestic animal
use that may be expected in a given area.

5. Species Composition. Species composition is a term relating the relative

abundance of one plant species to another using a common measurement. It is defined
as the proportion (percentage) of various species in relation to the total on a given
area. It may indicate the relative importance or influence of one species to another in a
specific physical setting and is a reflection of structure and hence wildlife habitat.
Composition can be determined by weight, cover, number, or other basic variables, and
should be reported as such, e.g., percent composition by weight. Generally, it is an
attribute arrived at indirectly. The attribute sampled may have been cover, but in order
to understand the relationship of species to one another, the relative percents by
species are calculated based on cover measurements. In a burn situation, data may be
somewhat misleading if caution is not exercised in interpretation. For instance, the
entire shrub or tree component may be eliminated, resulting in a dramatic increase in
the herbaceous component on a percent composition basis, although no real increase
in number of plants or volume of plant material produced may have been realized.

6. Number. Number is the total population of a species or classification category in a

delineated unit and is a measure of its abundance. The attribute is most valuable when
dealing with small numbers or particulars. In fire situations, actual counts may be
important to know for a scarce resource that may be affected by the fire. Threatened or
endangered species may be counted in their entirety, or numbers of snags before and
after prescribed burns may be noted. Because the attribute does not involve a sample
but rather the entire population, no sampling techniques except sheer counting can be
described.

7. Height. Height is the vertical measurement of vegetation from the top of the crown to
ground level. In herbaceous vegetation, it may be an indirect indication of productivity.
(See 4. Weight) Changes in height and hence changes in structure are some of the
most important vegetation characteristics used in determining suitability of areas for
various kinds of wildlife. Methods used for measuring height include the Biltmore stick,
clinometer, Abney level, and Relaskop.

8. Vigor. Vigor relates to the relative robustness of a plant in comparison to other

individuals of the same species and may vary with site, climatic conditions and age. It
can be a subjective assessment of the health of individual plants in similar site and
growing conditions based on general observations, or it can be more completely
defined with some kind of "measurement" of vigor, e.g., references to seed stalk
production per plant or unit area, number of tillers produced per plant or unit area,
number of leaves or stems, and so forth. Vigor also can be reflected by the size of a
plant and its parts in relation to its age and the environment in which it is growing. To
be a useful attribute to measure for fire effects, definition of what is to be measured
must be made prior to the fire, so that the term maintains a modicum of objectivity. The
phenological phase of the species under observation must be the same during each
evaluation in order to accurately assess and compare vigor. No matter how carefully
measurements are standardized, vigor is considered subjective and is based generally
on indirect measurements which may or may not relate to the actual vigor of the plant.

9. Growth Stages and Age Classes. Growth stages are the relative ages of
individuals of a species usually expressed in categories. Examples of such categories
are seedlings, juvenile (young), mature, and decadent plants. Age classes define in
more discrete units the ages of individuals, such as 0 to 5 years, or 6 to 20 years. Age
classes may be difficult to determine in herbaceous vegetation, succulents, and any
vegetation that does not produce definable growth rings. Both density and frequency
measurements outlined above may be made within the parameters of growth stages or
age classes, so that the observer may catalogue postburn changes in reproductive
capabilities of a site or in effects of the fire on the diversity reflected in different age
structures.

10. Phenology. Phenology refers to the timing of various growth and reproductive
phases of vegetation. It is based on yearly growth patterns of individual species. A wide
variety of phases may be described and then traced for individual species as the
growing season progresses (West and Wein 1971). For example, recording the time of
initiation of spring growth may be valuable to assess the effects of fire on early growth
before and after burning. Other phenological phases frequently recorded include time
of blooming, time of seed set, initiation of new terminal bud (signalling the end of
seasonal stem or leader growth) and time of dormancy. Mechanics of tracking
phenology simply involve delineating the growth phases one is most interested in and
then charting them as the season progresses.

11. Injury to Trees. As described earlier in this chapter (B.1.a.), percent crown scorch
and percent crown consumption can be good predictors of mortality for many tree
species. These can be assessed by estimating or measuring (as with an Abney level)
the total length of the tree crown and the length of crown scorched or consumed.
Monitoring of damage to tree stems may be needed to better understand the cause of
tree death. Height of stem char is measured on all sides of the tree. Depth of char
might also be a useful measure of injury on thick barked trees. If bark was consumed or
is sloughing off, this should also be noted.

12. Plant Mortality. The cataclysmic effects of fire frequently result in mortality of
vegetation. Conversely, in many situations it is of great value to know if plant mortality
has been slight following fire.

a. Tagged individuals. A quick and easy way to assess mortality is to tag individuals
prior to the fire. Pieces of tin, numbered metal tags, metal stakes, and other
nonflammable materials should be used adjacent to the individuals to be checked
postburn. A mapped layout of the plot will permit rapid relocation following the fire. It
may be necessary to monitor mortality for several years because it may take that long
for injured plants to die. Mortality of aboveground portions of shrubs, grasses, forbs,
and some species of trees can be visually determined. For many nonsprouting species,
death of the main stem, such as of a big sagebrush, is readily apparent, and indicates
that the entire plant is dead.

Some species on the burned area are capable of producing vegetative regrowth from
buried plant parts. A plant that appears to be dead immediately after the fire may sprout
the following growing season. In some cases, new growth must be excavated to
determine if a plant has vegetatively reproduced or if seedlings have established.

b. Chemical tests. Chemical tests can be performed to assess the death or survival of
individuals of important species.

(1) Tetrazolium. Tetrazolium tests were developed to determine the degree of seed
viability, but have been a standard mortality test for range situations. This chemical
tests for hydrogen (dehydrogenase) that is released by plant tissues during respiration.
Strong, healthy tissues develop a red stain; dead tissues remain their original color.
Detailed procedure for testing grass tissue and grass seeds are given in Stanton
(1975). The basic procedure is to soak the tissue of interest in a one percent
tetrazolium solution, and place it in the dark. Results show up in 5 to 6 hours or
overnight. Although this test has long been used, there are associated problems.
Procedures for seeds vary by species, and are best performed by experienced
analysts. Any sample being tested must be put in a closed container in the dark,
because bright sun can affect tetrazolium and cause the same color change as occurs
in the presence of dehydrogenase. Results take several hours to appear. On dark
tissue, such as Idaho fescue meristems, the color change may not be visible. The
interpretation of results can be a problem, because red stain may indicate something
other than active metabolism.

(2) Orthotolodiene-peroxide. The chemicals orthotolodiene and peroxide are used

sequentially to test for the enzyme peroxidase, found in most living plant cells. This test
has been successfully used on trees, and should work well on shrubs. While no
documentation of its use on grasses has been found, peroxidase should be present.
However, it is not known if peroxidase is present in sufficient quantities in dormant or
quiescent grass tissues to stain blue. The basic procedure is described in Ryan (1983).
For trees and shrubs, a piece of cambium is extracted with an increment borer (the
preferred approach), or exposed by scraping away the bark. An eyedropper is used to
cover the sample with a one percent orthotolodiene solution, and then peroxide is
applied. Live tissue will turn bright blue within a few moments. A reddish purple color,
followed by the appearance of a blue color, also indicates life. A greenish blue color
probably means dead tissue. After using this technique for a while, the colors that
indicate dead or live cambium become readily recognizable. This test is preferable to
the tetrazolium test because it can be used in the field and provides almost immediate
results. However, caution is necessary because orthotolodiene has been found to be
carcinogenic in laboratory studies. Gloves should be worn as a precaution.

(3) General comments. Metabolic by-products being tested for may not break down
until a few days after a fire, even though the plant is dead. The proper location on the
plant must be tested to determine mortality. On coniferous trees with living foliage, the
cambium should be checked, and also the roots, if much heating occurred at the base
of the tree. Trees and shrubs sprout from different locations on stems, root crowns, and
roots, and it is these sprouting sites which should be tested to indicate whether the
shrub may sprout. Grass crowns should be tested where the buds and reproductive
meristems are found.

More than one test may be necessary per plant, because plants can survive some
amount of fire damage. Tree cambium requires a test on all sides, and a shrub at
several sprouting sites. Unburned meristems and buds of bunchgrasses should be
tested at both the center and edges of each plant, and at different depths below the
surface if the buds occur below the ground surface.

13. Burn Severity. Burn severity (discussed in II B.6.e), while not an attribute of
vegetation, is an exceptionally good predictor of fire effects on vegetation. It indirectly
measures the heat pulse below the surface, and provides an indicator of fire impacts on
buried plant parts. Burn severity classes can be developed that apply to the type of
vegetation and soil organic layer characteristics on the site being investigated. The
degree of burn severity can be assigned to one of five classes, including "unburned",
"scorched", and "light", "moderate", and "high" severity. Definitions for the latter three
classes can be based upon the information in section B.2.c. in this chapter. Burn
severity can be described as a percentage of area on plots of a specific size, or related
to specific inventory points. Although a qualitative measure, this descriptor can be
related to plant mortality, and amount and mode of reproduction, such as by rhizome
sprouting or seed germination.

Burn severity classes have been developed for bunchgrass plants. (See VI.B.2.d.(4))
Monitoring the relationship of these classes to postfire mortality or production of
specific species can provide a valuable tool for predicting postfire grass response when
considering emergency fire rehabilitation, or developing prescriptions for prescribed fire
use.

14. Moisture Conditions. The heat regime of a fire depends on the amount and
condition of the fuel on the site, how it burns, and the duration of burning. In order to
build a database that can be used to predict plant response to fire, moisture conditions
at the time of a fire must be documented, because moisture levels are a key regulator
of heat release during a fire. See Chapter III.B. and III.D. for a more complete
discussion of fuel moisture content and how it is measured.
15. Postfire Weather. Vegetation response to fire can be dramatically affected by
postfire weather, particularly in regions with arid or semiarid climates. Knowledge of
postfire weather, especially precipitation, can often explain much of the measured or
observed variation in postfire effects.

E. Summary

Plant response to fire is a result of the interaction of the behavior and characteristics of
a fire with the characteristics of a plant. Plant community response is a product of the
responses of all plants on a burned area. The response of an individual species of
plant, or plant community, can vary among fires or within different areas of one fire.
This is because of variation in fuels, fuel moisture conditions, topography, windspeed,
and structure of the plant community itself, causing the heat regime of a fire to vary
significantly in time and space. The immediate effects of fire can be modified by postfire
weather and animal use. Fire can cause dramatic and immediate changes in
vegetation, eliminating some species or causing others to appear where they were not
present before the fire. Monitoring techniques that are used to detect trend in
vegetative communities are often not appropriate, either because they are not sensitive
enough to detect the changes that have occurred, or provide statistically inadequate
samples. Fire effects on plants, and plant response to fire treatments are predictable if
the principles and processes governing plant response are understood. If burning
conditions, the fire treatment, and vegetation response are properly monitored, the fire
effects that are observed can be interpreted, and our ability to predict fire effects on
plants will increase.

This page was last modified 05/31/01

|Disclaimer| | Privacy| | Copyright| |USFWS Main Page| |Webmaster|

 National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 05/31/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home CHAPTER VII - TERRESTRIAL WILDLIFE AND HABITAT

Preface
Objectives By Loren Anderson
Fire Behavior
Fuels
A. Introduction
Air Quality
Soils & Water
Plants Fire effects on terrestrial wildlife and their habitat are addressed in this
Wildlife chapter. Too many variables are involved with fire, wildlife and wildlife
Cultural Res. habitat to allow a "cookbook" approach. The underlying ecological
Grazing Mgmt. relationships of vegetation and wildlife are briefly described. A
Evaluation discussion on how fire may subsequently influence those relationships
Data Analysis through effects on food, cover, water, and space follows.
Computer Considerations for managing and monitoring fire effects on wildlife
Soft. habitat are also offered.
Glossary
Bibliography There is an increasing literature base available regarding fire-wildlife
Contributions relationships. Four publications the reader may find of particular interest
are: Effects of Fire on Birds and Mammals (Bendell 1974); Effects of
Fire on Fauna (Lyon et al. 1978); Fire: Its Effects on Plant Succession
and Wildlife in the Southwest (Wagle 1981) and Fire in North American
Wetland Ecosystems and Fire-Wildlife Relations: an Annotated
Bibliography (Kirby et al. 1988). This last document includes an
extensive bibliography of all literature on fire-wildlife relations indexed in
Wildlife Review 1935 to 1987.

The subject of wildlife, habitat and fire is burdened with generalizations

and ambiguities. The definition of wildlife varies from a taxon of
invertebrates to an entomologist, a Life Form (Thomas 1979) or Guild
(Short 1982) to an ecologist, or a full curl bighorn ram (Ovis canadensis)
to a hunter. Evaluation of the quality of habitat varies depending on the
perspective. Fire is frequently given the anthropomorphic rating of
"good" only because fire is considered a natural phenomenon. There is
a segment of the public which confuses the fact of fire with the effect of
fire and envisions only death, destruction, and loss. Cool burn, hot burn,
spring burn, and fall burn are terms that have definitive meaning only to
the person using them. Generalizations regarding fire effects on
vegetation also can be misleading. Species such as bitterbrush
(Purshia tridentata) are frequently credited with being so severely
harmed by fire that they should be given complete protection.
Ultimately, however, many of them are dependent on fire or some
similar disturbance (Bunting et al. 1984). Diversity and mosaic are two
other commonly used terms that frequently generalize to the point of
being meaningless. The widely held assumption that increased edge is
always beneficial is not uniformly accurate (Reese and Ratti 1988). As
there are no averages in nature (Vogl 1978), there are no
generalizations that can stand alone. Terms must be clearly defined
and qualified appropriately.

Oversimplification of fire effects commonly occurs when data,

knowledge, time or initiative are lacking. Conversely, there is the
attitude that if it is not complicated and difficult to understand, it is
invalid (Szaro 1986, Vogl 1978). Providing for (as with prescribed fire)
and assessing fire impacts on wildlife habitat consistent with general
ecological concepts can assist in addressing many complex
relationships in a more simplified manner while still retaining a level of
validity (Vogl 1978).

B. Principles and Processes of Fire Effects

Adhering to ecological principles and processes is recognized as

essential in preserving functional systems (Dubos 1972, Wilson 1985).
More accurate effect prediction and assessment, increased assurance
of success with prescribed fire, overall cost, and production efficiency
are a few of the other benefits derived from thinking and acting as
"ecologically" as current knowledge allows (Allen 1987, Chase 1988,
Graul and Miller 1984, Savory 1988, Yoakum 1979).

Implicit is the caveat to keep all the pieces (Allen 1987, Chase 1988,
Lyon and Marzluff 1984). What we may initially deem insignificant (e.g.,
mycorrhizal fungi) could be of ultimate importance in maintaining the
stability of the ecosystem (Watt 1972, Wilson 1985). How important may
lichens (Rhizoplaca spp.) be to the nutritional health of species such as
the pronghorn (Antilocapra americana)? What does the insidious loss of
forb diversity and abundance portend? Our imperfect knowledge of
ecosystem dynamics and the ramifications of our actions make it
imperative we retain the pieces and therefore, in many cases, options.
Disruption and loss of indigenous ecosystems have occurred over
extensive areas of the west through a combination of inadequate
management, fire, and highly competitive exotics such as cheatgrass
brome (Bromus tectorum) (Wright and Bailey 1982).

1. Floral Response.

a. Ecological basis. Much of the literature regarding fire effects, wildlife

and wildlife habitat revolves around successional theory, thus, it is
important to understand the concept. Floral succession - that somewhat
orderly progression of occupancy by successively higher ecological
order plant communities - is one of the primary descriptors of the natural
environment. The progression, however, from a "pioneer" stage through
various seral stages to that mostly esoteric end point called "climax," is
more easily addressed in theory than fact. Although the successional
trajectory is often portrayed as following a predictable path and
timeframe, there are many who question that view (Bendell 1974,
Dubos 1972, Leopold and Darling 1953, MacMahon 1980, Watt 1972).
The observation that some species such as gambel oak (Quercus
gambelii) are successional in one environment and climax in another
further complicates the picture (Harper et al. 1985, Whittaker 1975).
The "orderliness" of succession we like to envision is commonly
disrupted and altered by stochastic events we can neither anticipate nor
control (Rosentreter 1989). The apparent controversy does not
invalidate successional theory, but it does point out the limitations of our
knowledge and the need for being inquisitive and prudent.

b. Structural development. Habitat structure follows successional

trends in most plant communities. Short fire intervals tend to maintain or
promote early successional conditions typified primarily by herbaceous
species and comparatively limited structural diversity. Long fire intervals
favor community development along the successional trajectory. This
normally results in increased woody species development and greater
horizontal and/or vertical structural diversity.

c. Postburn plant community.

(1) For the most part, preburn plant composition and the individual plant
species response to fire determine membership of the initial postburn
floral community. Understanding plant survival mechanisms (see VI.B.,
this Guide) is essential for assessing wildfire effects on habitat and in
providing for desired prescribed fire effects.

(2) Plants stressed through drought, disease, insect infestations,

overgrazing, old age or a combination of these factors are likely to be
negatively impacted by burning regardless of how they would respond if
healthy (Bunting 1984, DeByle 1988b, Wright and Britton 1982).
However, trees that die as a result of fire can provide an important
habitat component for certain species.

(3) Herbaceous production increases occur most often on range sites in

high Fair or better ecological condition (Bunting 1984, West and Hassan
1985).

(4) Various animal species disseminate seed and can influence

subsequent floristic makeup of an area. Pinyon jays (Gymnorhinus
cynanocephalus) have been credited with the ability to

transport upwards of 30,000 pinyon (Pinus spp.) seeds per day up to 6

miles (9.7 kilometers). Both birds and small mammals are considered
instrumental in the expansion of pinyon-juniper woodland (Evans 1988).
Small mammals, such as chipmunks (Tamias spp.), disperse spores of
mycorrhizal fungi into burned areas, an essential component for the
establishment and survival of many plant species. (See V.B.c.(4).)

(5) Increases in plant nutrient density, palatability and earlier "greenup"

are not unusual occurrences (Bendell 1974, Daubenmire 1968a, Leege
and Hickey 1971, Wagle and Kitchen 1972). Earlier greenup of burned
areas is largely a function of heat absorption by the dark ash and
resultant increases in soil temperature. Reduced soil shading is also a
factor. Plants surviving a fire take advantage of higher nitrogen levels
provided by the ash and remain greener, more nutritious and more
palatable for a longer period of time (Brown 1989). However, this
phenomenon is normally shortlived. Commonly, available nitrogen
returns to normal levels within two to three growing seasons. Personal
observation (Anderson 1983) indicated elevated crude protein levels for
7 years following burning on a bighorn sheep winter/spring range. The
extended period of increased protein levels was believed due to
subsequent utilization made by bighorns rather than from burning,
however.

2. Faunal Response.
a. Ecological basis. Faunal succession follows floral succession, i.e., a
given set of vegetational conditions provides habitat for a more or less
distinct collection of wildlife species (Bendell 1974, Burger 1979,
Dasmann 1978, Evans 1988, Huff et al. 1984, Smith et al. 1984, Wolf
and Chapman 1987). Maser et al. (1984), however, pointed out that
ecological distinctions between plant communities do not uniformly
correlate with differences in animal communities. That apparent
inconsistency is explained by the observation that wildlife species most
often select habitat on the basis of structure rather than plant species
composition. Thus, most of the literature discusses faunal response in
terms of general vegetational structure and successional stage.

Fires which set succession back to a grass/forb stage primarily benefit

herbivores (vertebrates and invertebrates) and those species for which
herbaceous vegetation is desirable for cover. Various vertebrate and
invertebrate predators of herbivores also may benefit (Beck and Vogl
1972, Bendell 1974, Hansen 1986, Huff et al. 1984, Lyon and Marzluff
1984, Lyon et al. 1978, McGee 1982). Red foxes (Vulpes fulva), gray
foxes (Urocyon cinereoargenteus), and weasels (Mustela spp.) are
associated with early to mid-successional stages and the ecotones
between these stages and climax vegetation communities (Allen 1987).
Whitetail deer (Odocoileus virginianus), bobwhite quail (Colinus
virginianus) and cottontails (Sylvilagus spp.) are common to early-mid
stages (Dasmann 1978). Fagen (1988) notes a dependency on old
growth timber, especially during years of heavy snowfall, by Sitka black-
tailed deer (Odocoileus hemionus sitkensis). Muskrats (Ondatra
zibethicus) are favored by early successional conditions (Allen 1987).
The spotted owl (Strix occidentalis) is primarily an inhabitant of old
growth forests (USDI-Fish and Wildlife Service 1989). As broad as the
preceding descriptions may be, they do provide a relative perspective of
floral/faunal relationships. Habitat manipulation almost invariably
involves efforts to either set back, retard or accelerate plant succession
(Burger 1979, Huff et al. 1984, Wolf and Chapman 1987). Most animal
species are more cosmopolitan in their use of various successional
stages and structural conditions for feeding than for cover, escape or
reproduction (Dealy et al. 1981, Maser et al. 1984).

b. Structure.

(1) The consistency with which the value of structure is referred to in the
literature (e.g., Allen 1987, Geier and Best 1980, Harris and Marion
1981, McAdoo et al. 1989, McGee 1982, Smith et al. 1984) gives
credence to the assumption that structure is possibly the single most
important habitat clue to which wildlife responds. Structure may indicate
a feeding site for one animal. For another, that same structure may be
sought as nesting cover. The perceptual environment differs from
species to species and a structural "clue" for one may mean something
entirely different to another species (Dubos 1972). Although specific
plant species are frequently insignificant (Johnsgard and Rickard 1957,
McAdoo et al. 1989), vegetational complexity associated with
developing structural diversity is significant, particularly when faunal
species richness is the objective (Campbell 1979, Germano and
Lawhead 1986, Johnsgard and Rickard 1957, Short 1983, Willson
1974). A variety of "clues" accommodates interspecies variance and
automatically denotes a variety of potential habitat niches. Addressing
Great Basin habitats in southwestern Utah, Germano and Lawhead
(1986) found postbreeding bird diversity significantly correlated with
vertical habitat layering and that diversity of both rodents and lizards
correlated with horizontal habitat heterogeneity. Birds make more
efficient use of habitat volume or vertical space than other classes of
wildlife. Although structure is not height limited, Hall (1985) equated
herbaceous stubble of 1 to 2 inches (2.5 to 5.1 centimeters) to bare
ground for many species that nest and feed on the ground.

(2) The addition of a single structural element to a plant community can

greatly enhance faunal species diversity (Germano and Lawhead 1986,
Willson 1974). Maser and Gashwiler (1978) reported that the presence
of pinyon and juniper trees provided habitat for four additional Life
Forms that would otherwise not be present. Conversely, the loss or lack
of a single structural component can eliminate some species and be
just as lethal as a direct mortality factor (Knopf et al. 1988, Lyon and
Marzluff 1984, Wecker 1964).

(3) Habitat and community structure can include dead as well as living
components. Large diameter logs provide habitat in the form of travel
routes, as well as feeding, nesting, and reproduction (Bartels et al.
1985). Snags are critical nesting and feeding habitat for many species
of birds, as well as mammals and amphibians (Neitro et al. 1985).
Insectivorous birds that inhabit snags not only harvest insects in
recently burned areas, but help regulate populations of insects in
adjacent unburned areas, and in the newly developing forest (Wiens
1975 in Neitro et al. 1985).
(4) The interplay between only one or two structural components and
subsequent wildlife use is illustrated in the following examples. In a
sage (Artemisia spp.) shrubsteppe, horned larks (Eremophila alpestris)
and meadow larks (Sturnella neglecta) are favored primarily by a grass
stage. Sage (Amphispiza belli) and Brewer's sparrows (Spizella
breweri) are favored by a shrub stage. All four species can exist in
nearly a 1:1 ratio in a mixed grass/shrub type. Lark sparrows
(Chondestes grammacus) appear to benefit more from the mixture than
either the grass or shrub stage alone (McAdoo et al. 1989, Rotenberry
and Wiens 1978). Rotenberry and Wiens (1978) noted that horned larks
replaced sage sparrows as dominant following a burn that eliminated
sagebrush. Pronghorn have a somewhat similar potential response to
the lark sparrow's. They, too, are benefitted most by a combination of
herbaceous and shrub components. Their habitat quality drops rapidly if
either component is depauperate or, in the case of sagebrush, too tall
or dense (Yoakum 1980).

(5) The Life Form (Thomas 1979, Maser et al. 1984) and Habitat Guild
(Short 1982) concepts are organized around the relationship between
plant succession and resultant structural conditions, and wildlife. The
basic premise is that there are groups of animals with similar ecological
requirements that are met by similar successional stages of the plant
community. Not unlike successional theory, there are some who
question aspects of that approach (Block et al. 1987, Szaro 1986). The
Life Form approach to assessing and predicting effects (see Maser et
al. 1984) can be of considerable value, however, when large scale fires
are addressed. It is important to recognize that within a given Life Form,
a full range of species adaptability and species-specific niche
requirements may be encountered. The habitat needs of individual
species in a given Life Form may have to be scrutinized to assure
adequate consideration is given sensitive animals.

c. Species adaptability. Animals that are broadly adaptable

behaviorally and in their habitat preferences can accommodate change
more efficiently -- they have more options (Dubos 1972, Knopf et al.
1988, Vogl 1978, Watt 1972, Wecker 1964). These species are
frequently termed generalists, species of high versatility (Maser et al.
1984) or eurytopic (Knopf 1988). Deer mice (Peromyscus maniculatus)
and the ubiquitous coyote (Canis latrans) are two common generalists.
At the opposite end are species that have a narrow range of
environmental conditions under which they can survive and flourish. At
the extreme are the species that appear on "sensitive" or threatened
and endangered species lists. Species of specialized adaptability are
commonly termed specialists, of low versatility (Maser et al. 1984),
obligates (Kindschy 1986) of a particular habitat component, or
stenotopic (Knopf et al. 1988). Hammond's flycatcher (Empidonax
hammondii) has very narrow food and cover requirements (Maser et al.
1984). The sage grouse (Centrocercus europhasianus) and pinyon jay
are considered obligates of sagebrush and pinyon-juniper woodland
respectively (Kindschy 1986, Hardy 1945). Allen (1987) notes that many
furbearers of forested and wetland cover types have specific habitat
requirements and are less resilient in adapting to habitat modifications.
Specialists commonly can be eliminated by loss of a single habitat
component. Species of intermediate adaptability such as robins (Turdus
migratorius) and red-wing blackbirds (Agelaius phoeniceus) are referred
to as mesotopic (Knopf et al. 1988). Knopf et al. (1988) separated
riparian avifauna into eurytopic, mesotopic, and stenotopic guilds to
accommodate variations in habitat sensitivity.

d. Food and cover. The two most visually obvious determinants of

habitat suitability are food and cover.

(1) Dense timber travel lanes are frequently preferred by elk (Cervus
elaphus). Sagebrush can reach heights and densities that inhibit or
prevent pronghorn movement. Voles (Microtus montanus;
Clethrionomys gapperi) require certain litter layer or woody debris
habitat components. Birds require various structural conditions for nest
sites, hunting and song perches. Some species (e.g., the white-tail
deer) select for denser woody vegetation. Species such as bighorn
sheep or pronghorn normally select against it (Lyon et al. 1978, McGee
1982, Yoakum 1980).

(2) One species may consume primarily grass and forbs (e. g.,
grasshoppers or elk), another mostly forbs and shrubs (e.g., sage
grouse or pronghorn) and yet another, such as turkey (Meleagris
gallopavo), may make heavy use of mast. Hobbs (1989) provides a
strong case that the habitat mule deer (Odocoileus hemionus) need for
thermal cover correlates well with the nutritional plane of the animal.
Hobbs and others have noted the inherent value of forage diversity and
availability. Hobbs and Spowart (1984) found substantially improved
winter diet quality for both deer and bighorn sheep as a result of burning
although there were only relatively small changes in the quality of
individual forages. They attributed the diet quality increase to improved
availability of forage items and enhanced forage selection opportunities.
It is commonly assumed that increased herbaceous production
automatically occurs after burning.

There is sufficient documentation, however, indicating production

increases are not a foregone conclusion (Peek et al. 1979, Yeo 1981
and others). Certainly, some highly valued food items can be eliminated
if burning is too frequent. For example, McCulloch et al. (1965),
determined that Gambel oak produce very few acorns before reaching 2
inches dbh (diameter at breast height) (5.1 centimeters) and that
maximum production of mast probably does not occur until healthy
stems are 12 to 14 inches (30 to 36 cm) in diameter. Management of
Gambel oak for both mast production and deer winter browse may not
be possible on a given site.

(3) The differential effect fire has on wildlife is typically noted in a

realignment of species as some species become favored over others as
a result of changes in abundance of food and cover. For some classes
of wildlife (e.g., birds and small mammals), stability of total numbers has
beennoted, even though species composition changed (Lyon and
Marzluff 1984, Lyon et al. 1978, McGee 1982). Bendell (1974)
explained changes in the kind and frequency of parasitic infections
following burning as likely a result of alteration of habitat structure and
cover favored by intermediate hosts. He noted a trend towards more
species of parasites in greater frequency of infection with longer time
after burning. He stressed, however, that there is a broad response
range and that any blanket statement regarding fire effects on parasites
must be qualified.

(4) Various efforts have been made to "codify" vegetational

requirements for a few species. Yoakum (1980) outlines some fairly
specific vegetational conditions that constitute quality pronghorn range
in shrub-grasslands. Autenrieth et al. (1982) and others have developed
similar recommendations for sage grouse. Allen (1987) and Parker et al.
(1983) note vegetational criteria for pine marten (Martes americana)
and lynx (Lynx canadensis) respectively. A growing number of plant
community/wildlife association listings are available. Most are built
around the Life Form (Thomas 1979) or Habitat Guild (Short 1983)
concepts that relate wildlife use to structural conditions. Maser et al.
(1978) address wildlife species of the western juniper type; Maser et al.
(1984), fauna of the Northern Great Basin; Allen (1987), furbearers;
Harper et al. (1985), wildlife of the oak brush type, and Thomas (1979)
and Brown (1985) address coniferous forest wildlife. Brown (1985) is
somewhat unique in the level with which fish and amphibians are
addressed.

e. Influence of time.

(1) Each wildlife species has a unique reaction to fire and the
subsequent ecological changes caused by fire. Some species exhibit an
almost immediate reaction. For others, behavioral time lags and site
tenacity may extend the response time nearly indefinitely (Wiens et al.
1986). Eurytopic or generalist species can likely accommodate the
change more efficiently than stenotopic or obligate species (Lyon et al.
1978).

(2) Immediate postburn effect assessments can be a poor reflection of

long-term animal use. For example, McGee (1982) noted that mountain
voles (Microtus montanus) could not sustain populations on severe fall
burns. Repopulation was contingent on the subsequent development of
an adequate herbaceous mulch layer. Red-back voles (Clethrionomys
gapperi) require woody cover and moisture conditions (Getz 1968) that
may be eliminated by fire. Four years following a fire, however, red-back
voles may be the most common mammal present (Moore 1989). A
decadent stand of bitterbrush under a pinyon/juniper overstory may be
providing valuable deer forage. Burning that habitat may be the only
hope for assuring any bitterbrush is available for some future generation
of deer (Bunting et al. 1984). Huff et al. (1985) found that the highest
diversity of birds occurred in a 19-year-old forest in the Olympic
Mountains of Washington. Moore (1989) stated that burned forest 5 to
10-years-old becomes prime habitat for chipmunks, possibly even better
than the original forest. He associated improved chipmunk habitat with
development of complex shrub layers.

Bunting et al. (1984) observed managers must not be so concerned

with the short-term effects that they lose sight of the future needs of
species. A fire-damaged tree, for example, may not die for several
years, but then provides important habitat for cavity nesting birds for a
long time. The species that use the snag change over time as the snag
decomposes. In many more years the snag falls, becoming a forest
floor log that is habitat for many other species.

f. Extrinsic/intrinsic influences. Both extrinsic and intrinsic factors

direct individual species response to fire effects. Extrinsic factors
include such items as food, cover, water, predators, and other elements
of the external environment. A noticeable population increase,
decrease, or shift in use patterns would indicate burning had modified
some extrinsic factor(s), creating or removing a limiting element, or
creating a more desirable (not necessarily required) condition.
Physiological, behavioral andgenetic characteristics are intrinsic factors
that play a large role in determining how a species responds to a
burned area. When no particular response occurs it may be assumed
the impact of fire was inconsequential or that there were overriding
intrinsic factors that inhibited or prevented a response (Bendell 1974,
Moen 1979, Wolf and Chapman 1987). The presence or absence of
one species can influence the presence, absence or habitat utilization
pattern of another (Bendell 1974, Peek et al. 1984). Extrinsic and
intrinsic factors are not mutually exclusive and the seemingly endless
combinations of the two can easily make an absolute determination of
fire effects virtually impossible in many cases. Peek et al. (1984) found
positive results from fire on seven different bighorn sheep ranges. They
indicated however, that in each case, definite proof was lacking
because they were unable to isolate the effects of fire from other
potential factors. Wiens et al. (1986) also noted confounding elements
associated with trying to evaluate the effects of small-scale burning on
shrubsteppe avifauna.

3. Burn Characteristics: Influence on Potential Faunal Response.

a. Size of burned area. A number of small burns produces more edge

than a single large burn. More edge is commonly assumed to provide
more benefit to more species of wildlife (Odum 1966, Thomas et al.
1979, and others). When particular species are considered, however,
the picture is not that clear. A number of 5 to 10-acre (2 to 4-hectare)
burns in a pinyon-juniper/ sagebrush-bunchgrass type might be relished
by mule deer but be of little or no value to pronghorn. Some species,
such as the western flycatcher (Empidonax difficilis) and brown creeper
(Certhia americana), are seldom found associated with edge and
actually may be harmed if edge is increased at the expense of
adequate forest interior (Rosenberg and Raphael 1986). The creation of
fragmented habitat for some species is a potential concern, especially
with large scale fires. Burned areas are a considerable attraction for
many herbivores. Larger species such as elk, moose (Alces alces),
bighorn sheep, domestic cattle (Bos taurus), and wild horses (Equus
caballus) are capable of overutilizing burns of insufficient size to
accommodate their demand.
b. Burned area configuration.

(1) Patchy or irregular burns can enhance habitat diversity, particularly

in an area with only one or a few communities all in the same structural
condition. Increased diversity and resultant increases in edge effect
makes more niches available for partitioning. Edge length, width,
configuration, contrast, and stand size largely determine the degree of
benefits. It must be recognized that diversity and edge cannot be
increased indefinitely. Beyond some threshold, the pieces become
sufficiently small and mixed that they assume a sameness or
homogeneity (Thomas et al. 1979). Also, increased fragmentation of
habitat components caused by maximizing edge can eliminate those
species requiring larger tracts or that inhabit stand interiors (Reese and
Ratti 1988). The amount and type of diversity sought with prescribed
fire is determined by the management goals and objectives. Maximum
diversity provides for species richness but is incompatible with an
objective to maximize a particular species. The reader is referred to
"Edges" by Thomas et al. (1979) for one of the more concise and
understandable treatments of diversity and edge. "Edge Effect: a
Concept Under Scrutiny" (Reese and Ratti 1988) is a good companion
treatise offering qualifying considerations of the edge concept.

(2) Areal extent, composition, and orientation of habitat components

(food, cover, water, space) determines habitat suitability for individual
species and/or groups of species. Juxtaposition, interspersion,
complexity, diversity, and mosaic are terms commonly used to reflect
the physical mix and patterning of edges, structural components, plant
communities, and seral stages. These somewhat generic terms give a
relative idea of the variety and positioning of resources within a given
area. Nearly mystical qualities have been attached to these terms and
they are frequently (and improperly) used without qualification. The
ideal mosaic for a soil-surface invertebrate obviously sayslittle about the
optimum mosaic for sage grouse. A sage grouse habitat mosaic has no
relationship to a mosaic promoting coniferous forest avifauna species
richness. A "good" mosaic is meaningful only within the context of a
specific management goal or objective (Thomas et al. 1979).

(3) The scale of the postburn configuration or mosaic is a major

consideration. Whether the components (structure, cover, openings) are
measured in square inches, acres or miles can dictate which species
benefit. Thomas et al. (1979) suggest that wildlife species richness
should be approaching maximum in rangeland settings where "average"
habitat size is approximately 200 acres (81 hectares). It is recognized
that habitat requirements of individual species determine the relative
potential benefit of a particular size burn where species richness is not
the primary objective.

A seldom discussed aspect of habitat quality is that of habitat

fragmentation (Reese and Ratti 1988, Rosentreter 1989). Small isolated
islands of shrubs and trees following prescribed or wildfire are examples
of fragmented habitat. Many wildlife species exhibit high extinction rates
in fragmented habitat (Wilcox 1980). Fragmented habitat fails to provide
areal extent and linkages between and among components that are
implied with "quality" mosaic, juxtaposition, interspersion and diversity
for a given species or collection of species. Adequate linkage of habitat
components (e.g., a stringer of cover connecting larger areas of escape
cover) is a determining factor for many species. As with other
wildlife/habitat considerations, some species are favored over others by
a particular mosaic or juxtaposition of elements and a few species may
be eliminated entirely.

c. Burned area location. The location of a burn relative to animal use

patterns can have a major influence on subsequent use. The proximity
of propagules or potential inhabitants has an influence on what species
may occur on a burn site. Species mobility also plays a part. A young
bighorn ram may travel miles, a shrew (Sorex spp.), hardly any.
Whether potential inhabitants can see, smell or be expected to wander
across a burned area may dictate the presence or absence of a
particular species. An otherwise "excellent" burn (prescribed or wild)
that is too distant from traditional use areas (e.g., bighorn) may not be
utilized in any reasonable timeframe. Proximity of the burn to a critical
habitat component such as water or cover also determines use. Sage
grouse exhibit a reluctance to use water sources devoid of adequate
surrounding cover. In contrast, pronghorn generally avoid water sources
screened with tall dense vegetation. Loss of a critical habitat component
and how soon - if ever - that element is replaced may be of paramount
importance. Burn location can strongly influence ultimate vegetational
establishment. A small stand of Douglas-fir (Pseudotsuga menziesii) on
a steep south exposure may - from a practical standpoint - never
regenerate whereas a similar stand on a more moist north exposure
may be restocked in comparatively few years. Slope, aspect, and
elevation affect snow deposition, snow crusting, thermal patterns, and
wind conditions on burned areas. All of these factors have a bearing on
habitat quality for a given species.
d. Completeness of burn. A fire of low burn severity and low fireline
intensity, which consumes comparatively little of the existing plant
community, may have no perceptible impact on wildlife. However, the
high severity, high intensity fire can significantly alter habitat makeup,
sometimes for an extended period of time. The first example may have
influenced plant community succession very little. The second could
result in a major adjustment of seral position potentially resulting in
significant changes in structure, cover and the forage base. It is
common to see both examples and many variations of the two on a
single fire.

e. Timing of burn. Timing of a fire relative to plant phenology is a

primary factor dictating the postburn plant community makeup.
Perennial herbaceous species, for example, are most resistant if burned
when completely dormant (Britton 1984, Bunting 1984). The size and
phenological stage of coniferous tree buds influences their resistance to
fire. Large buds such as on ponderosa pine (Pinus ponderosa) that
have scaled out are more resistant than small and/or scaleless buds
(Ryan 1988). (SeeVI.B.1.a., this Guide.) Negative impacts may occur if
a vital habitat element for a species is burned at the "wrong" time. For
example, a fall burn that consumes bighorn sheep winter range could
be disastrous, at least in the short term. If that fire occurred early
enough in the spring for regrowth, it may be beneficial (Peek et al.
1984). The effect of fire on birds nesting in residual herbaceous
vegetation can vary markedly depending on whether the fire occurred
before, during, or after nesting activities were completed.

4. Direct Mortality. Fire related mortality is popularly considered

insignificant and generally ignored. That generalization can be very
misleading - at least on a site specific basis. There is no reason to
believe various wildlife species have some superior capability to predict
fire behavior or to locate safety zones through dense superheated
smoke. Animals do die, apparently, most often through suffocation
(Lawrence 1966). At times, the number may be high. Quinn (1979)
reported that an intense burn eliminated all small mammal species but
the kangaroo rat (Dipodomys heermanni). Nelson (1973, p. 139) relates
a very graphic eyewitness account of large numbers of dead and dying
buffalo (Bison bison) that were caught in a prairie fire in the early
1800's. There are indications that severe burning can cause potentially
long-term reductions in some insects and other invertebrates of the soil
surface layer (Lyon et al. 1978). A long-term loss of these invertebrates
could be of particular significance in the altered and frequently
truncated ecosystems affected by man. Highly mobile species and
species that can escape underground or into rock crevices are least
subject to direct mortality (Beck and Vogl 1972, Lawrence 1966, McGee
1982, Starkey 1985). Direct mortality is unlikely to have much effect on
many species if the entire population or range of those species are
considered. In a particular geographic setting, however, the potential
significance of that loss should at least be considered.

5. Miscellaneous Considerations.

a. Noxious weeds and exotic plant species. Noxious weeds and

exotic plant species are an increasing concern. Any wild or prescribed
fire occurring or planned in areas subject to noxious plant invasion
should be evaluated from that standpoint.

b. Human effects on subsequent use. Fences, roads, human activity

and other similar factors, either on or off-site, can significantly influence
subsequent wildlife use.

c. Livestock use. Burns are an attraction to many animals and

livestock are no exception. Fires occurring on slopes less than about 35
to 40 percent may be subjected to heavy use by all classes of livestock.
Horses may make excessive use of burns regardless of location.
Livestock can easily influence potential wildlife values of a burned area
by altering plan responses, reducing herbaceous habitat structure,
removing forage, and merely by their presence. Control and
management of livestock is essential.

d. Snow crusting. Wind crusting of snow is a common problem on

some deer winter ranges. Carpenter (1976) documented that a
sagebrush canopy disrupts wind crusting in addition to providing
frequently melted out areas around larger plants. Sagebrush stands
approaching 20 inches (51 centimeters) in height have been found to
collect up to 1 inch (2.5 centimeters) more water in the form of snow
than open grassland (Hutchison 1965). Haupt (1979) noted that small
burns in climax coniferous forests accumulate more snow than
unburned forest. He found a similar result on larger burns if there was
residual tree cover. However, if burn size exceeded four times the
height of surrounding tree cover, snow accumulation could be reduced
through wind scour.
e. Indirect effects. The possibility of remote or very indirect wildlife-fire
effect relationships should be considered. For example, there is little
question the current concern and management efforts directed toward
aspen (Populus tremuloides) stand rejuvenation are valid. Robb (1987)
however, found that 85 to 90 percent of all gastropods infected with
bighorn sheep lungworm (Protostrongylus spp.) larvae in her study
area, resided in aspen stands and aspen edges. Could fire in an aspen
type in one area have potential implications for bighorn monitored in
another? As difficult as it may be, effort must be extended to look at the
whole system; to think and act within an ecological perspective (Savory
1988).

f. Stochastic events. Climatic and weather related events such as

drought, abnormally high precipitation, shifts in precipitation regimen,
and unusually hot or cold temperatures can have a marked effect on the
interpretation of fire effects. Large populations of insects such as the
Mormon cricket (Anabrus simplex) can consume remarkable amounts of
vegetation in a relatively short time thus clouding fire effects
evaluations.

C. Resource Management Considerations

Understanding existing management goals and objectives for an area is

essential. It is recognized that site specific objectives may not be in
place to adequately address every wildfire situation. Some objectives
may even be mutually exclusive. It is recommended that a familiarity
with fire terminology be developed to facilitate communication on
wildfires and in planning for prescribed fire. The Fire Effects Information
System (F.E.I.S.) contains a wealth of information and is an invaluable
aid in predicting and assessing fire effects. (See XII.D.4., this Guide.)

Every fire is a "wildlife" burn. Only through careful consideration,

accommodation and management of factors influencing the results of
fire can wildlife goals and objectives hope to be met.

1. Define Terrestrial Wildlife Habitat Goals and Objectives.

2. Standard Considerations for Fire Suppression.

a. Protection of habitat improvement projects. Protect habitat

improvement projects such as guzzlers, nest structures, browse
plantations, fences, and recent prescribed burns.

b. Water. Water quality and flow considerations are of vital importance

to many species of wildlife. Efforts to protect water include, but are not
limited to, the following:

(1) Prohibiting the washing or rinsing of any container or equipment

containing potentially harmful substances in or near any spring, stream,
pond or lake. Containers would include such items as helicopter
buckets, retardant tanks, engine tanks, backpack pumps, and other
such items. Potentially harmful substances include, but are not limited
to, wet water, foaming agents, and petrochemicals in any form.

(2) Avoid the dropping or spraying of retardant, wet water, or foaming

agents directly on, or immediate to, wetlands, springs, streams, ponds
or lakes.

(3) Avoid alteration or damming of stream courses.

c. Control of vehicle and heavy equipment use.

(1) Restrict travel to existing roads to the extent possible.

(2) Avoid any travel in or across streams or wet meadows, or through

unique or limited habitats.

(3) Physically close and rehabilitate all firelines that potentially offer
ORV access.

d. Control of aircraft use.

(1) Establish low-level flight routes that avoid important habitat areas
such as bighorn sheep summer range, raptor nest sites, and waterfowl
nesting areas.

(2) Avoid harassment of big game or other species of wildlife.

e. Wildlife barrier management.

(1) Bulldozer-line windrows through timber or heavy brush should be

broken up, lopped and scattered. At a minimum, they should be
breached at all drainage crossings and at intervals between drainages
to facilitate movement of big game.

(2) Where trees have been slashed around ponds or other bodies of
water used to facilitate bucket drops, the slash should be reduced to a
depth of no more than 18 inches (46 centimeters) and/or travel lanes
cut through for wildlife access to the water.

3. Species Habitat Requirements.

a. Structure. Species structural requirements for feeding, hiding cover,

reproductive cover, thermal cover and ease of movement should be a
primary consideration. It is important to keep in mind that structure per
se is not defined by height (i.e., a short-grass meadow is just as much a
structural component as an impressive stand of grand fir). A number of
species-specific habitat management guides are available which
address structural requirements. Life Form, Habitat Guild, and other
similar listings also can be of assistance.

(1) How do current structural conditions compare with the perceived

optimum for a featured species or management for species richness?
The areal extent, shape, height, age, density, and orientation of
structural components, and the necessary linkages between and among
those components should be addressed. If structural conditions of
vegetation are at or near the perceived optimum and in a healthy
condition, fire would not be of benefit. It is important to remember the
value of nonliving structural components such as snags and downed
logs. Fire may have a positive effect on these features, such as when
fire creates snags, or a negative effect when a severe fire consumes
most downed woody debris.

(2) How adequate are structural conditions adjacent to the proposed

burn or wildfire? A number of species (e.g., elk, mule deer, and others)
more readily use burned areas if their cover requirements are met in
close proximity to the burned area. In some vegetation types (e.g.,
sagebrush-grass), the "best" habitat frequently burns because it has the
highest fine fuel loading.

(3) What structural conditions and orientation are desired within the fire
area? For example, some species may require certain cover
characteristics along drainage courses, from drainage courses to ridge
lines and/or along ridgelines to make efficient use of burned areas.
(4) What postburn timelags for structural development are tolerable? If
no timelag is acceptable, that structural condition should be afforded
protection from fire. Threatened or Endangered species, species
classed as "sensitive," and species that are obligates of late seral
conditions are species likely intolerant of any habitat loss duration if that
loss is of sufficient size. The opportunity for accepting short-term
structural losses for long-term gain should be explored, however.

Factors influencing plant response time include:

(a) Plant survival mechanism;

(b) Plant health;

(c) Phenological stage;

(d) Preburn and postburn management;

(e) Stochastic events such as drought, torrential rains, insect or disease

outbreaks and other unpredictable and largely uncontrollable
occurrences.

b. Behavior. Behavioral attributes may influence species response.

Nominal home range size, territory, interspecific compatibility, sensitivity
to human disturbance, site fidelity, preference for open vistas (e.g.,
antelope, bighorn sheep), preference for denser cover (e.g., white-tail
deer, ruffed grouse), and other aspects of behavior can have a
profound effect.

c. Food habits. Food habits must be considered. An animal that

depends on a comparatively few select food items is more sensitive to
fire effects than a species with more cosmopolitan dietary requirements.
The following is oriented toward herbivorous species with the
understanding that predators and scavengers are indirectly affected
through their prey base.

(1) What shift in available food items may occur as a result of burning?
Fire of sufficient fireline intensity and/or burn severity could benefit
grazers at the expense of browsers. Browsers could benefit from a fire
in a habitat type such as oak brush but species dependent on mast
produced by that oakbrush may be negatively impacted for many years.

(2) Will food items be available when the species requires them? For
example, a fall burn could eliminate critically needed forage for
wintering herbivores. A fire on that same site early enough in the spring
to promote substantial regrowth, could be beneficial.

(3) Is fire-sensitive vegetation involved in the food base? Long-term

negative impacts can be incurred if an animal is dependent on a fire-
sensitive species that burns. The opportunity, however, for accepting
short-term forage loss for long-term enhancement of the food base
should be considered.

d. Water availability. Free water availability dictates the presence or

absence of many species following a fire.

(1) Is water present on or close to the burned area? Some species will
travel miles for water. Others, however, need it immediately available.

(2) Is the water present yearlong or on a seasonal basis?

(3) What cover characteristics are present immediate to the water?

Some species (e.g., sage grouse, white-tail deer) show a reluctance to
use a water source deficient of adjacent cover. Other species, such as
antelope, prefer good visibility.

(4) The potential adverse effects of fire on water quality -- both onsite
and off-site -- should be addressed. Water quality can have a major
influence on food chain relationships.

4. Miscellaneous Considerations.

a. Size of burned area. Is the burn of sufficient size to accommodate

the forage demand of large herbivores? Deer, elk, bighorn sheep,
livestock, and a number of other species are capable of making
excessive use of burned areas that are of insufficient size. Many of
these species exhibit a strong affinity for burned areas. Options to
consider include:

(1) Burn additional similar size areas;

(2) Increase the size of the prescribed fire unit;

(3) Protect burned areas with fencing or by herding (as with domestic
sheep);

(4) Reduce numbers of animals generating the demand;

(5) Unless otherwise accommodated by management, conduct

prescribed burns for wildlife on cattle allotments on areas not readily
accessible to livestock, such as steep slopes.

b. Noxious weeds and exotic plants. Potential for noxious weed

and/or exotic plant invasion should be addressed and management
adjusted as necessary.

c. Potential changes in human access. Is the burned area in or near

zones of human activity? Human activity associated with roads,
campgrounds, ORV use, and other sources can strongly influence the
presence, absence or habitat use efficiency of many species. Bulldozed
firelines may create undesirable access. Burned areas may attract
recreational use by snowmachiners, skiers and others. Closures to
protect wildlife may be necessary.

d. Snow. Will the habitat quality of some species be altered by changes

in snow deposition and crusting factors as a result of burning?

e. Snags.

(1) Adequate snag protection should be incorporated into prescribed fire

plans. The opportunity to protect snags under certain wildfire situations
may also be present. Damage to snags from fire can be limited by such
measures as hand pulling or machine piling fuel away from their base,
and applying fire retardant foams around the bottom and along the
bottom part of important snags. Leave some living trees with broken
tops that will eventually become snags, and provide raptor nesting
habitat in the interim.

(2) When conducting postfire salvage logging, leave some dead or

dying trees to become future wildlife trees.

f. Downed logs. When prescribed burning, ensure that a certain

number of logs of a minimum specified diameter are left onsite. If
broadcast burning, prescribe moisture contents high enough in the
larger diameter material that it does not burn. If pile burning, leave logs
out of the piles.

D. Methods to Monitor Fire Effects

This section views some considerations for monitoring fire effects rather
than specific techniques.

1. Animal Population Changes. It is suggested that monitoring of

animal population changes be avoided unless there are overriding
reasons to do so. As noted previously, intrinsic factors and synergistic
relationships between and among plants and animals can easily
confound cause/effect assessment of fire effects on populations. The
sophistication of study design and execution, time and cost of such
studies is normally beyond field office capabilities. Contract studies or
support of cooperating agencies, organizations or institutions should be
investigated when such information is required.

2. Objectives.

a. Well-defined and measurable. Well-defined, measurable objectives

that describe essential plant community characteristics greatly facilitate
monitoring. For example, an objective to increase herbaceous
production for elk forage would require much more monitoring effort if it
were written to increase specific plant species rather than the total
production of herbaceous plants elk utilize. Yoakum (1980) indicates a
variety of forbs is required on quality pronghorn range. Specific forbs
are apparently of little importance. An objective to increase the variety
of forbs is much simpler to monitor than one that requires the absolute
determination of plant species. The same consideration holds for
objectives relating to structure.

b. Expressed in absolute terms. Objectives expressed as percent

composition are essentially meaningless unless accompanied by an
element addressing ground cover, pounds production, or some other
absolute. For example, 50 percent grass and 50 percent sagebrush on
a rangeland could mean anything from one grass plant and one
sagebrush plant to 500 pounds (227 kilograms) production of grass and
500 pounds of sagebrush.
3. Monitoring Level. The level of monitoring required needs to be
defined. There is a temptation to set up monitoring procedures of detail
or sophistication that are unwarranted for management purposes.
Monitoring procedures should match the issue sensitivity. For example,
if only an index of bighorn sheep preference for a burned area is
needed, a simple grazed-ungrazed plant transect inside and outside of
the burn may suffice. An effort that addressed factors such as
comparative production by plant species, chemical analysis of forage,
and pounds of various plant species utilized by bighorns inside and
outside the burn would be "overkill." The latter approach might be
appropriate, however, if bighorn sheep use of a burn was a
controversial issue or some detailed information regarding their ecology
was needed.

4. Consistency of Technique. Maintaining consistency of technique is

essential. The plot that is estimated one year, measured the next, and
perhaps photographed the next, does not allow for any meaningful
comparisons among years.

5. Observations. The value of observational information should not be

overlooked. Frequently, this subjective information provides the critical
links with more formal data to clarify what actually transpired due to fire
or whether the apparent effects were due to another reason. Anyone
who has had an occasion to be on the site, either during or following the
fire, is a potential source of information. Observations should be
properly documented and filed. Subjective information alone seldom
provides the confidence needed to make politically sensitive
management decisions.

6. Overall Effect. Addressing overall effect of fire on wildlife for a given

area that has burned is most easily approached by going from the
general to the specific.

a. What was the original successional stage and structural condition?

b. What Life Forms or Guilds were associated with those preburn

conditions?

c. What will the new successional stage and structural makeup be?

d. What Life Forms or Guilds will be favored by the new conditions?

e. How were species of management or public interest affected?

f. How may any obligate or otherwise sensitive species have been

affected?

7. Sources. An increasing number of sources for assistance in

developing monitoring programs are available. Only a few of the more
readily available are listed here. Inventory and Monitoring of Wildlife
Habitat (Cooperrider et al. 1986) contains a wealth of information.
Species specific habitat guidelines have been developed for pronghorn
(Yoakum 1980), sage grouse (Autenrieth et al. 1982) and a number of
other animals. Thomas et al. (1979) outlines simple procedures for
measuring and evaluating edge diversity. Estimating Wildlife Habitat
Variables (Hays et al. 1981) is an excellent field-oriented guide that not
only addresses procedures but offers estimates of time and cost
involved with various techniques. Chapters VI and X of this Guide
should be referred to for further considerations and direction on
monitoring and evaluation.

E. Summary

Fire is a shock - frequently, nearly instantaneous - to the ecological

setting involved (Huff et al. 1984, Lyon et al. 1978). Some wildlife
species are able to adapt to the rapid change in environment and some
cannot (Lyon and Marzluff 1984, Parker et al. 1983, Rotenberry and
Wiens 1978, Wecker 1964). The habitat for some species is greatly
improved, while for others it may be degraded if not eliminated, and
there will be endless variation in between (Beck and Vogl 1972, Bendell
1974, Evans 1988, McGee 1982, Wolf and Chapman 1987). No fire -
either wild or prescribed - is uniformly "good" or "bad." Effects are
differentially imposed.

A righteous attempt at providing for desired fire effects through

prescribed burning or evaluating wildfire effects on wildlife and its
habitat requires an integrated effort of disciplines. An appreciation of the
historical perspective can be invaluable. Contributions by plant or fire
ecologists are essential - individuals may have the talent but not the
title. Input from those with a thorough knowledge of fire is certainly
important. Postburn management is absolutely critical. Obtaining good
management necessarily requires close coordination with and
commitment from specialists in range, forestry, recreation, and others.
Without adequate monitoring and evaluation, little knowledge can be
gained and even less, shared.

This page was last modified 05/31/01

|Disclaimer| | Privacy| | Copyright| |USFWS Main Page| |Webmaster|

 National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 05/31/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home CHAPTER VIII - CULTURAL RESOURCES

Preface
Objectives By Dr. Richard C. Hanes
Fire Behavior
Fuels
A. Introduction
Air Quality
Soils & Water
Plants Cultural resources include a range of different resource types. These
Wildlife resources include locations containing archaeological and architectural
Cultural Res. remains resulting from human activity in the prehistoric and historic
Grazing Mgmt. periods; and locations of continued traditional use activities, primarily
Evaluation associated with areas of religious or traditional subsistence concern to
Data Analysis Native Americans.
Computer
Soft. Prehistoric archaeological sites include artifact scatters at locations
Glossary where tools were made, a series of depressions in the soil surface
Bibliography representing a pithouse village, pueblo ruins, and rock art panels where
Contributions figures were carved or painted centuries ago. Historic period sites
pertain to more recent activities. Examples include old cabins, early
homesteads, trails, battlegrounds, early mining remains, and logging
camps. The second category of cultural resources noted above includes
traditional areas where soil or plants are collected, or ceremonies
conducted for secular or religious purposes. In some cases, these
areas coincide with locations where archaeological remains are found,
but they are just as likely to be a spring, mountain, or other geographic
feature not containing tangible reminders of past activities.

B. Principles and Processes of Fire Effects

Particularly important information concerning fire effects on cultural

resource values has been developed by Peter Pilles (1982) of the U.S.
Forest Service and by National Park Service programs (Kelly and
Mayberry 1980). Much of the following information draws from those
sources as well as others noted.

It is difficult to accurately assess the effects of prescribed fires or

wildfires on cultural resources. One important factor is the widely
varying responses of vegetation and soils to fire within the same burn
area, or under the same prescription in different burn areas. At most
cultural resource sites on public lands, artifacts are distributed on the
soil surface, or buried within the soil, even when historic or prehistoric
structures are also present. The amount of surface and subsurface
heating depends upon the peak temperatures reached and the duration
of all phases of combustion. The amount of subsurface heating is a
function of a number of variables, including soil moisture content and
coarseness, amount and distribution of woody fuels, occurrence of duff
layer or other accumulations of organic litter, weather conditions, and
fuel and duff moisture content. (See Chapter III.B.6., this Guide for a
more detailed discussion of burn severity.) Fire behavior studies have
shown that no clear relationship is currently known between surface
temperatures attained in a fire and temperatures conducted below the
soil surface. It appears that only a small percentage of surface heat
penetrates the soil deposits, because soil temperature during a fire can
decrease dramatically in just a few inches of depth. Smoldering and
glowing combustion of both surface and subsurface fuels can be
important contributors to heating of buried artifacts.

Few formal studies of the effects of fire on archaeological sites have

been performed. Studies in the 1970's involved the Radio Fire in the
Coconino National Forest near Flagstaff, Arizona (Pilles 1982),the La
Mesa Fire near Bandelier National Monument, New Mexico (Armistead
1981; Traylor 1981), the Moccasin Mesa Fire at Mesa Verde National
Park, Colorado (Switzer 1974), and the Dutton Point Fire at Grand
Canyon National Park (Jones and Euler 1986). These were all wildfires,
not prescribed fires. A handful of studies and recorded observations
have been conducted in the 1980's, including experimentation with
prescribed fires in the Cleveland National Forest in California (Pidanick
1982; Welch and Gonzalez 1982). The literature addressing the effects
of fire and heat on archaeological materials is still very meager in
quantity and largely unpublished. However, enough information exists
to indicate that the effects are variable, depending upon the material
that is heated and the level of heating attained.

Arguments are made that many of the prehistoric cultural resource sites
have been exposed to fires, perhaps repeatedly, in the past. So why the
concern over fires today except in the more obvious cases where
perishable structures are involved? In addition to the rebuttal that we
are likely dealing with cumulative information loss from repeated
impacts, other important factors must be considered, including relative
burn severity between historic and prehistoric fires, recent surface
exposure of some ancient sites, and cumulative changes in erosion
patterns.

A growing body of information, based on early historic accounts,

ethnographic studies, and field fire history studies suggest fire was
much more prevalent prior to implementation of fire suppression policies
earlier this century. The higher frequency of fires earlier in time is at
least partly attributed to aboriginal burning practices (Lewis 1973, 1985;
Barrett and Arno 1982; Arno 1985; Gruell 1985). In areas where fires
occurred frequently, the exclusion of fire has often led to levels of fuels
higher than would have "naturally" occurred. Fires which occur now
potentially have greater impacts on cultural resources because of the
increased amounts of heat that can be released during burning.

Archaeological sites often contain a variety of cultural-related materials,

including stone, bone, shell, ceramics, metal, glass, wood, leather, and
other substances. The resultant combination depends on the
technologies of the inhabitants, the specific activities being performed at
that location, and the preservational characteristics of the setting.

1. Potential for Physical Damage to Materials. The effects of fire on

these various materials, whether buried or on the surface, can vary
significantly because of different inherent properties and locations
where materials occur.

a. Stone. A commonly observed result of intense, high temperature

fires is the occurrence of heat damaged stone artifacts. An example of
such damage was presented by a very hot prescribed fire in California's
chamise chaparral that resulted in temperatures over 700 F (371 C) at
one location where archaeological materials had been placed to assess
the effects (Pidanick 1982). Some chert chipped stone implements had
shattered; other artifacts including obsidian items and grinding stones
were heavily smudged. Artifacts placed at other locations where
temperatures never reached 400 F (204 C) were unbroken and only
lightly smudged. Laboratory experiments have demonstrated that the
crystalline structure of many forms of silica-rich stone changes when
heated above 700 F (371 C) (Purdy and Brooks 1971; Mandeville
1973). Beyond that temperature, stone will spall, crack, shatter, oxidize,
or simply break from direct exposure to heat. Extensive heat spalling of
lithic artifacts was observed in hot "spots" of the 1974 Day Burn in the
Apache-Sitgreaves National Forest.

Many masonry pueblos in the Southwest are constructed with dressed

sandstone blocks. Exposure to intense heat can lead to color changes
through oxidation, severe cracking, spalling, and even crumbling as a
result of burning away vegetation that has aided in stabilizing the
remains through time (USDD-COE 1989; Traylor 1981).

Rock art sites, including those where designs are painted on stone
(pictographs) and others that are pecked into stone (petroglyphs), are
especially susceptible to damage by fire. When exposed to intense
heat, painted designs can be soot blackened, scorched or completely
burned away while petroglyphs on friable stone, such as sandstone or
limestone, can exfoliate (Pilles 1982; Noxon and Marcus 1983).

b. Ceramics. Pottery is another class of artifact that may be seriously

affected by fire (Burgh 1960). Pottery is made by subjecting fabricated
clay vessels to intense heat where sintering and other physical and
chemical changes give it clastic properties. However, fire of long
duration or high intensity may refire prehistoric or historic ceramics,
which can recombine the constituents, oxidize certain elements, burn
out carbon paints and cause increased brittleness. Smudging or the
deposition of surface carbon residue also may make it difficult to identify
and date ceramics.

In the Apache-Sitgreaves fire, ceramics became highly vitrified and

appeared as hard black sponges. Similarly, after the 1972 Mesa Verde
National Park wildfire, spalling and discoloration of ceramics was noted.
The primary impact observed at the Dutton Point Fire was smudging of
shards that may disappear naturally after years of weathering. In
summary, fire can burn pot shards, affect their chemical composition,
change their colors, and alter their decorative paints and glazes, making
identification of styles and manufacturing techniques difficult in some
cases. (See also Pilles 1982, p. 6.) The more substantial changes begin
to occur with temperatures of 925 F (496 C), a threshold higher than for
stone.

c. Organics. Objects made or manipulated by a site's occupants are

not the only materials used to reconstruct a scenario of activities that
were performed at a given location. Many sites contain shell and bone,
giving evidence about the nature of prehistoric diets. When exposed to
a high level of heat, shell will become calcined and very friable. The
effects of fire on bone have yet to be thoroughly investigated, but at
Custer's Battlefield, old bovine bone fared very poorly compared to
stone and metal objects (Scott 1987). Pollen grains, used for
paleoenvironmental as well as dietary studies, are destroyed at
temperatures above 600 F (316 C) (Traylor 1981). Artifacts made of
organic materials such as woven baskets, wooden digging sticks,
rawhide cordage, and fur clothing are usually very fragile in the
archaeological record and are highly susceptible to charring and
consumption at very low temperatures (Seabloom, Sayler, and Ahler
1991).

d. Metal and glass. Not much is known about the effect of wildland fire
on inorganic materials largely associated with historic period sites in the
West, such as metal implements and glass bottles or beads. A recent
North Dakota prairie fire study found that small lead and glass items
became fused or melted when subjected to ground surface heating
(Seabloom, Sayler, and Ahler 1991).

2. Effects on Dating Techniques. The archaeologist today has several

techniques available for deriving absolute dates of site occupation
(Michels 1973). In addition to the impacts to artifacts and material types
noted above, materials used for several dating techniques also may be
affected by fires (Traylor 1981, Pilles 1982).

a. Tree rings. Tree ring records preserved in wooden beams or other

construction materials used in dendrochronologic studies are highly
susceptible to fire.

b. Radiocarbon. Charcoal samples used for radiocarbon dating can

become contaminated from ash and charcoal produced by a fire, and
could yield a date more recent than the true date of the sample.

c. Thermoluminescence. Pottery fragments, when subjected to

thermoluminescent dating techniques, could provide significantly
younger dates than expected after being exposed to high heating
episodes.

d. Obsidian hydration. Obsidian hydration is a dating technique that

measures the amount of moisture present in the external surface of an
obsidian artifact. Moisture is absorbed from the atmosphere by freshly
flaked obsidian surfaces at a constant rate. Heat from a fire (apparently
at only high levels of heating) can alter the moisture content, thus
yielding an inaccurate date or erase the record altogether.

e. Archaeo-Magnetic. Archaeo-magnetic dating measures the

orientation of electrons in stones from prehistoric hearths and compares
this data to changes in the earth's magnetic field over the past several
thousand years. If these features are subjected to temperatures above
975 F (524 C), they can give erroneous information by releasing
electrons to realign with the current magnetic fields of the earth.

f. Cation-Ratio. Cation-ratio dating is a new technique for dating rock

art through chemical analysis of surface varnish. It is possible that
smoke from a fire could alter the ion structure of these features, thereby
preventing accurate dating.

3. Impacts of Burn Area Preparation and Mechanical Suppression.

The most dramatic and predictable effects of fire activities on cultural
resources result from the use of equipment in burn area preparation,
fire suppression, or burn area rehabilitation work. Impacts from these
activities are also the most preventable. Pilles (1982, p. 6) has noted
that:

Studies of the La Mesa Fire at Bandelier National Monument and the

1977 fire on the Coconino National Forest found that heavy equipment
used during suppression activities and mop-up operations had a greater
effect to archaeological sites than did the actual fire itself. Artifacts are
broken and displaced and small sites can be completely destroyed by
one pass of a bulldozer blade. Depending on the depth of a blade cut
and the proximity of site features to the surface of the ground, buried
features, such as caches, burials, and firepits, can also be destroyed by
bulldozer work. During both the Radio and La Mesa forest fires, about
15 percent of the archaeological sites in the area were damaged by
heavy equipment. In the La Mesa Fire, however, most sites were
damaged during mop-up and restoration activities after being initially
avoided by fire suppression activities. Both instances point out the
importance of timely planning and continued coordination with
archaeologists for projects involving the use of heavy equipment.

Obviously, burn area preparation could cause damage to cultural

resources if mechanical equipment is used. Additionally, construction of
heliports, vehicular traffic, and hand construction of firelines can impact
cultural values. Postfire erosion control measures such as mechanical
seedings, contour trenching and furrowing, and construction of
sediment traps are restoration activities that pose significant threats to
archaeological sites.

4. Erosion and Looting. Loss of ground cover normally leads to greatly

enhanced visibility. In many regions of the West, wildfires have long
been noted for their propensity to expose sites previously difficult to
find; consequently large numbers of people can be found cleaning the
surface of diagnostic tools and excavating sites where archaeologically
rich deposits are discovered following fires. Similar behavior has been
noted of fire crews who had not previously been advised of the
significance of such activities (Traylor 1981). This is of increasing
concern, as the illegal collection and excavation of archaeological
materials has escalated during the past 30 years. The water holding
capabilities of litter, duff and surface soils are also reduced by fire,
which sometimes generates erosion hazards.

C. Resource Management Considerations

The preceding sections briefly describe a diverse array of impacts that

fire and associated fire management activities pose for cultural resource
values. However, many of the heating effects only occur at significantly
high temperatures and many associated on-the-ground activities can be
planned ahead of time. Consequently, the fire process can be managed
to minimize harmful effects and serve as a useful tool in managing
cultural resources.

1. Fire Planning. The most effective means of addressing fire effects is

through development of a management plan that takes the above
concerns into account. (See Anderson 1985.) Various facets of the land
management planning process may be used. The cultural information
may be provided in a prescribed fire plan, a wilderness management
plan, a general resource management plan, or, for areas that are of
particularly high cultural resource values, a cultural resource
management plan. All agency cultural resource management plans
should include a section on the effects of fire suppression. Regardless
of the type of plan employed, it should provide information about the
number, type and distribution of cultural resources, known or predicted
to occur, in a proposed project area (Pilles 1982, p. 8.) and how
susceptible these resources are to impacts from fire. Are there
abundant cultural resources in the area? Are there historic settler's
cabins or sawmills present that could be destroyed by fire? Has the
area ever been examined by a professionally qualified archaeologist?
Are there any Native American concerns that might be affected by a
prescribed burn? Are there any areas considered highly religious?
Would burning at a particular time of the year disrupt traditional religious
pilgrimages, plant collecting, or hunting practices in the area?

Prior to prescribed fires or the next wildfire season, baseline cultural

resource information should be gained minimally through an updated
synthesis of existing information, contacts with the appropriate Indian
tribes, coordination with the State Historic Preservation Office, and
inclusion of information from other knowledgeable sources. Further
information may be gathered through field reconnaissances, sample
field surveys, or detailed individual site assessments. From the
information gathered, areas of unusual sensitivity or highly significant
sites may be identified on maps and their vulnerability to fire effects
assessed. Management direction regarding fire activities may then be
established and the resulting information provided to those in charge of
planning and directing field activities. The management direction for
wildfire suppression and prescribed fire projects should be coordinated
with the State Historic Preservation Office, again, to streamline any
required Section 106 consultation needs that may arise when fire
activities are imminent. The plans should include procedures for training
fire crews about the illegality of artifact collecting and the associated
stiff penalties, and for educating crews to identify sites so that damage
to these resources can be avoided during fire suppression activity.

2. Maintaining Historic Plant Communities. Constructive use of fire

also can be identified in terms of reestablishing the historic
environmental context of important cultural resources and maintaining
certain Native American traditional practices (Larson and Larson 1988).
Examples of the former case includes restoration of grassland from
recent pinyon-juniper invasion at a historic fort site and removal of
brush thickets from historic trails, thus opening them to recreational use
(Pilles 1982).

In the case of Native American needs, burning can be used to promote

the growth of certain plants used for food, medicine, or craft
manufacture. An outstanding case is presented in California where
prescribed burning by the U.S. Forest Service allows growth of new
plant shoots. This new growth has the proper strength and resiliency for
the Yurok tribe to use in weaving baskets and hats for tribal ceremonies
and as traditional apparel (Pilles 1982). Such activity is helping
revitalize certain areas of traditional Indian culture.

3. Use of Prescribed Fire to Minimize Potential Damage from

Wildfire. As noted above, the heating effects of low temperature
prescribed fires appear to be substantially less than the effects of much
hotter wildfires. Archaeologists can learn much from fire history studies
and effects on soil properties. Knowledge of the frequency of prehistoric
fires would likely indicate the possible cumulative effects on cultural
resources, with high fire frequencies likely associated with "cool" fires
and minimal impacts on resources.

a. Subsurface resources. During prescribed fires, effects of heating

are usually not severe. Most artifacts are insulated from the heat of a
fire by an earth cover and ideal temperatures for most prescribed fires
are less than those that critically affect artifacts. If a condition is present
of low fuel loads or a fire occurs with higher duff or soil moisture
content, there is less potential for heating. However, if fire burns with
high heat per unit area, then damage to surface artifacts is likely. Also,
if a fire is of long duration, damage to buried artifacts is more likely.

In some areas fire suppression policies of the past century have led to
"artificially" high fuel loads, thus increasing the potential for damaging
cultural resources through severe heating when fires occur.
Additionally, encroachment of shrubs and/or trees allows deep litter
layers to accumulate and increases the potential for longer duration
fires. In some situations, an agency may need to use prescribed
burning or some other fuel load reduction strategy to attain its
mandated mission for protecting cultural resources.

b. Aboveground structures. Historic period and prehistoric

architectural sites pose special concerns. The historic period sites were
created either during, or just before, the period of enforcing strict fire
suppression policies without the augmentation of prescribed burns.
Consequently, the sites have not been subjected to any form of fire, and
preservation in many cases may be very good. Old buildings and ruins
constructed of wood are obviously susceptible to destruction by fire.
Burn prescriptions will need to be designed to avoid impacts on these
types of cultural resources yet reduce the fuel load buildup around
them. Possible prescriptions might be to put a fire line around sites with
wooden structures so they are not burned; modify project boundaries to
avoid rock outcrops where rock art is located; remove combustible
materials from the surface of a site so the fire will either burn around it
or burn with "cool" temperatures above it; shift the location of a control
line, staging area, or utilized water source to avoid a site; change the
dates of the burn so it does not impact Native American use of the area;
or simply have the archaeologist monitor the burn while it is in progress.

It can be concluded that by burning under favorable conditions where

burn severity may be controlled and monitored, management of
vegetative communities through fire can be pursued while enhancing
the agency's ability to protect cultural resources.

D. Methods to Monitor Fire Effects

The above discussion briefly addresses a number of complex issues for

which few objective data are available. Most information thus far has
been collected in association with major wildfires occurring in areas of
heavy fuel buildup. Very little quantification of the effects of fire on
archaeological sites has been documented, and little has been reported
of prefire site conditions. There is a great need for experimentation,
particularly utilizing prescribed fire conditions. In addition to recording
artifactual and site preburn and postburn information, detailed
documentation of fuel load, fuel and soil moisture, weather, fire rate of
spread, temperature at and below the soil surface during combustion,
duration of heating, and other fire factors needs to be accomplished. In
anticipation of a fire, half of a site could be excavated prior to the fire,
and the remainder excavated afterwards. By employing such
procedures, the effects of the burn on various aspects of the
archaeological record could be evaluated, and correlated with the
behavior and severity of the fire.

More study is needed to determine the full range of effects to cultural

resources by wildfire and, especially, controlled burning projects. In
order to accomplish this, a variety of experiments in different
environmental settings with different kinds of cultural resources needs
to be done. Only by observing fire effects in a variety of conditions can
we know for certain that there is no significant or irreplaceable loss to
cultural resources as a result of prescribed fire programs or the degree
of damage posed by wildfire. To accomplish this goal, land
management plans can commit specific cultural resource sites to
"management use" for fire effects experimentation.
In addition, furnace tests on archaeological materials are needed to
establish controls for comparison of field test results. Furnace tests can
assess the effects of different temperature levels and heating periods
on specimens of ceramics, metal, glass, bone, shell, and stone. Such
information can aid in documenting actual fire effects on artifact
friability, weight, and visible characteristics (e.g., color, form, decorative
patterns, and trademarks).

Experiments, such as those described above, would provide an ideal

opportunity for interagency cooperation and could assess the relative
success of the fire management procedures and philosophies of
different agencies. The value of such cooperation has already been
demonstrated by studies of the Radio Fire and the La Mesa Fire,
conducted by the Coconino National Forest and the Southwestern
Regional Office of the National Park Service. These two studies are the
main data sources available for assessing the impacts of fire and fire
management activities on cultural resources. An interagency
clearinghouse for assembling data and reports pertaining to fire effects
on cultural resources could greatly assist the otherwise disjointed
approach taken by the various organizations.

As acknowledged above, visibility of cultural resources is greatly

enhanced by burning away the ground cover. A 1977 tundra fire in
Alaska removed shrub growth, revealing prehistoric stone-lined pits
where none were previously visible (Racine and Racine 1979). Wildfires
in the early 1980's in pinyon-juniper woodland areas near Las Vegas
and Carson City similarly resulted in the discovery of prehistoric rock
ring features where none were previously known. The Las Vegas rings
were likely associated with past pinyon nut caches; the Carson City
features are possibly remains of habitation structures. In Mesa Verde
National Park, prehistoric farming terraces were revealed when fires
burned away the dense underbrush. Before this, archaeologists were
not aware of the abundant existence of such features in the Mesa
Verde area. Consequently, the coordination of prescribed fires with
postfire archaeological surveys would be beneficial to the agencies in
achieving goals for two programs. The benefits to the cultural resources
program are obvious in the greater efficiency achieved in conducting
inventory efforts. Also, patterns identified in heat damage to artifacts by
the above furnace and field studies have promise to aid in the
interpretation of heat damage sustained by artifacts in early historic and
prehistoric times (Seabloom, Sayler, and Ahler 1991).
E. Summary

Damage to cultural resources posed by wildfires and prescribed fires

can be severe, ranging from chemical alteration of cultural materials to
exfoliation of building materials and rock art panels. However, almost all
impacts can be avoided through advanced planning. Protective
measures can include removal of high fuel loads by hand or prescribed
fire, careful use of fire breaks for avoiding fire effects on wooden
structures and other highly susceptible resource values, and use of
archaeological monitors on wildfires in sensitive areas to avoid fire
suppression damage.

The experiments and observations thus far conducted indicate that

cultural materials below the surface, unless directly exposed to a
burning duff layer or burning underground roots, normally do not sustain
significant damage, if any at all (Traylor 1981). Though the Cleveland
National Forest found that many surface artifacts were damaged by a
prescribed fire in chamise chaparral, no subsurface artifacts were
affected (Pidanick 1982). Measurements taken at the prescribed fire
documented temperatures in excess of 800 F (427 C) at the ground
surface, but only 100 F (38 C) at 5 centimeters (2 inches) below the
surface. Obviously, the magnitude of fire effects on the soil and its
contents is proportional to heat penetration. In conifer forests,
temperatures of 200 F (93 C) have been recorded one/half inch (1.3
centimeters) deep in the soil, with duff layers considerably above that
figure. Obviously, such heating depends on the thickness of the duff
layer, duff moisture content, amount and moisture content of large
diameter dead woody fuels, and soil type and its moisture content.
Given current knowledge of fire effects on cultural resources, it is
apparent that fires involving larger fuel loads, longer duration burns, and
large total heat release pose significantly greater hazards to cultural
resources, than fires with short duration "cool" combustion
temperatures.

This page was last modified 05/31/01

|Disclaimer| | Privacy| | Copyright| |USFWS Main Page| |Webmaster|

 National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 05/31/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home CHAPTER IX - PREFIRE AND POSTFIRE GRAZING MANAGEMENT

Preface
Objectives By Ken Stinson
Fire Behavior
Fuels
A. Introduction
Air Quality
Soils & Water
Plants The impacts of grazing management before and after a fire have a dramatic effect
Wildlife on the response of vegetation to the fire, and what one can expect in the long term.
Cultural Res. The history of management on burned areas has included such things as:
Grazing Mgmt.
Evaluation 1. Seeding introduced species for increased livestock or wildlife forage that
Data Analysis resulted in additional management conflicts.
Computer Soft.
Glossary 2. Short-term rest from livestock grazing to allow seedling establishment.
Bibliography
Contributions 3. Some followup long-term grazing management but with heavy to severe
utilization rates.

4. Conducting prescribed fires to temporarily increase production on an allotment to

avoid adjustment in grazing use.

5. Temporarily increasing utilization by improving palatability of rank plant species

such as Tobosa grass (Hilaria mutica), sacaton (Sporobolus spp.), and sprouting
browse species.

6. Grazing use in an area during a prefire rest period required to accumulate

adequate fuel to carry a fire, resulting in a subsequent prescribed fire that could not
meet objectives.

The need for increased intensity of grazing management on burned areas can be
understood by realizing the potential change in the plant community and
associated animal response that can result from a burn. If one is not willing to
commit to long-term grazing management, prescribed fire should not be considered
or approved.

B. Principles of Prefire and Postfire Grazing Management

1. General Need for Improved Management. "Prescribed fire should not be a
substitute for good range management. A problem rooted in inappropriate range
management practices may not be corrected by vegetation treatment. In these
instances management should be altered prior to application of prescribed fire. If
livestock have premature access to the burn, the full benefits of the prescribed fire
may not be realized and negative impacts may occur unless management of the
livestock is included in the plan" (Bunting et al. 1987). "Followup management is
the most important aspect of a controlled burn and must be provided for in the
overall management plan" (Smith 1981). "Grazing management following burning
may significantly affect the degree of change in forage species productivity and
possibly the composition of the postburn vegetation" (Smith et al. 1985). The need
for management of livestock use on a burned area is most critical the first growing
season after fire, particularly in plant communities of arid and semiarid regions
(Trlica 1977). Livestock use must be managed on the sites of both prescribed fires
and wildfires.

Fire results in changes of animal behavior including grazing pattern, preferences,

utilization rates, forage consumption, and frequency of grazing use. Wild and
domestic animals are attracted to recently burned areas resulting in greater
utilization of the burned area than surrounding vegetation (Pase and Granfelt 1971;
Bunting et al. 1987). Cattle, horses, and sheep usually have the greatest impact.
Grazing animals frequently concentrate on a burn because the herbage or browse
is more accessible, palatable, and nutritious (Wright and Bailey 1982). Plant
growing points may also be exposed, increasing the likelihood of damage from a
foraging animal. Carbohydrate reserves of sprouting plants are usually depleted
because of energy required to regenerate after a fire. Repeated use of these plants
can cause considerably reduced vigor, and sometimes death of key forage or
browse species. (See VI.B.4.b. (1) and (2), this Guide, for a more detailed
discussion of carbohydrate reserves.)

Grazing in forested areas can help forest regeneration if competing plant species
are grazed, or hinder regeneration if tree seedlings or sprouts are eaten or
trampled. Extensive damage to young conifers from trampling has occurred in
clearcut areas that were seeded to grasses (McLean and Clark in Urness 1985),
and has been observed in burned clearcuts where postfire growth of grasses and
sedges attracted livestock (Zimmerman 1990). The presence of larger diameter
logging slash can discourage livestock and big game use.

2. Rest and Deferment.

a. Prefire. Prefire rest from grazing is required on many range sites to allow the
accumulation of enough fine fuel to carry the fire. This is important in shrub/grass
and pinyon-juniper types as well as in forested areas, particularly aspen
ecosystems where grass and shrub litter may be the main carrier fuels (Jones and
DeByle 1985). Allowing grazing, sometimes even for a short period of time during
the year before the fire, can remove enough fuel to limit fire spread. A patchy fire
may occur, or the fire may not be able to carry at all, and in both cases fire
treatment objectives are not met. Prefire rest may also be required to restore levels
of plant carbohydrate reserves on heavily grazed sites, shrub dominated sites, or
where shrubs are very old and in poor condition. More than one year of prefire rest
from grazing may be required to obtain adequate fuel to carry the fire, or to achieve
the desired postfire response, especially in areas with severely depleted
understories.

b. Postfire. The amount of nonuse necessary after a fire varies considerably with
the vegetal composition, site conditions, and objectives of the burn (Bunting et al.
1987). The initial concern following burning is the restoration of plant vigor and
seed production. Generally, at least two growing seasons rest are recommended
(Pase et al. 1977; Wright et al. 1979; Blaisdell et al. 1982), both to allow
reestablishment of preferred species and to deter reinvasion of shrubs. Bunting
(1984) usually recommends rest for one year and deferment of grazing until after
seeds have ripened the second year, if the range is otherwise in fairly good
condition. Some species of sprouting shrubs take much longer than two years to
recover, such as bitterbrush, and rest for a longer time period is necessary if
reestablishment of browse species is an objective.

"Anticipated results from the best prescriptions for a burn may be seriously
modified if destructive grazing practices are allowed afterwards. Only a small
amount of forage is produced the first year, and grazing may cause serious
damage to soil and desirable perennials. Despite the apparent abundance of green
herbage, most plants are low in vigor and will be further weakened or destroyed by
grazing. Furthermore, grazing will disturb the inadequately protected soil and allow
increased water and wind erosion. Protection through the second growing season
will allow restoration of vigor and the typical heavy seed production of perennial
grasses and forbs. However, after seed dissemination, light grazing may serve a
useful purpose in helping to plant the seed" (Blaisdell et al. 1982). Early
establishment of a good grass cover, and subsequent conservative management,
virtually assures soil stability and low sediment yields on moderate slopes (Pase et
al. 1977).

Grazing in the early growing seasons immediately following burning may

accelerate sagebrush reestablishment. This is particularly true when areas with
dense sagebrush and low production of grasses are burned (Laycock 1979; Smith
et al. 1985). This may be desirable if sagebrush is an important habitat component
for wildlife species. Grazing systems that provide for periodic rest during the
growing season will extend the useful lifetime of the project (Britton and Ralphs
1979; Smith et al. 1985).

Evans (1988) gives a general rule that newly seeded areas should not be grazed
for at least two years following seeding. Low potential sites and those seeded with
slow developing and slow growing species may require as many as four seasons of
nonuse to develop into productive stands. Below average amounts of postfire
rainfall also can retard the recovery of a site.

3. Proper Followup Grazing Management. "Improvement of an overgrazed range-

-that is, improvement in range condition--starts with a decision to stock the pasture
at a rate to permit improvement" (Dyksterhuis 1958). Burning an area that is in
poor condition because of overgrazing can temporarily increase production of
desired species. However, the improvement will be short-lived if grazing practices
remain unchanged. "Various combinations of rotation land deferment. . . have all
proven to be successful where such factors as range condition, kind of livestock,
stocking rate, season, and intensity were given proper consideration. Rate of
stocking--balancing numbers and time of grazing animals with forage resources--is
the most important part of good grazing management ... Seemingly there has been
over-optimism in judging grazing capacity and allowable use, which has been an
important factor in range deterioration . . . It has become increasingly apparent that
former utilization standards are often several times more than can be tolerated
continuously, and that reduction in livestock numbers is often necessary to correct
unsatisfactory conditions" (Blaisdell et al. 1982).

Holechek (1988) researched and published utilization guides by precipitation zone

for different range types in the USA. The recommended average degree of use of
the key species varies from 20 to 50 percent with the upper levels only on good
condition ranges or for dormant season grazing. Heavy grazing invariably leads to
a gradual loss in forage productivity and vigor, high death losses, and higher costs
for supplemental feed in drought years. Pechanec, Stewart, and Blaisdell (1954)
found in the sagebrush-grass type that proper followup management, i.e.,
protection from livestock use the first year, and light grazing the second year with
proper stocking thereafter, resulted in increased grazing capacity 9 years later.
Capacity on this area was increased by 83 to 106 percent, but there was only a 4
percent increase without this management. The area with proper management had
five sagebrush plants per 100 square feet (9.3 square meters), compared to 55
plants per 100 square feet on the area without the above management.

On desert grasslands postfire rest must occur, and careful, conservative

management followed until the weakened grass cover has completely recovered
(Pase et al. 1977). Postfire recovery of browse species in these arid areas may
take much longer than on more mesic sites (ibid.) A common goal for all grazing
systems should be reduction of damage from grazing while promoting beneficial
effects, and many systems appear equally effective (Blaisdell et al. 1982). Many
combinations have proven to be successful where such factors as range condition,
kind of livestock, stocking rate, season, and intensity were given proper
consideration.

4. Economic Considerations. One of the primary reasons for the interest in using
prescribed fire and limited control of wildfire is the perception that fire is a cheap
brush control treatment. Smith (1981) states that prescribed burning provides an
inexpensive brush control method, but labor will greatly increase the cost of
prescribed burning, so the planning process should emphasize practices (such as
fuel breaks) that will reduce labor needs to a minimum. He also lists one
disadvantage as the risk of fire escaping and consuming valuable forage, ensuing
property damage and danger to lives, resulting in expensive suppression costs and
civil suits.
Bunting et al. (1987) suggest that selection of area to be burned will dictate many
of the economic variables such as fire prescription and characteristics and whether
it achieves its objectives . . . the higher potential sites produce the highest benefit.
He also states that burning during the spring with snow lines and increased fuel
moisture on varying aspects adjacent to the proposed treatment area may aid in
fire control and reduce overall cost. The limited burn size, however, may increase
the amount of time and personnel required for ignition resulting in higher average
costs per acre or not achieving the planned objectives. Bunting says that
economics is also a factor in determining the size of fires. The costliest portion of
conducting prescribed fires is establishing and burning out the fire lines. The
smaller the size, the greater the perimeter per unit area. Without natural fuel
breaks, an extensive system of fire lines may have to be established to restrict the
fires to the desired size. This often makes the prescribed burns economically
unfeasible.

From an economic standpoint, spring burning is cheaper as it can be accomplished

with fewer individuals and without firebreaks in some situations (Blaisdell et al.
1982). West and Hassan (1985) state that the highest potential for prescribed
burns is on sites in good condition. Haslem (1983) provides several guidelines to
maximize returns from burning including realistic prescriptions, treating
manageable units, using livestock use for controlling escapes, use of natural
control barriers, and use of test burns. Smith (1981) states that followup
management is essential in extending the fire's useful lifetime.

Young and Evans (1978) state a general rule that one must be able to step from
one bunch grass plant to another to have a reasonable chance of enhancing the
site by recovery of existing plants. Bunting (1984) also notes the bluebunch
wheatgrass response is from existing plants for the first 3 or 4 years after a fire.
Dramatic increases in numbers of plants of exotic annual species can occur after
fire, particularly if the existing bunchgrass community was in poor condition or
many of the plants were killed by the fire. This potential for site invasion must be
considered along with the above guidelines when deciding if a site can recover
without artificial seeding.

5. Examples of Different Intensities of Grazing Management. Many different

grazing management strategies have been implemented after burns, however, few
have been intensively monitored to determine their impacts on fire effects. Two
prescribed burns in northwest Wyoming, which escaped into adjoining grazing
allotments, were monitored during 1987 to provide data on effects of postfire
grazing management on vegetative response.

a. Blue Creek Coordinated Resource Management Plan (CRMP). In 1984, the

operators agreed to a CRMP with the Wyoming Game and Fish Department and
the Bureau of Land Management. The area is located south of Meeteetse,
Wyoming, in the 15 to 19 inch (38 to 48 centimeters) precipitation zone at the 7,800
foot (2,377 meters) elevation. The vegetation type is composed of limber pine
(Pinus flexilis) and mountain big sagebrush (Artemisia tridentata vaseyana) on a
shallow loamy range site. Key graminoid species include Idaho fescue (Festuca
idahoensis), green needlegrass (Stipa viridula), and rhizomatous wheatgrasses.
The growing season in this area is from mid-May until about September 1. The
adjoining allotment received heavy livestock use from the first of July until snowfall
each year. The unburned site had a mountain big sagebrush canopy cover of 60
percent and produced an estimated 700 pounds per acre (785 kilograms per
hectare) of annual sagebrush growth.

The planned actions started in 1981, including nonuse of livestock grazing,

conducting prescribed burns, and fencing for grazing strategy implementation. In
the year of the prescribed fire, snow left the area earlier than normal. The
prescribed burn was conducted on April 3, 1985. The fire escaped across the
allotment boundary fence into the adjoining allotment, burning about 20 acres (9
hectares) of mountain sage type. About 15 acres (7 hectares) of the escaped fire
area were deferred from grazing for two growing seasons by a temporary electric
fence, that is, no grazing occurred during the growing season. The original intent
was to exclude animal use entirely but elk tore down the electric fence in
September both years.

Herbaceous production data was collected during July 1987 in four adjoining sites.
Weight estimates were made on ten plots, and two plots were clipped to obtain a
correction factor. The study included one transect in the Blue Creek allotment that
received no grazing for 4 years before the prescribed fire, and no use in the 27
months after the prescribed fire when the production data was collected. There
were three areas studied in the adjoining allotment, all of which were grazed in the
years before the prescribed fire. One area was unburned; the second received
season long grazing in 1985 and 1986; and the third site was the fenced area
where grazing was deferred throughout two growing seasons. Table IX-1 details
the site and species production data that was collected.

Table IX-1: Herbaceous Production (pounds/acre dry weight) - Blue Creek.

Species Unburned No Rest Deferred Nonuse

Idaho Fescue 280 74 568 1,056

Rhizomatous 19 447 210 762

wheatgrasses
Green needlegrass 24 51 182 103

Other Grasses and 590 920 828 713

Forbs

Forbs NA 579 1140 859

TOTAL 913 2071 2828 3493
All burned areas had higher grass production than the unburned area. However,
the areas with deferred grazing and nonuse had much higher production of Idaho
fescue, the preferred species. The nonuse area had twice the fescue production of
the area that was grazed the year before the fire and deferred for two seasons
afterwards. The unrested burned area had one-quarter of the Idaho fescue as the
unburned area. Higher postfire palatability of this preferred species likely resulted
in higher utilization rates by livestock, causing it to all but disappear from the site.

b. Orchard-Woods allotments. The second study area is located south of Ten

Sleep, Wyoming, in the upper 10 to 14 inch (25 to 36 centimeters) precipitation
zone at the 6,800 foot (2,073 meters) elevation. Vegetation was dominated by
mountain big sagebrush, bluebunch wheatgrass (Pseudoroegneria spicata), green
needlegrass, and Idaho fescue. The permittee for the Orchard Ranch was
conducting a prescribed burn on September 14, 1983, and the fire escaped into the
adjoining Woods Allotment. The Orchard pasture had received light to moderate
grazing with a deferred rotation strategy for two years prior to the fire and was
rested for two growing seasons afterwards. The adjacent Woods pasture received
heavy season-long grazing prior to the fire, and was deferred until seed ripe the
first two seasons after the burn. Herbaceous production data was collected on July
30, 1987, from both sides of the division fence by clipping ten plots (Table IX-2).

Table IX-2: Herbaceous Production (pounds/acre dry weight) - Orchard-Woods.

Key Species Woods Orchard

(Deferred) (Rested)
Green needlegrass, Bluebunch wheatgrass, 131 447
Idaho fescue
Other Grasses and Forbs 1,648 864
TOTAL 1,779 1,311

Total production on the deferred Woods allotment was higher than on the Orchard
area. However, much of this production was weedy grasses and forbs. Production
of the three preferred grass species on the Woods allotment was only one third of
that on the Orchard allotment which had been rested after the prescribed fire.

c. Management implications. The primary implication of the preceding examples

is that in order to increase production of late successional species, there must be a
commitment to rest after the burn and proper grazing management in the long-
term. The two growing-season rest after a burn greatly speeds the recovery and
improvement of the key species. If a burned area is not rested, the extra moisture
available after the sagebrush or other shrub and tree species are eliminated is
used by rhizomatous grasses, or early successional grasses and forbs, depending
upon what species are present before burning. If there are sprouting shrubs
present that are not used by livestock such as rabbitbrush, they can become
dominant in the community. Lupine, which is a legume, will take up the extra
moisture in the mountain sage type for the first two seasons if the late successional
grass species are damaged or lacking in the understory. The lack of sagebrush
invasion into the burn sites on Blue Creek shows the importance of maintaining the
maximum vegetative ground cover and vigor to help slow the recovery of
sagebrush seedlings on the site. It can be expected that forage would continue to
increase on these sites because peak production on a sage site generally does not
occur until the third to fifth year after a burn. An additional implication for prescribed
fire is that increased production of herbaceous species does not necessarily mean
that the site is enhanced, if, as on the Woods pasture, much of the production is
composed of annual weeds and rhizomatous grasses.

C. Resource Management Considerations

1. Fire Effects. The effects of fire on plants and their response characteristics are
described in detail in Chapter VI, this Guide. The following impacts and changes
must be considered in planning proper site management to obtain the desired fire
effects.

a. Damage to key forage and browse species by repeated heavy utilization by

animals or burning is very similar; therefore, many areas are impacted by fire and
then again by grazing and/or browsing animals.

b. Increased palatability and accessibility of grasses, forbs, and shrubs, influenced

by the green period, nutrient content, growth form, and removal of dead material,
occurs during the first few growing seasons after a burn.

c. Carbohydrate reserves of burned plants are lowered the first few seasons after a
burn.

d. Fire effects result in changes of animal behavior including distribution, utilization

rates, forage consumption, and frequency of use.

e. Prefire and postfire grazing management largely determines the benefit/cost

ratio because of its considerable influence on the life of the beneficial aspects of
the burn.

2. Plant Maintenance Factors. Little research has been conducted to determine

the best long-term management practices for burned areas. However, the findings
from plant ecology studies conducted over the past 60 years can be applied. The
following factors should be considered to determine grazing management
requirements:

a. Plant community health (carbohydrate reserves for plant vigor and recovery
potential).

b. Composition of plant species that occupy the site and their successional
position.
c. Recovery period required by species after burning for vegetative, root, and
reproductive growth.

d. Utilization limits for species and season of grazing. A key factor in range
deterioration has been overly optimistic estimates about the amount of allowable
use that an area can sustain. The research literature gives a range of 20 to 40
percent utilization of annual growth during the critical growth period and 40 to 60
percent during the plant dormant period depending upon species and management
objective. Use at higher levels can cause deterioration of the plant community. e.
Site potential, which considers the range site or habitat type description.

f. Plant species morphology and reproductive mechanism.

g. The type of grazing and browsing animals that use the site, both wild and
domestic, and rodent or insect use.

3. Length of Postfire Rest Period. The length of the period of postfire rest from
livestock use depends on these factors:

a. Ecosystem type and ecological condition before burning.

b. Vigor of vegetation prior to fire.

c. Season of fire.

d. Growing season conditions, including temperature and precipitation, prior to and

following burning.

e. Whether establishment of new plants is by seed or sprouting. Seedlings require

a longer rest period to become resistent to grazing damage. Browse species often
require second year growth before producing seeds.

f. The management objectives for the area. For example, the length of postfire rest
may vary significantly for the same area depending upon whether the area is being
managed primarily for range, wildlife, forestry, or recreation, or a mix of these
activities.

4. Summary Recommendation for Rest. Postfire rest from grazing is required on

both prescribed fires and wildfires, on seeded areas and unseeded areas. The
length of the period of rest from grazing is dependent upon accomplishment of
measured objectives for the area. General recommendations are:

a. Prefire.

(1) At least one growing season of rest before a prescribed burn is needed on
many sites to increase both root reserves and fine fuels needed to carry fire.
(2) Severely depleted sites may require several years of rest before burning in
order for key plants to regain vigor and reproductive capabilities.

b. Postfire.

(1) Burned areas should be rested until a good ground cover and a litter layer are
present to provide soil and watershed protection.

(2) One growing season of rest may be adequate on some highly productive or
high condition sites after a low severity fire. A minimum of two growing seasons
rest is recommended for most burned areas.

(3) Areas under intensive grazing management can be grazed immediately after a
fall-winter burn to harvest some unburned vegetation within the unit. This short use
period prior to regrowth could assist in nutrient cycling, roughing of the soil surface,
and breaking of any sealed ash layers. Because this use can also result in physical
damage to the plant roots and crowns, it should be closely monitored and occur
during a short time period. Time control grazing may be used to help accomplish
postfire objectives if proper consideration is given to animal impacts and plant
community recovery periods.

(4) Sites that are burned to increase utilization of unpalatable species should be
grazed immediately after the fire while the undesirable plants are young and
actively growing.

(5) Heavy utilization grazing of an area by wildlife or wild horses will require further
rest from permitted livestock if their numbers cannot be controlled. Wild horses are
difficult to move from their normal territories but any water close to the burn area
can be regulated, especially during the first few growing seasons. Wildlife numbers
can be controlled by harvesting excess animals, although this is only a feasible
management strategy in the event of an extremely large wildfire.

(6) Below normal precipitation may delay vegetative recovery. A burned area
should be inspected to determine if it is ready for grazing. It cannot be assumed
that vegetation is ready for grazing just because the prescribed period of rest has
occurred. The degree of accomplishment of measured objectives should be the
most important criteria in determining the length of the postfire rest period.

(7) If recovery of wildlife browse is a key objective, the length of the postfire rest
period may need to be much longer, and permitted levels of utilization by livestock
may be much lower, than if target species for an area are grasses and forbs to be
used by livestock.

c. After postfire rehabilitation. If postfire rehabilitation has been conducted,

livestock should not be allowed back onto the burned area without evaluating
whether rehabilitation objectives have been met. Postfire rehabilitation
considerations are discussed in Chapter VI.C.6.a., this Guide.

5. Effects of Management Strategy. The commitment to long-term grazing

management on burned areas is vital for real long-term improvement in plant
productivity and composition. Determination of desired plant community objectives
is necessary before the grazing strategy can be decided. Many sites proposed for
burning are in poor condition or have mature stands of trees with little understory
vegetation. These sites require long periods of time to progress through
successional changes to meet objectives. If introduced plant species are seeded
on the burned area, a grazing strategy must be designed to meet their survival
needs. Seeded sites may have to be fenced separately from native rangeland due
to phenology and palatability differences.

Effects of fire and grazing management cannot be easily separated; therefore, fire
effects on vegetation should be monitored and evaluated both in the short term
(prior to start of grazing or at the end of second growing season) and the long term
(after two cycles of the grazing strategy or 8 to 10 years). The short-term
evaluation assesses achievement of fire treatment objectives, while the long-term
evaluation considers the attainment of desired plant community objectives.

Specific grazing strategies must be designed to meet key species requirements

and land use planning objectives. These general principles should be kept in mind
in post fire management of burned areas.

a. Type of grass. Bunchgrasses require lighter utilization rates and longer rest
periods than do rhizomatous or annual plant species.

b. Palatable shrubs.

(1) Highly palatable sprouting shrubs often require long rest periods after fire to
allow restoration of carbohydrate reserves, to produce seeds, and permit seedling
establishment.

(2) Upland shrubs such as sagebrush require bare soil surface and minimal
herbaceous competition to enhance reestablishment. Therefore on many of the
drier and lower snowfall sites heavy spring grazing could promote sagebrush
establishment, if that is the management objective.

c. Riparian communities. Riparian communities with shrubs and trees require

long-term rest to recover and light utilization of key shrubs to maintain a healthy
community. It may be necessary to fence riparian areas for several years to allow
initial recovery.

d. Forb-dominated communities. In order to maintain a forb-dominated plant

community, the grasses must receive heavy grazing pressure or be burned more
frequently.
6. Economic Factors. The following general guidelines can be used when
analyzing fire effects that have an influence on economics.

a. Site selection criteria. The first step is to consider the site potential and set
objectives accordingly. These selection criteria can be used for analysis of fire
effects on either prescribed burns or wildfire rehabilitation.

(1) Ecological condition. Sites in high-fair or better condition should be selected for
prescribed burning if increased herbaceous production of desirable species is a
short-term objective and reseeding is not planned.

(2) Presence of desirable species. If improvement of native species is the

objective, it must be determined if there are enough desirable remnant plants to
make the burn worth conducting. Seeding is very costly and the native species are
more adapted to the sites.

(3) Invader species. Are undesirable species or noxious weeds present on the site
that might be favored by fire? Species such as cheatgrass, rabbitbrush, and
horsebrush are common problems on arid and semiarid rangelands. Any reseeding
must be completed before the first growing season to avoid dominance by
introduced annual species such as cheatgrass. Noxious weeds may require
treatment if there is a significant number present.

(4) Burn size. Is the burn acreage within the management unit large enough to
avoid livestock and wildlife concentrations, thus negating the positive fire effects?
Refer to Chapter VII.B.3.a. and b., this Guide, for a discussion of the impacts of
burn size and configuration on wildlife habitat.

b. Factors to consider in analyzing benefit/cost. The following points are given

to consider in planning prescribed burns to improve native herbaceous vegetation.
If the objective is to prepare a site for seeding or removing undesirable species,
then other principles could be involved.

(1) Conduct burns during the spring period or other seasons that require minimum
fire control efforts. The three major costs that can be reduced are equipment,
personnel, and fuel break construction. Offsetting factors include ignition problems
and objective accomplishment because fuel or soil moisture or weather limit the
likelihood of a successful fire.

(2) Burn the largest acreage possible within the constraints of the objectives for the
burn. The impact on animal species and other resources will determine the
maximum size.

(3) Sites with higher ecological condition and plant vigor respond more quickly and
favorably to burning. Greater production increases, a quicker recovery, and better
chance for seeding establishment occur when a given ecological site in the mid to
upper precipitation zone is burned and less damage to desirable species can be
expected.

(4) Extending the life of the fire effects through long-term grazing management can
improve the long-term cost effectiveness of a burn project.

E. Summary

Proper site management based on specific objectives and plant species is

essential in the management of fire effects. Improper grazing management can
easily nullify efforts put into prescription burning or wildfire rehabilitation, as well as
impede natural vegetative recovery after wildfire. Impacts of long-term grazing
management before and after a fire can be easily overlooked; therefore, proper
grazing management including the appropriate kind of livestock, the stocking rate,
the season and the intensity of utilization, and the length and frequency of use are
most important.

The period of nonuse by livestock necessary after a fire varies considerably with
the vegetative composition, site conditions, resource conflicts, and objectives of the
burn. Grazing closures apply to prescribed fires and wildfires, whether they are
artificially reseeded or recovery is by natural means. In some situations, the only
way to ensure nonuse of critical areas after a fire is to construct fences.

Proper grazing management before and after a fire has a major impact on fire
effects, vegetation changes, economics, and rehabilitation success. In analyzing
fire effects, several site selection criteria should be considered including the site
potential, the ecological condition, the presence of desirable and invader plant
species, the acreage of burn within the management unit, and the livestock
management. The consideration and implementation of these factors determines
the benefit/cost ratio and the success of a burn project or postfire rehabilitation
effort.

This page was last modified 05/31/01

|Disclaimer| | Privacy| | Copyright| |USFWS Main Page| |Webmaster|

 National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 05/31/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home CHAPTER X - EVALUATION

Preface
Objectives By Ken Stinson and Melanie Miller
Fire Behavior
Fuels
A. Introduction
Air Quality
Soils & Water
Plants Documentation is the collection, organization, and storage of all
Wildlife information pertinent to a specific project, in order to maintain a long-
Cultural Res. term record, and to facilitate project evaluation. Evaluation is both an
Grazing Mgmt. objective and intuitive process of examining and assessing data and
Evaluation observations to determine if planned objectives were met. Fire effects
Data Analysis documentation and evaluation are important in order to replicate
Computer positive fire effects and avoid duplication of negative fire effects in
Soft. future prescribed fires; to assess postfire management of land uses on
Glossary both wild and prescribed fires; to assess the effectiveness of postfire
Bibliography rehabilitation in preserving site quality; to provide rationale for fire use to
Contributions the public; to meet legal requirements and liability; to permit evaluation
of objectives; and to provide a basis for improvements in project
planning, implementation, and management. Fire effects monitoring and
evaluation should be included in area monitoring and evaluation plans.

1. Documentation. All documents shall be labeled by project name and

number, date, and legal description. A detailed list of types of data that
can document a wildland fire are listed in C.3.a., this chapter, pages 3
to 5. The following are general classes of information used to evaluate
fire effects, and should be included in project records.

a. A complete description or reference shall be documented for the

method and technique used to collect the monitoring data. Monitoring
studies appropriate to evaluate land-use objectives should be used.

b. Photographs properly labeled, dated, and filed, including compass

direction of photo.

c. Observations before, during, and after a prescribed fire or wildfire.

d. Schedules and responsibility of evaluation and documentation.

2. Organization and Maintenance of Documents. Because fire effects

information requires long evaluation periods and the data accumulates
over time, it is important that fire monitoring and evaluation records are
placed in a permanent file. Records may be needed many years after
the original incident to study the site, or for comparison with more
recent events on similar sites. The data should be stored in the
appropriate permanent project file, such as allotment/ management unit
files, or timber sale file. Records should be kept indefinitely or archived
and labelled as "Permanent File - Do Not Destroy."

B. Considerations for Fire Effects Evaluation

Evaluation is both an objective and intuitive process of examining and

assessing data and observations to determine whether planned
objectives were met. There are two aspects of postfire evaluation,
evaluation of fire effects, and an evaluation of the effectiveness of
postfire management actions. A "cookbook" approach to evaluating fire
effects should not and can not be applied in all situations (Pellant 1989).

1. Prescribed Fire versus Wildfire Evaluations.

a. Fire effects. This evaluation provides the decision maker with the
necessary information to make sound decisions regarding postfire
rehabilitation and management actions.

(1) Prescribed fire. For a prescribed fire, fire effects are evaluated and
compared to the fire treatment objectives. The fire treatment is
assessed to determine whether it indeed enhanced resource
management objectives for the site. Because the event occurs under
pre-planned conditions, site-specific prefire and postfire monitoring can
occur. If desired effects are not attained, then an evaluation must
determine why they were not. If postfire rehabilitation is planned, an
assessment must be made whether the fire treatment suitably prepared
the site for seeding or planting. A decision is made whether postfire site
management objectives that were developed before the prescribed fire
are still suitable.

(2) Wildfire. The area in which a wildfire occurs may have established
resource objectives, and fire treatment objectives may have been
prepared in the Escaped Fire Analysis. However, because wildfires are
random events, any prefire monitoring with the specific intent of
documenting the effects of that fire can rarely be done. Some
monitoring data may be available that was performed for other resource
management programs or projects, and some of these data may be
suitable for postfire comparisons.

After a wildfire, an assessment is made of the degree to which fire may

have affected the ability of the land to meet any resource management
objectives that may have been established for the area. In particular,
the expected response of the vegetation on the site, and the potential
for site erosion, must be evaluated. A plan to take actions paid by
Emergency Fire Rehabilitation funds may be written if action is required
to mitigate the effects of the wildfire.(1) The decision to use EFR funds is
very important; therefore, appropriate personnel should be utilized and
the evaluation process should be initiated as soon as possible. Fire
effects evaluation can begin long before the fire is declared out.
Recommendations are made whether a continuation of present site
management practices is acceptable, or if changes are needed, such as
in grazing management or use by all terrain vehicles.

b. Effectiveness of postfire actions. An evaluation should be

conducted to determine whether postfire site rehabilitation treatments
limited negative effects of fire, to assess the impacts of postfire
activities such as salvage logging, and to establish the effectiveness of
postfire site management actions, such as grazing restrictions, in
preserving or improving site quality.

c. Effect of fire and postfire actions on attainment of site

objectives. Depending on whether a wildfire or prescribed fire is being
evaluated, and the level of land use goals and objectives that existed
for the site before the fire occurred, an assessment must be made of
whether resource management, fire treatment, and/or rehabilitation
objectives were met.

2. Degree of Evaluation. The degree of evaluation that is required

depends on:
a. The complexity of the project; several resources may have been
monitored and documented.

b. Whether the resource and/or treatment objectives were tied to

ecosystem effects or individual species responses.

c. The potential for controversy, involving such factors as designated

Wilderness, critical wildlife or watershed values, Areas of Critical
Environmental Concern, or political considerations.

d. Experience with or understanding of site specific fire effects, or

rehabilitating or managing similar areas.

e. Time and funding availabilities.

f. Availability of existing data bases. Other disciplines may have

extensive records, such

as streamflow, weather data, or soils inventories, that can be

incorporated into the evaluation.

3. Steps in the Evaluation Process. Existing agency policies and

procedures may require certain steps to occur in specific order. The
following steps are suggested for postfire evaluation.

a. Identify parties responsible for conducting the evaluation process.

b. Review objectives.

c. Assemble data.

d. Interpret data/observations.

e. Determine if and to what degree objectives were met.

f. Prepare evaluation report, which includes recommendations, for

decision maker.

g. Disseminate new findings to colleagues.

h. Observe long-term changes.

C. Evaluation Procedure

1. Identify Parties Responsible for Conducting the Evaluation. An

interdisciplinary team approach is used to evaluate both prescribed fires
and wildfires. Managers designate both a team and a team leader. All
resource specialists involved in planning a prescribed fire or who had
the lead in conducting it (Burn Boss or Project Coordinator), should be
involved with a prescribed fire evaluation. Any resource specialist
involved in the suppression of a wildfire, particularly the Resource
Advisor or Environmental Specialist, should be included on the team
that assesses the effects of a wildfire. It is very important that
individuals with fire effects experience, particularly in similar vegetation
types, be involved in the evaluation.

2. Review Objectives. Review any general or specific objectives

developed for the site being evaluated. These can include land use
decisions, fire planning objectives, resource management objectives,
prescribed fire objectives, fire treatment objectives, and rehabilitation
objectives. Rehabilitation objectives referred to here are those
objectives established as goals for the rehabilitation treatment, such as
reducing soil loss, or preventing invasion by exotic plant species. Other
objectives that are standard operating procedures are also evaluated,
such as keeping a prescribed fire within the prescribed fire target area,
meeting safety concerns, protecting cultural resources, and staying
within the cost target for the project.

3. Assemble Data.

a. Data sources.

(1) Project file.

(2) Project plan such as Allotment Management Plan, or Timber Sale

Plan.

(3) Site specific plan such as Prescribed Burn Plan or Rehabilitation

Plan.

(4) Records of any onsite evaluations conducted by Interdisciplinary

teams, such as preparation for a prescribed fire, or observations made
during or shortly after a wildfire.

(5) Any reports associated with the occurrence of a wildfire being

evaluated, such as the Escaped Fire Situation Analysis, Burned Area
Report, or reports by the Fire Behavior Analyst, Incident Commander, or
Resource Advisor.

(6) Fire effects data or evaluations from similar projects, such as other
prescribed fire or wildfire evaluations.

(7) Climatological data.

(a) National Oceanographic and Atmospheric Administration (NOAA),

e.g., monthly summaries.

(b) Remote Automated Weather Stations (RAWS), including archived

data.

(c) Local manual weather stations.

(d) State climatologist.

(e) Soil Conservation Service (SCS), e.g., snow surveys and soil
moisture indices.

(8) Weather, fuel, or soil moisture data collected at the time of the fire.

(9) Air quality permits, smoke observations, or data collected at the time
of the fire.

(10) Any site specific monitoring information, such as preburn and

postburn fuel, soil, vegetation, or fuels data. For a wildfire, some
information on prefire vegetation composition may be obtained from any
resource management or activity plan for the area, or from rangeland
inventory, trend, and utilization studies.

(11) Aerial photographs.

(12) Resource maps, such as vegetation, soil, timber, or cultural site
locations.

(13) Records that document uses of the area.

(a) Timber sales, regeneration surveys, and records of post-logging

treatment, such as slashburning, pile burning, scarification, or no
treatment.

(b) Actual use and utilization, and season of use by licensed livestock
or wild horses; utilization levels on key species; unauthorized use
records (trespass file).

(c) "On-the-ground" observations, such as extensive amounts of wildlife

use during a particular period of time.

(d) Use by other public land users, such as sportsmen or conservation

groups, fish and game agencies.

(e) Observations on extreme weather events, anomalous climatic

trends, grasshopper or rodent infestations, wildlife use, and ORV
impacts.

b. Data adequacy. Compare the amount of monitoring data collected

with the amount planned for collection. If data collection was
inadequate, determine the reasons why planned monitoring was not
carried out.

c. Statistical analysis. Determine if the appropriate statistical analyses

have been completed, or if inadequate data were collected to conduct
the analysis. (See Data Analysis.) If data were collected in such a
fashion that statistical analysis is not possible, or is inadequate, note
should be taken so future data are properly gathered.

4. Interpretation of Data/Observations.

a. Uncertainty and reliability. Fiscal and time constraints do not allow

for collecting enough data for managers to make risk-free decisions,
and some uncertainty will remain, even if a statistical analysis is
conducted. Decide what risk level to accept, and whether any data
inadequacy is the result of uncontrollable factors such as acts of nature
(floods, hailstorms), vandalism, or equipment failures. If poor study
design prevents proper data analysis, document this fact. Be sure the
decision maker is informed of the reliability of data; i.e., distinguish
between use of complete and properly collected data (hard data) and
assumptions made where data gaps exist (soft data).

b. Results of data analysis. Explain data analysis results in terms that

non-statisticians can understand. Integrate any quantitative or
qualitative, onsite data with observations and results from other studies.
Be wary of interpreting data mechanically without considering the
possibilities of undocumented causes for change. If necessary, use
expertise from other offices to decipher complex interrelationships.
Prepare a summary report of the findings. 5. Determine If and To What
Degree Objectives Were Accomplished.

a. Indicate whether each planned objective was met and to what degree
it was accomplished.

b. If objectives were not met, identify reasons why. Some general areas
to consider are:

(1) Objectives were unattainable or mutually exclusive.

(2) Timeframes for objective accomplishment were unrealistic.

(3) Operational procedures used to implement the project were not

appropriate, such as an ignition method or ignition pattern that caused a
more severe treatment than needed to produce the desired fire effects.

(4) The operational plan and/or procedures outlined in the project plan
were not followed.

(5) Followup site management was inadequate; e.g., postfire grazing

occurred too quickly or at too high an intensity, or off road vehicle use
occurred on a burned area before vegetative recovery could sustain
ORV use.

(6) Monitoring techniques or sampling intensity were not adequate to

detect change.

6. Prepare Evaluation Report with Summary and

Recommendations.

a. Prescribed fire.

(1) Brief summary of actions taken and burning conditions.

(a) Date, time of day fire occurred, acreage burned, and ignition pattern
and method.

(b) Weather, fuel, and soil moisture conditions during the fire.

(c) Observed fire behavior and characteristics.

(d) Observed smoke production and characteristics.

(e) Any differences between fire prescription and actual burning

conditions.

(f) Map of the proposed fire area and the area that actually burned.

(2) List of objectives of prescribed fire.

(3) Summary of monitoring results and observations.

(4) Describe which and to what degree objectives were met (expected
results versus actual results).

(5) If objectives were not met, briefly describe those factors that were
responsible for the lack of achievement.

(6) Make recommendations to management.

(a) Immediate changes needed in management strategies.

(b) Changes in how future prescribed fires could be planned or

implemented, including changes in prescription or ignition that would
lead to better results.

(c) Changes in postfire site management.

(d) Changes in monitoring procedures.

(7) After management review and decision(s), evaluation report should

be filed in the appropriate file(s) and required followup actions assigned
and initiated. This report should be signed and dated by the preparer
and the reviewing official.

(8) Followup on management decision(s) - repeat evaluation process.

b. Wildfire - fire effects evaluation.

(1) Brief summary of wildfire extent and effects.

(a) Date and acreage of public lands burned.

(b) Prefire vegetation.

(c) Map of the burned area including soil mapping units.

(d) Multiple-use objectives identified in land use plan, e.g., watershed

value, wildlife habitat, livestock forage.

(e) Interdisciplinary team findings.

i. Estimated percent survival of key plant species.

ii. Potential for postfire erosion and sedimentation.

iii. Potential for invasion of site by exotic species.

(2) Recommendation for postfire actions.

(a) Postfire rehabilitation actions, e.g.,

i. Seeding, including recommended species and seeding rates.

ii. Contour falling of trees.

iii. Construction of instream structures to control channel erosion.

(b) Feasibility of conducting salvage logging.

i. Locations within burned area.

ii. Methods that will not cause negative impacts on soil.

(c) Changes in postfire land uses, e.g.,

i. Temporary restriction or exclusion of grazing to allow recovery of

perennial plants or establishment of seeded or planted vegetation.

ii. Road or all terrain vehicle closures.

(d) Need for fence construction to exclude livestock and/or wildlife.

(3) Specific objectives for postfire management: specify desired

condition, e.g.,

(a) Limit erosion to a specified amount.

(b) Restore site to a productive state by salvaging merchantable timber

and replanting trees, and leaving a specified number of wildlife trees
that do not pose a safety hazard.

(c) Obtain a specific level of plant cover or productivity.

c. Wildfire - rehabilitation and/or management evaluation.

(1) Prepare a brief summary of actions taken.

(a) List management objectives and planned actions.

(b) Describe actions taken and date.

(c) Include a map of treated or rested areas.

(2) Summary of monitoring results and observations.

(3) Describe which and to what degree objectives were met.

(4) If objectives were not met, briefly summarize what factors were the
cause for lack of achievement.

(5) Make recommendations to management.

(a) Immediate changes required in management strategies.

(b) Recommendations for improvement in rehabilitation and postfire site

management practices.

d. Post evaluation actions.

(1) After management review and decision(s), the evaluation report

should be filed in the appropriate file(s) and the required followup
actions assigned and initiated. This report should be signed and dated
by the preparer and the reviewing official.

(2) Followup on management decision(s) - repeat evaluation process if

appropriate.

(3) If no management action was taken (e.g., the burned area was
grazed, or no erosion control efforts were made), document any
adverse impacts that occurred and what future management actions
need to be taken under similar conditions.

7. Disseminate New Findings To Other Interested Parties.

a. Fire Effects Information System.

b. Internal or interagency newsletters.

c. Professional meetings.

d. Local, state or regional workshops.

e. Agency technical publications.

f. "Expert Systems" or relational databases.

g. Office-to-office memos.
h. Professional journals.

i. Videos and other visual media.

j. Electronic bulletin boards.

8. Observe Long-Term Changes.

a. Within what timeframe did the original target species reestablish?

b. Are noxious weeds or unwanted species increasing in the postfire

environment?

c. Have cyclic climatic events, such as drought or a series of wet years,

affected postfire vegetative recovery? What effect have they had on
plant community composition and characteristics?

d. Are postfire management actions still having the desired effects?

e. Is the level or type of postfire management adversely affecting

desirable native vegetation or seeded species?

f. How have the different species that were originally seeded persisted?

g. Is cumulative acreage burned by wildfires large enough to affect

whether future prescribed burns should be carried out?

D. Summary

Evaluation of both monitoring data and the impacts of postfire activities

must be conducted in order to ensure that lands receive the best
possible fire treatment, rehabilitation, and postfire management. Once
we have monitored and evaluated enough projects and management
actions on similar sites, and adjusted our actions based on these
results, we can become more confident that the proper treatment is
being implemented. The same degree of monitoring and evaluation
need not be carried out on all subsequently treated areas if vegetation
type, soil type, and treatment prescription are similar to that of other
successful treatments. However, it is professionally unacceptable to
conduct no prefire or postfire monitoring or site observation, to assume
that an area is ready for grazing because the designated length of time
has passed since the occurrence of a prescribed or wildfire, or to
conduct no evaluation of the implementation or effectiveness of postfire
site management in preserving or enhancing site quality. Without some
check on the results of our activities, accumulated assumptions can
lead to land treatments that do not meet resource management goals
and objectives, and lead to deterioration, instead of enhancement, of
site quality.

1. See BLM Manual Handbook H-1742-1 (USDI-BLM 1985b) for a

discussion of EFR planning procedures on BLM land.

This page was last modified 05/31/01

|Disclaimer| | Privacy| | Copyright| |USFWS Main Page| |Webmaster|

 National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 05/31/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home CHAPTER XI - DATA MANAGEMENT

Preface
Objectives By Dr. Robert Clark
Fire Behavior
Fuels
A. Introduction
Air Quality
Soils & Water
Plants Managers can gain a number of benefits from the analysis of their data.
Wildlife The results of the analysis helps managers make informed decisions
Cultural Res. leading to favorable outcomes. Analyzing and manipulating the data
Grazing gives the manager a special familiarity with the data and a feel for how
Mgmt. precise and repeatable the sampling is likely to be. Describing and
Evaluation defending the credibility of the data is easier when it has been
Data Analysis appropriately analyzed and described in standard statistical terms.
Computer Practice and familiarity with the most common statistical analyses
Soft. provides a basis for better understanding, interpretation, and proper
Glossary application of results presented in professional fire literature.
Bibliography
Contributions This chapter will acquaint fire managers with some elementary statistical
methods that can be used to help make decisions. The methods are
demonstrated by example. Managers interested in more detailed
discussion or more sophisticated methods are referred to Freese (1962,
1967), Little and Hills (1978), and Eshelman et al. (1986). Managers are
also encouraged to contact a statistician before data are collected to
ensure the data will be usable and appropriate, and that assumptions of
analysis are not violated.

Statistical terms used in this chapter are included in the Glossary of this
Guide. Additional discussion is available in statistics texts.

B. Principles and Procedures of Data Analysis

1. Determination of Sample Size (Number of Observations).

Samples (observations) cost money, but so do decisions based on
inadequate data. Therefore, it is helpful to collect an appropriate number
of observations. One useful method to determine the appropriate sample
size for continuous data is (Freese 1967, p. 12):

n = (t2s2) / E2

where:

n = the required number of observations.

t = tabulated Student's t available in most statistics texts.
s = sample standard deviation (and s2 = sample variance).
Note the unfortunate situation here, that some estimate of
variance is required to calculate sample size, i.e., there is
one equation with several unknown values. This is usually
handled by collecting a preliminary sample from which to
estimate a variance and a mean, which in turn are used to
calculate the sample size.
E = desired precision, expressed as a proportion of the
sample mean.

Example: Suppose the manager needed to know the foliar moisture

content of basin big sagebrush on a proposed prescribed fire area.
Because fire behavior in sagebrush is closely related to foliage moisture
content, the manager decided to sample such that the moisture content
was estimated within 10 points of the true moisture content level at the
90 percent confidence level. This indicated that the manager wanted to
know the moisture content on the site within plus or minus 10 points of
the true average value (precision level), and wanted the estimate to be
correct 9 out of 10 times (confidence level). 10 points of error was
selected because fire behavior in sagebrush can begin to change quite
significantly over a range of 20 percent foliar moisture content (plus or
minus 10 percent).

The manager collected foliage from 12 different plants, dried and

weighed the samples, calculated the moisture content, and recorded the
following results. What sample size is required to meet the manager's
needs? Was the preliminary sample adequate?

x = moisture
plant x2
content (%)
 1 70 4900
2 90 8100
3 80 6400
4 70 4900
5 60 3600
6 80 6400
7 90 8100
8 100 10000
9 90 8100
10 80 6400
11 70 4900
12 60 3600
n= Ê x2 =
Êx = 940
12 75,400

Note that the Ê symbol means summation.

a. Step 1. Look up t value from t table in most statistical texts (e.g.,

Freese 1967, p. 77). Degrees of freedom (df) are n - 1 (12 - 1 = 11). The
appropriate t value at the 90 percent confidence level (or 10 percent
error rate) is 1.796.

b. Step 2. Calculate the variance (s2) of the preliminary, 12-observation

sample according to [note that this calculation results in the sample
variance (s2) rather than the sample standard deviation (s)]:

s2 = Êx2 - ((Êx)2/n) / n-1 = 75,400 - ((940)2 / n) / 11 = 160.6

c. Step 3. Decide how much error can be tolerated. In this example, the
manager wanted the estimate to be within 10 percentage points of the
true mean moisture content. Therefore, E = 10.

d. Step 4. Calculate the appropriate sample size according to the

formula described above:

n = (t2 s2) / E2 = (1.796)2 (160.6) / (10)2 = 5.18 = 6 observations

e. Conclusion. The preliminary sample of 12 observations was
adequate. If the preliminary sample of 12 observations had been
inadequate, the manager had several options. First, more observations
could have been taken. Second, the acceptable precision level could
have been lowered. Third, a lower confidence level could have been
accepted and some added risk incurred. Or fourth, the fire prescription
might have been altered (e.g., faster windspeed or lower fuel moisture
content) to compensate for the added uncertainty. The important point is
that the manager had some quantitative information on which to base a
decision.

f. Note. The size of the preliminary sample depends on the variable

being measured, on the objective for measuring the variable, and on
funds, time, and work force constraints. Inherent variability associated
with most natural resource sampling indicates that 10 or more
observations may usually be required to obtain a reasonable estimate of
variance. One observation is never adequate because degrees of
freedom would be zero and no analysis is possible.

2. t-Test for Paired Observations (Plots). This test is especially useful

for comparing effects between two fire treatments, such as burned vs.
unburned, or backing fires vs. heading fires. An assumption of the test is
that pairs of plots are established prior to the treatment, each group has
the same population variance, the population of observations follows the
normal distribution, and that treatments are randomly assigned to each
individual plot. It is possible, however, to establish plot pairs adjacent to
firelines (one plot on either side) such that these assumptions are not
violated. A similar test for unpaired plots (Freese 1967, p. 24) should be
used where plots are not paired; that test, however, is slightly more time
consuming and a slightly greater loss of sensitivity may occur. Only the t-
test for paired plots is illustrated here.

Example: A fire manager suspected that fire residence time might be an

important factor affecting postburn Idaho fescue (Festuca idahoensis)
production. The manager designed an experiment to evaluate the
suspicion. An area was selected where part of the burning would be
done by a backing fire and part by a heading fire. A series of 10 plots
were clipped, oven-dried, and weighed preburn to establish that the two
areas were from the same population and had similar variances, and to
determine preburn fuel loading. The heading and backing fires were
then conducted on the same day under similar environmental conditions.
Sufficient rate of spread and flame depth measurements were made
during the fires to establish that residence time was significantly (p =
0.05) different between the two treatments. One year later the fire
manager clipped, oven-dried, and weighed Idaho fescue standing crop
to compare "production" on the two treatments. Based on the following
results, and assuming that residence time (and its implications) was the
only difference between the heading and backing fires, does residence
time influence postburn production?

d=
Heading
Quadrat Heading Backing d2
-
Backing
1 22 24 -2 4
2 15 13 +2 4
3 19 19 0 0
4 19 12 +7 49
5 21 17 +4 16
6 20 17 +3 9
7 18 19 19 1
8 20 12 12 64
9 21 15 15 36
10 14 14 14 0
n = 10 Êh=189 Êb=162 Êd=27 Êd2=183

xh = 18.9 xb = 16.2

a. Step 1. Note that by inspection the two means appear to be different

(18.9 appears greater than 16.2). The null hypothesis, Ho, is that the two
means are equal, i.e., xh is equal to xb. The alternate hypothesis, Ha, is
set up to allow rejection, i.e., xh is not equal to xb. This hypothesis will
result in a "two-tailed" test; one-tailed tests also can be established by
setting the means in the hypotheses "greater than" or "less than" rather
than "not equal to." Note that one-tailed tests require different t values
than two-tailed tests.
b. Step 2. Establish the confidence level for testing. By convention,
testing in this example was set at the alpha = 0.05 level (this is the same
as the 95 percent confidence level, or that the acceptable risk is to be
wrong one chance in 20 due to chance alone). From the t table, with n-1
degrees of freedom and an error rate of 0.05, t = 2.262. This is the
tabulated value below which the difference between sample means is
likely to be due to chance alone. A calculated value larger than 2.262
would suggest that a real difference exists between the two means, and
on the average, this conclusion would be correct 19 of 20 times.

c. Step 3. Calculate the variance of the difference between sample

means according to:

sd2 = Êd2 - (Ê(d)2 / n) / n - 1 = 183 - ((27)2/10) / 9 = 12.2333

d. Step 4. Calculate t according to:

t = (xh - xb) / §[sd2 / n] = (18.9 - 16.2) / §[12.2333 / 10] = 2.4412

§ = square root

e. Conclusion. The calculated value of 2.4412 is larger than the

tabulated t value of 2.262. This suggests that a real difference exists
between the two means. The fire manager may report to the supervisor
that postburn Idaho fescue production was different (p = 0.05) between
the heading fire and backing fire sites, and that more production
occurred on the sites burned with a headfire. If it is certain that the only
difference between treatments is residence time, it is also reasonable to
assume that long residence time was more detrimental than short
residence time. This conclusion is likely to be correct 19 out of 20 times.
Note that this statistical significance does not necessarily imply
biological significance. Also note that this "study" was not replicated, nor
was a "control" treatment used; both are strongly encouraged. Also note
that the method and level of testing, and the hypothesis, were
established before any sampling or burning occurred.

3. Chi-Square Analysis of Counts. Chi-square is a nonparametric

method that is useful for analyzing binary enumeration data that fall into
two categories, such as scorched or not scorched, alive or dead, scarred
or unscarred, or sprouted or not sprouted. These types of data are
common in fire management. Although several procedures are
available, the following example illustrates a useful method for many fire
management data.

Example: One half of a plant community containing bitterbrush (Purshia

tridentata) plants was burned in the fall and the other half was burned in
the spring. The fire manager tagged 20 randomly located bitterbrush
plants in each area, before burning, to estimate mortality. One year after
burning, the fire manager found 11 tagged plants alive on the spring
burned site and eight tagged plants alive on the fall burned site. At the
90 percent confidence level, was there a differential response between
fall and spring burned plants?

a. Step 1. Set up a 2 x 2 contingency table (2 rows and 2 columns of

observations):

Spring Fall
Total
Alive 11 8 19
Dead 9 12 21
Total 20 20 40

b. Step 2. Calculate chi-square. The general procedure, based on a 2 x

2 contingency table, is:

I II Total
1 a b a+b
2 c d c+d
Total a+c b+d a+b+c+d

chi-square = ([(| ad - bc |) - 1/2 n]2 n) / (a + c) (b + d) (a + b) (c + d)

chi-square = [(|(11 x 12) - (9 x 8)| - 1/2 (40)]2 (40)) / (20) (20) (19) (21)
= 0.4010

Note that the vertical bar ( | ) indicates absolute value, so it is irrelevant

which cross-multiplication product is subtracted from the other (i.e., 11 x
12 - 9 x 8, or 9 x 8 - 11 x 12). Use of the Yates Correction, [- 1/2 (n)] and
[1/2(40)] found in the numerators of the above equations, decreases the
chi-square value, and thus reduces the chance of declaring a significant
difference when one does not exist.

c. Step 3. Look up the tabular value for chi-square in a chi-square table,

such as Freese (1967, p. 82). Values in a chi-square table are different
than those in a t-table, and a t-table should never be used to obtain
values for a chi-square test. Degrees of freedom (df) for the 2 x 2
contingency table are calculated according to: [(rows - 1) x (columns -
1)]. In this example, df = (2-1)(2-1) = 1. Since the fire manager decided
to test at the 90 percent confidence level (10 percent error level), the
appropriate chi-square value to test this example is 2.71.

d. Conclusion. Since all values below 2.71 are below the 10 percent
error threshold, the calculated value of 0.4010 is not significant. The
manager concluded that, in this example, the sprouting of bitterbrush
was not different between spring and fall fire treatments. Note that no
cause and effect was implied. If a statistically significant difference had
been observed, the difference could have been due to environmental
factors or other unknown causes. This point is especially important for
data gathered during different time periods.

4. Other Tests. Two additional statistical testing methods that are

beyond the scope of this Guide but may be useful in routine fire
management work include analysis of variance (often abbreviated AOV
or ANOVA), and linear correlation and regression.

a. Analysis of variance. ANOVA is a method used to separate sources

of variation within treatments and among treatments and may be used
with several different experimental designs such as the completely
randomized design or the randomized complete block design. ANOVA
produces an "F" test of significance, and is often used with several mean
separation tests, such as Duncan's New Multiple Range Test, or
Orthogonal Comparisons. Although ANOVA has many potential
applications in fire management, a statistician should be consulted
before its use. Several texts, for example, Little and Hills (1978), provide
excellent discussions of ANOVA.

b. Linear correlation and regression. Linear correlation and

regression analysis are mathematical methods used to describe how
independent variables relate to each other, and how dependent
variables relate to independent variables. These approaches to
describing and understanding relationships common in fire management
work do not imply any cause and effect; they merely describe
associations and relationships. Because the potential for misuse is so
great, many statisticians discourage their use by apprentices. They are
especially useful, however, when used with the aid of a reputable
statistician.

5. Crunching Numbers. Development of microchip technology has

made the manipulation of large data sets rapid and easy for fire
managers. The three simple tests (sample size, t-test, and chi-square)
that are described above can all be easily calculated on simple, 4-
function, hand held calculators. Many small, hand held calculators are
commercially available that have internal, preprogrammed statistical
functions that quickly calculate the sample mean, variance, and
sometimes, even complete linear regression. One caution is that the
user should be aware whether the calculator in question uses n or n - 1
for degrees of freedom in the variance and standard deviation, and
should understand the implications.

The Hewlett-Packard HP-71B hand held calculator, which is routinely

used for fire behavior calculations, has several Custom Read Only
Memory (CROM) modules available. One CROM (American Micro
Products, Inc. 1984) contains many parametric and nonparametric tests
and is powerful enough to handle relatively large data sets. Such
devices make statistical analyses of fire management data routine.

Many statistical packages have been developed and are available for
micro and mini-computers (e.g., StatSoft, Inc. 1987). Some packages
include graphics capability, and many are available for several hundred
dollars or less. Fire managers who anticipate the collection of large
amounts of data for statistical manipulation are encouraged to
investigate statistical packages that are available.

Mainframe statistical packages (Dixon 1985, SAS Institute Inc. 1985) are
designed to handle very large data sets and complete an enormous
number of statistical tests. These large machines are usually restricted
to research institutions; however, fire managers should be aware that
most colleges, universities, and experiment stations have access to
mainframe computers. The use of such machines might be cost effective
if used only when infrequent but large data sets must be manipulated.

C. Resource Management Considerations

1. It is essential to know what the questions are before sampling and
data analysis can be designed to obtain answers. Sampling and data
analysis must be objective driven. Further, objectives should be
developed with sampling and data analysis in mind so that the
objectives are reasonable, measurable, and lend themselves to
analysis.

2. Sampling design and intensity (number of observations) should be

determined before initiating any data collection to ensure sampling and
analysis procedures are appropriate. The number of required
observations depends on desired precision and confidence levels; for
most management purposes it is usually adequate to sample within 20
percent of the mean at the 80 percent confidence level.

3. Experienced statisticians should be consulted before data are

collected and after data are analyzed.

4. It is not necessary, nor feasible in most cases, to sample and analyze

data from every community on every wild or prescribed fire. Usually it is
better to do an adequate job on one community than an inadequate job
on two or more communities. Further, it is often possible to design a
series of prescribed fires with a sampling and data analysis scheme
such that one fire or one stratum is emphasized and the remainder are
spot checked. Biologically oriented statisticians can provide advice on
how best to sample and analyze data when time and funding are
constrained.

5. Inadequate sampling is often more expensive than excessive

sampling; optimum sampling is usually the most cost effective.

6. Replicates are necessary to determine "sample error," and untreated

"controls" are necessary to isolate fire effects from other effects.

7. After data are collected and analyzed, consider both statistical

significance and biological significance; it is possible to establish
statistical significance that has no biological significance.

8. Place measured trust in results of data analysis. An unexpected or

undesired result is not a valid reason for discarding results and
"massaging" the data. Data analysis should be used to enhance
understanding as well as provide support.
9. It is tempting to draw inappropriate conclusions from analyzed data.
Caution is advised.

D. Methods to Monitor Fire Effects

1. The appropriate sample size for each data set should be determined
using the method described in B.1, or another statistically acceptable
method. A more detailed description of this method to determine proper
sample size is given in Norum and Miller (1984).

2. The t-test described in B.2 is especially useful for comparing the

means of several treatments, such as burned vs. unburned, or fall
burned vs. spring burned. Means that are helpful to compare include
"production," cover, and density. The test assumes that identical
treatments were applied; therefore, the comparison of spring burned vs.
fall burned treatments should be approached with caution.

3. Chi-square is a nonparametric test that is especially useful for

analyzing counts of binary data; such data fall into two convenient
categories, such as alive or dead, or sprouted or not sprouted.

4. A statistician should be consulted for additional data analysis

methods. Biometricians and statisticians who can provide assistance
may be located at agency offices, national and regional service centers
such as the BLM Service Center at Denver, experiment stations and
universities.

E. Summary

Statistical analysis of data and interpretation of results are helpful for

understanding fire effects and provide an essential tool for the decision
making process. Calculation of the appropriate sample size is essential,
and is based on desired precision and confidence levels. The t-test for
paired plots, and chi-square analysis of counts, are particularly useful for
understanding fire effects. Other, more sophisticated techniques may
require assistance of a statistician.

This page was last modified 05/31/01

|Disclaimer| | Privacy| | Copyright| |USFWS Main Page| |Webmaster|
 National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 05/31/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home CHAPTER XII - COMPUTER SOFTWARE

Preface
Objectives By Melanie Miller
Fire Behavior
Fuels
A. Introduction
Air Quality
Soils & Water
Plants Many packaged computer programs have been developed that can
Wildlife provide information useful for managing or interpreting fire effects. Some
Cultural Res. programs are accessible on mainframe computers, while others are
Grazing available on floppy disks that can be loaded onto personal computers.
Mgmt. Additional information about programs described here can be obtained
Evaluation by reading the referenced publications, or contacting specific offices,
Data Analysis where indicated.
Computer
Soft. B. Weather Analysis
Glossary
Bibliography Weather records can be used to manage fire effects by helping to
Contributions establish seasons and parameters for appropriate fire prescriptions, and
to document prefire weather trends and weather conditions at the time of
the fire. Records of postfire weather can be used to make inferences
about causes for observed fire effects. Long-term records are required
to establish a statistical basis for occurrence of seasonal trends in
elements such as temperature or rainfall, to detect deviations from
seasonal trends, or to document the likelihood of occurrence of specific
sets of weather conditions, such as combinations of temperature and
relative humidity.

1. Computerized Databases. Sources for data on which computerized

searches and analyses can be performed are the network of Remote
Automated Weather Stations, and the National Interagency Fire
Management Integrated Database.
a. Remote Automated Weather Stations (RAWS). The Remote
Automated Weather Station (RAWS) System is a program that provides
current remotely sensed weather data from stations established on
Federally administered lands (German 1988). The RAWS network
contains about 350 weather stations on BLM lands in the lower 48
States, as well as about 250 weather stations established on other
agency lands. There are approximately 40 RAWS stations in Alaska.
Over the next 5 years, approximately 60 to 100 additional RAWS
stations will be established.

The RAWS system uses self-contained meteorological collection

platforms and a mini-computer controlled satellite receiving station
located at the National Interagency Fire Center (NIFC). The weather
stations collect weather data, summarize it on an hourly basis, and then
transmit data on a one hour or three hour basis to NIFC through the
GOES satellite system. A computer at NIFC immediately distributes the
weather data to all users on the Initial Attack Management System
(IAMS), to the Weather Information Management System (WIMS) (see
item 1.b. below), and to the National Weather Service. BLM users
access this "real-time" weather data through the IAMS computer located
in most BLM District Fire Management offices. Other agency users
acquire data through WIMS.

NIFC archives all weather data for permanent storage. A copy of this
archived information is distributed quarterly and annually to the Desert
Research Institute at the University of Nevada at Reno. Weather
information can be obtained in a quarterly summary report for each
station, and through specially requested reports and studies. BLM users
request these reports and studies from BLM staff at NIFC, while other
agency users obtain these data from the Desert Research Institute.

b. National Interagency Fire Management Integrated Database

(NIFMID). The National Interagency Fire Management Integrated
Database contains the historic fire weather information previously
contained in the obsolete National Fire Weather Data Library. It is
located at the National Computer Center - Kansas City (NCC-KC),
managed by the U.S. Department of Agriculture. The library contains
daily observations from all fire weather stations, collected at the time of
peak burning conditions (1300 hours, local standard time) throughout
the fire season. The Weather Information Management System
database (WIMS), a replacement for AFFIRMS, stores the1300-hour
observations from automated and manual fire weather stations, as well
as 60 days of 24-hour observations from RAWS stations. The long-term
historical weather database of NIFMID stores 1300-hour observations.

Access to NIFMID is available through computer software packages that

reside in the computer library at the Kansas City Computer Center. In
order to use these programs, users must establish an account. Once an
account is opened, all access information is provided to the user.
Telecom- munications packages on most computers can be used to
communicate with this USDA computer. See item 2. Computer
Programs, for descriptions of computer programs available through NCC-
KC.

2. Computer Programs.

a. KCFAST. KCFAST is a program that facilitates use of data stored in

NIFMID. KCFAST is a utility that assists the casual user of this database
by removing the need to remember all of the commands required to
extract and format weather data stored for a particular weather station.
KCFAST itself does not perform analyses on data sets from the fire
weather data library, but it makes it easy to transmit these data to a
personal computer, where weather analysis can be performed. It
facilitates operation of the other programs described in this section, as
well as other software such as that used for fire planning. KCFAST is
available in versions that operate on the U. S. Forest Service Data
General system, as well as IBM compatible personal computers. To use
the PC version, communication software is required that can transmit an
ASCII file to a remote computer and log screen activity to a PC file
(Bradshaw and Andrews 1990).

b. FIREFAMILY. Three different computer programs were consolidated

into FIREFAMILY: FIRDAT, SEASON, and FIRINF. These programs use
historic weather data to predict future fire management needs, and
integrate fire management with other land management activities. The
weather data used is that stored in the NIFMID (1.b., above), and
FIREFAMILY software is on the USDA Kansas City computer. These
programs are fully described in the publication by Main et al. (1988).

(1) FIRDAT. FIRDAT can combine up to 100 years of historical weather

records with National Fire Danger Rating System (NFDRS) equations to
produce frequency distributions and graphs of NFDRS indices and
components. This can include the NFDRS calculated fuel moisture.
FIRDAT can also produce lists of daily weather observations.
(2) SEASON. SEASON uses FIRDAT data to summarize variations in
NFDRS inputs and fire danger severity during a fire season, and
seasonal patterns of fire danger over many years. Fuel moisture values,
fire danger indices, or fire behavior components can be presented in
tabular and graphic form.

(3) FIRINF. FIRINF allows the analysis of the co-occurrence of any two
NFDRS indices over 10-year periods.

c. PCFIRDAT. PCFIRDAT is a version of FIRDAT that has been

rewritten for use on IBM compatible computers. Using the PC version of
KCFAST, data are transferred from the NIFMID to a personal computer,
and PCFIRDAT is then used to perform data analyses. PCFIRDAT, as
well as PC versions of SEASON and FIRINF are in the final stages of
development. Work is being performed by the California Department of
Forestry.

d. PRESCRB. This program provides a climatological summary of

specific fire weather occurrences. The program counts the occurrences
of days on which all variables specified in a prescription are met, and
gives the probable number of days in any year on which the prescribed
conditions specified are likely to occur. The average number of days of
occurrence of prescribed conditions during successive 10-day periods of
the year is given. The average number of days in each burning period
also can be obtained, giving the prescribed fire planner an idea of
whether days meeting prescribed conditions are likely to occur singly, or
in a series. Furman (1979) documents use of the PRESCRB program.
An advantage of this program is that it can use weather data from other
sources than NIFMID.

e. RXWTHR. RXWTHR (Prescribed Fire Weather) provides

climatological summaries and co-occurrence frequencies of user
selected fire weather and fire danger rating parameters (Bradshaw and
Fischer 1981a, 1981b). The data base is the computerized fire weather
records within NIFMD. Simultaneous occurrence of two or three of
fifteen prescription parameters can be summarized, and shown in
tables. For example, tables can include summaries of temperatures, two
way co-occurrence tables for wind direction and wind speed, and three
way co-occurrence tables for temperature, relative humidity, and wind
speed. Once a user has obtained a feeling for the general pattern of
occurrence of desired prescribed conditions, screening out those that
have a low probability of occurring simultaneously, a more detailed
analysis can be conducted using RXBURN. Please note that a sixteenth
prescription parameter available for analysis in the program is duff
moisture. This value is based upon a duff moisture model that was
offered on an experimental basis in RXWTHR and RXBURN. This model
does not provide accurate results, and the duff moisture output
produced by these two programs should not be used.

f. RXBURN. RXBURN (Prescribed Fire Conditions) provides an analysis

of the frequency of occurrence of a set of prescription conditions, also
using the NIFMID as a database. Up to 15 parameters can be used in a
single prescription. Users define a preferable range of prescribed
conditions, and a broader range of conditions that is still acceptable.
Output tables include a summary table of the percentage of weather
conditions that are preferable, acceptable, and unacceptable; a table
that shows frequency of occurrence of preferable, acceptable, and
unacceptable conditions in each successive 10-day period, and by
month; the number of successive days that prescribed conditions have
occurred in each 10-day period and each month; and the probability of
meeting the prescription 1, 2, and 3 days in the future for each month.
As discussed under 2.d., RXWTHR, the duff moisture model that
provides a basis for parameter 16 is inaccurate; this parameter should
not be used.

There are two limitations to the use of RXWTHR and RXBURN. Weather
observations are rarely recorded for more than the five months of the
year that are the normal "fire season," and many prescribed fires are
staged before and after this period. The single observation per day that
must represent the entire day's weather is taken during average worst
case conditions, 1300 hours (Standard Time), southwest aspect,
midslope, and open canopy. This may poorly represent the weather at
the time of the day when a prescribed fire would be implemented.

g. Other. An additional set of climatological software is described by

Bradshaw and Fischer (1984). Eight computer programs for extensive
climatic summaries of weather variables, temperature, relative humidity,
wind, and precipitation, records are available. Five basic climatology
programs analyze NIFMID records by 10-day periods and by month.
Three averaging programs adjust results from the climatology programs
to smooth variances caused either by short periods of record or
incomplete station data.
C. Fire Behavior

1. BEHAVE. BEHAVE is a set of interactive, user friendly computer

programs that are used for estimating behavior of wild and prescribed
fires. BEHAVE is an integral part of the Fire Behavior Prediction System
that is used by Fire Behavior Analysts to estimate fire potential under
various fuels, weather, and topographic situations (Burgan and
Rothermel 1984). BEHAVE predicts the behavior of a steady state fire
advancing in surface fuels along a front. It cannot be used to estimate
long-term fuel burnout, non-flaming combustion, or extreme fire
behavior. The BEHAVE system consists of two subsystems: BURN and
FUEL.

a. FUEL subsystem. The FUEL subsystem has two programs:

NEWMDL and TSTMDL (Burgan and Rothermel 1984). NEWMDL
("NEW MODEL") is used to construct fuel models for specific application
when the existing 13 stylized fuel models in the Fire Behavior Prediction
System do not adequately describe fuels at a particular location.
TSTMDL ("TEST MODEL") is used to test the fuel models developed in
NEWMDL. The FUEL subsystem is rarely used except by trained Fire
Behavior Analysts or for research purposes. Please note that these fuel
models are not the same as those used in the National Fire Danger
Rating System.

b. BURN subsystem. The BURN subsystem is frequently used on wild

and prescribed fires. This subsystem has three components, FIRE 1,
FIRE 2, and RXWINDOW (see section C.2.). FIRE 1 contains the
wildland fire behavior prediction model developed by Rothermel (1972).
It includes modules that allow prediction of fire behavior (DIRECT and
SITE), fire growth (SIZE), containment requirements (CONTAIN), and
spotting distance (SPOT). Fire effects on the tree overstory are
predicted in SCORCH and MORTALITY, described in sections D.1.a.,
and D.1.b., this chapter. FIRE 2 allows calculation of fine dead fuel
moisture (MOISTURE), the probability of ignition (IGNITE), and relative
humidity (RH). Contents and operation of the BURN Subsystem are
described in Andrews (1986) and Andrews and Chase (1989). Users
require training for optimum application of this software to wildland fire
situations.

2. RXWINDOW. RXWINDOW, a component of the BURN subsystem in

BEHAVE, essentially runs DIRECT backwards, enabling prescribed fire
planners to obtain detailed windows of required fuel moisture and wind
conditions based upon desired fire behavior (Andrews and Bradshaw
1990). Desired fire behavior such as flame length or rate of fire spread
are entered, and the program calculates which combinations of fuel and
weather parameters would result in the fire behavior specified. The
program uses one of the 13 standard fire behavior fuel models, or a
custom model developed in the FUEL Subsystem. Other input values
include slope of the burn area and exposure of fuels to the wind. The
user can optionally set limits on 1-hour, 10-hour, 100-hour, live woody,
and live herbaceous moisture content, and effective windspeed.

a. FIRE. The FIRE module generates tables that display combinations of

effective windspeed and weighted fuel moisture that result in the desired
fire behavior. It indicates those pairs of wind and dead fuel moisture that
yield a fire within the fire behavior prescription. For those fuel models
where live fuel moisture is a component, the program prints ranges of
live fuel moisture for each appropriate wind and dead fuel moisture pair.
b. WIND. The WIND module requires input of site description
parameters, and whether the prescribed fire will be a headfire, backfire,
or flanking fire. Based on a constant slope, the program prints a table of
20-foot windspeeds coming from different directions with respect to
slope that will produce a range of effective windspeeds. WIND can use
either effective windspeeds identified in the FIRE output table or
windspeeds selected by the user.

c. MOISTURE. The MOISTURE module of RXWINDOW displays

ranges of moisture contents for the 10-hour fuel size class that results in
a specific weighted fuel moisture, for a given 1-hour fuel moisture.
MOISTURE also produce a table that shows for a given herbaceous
moisture content, the range of woody fuel moisture that results in a
specific weighted live fuel moisture.

3. Availability. Contact agency fire management staff for information on

obtaining access to FIRE 1, NEWMDL, and TSTMDL. A copy of the
complete BEHAVE program (including FIRE 2 and RXWINDOW) that
runs on an IBM compatible computer can be purchased through a
government contract at a low cost. The program disks will become
available through the Publications Management System at the National
Interagency Fire Center.

D. Fire Effects

1. Models within BEHAVE. There are few fire effects that relate directly
to fire behavior, the activity of the flaming front of the fire. However, two
programs within the BEHAVE system predict two effects, crown scorch
height, and tree mortality, that can be directly or indirectly related to fire
behavior. Access to both of these programs is through the BEHAVE
system, described in section C. of this chapter.

a. SCORCH. A module of the FIRE 1 program of BEHAVE, SCORCH

predicts lethal crown scorch heights from flame length, ambient air
temperature, and midflame windspeed. This model estimates the
maximum height in the convection column at which the lethal
temperature for live crown foliage is reached, assumed to be 140 F (60
C). Scorch heights can be estimated during a prescribed fire operation,
based on observations of the flame length, and ignition can be adjusted
accordingly if desired scorch heights are not being achieved. SCORCH
can be run by linking it to outputs from DIRECT. A more detailed
description of this model can be found in Andrews and Chase (1989).

The SCORCH model has several limitations. It was developed for flat
ground, so should be used on slopes with care. The model may not be
valid outside the range of conditions for which it was developed, with
fireline intensities ranging from 19 to 363 Btu per foot per second,
equivalent to flame lengths of 1.8 to 6.8 feet (0.5 to 2.1 meters). Air
temperatures are 73 to 88 F (23 to 31 C), and midflame windspeeds are
1.5 to 3 miles per hour (3.4 to 4.8 kilometers per hour). Under these
conditions, scorch heights ranged from 6.5 to 56 feet (2 to 17 meters).
Also, this model considers the heat released by the flaming front of the
fire, not from the long-term burnout of large fuels. If significant amounts
of residual burnout of large diameter fuels is anticipated, expected or
observed flame length resulting from this burnout can be entered
directly, instead of using the flame length calculated by BEHAVE in
DIRECT.

b. MORTALITY. Tree mortality is predicted by the MORTALITY module

of BEHAVE, also located in the FIRE 1 program. The model was
developed by Reinhardt and Ryan (1988b), and its use within the
BEHAVE program is described in Andrews and Chase (1989).
MORTALITY predicts the percentage of tree mortality from estimates of
crown and bole damage for a specific species, as different species have
varying abilities to survive a set amount of damage. Data on mortality
was collected from the following species, which can be specified when
running the model: western larch, Douglas-fir, western hemlock,
Engelmann spruce, western red cedar, lodgepole pine, and subalpine fir.
One can select one of these species to represent a species not listed if
the bark thickness is similar. Inputs required to run this program are
scorch height, tree height, crown ratio, and bark thickness, which is
calculated from the species of tree and its diameter at breast height.
Percentage of crown volume scorched is calculated from scorch height,
tree height, and crown ratio. Bole damage is assumed to be proportional
to bark thickness. The output is given in percent mortality. A 30 percent
mortality means that 30 of 100 trees would be expected to die if
subjected to the same fire, or that there is a 30 percent probability that
any individual tree would die.

A linked run can be made from DIRECT to SCORCH to MORTALITY.

Flame length can be calculated for a range of windspeeds in DIRECT.
From these values, SCORCH will calculate a range of scorch height
values, which provide one of the inputs to MORTALITY.

The model is limited by the assumption that the fire is of an average

duration. Mortality may be under predicted if a fire of long duration
occurs, caused by consumption of extremely dry duff or large diameter
fuel. If fuel is very light or patchy, mortality may be over estimated.

2. First Order Fire Effects Model (FOFEM). The FOFEM program

computes duff and woody fuel consumption, mineral soil exposure, fire-
caused tree mortality, and smoke production for forest stands (Keane et
al. 1990). A current version of the program is being tested by field users.
Future versions will allow prediction of soil heating and successional
changes. Default input values are derived from fuel models provided for
natural and activity fuels by many forest cover types. For further
information, contact the Fire Effects Research Work Unit at the
Intermountain Fire Sciences Laboratory, Missoula, Montana.

3. CONSUME. CONSUME is a PC based software program that

predicts the amount of fuel consumption on logged units based on
weather data, the amount and moisture content of fuels, and other
factors that describe a burn unit (Ottmar et al. 1993). The program
allows a resource manager to determine when and where to conduct a
prescribed fire to achieve desired objectives, while reducing impacts on
other resources. CONSUME can be used for most broadcast and
understory burns on forest lands where the dead woody fuels are
relatively homogeneous. The program applies to western forests
dominated by Douglas-fir, western hemlock, red alder, lodgepole pine,
or mixed conifer species. Program disks and the users guide are
available through the Publications Management System at NIFC.

4. Fire Effects Information System. The Fire Effects Information

System (FEIS) is a computerized information storage and retrieval
system that contains detailed information about the effects of fire on
specific plants, plant communities, and wildlife species. Plant species
information, for example, is organized into sub-categories of distribution
and occurrence; value and use; botanical and ecological characteristics;
fire ecology; and fire effects. Descriptions of the results of fire effects
case studies are included if available. The FEIS is not a typical
bibliographic data base that lists citations with key words and abstracts.
The Fire Effects System provides information in a text format, providing
reviews of the key facts in the literature, summarized into appropriate
sub-categories. Numerical codes in the text refer to citations listed in a
references section included in each species write-up. Where conflicting
information about plant response has been found, interpretations are
made if differences in season, burning conditions, or ecotype, for
example, can explain why observed variations occurred. Information
about wildlife species and ecosystems is handled in a similar fashion.

The system was developed by the U.S. Forest Service Intermountain

Fire Sciences Laboratory in Missoula, Montana, and money for
development of the prototype data base was provided by the BLM. Many
plant species of ecosystems managed by the Bureau are presently in
the system. Interagency funding is being used to expand the database
to include species of the eastern U.S. and Alaska, as well as additional
species of the western U.S.

The Citation Retrieval System (CRS) is an associated program that

contains all of the references used in compiling species and plant
community information for the FEIS. The CRS can be searched for a
specific citation, author, or key word. It can prepare a bibliography from
a list of citation index numbers selected by the user from a species write-
up.

The FEIS and CRS are available on the Forest Service Intermountain
Region computer in Ogden, Utah. Access is through the Forest Service
Data General System, or by phone modem from any personal computer
with software that allows communication with a Data General computer.
There is no cost for PC users other than telephone line charges. Data
can be saved to a temporary file on the main frame computer, and sent
over phone lines to the user's computer. Department of Interior
employees can contact their national fire management offices to obtain
access information. U.S. Forest Service employees can obtain
assistance from their regional FEIS coordinator. States provide
information through their State FEIS coordinator.

E. Smoke Modeling.

1. SASEM. The Simple Approach Smoke Estimation Model (SASEM) is

a tool for the analysis of smoke dispersion from prescribed fires (Sestak
and Riebau 1988). It is a screening model, in that it uses simplified
assumptions and tends to over predict impacts, yielding conservative
results. If violations of air quality standards are not predicted by SASEM,
it is unlikely that they will occur. Inputs to the model include basic
descriptions of the fuels, such as type and loading, expected fireline
intensity, and expected burn duration. Windspeed and direction,
dispersion conditions, and average mixing height are considered, as well
as distance and direction of the fire from sensitive receptors. The model
calculates fuel consumption and particulate emission factors from fuel
loading and expected fireline intensity. Model outputs include maximum
particulate concentration and the distance from the fire at which it will
occur, ranges of distances from the fire at which any primary or
secondary particulate standards would be violated, and the reduction in
visual range at selected receptors. Outputs are given in tabular fashion
for a range of dispersion and windspeed conditions.

SASEM is extremely simple to use, and requires no data inputs that are
not normally acquired as part of the prescribed fire planning process.
The program is available on floppy disks for operation on IBM
compatible machines. For further information, contact agency air
resource or fire management specialists, or the air quality staff at the
Environmental Science and Technology Center, National Biological
Survey, Fort Collins, Colorado.

2. TSARS. The following discussion is taken from Sestak (et al. 1991).
The Tiered Smoke/Air Resources System, TSARS, allows fire
management field officers to test fire prescriptions for smoke
management problems. Models with a high degree of rigor can be used
to solve more complex problems, however, higher user proficiency is
required.

a. Existing components. Models currently in the system include

SASEM, explained in item E.1. above, EPM, and VALBOX.
(1) The Emission Production Model, EPM, is a more elaborate model of
heat and particulate production than SASEM. Originally designed for
forest fuels, particularly logging slash, data are presently being collected
to broaden the model to other fuel types, particularly rangeland fuels.

(2) VALBOX models an airshed in complex terrain. Air in a mountain

valley is divided into a series of connecting boxes, with dimensions
calculated from topographic map data. This model best describes
conditions when an inversion exists and air is stagnant.

b. Proposed components. BEHAVE, described in Section C. of this

chapter, when added to the TSARS system, will expedite the running of
EPM and SASEM because many of their input values are contained in
the standard fire behavior fuel models used in BEHAVE. Use of these
models together will allow consideration of smoke impacts when
developing prescribed fire prescriptions.

TAPAS, the Topographic Air Pollution Analysis System, is a set of

meteorological and pollution dispersion models suitable for use with
multiple emission sources in complex terrain. Two models contained in
TAPAS, WINDS and CITPUFF, will be specialized in the TSARS
program for use in prescribed fire planning. WINDS is a two dimensional
wind field model. A three dimensional wind model in TAPAS can be
used by if elevation grid databases are created, information that is
generally available in in Geographic Information Systems.

Wind fields created by the two or three dimensional models can be used
with a final proposed component of TSARS. CITPUFF approximates the
dispersion of a pollutant as it follows a path across a simulation area.
The emission information required by this model would be provided by
SASEM or EPM emission calculations.

It is intended that TSARS be available on the second generation IAMS

system, and also on IBM compatible personal computers. For more
information, contact air quality staff, Environmental Science and
Technology Center, National Biological Survey, Fort Collins, Colorado.

F. Library Services

The literature search services maintained by Federal Libraries are

valuable sources of computerized information helpful for managing fire
effects. The Bureau of Land Management Library at the Denver Federal
Center, for example, has the capability of making on-line searches of
several hundred data bases that include periodicals, books, reports, and
other publications. Some data bases provide abstracts along with full
citations. The user works with a Library staff person by defining the kind
of information desired, as well as specifying key words to be used in the
search. Bibliographies can be requested by subject area, author, report
number, or other category. The library can also provide copies of
articles, and loan books in their collection and through inter-library loan.
Any BLM employee can contact the Library at the BLM Service Center in
Denver for assistance. The phone number is 303/236-6646.

The Alaska Resources Library provides a full range of library services to

all Federal employees, particularly in the natural resources field. The
Library is located in the Federal Building in Anchorage, where it is
administered by the BLM. The phone number is 907/271-5025; FAX is
907/271-5965.

The Department of the Interior Natural Resources Library provides a full

range of library services to USDI employees in response to telephone
and written requests. They are located in the Main Interior Building,
Washington, D.C.; their phone number is 202/208-5815.

National Park Service employees can contact the NPS Service Center at
Harper's Ferry, Virginia or Denver for library assistance. The phone
number at Harper's Ferry is 304/535-6371; Denver is 303/969-2100.

U. S. Forest Service employees can obtain a full range of library

services through FS INFO, available to them on the Data General
System through the Information Center process. Both Forest Service
Research and National Forest System Employees can seek assistance
from the FS INFO center located in at the Forest and Range Experiment
Station in their geographic region.

G. Summary

Computer technology and applications are developing so quickly that

any list of software is incomplete as soon as it is published. Specialized
computer programs, called expert systems, may be available in the next
few years. Expert systems are being developed or planned that can
assist in the development of fire prescriptions to meet specific resource
objectives, and to achieve specific fire effects. Agency fire management
and air quality specialists can be contacted for information about future
computer software development.

This page was last modified 05/31/01

|Disclaimer| | Privacy| | Copyright| |USFWS Main Page| |Webmaster|

 National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 05/31/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home GLOSSARY OF TERMS

Preface
Objectives
Fire Behavior
-A-
Fuels
Air Quality absolute value: the absolute value represents the distance that a
Soils & Water positive or negative number is from zero, when numbers are arrayed on
Plants a line with negative numbers to the left of zero and positive numbers to
Wildlife the right of zero. The absolute value of a positive number is the number
Cultural Res. itself, whereas the absolute value of a negative number is the opposite
Grazing (positive) number (Batchelet 1976).
Mgmt.
Evaluation accelerated erosion: erosion much more rapid than normal, natural, or
Data Analysis geologic erosion, primarily as a result of the influence of the activities of
Computer humans, or, in some cases, of other animals or natural catastrophes that
Soft. expose bare surfaces, for example, fires.
Glossary
Bibliography
accuracy: the closeness of a measured or computed value to its true
Contributions
value. Accuracy cannot be determined from a sample, and usually
remains unknown. (See precision.)

active crown fire: a fire in which a solid flame develops in the crowns of
trees, but the surface and crown phases advance as a linked unit
dependent on each other.

active layer: soil layer that overlies permafrost that thaws every
summer (Viereck and Schandelmeier 1980).

activity fuels: fuels resulting from, or altered by, forestry practices such
as timber harvest or thinning, as opposed to naturally created fuels.
(See natural fuels.)

aeolian (eolian): movement of material, such as soil, through wind

action.

aerial fuels: the layer of fuels that is above the surface fuels, including
living tree and shrub crowns, mosses, lichens, vines, and dead branch
material.

age class: classes that define the ages of individuals in discrete units,
such as 0 to 5 years, or 6 to 20 years.

algal bloom: proliferation of living algae on the surface of lakes,

streams, or ponds that is stimulated by nutrient, especially nitrogen or
phosphate, enrichment.

alkaline soil: any soil with a pH value greater than 7.0. Often used
interchangeably with "basic soil."

allelopathy: inhibition by a plant of the germination, establishment, and

growth of seedlings.

ammonification: the biochemical process whereby ammoniacal

nitrogen is released from nitrogen containing organic compounds.

analysis of covariance (ACOVA): test that combines regression with

analysis of variance. In general, variation in y that is associated with x is
removed from the error variance, which results in more precise
estimates from more powerful tests. An example in fire management is
the case where plants of different sizes are measured before burning;
because plant size may affect postburn response, the effect of plant size
is "controlled" so that differences due to burning will be apparent.

analysis of variance (ANOVA or AOV): test used when statistical

inferences are to be made about more than two means. Test can
"control" one or more sources of variation, depending on model and
experimental design.

anion: negatively charged ion.

annual plant: a plant completing its life cycle in a year or less (Benson
1967).

Aridisol: soil with pedogenic horizons, low in organic matter, that are
never moist as long as three consecutive months. They also have one or
more of the following diagnostic horizons, including argillic, natric,
cambic, calcic, petrocalcic, gypsic, or salic, and may have a duripan.
Aridisols are approximately equal to Desert, Reddish Desert, Sierozem,
Solonchak, some Brown and Reddish Brown, and associated Solonetz
soils of the pre-1966 classification scheme.

arithmetic mean: the mean (or average) of sample data; it provides an

unbiased estimate of the parametric mean; designated by x. (NOTE:
There are other kinds of means, such as harmonic, geometric, and
parametric.)

arrangement: see fuel arrangement.

ASCII: American Standard Code for Information Interchange. A code for

transmitting information asynchronously on local and long distance
communication lines; representing a standard set of letters, numbers,
and control characters.

attributes: those variables that cannot be measured, but must be

expressed qualitatively. (See variable.)

available fuel: the portion of the total fuel on the site that would actually
burn under a given set of environmental conditions.

avoidance: a smoke emission control strategy that considers

meteorological conditions when scheduling prescribed fires in order to
avoid incursions into smoke sensitive areas (Mathews et al. 1985).

axil: the upper side of the point where a leaf meets a stem, or a branch
meets another branch or the main stem of a plant.

axillary bud: a bud in a leaf axil.

-B-

backing fire: a fire, or that part of a fire, spreading or set to spread into
the wind, or down a slope.

basal cover: the vertical projection of the root crown onto the ground.
(See cover and foliar cover.)
BEHAVE: a system of two interactive computer programs for modelling
fuel and fire behavior.

bias: systematic distortion that may be due to measurement error,

method of selecting the sample,

etc. An example is the use of an uncalibrated balance that consistently

overestimates mass by 10 grams, or neglecting to subtract tare weight
(weight of the container) from packaged samples.

biennial plant: a plant that completes its life cycle in two years. Biennial
plants usually produce only

basal leaves above ground the first year and both basal leaves and
flowering stems the second (Benson 1967).

broadcast burning: allowing a prescribed fire to burn over a designated

area within well-defined boundaries for reduction of fuel hazard, as a
resource management treatment, or both.

Btu: British thermal unit. The amount of heat needed to raise the
temperature of one pound of water (one pint) one degree Fahrenheit.

bud: a vegetative growing structure at the tip of a stem or branch with

the enclosing scale leaves or

immature leaves; a young flower bud that has not yet opened. A
vegetative bud may also be located along the surface of roots or
rhizomes, or buried in woody stem or root tissue (Benson 1967).

bud primordia: a cluster of plant cells with the physiological potential to

develop into a bud or actively growing shoot.

bulb: an underground bud covered by fleshy scales, the coating formed

from the bases of leaves (Benson 1967).

bulk density: weight per unit volume. For fuels, this is usually
expressed as pounds per cubic foot; for soils, grams per cubic
centimeter.
burl: a mass of woody tissue from which roots and stems originate, and
which is often covered with dormant buds (James 1984).

burn severity: a qualitative assessment of the heat pulse directed

toward the ground during a fire. Burn severity relates to soil heating,
large fuel and duff consumption, consumption of the litter and organic
layer beneath trees and isolated shrubs, and mortality of buried plant
parts.

-C-

cambium: a layer of dividing plant cells which add during each growing
season a layer of woody material (largely xylem) on their inner side
toward the center of the stem or root and a layer of bark (phloem and
associated tissues) on the outer side (Benson 1967).

carbohydrates: starches and sugars manufactured by a plant and used

to provide energy for metabolism, and structural compounds for growth
(Trlica 1977).

carrier fuels: the fuels that support the flaming front of the moving fire.

cation: positively charged ion. Common soil cations include calcium,

magnesium, sodium, potassium, and hydrogen.

caudex: a largely underground stem base which persists from year to

year and each season produces leaves and flowering stems (Benson
1967).

chain: unit of measure in land survey, equal to 66 feet (20 meters) (80
chains equal one mile). Commonly used to report fire perimeters and
other fireline distances, chains can be easily converted to acreage (e.g.,
10 square chains equal one acre).

Class 1 Area: geographic areas designated by the Clean Air Act where
only a very small amount or increment of air quality deterioration is
permissible. Class 1 Areas include specified National Parks, Wilderness
Areas, and certain Indian reservations.

clastic: being able to readily break into small fragments or pieces.

climax: the highest ecological development of a plant community
capable of perpetuation under the prevailing climatic and edaphic
conditions (Range Term Glossary Committee 1974).

clone: a group of individuals propagated vegetatively from a single

individual of seedling origin (Barnes 1966 in Jones 1985). Individuals in
a clone are genetically the same plant.

colonizer: species that establish on a burned (or otherwise denuded)

site from seed (Stickney 1986).

combustion: consumption of fuels by oxidation, evolving heat, flame,

and/or incandescence.

combustion efficiency: the relative amount of time a fire burns in the

flaming phase of combustion, as compared to smoldering combustion. A
ratio of the amount of fuel that is consumed in flaming combustion
compared to the amount of fuel consumed during the smoldering phase,
in which more of the fuel material is emitted as smoke particles because
it is not turned into carbon dioxide and water.

confidence level: the percentage confidence that a statement is true. If

data are normally distributed about an average value, the statement
'within 10 percent of the mean at the 80 percent confidence level' would
indicate that a sample is likely to be within 10 percent of the true
average value of the population 80 percent of the time.

continuity: see fuel continuity.

continuous variables: those variables which, at least theoretically, can

assume an infinite number of values between any two fixed points. An
example is fuel load, which theoretically can be anywhere between zero
and infinity. (See discrete variables.)

control strategy: a way of implementing a prescribed fire that manages

smoke output. (See avoidance, dilution, and emission reduction.)

convection column: the thermally induced ascending column of gases,

smoke, water vapor, and particulate matter produced by a fire.

cookbook approach: to follow a particular procedure without deviation.

coordinated resource management: a process that directly involves
everyone concerned with resource management in a given planning
area.

corm: a bulb-like structure formed by enlargement of the stem base,

sometimes coated with one or more membranous layers (Benson 1967).

cover: the area on the ground covered by the combined aerial parts of
plants expressed as a percent of the total area. (See basal cover and
foliar cover.)

criteria pollutants: those air pollutants designated by the

Environmental Protection Agency as poten- tially harmful and for which
ambient air standards have been set to protect the public health and
welfare. The criteria pollutants are carbon monoxide, particulate matter,
sulfur dioxide, nitrogen dioxide, lead, and ozone.

crown consumption: combustion of the twigs, and needles or leaves of

a tree during a fire.

crown fire: a fire that advances by moving among crowns of trees or

shrubs.

crown ratio: the ratio of live crown to tree height.

crown scorch: causing the death of tree foliage by heating it to lethal

temperature during a fire, although the foliage is not consumed by the
fire. Crown scorch may not be apparent for several weeks after the fire.

crown scorch height: the height above the surface of the ground to
which a tree canopy is scorched.

crowning potential: a probability that a crown fire may start, calculated

from inputs of foliage moisture content and height of the lowest part of
the tree crowns above the surface.

-D-

data: the facts collected during observations; used as a plural noun, i.e.,
"data are," or "data were." (See datum.)
datum: a single fact collected during observation; the singular of "data."
(See data.)

dead fuels: naturally occurring fuels without living tissue, in which the
moisture content is governed almost entirely by absorption or
evaporation of atmospheric moisture (relative humidity and
precipitation).

decreaser species: plant species of the original vegetation that

decrease in relative amount under overuse by grazing or browsing
animals. Commonly termed decreasers.

degrees of freedom: the quantity n or n-1, where n is the number of

observations (population or sample size) upon which a variance has
been based; degrees of freedom is usually designated by the
abbreviation "df."

density: the number of plants or parts of plants per unit area.

designated area: those areas identified as principal population centers

or other areas requiring protection under state or federal air quality laws
or regulations.

desired plant community: a plant community which produces the kind,

proportion, and amount of vegetation necessary for meeting or
exceeding the land use plan goals and activity plan objectives
established for the site.

diffusion: the net movement of gas molecules from areas with higher
concentration of that gas to areas with lower concentration.

dilution: a control strategy used in managing smoke from prescribed

fires in which smoke concentration is reduced by diluting it through a
greater volume of air, either by scheduling during good dispersion
conditions or burning at a slower rate (Mathews et al. 1985).

disclimax: (disturbance climax) a stable plant community which is not

the climatic or edaphic climax and which is perpetuated by man or his
domestic animals (Odum 1966).
discrete variables: those variables that have only certain fixed
numerical values, with no intermediate values possible in between; also
known as discontinuous or meristic variables. An example is the number
of offspring per litter, where only integers (whole numbers) are possible.
(See continuous variables.)

diversity: the relative degree of abundance of wildlife species, plant

species, communities, habitats, or habitat features per unit of area
(Thomas 1979).

duff: the partially decomposed organic material of the forest floor that
lies beneath the freshly fallen twigs, needles and leaves. The
fermentation and humus layers of the forest floor (Deeming et al. 1977).

-E-

ecological condition (range): the existing state of the vegetation on a

site compared to the natural potential (climax) plant community for that
site. This term is used interchangeably with "range condition" and
describes the deviation from the climax condition according to four
arbitrary condition classes. Not synonymous with "forage condition"
which does not relate to site potential.

ecological niche: the role or function a particular organism plays in the

environment (Hanson 1962).

ecological site: a distinctive geographic unit that differs from other

kinds of geographic units in its ability to produce a characteristic natural
plant community. An ecological site is the product of all the
environmental factors responsible for its development. It is capable of
supporting a native plant community typified by an association of
species that differs from that of other ecologic sites in the kind or portion
of species or in total production.

ecology: the study of the interrelationships of organisms with one

another and with the environment (Hanson 1962).

ecosystem: an interacting natural system including all the component

organisms together with the abiotic environment (Hanson 1962).

ecotone: the area influenced by the transition between plant

communities or between successional stages or vegetative conditions
within a plant community (Thomas 1979).

edge: the place where plant communities meet or where successional

stages or vegetative conditions within plant communities come together
(Thomas 1979).

edge effect: the increased richness of flora and fauna resulting from the
mixing of two communities where they join (Thomas 1979).

effective windspeed: a value that combines the speed of the wind with
the additive effect of slope when a fire is burning up or across a slope.

emission: a release into the outdoor atmosphere of air contaminants

such as smoke.

emission factor: the mass (weight) of particulate matter produced per

unit mass of fuel consumed (expressed as grams per kilogram or
pounds per ton).

emission inventory: a listing by source of the amounts of air pollutants

discharged into the atmosphere.

emission reduction: a strategy for controlling smoke from prescribed

fires that minimizes the amount of smoke output per unit area treated.

equilibrium moisture content: the moisture content that a fuel particle

will attain if exposed for an indefinite period in an environment of
specified constant temperature and humidity. When a fuel particle has
reached its EMC, the net exchange of moisture between it and its
environment is zero (Deeming et al. 1977).

eurytopic: having a wide range of suitable ecological conditions

(Pennak 1964).

eutrophication: the process whereby water becomes excessively rich

in nutrients and correspondingly deficient, at least seasonally, in oxygen.
Often accompanied or followed by algal blooms.

exfoliation: the separation of concentric layers of rock from the original

rock mass.
experimental design: the process of planning an experiment or
evaluation so that appropriate data will be collected, which may be
analyzed by statistical methods resulting in valid and objective
conclusions. Examples include Completely Random Design,
Randomized Block Design, Latin Square, Factorial Experiment, Split-
Plot Design, and Nested Design.

experimental error: a measure of the variation which exists among

observations on experimental units that are treated alike; "natural"
variation.

-F-

feeder roots: small diameter roots that collect most water and nutrients
for a plant, usually located near the soil surface.

fine fuels: small diameter fuels such as grass, leaves, draped pine
needles, and twigs, which when dry, ignite readily and are rapidly
consumed.

fire behavior: the manner in which a fire burns in response to the

variables of fuel, weather, and topography.

fire intensity: see fireline intensity.

Fire Behavior Prediction System: a system that uses a set of

mathematical equations to predict certain aspects of fire behavior in
wildland fuels when provided with data on fuel and environmental
conditions (Rothermel 1983).

fire regime: periodicity and pattern of naturally occurring fires in a

particular area or vegetative type, described in terms of frequency,
biological severity, and areal extent (Tande 1980).

fire severity: see burn severity.

fire spread model: a set of physics and empirical equations that form a
mathematical representation of the behavior of fire in uniform wildland
fuels (Rothermel 1972).
fire treatment: the use of prescribed fire to accomplish a specified
objectives.

fire whirl: a spinning, vortex column of ascending hot air and gases
rising from a fire and carrying aloft smoke, debris, and flame. Fire whirls
range from a foot or two in diameter to small tornadoes in size and
intensity. They may involve the entire fire area or only a hot spot within
the area.

fireline intensity: the heat released per unit of time for each unit length
of the leading fire edge. The primary unit is Btu per lineal foot of fire front
per second (Byram 1959 in Albini 1976).

firing pattern: see ignition pattern.

firing technique: see ignition pattern.

first order fire effects (FOFE): the direct and immediate effects of fire.

flame length: the average length of flames when the fire has reached
its full, forward rate of spread, measured along the slant of the flame
from the midpoint of its base to its tip.

flaming phase: the phase of combustion in which gases distilled from

fuels rapidly combine with atmospheric oxygen, producing visible
flames.

FLPMA: the Federal Land Policy and Management Act of 1976 (Public
Law 94-579, 90 Stat. 2743, 43 USC 1701).

foliar cover: the projection of all plant parts vertically onto the ground.
(See cover, and basal cover.)

forb: a plant with a soft, rather than permanent woody stem, that is not a
grass or grasslike plant.

forward rate of spread: the speed with which a fire moves in a

horizontal direction across the landscape, usually expressed in chains
per hour or feet per minute.

frequency of occurrence: a quantitative expression of the presence or

absence of individuals of a species in a population; the ratio between the
number of sample units that contain a species and the total number of
sample units.

fuel: combustible plant material, both living and dead that is capable of
burning in a wildland situation.

fuel arrangement: the spatial distribution and orientation of fuel

particles within a fuel bed.

fuel bed: an array of fuels usually constructed with specific loading,

depth, and particle size, to meet experimental requirements; also
commonly used to describe the fuel composition in natural settings.

fuel bed depth: average height of surface fuels contained in the

combustion zone of a spreading fire front.

fuel continuity: the degree or extent of continuous or uninterrupted

distribution of fuel particles in a fuel bed, a critical influence on a fire's
ability to sustain combustion and spread. This applies both to aerial fuels
and surface fuels.

fuel depth: see fuel bed depth.

fuel loading: the weight of fuels in a given area, usually expressed in

tons per acre, pounds per acre, or kilograms per square meter.

fuel model: a characterization of fuel properties of a typical field

situation. A fuel model contains a complete set of inputs for the fire
spread model.

fuel moisture content: the amount of water in a particle of fuel, usually

expressed as a percentage of the oven dry weight of the fuel particle.

fuel size class: a category used to describe the diameter of down dead
woody fuels. Fuels within the same size class are assumed to have
similar wetting and drying properties, and to preheat and ignite at similar
rates during the combustion process.

-G-
glowing phase: phase of combustion in which a solid surface of fuel is
in direct contact with oxygen, and oxidation occurs, usually
accompanied by incandescence, and little smoke production.

graminoid: grasslike plant, including grasses, sedges, rushes, reeds,

and cattails.

gravimetric: of, or pertaining to, measurement by weight.

grazing management (strategy): the manipulation of the grazing use

on an area in a particular pattern, to achieve specific objectives.

grazing pattern: dispersion of livestock grazing within a management

unit or area.

ground fire: fire that burns the organic material in the soil layer (e.g. a
"peat fire") and often also the surface litter and low-growing vegetation.

ground fuels: all combustible materials below the surface litter layer,
including duff, tree and shrub roots, punky wood, dead lower moss and
lichen layers, and sawdust, that normally support glowing combustion
without flame.

growth stage: the relative ages of individuals of a species, usually

expressed in categories such as seedlings, juvenile, mature, and
decadent.

guild: see habitat guild.

-H-

habitat: the sum total of environmental conditions of a specific place

occupied by a wildlife species or a population of such species (Thomas
1979).

habitat guild: a group of species having similar ecological requirements

and/or foraging strategies and therefore having similar roles in the
community (Cooperrider et al. 1986).

heading fire: a fire front spreading, or ignited to spread with the wind,
up a slope, or influenced by a combination of wind and slope.
heat content: the net amount of heat that would be given off if fuel
burns when it is absolutely dry, noted as Btu per pound of fuel.

heat per unit area: total amount of heat released per unit area as the
flaming front of the fire passes, expressed as Btu/square foot; a
measure of the total amount of heat released in flames.

heavy fuels: dead fuels of large diameter (3.0 inches or larger) such as
logs and large branchwood.

height: the vertical measurement of vegetation from the top of the

crown to ground level.

herbivorous: feeding on plants; phytophagous (Cooperrider et al.

1986).

humidity: see relative humidity.

hydrophobicity: resistance to wetting exhibited by some soils, also

called water repellency. The phenomena may occur naturally or may be
fire-induced. It may be determined by water drop penetration time,
equilibrium liquid-contact angles, solid-air surface tension indices, or the
characterization of dynamic wetting angles during infiltration.

-I-

ignition method: the means by which a prescribed fire is ignited, such

as hand-held drip torch, heli- torch, and backpack propane tanks.

ignition pattern: the configuration and sequence in which a prescribed

fire is ignited. Patterns include, for example, spot fire, strip-head fire, and
ring fire (same as ignition technique).

ignition technique: see ignition pattern.

illuviation: soil development process by which materials are

translocated from an upper soil horizon and immobilized in a soil horizon
at a lower level in the soil profile.

imbibe: to absorb liquid or moisture.

Inceptisol: soil that is usually moist and has pedogenic horizons of
alteration of parent material but not of illuviation. Generally, the direction
of development is not evident, or is too weak to classify in another soil
order. Inceptisols are approximately equal to Ando, Sol Brun Acide,
some Brown Forest, Low-Humic Gley, and Humic Gley soils of the pre-
1966 classification scheme.

increaser species: plant species of the original vegetation that increase

in relative amount, at least for a time, under overuse. Commonly termed
increasers.

independent crown fire: a fire that advances in the tree crowns alone,
not requiring any energy from the surface fire to sustain combustion or
movement.

intensity: see fireline intensity.

interspersion: the intermixing of plant species and plant communities

that provide for animals in a defined area (Thomas 1979).

introduced plant species: a species not a part of the original fauna or

flora of an area.

invader species: plant species that were absent in undisturbed portions

of the original vegetation and will invade under disturbance or continued
overuse. Commonly termed invaders.

-J-

juxtaposition: the arrangement of stands of vegetation in space

(Thomas 1979).

-K-

Kcal: a kilogram-calorie is the amount of heat needed to raise the

temperature of one kilogram of water (1 liter) by 1 degree Celsius.

key forage species: forage species of particular importance in the plant

community or which are important because of their value as indicators of
change in the community.
 -L-

ladder fuels: fuels that can carry a fire from the surface fuel layer into
the aerial fuel layer, such as a standing dead tree with branches that
extend along its entire length.

leach: removal of soluble constituents from ashes or soil by percolation

of water.

life-form: a group of wildlife species whose requirements for habitat are

satisfied by similar successional stages within given plant communities
(Thomas 1979).

lignotuber: a mass of woody tissue from which roots and stems

originate, which often covered with dormant buds (James 1984); same
as root crown.

litter: the top layer of forest floor, typically composed of loose debris
such as branches, twigs, and recently fallen leaves or needles; little
altered in structure by decomposition. The L layer of the forest floor
(Deeming et al. 1977). Also loose accumulations of debris fallen from
shrubs, or dead parts of grass plants laying on the surface of the
ground.

live fuel moisture content: ratio of the amount of water to the amount
of dry plant material in living plants.

live fuels: living plants, such as trees, grasses, and shrubs, in which the
seasonal moisture content cycle is controlled largely by internal
physiological mechanisms, rather than by external weather influences.

live herbaceous moisture content: ratio of the amount of water to the

amount of dry plant material in herbaceous plants, i.e., grasses and
forbs.

live woody moisture content: ratio of the amount of water to the

amount of dry plant material in shrubs.

-M-
mast: the fruit of trees suitable as food for livestock and wildlife (Ford-
Robertson 1971).

mean: see arithmetic mean.

meristem: growing points of grasses, from which leaf blade elongation

occurs during active growing periods.

mesotopic: having an intermediate range of suitable ecological

conditions.

mho: meter/kilogram/second unit of electrical conductance, equal to the

conductance of a conductor in which a potential difference of one volt
maintains a current of one ampere.

midflame windspeed: the speed of the wind measured at the midpoint

of the flames, considered to be most representative of the speed of the
wind that is affecting fire behavior.

millimho: a unit of electrical conductance, equal to 0.001 mho.

moisture content: see fuel moisture content.

moisture of extinction: the moisture content of a specific fuel type

above which a fire will not propagate itself, and a firebrand will not ignite
a spreading fire.

Mollisol: soil with nearly black, organic-rich surface horizons and high
supplies of bases; they may accumulate large amounts of organic matter
in the presence of calcium. They have mollic epipedons and base
saturation greater than 50% in any cambic or argillic horizon and are
approximately equal to Chestnut, Chernozem, Brunizem, Rendzina,
some Brown, Brown Forest, and associated Solonetz and Groundwater
Podzols of the pre-1966 classification scheme.

mosaic: the intermingling of plant communities and their successional

stages in such a manner as to give the impression of an interwoven
design (Ford-Robertson 1971).

muck: a highly decomposed layer of organic material in an organic soil

(Buckman and Brady 1966).
mycorrhiza (pl. mycorrhizae): a mutually beneficial (symbiotic)
association between a plant root and a fungus, that enhances the ability
of the root to absorb water and nutrients.

-N-

National Fire Danger Rating System: a multiple index scheme

designed to provide fire and land management personnel with a
systematic means of assessing various aspects of fire danger on a day-
to-day basis.

native species: a species which is a part of the original fauna or flora of

the area in question.

natural fuels: fuels resulting from natural processes and not directly
generated or altered by land management practices. (See activity fuels.)

NFDRS: see National Fire Danger Rating System.

niche: (habitat niche) the peculiar arrangement of food, cover, and

water that meets the requirements of a particular species.

NIFMID: National Interagency Fire Management Integrated Database.

nonparametric tests: statistical testing techniques that are not

dependent on a given distribution. (See parametric tests.)

normal distribution: a continuous frequency distribution whose graphic

representation is a bell-shaped curve that is symmetrical about the
mean, which by definition has a mean of 0 and a variance of 1. Many
other types of distributions exist that have different shaped curves, such
as hypergeometric, Poisson, and binomial.

number: the total population of a species or classification category in a

delineated unit, a measure of its abundance.

nutrient: elements or compounds that are essential as raw materials for

organism growth and develop ment, such as carbon, oxygen, nitrogen,
and phosphorus. There are at least 17 essential nutrients.
 -O-

obligate: restricted to a particular condition of life, as an obligate

seeder, a plant that can only reproduce by seed.

off-site colonizers: plants that germinate and establish after a

disturbance from seed that was carried from off of the site (Stickney
1986).

onsite colonizers: plants that germinate and establish after a

disturbance from seed that was present on the site at the time of the
disturbance (Stickney 1986).

one-hour timelag fuels: dead fuels consisting of dead herbaceous

plant material and roundwood less than 0.25 inches (0.64 cm) in
diameter, expected to reach 63 percent of equilibrium moisture content
in one hour or less.

one-hundred hour timelag fuels: dead fuels consisting of roundwood

in the size range from 1.0 to 3.0 inches (2.5 to 7.6 cm) in diameter,
estimated to reach 63 percent of equilibrium moisture content in one
hundred hours.

one-thousand hour timelag fuels: dead fuels consisting of roundwood

3.0 to 8.0 inches (7.6 to 20.3 cm) in diameter, estimated to reach 63
percent of equilibrium moisture content in one thousand hours.

organic matter: that fraction of the soil that includes plant and animal
residues at various stages of decomposition, cells and tissues of soil
organisms, and substances synthesized by the soil population.

organic soil: a soil with a percentage content of organic matter greater

than about 20 to 25 percent (Buckman and Brady 1966).

-P-

packing ratio: the percentage of a fuel bed that is composed of fuel

particles, the remainder being air space among the individual particles
(Burgan and Rothermel 1984); the fuel volume divided by fuel bed
volume.
palatability: the relish that an animal shows for a particular species,
plant or plant part; how agreeable the plant is to the taste.

parameter: a variable which can be measured quantitatively;

sometimes, an arbitrary constant; associated with populations. One of
the unknown values that determine a model. (See statistic.)

parametric tests: statistical tests that are based on normal distributions.

(See nonparametric tests).

particle size: the size of a piece of fuel, often expressed in terms of size
classes.

particulate matter: any liquid or solid particles present in the

atmosphere. Particulate matter diameter is measured in microns.

passive crown fire: a fire in the crowns of trees in which trees or

groups of trees torch, ignited by the passing front of the fire. The
torching trees reinforce the spread rate, but these fires are not basically
different from surface fires.

peat: a deposit of slightly or non-decayed organic matter (Buckman and

Brady 1966).

percolation: passage of liquid through a porous body, as movement of

water through soil.

perennial plant: a plant that continues to grown year after year (Benson
1967). (See annual plant and biennial plant.)

permafrost: a short term for "permanently frozen ground"; any part of

the earth's crust, bedrock, or soil mantle that remains below 32 F (0 C)
continuously for a number of years (Brown 1970 in Viereck and
Schandelmeier 1980). (See active layer.)

petroglyph: a type of rock art in which a design is pecked into stone.

pH: the negative logarithm (base =10) of the hydronium ion

concentration, in moles per liter. It is a numerical measure of acidity or
alkalinity on a scale of 1 to 14, with the value of 7.0 being neutral.
phenology: the relationship of the seasonal sequence of climatic factors
with the timing of growth and reproductive phases in vegetation, such as
initiation of seasonal growth, time of blooming, time of seed set, and
development of new terminal buds (Daubenmire 1968b).

phreatophyte: a plant that derives its water from subsurfaces, typically

having roots that reach the water table, and is therefore somewhat
independent of precipitation. Obligate phreatophytes require this
situation, whereas facultative phreatophytes merely take advantage it.

pictograph: a type of rock art in which a design is painted onto stone.

pile burning: burning of logging slash that has been arranged into
individual piles. (See broadcast burning.)

PM10: particles with an aerodynamic diameter smaller than or equal to a

nominal ten micrometers.

PM2.5: particles with an aerodynamic diameter smaller than or equal to a

nominal 2.5 micrometers.

population: all possible values of a variable; the entire group that is

examined. (See sample.)

precision: the closeness of repeated measurements of the same

quantity. (See accuracy.)

preignition phase: preliminary phase of combustion in which fuel

elements ahead of the fire are heated, causing fuels to dry. Heat
induces decomposition of some components of the wood, causing
release of combustible organic gases and vapors.

prescribed burning: controlled application of fire to wildland fuels in

either their natural or modified state, under specified environmental
conditions that allows the fire to be confined to a predetermined area,
and produce the fire behavior and fire characteristics required to attain
planned fire treatment and resource management objectives.

prescribed fire: an intentionally or naturally ignited fire that burns under

specified conditions that allow the fire to be confined to a predetermined
area and produce the fire behavior and fire characteristics required to
attain planned fire treatment and resource management objectives.

prescription: a written statement defining the objectives to be attained

as well as the conditions of temperature, humidity, wind direction and
speed, fuel moisture, and soil moisture, under which a fire will be
allowed to burn. A prescription is generally expressed as acceptable
ranges of the prescription elements, and the limit of the geographic area
to be covered.

prevention of significant deterioration: a provision of the Clean Air

Act with the basic intent to limit degradation of air quality, particularly in
those areas of the country where the air quality is much better than
standards specified in the Law.

probability: a measurement that denotes the likelihood that an event

occurred simply by chance.

probability of ignition: the chance that a firebrand will cause an

ignition when it lands on receptive fuels.

productivity: weight of dry matter produced in a given period by all the

green plants growing in a given space (Daubenmire 1968b).

PSD: see prevention of significant deterioration.

pyrolysis: the thermal or chemical decomposition of fuel at an elevated

temperature. This is the pre- ignition phase of combustion during which
heat energy is absorbed by the fuel that, in turn, gives off flammable
tars, pitches, and gases.

-R-

ramet: an individual member of a clone. For example, every individual

stem in an aspen clone is a ramet.

random: the assignment of treatments to experimental units, or the

selection of samples, such that all units or samples have an equal
chance of receiving the treatment being estimated. It serves to assure
unbiased estimates of treatment means and experimental error.

rate of spread: see forward rate of spread.

RAWS: see Remote Automated Weather Station.

reaction intensity: the rate of heat release, per unit area of the fire
front, expressed as heat energy/area/time, such as Btu/square
foot/minute, or Kcal/square meter/second.

regeneration: see vegetative regeneration.

regional haze: atmospheric haze over a large area with no attributable

source.

relative frequency: see frequency of occurrence.

relative humidity: the ratio, in percent, of the amount of moisture in a

volume of air to the total amount which that volume can hold at the given
temperature and atmospheric pressure. Relative humidity is a function of
the actual moisture content of the air, the temperature, and the
atmospheric pressure (Schroeder and Buck 1970).

Remote Automatic Weather Station (RAWS): a self contained

meteorological platform that automatically acquires, processes, and
stores local weather data for subsequent transmission through a satellite
to an earth receiving station.

reproduction: see vegetative reproduction.

residence time: the time required for the flaming zone of the moving
front of a fire to pass a stationary point; the total length of time that the
flaming front of the fire occupies one point.

residual colonizers: plants that germinate after a disturbance from

seed that was present on the site (Stickney 1986).

respiration: oxidation of food in living cells, with the resulting release of

energy; part of the energy is transferred to other compounds and some
is used in the activation of certain cell processes (Meyer et al. 1973).

rhizome: a horizontal plant stem, growing beneath the surface, and

usually covered with dormant buds.
root crown: a mass of woody tissue from which roots and stems
originate, and which are often covered with dormant buds (James 1984);
same as lignotuber.

running crown fire: a fire moving in the crowns of trees, dependent

upon, or independent from the surface fire.

-S-

sample: part of a population; that portion of the population that is

measured.

sample size: the number of items or observations in a sample; usually

denoted by lower case letter n.

SASEM: Simple Approach Smoke Estimation Model (SASEM), a

computer model for the analysis of smoke dispersion from prescribed
fires. It is a screening model, in that it uses simplified assumptions and
tends to over predict impacts, yielding conservative results.

second order fire effects (SOFE): the indirect effects of fire treatment
that occur over the longer term.

seedbank: the supply of viable seeds present on a site. Seeds include

those recently dispersed by plants, long-lived seeds buried in organic
and soil layers, or those stored in cones in a tree canopy.

semi-serotinous: cones of coniferous trees that open and release their

seeds over a period of years (Zasada 1986).

senescence: period of declining productivity after the period of most

active growth, referred to both in terms of the seasonal life cycle of a
plant, and the total life of a perennial plant.

seral: pertaining to a succession of plant communities in a given habitat

leading to a particular climax association; a stage in a community
succession (Cooperrider et al. 1986).

sere: the stages that follow one another in an ecological succession

(Hanson 1962).
serotiny: storage of coniferous seeds in closed cones in the canopy of
the tree. Serotinous cones of lodgepole pine do not open until subjected
to temperatures of 45 to 50 C (113 to 122 F), causing the melting of the
resin bond that seals the cone scales.

severity: see burn severity.

short-life species: a plant that grows several years before being

replaced by a species more adapted to the changing site conditions.

simulation: a realistic visual portrayal which demonstrates the

perceivable changes in landscape features caused by a proposed
management activity. This is done through the use of photography, art
work, computer graphics and other such techniques.

sinter: clustering of clay particles that occurs when pottery is fired.

SIP: see State Implementation Plan.

slash: concentrations of wildland fuels resulting from human activities

such as logging, thinning, and road construction, and natural events
such as wind. Slash is composed of branches, bark, tops, cull logs,
uprooted stumps, and broken or uprooted trees.

smoldering phase: a phase of combustion that can occur after flames

die down because the reaction rate of the fire is not high enough to
maintain a persistent flame envelope. During the smoldering phase,
gases condense because of the cooler temperatures, and much more
smoke is produced than during flaming combustion.

soil structure: the combination or arrangement of primary soil particles,

units, or peds. Examples include platy, prismatic, columnar, blocky,
angular blocky, subangular granular, and crumb.

soil texture: the relative proportions of the various soil separates,

primarily sand, clay, and silt. Subdivisions of the three basic separates,
such as very fine sand, are often used.

spall: disintegration of a rock by breaking away of an outer layer.

species composition: a term relating the relative abundance of one

plant species to another using a common measurement; the proportion
(percentage) of various species in relation to the total on a given area.

species richness: a measurement or expression of the number of

species of plants or animals present in an area; the more species
present, the higher the degree of species richness (Thomas 1979).

spot fire: fire caused by flying sparks or embers outside the perimeter
of the main fire.

spot forecast: a customized prediction of atmospheric conditions at a

specific site that is issued by the National Weather Service, usually
requested in connection with a wildfire incident or a prescribed fire.

spot weather forecast: see spot forecast.

spotting: production of burning embers in the moving fire front that are
carried a short distance ahead of the fire, or in some cases are lofted by
convective action or carried by fire whirls some distance ahead.

standard deviation: a measure of the variation, or spread, of individual

measurements; a measurement which indicates how far away from the
middle the statistics are; usually denoted by the lower case s for sample
data; mathematically equal to the square root of variance.

State Implementation Plan: a plan that describes how a State intends

to achieve Federal and State standards relative to the Clean Air Act,
usually containing State regulations related to maintenance of air quality.

statistic: the number that results from manipulating raw data according
to a specified procedure; associated with samples. (See parameter.)

statistics: the scientific study of numerical data based on natural

phenomena.

stenotopic: having a narrow range of suitable ecological conditions

(Pennak 1964).

stochastic: of, or pertaining to, randomness.

stolon: a branch of a plant which grows along the surface of the ground
and produces plants and roots at intervals.

structure (vegetative): the form or appearance of a stand; the

arrangement of the canopy; the volume of vegetation in tiers or layers
(Thomas 1979).

succession: the process of vegetational development whereby an area

becomes successively occupied by different plant communities of higher
ecological order (Range Term Glossary Committee 1974).

successional change: see succession.

sum: the amount obtained by adding numbers or quantities; total;

usually denoted by an upper case Greek sigma, .

surface area to volume ratio: the ratio between the surface area of an
object, such as a fuel particle, to its volume. The smaller the particle, the
more quickly it can become wet, dry out, or become heated to
combustion temperature during a fire.

surface fire: fire that burns surface litter, dead woody fuels, other loose
debris on the forest floor, and some small vegetation.

surface fuels: fuels that contact the surface of the ground, consisting of
leaf and needle litter, dead branch material, downed logs, bark, tree
cones, and low stature living plants.

survivors: plant species with established plants on the site that can
vegetatively regenerate after the fire (Stickney 1986).

-T-

TAC: total available carbohydrates.

ten-hour timelag fuels: dead fuels consisting of roundwood 0.25 to 1.0

inches (0.6 to 2.5 cm) in diameter, estimated to reach 63 percent of
equilibrium moisture content in ten hours.

thermoluminescence: a property of fired materials, such as ceramics,

causing them to become luminous when gently heated again. Because
this property decays at a known rate, the age of a ceramic artifact can
be estimated by heating it and measuring the amount of
phosphorescence.

tiller: new growth in a graminoid that originates from dormant axillary

buds in the plant crown or on rhizomes (Dahl and Hyder 1977).

tillering: process of producing new grass growth from dormant axillary

buds in the plant crown or on rhizomes (Dahl and Hyder 1977).

timelag: the time necessary for a fuel particle to lose or gain

approximately 63 percent of the difference between its initial moisture
content and its equilibrium moisture content.

torch: ignition and subsequent envelopment in flames, usually from

bottom to top, of a tree or small group of trees.

total available carbohydrates (TAC): carbohydrates that are in a form

that can be utilized as a readily available source of energy by a plant,
including sugars, starch, dextrins, and fructosans (Smith et al. 1964 in
Trlica 1977).

total fuel: all plant material, both living and dead, on a site.

trachea: in air breathing vertebrates, the tube that serves as the

principal passage for conveying air to the lungs.

treatment: a procedure whose effect can be measured and compared

with the effect of other procedures. Examples include a fall burned
prescribed fire, an unburned "control", or an area burned with a specific
ignition method or pattern.

-U-

underburning: prescribed burning in activity-created or natural fuels

beneath a forest canopy, usually with the objective of preserving the
dominant overstory trees.

utilization rates (limits): the proportion of the current year's forage

production that is removed by grazing or browsing animals. It may refer
to particular species or to the entire plant community and is usually
expressed as a percentage.
 -V-

vapor pressure: the contribution to total atmospheric pressure due to

the presence of water molecules in the air (Schroeder and Buck 1970).

variable: any changing characteristic; in statistics, a measurable

characteristic of an experimental unit.

variance: the sum of the squares of the deviates divided by one less
than the total number of deviates; a measure which indicates how far
away from the middle the statistics are; usually denoted by the lower
case s2. Variance is the standard deviation squared. In practice, it is
easier to compute the variance, then take the square root to obtain the
standard deviation. (See standard deviation.)

vegetative regeneration: development of new aboveground plants from

surviving plant parts, such as by sprouting from a root crown or
rhizomes. Even if plants form their own root system, they are still
genetically the same as the parent plant (Zasada 1989).

vegetative reproduction: establishment of a new plant from a seed that

is a genetically distinct individual (Zasada 1989).

vigor: a subjective assessment of the health of individual plants in

similar site and growing conditions; or a more specific measure based
upon a specific facet of growth, such as seed stalk or tiller production
per plant or per unit area.

-W-

weight: as used in vegetation inventory and monitoring, the total

biomass of living plants growing above the ground in a given area at a
given time.

wildfire: a free burning and unwanted wildland fire requiring a

suppression action.

This page was last modified 05/31/01

|Disclaimer| | Privacy| | Copyright| |USFWS Main Page| |Webmaster|
 National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 05/31/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home BIBLIOGRAPHY
Preface
Objectives Agee, J. K. 1993. Sitka spruce, coast redwood, and western hemlock
Fire Behavior forests, p. 187-225. IN Fire ecology of Pacific Northwest Forests. Island
Fuels
Press. Washington, DC.
Air Quality
Soils & Water
Albini, Frank A. 1976. Estimating wildfire behavior and effects. USDA,
Plants
Wildlife
For. Serv. Gen. Tech. Rep. INT-30. Intermt. For. and Range Exp. Sta.,
Cultural Res.
Ogden, UT. 92 p.
Grazing Mgmt.
Evaluation Allen, Arthur W. 1987. The relationship between habitat and
Data Analysis furbearers, p. 164-179. IN Milan Novak, James A. Baker, Martyn E.
Computer Obbard, and Bruce Malloch (eds.). Wild furbearer management and
Soft. conservation in North America. Ontario Trappers Assoc., Ontario,
Glossary Canada.
Bibliography
Contributions American Micro Products, Inc. 1984. AMPI statistics library for the
Hewlett-Packard HP-71B. AMPI, U.S.A. 217 p.

Anderson, Bruce A. 1985. Archaeological consideration for park and

wilderness fire management planning, p. 145-148. IN James E. Lotan,
Bruce M. Kilgore, William C. Fischer, and Robert W. Mutch (tech.
coord.). Proc. - Symp. and workshop on wilderness fire. USDA, For.
Serv. Gen. Tech. Rep. INT-182. Intermt. For. and Range Exp. Sta.,
Ogden, UT.

Anderson, Hal E. 1978. Graphic aids for field calculation of dead, down
forest fuels. USDA, For. Serv. Gen. Tech. Rep. INT-45. Intermt. For.
and Range Exp. Sta., Ogden, UT. 19 p.

Anderson, Hal E. 1982. Aids to determining fuel models for estimating

fire behavior. USDA, For. Serv. Gen. Tech. Rep. INT-122. Intermt. For.
and Range Exp. Sta., Ogden, UT. 22 p.

Anderson, Hal. 1983. Burnout of large-sized woody fuels, p. 164-169.

IN Proc. 17th Asilomar conf. on fire and blast effects of nuclear
weapons. Lawrence Livermore Laboratory. Livermore, CA.

Anderson, Hal E. 1990. Moisture diffusivity and response time in fine

forest fuels. Can. J. For. Res. 20: 315-325.

Anderson, Loren. 1983. Personal observation. Wildlife biologist.

Salmon District. USDI, Bur. Land Manage. Salmon, ID.

Andrews, Patricia L. 1986. BEHAVE: Fire behavior prediction and fuel

modeling system - BURN subsystem, part 1. USDA, For. Serv. Gen.
Tech. Rep. INT-194. Intermt. Res. Sta., Ogden, UT. 130 p.

Andrews, Patricia L. and Larry S. Bradshaw. 1990. RXWINDOW: A

program for defining windows of acceptable burning conditions for
prescribed fire based on desired fire behavior. USDA, For. Serv. Gen.
Tech. Rep. INT-273. Intermt. Res. Sta., Ogden, UT. 54 p.

Andrews, Patricia L. and Carolyn H. Chase. 1989. BEHAVE: Fire

behavior prediction and fuel modeling system - BURN subsystem, part
2. USDA, For. Serv. Gen. Tech. Rep. INT-260. Intermt. Res. Sta.,
Ogden, UT. 93 p.

Armistead, John C. 1981. The Bandelier blowup. Amer. For. 87(10):29-

62.

Arno, Stephen F. 1985. Ecological effects and management

implications of Indian fires, p. 81-86. IN James E. Lotan, Bruce M.
Kilgore, William C. Fischer, and Robert W. Mutch (tech. coord.). Proc. -
Symp. and workshop on wilderness fire. USDA, For. Serv. Gen. Tech.
Rep. INT-182. Intermt. For. and Range Exp. Sta., Ogden, UT.

Autenrieth, Robert, William Molini, and Clait Braun. 1982. Sage

grouse management practices. Western States Sage Grouse Commit.
Tech. Bull. 1. Twin Falls, ID. 42 p.

Barker, W. G. and W. B. Collins. 1963. The blueberry rhizome: In vitro

culture. Can. J. Bot. 41:1325- 1329.
Barney, Milo A. and Neil C. Frischknecht. 1974. Vegetation changes
following fire in the pinyon- juniper type of west-central Utah. J. Range
Manage. 27(2):91-96.

Barrett, Stephen W. and Stephen F. Arno. 1982. Indian fires as an

ecological influence in the Northern Rockies. J. Forest. 80:647-651.

Barro, Susan C. and Susan G. Conard. 1987. Use of ryegrass

seeding as an emergency revegetation measure in chaparral
ecosystems. USDA, For. Serv. Gen. Tech. Rep. PSW-102. Pacif.
Southw. For. and Range Exp. Sta., Berkeley, CA. 12 p.

Bartels, Ronald, John D. Dell, Richard L. Knight, and Gail Schaefer.

1985. Dead and down woody material, p. 171-186. IN E. Reade Brown
(ed.). Management of wildlife and fish habitats in forests of western
Oregon and Washington. Part 1 - Chapter narratives. USDA, For. Serv.
Publ. No. R6-F&WL-192-1985. Pacif. Northw. Reg., Portland, OR.

Batschelet, Edward. 1976. Introduction to mathematics for life

scientists. Second edition. Springer- Verlag. New York. 643 p.

Beck, A. M. and Richard J. Vogl. 1972. The effects of spring burning

on rodent populations in a brush prairie savannah. Mammalogy 53:336-
346.

Bendell, J. F. 1974. Effects of fire on birds and mammals, p. 73-138. IN

T. T. Kozlowski, and C. E. Ahlgren (eds.). Fire and ecosystems.
Academic Press, New York.

Benson, Lyman. 1967. Plant classification. D.C. Heath and Company,

Boston, MA. 688 p.

Bilderback, David. 1974. Personal conversation with Melanie Miller.

Black, C. A. (ed.). 1965. Methods of soil analysis. Part 1: Physical and

mineralogical properties, including statistics of measurement and
sampling, 770 p., and Part 2: Chemical and microbiological properties,
1572 p. Amer. Soc. Agron., Madison, WI.

Blackmarr, W. H. and William B. Flanner. 1968. Final Report.

Seasonal variation in moisture content of some common shrubs of the
eastern North Carolina organic soils area. USDA, For. Serv. Rev. Draft
FS-SE-2106-1-2. Southern For. Fire Lab. Southeast. For. Exp. Sta. 15
p. plus illus.

Blaisdell, J. P. and Walter F. Mueggler. 1956. Sprouting of bitterbrush

following burning. Ecology 37(2):365-370.

Blaisdell, J. P., R. B. Murray, and E. D. McArthur. 1982. Managing

Intermountain rangelands - Sagebrush-grass ranges. USDA, For. Serv.
Gen. Tech. Rep. INT-134. Intermt. For. and Range Exp. Sta., Ogden,
UT. 41 p.

Blaney, D.G. and G.E. Warrington. 1983. Estimating soil erosion using
an erosion bridge. WSDG-TP- 00008. USDA, For. Serv. Watershed
Develop. Group, Ft. Collins, CO.

Block, William M., Leonard A. Brennan, and R.J. Gutierrez. 1987.

Evaluation of guild-indicator species for use in resource management.
Environ. Manage. 11:265-269.

Blonski, Kenneth S. and John L. Schramel. 1981. Photo series for

quantifying natural forest residues: Southern Cascades, Northern Sierra
Nevada. USDA, For. Serv. Gen. Tech. Rep. PSW-56. Pacif. Southw.
For. and Range Exp. Sta., Berkeley, CA. 145 p.

Bower, C. A. and L. V. Wilcox. 1965. Soluble salts, p. 933-951. IN C.

A. Black (ed.). Methods of soil analysis. Part 2: Chemical and
microbiological properties. Amer. Soc. Agron., Madison, WI.

Brackebusch, Arthur P. 1975. Gain and loss of moisture in large forest

fuels. USDA, For. Serv. Res. Pap. INT-173. Intermt. For. and Range
Exp. Sta., Ogden, UT. 50 p.

Bradshaw, Larry S. and Patricia L. Andrews. 1990. FCFAST: Fort

Collins fire access software. Fire Management Notes 51(4):26-27.

Bradshaw, Larry S. and William Fischer. 1981a. A computer system

for scheduling fire use. Part 1: The system. USDA, For. Serv. Gen.
Tech. Rep. INT-91. Intermt. For. and Range Exp. Sta., Ogden, UT. 63
p.
Bradshaw, Larry S. and William Fischer. 1981b. A computer system
for scheduling fire use. Part II: Computer terminal operator's manual.
USDA, For. Serv. Gen. Tech. Rep. INT-100. Intermt. For. and Range
Exp. Sta., Ogden, UT. 34 p.

Bradshaw, Larry S. and William Fischer. 1984. Computer programs

for summarizing climatic data stored in the National Fire Weather Data
Library. USDA, For. Serv. Gen. Tech. Rep. INT-164. Intermt. For. and
Range Exp. Sta., Ogden, UT. 39 p.

Brand, David G. 1986. Competition induced changes in developmental

features of planted Douglas-fir in southwestern British Columbia. Can.
J. For. Res. 16:191-196.

Branson, F. A., G. F. Gifford, K. G. Renard, and R. F. Hadley. 1981.

Rangeland hydrology. 2nd ed. Kendall/Hunt Pub. Co., Dubuque, IA. 340
p.

Britton, C. M. 1984. Personal conversation with Loren Anderson.

Britton, C. M., B. L. Karr, and F.A. Sneva. 1977. A technique for

measuring rate of fire spread. J. Range Manage. 30(5):395-397.

Britton, Carlton M. and M. H. Ralphs. 1979. Use of fire as a

management tool in sagebrush ecosystems, p. 101-109. IN The
sagebrush ecosystem: A symposium. Utah State Univ., Logan.

Brown, Arthur A. and Kenneth P. Davis. 1973. Forest fire: Control

and use. McGraw-Hill Book Company, New York. 686 p.

Brown, D. 1954. Methods of surveying and measuring vegetation.

Commonwealth Bur. Pasture and Field Crops. Bull. 42. Hurley Berks,
England. 233 p.

Brown, E. Reade (ed.). 1985. Management of wildlife and fish habitats

in forests of western Oregon and Washington, parts 1 and 2. USDA,
For. Serv. Publ. R6-F&WL-192. Pacif. Northw. Reg., Portland, OR. 332
p. and 302 p.

Brown, James K. 1974. Handbook for inventorying downed woody

material. USDA, For. Serv. Gen. Tech. Rep. INT-116. Intermt. For and
Range Exp. Sta., Ogden, UT. 24 p.

Brown, James K. 1975. Fire cycles and community dynamics in

lodgepole pine forests, p. 429-456. IN D. M. Baumgartner (ed.).
Management of lodgepole pine ecosystems. Coop. Extension Serv.
Wash. State Univ., Pullman, WA.

Brown, James K. 1982. Fuel and fire behavior prediction in big

sagebrush. USDA, For. Serv. Res. Pap. INT-290. Intermt. For and
Range Exp. Sta., Ogden, UT. 10 p.

Brown, James K. 1987. Effects of fire on fuels. Lesson Plan. IN

Managing Fire Effects. Boise Interag. Fire Center, Boise, ID. 26 p. and
illus.

Brown, James K. 1989. Personal conversation with Loren Anderson.

Brown, James K., G. D. Booth, and D. G. Simmerman. 1989.

Seasonal change in live fuel moisture of understory plants in western
U.S. aspen, p. 406-412. IN D. C. MacIver, H. Auld, and R. Whitewood
(eds.). Proc. 10th Conf. Fire and For. Meteorol. Environment Canada,
Forestry Canada, Ottawa, ON, Canada.

Brown, James K., Michael A. Marsden, Kevin C. Ryan, and

Elizabeth D. Reinhardt. 1985. Predicting duff and woody fuel
consumed by prescribed fire in the northern Rocky Mountains. USDA,
For. Serv. Res. Pap. INT-337. Intermt. For. and Range Exp. Sta.,
Ogden, UT. 23 p.

Brown, James K., Rick D. Oberheu, and Cameron M. Johnston.

1982. Handbook for inventorying surface fuels and biomass in the
Interior West. USDA, For. Serv. Gen. Tech. Rep. INT-129. Intermt. For.
and Range Exp. Sta., Ogden, UT. 48 p.

Brown, James K., Elizabeth D. Reinhardt, and William C. Fischer.

1991. Predicting duff and woody fuel consumption in northern Idaho
prescribed fires. For. Sci. 37(6): 1550-1566.

Brown, James K. and Thomas E. See. 1981. Downed and dead

woody fuel and biomass in the northern Rocky Mountains. USDA, For.
Serv. Gen. Tech. Rep. INT-117. Intermt. For. and Range Exp. Sta.,
Ogden, UT. 48 p.

Brown, James K. and Dennis G. Simmerman. 1986. Appraising fuels

and flammability in western aspen: A prescribed fire guide. USDA, For.
Serv. Gen. Tech. Rep. INT-205. Intermt. Res. Sta., Ogden, UT. 48 p.

Buckhouse, John C. and Gerald F. Gifford. 1976. Grazing and debris

burning on pinyon-juniper sites -- Some chemical water quality
implications. J. Range Manage. 29:299-301.

Buckman, Harry O. and Nyle C. Brady. 1966. The nature and

properties of soils. A college text of edaphology. 6th Ed. The Macmillan
Co., New York. 567 p.

Bunting, S. C. 1984. Personal conversation with Loren Anderson.

Bunting, Stephen C. 1985. Fire in sagebrush-grass ecosystems:

Successional changes, p. 7-11. IN Ken Sanders and Jack Durham
(eds.). Rangeland fire effects: A symposium. USDI, Bur. Land Manage.,
Idaho State Office, Boise.

Bunting, Stephen C., Bruce M. Kilgore, and Charles L. Bushey.

1987. Guidelines for prescribed burning sagebrush-grass rangelands in
the Northern Great Basin. USDA, For. Serv. Gen. Tech. Rep. INT-231.
Intermt. Res. Sta., Ogden, UT. 33 p.

Bunting, Stephen C., Leon F. Neuenschwander, and George Gruell.

1984. Ecology of antelope bitterbrush in the northern Rocky Mountains,
p. 48-57. IN James E. Lotan and James K. Brown (eds.). Proc. - Fire's
effects on wildlife habitat. USDA, For. Serv. Gen. Tech. Rep. INT-186.
Intermt. For. and Range Exp. Sta., Ogden, UT.

Buol, S. W., F. D. Hole, and R. J. McCracken. 1973. Soil genesis and

classification. The Iowa State Univ. Press. Ames, IA. 360 p.

Burgan, Robert E. 1993. Personal conversation with Melanie Miller.

Burgan, Robert E. and Richard C. Rothermel. 1984. BEHAVE: Fire

behavior prediction and fuel modeling system - FUEL Subsystem.
USDA, For. Serv. Gen. Tech. Rep. INT-167. Intermt. For. and Range
Exp. Sta., Ogden, UT. 126 p.

Burger, George V. 1979. Principles of wildlife management, p. 89-97.

IN Richard D. Teague and Eugene Decker (eds.). Wildlife conservation:
Principles and practices. The Wildlife Society, Washington, D.C.

Burgh, Robert F. 1960. Potsherds and forest fires in the Pueblo

country. Plateau 33:54-56.

Byram, George M. 1959. Combustion of forest fuels, p. 61-89. IN

Davis, Kenneth P. Forest fire: Control and use. McGraw-Hill Book
Company, Inc. New York.

Caldwell, M. M., J. H. Richards, D. A. Johnson, and R. S. Dzurec.

1981. Coping with herbivory: Photosynthetic capacity and resource
allocation in two semiarid Agropyron bunchgrasses. Oecologia (Berlin)
50:14-24.

Campbell, T. M., III. 1979. Short-term effects of timber harvests on pine

marten ecology. M.S. Thesis, Colo. St. Univ., Ft. Collins. 76 p.

Carpenter, L. H. 1976. Nitrogen-herbicide effects on sagebrush deer

range. Ph.D. Thesis, Colo. St. Univ., Ft. Collins. 159 p.

Chambers, Jeanne C. and Ray W. Brown. 1983. Methods for

vegetation sampling and analysis on revegetated mined lands. USDA,
For. Serv. Gen. Tech. Rep. INT-151. Intermt. For. and Range Exp. Sta.,
Ogden, UT. 37 p.

Chandler, Craig, Phillip Cheney, Philip Thomas, Louis Trabaud,

and Dave Williams. 1983. Fire in forestry, Volume I: Forest fire
behavior and effects. John Wiley & Sons, New York. 450 p.

Chase, Alston. 1988. Scientific breakdown: The cultural weakness

behind our ecological failures. Outside (Nov.):45-46.

Christensen, Norman L. and Cornelius H. Muller. 1975a. Relative

importance of factors controlling germination and seedling survival in
Adenostoma chaparral. Amer. Midl. Natur. 93(1):71-78.

Christensen, Norman L. and Cornelius H. Muller. 1975b. Effects of

fire on factors controlling plant growth in Adenostoma chaparral.
Ecolog. Monog. 45:29-55.

Chrosciewicz, Z. 1986. Foliar moisture content variations in four

coniferous tree species of central Alberta. Can. J. For. Res. 16:157-162.

Clark, Robert G. 1986. Personal conversation with Melanie Miller.

Clark, Robert G. 1989. Personal conversation with Melanie Miller.

Cohen, Warren B., Philip N. Omi, and Merrill R. Kaufmann. 1990.

Heating-related water transport to intact lodgepole pine branches. For.
Sci. 36(2): 246-254.

Conrad, C. Eugene and Charles E. Poulton. 1966. Effect of wildfire

on Idaho fescue and bluebunch wheatgrass. J. Range Manage. 19:138-
141.

Cook, C. Wayne. 1966a. Carbohydrate reserves in plants. Utah Agr.

Exp. Sta. Resour. Ser. No. 31. 47 p.

Cook, C. Wayne. 1966b. The role of carbohydrate reserves in

managing range plants. Utah Agr. Exp. Sta. Mimeo Ser. 499. 11 p.

Cook, C. Wayne and James Stubbendieck (eds.). 1986. Range

research: Basic problems and techniques. Soc. Range Manage.,
Denver, CO. 317 p.

Cooperrider, A. Y., R. J. Boyd, and H. R. Stuart (eds.). 1986.

Inventory and monitoring of wildlife habitat. USDI, Bur. Land Manage.,
Service Center, Denver, CO. 858 p.

Countryman, Clive M. 1972. The fire environment concept. USDA, For.

Serv. Pacif. Southw. For. and Range Exp. Sta., Berkeley, CA.

Countryman, Clive M. 1976. Heat--It's role in wildland fire--Part 3. Heat

conduction and wildland fire. USDA, For. Serv. Pacif. Southw. For. and
Range Exp. Sta., Berkeley, CA. 12 p.

Countryman, Clive M. and William M. Dean. 1979. Measuring

moisture content in living chaparral: A field user's manual. USDA,
Forest Service. Gen. Tech. Rep. PSW-36. Pacif. Southw. For. and
Range Exp. Sta., Berkeley, CA. 27 p.

Countryman, Clive M. and Charles W. Philpot. 1970. Physical

characteristics of chamise as a wildland fuel. USDA, For. Serv. Res.
Pap. PSW-66. Pacif. Southw. For. and Range Exp. Sta., Berkeley, CA.
16 p.

Dahl, B. E. and D. N. Hyder. 1977. Developmental morphology and

management implications, p. 257- 290. IN Ronald E. Sosebee, and 9
others. Rangeland plant physiology. Soc. for Range Manage. Denver,
CO.

Dasmann, Raymond F. 1978. Wildlife and ecosystems, p. 18-27. IN H.

P. Brokaw (ed.). Wildlife and America. Counc. Environ. Qual.,
Washington, D.C.

Daubenmire, Rexford. 1959. A canopy-coverage method of

vegetational analysis. Northw. Science 33:43-64.

Daubenmire, Rexford 1968a. Ecology of fire in grasslands. Adv. Ecol.

Res. 5:209-266.

Daubenmire, Rexford. 1968b. Plant communities. A textbook of plant

synecology. Harper & Row. New York. 300 p.

Daubenmire, Rexford. 1975. Plant succession on abandoned fields,

and fire influences in a steppe area in southeastern Washington.
Northw. Sci. 49(1):36-48.

Davis, Edwin A. 1984. Conversion of Arizona chaparral to grass

increases water yield and nitrate loss. Water Resour. Res. 20:1643-
1649.

Davis, Kenneth P. 1959. Forest fire: Control and use. McGraw-Hill

Book Company, Inc. New York. 584 p.

Dealy, Edward J., Donavin A. Leckenby, and Diane M. Concannon.

1981. Plant communities and their importance to wildlife. 66 p. IN
Wildlife habitats in managed rangelands - The Great Basin of
southeastern Oregon. USDA, For. Serv. Gen. Tech. Rep. PNW-120.
Pacif. Northw. For. and Range Exp. Sta., Portland, OR.

DeBano, Leonard F. 1969a. Observations on water repellent soils in

the western United States, p.7- 29. IN Proc. symp. on water repellent
soils. Univ. Calif., Riverside.

DeBano, Leonard F. 1969b. Water repellent soils: A worldwide

concern in management of soil and vegetation. Agric. Sci. Rev. 7:11-18.

DeBano, Leonard F. 1977. Fire's effect on physical and chemical

properties of chaparral soils, p. 65- 74. IN Harold A. Mooney and C.
Eugene Conrad (eds.). Proc. symp. on environmental consequences of
fire and fuel management in Mediterranean ecosystems. USDA, For.
Serv. Gen. Tech. Rep. WO-3. Washington, D.C.

DeBano, Leonard F. 1979. Effects of fire on soil properties, p. 109-118.

IN California forest soils. Univ. Calif. Div. Agric. Sci. Pub. 4094,
Berkeley, CA.

DeBano, Leonard F. 1981. Water repellent soils: A state-of-the art.

USDA, For. Serv. Gen. Tech. Rep. PSW-46. Pacif. Southw. For. and
Range Exp. Sta., Berkeley, CA. 21 p.

DeByle, Norbert V. 1988a. Personal conversation with Melanie Miller.

DeByle, Norbert V. 1988b. Personal conversation with Loren

Anderson.

DeByle, Norbert V. and Paul E. Packer. 1972. Plant nutrient and soil
losses in overland flow from burned forest clearcuts, p. 296-307. IN
Watersheds in transition. Amer. Water Resour. Assoc. and Colo. State
Univ., Ft. Collins.

Deeming, John E., Robert E. Burgan, and Jack D. Cohen. 1978. The
National Fire-Danger Rating System - 1978. USDA, For. Serv. Gen.
Tech. Rep. INT-39. Intermt. For. and Range Exp. Sta., Ogden, UT. 63
p.

Dell, John D. and Charles W. Philpot. 1965. Variations in the moisture

content of several fuel size components of live and dead chamise.
USDA, For. Serv. Res. Note PSW-83. Pacif. Southw. For. and Range
Exp. Sta., Berkeley, CA. 7 p.

Dixon, W. J. (ed.). 1985. BMDP statistical software. Univ. Calif. Press,

Berkeley.

Dost, F. N. 1990. Acute toxicology of components of vegetation smoke.

Reviews of Environmental Contamination and Toxicology Vol. 119:1-46.

Dubos, Rene. 1972. A God within. Charles Scribner's and Sons. p. 155-
210.

Dyksterhuis, E. J. 1958. Range conservation as based on sites and

condition classes. J. Soil and Water Conserv. Vol. 13, No. 4. 1 p.

Dyrness, C. T. and Rodney A. Norum. 1983. The effects of

experimental fires on black spruce forest floors in interior Alaska. Can.
J. For. Res. 13(5):879-893.

EPA. 1976. Quality criteria for water. U.S. Environ. Prot. Agency,
Washington, D.C. 256 p.

EPA. 1979. Methods for chemical analysis of water and wastes. EPA-
600 4-79-020. U.S. Environ. Prot. Agency, Office of Res. and Develop.,
Cincinnati, OH. 460 p.

Eshelman, Kris, Shirley Hudson, Bob Mitchell, Mike Pellant, and

Kay Thomas. 1986. The lighter side of statistics. (Revised). USDI, Bur.
Land Manage., Service Center, Denver, CO. 36 p.

Evans, Raymond A. 1988. Management of pinyon-juniper woodlands.

USDA, For. Serv. Gen. Tech. Rep. INT-249. Intermt. Res. Sta., Ogden,
UT. 34 p.

Everett, Richard L. 1987a. Allelopathic effects of pinyon and juniper

litter on emergence and growth of herbaceous species, p. 62-67. IN
Gary W. Frasier and Raymond A. Evans (eds.). Proc. - Symp. seed and
seedbed ecology of rangeland plants. USDA, Agric. Res. Serv.

Everett, Richard L. 1987b. Plant response to fire in the pinyon-juniper

zone, p. 152-157. IN Richard L. Everett (comp.). Proc. - Pinyon-juniper
conf. USDA, For. Serv. Gen. Tech. Rep. INT-215. Intermt. Res. Sta.,
Ogden, UT.

Fagen, Robert. 1988. Population effects of habitat change. J. Wildl.

Manage. 52:41-46.

Farnsworth, R. B., E. M. Romney, and A. Wallace. 1978. Nitrogen

fixation by microfloral-higher plant associations in arid and semiarid
environments, p. 17-19. IN Neil E. West and John J. Skujins (eds.).
Nitrogen in desert ecosystems. US/IBP Synthesis Ser. 9. Dowden,
Hutchinson, and Ross, Inc., Stroudsburg, PA.

Feller, M. C. and J. P. Kimmins. 1984. Effects of clearcutting and

slash burning on streamwater chemistry and watershed nutrient
budgets in southwestern British Columbia. Water Resour. Res. 20:29-
40.

Ferguson, Robert. 1988. Personal conversation with Melanie Miller.

Finklin, Arnold I. and William C. Fischer. 1990. Weather station

handbook--An interagency guide for wildland managers. Natl. Wildf.
Coord. Grp. NFES No. 1140. PMS No. 426-2. Natl. Interag. Fire Center,
Boise, ID. 237 p.

Fischer, William C. 1981a. Photo guides for appraising downed woody

fuels in Montana forests: How they were made. USDA, For. Serv. Res.
Note INT-299. Intermt. For. and Range Exp. Sta., Ogden, UT. 12 p.

Fischer, William C. 1981b. Photo guide for appraising downed woody

fuels in Montana forests: Grand fir - larch - Douglas-fir; western hemlock
- western redcedar; and western redcedar cover types. USDA, For.
Serv. Gen. Tech. Rep. INT-96. Intermt. For. and Range Exp. Sta.,
Ogden, UT. 53 p.

Fischer, William C. 1981c. Photo guide for appraising downed woody

fuels in Montana forests: Interior ponderosa pine; ponderosa pine -
larch - Douglas-fir; larch - Douglas-fir; and interior Douglas-fir cover
types. USDA, For. Serv. Gen. Tech. Rep. INT-97. Intermt. For. and
Range Exp. Sta., Ogden, UT. 133 p.

Fischer, William C. 1981d. Photo guide for appraising downed woody

fuels in Montana forests: Lodgepole pine, and Engelmann spruce -
subalpine fir cover types. USDA, For. Serv. Gen. Tech. Rep. INT-98.
Intermt. For. and Range Exp. Sta., Ogden, UT. 143 p.

Flinn, Marguerite A. and Ross W. Wein. 1977. Depth of underground

plant organs and theoretical survival during fire. Can J. Bot. 55:2550-
2554.

Floyd, Donald A. and Jay E. Anderson. 1987. A comparison of three

methods for estimating plant cover. J. Ecol. 75:221-228.

Ford-Robertson, F. C. (ed.). 1971. Terminology of forest science

technology practice and products: English-language version. Soc.
Amer. Forest. Multilingual For. Term. Ser. 1., Washington, D.C.

Fosberg, Michael. 1975. Heat and water vapor flux in conifer forest
litter and duff: A theoretical model. USDA, For. Serv. Res. Pap. RM-
152. Rocky Mt. For. and Range Exp. Sta., Ft. Collins, CO. 23 p.

Frandsen, William H. 1987. The influence of moisture and mineral soil

on the combustion limits of smoldering forest duff. Can J. For. Res. 17:
1540-1544.

Frandsen, William H. 1993. Personal conversation with Melanie Miller.

January 28, 1993.

Frandsen, W. H., R. H. Hungerford, and K. C. Ryan. 1993. Heat

transfer into the duff and organic soil. Progress Report. Cooperative
Agreement 14-48-0009-92-962 DCN:98210-2-3927. USDI, Fish and
Wildl. Serv., and USDA, For. Serv. 26 p.

Frandsen, William H. and Kevin C. Ryan. 1986. Soil moisture reduces

belowground heat flux and soil temperatures under a burning fuel pile.
Can. J. For. Res. 16:244-248.

Freese, Frank. 1962. Elementary forest sampling. USDA, Agric.

Handbk. 232. 91 p. (1)

Freese, Frank. 1967. Elementary statistical methods for foresters.

USDA, Agric. Handbk. 317. 87 p. 1

Frischknecht, Neil C. 1978. Effects of grazing, climate, fire, and other

disturbances on long-term productivity of sagebrush-grass ranges, p.
633-735. IN Proc. First Internatl. Rangeland Congress.

Furman, R. William. 1979. Using fire weather data in prescribed fire

planning: Two computer programs. USDA, For. Serv. Gen. Tech. Rep.
RM-63. Rocky Mt. For. and Range Exp. Sta., Ft. Collins, CO. 11 p.

Gardner, Walter H. 1965. Water content, p. 82-127. IN C. A. Black

(ed.). Methods of soil analysis. Part 1: Physical and mineralogical
properties, including statistics of measurement and sampling. Amer.
Soc. Agron., Madison, WI.

Gary, Howard L. 1971. Seasonal and diurnal changes in moisture

contents and water deficits of Engelmann spruce needles. Bot. Gaz.
132(4):327-332.

Geier, A. R. and Louis B. Best. 1980. Habitat selection by small

mammals of riparian communities: Evaluating effects of habitat
alterations. J. Wildl. Manage. 44:16-24.

German, Stephen C. 1988. Initial Attack Management System (IAMS).

Information package. Bur. Land Manage., Div. Inform. Systems
Manage. Boise Interag. Fire Center, ID. 33 p.

Germano, David J. and David N. Lawhead. 1986. Species diversity

and habitat complexity: Does vegetation organize vertebrate
communities in the Great Basin? Great Basin Natur. 46:711-720.

Getz, L. L. 1968. Influence of water balance and microclimate on the

local distribution of the red-back vole and white-footed mouse. Ecology
49:276-280.

Golterman, H. L. and R. S. Clymo. 1969. Methods for chemical

analysis of fresh waters. IBP Handbk. 8. Blackwell Sci. Pub., Oxford,
GB.

Gratkowski, H. 1973. Pregermination treatments for redstem

ceanothus seed. USDA, For. Serv. Res. Pap. PNW-156. Pacif. Northw.
For. and Range Exp. Sta., Portland, OR. 10 p.

Graul, Walter D. and G. C. Miller. 1984. Strengthening ecosystem

management approaches. Wildl. Soc. Bull. 12:282-289

Greig-Smith, P. 1983. Quantitative plant ecology. 3rd ed. Univ. Calif.

Press, Berkeley. 359 p.

Gruell, George E. 1985. Indian fires in the Interior West: A widespread

influence, p. 68-74. IN James E. Lotan, Bruce M. Kilgore, William C.
Fischer, and Robert W. Mutch (tech. coord.). Proc. - Symp. and
workshop on wilderness fire. USDA, For. Serv. Gen. Tech. Rep. INT-
182. Intermt. For. and Range Exp. Sta., Ogden, UT.

Haeussler, S. and D. Coates. 1986. Autecological characteristics of

selected species that compete with conifers in British Columbia: A
literature review. B.C. Ministry of Forests and Lands. FRDA Rep. 001.
Victoria, BC, Canada. 180 p.

Hall, Frederick C. 1985. Management practices and options, 16 p. IN

Wildlife habitats in managed rangelands - The Great Basin of
southeastern Oregon. USDA, For. Serv. Gen. Tech. Rep. PNW-189.
Pacif. Northw. For. and Range Exp. Sta., Portland, OR.

Hanks, R. J. and G. L. Ashcroft. 1980. Applied soil physics. Springer-

Verlag, Berlin. 159 p.

Hansen, James D. 1986. Comparison of insects from burned and

unburned areas after range fire. Great Basin Natur. 46:721-727.

Hanson, Harold Christian. 1962. Dictionary of ecology. Philos. Libr.,

New York. 382 p.

Hardy, Colin. 1990. Personal conversation with Larry Mahaffey.

Hardy, R. 1945. Breeding birds of the pygmy conifers in the Book Cliff
regions of eastern Utah. Auk 62:523-542.

Harniss, Roy O. and Robert B. Murray. 1973. Thirty years of vegetal

change following burning of sagebrush-grass range. J. Range Manage.
26(5):322-325.

Harper, Kimball T., Fred J. Wagstaff, and Lynn M. Kunzler. 1985.

Biology and management of the Gambel oak vegetative type: A
literature review. USDA, For. Serv. Gen. Tech. Rep. INT-179. Intermt.
For. and Range Exp. Sta., Ogden, UT. 31 p.

Harrington, Michael G. 1987. Predicting reduction of natural fuels by

prescribed burning under ponderosa pine in southeastern Arizona.
USDA, For. Serv. Res. Note RM-472. Rocky Mt. For. and Range Exp.
Sta., Ft. Collins, CO. 4 p.

Harrington, Michael G. and Stephen S. Sackett. 1990. Using fire as a

management tool in southwestern ponderosa pine, p. 122-133. IN J. S.
Krammes (tech. coord.). Effects of fire management of southwestern
natural resources. Proc. of the symp. USDA, For. Serv. Gen. Tech.
Rep. RM-191. Rocky Mt. For. and Range Exp. Sta., Ft. Collins, CO.

Harris, Grant A. 1977. Root phenology as a factor of competition

among grass seedlings. J. Range Manage. 30(3):172-177.

Harris, L. D. and W. R. Marion. 1981. Forest stand scheduling for

wildlife in the mutiple-use forest, p. 209-214. IN Proc. conf. of the Soc.
of Amer. Foresters. Orlando, FL.

Hartford, R. A. 1989. Smoldering combustion limits in peat as

influenced by moisture, mineral content, and organic bulk density, p.
282-286. IN D. H. MacIver, H. Auld, and R. Whitewood (eds.). Proc.
10th Conf. Fire and For. Meteorol. Environment Canada, Forestry
Canada, Ottawa, ON, Canada.

Hartford, Robert A. 1993. Personal conversation with Melanie Miller.

Hartford, Roberta A. and William H. Frandsen. 1992. When it's hot,

it's hot . . . or maybe it's not. (Surface flaming may not portend
extensive soil heating). Intl. J. Wildl. Fire 2(3):139-144.

Hartford, Roberta A. and Richard C. Rothermel. 1991. Fuel moisture

as measured and predicted during the 1988 fires in Yellowstone Park.
USDA, For. Serv. Res. Note INT-396. Intermt. Res. Sta., Ogden, UT. 7
p.

Harvey, Alan E., Martin F. Jurgensen, Michael J. Larsen, and

Russell T. Graham. 1987. Decaying organic materials and soil quality
in the inland Northwest: A management opportunity. USDA, For. Serv.
Gen. Tech. Rep. INT-225. Intermt. Res. Sta., Ogden, UT. 15 p.

Harvey, Alan E., Martin F. Jurgensen, Michael J. Larsen, and Joyce

A. Schlieter. 1986. Distribution of active ectomycorrhizal short roots in
forest soils of the Inland Northwest: Effects of site and disturbance.
USDA, For. Serv. Res. Pap. INT-374. Intermt. Res. Sta., Ogden, UT. 8
p.

Haslem, Jack. 1986. Rangeland burning for maximum returns, p. 203-

208. IN Proc. fire management: The challenge of protection and use.
Dept. For. Resour., Utah. State Univ., Logan, UT.

Hatton, Thomas J., Neil R. Viney, E. A. Catchpole, Neville J. De

Mestre. 1988. The influence of soil moisture on eucalyptus leaf litter
moisture. For. Sci. 34(2): 292-301.

Haupt, H. F. 1979. Local climatic and hydrologic consequences of

creating openings in climax timber of north Idaho. USDA, For. Serv.
Res. Pap. INT-223. Intermt. For. and Range Exp. Sta., Ogden, UT. 43
p.

Hays, Robert L., Cliff Summers, and William Seitz. 1981. Estimating
wildlife habitat variables. USDI, Fish and Wildl. Serv. FWS/OBS-81-47.
Washington, D.C. 111 p.

Heilman, P. E. 1966. Change in distribution and availability of nitrogen

with forest succession on north slopes in interior Alaska. Ecology
47:826-831.

Heilman, P. E. 1968. Relationship of availability of phosphorus and

cations to forest succession and bog formation in interior Alaska.
Ecology 49(2):331-336.

Helvey, J. D., A. R. Tiedemann, and T. D. Anderson. 1985. Plant

nutrient losses by soil erosion and mass movement after wildfire. J. Soil,
Water Conserv. 40:168-173.

Hem, John D. 1970. Study and interpretation of the chemical

characteristics of natural water. Water Supply Pap. 1473. USDI, Geol.
Surv., Washington, D.C. 363 p.
Hermann, Sharon M., Ronald A. Phernetton, Allen Carter, and Tony
Gooch. 1991. Fire and vegetation in peat-based marshes of the
Coastal Plain: Examples from the Okefenokee and Great Dismal
Swamps, p. 217-234. IN High intensity fire in wildlands: Management
challenges and options. Proc. 17th Tall Timbers Fire Ecol. Conf. Tall
Timbers Res. Sta. Tallahassee, FL.

Hibbert, Alden R. 1983. Water yield improvement potential by

vegetation management on western rangelands. Water Resour. Bull.
19:375-381.

Hobbs, N. T. and R. W. Spowart. 1984. Effects of prescribed fire on

nutrition of mountain sheep and mule deer during winter and spring. J.
Wildl. Manage. 48:551-560.

Hobbs, N. Thompson. 1989. Linking energy balance to survival in

mule deer: Development and test of a simulation model. J. Wildl.
Manage. Wildl. Monogr. 101. 39 p.

Holechek, Jerry L. 1988. An approach for setting the stocking rate.

Rangelands 10(1):10-14.

Huber, D. M., H. L. Warren, D. W. Nelson, and C. Y. Tsai. 1977.

Nitrification inhibitors - New tools for food production. BioScience
27:523-529.

Huff, Mark H., James K. Agee, and David A. Manuwal. 1984. Postfire
succession of avifauna in the Olympic Mountains, Washington, p. 8-15.
IN James E. Lotan and James K. Brown (eds.). Proc. - Fire's effects on
wildlife habitat. USDA, For. Serv. Gen. Tech. Rep. INT-186. Intermt.
For. and Range Exp. Sta., Ogden, UT.

Hungerford, Roger. 1989. Modeling the downward heat pulse from fire
in soils and in plant tissue, p. 148-154. IN D. C. MacIver, H. Auld, and
R. Whitewood (eds.). Proc. - 10th Conf. on Fire and Forest Meteorol.
Environment Canada, Forestry Canada, Ottawa, ON, Canada.

Hutchings, Selar S. and Jack E. Schmautz. 1969. A field test of the

relative-weight estimate method for determining herbage production. J.
Range Manage. 22(6): 408-411.
Hutchison, B. A. 1965. Snow accumulation and disappearance
influenced by big sagebrush. USDA, For. Serv. Res. Note RM-46.
Rocky Mt. For. and Range Exp. Sta., Ft. Collins, CO.

James, Susanne. 1984. Lignotubers and burls - Their structure,

function and ecological significance in Mediterranean ecosystems. Bot.
Rev. 50(3):225-266.

Johansen, Ragnar, John Deeming, Mike Long, and Darold Ward.

1985. Chapter II. Smoke production characteristics and effects, p. 5-10.
IN Prescribed fire smoke management guide. National Wildf. Coord.
Grp. Prescr. Fire and Fire Effects Working Team. NFES No. 1279.

Johnsgard, P. A. and W. H. Rickard. 1957. The relationship of spring

bird distribution to vegetative mosaic in southern Washington. Ecology
38:171-174.

Jones, Anne T. and Robert C. Euler. 1986. Effects of forest fires on

archaeological resources at Grand Canyon National Park. N. Amer.
Archaeol. 7:243-254.

Jones, John R. 1985. Distribution, p. 9-18. IN Norbert V. DeByle and

Robert P. Winokur (eds.). Aspen: Ecology and management in the
western United States. USDA, For. Serv. Gen. Tech. Rep. RM-119.
Rocky Mt. For. and Range Exp. Sta., Ft. Collins, CO.

Jones, John R. and Norbert V. DeByle. 1985. Fire, p. 77-81. IN

Norbert V. DeByle and Robert P. Winokur (eds.). Aspen: Ecology and
management in the western United States. USDA, For. Serv. Gen.
Tech. Rep. RM-119. Rocky Mt. For. and Range Exp. Sta., Ft. Collins,
CO.

Keane, Robert E., James K. Brown, Elizabeth D. Reinhardt, and

Kevin C. Ryan. 1990. Predicting first order fire effects in the United
States. Compiler 8(4):11-15.

Keeley, Jon E. 1987. Role of fire in seed germination of woody taxa in

California chaparral. Ecology 68(2):434-443.

Kelleyhouse, David G. 1980. Fire/wildlife relations in Alaska. Unpub.

manuscript. Alaska Dept. of Fish and Game, Fairbanks. 19 p.
Kelly, Roger E. and Jim Mayberry. 1980. Trial by fire: Effects of NPS
burn programs upon archaeological resources, p. 603-610. IN Proc.
second conf. on scientific research in the Natl. Parks, Volume 1. USDI,
Nat. Park Serv., West. Reg., San Francisco, CA.

Kender, Walter J. 1967. Rhizome development in the lowbush

blueberry as influenced by temperature and photoperiod. Am. Soc.
Hort. Sci. 90:144-148.

Kindschy, Robert R. 1986. Rangeland vegetative succession -

Implications to wildlife. Rangelands 8:157-159.

Kirby, Ronald E., Stephen J. Lewis, and Terry N. Sexon. 1988. Fire
in North American wetland ecosystems and fire-wildlife relations: An
annotated bibliography. USDI, Fish Wildl. Serv. Bio. Rep. 88(1).
Washington, D.C.

Knopf, Fritz L., James A. Sedgewick, and Richard W. Cannon.

1988. Guild structure of a riparian avifauna relative to seasonal cattle
grazing. J. Wildl. Manage. 52:280-290.

Koski, Wayne H. and William C. Fischer. 1979. Photo series for

appraising thinning slash in north Idaho: Western hemlock, grand fir,
and western redcedar timber types. USDA, For. Serv. Gen. Tech. Rep.
INT-46. Intermt. For. and Range Exp. Sta., Ogden, UT. 50 p.

Krammes, J. S. and J. Osborn. 1969. Water-repellent soils and

wetting agents as factors influencing erosion, p. 177-187. IN L. F.
DeBano and J. Letey (eds.). Proc. symp. on water-repellent soils. Univ.
Calif., Riverside.

Lancaster, James W. 1970. Timelag useful in fire danger rating. Fire

Control Notes 32(3): 6-8, 10.

Lance, J. C., S. C. McIntyre, J. W. Naney, and S. S. Rousseva. 1986.

Measuring sediment movement at low erosion rates using Cesium-137.
J. Soil Sci. Soc. of Amer. Vol. 50: 1303-1309.

Larson, L. L. and P. A. Larson. 1988. Historic landscape restoration:

Views from 3 angles. Park Sci. 8(3):6-9. USDI, Natl. Park Serv.,
Washington, D.C.

Lawrence, G. E. 1966. Ecology of vertebrate animals in relation to

chaparral fire on the Sierra Nevada foothills. Ecology 47:278-291.

Laycock, W. A. 1979. Management of sagebrush. Rangelands 1: 207-

210.

Leege, Thomas A. 1968. Prescribed burning for elk in northern Idaho,

p. 235-253. IN Proc. Tall Timbers Fire Ecol. Conf. No. 8. Tall Timbers
Res. Sta., Tallahassee, FL.

Leege, Thomas A. and William O. Hickey. 1971. Sprouting of

northern Idaho shrubs after prescribed burning. J. Wildl. Manage.
35:508-515.

Leopold, A. S. and F. F. Darling. 1953. Effects of land use on moose

and caribou in Alaska. Trans. North. Amer. Wildl. Conf. 18:553-562.

Lewis, Henry T. 1973. Patterns of Indian burning in California: Ecology

and ethnohistory. Ballena Press Anthropol. Papers, No. 1. Ramona,
CA. 101 p.

Lewis, Henry T. 1985. Why Indians burned: Specific versus general

reasons, p. 75-80. IN James E. Lotan, Bruce M. Kilgore, William C.
Fischer, and Robert W. Mutch (tech. coord.). Proc. - Symp. and
workshop on wilderness fire. USDA, For. Serv. Gen. Tech. Rep. INT-
182. Intermt. For. and Range Exp. Sta., Ogden, UT.

Littke, W. R. and R. I. Gara. 1986. Decay of fire-damaged lodgepole

pine in south-central Oregon. For. Ecol. & Manage. 17:279-287.

Little, Susan N., Roger D. Ottmar, and Janet L. Ohmann. 1986.

Predicting duff consumption from prescribed burns on conifer clearcuts
in western Oregon and western Washington. USDA, For. Serv. Res.
Pap. PNW-362. Pacif. Northw. Res. Sta., Portland, OR.

29 p.

Little, Thomas M. and F. Jackson Hills. 1978. Agricultural

experimentation. John Wiley & Sons, New York. 350 p.
Loomis, Robert M. and Richard W. Blank. 1981. Summer moisture
content of some northern lower Michigan understory plants. USDA, For.
Serv. Res. Note NC-263. N. Central For. Exp. Sta., St. Paul, MN. 4 p.

Lotan, James E. 1976. Cone serotiny - fire relationships in lodgepole

pine, p. 267-278. IN Proc. Tall Timbers Fire Ecol. Conf. No. 14. Tall
Timbers Res. Sta., Tallahassee, FL.

Lyon, J. G., J. F. McCarthy, and J. T. Heinen. 1986. Video digitation

of aerial photographs for measurement of wind erosion damage on
converted rangeland. Photogram. Enginrg. and Remote Sensing 52(3):
373-377.

Lyon, L. Jack, Hewlette S. Crawford, Eugene Czuhai, Richard L.

Fredricksen, Richard F. Harlow, Louis J. Metz, and Henry A.
Pearson. 1978. Effects of fire on fauna: A state-of-knowledge review.
USDA, For. Serv. Gen. Tech. Rep. WO-6. Washington, D.C. 41 p.

Lyon, L. Jack and John M. Marzluff. 1984. Fire's effects on a small

bird population, p. 16-22. IN James E. Lotan and James K. Brown
(eds.). Proc .- Fire's effects on wildlife habitat. USDA, For. Serv. Gen.
Tech. Rep. INT-186. Intermt. For. and Range Exp. Sta., Ogden, UT.

MacMahon, J. A. 1980. Ecosystems over time: Succession and other

types of change, p. 27-58. IN R. H. Waring (ed.). Forests: Fresh
perspectives from ecosystem analysis. Oregon St. Univ. Press,
Corvallis.

Main, William A., Robert G. Straub, and Donna M. Pannanen. 1988.

FIREFAMILY 1988. USDA, For. Serv. Gen. Tech. Rep. NC-138. N.
Central For. Exp. Sta., St. Paul, MN. 35 p.

Mandeville, M. D. 1973. A consideration of the thermal pre-treatment of

chert. Plains Anthropol. 18:177-202.

Martin, Robert E. 1963. A basic approach to fire injury of tree stems, p.

151-162. IN Proc. Tall Timbers Fire Ecol. Conf. No. 2. Tall Timbers Res.
Sta., Tallahassee, FL.

Martin, Robert E., Hal E. Anderson, William D. Boyer, John H.

Dieterich, Stanley N. Hirsch, Von J. Johnson, and W. Henry McNab.
1979. Effects of fire on fuels. A state-of-knowledge review. USDA, For.
Serv. Gen. Tech. Rep. WO-13. 64 p.

Martin, Robert E., David W. Frewing, and James L. McClanahan.

1981. Average biomass of four Northwest shrubs by fuel size class and
crown cover. USDA, For. Serv. Res. Note PNW-74. Pacif. Northw. For.
and Range Exp. Sta., Portland, OR. 6 p.

Maser, Chris and Jay S. Gashwiler. 1978. Interrelationships of wildlife

and western juniper, p. 37-82. IN Robert E. Martin, J. Edward Dealy,
and David L. Caraher (eds.). Proc. - Western juniper ecology and
management workshop. USDA, For. Serv. Gen. Tech. Rep. PNW-74.
Pacif. Northw. For. and Range Exp. Sta., Portland, OR.

Maser, Chris, Jack Ward Thomas, and Ralph G. Anderson. 1984.

The relationship of terrestrial vertebrates to plant communities, part 1
and 2. 25 p. and 237 p. IN Wildlife habitats in managed rangelands -
The Great Basin of southeastern Oregon. USDA, For. Serv. Gen. Tech.
Rep. PNW-172. Pacif. Northw. For. and Range Exp. Sta., Portand, OR.

Mathews, Ed, Lee Lavdas, Larry Mahaffey, Tom Nichols, Dave

Sandberg, and Mike Ziolko. 1985. Chapter III. Smoke management, p.
11-18. IN Prescribed fire smoke management guide. Natl. Wildf. Coord.
Grp. Prescr. Fire and Fire Effects Working Team. NFES No. 1279.

Maxwell, Wayne G. and Franklin R. Ward. 1976a. Photo series for

quantifying forest residues in the coastal Douglas-fir - hemlock type;
coastal Douglas-fir - hardwood type. USDA, For. Serv. Gen. Tech. Rep.
PNW-51. Pacif. Northw. For. and Range Exp. Sta., Portland, OR. 103 p.

Maxwell, Wayne G. and Franklin R. Ward. 1976b. Photo series for

quantifying forest residues in the ponderosa pine type; ponderosa pine
and associated species type; lodgepole pine type. USDA, For. Serv.
Gen. Tech. Rep. PNW-52. Pacif. Northw. For. and Range Exp. Sta.,
Portland, OR. 73 p.

Maxwell, Wayne G. and Franklin R. Ward. 1979. Photo series for

quantifying forest residues in the Sierra mixed conifer type, Sierra true
fir type. USDA, For. Serv. Gen. Tech. Rep. PNW-95. Pacif. Northw. For.
and Range Exp. Sta., Portland, OR.
Maxwell, Wayne G. and Franklin R. Ward. 1980. Photo series for
quantifying natural forest residues in common vegetation types of the
Pacific Northwest. USDA, For. Serv. Gen. Tech. Rep. PNW-105. Pacif.
Northw. For. and Range Exp. Sta., Portland, OR. 229 p.

McAdoo, J. Kent, William S. Longland, and Raymond A. Evans.

1989. Nongame bird community responses to sagebrush invasion of
crested wheatgrass seedings. J. Range Manage. 53:494-502.

McCammon, Bruce P. 1976. Snowpack influences on dead fuel

moisture. For. Sci. 22(3): 323-328.

McConnell, B. R. and G. A. Garrison. 1966. Seasonal variations of

available carbohydrates in bitterbrush. J. Wildl. Manage. 30(1):168-172.

McCreight, Richard W. 1981. Microwave ovens for drying live wildland

fuels: An assessment. USDA, For. Serv. Res. Note PSW-349. Pacif.
Southw. For. and Range Exp. Sta., Berkeley, CA. 5 p.

McCulloch, C. Y., D. C. Wallmo, and P. F. Ffolliott. 1965. Acorn yield

of Gambel oak in northern Arizona. USDA, For. Serv. Res. Note RM-48.
Rocky Mt. For. and Range Exp. Sta., Ft. Collins, CO. 2 p.

McGee, John M. 1982. Small mammal populations in an unburned and

early fire successional sagebrush community. J. Range Manage.
35:177-179.

McMahon, C. K., C. W. Adkins, and S. L. Rodgers. 1987. A video

image analysis system for measuring fire behavior. Fire Manage. Notes
47(1):10-15.

McMurray, Nancy E. 1988. Pinus ponderosa var. scopulorum. IN

William C. Fischer. Fire Effects Information System. [Data base].
Missoula, Montana. USDA, For. Serv. Intermt. Res. Sta. Intermt. Fire
Sciences Lab. Magnetic tape reels; 9 track; 1600 bpi; ASCII with
Common LISP present.

McPherson, J. K. and C. H. Muller. 1969. Allelopathic effect of

Adenostoma fasciculatum, "chamise," in the California chaparral. Ecol.
Monogr. 39:177-198.
McRae, Douglas J., Martin E. Alexander, and Brian J. Stocks. 1979.
Measurement and description of fuels and fire behavior on prescribed
burns: a handbook. Canad. Forest. Serv., Dept. Environ. Rep. O-X-287.
Great Lakes For. Res. Centre, Sault Ste. Marie, ON. 58 p.

Meyer, Bernard S., Donald B. Anderson, Richard H. Bohning, and

Douglas G. Fratianne. 1973. Introduction to plant physiology. D. Van
Nostrand Company. New York. 565 p.

Michels, Joseph. 1973. Dating methods in archaeology. Seminar

Press, New York. p. 193.

Miller, Melanie. 1976. Shrub sprouting response to fire in a Douglas-

fir/western larch ecosystem. M.S. thesis. Univ. Mont., Missoula. 124 p.

Miller, Melanie. 1977. Response of blue huckleberry to prescribed fires

in a western Montana larch-fir forest. USDA, For. Serv. Res. Pap. INT-
188. Intermt. For. and Range Exp. Sta., Ogden, UT. 33 p.

Miller, Melanie. 1978. Effect of growing season on sprouting of blue

huckleberry. USDA, For. Serv. Res. Note INT-240. Intermt. For. and
Range Exp. Sta., Ogden, UT. 8 p.

Miller, Melanie. 1981. Personal observation. Fire ecologist. USDI, Bur.

Land Manage., Fairbanks District Ofice, Fairbanks, AK.

Miller, Melanie. 1988. Personal observation. Fire ecologist. USDI, Bur.

Land Manage., Boise Interag. Fire Center, Boise, ID.

Moen, Aaron N. 1979. Animal behavior, p. 107-116. IN Richard D.

Teague and Eugene Decker (eds.). Wildlife conservation: Principles
and practices. The Wildlife Society, Washington, D.C.

Moore, Robert. 1989. Personal conversation with Loren Anderson.

Morgan, Penelope and Leon F. Neuenschwander. 1988. Shrub

response to high and low severity burns following clear-cutting in
northern Idaho. West. J. Appl. For. 3(1):5-9.
Mueggler, Walter F. 1976. Ecological role of fire in western woodland
and range ecosystems, p. 1-9. IN Frank E. Busby and Edward Storey
(eds.). Use of prescribed burning in western woodland and range
ecosystems: A symposium. Utah Agric. Exp. Sta., Utah State Univ.,
Logan.

Mueggler, Walter F. 1983. Variation in production and seasonal

development of mountain grasslands in western Montana. USDA, For.
Serv. Res. Pap. INT-316. Intermt. For. and Range Exp. Sta., Ogden,
UT. 16 p.

Mueller-Dombois, D. and H. Ellenberg. 1974. Aims and methods of

plant ecology. John Wiley and Sons, New York. 547 p.

Muller, C. H., R. B. Hanawalt, and J. K. McPherson. 1968.

Allelopathic control of herb growth in the fire cycle of California
chaparral. Bull. Torrey Bot. Club 95:225-231.

Mutch, R. W. and O. W. Gastineau. 1970. Timelag and equilibrium

moisture content of reindeer lichen. USDA, For. Serv. Res. Pap. INT-76.
Intermt. For. and Range Exp. Sta., Ogden, UT. 8 p.

Neitro, William A., Virgil W. Binkley, Steven P. Cline, R. William

Mannan, Bruce G. Marcot, Douglas Taylor, and Frank F. Wagner.
1985. Snags (wildlife trees), p. 129-169. IN E. Reade Brown (ed.).
Management of wildlife and fish habitats in forests of western Oregon
and Washington. Part 1 - Chapter narratives. USDA, For. Serv. Publ.
No. R6-F&WL-192-1985. Pacif. Northw. Reg., Portland, Oregon.

Nelson, J. G. 1973. The last refuge. Harvest House, Montreal, Quebec,

Canada. 230 p.

Nissley, S. D., R. J. Zasoski, and R. E. Martin. 1980. Nutrient

changes after prescribed surface burning of Oregon ponderosa pine
stands, p. 214-219. IN R. E. Martin and six others (eds.). Proc. sixth
conf. on fire and forest meterol. Soc. Amer. Forest., Washington, D.C.

Norum, Rodney A. 1975. Characteristics and effects of understory fires

in western larch/Douglas-fir stands. Ph.D. Dissertation. Univ. of
Montana, Missoula. 155 p.
Norum, Rodney A. 1977. Preliminary guidelines for prescribed burning
under standing timber in western larch/Douglas-fir forests. USDA, For.
Serv. Res. Note INT-229. Intermt. For. and Range Exp. Sta., Ogden,
UT. 15 p.

Norum, Rodney A. 1987. Ignition and firing: How they meet and/or
interact with other prescription variables. Lesson Plan. IN Fire
Prescription Writing. Boise Interag. Fire Center, Boise, ID.

Norum, Rodney A. 1992. Personal conversation with Melanie Miller.

Norum, Rodney A. and William C. Fischer. 1980. Determining the

moisture content of some dead forest fuels using a microwave oven.
USDA, For. Serv. Res. Note INT-177. Intermt. For. and Range Exp.
Sta., Ogden, UT. 7 p.

Norum, Rodney A. and Melanie Miller. 1981. Unpublished data on file

at USDI, Bur. Land Manage., Div. Fire and Aviat. Pol. and Manage.,
Boise, ID.

Norum, Rodney A. and Melanie Miller. 1984. Measuring fuel moisture

content in Alaska: Standard methods and procedures. USDA, For. Serv.
Gen. Tech. Rep. PNW-171. Pacif. Northw. For. and Range Exp. Sta.,
Portland, OR. 34 p.

Noste, Nonan V. and Charles L. Bushey. 1987. Fire response of

shrubs of dry forest habitat types in Montana and Idaho. USDA, For.
Serv. Gen. Tech. Rep. INT-239. Intermt. Res. Sta., Ogden, UT. 22 p.

Noxon, J. S. and D. A. Marcus. 1983. Wildfire-induced cliff face

exfoliation and potential effects on cultural resources in the Needles
District of Canyonlands National Park, Utah. Southwestern Lore 49(2):1-
8.

Odum, Eugene P. 1966. Fundamentals of ecology. 2nd ed. W. B.

Saunders Co., Philadelphia, PA. 546 p.

Ottmar, Roger E., Mary F. Burns, Janet N. Hall, and Aaron D.

Hanson. 1993. CONSUME Users Guide. USDA, For. Serv. Gen. Tech.
Rep. PNW-GTR-304. Pacif. Northw. Res. Sta., Portland, OR. 118 p.
Ottmar, Roger D., Colin C. Hardy, and Robert E. Vihnanek. 1990.
Stereo photo series for quantifying forest residues in the Douglas-fir-
hemlock type of the Willamette National Forest. USDA, For. Serv. Gen.
Tech. Rep. PNW-GTR-258. Pacif. Northw. For. and Range Exp. Sta.,
Portland, OR. 63 p.

Ottmar, Roger D. and David V. Sandberg. 1985. Calculating moisture

content of 1000-hour timelag fuels in western Washington and western
Oregon. USDA, For. Serv. Res. Pap. PNW-336. Pacif. Northw. For. and
Range Exp. Sta., Portland, OR. 16 p.

Owensby, Clenton E. and John Bruce Wyrill, III. 1973. Effects of

range burning on Kansas Flint Hills soil. J. Range Manage. 26:185-188.

Parker, G. R., J. W. Maxwell, L. D. Morton, and G. E. J. Smith. 1983.

Ecology of the lynx (Lynx canadensis) on Cape Breton Island. Can. J.
Zool. 61:770-786.

Parker, V. Thomas. 1984. Correlation of physiological divergence with

reproductive mode in chaparral shrubs. Madrono 31(4):231-242.

Parker, V. Thomas. 1987. Can native flora survive prescribed burns?

Fremontia 15(2):3-6.

Parker, V. Thomas. 1989. Maximizing vegetation response on

management burns by identifying fire regimes, p. 87-91. IN Neil H. Berg
(tech. coord.). Proc. symp. on fire and watershed management. USDA,
For. Serv. Gen. Tech. Rep. PSW-109. Pacif. Southw. For. and Range
Exp. Sta., Berkeley, CA.

Pase, Charles P. and Carl E. Granfelt. 1977. The use of fire on

Arizona rangelands. Arizona Interag. Range Comm. Pub. No. 4. 15 p.

Paysen, Timothy E. and Jack D. Cohen. 1990. Chamise chaparral

dead fuel fraction is not reliably predicted by age. W. J. Appl. Forest.
5(4):127-131.

Pechanec, J. F., G. Stewart, and J. P. Blaisdell. 1954. Sagebrush

burning - Good and bad. USDA Farmer's Bull. 1948. 34 p.

Peech, Michael. 1965. Hydrogen-ion activity, p. 914-932. IN C. A.

Black (ed.). Methods of soil analysis. Part 2: Chemical and
microbiological properties. Amer. Soc. Agron., Madison, WI.

Peek, James M., Dennis A. Demarchi, and Donald E. Stucker. 1984.

Bighorn sheep and fires: Seven case histories, p. 36-43. IN James E.
Lotan and James K. Brown (eds.). Proc. - Fire's effects on wildlife
habitat. USDA, For. Serv. Gen. Tech. Rep. INT-186. Intermt. For. and
Range Exp. Sta., Ogden, UT.

Peek, J. M., R. A. Riggs, and J. L. Lauer. 1979. Evaluation of fall

burning on bighorn sheep winter range. J. Range Manage. 32:430-432.

Pellant, Mike. 1989. Evaluation. Lesson Plan. IN Fire Effects on Public

Lands. Boise Interag. Fire Center. Boise, Idaho.

Pennak, Robert W. 1964. Collegiate dictionary of zoology. Ronald

Press Co., New York. 583 p.

Perry, D. A., R. Molina, and M. P. Amaranthus. 1987. Mycorrhizae,

mycorrhizospheres, and reforestation; current knowledge and research
needs. Can. J. For. Res. 17:929-940.

Petersburg, Stephen. 1989. Personal conversation with Melanie Miller.

Peterson, David L. 1985. Crown scorch volume and scorch height:

Estimates of postfire tree condition. Can. J. For. Res. 15:596-598.

Peterson, Terry D. 1989. Characteristics of Ceanothus. IN Forest

vegetation management workshop. Oreg. State Univ., Corvallis.
Unpaged.

Philpot, Charles W. and Robert W. Mutch. 1971. The seasonal trends

in moisture content, ether extractives, and energy of ponderosa pine
and Douglas-fir needles. USDA, For. Serv. Res. Pap. INT-102. Intermt.
For. and Range Exp. Sta., Ogden, UT. 21 p.

Pidanick, Bill. 1982. Prescribed fire/cultural artifacts: Investigating the

effects. Pacif. Southw. Log. USDA, For. Serv., Pacif. Southw. Reg., San
Francisco, CA. 2 p.

Pieper, R. D. 1973. Measurement techniques for herbaceous and

shrubby vegetation. Dept. Anim., Range, and Wildl. Sci., New Mex.
State Univ., Las Cruces, NM. 187 p.

Pilles, Peter J. 1982. Prescribed fire management and cultural

resource management. Manuscript. USDA, For. Serv. Southw. Reg.,
Coconino Nat. For., Flagstaff, AZ. 11 p.

Platts, W. S., W. F. Megahan, and G. W. Minshall. 1983. Methods for

evaluating stream, riparian, and biotic conditions. USDA, For. Serv.
Gen. Tech. Rep. INT-138. Intermt. For. and Range Exp. Sta., Ogden,
UT. 70 p.

Pritchett, William L. 1979. Properties and management of forest soils.

John Wiley & Sons, New York. 500 p.

Purdy Barbara A. and H. K. Brooks. 1971. Thermal alteration of silica

minerals: An archaeological approach. Science 173: 322-325.

Quinn, R. D. 1979. Effects of fire on small mammals in the chaparral.

Calif. Nev. Wildl. Trans. 1979:125-133.

Raabe, O. G. 1984. Site selective sampling criteria for the thoracic and
respirable mass fractions. Ann. Am. Conf. Ind. Hyg. 11:53-65.

Racine, Charles H. and Marilyn M. Racine. 1979. Tundra fires and

two archaeological sites in the Seward Peninsula, Alaska. Arctic 32(1):
76-79.

Raison, R. J. 1979. Modification of the soil environment by vegetation

fires, with particular reference to nitrogen transformations: A review.
Plant and Soil 51:73-108.

Range Term Glossary Committee. 1974. A glossary of terms used in

range management. Soc. Range Manage., Denver. CO. 36 p.

Ranger, G. E. and F. F. Frank. 1978. The 3-F erosion bridge - A new

tool for measuring soil erosion. Calif. Dept. Forest. and Fire Prot. Pub.
23. Sacramento, CA.

Rasmussen, G. Allen and Henry A. Wright. 1988. Germination

requirements of flameleaf sumac. J. Range Manage. 41(1):48-52.
Reaves, Jimmy L., Charles G. Shaw, III, and John E. Mayfield. 1990.
The effects of Trichoderma spp. isolated from burned and non-burned
forest soils on the growth and development of Armillaria ostoyae.
Northw. Sci. 64(1):39-44.

Reese, Kerry P. and John T. Ratti. 1988. Edge effect: A concept

under scrutiny. Trans. N. Amer. Wildl. Conf. 53:127-136.

Reifsnyder, William E. 1961. Seasonal variation in the moisture

content of green leaves of mountain laurel. For. Sci. 7(1): 16-23.

Reinhardt, Elizabeth, James K. Brown, William C. Fischer, and

Russell T. Graham. 1991. Woody fuel and duff consumption by
prescribed fire in northern Idaho mixed conifer logging slash. USDA,
For. Serv. Res. Pap. INT-443. Intermt. Res. Sta., Ogden, UT. 22 p.

Reinhardt, Elizabeth D. and Kevin C. Ryan. 1988a. Eight-year tree

growth following prescribed underburning in a western Montana
Douglas-fir/western larch stand. USDA, For. Serv. Res. Note INT-387.
Intermt. Res. Sta., Ogden, UT. 6 p.

Reinhardt, Elizabeth D. and Kevin C. Ryan. 1988b. Estimating tree

mortality resulting from prescribed fire. Fire Manage. Notes 49(4): 30-
36.

Reinhardt, T. E. 1989. Firefighter smoke exposure at prescribed burns.

A study and action recommendation. Unpubl. Rep. USDA, For. Serv.
Pacif. Northw. For. and Range Exp. Sta., Seattle, WA. 86 p. plus appen.

Renard, K. G., G. R. Foster, G. A. Weesies, and J. P. Porter. 1991.

RUSLE: Revised universal soil loss equation. J. Soil and Water
Conserv. 46(1):30-33.

Richards, J. H. and M. M. Caldwell. 1985. Soluble carbohydrates,

concurrent photosynthesis and efficiency in regrowth following
defoliation: A field study with Agropyron species. J. Appl. Ecol. 22:907-
920.

Richards, Leon W. 1940. Effect of certain chemical attributes of

vegetation on forest inflammability. J. Agric. Res. 60(12): 833-838.
Richter, D. D. and C. W. Ralston. 1982. Prescribed fire: Effects on
water quality and forest nutrient cycling. Science 215(4533):661-663.

Riedel, A. L. and S. J. Petersburg. 1989. Live fuel moisture in

Wyoming big sagebrush (Artemisia tridentata wyomingensis) in
Dinosaur National Monument. Manuscript on file at Dinosaur Natl. Mon.,
Dinosaur, CO. 19 p.

Rietveld, W. J. 1976. Cone maturation in ponderosa pine foliage

scorched by wildfire. USDA, For. Ser. Res. Note RM-317. Rocky Mt.
For. and Range Exp. Sta., Ft. Collins, CO. 7 p.

Robb, L. A. 1987. Gastropod intermediate host of lungworm

(Nematoda: Protostrongylidae) on a bighorn sheep winter range:
Aspects of transmission. M.S. thesis, Univ. of Alberta, Edmonton. 111
p.

Rodriguez-Barrueco, C. and G. Bond. 1968. Nodule endophytes in

the genus Alnus, p. 185-192. IN J. M. Trappe, J. F. Franklin, R. F.
Tarrant, and G. M. Hanson (eds.). Biology of alder. Pacif. Northw. For.
and Range Exp. Sta., Portland, OR.

Rogers, G. F. and M. K. Vint. 1987. Winter precipitation and fire in the

Sonoran desert. J. Arid Environ. 13:47-52.

Rosenberg, K. V. and M. G. Raphael. 1986. Effects of forest

fragmentation in Douglas-fir forests, p. 263-272. IN J. Verner, M. L.
Morrison, and C. J. Ralphs (eds.). Wildlife 2000: Modeling habitat
relationships of terrestrial vertebrates. Univ. Wisc. Press., Madison.

Rosentreter, Roger. 1989. Personal conversation with Loren

Anderson.

Rotenberry, John T. and John A. Wiens. 1978. Nongame bird

communities in northwestern rangelands, p. 32-46. IN R. M. DeGraaf
(tech. coord.). Proc. nongame bird habitat management in the
coniferous forests of the western U.S. USDA, For. Serv. Gen. Tech.
Rep. PNW-64. Pacif. Northw. For. and Range Exp. Sta., Portland, OR.

Rothermel, Richard C. 1972. A mathematical model for predicting fire

spread in wildland fuels. USDA, For. Serv. Res. Pap. INT-114. Intermt.
For. and Range Exp. Sta., Ogden, UT. 40 p.

Rothermel, Richard C. 1980. Personal letter to Rodney A. Norum.

September 16, 1980.

Rothermel, Richard C. 1983. How to predict the spread and intensity

of forest and range fires. USDA, For. Serv. Gen. Tech. Rep. INT-143.
Intermt. For. and Range Exp. Sta., Ogden, UT.

161 p.

Rothermel, Richard C. 1991. Predicting behavior and size of crown

fires in the Northern Rocky Mountains. USDA, For. Serv. Res. Pap. INT-
438. Intermt. Res. Sta., Ogden, UT. 46 p.

Rothermel, Richard C. and J. E. Deeming. 1980. Measuring and

interpreting fire behavior for correlation with fire effects. USDA, For.
Serv. Gen. Tech. Rep. INT-93. Intermt. For. and Range Exp. Sta.,
Ogden, UT. 4 p.

Rothermel, Richard C., Ralph A. Wilson, Jr., Glen A. Morris,

Stephen S. Sackett. 1986. Modeling moisture content of fine dead
wildland fuels: Input to the BEHAVE Fire Prediction System. USDA, For.
Serv. Gen. Tech. Rep. INT-359. Intermt. Res. Sta., Ogden, UT. 4 p.

Roundy, Bruce A., James A. Young, Greg J. Cluff, and Raymond A.

Evans. 1983. Measurement of soil water on rangelands. USDA, Agric.
Res. Serv. Res. Results West. Ser. No. 31. Oakland, CA. 27 p.

Rowe, J. S. 1983. Concepts of fire effects on plant individuals and

species, p. 135-151. IN R. W. Wein and D. A. Maclean (eds.). The role
of fire in northern circumpolar ecosystems. SCOPE Rep., No. 18. John
Wiley and Sons. New York. 322 p.

Ryan, Kevin C. 1982. Evaluating potential tree mortality from

prescribed burning, p. 167-178. IN David M. Baumgartner (comp.).
Symp. - Site preparation and fuels management in steep terrain. Wash.
State Univ. Coop. Extension, Pullman.

Ryan, Kevin C. 1983. Techniques for assessing fire damage to trees,

p. 1-11. IN James E. Lotan (ed.) Proc. Symp. on Fire - Its field effects.
Intermountain Fire Council. Missoula, MT.

Ryan, Kevin C. 1987. Techniques for writing fire prescriptions to

minimize or enhance tree mortality. Lesson Plan. IN Fire Prescription
Writing. Boise Interagency Fire Center, Boise, ID.

Ryan, Kevin C. and Nonan V. Noste. 1985. Evaluating prescribed

fires, p. 230-238. IN Proc. - Symp. and workshop on wilderness fire.
USDA, For. Serv. Gen. Tech. Rep. INT-182. Intermt. For. and Range
Exp. Sta., Ogden, UT.

Ryan, Kevin C., David L. Peterson, and Elizabeth D. Reinhardt.

1988. Modeling long-term fire-caused mortality of Douglas-fir. For. Sci.
34(1):190-199.

Sackett, Steve. 1981. Designing fuel moisture sampling systems.

Lesson plan. IN Fire Management for Managers. Natl. Adv. Resour.
Tech. Center. Marana, AZ.

Samuel, M. J. and E. J. DePuit. 1987. Competition and plant

establishment, p. 138-148. IN Gary W. Frasier and Raymond A. Evans
(eds.). Proc. symp. seed and seedbed ecology of rangeland plants.
Tucson, AZ.

Sandberg, David. 1980. Duff reduction by prescribed underburning in

Douglas-fir. USDA, For.Serv. Res. Pap. PNW-272. Pacif. Northw. For.
and Range Exp. Sta., Portland, OR. 18 p.

Sandberg, D. V. 1987. Progress in reducing emissions from prescribed

forest burning in western Washington and western Oregon, 13 p. IN
Proc. 23rd ann. meeting Air Pollution Control Assoc. Pacif. Northw.
Internatl. Sect., Pittsburgh, PA.

Sandberg, David V. and Frank N. Dost. 1990. Effects of prescribed

fire on air quality and human health, p. 191-218. IN John D. Walstad,
Steven R. Radosevich, and David V. Sandberg (eds.). Natural and
prescribed fire in Pacific Northwest forests. Oreg. State Univ. Press,
Corvallis.

Sandberg, D. V. and J. Peterson. 1987. Daily slash burn emissions

inventory design, part II. U.S. Environ. Prot. Agency. Fin. Rep. EPA
Contr. IAG EPA 83-291. Office Air Prog. - Reg. X, Seattle, WA.

Sandberg, D. V., J. M. Pierovich, D. G. Fox, and E. W. Ross. 1978.

Effects of fire on air: A state-of- knowledge review. USDA, For. Serv.
Gen. Tech. Rep. WO-9. Washington, D.C.

Sandberg, David V. and Franklin R. Ward. 1981. Predictions of fire

behavior and resistance to control for use with photo series for the
Douglas-fir - hemlock type and the coastal Douglas-fir - hardwood type.
USDA, For. Serv. Gen. Tech. Rep. PNW-116. Pacif. Northw. For. and
Range Exp. Sta., Portland, OR. 60 p.

SAS Institute. 1985. SAS user's guide: Statistics. SAS Institute, Inc.,
Cary, NC. 584 p.

Sauer, Ronald H. and Daniel W. Uresk. 1976. Phenology of steppe

plants in wet and dry years. Northw. Sci. 50(3):133-138.

Savory, Alan. 1988. Holistic resource management. Island Press,

Washington, D.C. 564 p.

Schier, G. A. 1972. Apical dominance in multishoot culture from aspen

roots. For. Sci. 18:147- 149.

Schier, George A. 1975. Deterioration of aspen clones in the middle

Rocky Mountains. USDA, For. Serv. Res. Pap. INT-170. Intermt. For.
and Range Exp. Sta., Ogden, UT. 14 p.

Schier, George A. 1983. Vegetative regeneration of gambel oak and

chokecherry from excised rhizomes. For. Sci. 29(3):499-502.

Schier, George A., John R. Jones, and Robert P. Winokur. 1985.

Vegetative regeneration, p. 29-33. IN Norbert V. DeByle and Robert P.
Winokur (eds.). Aspen: Ecology and management in the western United
States. USDA, For. Serv. Gen. Tech. Rep. RM-119. Rocky Mt. For. and
Range Exp. Sta., Ft. Collins, CO.

Schiff, Ashley L. 1962. Fire and water: Scientific heresy in the Forest
Service. Harvard Univ. Press, Cambridge, MA. 225 p.

Schmidt, Marcus. 1992. Great Basin live fuel moisture project. Bi-
weekly report. October 13, 1992. USDI, Bur. Land Manage., Nevada
State Office, Div. Fire and Aviat. Manage., Reno. Unpaged.

Schmidt, Wyman C. and James E. Lotan. 1980. Phenology of

common forest flora of the northern Rockies - 1928 to 1937. USDA, For.
Serv. Res. Pap. INT-259. Intermt. For. and Range Exp. Sta., Ogden,
UT. 20 p.

Schroeder, Mark J. and Charles C. Buck. 1970. Fire weather . . . A

guide for application of meteorological information to forest fire control
operations. USDA, For. Serv. Agric. Handb. 360. 229 p.

Scott, Douglas. 1987. Personal conversation with Dr. Richard Hanes.

Seabloom, Robert W., Rodney D. Sayler, and Stanley A. Ahler.

1991. Effects of prairie fires on archaeological artifacts. Park Science
11(1): 1-3.

Sestak, M. L. and A. R. Riebau. 1988. SASEM: Simple approach

smoke estimation model. USDI, Bur. Land Manage. Tech. Note 382.
BLM/YA/PT-88/003 + 7300. Serv. Center, Denver, CO. 31 p.

Sestak, M. L., A. R. Riebau, and M. Matthews. 1991. A tiered

approach to smoke management on government lands, p. 470-477. IN
Patricia L. Andrews and Donald F. Potts (eds.). Proc. 11th Conf. on Fire
and For. Meteor. Soc. of Amer. For.

Settergren, Carl D. 1969. Reanalysis of past research on effects of fire

on wildland hydrology. Coll. Agric. Exp. Sta. Res. Bull. 954. Univ.
Missouri, Columbia. 16 p.

Shafizadeh, Fred, Peter P. S. Chin, and William F. DeGroot. 1977.

Effective heat content of green forest fuels. For. Sci. 23(1):81-87.

Sharrow, S. H. and H. A. Wright. 1977. Proper burning intervals for

tobosagrass in west Texas based on nitrogen dynamics. J. Range
Manage. 30:343-346.

Shearer, Raymond C. 1975. Seedbed characteristics in western larch

forests after prescribed burning. USDA, For. Serv. Res Pap. INT-167.
Intermt. For. and Range Exp. Sta., Ogden, UT.
26 p.

Short, H. L. 1982. Techniques for structuring wildlife guilds to evaluate

impacts on wildlife communities. USDI, Fish and Wildl. Serv. Spec. Rep.
244.

Short, H. L. 1983. Wildlife guilds in Arizona desert habitats. USDI, Bur.

Land Manage. Tech. Note 362. Serv. Center, Denver, CO. 258 p.

Simard, Albert J. 1968. The moisture content of forest fuels - I. A

review of the basic concepts. For. Fire Res. Inst. Info. Rep. FF-X-14.
Canada Dept. of Forest. and Rural Developm., Ottawa, ON. 47 p.

Simard, Albert J., James E. Eenigenburg, and Richard W. Blank.

1984. Predicting fuel moisture in jack pine slash: A test of two systems.
Can. J. For. Res. 14: 68-76.

Simard, A. J. and W. A. Main. 1982. Comparing methods of predicting

jack pine slash moisture. Can. J. For. Res. 12: 793-802.

Smith, Graham W., Nicholas C. Nydegger, and Dana L. Yensen.

1984. Passerine bird densities in shrubsteppe vegetation. J. Field Bio.
55:261-264.

Smith, M. A. 1981. Prescribed burning: Effective control of sagebrush

in Wyoming. Wyo. Agric. Exp. Sta. Bull. No. RJ-165. Univ. of Wyo.,
Laramie. 12 p.

Smith, M. A., J. L. Dodd, and J. D. Rogers. 1985. Prescribed burning

on Wyoming rangeland. Agric. Ext. Serv. Bull. B-810, Univ. of Wyo.,
Laramie. 25 p.

Smith, Michael A., Henry A. Wright, and Joseph L. Schuster. 1975.

Reproductive characteristics of redberry juniper. J. Range Manage.
28(2):126-128.

Smith, S. D., S. C. Bunting, and M. Hironaka. 1986. Sensitivity of

frequency plots for detecting vegetation change. Northw. Sci. 60:279-
286.
Soil Survey Staff. 1975. Soil taxonomy. USDA Handbk. 436. Sup.
Doc., U.S. Govt. Print. Off., Washington, D.C. 754 p.

Sprent, Janet I. 1987. The ecology of the nitrogen cycle. Cambridge

Univ. Press, Cambridge, GB. 151 p.

Stanton, Frank W. 1975. Determining recovery potential of burned

plants following range fire. Rangeman's Journal 2(5): 152.

Starkey, Edward, E. 1985. Impact of fire on nongame wildlife, p. 48-51.

IN Ken Sanders and Jack Durham (eds.). Rangeland fire effects: A
symposium. USDI, Bur. Land Manage., Idaho

State Office, Boise.

StatSoft, Inc. 1987. CSS (Complete Statistical System). StatSoft, Inc.,

Tulsa, OK. 2 Vols., 1130 and 840 p.

Stein, William I. 1986. Regeneration outlook on BLM lands in the

Siskiyou Mountains. USDA, For. Serv. Res. Pap. PNW-349. Pacif.
Northw. For. and Range Exp. Sta., Portland, OR. 104 p.

Stickney, Peter F. 1986. First decade plant succession following the

Sundance forest fire, northern Idaho. USDA, For. Serv. Gen. Tech.
Rep. INT-197. Intermt. Res. Sta., Ogden, UT.

26 p.

Stone, E. C. and G. Juhren. 1953. The effects of fire on the

germination of the seed of Rhus ovata Wats. Amer. J. Bot. 38:368-372.

Striffler, W. D. and E. W. Mogren. 1971. Erosion, soil properties, and

revegetation following a severe burn in the Colorado Rockies, p. 25-36.
IN C. W. Slaughter, Richard J. Barney, and G. M. Hanson (eds.). Fire in
the northern environment - A symposium. USDA, For. Serv. Pacif.
Northw. For. and Range Exp. Sta., Portland, OR.

Sturges, David L. 1983. Long-term effects of big sagebrush control on

vegetation and soil water. J. Range Manage. 36:760-765.

Sweeney, J. R. 1956. Responses of vegetation to fire. Univ. Calif. Publ.

Bot. 28:143-231.

Switzer, Ronald R. 1974. The effects of forest fire on archaeological

sites in Mesa Verde National Park, Colorado. The Artifact 12(3):1-8.

Szaro, Robert C. 1986. Guild management: An evaluation of avian

guilds as a prediction tool. Environ. Manage. 10:681-688.

Tande, Gerald F. 1980. Interpreting fire history in Jasper National Park,

Alberta, p. 31-34. IN Marvin A. Stokes and John H. Dieterich (tech.
coord.). Proc. fire history workshop. USDA, For. Serv. Gen. Tech. Rep.
RM-81. Rocky Mt. For. and Range Exp. Sta., Ft. Collins, CO.

Tappeiner, John C., Timothy B. Harrington, and John D. Walstad.

1984. Predicting recovery of tanoak (Lithocarpus densiflorus) and
pacific madrone (Arbutus menziesii) after cutting or burning. Weed Sci.
32:413-417.

Thill, Donald C., K. George Beck, and Robert H. Callihan. 1984. The
biology of Bromus tectorum. Weed Sci. 32, Suppl. 1:7-12.

Thomas, Jack Ward (ed.). 1979. Wildlife habitats in managed forests:

The Blue Mountains of Oregon and Washington. USDA, For. Serv.
Agric. Handbk. No. 553. Sup. Doc. G.P.O. Washington, D.C. 512 p.

Thomas, Jack Ward, Chris Maser, and Jon E. Rodiek. 1979. Edges.
17 p. IN Wildlife habitats in managed rangelands - The Great Basin of
southeastern Oregon. USDA, For. Serv. Gen. Tech. Rep. PNW-85.
Pacif. Northw. For. and Range Exp. Sta., Portland, OR.

Thurston, Robert V., Rosemarie C. Russo, Carlos M. Fetterolf, Jr.,

Thomas A. Edsall, and Yates M. Barber, Jr. 1979. A review of the
EPA Red Book: Quality criteria for water. Amer. Fisheries Soc.,
Bethesda, MD. 313 p.

Tiedemann, Arthur R. 1987. Combustion losses of sulfur from forest

foliage and litter. For. Sci. 33:216-223.

Tiedemann, A. R., W. P. Clary, and R. J. Barbour. 1987.

Underground systems of gambel oak (Quercus gambelii) in central
Utah. Amer. J. Bot. 74(7):1065-1071.
Tiedemann, Arthur R., Carol E. Conrad, John H. Dieterich, James
W. Hornbeck, Walter F. Megahan, Leslie A. Viereck, and Dale D.
Wade. 1979. Effects of fire on water: A state-of-knowledge review.
USDA, For. Serv. Gen. Tech. Rep. WO-10. Washington, D.C. 28 p.

Tisdale, Samuel L. and Werner L. Nelson. 1975. Soil fertility and

fertilizers, 3rd ed. Macmillan and Sons., Inc. New York. 694 p.

Tomback, Diana F. 1986. Post-fire regeneration of Krummholz

whitebark pine: A consequence of nutcracker seed caching. Madrono
33(2):100-110.

Trappe, James M. 1981. Mycorrhizae and productivity of arid and

semiarid rangelands, p. 581-599. IN Advances in food producing
systems for arid and semi-arid lands. Academic Press, Inc., New York.

Traylor, Diane. 1981. Effects of La Mesa Fire on Bandelier's cultural

resources, p. 97-102. IN La Mesa fire symp. USDOE, Los Alamos Nat.
Lab., Los Alamos, NM.

Trevett, M. F. 1956. Observation on the decline and rehabilitation of

lowbush blueberry fields. Maine Agric. Exp. Sta. Misc. Publ. 626. Univ.
Maine, Orono. 21 p.

Trlica, M. J. 1977. Distribution and utilization of carbohydrate reserves

in range plants, p. 73-96. IN Ronald E. Sosebee, and 9 others.
Rangeland plant physiology. Soc. for Range Manage. Denver, CO.

Turner, Frederick B. and David C. Randall. 1987. The phenology of

desert shrubs in southern Nevada. J. Arid Environ. 13:119-128.

Urness, Phillip J. 1985. Managing lodgepole pine ecosystems for

game and range values, p. 297-304. IN David W. Baumgartner, Richard
G. Krebill, James T. Arnott, and Gordon F. Weetman (comp. and ed.).
Symp. proc.: Lodgepole pine, the species and its management. Coop.
Exten. Serv., Wash. State Univ., Pullman.

USDA-Forest Service, and Johns Hopkins University. 1989. The

effects of forest fire smoke on firefighters: A comprehensive study plan.
Third draft. Prepared for: Congressional Committee on Appropriations
for Title II-Related Agencies and the National Wildfire Coordinating
Group. Intermt. Res. Sta., Fire Chem. Res. Work Unit, Missoula, MT,
and School of Hygiene and Public Health, Johns Hopkins Univ.,
Baltimore, MD. 84 p. plus appen.

USDD-Corps of Engineers. 1989. Effects of forest fires and burn

programs on archaeological resources. Corps of Engin., Waterways
Exp. Sta., Archaeological Sites Protection and Preserv. Notebook,
Tech. Note I-8. Vicksburg, MS.

USDI-Bureau of Land Management, 1985a. Rangeland monitoring -

Trend studies. Tech. Ref. 4400-4. Bur. Land Manage., Serv. Center,
Denver, CO. 130 p.

USDI-Bureau of Land Management. 1985b. Emergency fire

rehabilitation, Bureau Manual Handbook H-1742-1. Bur. Land Manage.,
Washington, D.C. 17 p.

USDI-Fish and Wildlife Service. 1989. Protection proposed for the

northern spotted owl. Endangered Species Tech. Bull. Vol. XIV No. 7.
12 p.

USDI-National Park Service. 1992. Fire monitoring handbook. Natl.

Park Serv., Western Region. San Francisco, CA. 134 p. plus
appendices.

Van Wagner, C. E. 1977. Conditions for the start and spread of crown
fire. Can. J. For. Res. 7:23- 34.

Van Wagner, C. E. 1979. A laboratory study of weather effects on

drying rate of jack pine litter. Can. J. For. Res. 9: 267-275.

Viereck, L. A. and M. J. Foote. 1979. Permafrost, p. 17-21. IN L. A.

Viereck, and C. T. Dyrness (eds.). Ecological effects of the Wickersham
Dome fire near Fairbanks, Alaska. USDA, For. Serv. Gen. Tech. Rep.
PNW-90. Pacif. Northw. For. and Range Exp. Sta., Portland, OR.

Viereck, Leslie A. and Linda A. Schandelmeier. 1980. Effects of fire

in Alaska and adjacent Canada - A literature review. USDI, Bur. Land
Manage. Tech. Rep. 6. Alaska State Office, Anchorage. 124 p.
Viney, Neil R. 1991. A review of fine fuel moisture modelling. Intl. J.
Wildl. Fire 1(4): 215-234.

Vogl, Richard J. 1978. A primer of ecological principles - Book one.

Pyro Unlimited Pub., Cypress, Calif. 172 p.

Volland, Leonard A. and John D. Dell. 1981. Fire effects on Pacific

Northwest forest and range vegetation. USDA, For. Serv. Pacif. Northw.
Reg., Range Manage. and Aviat. and Fire Manage., Portland, OR. 23 p.

Wade, Dale D. 1986. Linking fire behavior to its effects on living plant
tissue. IN Proc. Ann. Conv. Soc. Amer. Foresters. Birmingham, AL.
Unpaged.

Wade, Dale D. and R.W. Johansen. 1986. Effects of fire on southern

pine: Observations and recommendations. USDA, For. Serv. Gen.
Tech. Rep. SE-41. Southea. For. Exp. Sta., Asheville, NC. 14 p.

Wagle, R. F. 1981. Fire: Its effects on plant succession and wildlife in

the southwest. Univ. of Ariz., Tucson. 82 p.

Wagle, R. F. and J. H. Kitchen, Jr. 1972. Influence of fire on soil

nutrients in a ponderosa pine type. Ecology 53:118-125.

Ward, Franklin R. and David V. Sandberg. 1981a. Predictions of fire

behavior and resistance to control for use with photo series for the
Sierra mixed conifer type and the Sierra true fir type. USDA, For. Serv.
Gen. Tech. Rep. PNW-114. Pacif. Northw. For. and Range Exp. Sta.,
Portland, OR. 48 p.

Ward, Franklin R. and David V. Sandberg. 1981b. Predictions of fire

behavior and resistance to control for use with photo series for the
ponderosa pine type, ponderosa pine and associated species type, and
lodgepole pine type. USDA, For. Serv. Gen. Tech. Rep. PNW-115.
Pacif. Northw. For. and Range Exp. Sta., Portland, OR. 46 p.

Ward, Peter. 1968. Fire in relation to waterfowl habitat of the Delta

marshes, p. 255-268. IN Proc. Tall Timbers Fire Ecol. Conf. No. 8. Tall
Timbers Res. Sta., Tallahassee, FL.

Watt, Kenneth E. F. 1972. Man's efficient rush toward deadly dullness,

p. 358-366. IN Alan Ternes (ed.). Ants, Indians and little dinosaurs. Am.
Museum of Natur. Hist., Washington, D.C.

Wecker, Stanley C. 1964. Habitat selection. Sci. Amer. 211:109-116.

Welch, Pat and Tirzo Gonzalez. 1982. Research design: Prescribed

burn impact evaluation upon cultural resources, LMDA and Thing
Mountain chaparral management projects. Manuscript. USDI, Bur. Land
Manage., El Centro, CA. 8 p.Wells, Carol G., Ralph E. Campbell,
Leonard F. DeBano, Clifford E. Lewis, Richard L. Frederiksen, E.
Carlyle Franklin, Ronald C. Froelich, and Paul H. Dunn. 1979.
Effects of fire on soil: A state-of-knowledge review. USDA, For. Serv.
Gen. Tech. Rep. WO-7. Washington, D.C. 34 p.

Wendel, George W. and Theodore G. Storey. 1962. Seasonal

moisture fluctuations in four species of pocosin vegetation. USDA, For.
Serv. Sta. Pap. No. 147. Southeast. For. Exp. Sta., Asheville, NC. 9 p.

West, Neil E. 1968. Rodent-influenced establishment of ponderosa

pine and bitterbrush seedlings in central Oregon. Ecology 49(5):1009-
1011.

West, Neil E. and M. A. Hassan. 1985. Recovery of sagebrush-grass

vegetation following wildfire. J. Range Manage. 38:131-134.

West, N. E. and R. W. Wein. 1971. A plant phenological index

technique. BioScience. 21:116-117.

Whitford, Walter G. 1986. Decomposition and nutrient cycling in

deserts, p. 93-117. IN Walter G. Whitford (ed.). Pattern and process in
desert ecosystems. Univ. New Mex. Press, Albuquerque.

Whittaker, Robert H. 1975. Communities and ecosystems. MacMillan

Publishing. Co., New York. 385 p.

Wiens, John A., John T. Rotenberry, and Beatrice Van Horne. 1986.
A lesson in the limitations of field experiments: Shrubsteppe birds and
habitat alteration. Ecology 67:365-367.

Wilcox, B. A. 1980. Insular ecology and conservation, p. 95-117. IN M.

E. Soule and B. A. Wilcox (eds.). Conservation biology: An evolutionary -
ecological perspective. Sinauer Associates, Sunderland, MA.

Willson, M. F. 1974. Avian community organization and habitat

structure. Ecology 55:1017-1029.

Wilson, Edward O. 1985. The biological diversity crises. BioScience

35:700-706.

Wolf, Michael L. and Joseph A. Chapman. 1987. Principles of

furbearer management, p. 101-112. IN Milan Novak, James A. Baker,
Martyn E. Obbard, and Bruce Malloch (eds.). Wild furbearer
management and conservation in North America. Ontario Trappers
Assoc., ON, Canada.

Wright, Henry A. 1970. A method to determine heat-caused mortality in

bunchgrasses. Ecology 51(4):582-587.

Wright, Henry A. 1971. Why squirreltail is more tolerant to burning than

needle-and-thread. J. Range Manage. 24(4):277-284.

Wright, Henry A. and Arthur W. Bailey. 1982. Fire ecology, United

States and southern Canada. John Wiley and Sons, Wiley-Interscience
Publication, New York. 501 p.

Wright, Henry A. and Carlton M. Britton. 1982. Fire in range and arid
lands. Lesson Plan. IN Advanced Fire Management, Natl. Adv. Resour.
Tech. Center, Marana, AZ.

Wright, Henry A., Francis M. Churchill, and W. Clark Stevens. 1976.

Effect of prescribed burning on sediment, water yield, and water quality
from dozed juniper lands in central Texas. J. Range Manage. 29:294-
298.

Wright, Henry A. and James O. Klemmedson. 1965. Effect of fire on

bunchgrasses of the sagebrush- grass region in southern Idaho.
Ecology 46(5):680-688.

Wright, Henry A., Leon F. Neuenschwander, and Carlton M. Britton.

1979. The role and use of fire in sagebrush-grass and pinyon-juniper
plant communities. USDA, For. Serv. Gen. Tech. Rep. INT-58. Intermt.
For. and Range Exp. Sta., Ogden, UT. 48 p.
Wright, H. E., Jr. 1981. The role of fire in land/water interactions, p.
421-444. IN H. A. Mooney, T. M. Bonnicksen, N. L. Christensen, J. E.
Lotan, and W. A. Reiners (eds.). Proc. conf. - Fire regimes and
ecosystem properties. USDA, For. Serv. Gen. Tech. Rep. WO-26.
Washington, D.C.

Wyant, James G., Richard D. Laven, and Philip N. Omi. 1983. Fire
effects on shoot growth characteristics of ponderosa pine in Colorado.
Can. J. For. 13:620-625.

Yarie, J. 1983. Environmental and successional relationships of the

forest communities of the Porcupine River drainage, interior Alaska.
Can J. For. Res. 13(5):721-728.

Yeo, Jeffrey J. 1981. The effects of rest-rotation grazing on mule deer

and elk populations inhabiting the Herd Creek allotment, East Fork
Salmon River, Idaho. M.S. thesis, Univ. Idaho, Moscow. 119 p.

Yoakum, James D. 1979. Habitat improvement, p. 132-139. IN Richard

D. Teague, and Eugene Decker (eds.). Wildlife conservation: Principles
and practices. The Wildl. Soc. Washington, D.C.

Yoakum, James D. 1980. Habitat management guides for the

American pronghorn antelope. USDI, Bur. Land Manage. Tech. Note
347. Serv. Center, Denver, CO. 74 p.

Young, J. A. and R. A. Evans. 1978. Population dynamics after

wildfires in sagebrush- grasslands. J. Range Manage. 31: 283-289.

Young, Richard P. 1986. Fire ecology and management of plant

communities of Malheur National Wildlife Refuge, southeastern Oregon.
Ph.D. thesis. Oreg. State Univ. 169 p.

Youngberg, C. T. 1981. Organic matter of forest soils, p. 137-144. IN

Paul E. Heilman, Harry W. Anderson, and David M. Baumgartner (eds.).
Forest soils of the Douglas-fir region. Wash. State Univ. Coop. Ext.
Serv., Pullman.

Zasada, John C. 1971. Natural regeneration of interior Alaska forests -

Seed, seedbed, and vegetative reproduction considerations, p. 231-
246. IN C. W. Slaughter, Richard J. Barney, and G. M. Hansen (eds.).
Fire in the northern environment - A symposium. USDA, For. Serv.
Pacif. Northw. For. and Range Exp. Sta., Portland, OR.

Zasada, John C. 1985. Production, dispersal, and germination of white

spruce and paper birch and first-year seedling establishment after the
Rosie Creek fire, p. 34-37. IN Glenn P. Juday, and C. Theodore
Dyrness (eds.). Early results of the Rosie Creek fire research project,
1984. Agric. and For. Exp. Sta., Misc. Pub. 85-2. Univ. of Alaska,
Fairbanks.

Zasada, John C. 1986. Natural regeneration of trees and tall shrubs on

forest sites in interior Alaska, p. 45-73. IN K. Van Cleve, F. S. Chapin,
III, P. W. Flanagan, L. A. Viereck, and C. T. Dyrness (eds.). Forest
ecosystems in the Alaskan taiga. Springer-Verlag. New York.

Zasada, John C. 1989. Personal conversation with Melanie Miller.

Zasada, John C., M. Joan Foote, Frederick J. Deneke, and Robert

H. Parkerson. 1978. Case history of an excellent white spruce cone
and seed crop in interior Alaska: Cone and seed production,
germination, and seedling survival. USDA, For. Serv. Gen. Tech. Rep.
PNW-65. Pacif. Northw. For. and Range Exp. Sta., Portland, OR. 53 p.

Zasada, John C., Rodney A. Norum, Robert M. Van Veldhuizen, and

Christian E. Teutsch. 1983. Artificial regeneration of trees and tall
shrubs in experimentally burned upland black spruce/feather moss
stands in Alaska. Can. J. For. Res. 13(5):903-913.

Zasada, John C. and George A. Schier. 1973. Aspen root suckering

in Alaska: Effect of clone, collection date, and temperature. Northw. Sci.
47(2):100-104.

Zimmerman, G. T. 1988. Monitoring fire behavior and characteristics.

Lesson plan. IN Fire Effects on Public Lands. Boise Interag. Fire
Center, Boise, ID.

Zimmerman, G. Thomas. 1990. Personal conversation with Melanie

Miller.

Zschaechner, Greg A. 1985. Studying rangeland fire effects: A case

study in Nevada, p. 66-84. IN Ken Sanders and Jack Durham (eds.).
Rangeland fire effects: A symposium. USDI, Bur. Land Manage., Idaho
State Office, Boise.

1.

1The two publications by Frank Freese are no longer printed by the

government, but have been reprinted under one cover by O.S.U.
Bookstores, Inc., Oregon State Univ., Corvallis, 97331.

This page was last modified 05/31/01

|Disclaimer| | Privacy| | Copyright| |USFWS Main Page| |Webmaster|

 National Wildfire Coordinating Group

Fire Effects Guide

This page was last modified 05/31/01
|Disclaimer| | Privacy| | Copyright| |Webmaster|

Home CONTRIBUTORS
Preface
Objectives Loren D. Anderson, Wildlife Biologist, Bureau of Land Management,
Fire Behavior Salmon District Office, Salmon, Idaho
Fuels
Air Quality
Dr. Robert G. Clark, Chief, Branch of Fire and Aviation, Bureau of
Soils & Water
Plants
Land Management, Idaho State Office, Boise, Idaho
Wildlife
Cultural Res. Jean Findley, Botanist, Bureau of Land Management, Vale District
Grazing Mgmt. Office, Vale, Oregon
Evaluation
Data Analysis Dr. Richard C. Hanes, Archaeologist, Bureau of Land Management,
Computer Oregon State Office, Portland, Oregon
Soft.
Glossary Larry Mahaffey, Smoke and Fuels Specialist, Bureau of Land
Bibliography Management, Division of Fire and Aviation Policy and Management,
Contributions Boise, Idaho

Melanie Miller, Fire Ecologist, Bureau of Land Management, Division of

Fire and Aviation Policy and Management, Boise, Idaho

Ken Stinson, District Range Conservationist, Bureau of Land

Management, Worland District Office, Worland, Wyoming

Dr. G. Thomas Zimmerman, Fire Management Specialist, National

Park Service, Branch of Fire, Boise, Idaho

CHAPTER CREDITS
Introduction - Dr. Bob Clark and Melanie Miller

I. Development of Objectives - Dr. Tom Zimmerman

II. Fire Behavior and Characteristics - Melanie Miller

III. Fuels - Melanie Miller

IV. Air Quality - Larry Mahaffey and Melanie Miller

V. Soils, Water, and Watersheds - Dr. Bob Clark

VI. Plants - Melanie Miller and Jean Findley

VII. Terrestrial Wildlife and Habitat - Loren Anderson

VIII. Cultural Resources - Dr. Richard C. Hanes

IX. Prefire and Postfire Grazing Management - Ken Stinson

X. Documentation and Evaluation - Ken Stinson and Melanie Miller

XI. Data Analysis - Dr. Bob Clark

XII. Computer Software - Melanie Miller

EDITOR

Melanie Miller

ACKNOWLEDGMENTS

This Guide was initiated to complete the process that began with the
development and staging of two sessions of the Bureau of Land
Management workshop, Fire Effects on Public Lands, in 1988 and
1989. We would like to give special recognition to Bill Harkenrider, who
died in a plane crash while travelling to the National Interagency Fire
Center to continue his contributions toward development and
finalization of the Fire Effects Workshop.

We wish to recognize the following individuals who, in addition to the

Guide authors, made substantial contributions to the workshop or the
Guide:

Sandy Bowers, Bureau of Land Management, Denver, Colorado

William Brookes, Bureau of Land Management, Portland, Oregon

Dr. Jim Brown, U. S. Forest Service, Missoula, Montana

Jim Cafferty, Bureau of Land Management, Boise, Idaho

Bill Clark, National Park Service, Boise, Idaho

Stan Coloff, National Biological Survey, Washington, D. C.

Chet Conard, Bureau of Land Management (Retired)

Dr. Norb DeByle, Forest Service (Retired)

Gardner Ferry, Bureau of Land Management, Boise, Idaho

Bill Fischer, U. S. Forest Service (Retired)

Stan Frazier, Bureau of Land Management, Oregon State Office

George Gruell, U. S. Forest Service (Retired)

Mick Harrington, U. S. Forest Service, Missoula, Montana

John Key, Bureau of Land Management, Riverside, California

Bob Kindschy, Bureau of Land Management, Vale, Oregon

Steve Lent, Bureau of Land Management, Prineville, Oregon

Roy Montgomery, Bureau of Land Management, Portland, Oregon

Tom Nichols, National Park Service, San Francisco, California

Dr. Rod Norum, National Park Service (Retired)

Dr. Phil Omi, Colorado State University, Fort Collins, Colorado

Mike Pellant, Bureau of Land Management, Boise, Idaho

Steve Petersburg, National Park Service, Dinosaur National Monument,

Colorado

Phil Range, Bureau of Land Management, Boise, Idaho

Al Riebau, National Biological Survey, Fort Collins, Colorado

Kirk Rowdabaugh, Bureau of Land Management, Phoenix, Arizona

Cheryl Ruffridge, Bureau of Land Management, Las Vegas, Nevada

Dr. Kevin Ryan, U. S. Forest Service, Missoula, Montana

Dr. Byron Thomas, Bureau of Land Management (Retired)

Dave Wolf, Bureau of Land Management, Las Vegas, Nevada

Jim Yoakum, Bureau of Land Management (Retired)

Greg Zschaechner, Bureau of Land Management, Missoula, Montana

This page was last modified 05/31/01

|Disclaimer| | Privacy| | Copyright| |USFWS Main Page| |Webmaster|