12.

1 The Role of Biota in Geomorphology: Ecogeomorphology

DR Butler, Texas State University-San Marcos, San Marcos, TX, USA
CR Hupp, US Geological Survey, Reston, VA, USA
r 2013 Elsevier Inc. All rights reserved.

12.1.1 Introduction to Ecogeomorphology 1

12.1.2 Chapter Sequence and Topics in this Volume 1
References 4

12.1.1 Introduction to Ecogeomorphology papers (Naylor, 2005; Corenblit et al., 2008; Corenblit and
Steiger, 2009; Viles et al., 2008; Phillips, 2009) have continued
Studies at the interface of ecology and geomorphology date this trend of advancements and extensions of the field of
nearly back to origins of the word ‘ecology’ first coined by Ernst ecogeomorphology.
Haeckel circa 1870 (Odum, 1971). Early ecologists clearly Paralleling the rise of ecogeomorphology/biogeomor-
documented the value of relating geomorphic form and pro- phology has been the rise of the allied subfield of ecosystem
cess as partial explanation of species distributions (e.g., Cowles, engineering (Jones et al., 1994; Buchman et al., 2007; Jones
1899), which fundamentally defines the field, ecology. Charles and Gutiérrez, 2007). Not all aspects of ecosystem engineering
Darwin’s final published work (Darwin, 1881), ‘The Formation deal, however, with the geomorphic environment, but none-
of Vegetable Mould through the Action of Worms With Ob- theless similarities exist among many of the topics examined
servation of Their Habits,’ examined the role of earthworms as a in ecogeomorphology and in ecosystem engineering. The 2011
bioturbational agent. These early examples from such giants in Binghamton Geomorphology Symposium, in October 2011,
the field as Darwin and Cowles did not, however, lead to an examines the topic of Zoogeomorphology and Ecosystem
immediate expansion in studies examining the interface of Engineering, and seeks to bring together scientists and prac-
ecology and geomorphology. Since the end of the nineteenth tioners from these closely related fields.
century and before the very late twentieth century, studies ex- This volume is loosely partitioned among phytogeo-
plicitly bridging the fields of ecology and geomorphology were, morphic-, biogeochemical-, and zoogeomorphic-oriented
instead, sparse but included the important papers of Olson chapters. In conformity with the title, our focus tends to be on
(1958) and Hack and Goodlett (1960). species distributions as limited by geomorphic forms and
Much more recently, a wide array of papers, volumes, and processes; with a secondary focus on the distinctive geo-
symposia has attempted to squarely integrate the fast emer- morphic impacts of animals on the landscape. The likelihood
ging, now increasingly distinct field of biogeomorphology as of a given species vigorously sustained on particular landforms
termed by Viles (1988). Ecogeomorphology is essentially is a function of suitability of the area for establishment and
synonymous with biogeomorphology and perhaps is better growth and the ambient environmental conditions of the area
reflective of the field (Hupp et al., 1995; Wheaton et al., 2011). that permit persistence at least until reproductive age. The
The editors of the present volume have had a relatively long distribution patterns may be limited by the tolerance of a
appreciation for the body of work relating biota and geo- species for specific disturbance or stress regimes, as well as by
morphic form and process (Butler, 1995; Hupp et al., 1995) tolerance for other more diffuse interactions including com-
and hope our efforts here continue to underscore the scientific petition, for which one set of factors drives the limits at one
value of ecogeomorphology. extreme, another set drives the other extreme (Hupp and
Ecogeomorphology (or biogeomorphology) as a sym- Bornette, 2003; Bornette et al., 2008).
posium topic began in 1990 with the British Geomorphic A diverse array of chapters from authors in several coun-
Research Group (Thornes, 1990). The following two decades tries, who have demonstrated the utility and need for inter-
have experienced a tremendous increase in the number of disciplinary science have been assembled. Further, we have
ecogeomorphic publications including symposia proceedings tried to include authors relatively early in their careers to
(Hupp et al., 1995; Bennett and Simon, 2004; Hession et al., provide a fresh aspect to this volume as well as authors at the
2010; Wheaton et al., 2011). This increase in ecogeomorphic other end of their careers to provide a seasoned, long-term
publications was documented by Hession et al. (2010) where perspective. Each chapter was submitted by the author(s) to
they searched in the Web of Science using the keywords the editors, fully refereed, and revised. The chapters offer, we
‘vegetation and geomorphology’ and found 7 publications hope, a strong and diverse snapshot of the international field
between 1977–86, 95 between 1987–96, 425 between of ecogeomorphology.
1997–2006, and 186 in just the three years 2007–09. Recent

12.1.2 Chapter Sequence and Topics in this Volume

Butler, D.R., Hupp, C.R., 2013. The role of biota in geomorphology:
Ecogeomorphology. In: Shroder, J. (Editor in chief), Butler, D.R.,
Hupp, C.R. (Eds.), Treatise on Geomorphology. Academic Press, San Diego, Species distributions are controlled by available habitats
CA, vol. 12, Ecogeomorphology, pp. 1–5. (ecology), which are constrained by physical attributes such as

Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00316-X 1

2 The Role of Biota in Geomorphology: Ecogeomorphology

landform, grain size, and hydraulics (geomorphology). places, critical habitat for aquatic communities. They highlight
Jacobson (see Chapter 12.2) shows that physical habitat is a the effects of large wood on channel morphology and fluvial
fundamental influence on riverine ecosystem structure and processes and the ecological and societal consequences
function. He uses habitat dynamics studies to illustrate the of wood in rivers. They finish with a discussion of the in-
value of detailed hydrogeomorphic parameterization (spatial corporation and implications of large wood in terms of river
and temporal) in determining the distribution of riverine management and restoration.
biota including the high ecological-interest pallid sturgeon, European rivers have experienced millennia of human
bigheaded carp, sedentary freshwater mussels, cottonwood disturbance both catchment-wide and in channel. The impacts
communities, and shorebirds. Riverine habitat dynamics are of river morphology and riparian ecosystems may have cul-
categorized into hydrodynamics and morphodynamics to minated in the twentieth century, particularly during recent
construct an innovative, interpretive framework for viewing decades, however major changes are predicted to continue
habitat regime much like flow regime. Understanding habitat in the near future through climate change. Rinaldi et al. (see
dynamics is increasing in importance as society shifts goals Chapter 12.4) discuss progress in understanding, quantifying,
toward restoration of riverine ecosystems and this type of and modeling channel processes emergent from recent Euro-
ecogeomorphic effort will be greatly valued toward this end. pean research on sediment transport, bank erosion, and the
Tree-ring studies (dendrochronology) have been particu- formation of various channel patterns. They present the major
larly useful in providing temporal information at scales not concepts linking the structure and function of river ecosystems
easily approached with other techniques. When combined and with fluvial processes from a European perspective. This is
with the dating of important geomorphic events (floods, slope followed by a discussion of the impacts of geomorphic pro-
failure) the study is termed dendrogeomorphology, which by cesses on connectivity, habitat heterogeneity, and a scale of
nature at once facilitates the understanding of both geo- flood disturbance. The authors strongly suggest that process-
morphic processes and woody plant ecology. Merigliano et al. oriented hydro-geomorphic approaches to reestablishing good
(see Chapter 12.10) review the use of tree rings to date flood ecological quality of European rivers including the mainten-
disturbance, channel change, and sediment deposition. They ance of biodiversity will be necessary for scientific river man-
emphasize semiarid areas of western US and include indepth agement and restoration.
discussions of riparian woody species establishment on de- Quantification of the relation between fluvial geomorphic
positional features. Distributional patterns and age of riparian processes and riparian vegetation patterns has been a relatively
trees are used to elucidate the development of fluvial landform recent research focus, largely beginning just a few decades ago.
and riparian ecosystem processes. Further, they show that as However, since then there was an almost exponential increase in
watershed area decreases and aridity increases, large floods the number of papers published on this topic. Bendix and Stella
have an increasingly greater impact on channel dynamics and (see Chapter 12.5) have prepared an extensive review of the
vegetation patterns. historical arc of research on ecogeomorphic interactions in ri-
The use of tree rings for dating geomorphic processes parian zone. They then report on an innovative, intriguing
(dendrogeomorphology) has advanced dramatically in recent examination of the past 20 years of published research in this
years, being extended from the fluvial landscape where it arena. Through a screening process based on quantification,
was largely focused at first to landscape types across the realm topic, and field orientation, the mass of literature was then coded
of geomorphic processes, climate types, and geographic by journal, region of study, biome, scale, environmental rela-
locations. Stoffel et al. (see Chapter 12.9) illustrate the growth tionships, and causal mechanisms. Among many interesting
in the field of dendrogeomorphology, and offer several ex- findings they note distinct research bias as to geographical areas
amples of how tree-ring analysis can now offer not only an- of study, vegetation versus geomorphic focus, location of re-
nual resolution of geomorphic processes, but intra-annual searchers, scale of study, journal outlet, and several other subjects.
precision. New techniques, especially the use of traumatic Quantification of roughness in fluvial systems has been,
resin ducts, have been broadly applied in the mountains and is still a challenging endeavor since the beginning of
of Europe. The chapter describes the application of tree rings quantitative geomorphic analyses in the mid twentieth cen-
to studies of snow avalanches, rockfalls, landslides, floods, tury, yet roughness remains a critical parameter for some of
earthquakes, wildfires, and several other processes. It illus- the most important fluvial hydraulic computations. In most
trates the breadth and diversity of applications in con- riverine systems, vegetation provides substantial roughness, in
temporary dendrogeomorphology and underlines the growing some places most. Research in this interaction is just begin-
potential to expand dendrogeomorphic research. ning. Hession and Curran (see Chapter 12.6) provide what
Studies of riparian large wood (previously termed large they believe is the first step in understanding the complex
woody debris, LWD) dynamics and impacts have progressed processes involved through investigation of feedbacks and
from nearly none to major efforts in just the past 40 years. linkages among channel flow, morphology, sediment dy-
Stream and river scientists now recognize large wood as a namics, and vegetation characteristics. They frame their dis-
significant structural and functional component of aquatic cussion through the location of vegetation as inchannel
ecosystems and critical to understanding fluvial geomorphic emergents and submergents, on banks, and on floodplains.
processes in forested watersheds. Le Lay et al. (see Chapter With each location, the influence of vegetation on roughness
12.3) provide a comprehensive international synthesis of our at the reach scale, hydraulics and turbulence, and sediment
knowledge of large wood characteristics, its effects on flow dynamics is discussed. They conclude with section on the
hydraulics, channel and valley form and evolution, storage complexities related to vegetation and fluvial processes and
and transfer of sediment and organic matter, and in some some of the research opportunities and challenges.
 The Role of Biota in Geomorphology: Ecogeomorphology 3

Early ecologists understood the need to document geo- vegetation, hydrogeomorphology, and nutrient biogeochem-
morphic form and process to explain plant species distri- istry. The defining characteristics of floodplain ecosystems are
butions. Although this relationship has been acknowledged determined by the many interactions among physical and
for over a century, with the exception of a few landmark biological processes. The chapter concludes stressing that the
papers, only the past few decades have experienced intensive valuable ecosystem services floodplains provide depends on
research on this interdisciplinary topic. Hupp and Osterkamp improved understanding and predictive models of interactive
(see Chapter 12.7) provide a summary of the intimate re- system controls and behavior, which will lead to better con-
lations between vegetation and geomorphic/process on hill- servation and restoration of this important ecosystem.
slopes and fluvial systems. These relations are separated into Peatlands cover 3% of the Earth’s surface, and play a sig-
systems (primarily fluvial) in dynamic equilibrium and those nificant role in water and carbon cycling. Evans notes that
that are in nonequilibrium conditions including the impacts peatlands have been relatively understudied in geomorph-
of various human disturbances affecting landforms, geo- ology in spite of their widespread distribution and import-
morphic processes, and interrelated, attendant vegetation ance, and provides a chapter that brings together the state of
patterns and processes. They finish with a conceptual model of knowledge on this topic. Evans examines the geomorphology
stream regime focusing on sediment deposition, erosion, and and hydrology of peatlands, and describes the potential for
equilibrium that can be expanded to organize and predict peatland degradation especially in permafrost regions as a
vegetation patterns and life history strategies. result of climatic warming. Peatland degradation in non-
Slope and streambank stability is affected by several im- permafrost environments is described in detail, and Evans
portant factors. Among them vegetation, where present, may notes the reduction in functionality of peatlands in associ-
represent a substantial influence. Mechanical strengthening of ation with their geomorphic degradation. He concludes with
geomorphic features may occur by soil reinforcement from an examination of the necessity for better management of the
roots, due to their tensile strength and their frictional and worldwide peatlands resource, particularly in the context of its
adhesional properties. Quantitative research on this topic has changing functionality resulting from climate change.
largely occurred in just the past two decades. Pollen-Bankhead Salt marshes and tidal flats are dynamic geomorphic and
et al. (see Chapter 12.8) provide a comprehensive review of this biogeochemical environments subject to degradation from
recent research with a focus on tensile strength and fiber-bundle human misuse. In this volume, Fagherazzi et al. offer two
models. They show that even with the successful modeling of (see Chapters 12.12 and 12.13) on these delicate coastal en-
root breakage, stretching, compression, and pullout several vironments. In their examination of salt marshes, they note
obstacles for root-reinforcement prediction remain. Current the critical necessity for coupling geomorphological and eco-
modeling efforts by the authors suggest, however, that roots can logical processes in this delicate environment, without which
be of significant importance in the stabilization of slopes and degradation and disappearance of the resource could occur.
streambanks affecting both the timing and magnitude of mass The authors present a broad overview of the feedbacks be-
failure and the geomorphic causes. tween biota and sedimentological processes in salt marshes,
Human alterations to the landscape such as dam con- including recent numerical models that have been utilized to
struction and stream channelization may lead to dramatic study the ecogeomorphic evolution of intertidal areas. They
changes to the fluvial environment and associated ecosystems. emphasize both plant and animal processes active in the salt
The results of channelization can be exacerbated by local marsh environment, and describe how some animal processes
geology and land use resulting in phytogeomorphic adjust- such as crab burrowing act to disrupt the sediment of the
ment including the formation of valley plugs and major shifts marshes, whereas other animals such as mussels and other
in vegetative composition. Pierce and King (see Chapter 12.14) bivalves stabilize marsh sediment by both slowing wave and
describe valley plugs, their formation in response to stream current velocities and binding sediment to the root mat. In
channelization, and vegetative responses. They show, in detail, their examination of tidal flats, the authors extend their
that altered stream regime (surface and subsurface hydrology examination of the interdependency of geomorphological and
and sedimentation) exert strong controls on species com- ecological processes into the more active and dynamic tidal
position, abundance, and successional processes. Given the and deltaic coastal environment.
considerable interest in restoring altered fluvial systems, the The intense heat produced by fire directly and indirectly
authors highlight the need for understanding valley affects both vegetation and geomorphic processes. In her
plug formation and consequences within a watershed context. chapter, Stine (see Chapter 12.15) focuses on the effects of fire
A discussion of various restoration approaches and examples on geomorphological processes (including erosion, runoff,
concludes the chapter. and weathering rates), and changes to soil properties and
Nutrient biogeochemistry is an emerging field with sub- hydrological processes. A variety of topics related to the
stantial real and potential implications for ecosystem manage- significance of fire as a geomorphic agent are addressed, in-
ment and restoration, yet its understanding is constrained by cluding soil hydrophobicity, infiltration, mineralization,
hydrogeomorphic and vegetative conditions. Floodplains offer and nutrients; surface erosion and mass movements; wea-
an opportunity to better understand hydrogeomorphic, vege- thering; runoff, streamflow and morphology; prehistorical
tative, and biogeochemical processes and their interactions, fire and geomorphology; and topographic influence on fire
feedbacks, and linkages. Noe (see Chapter 12.21) explores this behavior. Vegetation type and density, topographic features,
complex mosaic of relations along four dimensions: longi- and soil conditions may buffer, enhance, or influence the
tudinal, lateral, vertical, and temporal. He provides insightful patterns of geomorphic disturbance induced by fire on the
discussion of the dependencies in multiple directions among landscape.
4 The Role of Biota in Geomorphology: Ecogeomorphology

Zoogeomorphology is the study of the geomorphic effects Ants and termites can therefore be regarded as of major eco-
of animals, ranging from small invertebrates to large ver- geomorphological importance through their modifications of
tebrates such as elephants and bison. Butler et al. (see Chapter both soil and geomorphic processes.
12.16) present a chapter that provides a bridge into the re- The final chapter in this volume examines the hydrological
maining chapters in this volume by providing a general and geomorphological impacts of the two existing species of
overview of the geomorphic impacts of animals, including ecosystem engineers, Castor canadensis and Castor fiber, the
trampling, loading, digging/foraging for food, geophagy, North American and European beavers. Westbrook et al. (see
lithophagy, nest building, burrowing, mound building, and Chapter 12.20) examine the main hydrologic signature of
damming of streams by beaver (itself the specific focus of an beaver activities and how it varies with hydrogeomorphic
additional chapter by Westbrook et al. (see Chapter 12.18)). setting – along confined streams it is the pond formed up-
The authors offer a variety of examples of specific landforms stream of dams, along unconfined streams it is downstream
constructed or excavated by animals, as well as pathways both flooding on floodplains and terraces, and in preexisting wet-
direct and indirect by which animal activity impacts the geo- lands it is the formation of open-water bodies. Beaver dams
morphological landscape. The authors also suggest that par- moderate stream flows, increase surface water and riparian
ticularly in an era of rapidly changing climate, a fruitful groundwater storage, regulate hyporheic flows, and enhance
expansion of zoogeomorphological research could be under- evapotranspiration rates. Beavers also excavate canals on the
taken in ecotone environments such as at alpine treeline. margins of beaver ponds and create extensive burrow systems
When examining the role of both animals and plants as in riverbanks where damming is not possible. Bank burrows
geomorphic agents, it must be recalled that a great deal of are also common in beaver ponds. The authors also point out
geomorphic work is carried out at the microscopic level. Viles what is missing in the beaver hydrogeomorphological litera-
(see Chapter 12.17) illustrates that microbes are effective ture: clear linkages between affected hydrological processes
contributors to biogeochemical processes, make up a large and ecosystem functioning, especially at larger spatial and
part of global biomass and biodiversity, are found in virtually temporal scales.
all geomorphic environments, and as a consequence are of
great biogeomorphological significance. She describes how
microbes contribute to Earth surface processes through both References
bioerosive and bioprotective roles, and focuses on a variety of
extreme environments such as cold and hot arid environ- Bennett, S., Simon, A., 2004. Riparian Vegetation and Fluvial Geomorphology.
ments, rocky coasts and coral reefs, and ruiniform landscapes American Geophysical Union, Water Science and Application, Washington, DC,
vol. 8, 282 pp.
where the impacts and significance of microbial processes are
Bornette, G., Tabacchi, E., Hupp, C., Puijalon, S., Rostan, J.C., 2008. A model of
particularly apparent. The chapter concludes by posing a plant strategies in fluvial hydrosystems. Freshwater Biology 53, 1692–1705.
number of key questions concerning the overall impact and Buchman, N., Cuddington, K., Lambrinos, J., 2007. A historical perspective on
place of microbial processes in the field of geomorphology. ecosystem engineering. In: Cuddington, K., Byers, J.E., Wilson, W.G., Hastings, A.
The term ‘fossorial’ applies to animals that burrow, and (Eds.), Ecosystem Engineers – Plants to Protists. Elsevier, Amsterdam, pp. 25–46.
Butler, D.R., 1995. Zoogeomorphology. Cambridge University Press, Cambridge,
Butler et al. (see Chapter 12.18) provide an overview of a suite
239 pp.
of animals that impact the landscape through their burrowing Corenblit, D., Gurnell, A.M., Steiger, J., Tabacchi, E., 2008. Reciprocal adjustments
and denning activities. The authors note that a majority of the between landforms and living organisms: extended geomorphic evolutionary
zoogeomorphic research on fossorial animals has focused on insights. Catena 73, 261–273.
impacts such as soil erosion, slope failure or mass wasting Corenblit, D., Steiger, J., 2009. Vegetation as a major conductor of geomorphic
changes on the Earth surface: toward evolutionary geomorphology. Earth Surface
events, and biogeochemical alteration of soil. The transition Processes and Landforms 34, 891–896.
from quantifiable studies of animal geomorphology to the Cowles, H.C., 1899. The ecological relations of the vegetation on the sand dunes of
secondary impacts on ecosystems is a logical advance of the Lake Michigan. Botanical Gazette 27, 95–117, 167–202, 281–308, 361–391.
study of zoogeomorphology; however, such studies have Darwin, C., 1881. The Formation of Vegetable Mould, through the Action of
Worms, with Observations on Their Habitats. John Murray, London.
largely been lacking, and additional quantitative data on direct
Hack, J.T., Goodlett, J.C., 1960. Geomorphology and forest ecology of a mountain
geomorphic impacts created by fossorial animals are required region in the Central Appalachians. US Geological Survey Professional Paper
before such secondary impacts can be properly assessed and 347, Washington, DC.
contextualized. Hession, W.C., Wynn, T., Resler, L., Curran, J., 2010. Geomorphology and
A specific group of fossorial insects, ants and termites, are vegetation: interactions, dependencies, and feedback loops. Special issue.
Geomorphology 116, 203–205.
the focus of the chapter by Whitford and Eldridge (see Chapter Hupp, C.R., Bornette, G., 2003. Vegetation, fluvial processes and landforms in
12.19). Ants and termites are one of the most widespread insect temperate areas. In: Kondolf, M., Piegay, H. (Eds.), Tools in Geomorphology.
groups, and they occur on all continents except Antarctica. John Wiley and Sons, Chichester, UK, pp. 269–288.
Whitford and Eldridge illustrate how the soil-dwelling ants and Hupp, C.R., Osterkamp, W.R., Howard, A.D., 1995. Biogeomorphology – Terrestrial
and Freshwater Systems. Elsevier, Amsterdam, The Netherlands, 347 pp.
termites disturb surface and subsurface soils while constructing
Jones, C.G., Gutiérrez, J.L., 2007. On the purpose, meaning, and usage of the
vast nests and interconnecting tunnels. This soil movement has physical ecosystem engineering concept. In: Cuddington, K., Byers, J.E., Wilson,
major effects on soil turnover and development, clay mineral- W.G., Hastings, A. (Eds.), Ecosystem Engineers – Plants to Protists. Elsevier,
ogy, the retention and infiltration of water, and soil chemical Amsterdam, pp. 3–24.
properties. The effects of ant and termite activity appear to be Jones, C.G., Lawton, J.H., Shachak, M., 1994. Organisms as ecosystem engineers.
Oikos 69, 373–386.
greatest for structures that persist for many years, such as the Naylor, L.A., 2005. The contributions of biogeomorphology to the emerging
large mounds and termitaria that are constructed above-surface field of geobiology. Palaeogeography, Palaeoclimatology, Palaeoecology 219,
extensions of the more widespread nest burrows belowground. 35–51.
 The Role of Biota in Geomorphology: Ecogeomorphology 5

Odum, E.P., 1971. Fundamentals of Ecology, Third ed. W.B. Saunders Company, Viles, H., 1988. Biogeomorphology. Blackwell, Oxford, UK, 365 pp.
Philadelphia, USA, 574 pp. Viles, H.A., Naylor, L.A., Carter, N.E.A., Chaput, D., 2008. Biogeomorphological
Olson, J.S., 1958. Lake Michigan dune development 2. Plants as agents and tools disturbance regimes: progress in linking ecological and geomorphological
in geomorphology. Journal of Geology 66, 345–351. systems. Earth Surface Processes and Landforms 33, 1419–1435.
Phillips, J.D., 2009. Biological energy in landscape evolution. American Journal of Wheaton, J.M., Gibbins, C., Wainwright, J., Larsen, L., McElroy, B., 2011. Preface,
Science 309, 271–289. multiscale feedbacks in ecogeomorphology. Geomorphology 126, 265–268.
Thornes, J.B., 1990. Vegetation and Erosion. Wiley, Chichester, UK, 518 pp.

Biographical Sketch

Dr. David R Butler is the Texas State University System Regents’ Professor of geography, and a University Dis-
tinguished Professor at Texas State University-San Marcos where he has been on faculty since 1997. His research
interests are in the areas of mountain geomorphology, zoogeomorphology, biogeomorphology, and den-
drogeomorphology, focusing his work in the Glacier National Park region of Montana, USA. He has published
more than 150 refereed papers in journals and conference volumes, and more than 35 book chapters. He is the
author of ‘Zoogeomorphology – Animals as Geomorphic Agents’ (Cambridge University Press, 1995 and 2007),
and coeditor of ‘Tree Rings and Natural Hazards’ (Springer, 2010), ‘The Changing Alpine Treeline – The Example
of Glacier National Park, Montana’ (Elsevier, 2009), and ‘Mountain Geomorphology – Integrating Earth Systems’
(Elsevier, 2003). He has received the GK Gilbert Award for Excellence in Geomorphological Research from the
Geomorphology Specialty Group of the Association of American Geographers (AAG), and the Distinguished
Career Award and the Outstanding Recent Accomplishment Award from the AAG’s Mountain Geography Specialty
Group. David holds a BA and MSc in geography from the University of Nebraska at Omaha, and the PhD in
geography from the University of Kansas.

Dr. Cliff R Hupp is the Project Chief of the Vegetation and Hydrogeomorphology Project of the U.S. Geological
Survey, National Research Program in Reston, Virginia, where he has been since 1978. Presently, he is the Research
Advisor for the Geomorphology and Sediment Transport Group of the USGS National Research Program. He has
investigated fluvial geomorphology and riparian vegetation ecology in relation to landforms and hydrologic
processes for 33 years. Additional research includes studies on forested wetland biogeochemistry, channel evo-
lution, floodplain processes and forms, sedimentation dynamics, and carbon sequestration in riparian ecosystems
in the U.S. and Europe. He has published widely in refereed papers of journals, books, and symposia proceedings.
Cliff was a student of the late John T Hack at the George Washington University, where he received his doctorate in
1984 in plant ecology and geomorphology. He received his MS degree from George Mason University in plant
ecology (1979). Dr. Hupp is the 1993 recipient of the Ecological Society of America, WC Cooper Award for
outstanding research in physiographic ecology. He received the U.S. Department of Interior Superior Service
Award in 2006 and the Senior Research Award of the Association of Southeastern Biologists in 2010. He served as
section editor for the ESA journals ‘Ecology’ and ‘Ecological Monographs’ from 1999 until 2009.
 12.2 Riverine Habitat Dynamics
RB Jacobson, US Geological Survey, Columbia, MO, USA
Published by Elsevier Inc.

12.2.1 Introduction 6
12.2.1.1 Some Definitions and Concepts 6
12.2.1.1.1 Habitat 6
12.2.1.1.2 Spatial characteristics of habitat 7
12.2.1.1.3 Habitat dynamics 7
12.2.2 Habitat Dynamics of Selected Biota in Riverine Ecosystems 8
12.2.2.1 Pallid Sturgeon 9
12.2.2.2 Bigheaded Carps 12
12.2.2.3 Sedentary Organisms: Native Freshwater Mussels 12
12.2.2.4 Cottonwood Communities 12
12.2.2.5 Shorebirds: Terns and Plovers 13
12.2.3 Implications and Applications of Habitat Dynamics 13
12.2.3.1 Problem Diagnosis: Understanding Stressors, Habitat Grain, and Habitat Extent 13
12.2.3.2 Hydrodynamics and Habitat Regime 14
12.2.3.3 Morphodynamics and Habitat Regime 14
12.2.4 Conclusions 16
References 16

Abstract

The physical habitat template is a fundamental influence on riverine ecosystem structure and function. Habitat dynamics
refers to the variation in habitat through space and time as the result of varying discharge and varying geomorphology. Habitat
dynamics can be assessed at spatial scales ranging from the grain (the smallest resolution at which an organism relates to its
environment) to the extent (the broadest resolution inclusive of all space occupied during its life cycle). In addition to a
potentially broad range of spatial scales, assessments of habitat dynamics may include dynamics of both occupied and
nonoccupied habitat patches because of process interactions among patches. Temporal aspects of riverine habitat dynamics
can be categorized into hydrodynamics and morphodynamics. Hydrodynamics refers to habitat variation that results from
changes in discharge in the absence of significant change of channel morphology and at generally low sediment-transport
rates. Hydrodynamic assessments are useful in cases of relatively high flow exceedance (percent of time a flow is equaled or
exceeded) or high critical shear stress, conditions that are applicable in many studies of instream flows. Morphodynamics
refers to habitat variation resulting from changes to substrate conditions or channel/floodplain morphology. Morphodynamic
assessments are necessary when channel and floodplain boundary conditions have been significantly changed, generally by
relatively rare flood events or in rivers with low critical shear stress. Morphodynamic habitat variation can be particularly
important as disturbance mechanisms that mediate population growth or for providing conditions needed for reproduction,
such as channel-migration events that erode cutbanks and provide new pointbar surfaces for germination of riparian trees.
Understanding of habitat dynamics is increasing in importance as societal goals shift toward restoration of riverine eco-
systems. Effective investment in restoration strategies requires that the role of physical habitat is correctly diagnosed and that
restoration activities address true habitat limitations, including the role of dynamic habitats.

12.2.1 Introduction and time is critical to river management. The objectives of this
chapter are: (1) to explore the concept and scales of habitat
The concept of habitat is fundamental to the ecological dynamics as they operate in typical riverine systems, including
understanding of streams and rivers. Habitat not only defines the channel and floodplain (up to nominally a 0.01 annual
the template of resources and conditions of spaces that or- probability flood); and (2) to define the concepts of hydro-
ganisms occupy, but also is the link to most management and dynamics and morphodynamics, including discussion of how
restoration actions. Most of what society seeks to manage or they relate to understanding the limiting factors that need to
restore in riverine ecosystems relates directly to habitat. Hence, be addressed in order to restore aquatic habitats.
an understanding of habitat and how it is distributed in space
12.2.1.1 Some Definitions and Concepts

Jacobson, R.B., 2013. Riverine habitat dynamics. In: Shroder, J. (Editor in

12.2.1.1.1 Habitat
chief), Butler, D.R., Hupp, C.R. (Eds.), Treatise on Geomorphology. Aca- Habitat can be defined holistically as ‘‘the resources and con-
demic Press, San Diego, CA, vol. 12, Ecogeomorphology, pp. 6–19. ditions present in an area that produce occupancy – including

6 Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00318-3

 Riverine Habitat Dynamics 7

survival and reproduction – by a given organism’’ (Hall et al., landscape ecology, the concepts of grain and extent help de-
1997). Such broad definitions encompass all physical, chem- fine these scales. Grain is generally considered the smallest
ical, and biological conditions as being potentially important unit of resolution of a set of measurements and extent is
in producing successful occupancy. The definition of habitat considered the overall area encompassed by the measurements
is frequently simplified to emphasize abiotic (physical and (Wiens, 1989). To consider all relevant scales of habitat for
chemical) conditions, and the temporal characteristics of their particular species, the grain should be set at the finest scale at
occurrence (Jeffries and Mills, 1990; Gordon et al., 1992). which an organism interacts with its habitat during a par-
Hereafter in this chapter, emphasis is on the physical ticular life stage and extent should include the breadth of
aspects of habitat that relate most directly to the science interaction, encompassing all habitats occupied by an organ-
of geomorphology; however, the reader is cautioned that ism during a particular life stage, or, for some assessments, all
a broad definition of habitat includes chemical and biotic life stages. For migratory species, the extent can be quite large.
influences, which, in some cases, may be of overriding Although conditions commonly vary continuously within
importance in defining habitat (Lancaster and Downes, 2009). river–floodplain systems, habitats are typically discretized into
A highly relevant example is how vegetation communities patches, or areas or volumes with limited ranges of variability
(whose reproduction and survival depend, to a large extent, in specified conditions. Like the general concept of habitat,
on physical habitat) alter the hydraulic environment around how a patch is defined depends on the organism and
them and provide shading and cover, and thereby alter phys- questions under consideration (Pringle et al., 1988). Whether
ical, chemical, and biological habitat conditions experienced specified by a particular process, specific organism, naturally
by other organisms. occurring gradients, or simply by arbitrary classification, pat-
Some measures of habitat can be generic and applicable to ches can be analyzed in terms of their spatial organization,
many organisms and life stages. Generic measures may include functional relations among patches, and their temporal
water-quality conditions that would affect many organisms, such changes (Pickett and White, 1985). Recognition of the role of
as dissolved oxygen, turbidity, or water residence time. Other patches in the life histories of riverine organisms (Bornette
habitat measures can be specific to an organism or a life stage of et al., 2008) requires an expansion of the concept of habitat to
interest. A specific habitat can be defined, for example, as understanding the functional relations among habitat patches
spawning habitat for a trout, or foraging habitat for a shorebird. as they contribute to occupancy, survival, and reproduction of
Determination of links between biota and habitat can be an organism. Patch metrics – like total edge, juxtaposition,
challenging. Habitat may be critically necessary to complete a and patch diversity (McGarigal and Marks, 1995) – can be
specific life stage or simply supportive. For sedentary organ- useful measures of aquatic habitat spatial structure (Jacobson
isms, such as trees, habitat associations can be easily estab- and Galat, 2006; Jacobson et al., 2009; Reuter et al., 2003).
lished, by measuring the range of conditions occupied by Habitats are conventionally organized into hierarchies of
trees. For mobile organisms, the association can be more units related to geomorphology: macrohabitats, channel units,
difficult to establish from field observations of transient reaches, segments, and channel networks (Frissell et al., 1986).
occupancy and environmental conditions. Determination of Classification or ‘typing’ of channels is useful in assessments of
habitat affinities for fish, for example, is generally addressed by habitat dynamics because it clusters channels into classifications
measuring conditions in spaces occupied by the fish (habitat that should share similar dynamics and spatial structure. Such
use) and comparing with conditions that are available (habitat units have been called ‘process domains’ (Montgomery, 1999) or
availability). The ratio is a measure of habitat selection, which ‘functional process zones’ (Thorp et al., 2006).
can be formalized in selectivity coefficients (Lechowicz, 1982)
or probabilistic measures (Ahmadi-Nedushan et al., 2006). 12.2.1.1.3 Habitat dynamics
Field-based determinations of habitat selection are limited to Habitat dynamics refers to processes of change of habitat and
understanding selection from among available habitats and can apply to a single habitat patch or the mosaic of patches.
do not necessarily measure habitat preference, or the optimal Change can be measured by change in quality, availability, or
habitat for that life stage, if optimal conditions are not present. patch structure. The temporal characteristics of change can be
As a result of these challenges, some specific habitat affinities measured in terms of magnitude, frequency, duration, timing,
are well established, whereas others are much less well and rate of change, similar to the temporal characteristics
understood. Where specific habitat requirements or affinities of flow regimes (Richter et al., 1996). Habitat dynamics is
are not known, useful analysis of dynamics can still be especially important in the context of understanding how
accomplished with generic measures of hydraulics, hydro- disturbances restructure communities and populations.
period, or residence time (Jacobson et al., 2009). Similar to the ideas about ecologically acceptable states of
hydrologic alteration (Poff et al., 2009; Richter et al., 1996),
12.2.1.1.2 Spatial characteristics of habitat most riverine ecosystems can accommodate some rate of
Habitats can be measured in terms of spatial characteristics habitat dynamics; indeed, life strategies of many riverine or-
as well as quality. Spatial measures may be in terms of areas, ganisms depend on periodic or episodic changes in habitat
volumes, or the sizes, shapes, and spatial organization of availability or quality. Habitat change beyond a threshold,
patches of habitat. Scale is critically important to under- however, becomes a disturbance. Examples are extended
standing habitat dynamics and scale may vary considerably drought that dries up the channel and hyporheic refugia
depending on the biological context of a study. Scales (Fowler and Death, 2001; Palmer et al., 2005a), or bed scour
of interest are commonly bounded by the scales at which that wipes out populations of benthic algae (Power and
organisms interact with their physical environment. In Stewart, 1987).
8 Riverine Habitat Dynamics

Human alterations of ecosystems often have been identi-

fied as sources of disturbances outside the range of naturally
occurring habitat dynamics (Landres et al., 1999). Quantifying
natural rates of habitat dynamics is therefore critical to dia- Hydrodynamic
gnosing impaired riverine ecosystems and for defining resto- domain

Increasing critical
ration goals (Palmer et al., 2005b, 2007).

shear stress
By assessing habitat dynamics in terms of regimes, one can
explicitly address seasonality, sequence, and disturbance
potential, factors that are fundamental to ecological functions. Morphodynamic
Operationally, habitat regimes for riverine ecosystems are domain
compiled by developing discharge–habitat availability curves,
then assessing temporal variation in the discharge through
frequency or duration analysis. Discharge–habitat availability Decreasing flow exceedance
curves are typically computed using conventional one- Decreasing flood annual probability
dimensional or multidimensional hydraulic models (Bovee Figure 1 Conceptual model of hydrodynamic and morphodynamic
et al., 1998; Jacobson and Galat, 2006) or by field mapping of domains as functions of flow exceedance and critical shear stress.
habitat units over a range of discharges (Parasiewicz, 2007). The
use of multidimensional hydrodynamic models allows quan-
zero discharge, rise as discharge increases, and then typically
tification of many spatial attributes of habitat, including di-
decrease as discharge fills the channel. A generalized deep-
versity (Reuter et al., 2003; Pasternack et al., 2004), gradients
water channel class can be defined that also starts at zero area
between habitat units (Crowder and Diplas, 2006; Johnson
with zero discharge, but increases in area to an asymptote at
et al., 2006), patch dynamics (Bowen et al., 2003), and patch
bankfull. A low Froude number class rises quickly to a nearly
persistence (Bovee et al., 2004). Habitat simulation studies
steady value; in the case illustrated in Figure 2, low Froude
generally have been limited to consideration of a fixed bed,
number areas are associated with slow water behind wing
thus allowing for hydrodynamic assessments but not for
dikes and are relatively constant with discharge. A connected-
addressing habitat changes associated with changes to channel
floodplain class increases with discharge only after a threshold
morphology due to sediment transport, erosion, or deposition.
discharge at bankfull.
Hydrodynamic models that neglect geomorphic change are
Combined with flow duration, such generic discharge–habitat
appropriate for low-flow studies or studies on rivers with im-
curves can be used to define habitat–duration functions to char-
mobile beds. Studies that intend to consider the ecological
acterize how often specific habitats are available on an annual
effects of flows capable of geomorphic change need to in-
basis; habitat–duration plots on a daily basis (Figure 3) illustrate
corporate sediment transport and channel evolution in either
seasonal habitat dynamics in a form that can be compared
field measurements or appropriate morphodynamic models
directly with seasonal understanding of habitat requirements.
(McDonald et al., 2005; Schmeeckle and Nelson, 2003).
Assessment of habitat dynamics on a daily duration basis also
The different information that hydrodynamic and mor-
allows for consideration of flow-related events that could qualify
phodynamic models contribute to understanding habitat
as cues for important life-stage events (e.g., floodplain connecting
dynamics suggests a general framework for differentiating
pulses to allow for spawning of some fishes) or for assessment of
timescales. A hydrodynamic timescale can be defined as the
periodicity of flow-related disturbances that affect availability and
timeframe over which habitat dynamics is dominated by flow
diversity of habitat types (Figure 4).
regime interacting with a relatively stable channel–floodplain
configuration. A morphodynamic timeframe can be defined as
the timeframe or flow exceedance over which significant
sediment transport and geomorphic change take place. A 0.01 12.2.2 Habitat Dynamics of Selected Biota in
annual probability flood, for example, would likely be a Riverine Ecosystems
morphodynamic event but a 1.0 annual probability flood may
or may not be morphodynamic depending on resistance of Habitat dynamics can affect individuals, populations, and
the bed and banks to erosion. Hydrodynamic timescales ex- communities by influencing quality, quantity, patch arrange-
tend to lower-flow frequencies or exceedances as thresholds ment, and disturbance rates at a variety of scales. The fol-
for sediment-transport increase, for example, with bedrock- lowing examples are intended to illustrate processes and scales
bounded or cobble-gravel channels (Figure 1). For sandbed that relate to the understanding of habitat dynamics for dif-
channels where sediment transport and geomorphic change ferent types of riverine biota and their life stages, with an
are nearly continuous, hydrodynamic timescales would extend emphasis on biota from large rivers of the central United
only to flow exceedances below which significant sediment States. In each example, I consider the known habitat affinities
and channel change occur. of the biotic group or species, discuss the grain and extent of
Generic discharge–habitat curves applicable to habitat dy- habitat use, and consider the dynamics of these habitats in
namics of many rivers are illustrated in Figure 2. The emergent terms of hydrodynamics and morphodynamics. Although this
sandbar class decreases in area with increasing discharge. is a biased selection of biota, it illustrates the point that
A generalized shallow, slow current–velocity class can have hydrodynamics and morphodynamics need to be understood
substantial variation in form depending on values of depth over a broad range of scales depending on the species or life
and velocity used to define it, but all begin with zero area at stage of interest.
 Riverine Habitat Dynamics 9

160 1600

140 1400

Connected floodplain area (ha)

120 1200

Habitat area (ha)

100 1000

80 800

60 600

40 400

20 200

0 0
(a) 0 1 2 3 4 5
0.12

0.10
Discharge frequency

0.08

0.06

0.04

0.02

0.00
(b) 0 1 2 3 4 5
7

6
Sandbar
5 Shallow, slow water
Area × Frequency

Deep, fast water

4 Low froude number
Connected floodplain
3

0
0 1 2 3 4 5
(c) Discharge normalized by median discharge
Figure 2 Typical aquatic habitat class relations, Lower Missouri River. Discharge is normalized by median discharge. (a) Discharge-area curves for
five habitat classes showing typical variation with discharge. (b) Frequency distribution of discharges. (c) Area times frequency curves to indicate
habitat-effective discharges (Doyle et al., 2005), those discharges that are responsible for most of the availability of that class on average. Adapted
from Jacobson, R.B., Johnson, H.E., III, Dietsch, B.J., 2009. Hydrodynamic simulations of physical aquatic habitat availability for pallid sturgeon in
the Lower Missouri River, at Yankton, South Dakota, Kenslers Bend, Nebraska, Little Sioux, Iowa, and Miami, Missouri, 2006–07. US Geological
Survey Scientific Investigations Report 2009-5058. US Geological Survey, Reston, VA, 67 pp.

12.2.2.1 Pallid Sturgeon range of habitat types over broad spatial extent. Spawning
migrations as long as 1700 km have been documented within
The pallid sturgeon (Scaphirhynchus albus) is an endange- the Missouri River (Aaron DeLonay, US Geological Survey,
red fish that occurred historically throughout the Lower Mis- personal communication); under present-day conditions, this
sissippi, Middle Mississippi, and Lower Missouri rivers. extent is mostly a single-thread, channelized river, although
It is an example of fishes that migrate long distances to it includes complexity within channel training structures and
complete their reproductive cycles and therefore use a wide in numerous tributary confluences (Jacobson et al., 2009;
10 Riverine Habitat Dynamics

40
Preregulation
historical hydrology
1928−52

30
Sandbar area (ha)

At least this area of sandbar habitat:

10% of the time
20 25% of the time
75% of the time
90% of the time

10

40
Postregulation
historical hydrology
1967−2001

30
Sandbar area (ha)

At least this area of sandbar habitat:

10% of the time
20 25% of the time
75% of the time
90% of the time

10

0
Jan Mar May July Sept Nov
Month
Figure 3 Duration hydrograph plots of emergent sandbar habitat availability, Lower Missouri River, near Waverly, Missouri, for natural
and managed flow regimes. Habitat–duration plots illustrate habitat dynamics as a function of year-to-year hydrologic variability (shown in
duration bands) and seasonality. Reproduced from Jacobson, R.B., Laustrup, M.S., D’Urso, G.J., Reuter, J.M., 2004. Physical habitat
dynamics in four side-channel chutes, Lower Missouri River. US Geological Survey Open-File Report 2004-1071. US Geological Survey, Reston,
VA, 60 pp.

Reuter et al., 2008, 2009). The grain of habitat used ranges hypothesized that its life stages are not particularly sensitive to
from millimeters occupied by eggs and drifting larvae, up to hydrodynamic availability of habitat (DeLonay et al., 2007,
1–2-m length of the adult fish. The extent of an egg may be a 2009), and potential migratory and spawning habitats
few meters as they settle from release to adhere to substrate, have been shown to be relatively insensitive to discharge
whereas the extent of drifting larvae and migrating adults is over a hydrodynamic timescale (Jacobson et al., 2009). As a
hundreds of kilometers (Braaten et al., 2008; DeLonay et al., benthic fish, the pallid sturgeon has been strongly linked to
2009). As a main-channel obligate adapted to high current substrate types (Bramblett and White, 2001; Reuter et al.,
velocities and that rarely occupies floodplains, it has been 2009). Flow pulses have been hypothesized to be necessary
 Riverine Habitat Dynamics 11

20 000

16 000 Race

Habitat area (m2)

Riffle
12 000

8000

4000

0
16 000
14 000 Edgewater
Glide
12 000
Habitat area (m2)

Pool
10 000
8000
6000
4000
2000
0
1.6

1.4
Shannon diversity

1.2
index

1.0

0.8

0.6
180
160
Daily mean discharge,
crane bottom (m3 s−1)

140
120
100
80
60
40
20
0
1/1/99 7/1/99 1/1/00 7/1/00 1/1/01 7/1/01 1/1/02 7/1/02
Date
Figure 4 Examples of aquatic habitat dynamics. Time series of hydraulically modeled habitat class availability, plus habitat diversity index, Bear
Creek, northern Arkansas. This example illustrates seasonal, periodic hydrologic controls on habitat availability and episodes that may interact
with specific life-stage needs, such as spawning habitat availability or refuge from low flows.

behavioral cues for migration and spawning (US Fish and Morphodynamics have been hypothesized as an important
Wildlife Service, 2003); such cues might be felt as change or factor in the life history of pallid sturgeon in two particular
rate of change in water temperature or turbidity that would ways. First, flow pulses have been hypothesized to provide
constitute a transient condition within existing main-channel the additional velocity needed to flush fine sediment
habitat rather than making new habitat available. Flow pulses and condition spawning substrates for adhesive eggs. Geo-
have also been hypothesized to affect downstream drift rates morphic studies have confirmed that flow pulses provide
of free-embryo larvae by altering mainstem velocity, a con- sufficient velocities to scour the bed and significantly change
dition that would affect a millimeter-scale individual over channel morphology, but direct links to functional spawning
potentially hundreds to thousands of kilometers of river substrate have not been demonstrated (Elliott et al., 2009).
(DeLonay et al., 2009). Morphodynamics are also central to restoration activities
12 Riverine Habitat Dynamics

intended to re-engineer the river to provide additional shal- limited movement. For sedentary organisms such as the adult
low, slow-water habitat for rearing of juvenile sturgeon. The stage of mussels, the grain and extent of habitat are quite
general approach to re-engineering the channel to increase similar and stability of habitat is paramount. As a group,
this habitat class has been to excavate pilot channels for new freshwater mussels are tightly linked to substrate, although
chutes or to widen the main channel. The pilot excavations are the species can be variously associated with mud, sand, or
intended to enlarge to provide a future, expected quantity of gravel cobble. The key properties of substrate for mussels are
habitat. The functional relations among discharge, channel stability and a particle-size distribution amenable to bur-
enlargement, and rates of habitat development, although rowing (Strayer, 2008). Slow rates of movement leave mussels
poorly understood, are the central design issues in the res- very susceptible to habitat change on hydrodynamic time-
toration effort. scales, in particular low-water events that can threaten
them with desiccation. Unlike habitat assessment studies for
fishes, in which it can be assumed that most fish have the
12.2.2.2 Bigheaded Carps option to migrate within the channel to seek acceptable
habitat, hydrodynamic habitat assessments for mussels need
Bigheaded carps consist of three species of Asian carps that
to evaluate habitat patch persistence in a specific location
were imported into the US from China for aquaculture. All
(Bovee et al., 2004).
three species have escaped from aquaculture facilities and have
Because mussels are sensitive to substrate instability,
proliferated in the Mississippi, Missouri, and Illinois rivers.
erosion and sedimentation (morphodynamics) are a source
Among the three species, there is presently (2010) great con-
of disturbance. Geomorphically unstable channels are
cern that two of them, the bighead (Hypophthalmichthys nobilis)
commonly associated with diminished mussel populations
and the silver carp (Hypophthalmichthys molitrix), may migrate
(Hornbach, 2001; Strayer, 2008). Although bedload trans-
into the Great Lakes where they would enjoy a competitive
porting events have the capacity to extirpate mussel popu-
advantage and threaten native and sport fishes. Although in-
lations locally, mussels can recolonize as host fish migrate
habiting the same rivers, bigheaded carps have very different
through a reach and provide opportunities for recolonization
habitat affinities compared with pallid sturgeon. Bigheaded
of juveniles (Sietman et al., 2001). By recolonizing, mussel
carps can grow to several meters length and weigh over 100 kg.
populations can persist in morphologically dynamic rivers
Similar to sturgeon, these carps migrate to spawn in turbulent
(Miller, 2006).
water, commonly over rocky substrate, but their migration
distances are considerably shorter, ranging up to 100 km
(Kolar et al., 2007). In contrast to the adhesive eggs of pallid
sturgeon, bigheaded carps have buoyant eggs that need to float 12.2.2.4 Cottonwood Communities
for as much as 100 km before hatching, depending on velocity
Studies of cottonwoods (e.g., plains cottonwood, Populus
and temperature. Hence, the grain of bigheaded carp eggs
deltoides) have demonstrated the close coupling of hydro-
is similar to that of sturgeon, but the extent is much greater.
dynamics, morphodynamics, and reproduction of these
The grain and extent of larvae are similar. The grain of adult
riparian trees. The grain of habitat occupancy for a seed is
carp is 1–2 m; the extent of adult fish habitat use is not well
several millimeters, but the extent can be quite large because
defined at this time.
they can be dispersed long distances in the air and by water.
As bigheaded carps are pelagic and planktivorous, their
Once germinated, the grain and extent of individuals are
habitat affinities are not linked directly to substrate com-
limited to the scale of the diameter of stem.
position. The fact that they spawn on the rising limb of annual
For natural reproduction, conditions are set by morpho-
floods has been interpreted as an adaptation to allow larvae
dynamic processes that include lateral migration of the
access to floodplain habitats, and young-of-year carps have
channel and associated deposition of bare mineral soil on
been abundant on floodplains (Kolar et al., 2007; Schrank
point bars. Germination requires that seeds are available and
et al., 2001); hence, floodplain-connecting floods are probably
dispersed by air or water to settle on the bare surfaces. Typical
important to the growth and survival of these fish. Adults of
arrangement of cottonwoods in linear patches subparallel to
both carp species have been documented to not only select the
the channel (Everitt, 1968) demonstrates that seed deposition
slower, deeper habitats available in rivers, such as the Lower
or germination conditions are hydrodynamically controlled
Missouri, but also migrate into adjacent floodplains when
subsequent to deposition of the bar surface. As part of this
sufficient water exists (Johnson et al., 2006; Kolar et al., 2007).
morphodynamic process, erosion of the opposite bank typi-
Morphodynamics relevant to bigheaded carps are mainly
cally delivers older cottonwood and other trees to become
human-engineered events, either construction of low-head
habitat elements in the channel.
dams that provide abundant lotic habitat or reestablishment
Postgermination survival of cottonwoods has been closely
of flow through notches in L-head wing dikes, which is
tied to hydrodynamics of surface water and the related ele-
thought to diminish carp habitat.
vation of the groundwater table under the point bar: survival
of a seedling depends on having the water table near the
surface for initial germination; subsequently, growth and sta-
12.2.2.3 Sedentary Organisms: Native Freshwater Mussels
bility of the seedling depend on a groundwater recession rate
Freshwater mussels (family Unionidae) have complex life that allows root growth to keep pace with the water-
cycles in which their larval stage is parasitic on a mobile host table decline. If the groundwater (and surface water) recedes
fish; juveniles through adult life stages are capable of only too quickly, it can cause cavitation or desiccation in the
 Riverine Habitat Dynamics 13

seedling roots; if it recedes too slowly, the root may not extend washing nests away. Typically, nests that are initiated after the
deep enough to provide stability or access to water in sub- peak of snowmelt-driven seasonal hydrographs are successful,
sequent seasons (Mahoney and Rood, 1998). Survival of cot- although late-season floods can periodically wash out a year
tonwood seedlings in northern latitudes also depends on class, especially for birds that nested at elevations too close
germination at elevations where they are not affected by ice to the water.
scour during the winter. Finally, once established, riparian
vegetation communities produce a morphodynamic feedback
by increasing hydraulic roughness and affecting quantity and
12.2.3 Implications and Applications of Habitat
pattern of deposition (Griffin et al., 2005; McKenney et al.,
Dynamics
1995; Smith, 2007).
Recognition of the complexities of interactions of species with
12.2.2.5 Shorebirds: Terns and Plovers their habitats over hydrodynamic and morphodynamic time-
scales, as illustrated in the preceding examples, provides a
Reproduction, growth, and survival of many shorebirds are
basis for managing and restoring riverine ecosystems and the
also tightly coupled to river hydrodynamics and morphody-
species they support. In many cases, successful restoration
namics. Within the Central US, interior least terns (Sterna
depends on understanding the species’ life-history needs,
antillarum) and the northern Great Plains population of pip-
correctly diagnosing the river’s impairment, and restoring a
ing plovers (Charadrius melodus) are of interest because of their
range of hydrodynamic and morphodynamic habitat regimes
endangered and threatened status, respectively, and because
that support the species of interest without promoting per-
of their affinities for bare sandbars. Both of these species
sistence of exotic species.
overwinter in the Gulf Coast, but then fly north to nest on
sandbars of rivers such as the Mississippi, Missouri, and Platte
(National Research Council, 2004; US Fish and Wildlife
12.2.3.1 Problem Diagnosis: Understanding Stressors,
Service, 2000, 2003). When young and relatively sedentary
Habitat Grain, and Habitat Extent
(egg and nestling life stages), these birds interact with their
habitat at the scale of tens of millimeters to meters. Parents Habitats of many rivers worldwide have diminished in quality
choose to place nests in a fairly narrow range of sand-gravel and quantity due to land-use changes, impoundments, and
particle sizes, in bare areas with long sight distances (for channelization. These habitat changes have been associated
predator avoidance), and they select sites with heterogeneous with declines of many species, although the direct linkage
elements, such as drift wood, to act as camouflage (Cohen is not always clear, and, hence, in the absence of a correct
et al., 2009; Smith and Renken, 1991). As they age, tern chicks diagnosis of the linkage, restoration is likely to be ineffective.
tend to stay in their nests and are fed by parents who forage A prominent example is the endangered Delta smelt in the
over hundreds to thousands of meters (Kirsch, 1996), By Sacramento Delta. Because water diversions for agricultural
contrast, the more precocious plover chicks interact with tens production and public water supply have been substantial
of meters of sandbar habitat as they forage for insects (Le Fer and widespread in the Delta, initial diagnosis of the cause of
et al., 2008). During nesting season, parents interact with their the smelt decline identified loss of physical habitat, and po-
habitat over hundreds to thousands of meters as they forage tentially, degraded water quality; this diagnosis resulted in
for small fish (terns) and invertebrates (piping plovers). substantial investments in habitat restoration projects (Som-
Hence, relevant hydrodynamics and morphodynamics for mer et al., 2007). When improved hydrologic and habitat
terns and plovers during nesting season include the nesting conditions failed to result in species recovery, hypotheses
site as well as adjacent foraging habitat. For the entire life cycle moved toward more complex ecological explanations in-
of these birds, the extent of habitat ranges up to thousands cluding problems with the food web and competition by in-
of kilometers; survival may depend on habitat conditions at vasive species, instead of explanations related to physical
many places along their seasonal migration. habitat limitations. Similarly, habitat degradation has been
Most emphasis, however, has been on nesting success implicated in decline of many sturgeon species such as the
because reproductive success is fundamental to population pallid sturgeon (Dryer and Sandvol, 1993), yet the repro-
viability (Kirsch, 1996). Natural riverine nesting sites on ductive ecology of many sturgeon species also makes them
sandbars require deposition of bare sandbars of sufficient size particularly susceptible to overharvest. Hence, proper diag-
to provide long sight lines for predator avoidance. Sandbars nosis of stressors on species is crucial for effective restoration
that are detached from the shore and long distances from large strategies; restoration of elements of physical habitat is not
riparian trees provide additional protection from mammalian a panacea (Hilderbrand et al., 2005; Lancaster and Downes,
and bird predation. These conditions are created by relatively 2009).
rare floods, and suitable nesting habitats are favored in In cases where habitat availability or quality is correctly
wide, multithread channel reaches. Suitable nesting habitat diagnosed as the limiting factor in species viability, under-
can commonly decrease over years between large floods as standing of the role of hydrodynamic and morphodynamic
vegetation becomes established on the sandbars. Maintenance variability can be key to restoration and management strat-
of nesting habitat, therefore, requires episodic deposition egies. If holistic restoration is possible – that is, complete
or vegetation-scouring events. On hydrodynamic timescales, restoration of the full natural range of variability of flow
successful nesting during a season requires water levels that are regime, sediment regime, and channel constraints – the full
steady to decreasing through the nesting period to avoid range of habitat dynamics would presumably be restored.
14 Riverine Habitat Dynamics

In most impaired river systems, full restoration is not possible, Understanding of habitat dynamics on hydrodynamic time
and strategies focus on re-engineering components of habitat frames has been key to assessing effects of water-resources
(Gore and Shields, 1995; Rhoads et al., 1999). Restoration of development, with an emphasis on how reservoir regulation
components of a riverine ecosystem, or recovery of a limited and water diversions affect instream flows (Bovee et al., 1998;
range of ecosystem functions to meet multiple objectives, re- Bovee et al., 2004; Bowen et al., 2003; Freeman et al., 2001;
quires considerably more understanding of habitat dynamics Parasiewicz, 2007). If limiting habitats are correctly identified
compared to holistic restoration. and the assumptions of hydrodynamic approaches are valid-
ated, assessments of availability of habitat metrics as functions
of discharge provide a useful extension to consideration of the
12.2.3.2 Hydrodynamics and Habitat Regime
flow regime alone in evaluating impairment or restoration
The concept of flow regime provides a useful framework for options (Arthington et al., 2006; Poff et al., 1997, 2009).
understanding habitat dynamics. Habitat regime can be con-
sidered to have similar temporal properties of a flow regime:
12.2.3.3 Morphodynamics and Habitat Regime
magnitude (or area or volume), duration, frequency, timing, and
rate of change (Poff et al., 1997). Temporal aspects of habitat Hydrodynamic approaches to understanding habitat regime are
dynamics can also be described in terms of environmental flow valuable in quantifying much of the range of habitat dynamics
components (EFCs), or components of the continuum of that relates simply to how much water is in the river and how it
habitat availability associated with classes such as extreme low is hydraulically distributed. Hydrodynamic approaches alone,
flows, low flows, flow pulses, small floods, and large floods however, fail to address how habitats change over time and
(Mathews and Richter, 2007). If discharge-area curves can be space as sediment is transported and changes occur in bed
developed for a habitat type (e.g., Figure 2(a)), then a time series substrate, near-bed sediment fluxes, bedforms, and channel
of discharges can be used to generate a time series of availability morphology, all of which potentially affect habitat at various
of habitat (e.g., Figure 4), which can be analyzed in terms of scales. Quantitative understanding of morphodynamic effects
flow-regime variables. Analysis of the temporal dimensions of on habitat regime is considerably more challenging than
habitat allows for explicit assessment of timing (or seasonality), understanding of hydrodynamic effects, as it requires inte-
a potentially critical issue in synchronizing hydrologic drivers grating hydraulics with the mechanics of sediment erosion,
with life-history events (Figures 3 and 4). transport, and deposition, including complex interactions with
Habitat regime can also be addressed in terms of the vegetation and gravitational bank-failure processes. Direct
product of magnitude and frequency of habitat availability to geomorphic and sedimentologic measurements of habitat
determine effective habitat events, analogous to the idea of change remain the foundation of understanding, but compu-
effective flows in sediment transport and channel adjustments tational modeling capabilities are progressing to a point where
(Doyle et al., 2005; Wolman and Gerson, 1978). The effective- they can provide useful insights in habitat dynamics.
discharge approach uses the habitat–discharge relation as a Measurements of habitat dynamics can range from grain-
‘rating curve’ and calculates the product of magnitude of scale to drainage-basin scale. Most field-based measurements
habitat at a particular discharge and the proportion of time a are at patch or cross-sectional scale using methods such as
discharge interval exists to develop the magnitude  frequency scour chains (Haschenburger, 1999; Matthaei and Townsend,
curve; the peak of the curve is the habitat-effective discharge, 2000) or pressure transducers (Borg et al., 2007) to assess
the discharge that produces the most of that habitat avail- scour, and repeat surveys of cross sections (Elliott et al., 2009;
ability over time (Doyle et al., 2005). The habitat-effective McKenney, 2001; Reuter et al., 2003; Schmidt and Graf, 1990)
discharge will vary with the type of habitat (Figure 2(a)) and to assess net morphological change. Resurvey frequency varies
the shape of the frequency distribution of discharges considerably among such studies, from continuous capabil-
(Figure 2(b)). For example, effective discharges for habitat ities afforded by pressure transducers to seasonal and multi-
types typical of the Lower Missouri River, USA, vary from annual cycles. The challenges in habitat monitoring studies
0.3 times median discharge to 3 times median discharge, in- are to resurvey with sufficient frequency to capture relevant
dicating the dynamic shifts in habitat availability that occur events and to extend monitoring over sufficient timescales to
during a typical year. put morphodynamic events into temporal context.
Spatial arrangements of habitat patches present additional In combination, channel resurveys and hydrodynamic
complexity to the temporal assessments. If habitat types can modeling provide hydrodynamic habitat metrics in assess-
be defined as discrete units, the arrangement of patches can ments of morphodynamic changes (Jacobson et al., 2009).
be assessed and variables relating to ecologically important The time and resources needed to resurvey computational
patch arrangement can be evaluated aswell in the temporal meshes after geomorphic events are considerable, however,
domain. For example, total edge between habitat patches and and examples of this approach are limited.
patch density is often considered ecologically important Remote-sensing measurements of morphodynamic changes
(Jacobson et al., 2009; McGarigal and Marks, 1995); similarly, are common, ranging from use of low-altitude balloon-based
complexity of habitat can be measured by indices such repeat photography to quantify dune migration (McElroy and
as the Simpson’s diversity index and related to ecologic Mohrig, 2009; Mohrig and Smith, 1996), to common use of
potential. Relations between discharge and patch distribution aerial photography to map changes in channel features and
characteristics can be used to assess sensitivity of these channel migration (Elliott and Jacobson, 2006; Jacobson and
metrics to discharge variation on hydrodynamic timescales Pugh, 1997; Joeckel and Henebry, 2008; Shields et al., 2000;
(Figure 5). Wallick et al., 2008; Zanoni et al., 2008). Deployment of
 Riverine Habitat Dynamics 15

2000

Patch density, number per 100 ha

Patch density
Edge density
1500

Edge density (m ha−1) 1000

500

0
(a) 0.0 0.5 1.0 1.5 2.0 2.5

0.90
Simpson’s diversity index

0.85

0.80

0.75

0.70

0.65

0.60

0.55
0.0 0.5 1.0 1.5 2.0 2.5
(b) Discharge normalized by median discharge
Figure 5 Spatial habitat patch characteristics, Lower Missouri River, near Miami, Missouri. (a) Relations of patch density and edge density of
discrete aquatic habitats to normalized discharge. (b) Relation of Simpson’s diversity index calculated from habitat patch distributions to
normalized discharge. Reproduced from Jacobson, R.B., Johnson, H.E., III, Dietsch, B.J., 2009. Hydrodynamic simulations of physical aquatic
habitat availability for pallid sturgeon in the Lower Missouri River, at Yankton, South Dakota, Kenslers Bend, Nebraska, Little Sioux, Iowa, and
Miami, Missouri, 2006–07. US Geological Survey Scientific Investigations Report 2009-5058. US Geological Survey, Reston, VA, 67 pp.

high-resolution bathymetric and ground- and aerial-based time frames (Johannesson and Parker, 1989; Lancaster and
LiDAR mapping systems (LiDAR, light detection and ranging) Bras, 2002; Larsen et al., 2007; Larsen and Greco, 2002; Wallick
also allows for repeat three-dimensional measurements of et al., 2006); morphodynamic models at this scale have po-
geomorphic and habitat changes (Hazel et al., 1999; Heritage tential application to prediction of riparian vegetation com-
and Hetherington, 2007; Kinzel et al., 2006). munity dynamics (Lytle and Merritt, 2004).
Modeling of morphodynamics can extend understanding Models have also been constructed at the drainage-basin
gained from measurement and observation, and serve to test scale to explain distributions of habitats and disturbance
sensitivity of habitat dynamics to changes in inputs. At the processes throughout a drainage network. Such models are
patch scale, shear-stress models are typically used to predict bed necessarily highly simplified relative to those that capture
mobility and disturbance to the benthos (Lisle et al., 2000; detailed hydraulic and gravitational processes, but have been
Schwendel et al., 2010). At patch to reach scales, hydrodynamic shown to be useful. In high-relief drainage basins, where
models have been linked to sediment-transport models to riverine habitats can be affected by energetic flows that
predict bed evolution, hence changes in aquatic habitat transport large-caliber sediment and woody debris, landscape-
(Abad et al., 2008; Barton et al., 2005; McDonald et al., 2006; scale models have been constructed to provide spatial pre-
Pasternack et al., 2004; Wiele et al., 2007). Extending mor- dictions of channel disturbance and substrate composition
phodynamic modeling out of the channel requires incorpor- (Benda et al., 2007; Bigelow et al., 2007; Lancaster et al.,
ating additional complexity related to gravitational bank-failure 2003). In low-relief landscapes, models of aquatic habitat
processes (Langendoen et al., 2009) and the effects of bank availability and dynamics are typically statistical associations
vegetation and geometry (Griffin et al., 2005; Kean et al., 2009; or heuristic rather than being process based (Gergel et al.,
Pollen-Bankhead and Simon, 2010). Considerably more sim- 2002; Jacobson and Gran, 1999; Jones et al., 2006; Sowa et al.,
plified models based on channel curvature and its effects on 2007; Wang et al., 2006; Wondzell et al., 2007).
near-bank velocities have been constructed to simulate channel Predictive understanding of aquatic habitat dynamics is an
meandering and migration over long reaches and multidecadal area of special interest in the field of channel restoration.
16 Riverine Habitat Dynamics

Many aquatic restoration activities are channel re-engineering addition, the physical habitat template that provides resources
or reconfiguration projects intended to recover ecological to an organism often comprises that spatial arrangement of
processes by restoring habitat and habitat dynamics. On in- patches around it, in addition to the patch it occupies. A
tensively managed rivers characterized by large impound- comprehensive understanding of the functions of habitat may
ments, bank stabilization, and channel-training structures, require an extended understanding of the functional relations
channel morphology and sediment regime can be uncoupled among many indirectly related habitat patches.
from the flow regime, thereby limiting the ability of flow re- It can be useful to categorize riverine habitat dynamics into
gime restoration alone to restore ecological processes (Brown hydrodynamics and morphodynamics. Hydrodynamics refers
and Pasternack, 2008; Jacobson and Galat, 2006). Channel to habitat variation that results from changes in discharge in
re-engineering projects on such river systems typically seek to the absence of significant change of channel morphology and
increase habitat diversity and allow for some morphodynamic at generally low sediment-transport rates. Hydrodynamic as-
freedom. Some projects seek to minimize costs by con- sumptions are convenient for many assessments of habitat
structing an initial (or pilot) geometry that is intended to dynamics, especially for habitats in rivers with high critical
evolve to an equilibrium state with a more natural suite of shear stresses or in which low-flow habitat questions are em-
physical habitats and habitat dynamics (Jacobson et al., 2001; phasized. Habitat dynamics on hydrodynamic timescales can
Papanicolaou et al., 2010; Weber et al., 2009); other ap- be assessed using conventional hydraulic models to produce
proaches involve designing a river corridor within which the discharge–habitat functions, and, subsequently, time series of
river can be allowed to migrate freely and create dynamic habitat availability that can be evaluated in terms of magni-
habitats (Larsen et al., 2007; Piégay et al., 2005). Under both tude (area or volume), duration, frequency, rate of change,
of these approaches, a predictive understanding of how and timing. Morphodynamic assessments generally involve
habitat dynamics will evolve is desirable. Similarly, the rec- discharge events with lower frequencies that result in changes
ognition that many rivers downstream of dams have dimin- to substrate conditions or channel/floodplain morphology.
ished sediment supplies has resulted in restoration efforts Although such events are rare, they can be particularly im-
that involve augmentation of sediment; predicting the fate portant for life stages of some organisms, such as channel-
of sediment augmentations requires quantitative models for migration events that erode cutbanks and provide new
sediment transport and habitat evolution (Barlaup et al., pointbar surfaces for germination of riparian trees.
2008; Fotherby, 2007; Pasternack et al., 2004; Singer and Understanding of habitat dynamics is increasing in im-
Dunne, 2006). Finally, interest in scaling morphodynamic portance as societal goals shift toward restoration of riverine
models to landscape scales is increasing as agencies and in- ecosystems. Effective investment in restoration strategies re-
stitutions seek to assess regional-scale effects of disturbances quires that the role of physical habitat is correctly diagnosed
such as climate change, urbanization, and agricultural land and that the restoration activity addresses true habitat limi-
conversions (Palmer et al., 2008). tations, including potential needs for dynamic habitats.
Moreover, concerns about the potential effects of climate
change and other land-use stressors indicate a need to extend
12.2.4 Conclusions the understanding of habitat dynamics to watershed and
regional scales (Reuter et al., 2003).
Physical habitats are a fundamental influence on riverine
ecosystem structure and function. In addition to a diversity of References
physical conditions, riverine habitats provide those conditions
in spatial arrangements and with temporal variations that Abad, J.D., Buscaglia, G.C., Garcia, M.H., 2008. 2D stream hydrodynamic, sediment
interact with life-stage needs of riverine biota. Physical habitat transport and bed morphology model for engineering applications. Hydrological
is therefore best understood in terms of the specific species Processes 22(10), 1443–1459.
Ahmadi-Nedushan, B., St-Hilaire, A., Bérubé, M., Robichaud, É., Thiémonge, N.,
and life-stage requirements, for example, adult spawning Bobée, B., 2006. A review of statistical methods for the evaluation of aquatic habitat
habitat, seed germination habitat, chick foraging habitat, or suitability for instream flow assessment. River Research and Applications 22(5),
juvenile migratory habitat. 503–523.
Physical habitats have been highly altered in many river Arthington, A.A., Bunn, S.E., Poff, N.L., Naiman, R.J., 2006. The challenge of
systems by human actions, and restoration of physical habitat providing environmental flow rules to sustain river ecosystems. Ecological
Applications 16(4), 1311–1318.
has become a major emphasis in river restoration. Neverthe- Barlaup, B.T., Gabrielsen, S.E., Skoglund, H., Wiers, T., 2008. Addition of spawning
less, physical habitat is only a part of many abiotic and biotic gravel – a means to restore spawning habitat of atlantic salmon (Salmo salar
processes that interact to support reproduction, growth, and L.), and Anadromous and resident brown trout (Salmo trutta L.) in regulated
survival of riverine biota. Proper diagnosis that habitat has rivers. River Research and Applications 24(5), 543–550.
been impaired and is a limiting ecological factor is necessary Barton, G.J., McDonald, R.R., Nelson, J.M., Dinehart, R.L., 2005. Simulation of flow
and sediment mobility using a multidimensional flow model for the white
for effective restoration. sturgeon critical-habitat reach, Kootenai River near Bonners Ferry, Idaho. US
Habitat can be assessed at spatial scales ranging from the Geological Survey Scientific Investigations Report 2005-5230. US Geological
grain to the extent, that is, from the smallest scale of spatial Survey, Reston, VA, 54 pp.
resolution with which an organism interacts with its environ- Benda, L., Miller, D., Andras, K., Bigelow, P., Reeves, G., Michael, D., 2007.
ment to the largest scale of resolution. For mobile organisms, NetMap: a new tool in support of watershed science and resource management.
Forest Science 53, 206–219.
the grain and extent can differ by many orders of magnitude, Bigelow, P.E., Benda, L.E., Miller, D.J., Burnett, K.M., 2007. On debris flows, river
and a comprehensive assessment of physical habitat related to networks, and the spatial structure of channel morphology. Forest Science 53,
that organism would need to assess the entire range. In 220–238.
 Riverine Habitat Dynamics 17

Borg, D., Rutherfurd, I., Stewardson, M., 2007. The geomorphic and ecological Griffin, E.R., Kean, J.W., Vincent, K.R., Smith, J.D., Friedman, J.M., 2005. Modeling
effectiveness of habitat rehabilitation works: continuous measurement of effects of bank friction and woody bank vegetation on channel flow and
scour and fill around large logs in sand-bed streams. Geomorphology 89(1–2), boundary shear stress in the Rio Puerco, New Mexico. Journal of Geophysical
205–216. Research 110(F4), 1–15,http://dx.doi.org/10.1029/2005JF000322.
Bornette, G., Tabacchi, E., Hupp, C.R., Puijalon, S., Rostain, J.C., 2008. A model of Hall, L.S., Krausman, P.R., Morrison, M.L., 1997. The habitat concept and a plea
plant strategies in fluvial hydrosystems. Freshwater Biology 53, 1692–1705. for standard terminology. Wildlife Society Bulletin 25(1), 173–182.
Bovee, K.D., Lamb, B.L., Bartholow, J.M., Stalnaker, C.B., Taylor, J., Henriksen, J., Haschenburger, J.K., 1999. A probability model of scour and fill depths in gravel-
1998. Stream habitat analysis using the instream flow incremental methodology. bed channels. Water Resources Research 35, 2857–2869.
US Geological Survey Information and Technology Report USGS/BRD-1998- Hazel, J.E., Kaplinski, M., Parnell, R., Manone, M., Dale, A., 1999. Topographic and
0004. US Geological Survey, Reston, VA, 131 pp. bathymetric changes at thirty-three long-term study sites. In: Webb, R.H.,
Bovee, K.D., Waddle, T.J., Jacobson, R.B., 2004. Quantification of habitat patch Schmidt, J.C., Marzolf, G.R., Valdez, R.A. (Eds.), The Controlled Flood in Grand
persistence in rivers affected by hydropeaking. In: Lanfear, K.J., Maidment, D.R. Canyon. American Geophyscial Union Geophysical Monograph Series 110. AGU,
(Eds.), AWRA Spring Specialty Conference: GIS and Water Resources III, May Washington, DC, pp. 161–183.
2004. American Water Resources Association, Nashville, TN, pp. 1–10. Heritage, G., Hetherington, D., 2007. Towards a protocol for laser scanning in
Bowen, Z.H., Bovee, K.D., Waddle, T.J., 2003. Effects of flow regulation on shallow- fluvial geomorphology. Earth Surface Processes and Landforms 32, 66–74.
water habitat dynamics and floodplain connectivity. Transactions of the American Hilderbrand, R.H., Watts, A.C., Randle, A.M., 2005. The myths of restoration
Fisheries Society 132, 809–823. ecology. Ecology and Society 10(1), 19.
Braaten, P.J., Fuller, B.D., Holte, L.D., Viste, W., Brandt, T.F., Legare, R.G., 2008. Hornbach, D.J., 2001. Macrohabitat factors influencing the distribution of Naiads in
Drift dynamics of larval pallid sturgeon and shovelnose sturgeon in a natural the St. Croix River, Minnesota and Wisconsin, USA. In: Bauer, G., Wachtler, K.
side channel of the Missouri River, Montana. North American Journal of (Eds.), Ecology and Evolution of the Freshwater Mussels Unionoida, Ecological
Fisheries Management 28, 808–826. Studies. Springer, Berlin, pp. 213–230.
Bramblett, R.G., White, R.G., 2001. Habitat use and movements of pallid and Jacobson, R.B., Galat, D.L., 2006. Flow and form in rehabilitation of large-river
shovelnose sturgeon in the Yellowstone and Missouri Rivers in Montana and ecosystems: an example from the Lower Missouri River. Geomorphology
North Dakota. Transactions of the American Fisheries Society 130, 1006–1025. 77(3–4), 249–269.
Brown, R.A., Pasternack, G.B., 2008. Engineered channel controls limiting spawning Jacobson, R.B., Gran, K.B., 1999. Gravel routing from widespread, low-intensity
habitat rehabilitation success on regulated gravel-bed rivers. Geomorphology landscape disturbance, Current River Basin, Missouri. Earth Surface Processes
97(3–4), 631–654. and Landforms 24, 897–917.
Cohen, J.B., Wunker, E.H., Fraser, J.D., 2009. Substrate and vegetation selection by Jacobson, R.B., Johnson, H.E., III, Dietsch, B.J., 2009. Hydrodynamic simulations
nesting piping plovers. Wilson Journal of Ornithology 120(2), 404–407. of physical aquatic habitat availability for pallid sturgeon in the Lower Missouri
Crowder, D.W., Diplas, P., 2006. Applying spatial hydraulic principles to quantify River, at Yankton, South Dakota, Kenslers Bend, Nebraska, Little Sioux, Iowa,
stream habitat. River Research and Applications 22, 79–89. and Miami, Missouri, 2006–07. US Geological Survey Scientific Investigations
DeLonay, A.J., Jacobson, R.B., Papoulias, D.M., et al., 2009. Ecological Report 2009-5058. US Geological Survey, Reston, VA, 67 pp.
requirements for pallid sturgeon reproduction and recruitment in the Lower Jacobson, R.B., Laustrup, M.S., Chapman, M.D., 2001. Fluvial processes and
Missouri River – a research synthesis 2005–08. US Geological Survey Scientific passive rehabilitation of the Lisbon Bottom side-channel chute, Lower
Investigations Report 2009-5201. US Geological Survey, Reston, VA, 59 pp. Missouri River. In: Dorava, J.M., Montgomery, D.L., Fitzpatrick, F.A., Palcsak,
DeLonay, A.J., Papoulias, D.M., Wildhaber, M.L., Mestl, G.E., Everitt, D.W., B.B. (Eds.), Fluvial Processes and Physical Habitat. American Geophysical
Chojnacki, K.A., 2007. Movement, habitat use, and reproductive behavior of Union Water Science and Application Series, vol. 4. AGU, Washington, DC,
shovelnose sturgeon and pallid sturgeon in the Lower Missouri River. pp. 199–216.
In: Korschgen, C. (Ed.), Factors Affecting the Reproduction, Recruitment, Habitat, Jacobson, R.B., Laustrup, M.S., D’Urso, G.J., Reuter, J.M., 2004. Physical habitat
and Population Dynamics of Pallid Sturgeon and Shovelnose Sturgeon in the dynamics in four side-channel chutes, Lower Missouri River. US Geological
Missouri River. US Geological Survey Open-File Report 2007-1262. US Survey Open-File Report 2004-1071. US Geological Survey, Reston, VA, 60 pp.
Geological Survey, Reston, VA, pp. 23–102. Jacobson, R.B., Pugh, A.L., 1997. Riparian vegetation and the spatial pattern of
Doyle, M.C., Stanley, E.H., Strayer, D.L., Jacobson, R.B., Schmidt, J.C., 2005. stream-channel instability, Little Piney Creek, Missouri. US Geological Survey
Effective discharge analysis of ecological processes in streams. Water Resources Water Supply Paper 2494. US Geological Survey, Reston, VA, 33 pp.
Research 41(W11411), 1–16,http://dx.doi.org/10.1029/2005WR004222. Jeffries, M., Mills, D., 1990. Freshwater Ecology: Principles and Applications.
Dryer, M.P., Sandvol, A.J., 1993. Recovery Plan for the Pallid Sturgeon Belhaven Press, New York, NY.
(Scaphirhynchus albus). US Fish and Wildlife Service, Bismarck, ND, 55 pp. Joeckel, R.M., Henebry, G.M., 2008. Channel and island change in the lower
Elliott, C.M., Jacobson, R.B., 2006. Geomorphic classification and assessment of Platte River, Eastern Nebraska, USA: 1855–2005. Geomorphology 102(3–4),
channel dynamics in the Missouri National Recreational River, South Dakota and 407–418.
Nebraska. US Geological Survey Scientific Investigations Report 2006-5313. US Johannesson, H., Parker, G., 1989. Linear theory of river meanders. In: Ikeda, S.,
Geological Survey, Reston, VA, 66 pp. Parker, G. (Eds.), River Meandering. Water Resources Monograph 12. American
Elliott, C.M., Reuter, J.M., Jacobson, R.B., 2009. Channel morphodynamics in four Geophysical Union, Washington, DC, pp. 181–213.
reaches of the Lower Missouri River, 2006–07. US Geological Survey Scientific Johnson, H.E., Jacobson, R.B., Delonay, A.J., 2006. Hydroecological modeling of
Investigations Report 2009-5074. US Geological Survey, Reston, VA, 258 pp. the Lower Missouri River. In: Leavesly, G.H. (Ed.), Proceedings of the Third
Everitt, B., 1968. Use of the cottonwood on an investigation of the recent history of Federal Interagency Hydrologic Modeling Conference. Reno, NV, USA, pp. 1–8,
a flood plain. American Journal of Science 266, 417–439. 2–6 April 2006. Subcommittee on Hydrology of the Interagency Advisory
Fotherby, L.M., 2007. Platte River Habitat Recovery: FSM Restoration Actions. Committee on Water Information. ISBN 0-9779007-0-3.
ASCE, Tampa, FL, 537. Jones, K.L., Poole, G.C., Meyer, J.L., Bumback, W., Kramer, E.A., 2006. Quantifying
Fowler, R.T., Death, R.G., 2001. The effect of environmental stability on hyporheic expected ecological response to natural resource legislation: a case stuy of
community struture. Hydrobiologia 445, 85–95. riparian buffers, aquatic habitat, and trout populations. Ecology and Society
Freeman, M.C., Bowen, Z.H., Bovee, K.D., Irwin, E.R., 2001. Flow and habitat 11(2), 1–15.
effects on juvenile fish abundance in natural and altered flow regimes. Kean, J.W., Kuhnle, R.A., Smith, J.D., Alonso, C.V., Langendoen, E.J., 2009. Test of
Ecological Applications 11, 179–190. a method to calculate near-bank velocity and boundary shear stress. Journal of
Frissell, C.A., Liss, W.J., Warren, C.E., Hurley, M.D., 1986. A hierarchical framework Hydraulic Engineering 135(7), 588–601.
for stream habitat classification: viewing streams in a watershed conext. Kinzel, P.J., Nelson, J.M., Wright, C.W., 2006. Monitoring changes in the Platte
Environmental Management 10, 199–214. River riparian corridor with serial LiDAR surveys. US Geological Survey Fact
Gergel, S.E., Turner, M.G., Miller, J.R., Melack, J.M., Stanley, E.H., 2002. Landscape Sheet 2006–2063. US Geological Survey, Reston, VA, 4 pp.
indicators of human impacts to riverine systems. Aquatic Sciences – Research Kirsch, E.M., 1996. Habitat selection and productivity of least terns on the Lower
Across Boundaries 64(2), 118–128. Platte River, Nebraska. Wildlife Monographs 132, 1–48.
Gordon, N.D., McMahon, T.A., Finlayson, B.L., 1992. Stream Hydrology – An Kolar, C.S., Chapman, D.C., Courtenay, W.R.J., Housel, C.M., Williams, J.D.,
Introduction for Ecologists. Wiley, Chichester, 526 pp. Jennings, D.P., 2007. Bigheaded Carps:A Biological Synopsis and Environmental
Gore, J.A., Shields, F.D., Jr., 1995. Can large rivers be restored? Bioscience 45(3), Risk Assessment. Special Publication 33. American Fisheries Society, Bethesda,
142–152. MD, 204 pp.
18 Riverine Habitat Dynamics

Lancaster, J., Downes, B.J., 2009. Linking the hydraulic world of individual Palmer, M.A., Reidy Liermann, C.A., Nilsson, C., Florke, M., Alcamo, J.,
organisms to ecological processes: putting ecology into ecohydraulics. River Lake, P.S., Bond, N., 2008. Climate change and the world’s river basins:
Research and Applications 26(4), 385–403. anticipating management options. Frontiers in Ecology and the Environment
Lancaster, S.T., Bras, R.L., 2002. A simple model of river meandering and its 6(2), 81–89.
comparison to natural channels. Hydrological Processes 16(1), 1–26. Papanicolaou, A.N., Elhakeem, M., Dermisis, D., Young, N., 2010. Evaluation of the
Lancaster, S.T., Hayes, S.K., Grant, G.E., 2003. Effects of wood on debris flow Missouri River shallow water habitat using a 2D-hydrodynamic model. River
runout in small mountain watersheds. Water Resources Research 39(6), 1–21. Research and Applications, doi:10.1002/rra.1344.
Landres, P.B., Morgan, P., Swanson, F.J., 1999. Overview of the use of natural Parasiewicz, P., 2007. The MesoHabSim model revisited. River Research and
variability concepts in managing ecological systems. Ecological Applications Applications 23(8), 1–11.
9(4), 1179–1188. Pasternack, G.B., Wang, C.L., Merz, J.E., 2004. Application of a 2D hydrodynamic
Langendoen, E.J., Wells, R.R., Thomas, R.E., Simon, A., Bingner, R.L., 2009. model to design of reach-scale spawning gravel replenishment on the
Modeling the evolution of incised streams. III: model application. Journal of Mokelumne River, California. River Research and Applications 20, 205–225.
Hydraulic Engineering 135(6), 476–486. Pickett, S.T.A., White, P.S., 1985. The Ecology of Natural Disturbance and Patch
Larsen, E.W., Girvetz, E.H., Fremier, A.K., 2007. Landscape level planning in alluvial Dynamics. Academic Press, Orlando, FL, 472 pp.
riparian floodplain ecosystems: using geomorphic modeling to avoid conflicts Piégay, H., Darby, S.E., Mosselman, E., Surian, N., 2005. A review of techniques
between human infrastructure and habitat conservation. Landscape and Urban available for delimiting the erodible river corridor: a sustainable approach to
Planning 79(3–4), 338–346. managing bank erosion. River Research and Applications 21(7), 773–789.
Larsen, E.W., Greco, S.E., 2002. Modeling channel management impacts on Poff, N.L., Allan, J.D., Bain, M.B., et al., 1997. The natural flow regime. Bioscience
river migration: a case study of woodson bridge state recreation area, 47, 769–784.
Sacramento River, California, USA. Environmental Management 30(2), Poff, N.L., Richter, B.D., Arthington, A.H., et al., 2009. The ecological limits of
209–224. hydrologic alteration (ELOHA): a new framework for developing regional
Le Fer, D., Fraser, J.D., Kruse, C.D., 2008. Piping plover foraging-site selection on environmental flow standards. Freshwater Biology 55, 147–170.
the Missouri River. Waterbirds 31(4), 587–592. Pollen-Bankhead, N., Simon, A., 2010. Hydrologic and hydraulic effects of riparian
Lechowicz, M.J., 1982. The sampling characteristics of electivity indices. Oecologia root networks on streambank stability: is mechanical root-reinforcement the
52(1), 22–30. whole story? Geomorphology 226, 353–362.
Lisle, T.E., Nelson, J.M., Pitlick, J., Madej, M.A., Barkett, B.L., 2000. Variability of Power, M.E., Stewart, A.J., 1987. Disturbance and recoery of an algal assemblage
bed mobility in natural, gravel-bed channels and adjustments to sediment load following flooding in an Oklahoma stream. American Midland Naturalist 117,
at local and reach scales. Water Resources Research 36(12), 3743–3755. 333–345.
Lytle, D.A., Merritt, D.M., 2004. Hydrologic regimes and riparian forests: a Pringle, C.M., Naiman, R.J., Bretschko, G., et al., 1988. Patch dynamics in lotic
structured population model for cottonwood. Ecology 85(9), 2493–2503. systems: the stream as a mosaic. Journal of the North American Benthological
Mahoney, J.M., Rood, S.B., 1998. Streamflow requirements for cottonwood seedling Society 7(4), 503–524.
requirement – an integrative model. Wetlands 18, 634–638. Reuter, J.M., Jacobson, R.B., Elliott, C.M., 2003. Physical stream habitat dynamics
Mathews, R., Richter, B.D., 2007. Application of the indicators of hydrologic in Lower Bear Creek, Northern Arkansas. US Geological Survey Biological
alteration software in environmental flow setting. Journal of the American Water Sciences Report 2003–0002. US Geological Survey, Reston, VA, 49 pp.
Resources Association 43(6), 1400–1413. Reuter, J.M., Jacobson, R.B., Elliott, C.M., DeLonay, A.J., 2009. Assessment of
Matthaei, C.D., Townsend, C.R., 2000. Long-term effects of local disturbance history Lower Missouri River physical aquatic habitat and its use by adult sturgeon
on mobile stream invertebrates. Oecologia 125(1), 119–126. (genus Scaphirhynchus) 2005–07. US Geological Survey Scientific
McDonald, R.R., Barton, G.J., Nelson, J.M., Paragamian, V.L., 2006. Modeling Investigations Report 2009-5121. US Geological Survey, Reston, VA, 81 pp.
hydraulic and sediment transport processes in white sturgeon spawning habitat Reuter, J.M., Jacobson, R.B., Elliott, C.M., Johnson, H.E., III, DeLonay, A.J., 2008.
on the Kootenai River. Joint 8th Federal Interagency Sedimentation and 3rd Hydraulic and substrate maps of reaches used by sturgeon (genus
Federal Interagency Hydrologic Modeling Conference, Reno, NV, 8 pp. Scaphirhynchus) in the Lower Missouri River, 2005–07. Data Series Report 386.
McDonald, R.R., Nelson, J.M., Bennett, J.P., 2005. Multi-dimensional surface-water US Geological Survey, Reston, VA, 442 pp.
modeling system user’s guide. US Geological Survey Techniques and Methods Rhoads, B.L., Wilson, D., Urban, M.L., Herricks, E.E., 1999. Interaction between
6-B2. US Geological Survey, Reston, VA, 136 pp. scientists and nonscientists in community-based watershed management:
McElroy, B., Mohrig, D., 2009. Nature of deformation of sandy bed forms. Journal emergence of the concept of stream naturalization. Environmental Management
of Geophysical Research 114, 13. 24, 297–308.
McGarigal, K., Marks, B.J., 1995. FRAGSTATS – spatial pattern analysis program Richter, B.D., Baumgartner, J.V., Powell, J., Braun, D.P., 1996. A method for
for quantifying landscape structure: Portland, Oregon. US Department of assessing hydrological alteration within ecosystems. Conservation Biology 10(4),
Agriculture, Forest Service, General Technical Report PNW-351, 122 pp. 1163–1174.
McKenney, R., 2001. Comparison of channel changes and habitat diversity in a Schmeeckle, M.W., Nelson, J.M., 2003. Direct numerical simulation of bedload
warm-water, gravel-bed stream. In: Dorava, J.M., Montgomery, D.R., Palscak, B., transport using a local, dynamic boundary condition. Sedimentology 50,
Fitzpatrick, F.A. (Eds.), Geomorphic Processes and Riverine Habitat. American 279–301.
Geophysical Union, Water Science and Applications, Washington, DC, Schmidt, J.C., Graf, J.B., 1990. Aggradation and degradation of alluvial-sand
pp. 57–71. deposits, 1965 to 1986, Colorado River, Grand Canyon National Park, Arizona.
McKenney, R., Jacobson, R.B., Wertheimer, R.C., 1995. Woody vegetation and US Geological Survey Professional Paper 1493. US Geological Survey, Reston,
channel morphogenesis in low-gradient, gravel-bed streams in the Ozarks VA, 74 pp.
Region, Missouri and Arkansas. Geomorphology 13, 175–198. Schrank, S.J., Braaten, P.J., Guy, C.S., 2001. Spatiotemporal variation in density of
Miller, A.C., 2006. Experimental gravel bar habitat creation in the Tombigbee River, larval bighead carp in the Lower Missouri River. Transactions of the American
Mississippi. US Army Corps of Engineers, Engineering Research and Design Fisheries Society 130(5), 809–814.
Center, ERDC/TN EMRRP-ER-03, 11 pp. Schwendel, A.C., Death, R.G., Fuller, I.C., 2010. The assessment of shear stress
Mohrig, D., Smith, J.D., 1996. Predicting the migration rates of subaqueous dunes. and bed stability in stream ecology. Freshwater Biology 55, 261–281.
Water Resources Research 32, 3207–3217. Shields, F.D., Jr., Simon, A., Steffen, L.J., 2000. Reservoir effects on downstream
Montgomery, D.R., 1999. Process domains and the river continuum. Journal of the river channel migration. Environmental Conservation 27, 54–66.
American Water Resources Association 35, 397–410. Sietman, B.E., Whitney, S.D., Kelner, D.E., Blodgett, K.D., Dunn, H.L., 2001. Post-
National Research Council, 2004. Endangered and Threatened Species of the Platte extirpation recovery of the freshwater mussel (Bivalvia: Unionidae) fauna in the
River. National Academies Press, Washington, DC, 300 pp. upper Illinois River. Journal of Freshwater Ecology 16(2), 273–281.
Palmer, M., Allan, J.D., Meyer, J., Bernhardt, E.S., 2007. River restoration in the Singer, M.B., Dunne, T., 2006. Modeling the influence of river rehabilitation
twenty-first century: data and experiential knowledge to inform future efforts. scenarios on bed material sediment flux in a large river over decadal timescales.
Restoration Ecology 15(3), 472–481. Water Resources Research 42, 1–14.
Palmer, M.A., Bely, A.E., Berg, K.E., 2005a. Response of invertebrates to lotic Smith, J.D., 2007. Beaver, willow shrubs, and floods. In: Johnson, E.A., Miyanishi,
disturbance: a test of the hyporheic refuge hypothesis. Oecologia 89(2), 182–194. K. (Eds.), Plant Disturbance Ecology. Elsevier, New York, NY, pp. 603–672.
Palmer, M.A., Bernhardt, E.S., Allan, J.D., et al., 2005b. Standards for ecologically Smith, J.W., Renken, R.B., 1991. Least tern nesting habitat in the Mississippi River
successful river restoration. Journal of Applied Ecology 42(2), 208–217. Valley adjacent to Missouri. Journal of Field Ornithology 62, 497–504.
 Riverine Habitat Dynamics 19

Sommer, T., Armor, C., Baxter, R., et al., 2007. The collapse of pelagic fishes in the Wallick, J.R., Lancaster, S.T., Bolte, J.P., 2006. Determination of bank erodibility for
Upper San Francisco estuary. Fisheries 32(6), 270–277. natural and anthropogenic bank materials using a model of lateral migration and
Sowa, S.P., Annis, G., Morey, M.E., Diamond, D.D., 2007. A GAP analysis and observed erosion along the Willamette River, Oregon, USA. River Research and
comprehensive conservation strategy for riverine ecosystems of Missouri. Applications 22, 631–649.
Ecological Monographs 77(3), 301–334. Wang, L., Seelbach, P.W., Hughes, R.M., 2006. Introduction to landscape influences
Strayer, D.L., 2008. Freshwater Mussel Ecology – A Multifactor Approach to on stream habitats and biological assemblages. American Fisheries Society
Distribution and Abundance. Freshwater Ecology Series. University of California Symposium. The American Fisheries Society, Bethesda, MD, pp. 1–23.
Press, Berkeley, CA, 204 pp. Weber, C., Schager, E., Peter, A., 2009. Habitat diversity and fish assemblage
Thorp, J.H., Thoms, M.C., Delong, M.D., 2006. The riverine ecosystem synthesis: structure in local river widenings: a case study on a swiss river. River Research
biocomplexity in river networks across space and time. River Research and and Applications 25(6), 687–701.
Applications 22(2), 123–147. Wiele, S.M., Wilcock, P.R., Grams, P.E., 2007. Reach-averaged sediment routing
US Fish and Wildlife Service, 2000. Biological opinion on the operation of the model of a canyon river. Water Resources Research 43, 1–16.
Missouri River main stem reservoir system, operation and maintenance of Wiens, J.A., 1989. Spatial scaling in ecology. Functional Ecology 3(4), 385–397.
the Missouri River bank stabilization and navigation project, and operation of Wolman, M.G., Gerson, R., 1978. Relative scales of time and effectiveness
the Kansas River reservoir system. US Fish and Wildlife Service, Bismarck, ND., of climate in watershed geomorphology. Earth Surface Processes 3,
308 pp. 189–208.
US Fish and Wildlife Service. 2003. Amendment to the 2000 biological opinion on Wondzell, S.M., Hemstrom, M.A., Bisson, P.A., 2007. Simulating riparian vegetation
the operation of the Missouri River main stem reservoir system, operation and and aquatic habitat dynamics in response to natural and anthropogenic
maintenance of the Missouri River bank stabilization and navigation project, and disturbance regimes in the Upper Grande Ronde River, Oregon, USA. Landscape
operation of the Kansas River reservoir system. US Fish and Wildlife Service, and Urban Planning 80, 249–267.
Minneapolis, MN 308 pp. Zanoni, L., Gurnell, A., Drake, N., Surian, N., 2008. Island dynamics in a braided
Wallick, J.R., Grant, G.E., Lancaster, S.T., Bolte, J.P., Denlinger, R.P., 2008. Patterns river from analysis of historical maps and air photographs. River Research and
and controls on historical channel change in the Willamette River, Oregon, USA. Applications 24(8), 1141–1159.
In: Gupta, A. (Ed.), Large Rivers: Geomorphology and Management. Wiley, New
York, NY, pp. 491–516.

Biographical Sketch

Dr. Robert B. Jacobson received his PhD from the Department of Geography and Environmental Engineering,
Whiting School of Engineering, The Johns Hopkins University, and he holds an undergraduate degree in geology
from Carleton College. He has worked for the US Geological Survey for 25 years on subjects including geologic
hazards, neotectonics, paleoseismology, geomorphology, and surficial processes. Currently, he is chief of the River
Studies Branch, Columbia Environmental Research Center, Columbia, Missouri, USA, where he supervises a staff
of physical, chemical, and biological scientists engaged in interdisciplinary research in fundamental river pro-
cesses. His present research focuses on river corridor habitat dynamics, with an emphasis on improving the
scientific foundation for restoration and management of large, multipurpose rivers.
 12.3 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and
Processes: Problem Statements and Challenging Issues
Y-F Le Lay and H Piégay, Université de Lyon, Lyon, France
B Moulin, AEMGEO, Lyon
r 2013 Elsevier Inc. All rights reserved.

12.3.1 Introduction 21
12.3.2 Space–Time Framework of Wood Dynamics 22
12.3.2.1 Input Processes of LW 22
12.3.2.2 Controls on Transfer Processes of LW 23
12.3.2.2.1 Hydraulic processes controlling wood transport 23
12.3.2.2.2 Amount of in-channel wood and ability to move 24
12.3.2.2.3 Residence time and decomposition 25
12.3.2.3 Regional Characters Controlling Abundance, Distribution, and Residence Time of LW 25
12.3.3 LW Effects on Fluvial Processes, Channel Morphology, and Riparian Features 27
12.3.3.1 Wood Jam Effects on the Hydraulic Conditions and Transport of Material 27
12.3.3.2 LW Effects on Channel Morphology and Grain Size Pattern 27
12.3.3.3 Relative Influence of LW According to Channel Size 29
12.3.3.4 Geomorphological Effects of LW on Riparian Areas 30
12.3.4 In-Channel Wood and River Management 31
12.3.4.1 Ecological Benefits of LW 31
12.3.4.2 Risks and Nuisances Associated with LW 31
12.3.4.3 Applications of Knowledge in Terms of Restoration and Sustainable Management 33
References 33

Glossary in World Rivers (held in Oregon in October 2000) agreed

Ecological benefits Ecological benefits designate that the expression ‘‘woody debris,’’ commonly used over
environmental functions, goods, and services that humans the last 2 decades by scientists, now appeared inappropriate
and organisms enjoy from ecosystems. For instance, wood and negative and would be preferably replaced by
in streams and rivers has positive consequences on the ‘‘in-channel wood.’’
abundance and diversity of aquatic communities of fishes Maintenance The maintenance of streams and rivers aims
and benthic macroinvertebrates. to ensure appropriate conditions of channel and banks for
Geomorphic hazard A geomorphic hazard is a process or river and land uses. It generally enhances free-flow
event in the physical environment that has the potential to condition and consists in a set of tasks, including the
negatively affect humans, their activities or the (regular) clearing of channels, the removal of sediment and
environment. For instance, in-channel large wood gravel, the removal of dead and living trees, the removal of
may damage structures (roads, bridges, and culverts), any obstruction, the pruning of ligneous vegetation and the
raise the elevation of water surface, or increase lateral mowing of herbaceous vegetation, and the removal of
erosion. logjams and in-channel wood, floating or not.
Glide Glide is a habitat characterized by attributes of both Pool The pools designate deep stream bed features where
pools and riffles. It designates a stream reach with the water velocity is quite slow compared to other features.
moderately shallow water. Immediately downstream of a Different types of pools may be distinguished, such as
pool, glide begins where the flow velocity increases. secondary channel pools, backwater pools, lateral scour
Large wood (LW) Large wood includes a wide variety of pools, or plunge pools.
material types, including whole trees, snags, tree tops, logs, Restoration Restoration consists of a set of
chunks of wood, limbs, branches, stumps, and root wads. actions designed to improve a degraded or simplified
A piece of LW may be operationally defined as greater than environment due to human pressures on structures and
10 cm in diameter and more than 1 m long. As LW has functions. It is based on different conditions and objectives
significant hydraulic, geomorphic, and biological effects, (ecological improvement, ecosystem services, or human
participants at the First International Conference on Wood benefits).

Le Lay, Y.-F., Piégay, H., Moulin, B., 2013. Wood entrance, deposition,
transfer and effects on fluvial forms and processes: problem statements and
challenging issues. In: Shroder, J. (Editor in Chief), Butler, D.R., Hupp, C.R.
(Eds.), Treatise on Geomorphology. Academic Press, San Diego, CA, vol. 12,
Ecogeomorphology, pp. 20–36.

20 Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00320-1

 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and Processes 21

Riffle Riffles designate bed features with the steepest abundance reflects the balance between input and output
slopes and shallowest depths. Rapids and cascades have processes.
higher gradients than riffles. Wood jam When individual pieces of LW are transported
Stream habitat Stream habitat designates a spatial unit by flows and deposited against obstructions, they may
where an animal or plant lives or satisfies ecological gather and form accumulations, sometimes very massive, so
functions. Micro- and mesohabitats are usually called ‘‘jams.’’ A log jam is an obstacle formed by the
distinguished, the first one being point-based and described crowding together of a number of floating logs.
in terms of local hydraulic conditions (grain size, depth, Wood reintroduction Wood reintroduction is a method
velocity, and turbulence) whereas the second one increasingly used in restoration projects to improve the
corresponds to geomorphic features to which ecological hydraulic, morphological, and ecological status of degraded
functions are attributed. A typical stream contains three streams and rivers. The placement of stable wood structures
major types of mesohabitats that may be characterized by a may contribute to local bank protection and to the active
dominant platform or physical characteristics: riffles, runs, creation of in-stream habitats on a reach scale.
and pools. Some stream mesohabitats, such as plunge Wood transport In rivers, wood transport designates the
pools, may be caused by large woods or boulders. movement of large woods and occurs according to three
Wood budget Large wood in streams and rivers is a distinct regimes: (1) the logs move with rare piece-to-piece
consequence of the balance between wood recruitment contacts (uncongested transport); (2) the logs move as a
from terrestrial sources, hydraulic transport of wood into single mass (congested transport); and (3) logs move either
and out of the considered reach in relation to reach individually or in clumps (semi-congested transport). The
roughness and obstacles, and in situ alteration of wood due transport of large wood is a function of stream size and
to decomposition or physical abrasion. As wood budget increases with the flood magnitude.
derives from different components, the large wood

Abstract

Throughout the world before the 1970s, in-channel large wood (hereafter LW), was generally considered a nuisance or a
hazard to be avoided because of the hydraulic effects and consequential associated risks that it could produce. LW was
systematically removed from channels for a perceived benefit to human activities. Over the last four decades, LW has
received an increasing interest among scientists who recognize it as a significant structural and functional component of
aquatic ecosystems. Research in geomorphology has addressed LW characters and its effects on flow hydraulics, on channel
and valley forms and their evolution, on the storage and transfer of sediment and organic matter, as well as on associated
habitats for aquatic communities. After considering the space–time framework within which the wood dynamics can be
studied, we highlight the LW effects on channel morphology and fluvial processes, the ecological and societal consequences
of wood in rivers and implications in terms of river management and restoration. After centuries of wood removal, LW is
now introduced in rivers for ecological improvement in some areas of North America, Europe, Australia, and Japan.

12.3.1 Introduction debris (FOD) composed of such material as twigs and leaves.
Moreover, individual pieces of LW may gather and form small
All aquatic ecosystems are intimately associated with the sur- accumulations and massive jams when they are transported by
rounding terrestrial landscape (Naiman and Décamps, 1997). flows and deposited against obstructions.
Wood pieces introduced in channels of woodland rivers give Though streams and rivers commonly drain overall forested
obvious evidence of this relationship. As suggested by the floodplains at least in the temperate zone, Keller and Swanson
discussions that occurred during the second conference on (1979) established that ‘‘Most research in fluvial geomorph-
Wood in Rivers held in Scotland in August 2006, fluvial geo- ology has focused on interactions among flowing water,
morphologists do not continue to use the term ‘debris,’ pre- sediment concentration, and channel morphology. Little in-
ferring the terms ‘in-channel wood’, ‘large wood’ (LW) if a formation is available concerning the effects of various types of
large size is reached or only wood when there is no risk of large organic material on channel form and processes.’’ Never-
misunderstanding. Such wood then includes a wide variety of theless, since the end of 1960s, LW has received an increasing
material types and sizes. It consists of whole trees, snags, tree interest among scientists who finally recognized it as a signifi-
tops, logs, chunks of wood (resulting from the degradation of cant structural and functional component of aquatic eco-
logs and snags), limbs, branches, stumps, and root wads. systems. Pioneering research mainly focused on small streams
The size used by geomorphologists to define LW varies widely (Zimmerman et al., 1967) draining mountainous terrain
among studies achieved in different academic disciplines (Heede, 1972), particularly in the Pacific Northwest region of
and geographic areas. It makes exact comparisons difficult. North America (Swanson et al., 1976), where human impacts
A piece of LW may be operationally defined as greater than on the forest cover are relatively recent. More recent scientific
10 cm in diameter (e.g., Keller and Swanson, 1979; Bryant, literature has tended to weaken this regional bias (Gregory
1981) and more than 1 m long (e.g. Bilby and Ward, 1989). Its et al., 2003). Studies have mostly been pursued in the
dimensions are larger than fine wood (FW) and fine organic temperate zones, in other parts of North America, as well as
22 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and Processes

South America, Europe, South Africa, Australia, New Zealand, size, they have geomorphic effects on their environment where
and Japan, where river systems have experienced human dis- they are lodged in channel, locally influencing flow hydraulics.
turbance to a large extent. In 2000s, investigations have also The ‘wood-driven geomorphology’ is therefore complex in
stressed the tropical zone (Cadol et al., 2009; Wohl et al., 2009), space and time because of various conditions of material
the semi-arid zone (Pettit et al., 2005), and the sub-Antarctic availability, and multi-scaled conditions to introduce, trans-
zone (Mao et al., 2008). Over the last 40 years, research in port, and store it. In-channel wood loading strongly varies not
geomorphology has addressed LW characters and its effects on only at a point in time but also among different parts of
flow hydraulics, on channel and valley forms and their evo- network. At any site, the amount of in-channel LW is a con-
lution, on the storage and transfer of sediment and organic sequence of the balance between wood recruitment from ter-
matter, as well as on associated habitats for aquatic com- restrial sources, hydraulic transport of wood into and out of
munities. Management implications are considered not only the considered reach in relation to reach roughness and obs-
through empirical analyses and recommendations but also with tacles, and in situ decay of wood. Thus, LW abundance reflects
new research issues on restoration strategies or impact assess- the balance between input and output processes (Keller and
ment with experimentation on wood addition or removal. Swanson, 1979). Wood budgeting is therefore becoming an
This chapter aims to summarize the main results on the life important research issue used to understand all the cascading
cycle of the wood in rivers and its geomorphic role within transfer of wood from the source to the sea. Benda and his
world rivers. We consider the space–time framework within co-authors provided a quantitative framework to consider it
which the wood dynamics can be studied, the LW effects on at the catchment scale (e.g., Benda and Sias, 2002). MacVicar
channel morphology and fluvial processes, and the ecological et al. (2009) introduced a series of techniques to quantify the
and societal consequences of wood in rivers and implications components of a wood budget.
in terms of river management.
12.3.2.1 Input Processes of LW
12.3.2 Space–Time Framework of Wood Dynamics In-channel LW recruitment includes chronic forest mortality;
toppling of trees due to fires and windstorms; inputs from
The wood, similar to the bedload or the suspended load, is bank erosion; wood supplied by slope movements and pro-
an external material which is introduced, transported, and cesses such as landslides, debris flows, and snow avalanches;
deposited. Because some pieces are large relative to the river and exhumation of buried wood (Figure 1; Benda and Sias,

Alteration of
Input processes Output processes
standing large wood

Physical processes
• Blowdown
• Mass wasting
• Bank erosion
Chronic production agents

• Flotation

Biological processes
• Tree mortality due to:
Biological degradation Uncongested,
- old age
semi-congested
- insect infestations • Microbial decay
and congested
- diseases
• Invertebrate consumption transport
• Large litter fall
• Beaver activity
Physical breakdown Debris flow
• Debris flow
• Severe flood
• Forest fire Leaching
Episodic events

• Icing
during winter storms
• Blowdown
due to high winds
• Earthquake

Figure 1 Input and output processes of LW. Modified from Keller and Swanson, 1979; Swanson et al., 1982; Bisson et al., 1987; Keller and
Macdonald, 1995; Braudrick et al., 1997; Bilby and Bisson, 1998.
 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and Processes 23

2002). If the major part of the total wood input is generally channel, which may stimulate lateral erosion on both banks
derived from the lateral margins within a distance equal to the and further destabilize riverside trees. Undercutting of stream-
height of streamside trees (Murphy and Koski, 1989), up- banks may weaken the root system before causing living trees
stream slope processes also deliver a substantial quantity of tip into the channel. Bank erosion may also remobilize wood
LW to small, steep mountain channels where rates of bank stored on the floodplain and exhume wood buried in alluvium
erosion are likely to be minimal (May and Gresswell, 2003b). and colluviums. Floods transfer LW from upstream, remove it
LW comes from upland forests on steep slopes, the foot of from floodplains, and tear trees from the banks. Biological
adjacent hillslopes, or from near-stream riparian zones. Ob- processes are also significant with tree mortality due to old age,
servations on large rivers highlight the contribution to LW input insect infestations, diseases, large litter fall, and beaver activity.
by both floodplains and hillslopes. For instance, along the Stress induced by flooding may facilitate insect infestation.
Rhône River, France, LW trapped in the Génissiat Reservoir was Beaver may be an important production agent of small pieces of
comprised of 44% softwood species, 18% hardwood pieces, wood (Rosenfeld and Huato, 2003).
and 25% conifers (Figure 2). If the majority of LW comprised Temporal patterns are also complex, even if not well
of hardwood originated from the riparian corridor, coniferous known. Episodic events are very critical in terms of wood
fragments came from hillslopes (Moulin and Piégay, 2004). input. Vegetation is naturally prone to episodic processes
LW enters channels via physical/biological and chronic/ (Bisson et al., 1987; Bilby and Bisson, 1998). Large and in-
episodic processes (Keller and Swanson, 1979; Swanson frequent events may provide considerable amounts of LW to
et al., 1982; Keller and Macdonald, 1995). Geomorphological channels of all sizes in a short time. They include debris
processes may play an important role. Chronic production flows, severe floods, forest fires, icing during winter storms,
agents deliver single pieces or small number of trees at and blowdowns due to high winds. During the winter
short time intervals. Physical processes consist of blowdown, of 1811–12, the so-called New Madrid earthquake struck the
mass wasting, bank erosion, and flotation. Strong winds central Mississippi Valley so severely that ‘‘the river was
uproot entire trees, snap tree tops, and break branches. temporarily covered with wreckages and debris, snags and
Trees that grow on wet sites may be more susceptible to wind- sawyers multiplied, the banks caved, and islands disappeared’’
throw than those on dry sites because of shallow rooting (Fuller, 1912). Likewise, in a steep and highly dissected
(Harmon et al., 1986). Likewise, the configuration of con- watershed of the coastal Oregon, Reeves et al. (2003) under-
stricted valleys increases the Venturi effect and the damages due lined the predominant role of episodic events: their results
to wind. Slow mass movements on hillslopes include soil creep, show that 65% of the number of pieces and 46% of the esti-
slump, and earthflow. They tip trees, making them more sen- mated volume of wood come from upslope sources delivered
sitive to wind-throw, and they also cause riverside trees to by mass wasting, whereas streamside sources adjacent to the
topple over the channel. Moreover, mass wasting constricts the channel contributed about 35% of the number of pieces and
54% of the estimated volume of wood. Thus, the topographic
features of a watershed and channel size strongly influence the
50 relative contribution by wood sources as well as storage and
output processes (May and Gresswell, 2003b). In a moun-
45 tainous neotropical watershed of central Panama, a storm
provoked the introduction of wood and the formation of large
40
logjams ‘‘via bank erosion (a relatively minor source), land-
35 slides along the valley walls of the main channel, and tributary
inputs that include both landslides and bank erosion along
Frequency (%)

30 the tributary’’ (Wohl et al., 2009). These processes interact to

form complex cascades of disturbances that influence the
25
amount of wood entering a stream (Evans et al., 1993).
20
12.3.2.2 Controls on Transfer Processes of LW
15
Unlike sediment, little is known about when, how, and the
10 distance wood is transported in channels and where it is
5 deposited, given that large floods capable of moving LW are
relatively infrequent (Braudrick and Grant, 2001).
0
Riparian

Hardwood

(too decomposed)
species

12.3.2.2.1 Hydraulic processes controlling wood

species

Not determined
Conifer

transport
These processes are not well known; yet, they control wood
transport. Several approaches were considered: (1) the intro-
duction of wood in hydraulic modelling progressively identi-
fying the contribution of the different physical parameters in
Figure 2 Distribution of wood pieces trapped by the Genissat dam’s different hydraulic contexts (e.g. Shields and Gippel, 1995);
reservoir according to the state of their extremities. Modified from (2) the interactions of wood and flow during the transport
Moulin and Piégay (2004). in flume experiments (e.g., Young, 1991); and also (3) the
24 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and Processes

interactions of wood and sediments in debris flows (Lancaster wood-poor debris flows and deposit on steeper gradients
et al., 2003). Various characteristics of a piece of wood affect (Lancaster et al., 2003).
its likelihood of movement (e.g., wood density, buoyancy,
orientation, size, and form related to flow depth, velocity,
and roughness). By flume experiments, Braudrick et al. 12.3.2.2.2 Amount of in-channel wood and ability to
(1997) have shown three distinct wood transport regimes: move
uncongested, congested, and semi-congested. During un- Bilby and Bisson (1998) reported that the highest level of
congested transport, logs move with rare piece-to-piece wood biomass is about 180 kg m 2 in some streams flowing
contacts, typically occupy less than 10% of the channel area, through the redwood forests of northern California. In un-
and independently rotate or roll in the presence of an managed upland streams of Redwood Creek basin, California,
obstruction such as boulders. In congested transport, the logs wood storage ranges from 150 to 2500 t ha 1 (Keller and
move as a single mass, occupy more than 33% of the channel Macdonald, 1995). Larger river reaches present much lower
area, and cannot independently respond to obstacle due to amounts, because the abundance and distribution of in-
many piece-to-piece contacts. Semi-congested transport occurs channel LW are chiefly influenced by channel size and the
when logs occupy between 10% and 33% of the channel area ability of wood to move (Marcus et al., 2002; Piégay, 2003;
and move either individually or in clumps. Moulin et al., 2011). LW mobility/stability is controlled by
Though debris flows are commonly considered as two- many factors, including piece dimensions, wood density,
phase systems of sediment and water, wood may represent rootwad attachment, and the degree of anchoring to the bed
a third phase in forested environments. Wood content of and banks of channel. Most mobile pieces of wood are shorter
in-channel stored sediment may influence debris flow initi- than bank-full channel width. Pieces with large diameters re-
ation, runout lengths, and locations of deposition (Swanson quire a deeper flow to be entrained and travel less far (Bilby
et al., 1976; Brayshaw and Hassan, 2009). As a debris flow and Ward, 1989). Therefore, smaller LW moves farther than
moves down the channel, it entrains LW and living vegetation larger ones. Pieces oriented at 901 to flow direction are less
from streambed and the foot of adjacent hillslopes. Lancaster stable than those oriented at 01 (Bocchiola et al., 2006).
et al. (2003) ‘‘proposed that (1) entrainment of wood by Rootwads decrease LW mobility by anchoring it to the bed and
debris flows reduces velocity because momentum conser- banks and by increasing drag (Abbe and Montgomery, 1996).
vation requires that addition of wood mass be compensated All elements that can reduce flow velocity may also reduce
by a loss in velocity and (2) flow direction angle changes LW movement in channels. Thus, riparian vegetation on banks
(bends) reduce velocity because wood causes debris flows and in channel may hinder movement of large pieces of wood,
to behave more like objects colliding with valley walls facilitate mineral deposition, and anchor LW (Figure 3).
than like a fluid that flows between them.’’ Debris flows Wood accumulates wherever the roughness increases and is
highly laden with wood should travel shorter distances than deposited at characteristic locations, including the outside of

D. LWD in upper bars

deposited during falling stage
A. Isolated jam

C. Jams on concave G. Enchored trees

banks of overbank flow
E. Isolated trunk on channels
point bar

F. Strand line
A. Isolated jam
trapped by standing
shrubs

B. Debris
line on concave
bank

Piedmont Lowland

Braided river Free meandering river

Figure 3 Typology of large wood accumulations observed in large rivers. Reproduced from Gurnell, A.M., Piégay, H., Swanson, F.J., Gregory,
S.V., 2002. Large wood and fluvial processes. Freshwater Biology 47, 601–619. p. 615, with permission from Wiley.
 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and Processes 25

bends (concave bank of meanders, tributary junctions, and transfer downstream in solution play a minor role. LW has a
valley bends), as well as the head of vegetated islands and low surface-to-volume ratio and contains more polymeric
mid-channels bars, shallow riffles, entrances to overbank flow material than soluble substances. ‘‘As microbes transform
channels, and immobile obstacles such large boulders, trees, these polymers to soluble material, leaching may increase’’
logjams, hydraulic structures, and bridges (Piégay et al., 1999; (Harmon et al., 1986). By contrast, wood fragmentation
Braudrick and Grant, 2001; Moulin et al., 2011). (i.e., the physical breakdown by flowing water) is particularly
As channel size increases, LW abundance decreases whereas efficient in streams and is accelerated by biological activity.
average diameter, length, and volume of LW pieces increase Biological degradation of wood includes microbial decay and
(Bilby and Ward, 1989, 1991; Abbe and Montgomery, 1996). invertebrate consumption. Degradation of wood in aquatic
Fluvial transport of wood typically increases because indi- ecosystem is much slower than in terrestrial ecosystems
vidual pieces of wood are generally shorter than the bank-full because of anaerobic conditions (Keller and Swanson, 1979).
width (Bilby and Bisson, 1998). As a result, LW moves more The terrestrial decomposer guild is more diverse than the
frequently and the transport distance of mobile pieces in- aquatic guild (Martius, 1997). However, a previous sub-
creases in larger channels (Bilby and Ward, 1991). In large mersion accelerates the terrestrial decay. The imbibitions and
rivers, LW appears to be more often oriented downstream with alternation of immersion and emersion of wood form cave-
the flow – with the rootwad upstream – than in headwater like structures, enhancing wood surface exposed to exterior
streams (Gippel et al., 1996). Though the frequency of attacks and increasing its textural complexity. Decomposition
logjams decreases downstream, their size increases. Interjam generally occurs from the periphery toward the interior of
spacing also increases downstream (Martin and Benda, 2001). wood because the distribution and lignocellulolytic activity
of the microbial community is firstly restricted to the surface
12.3.2.2.3 Residence time and decomposition of submerged wood (Aumen et al., 1983).
If LW moves unhindered downstream, it may reside in channels Bulkier and harder to decompose and consume, wood is
for decades to centuries (Gippel et al., 1996). This is associated naturally more refractory than leaves, barks, seeds, and flowers.
with the channel capacity to retain wood and the properties of As wood is characterized by its high lignocellulose and low
wood to be decomposed and fragmented by abrasion processes. nitrogen content, in comparison with other plan tissues, it is
The amount of in-channel LW observed in a river channel may generally a poor nutritional resource. Nevertheless, it provides
be considered in light of the interrelationship between the carbon for organisms capable of digesting its principal com-
stream reach and its environment, notably the floodplain, the ponents (i.e., cellulose, hemicellulose, and lignin). Microbial
adjacent hillslopes, the upstream and downstream parts, and biomass of biofilm is higher on wood than on mineral surfaces
the regional context on a large scale. and even leaves (Spänhoff et al., 2001). More than bacteria,
LW accumulations are often surveyed over a short time lignicolous fungi play a dominant role in microbial wood
period, frequently only one survey, and it is difficult to de- processing in lotic ecosystems. However, among more than 600
termine the residence time of wood pieces and how often they species of freshwater fungi, only a limited group has the en-
move. Lammel noted in 1972 that ‘‘Normally, the zymatic capability to degrade cellulose and lignin (Wong et al.,
floatable material of these depositions will be flushed out 1998). The decay of submerged wood is a result of the fresh-
annually. However, it is not every year that the run-off pro- water fungi’s ability to form soft-rot cavities within lignified cell
duced by the annual winter storms is high enough to flush out walls. This lignocellulose degradation causes significant weight
the whole drainage and large debris may be moved only by loss and reduction in crushing strength.
floods of large size. Very often then, debris accumulates locally The microbial conditioning increases the protein content of
until such time as the stream discharge is high enough to wood, degrades and transforms surficial wood layers and the
move the material downward. This may be the case only very near-surface matrix, and thus allows freshwater benthic mac-
few times during a 100-year period.’’ Although the amount of roinvertebrates to obtain sufficient nitrogen and other nutrients
exported wood increases with the flood magnitude, it also to complete their life cycle. Xylophagous insects also contribute
varies as a function of the position of the flood event in the to wood degradation in streams by chewing, ingesting, and
hydrological series (Moulin and Piégay, 2004). LW transport excavating it. Gougers and tunnelers as well as surface scrappers
appears to be modest between the 10-year floods and and shredders exploit LW that microbial activity has enriched in
results from either the breakup of individual jams or the fungal, bacterial, and animal cells. Natural wood surfaces being
remobilization of LW delivered between flood events (Brau- never smooth, macroinvertebrates occupy and extend the
drick et al., 1997). During flood stages, logjams may be cracks, holes, grooves, protrusions, crevices, and the small de-
washed out (Heede, 1972). Overbank flow may carry it out pressions occurring on the wood surface (Mathooko and
of the channel and deposit it on the floodplain where Otieno, 2002). Consequently, it increases surface-to-volume
decomposition and fragmentation occur in the terrestrial ratio and provides fresh surfaces for microbial occupation,
environment, rather than import to the channel again (Benke which further weakens the wood.
and Wallace, 1990). In forested watersheds, channels are
commonly replete with LW. Low-gradient channels often
contain amounts of LW greater than their adjacent wetlands in
12.3.2.3 Regional Characters Controlling Abundance,
which the fate of most swamp wood appears to be in situ
Distribution, and Residence Time of LW
degradation (Benke and Wallace, 1990).
Without biological activity, water dissolves wood so slowly The relative importance of these different processes control-
that leaching of dissolved organic matter from LW and the ling wood input, movement, residence, in-channel amount
26 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and Processes

varies greatly with changes in the fluvial geomorphologic Braudrick and Grant, 2001). In high energy rivers, most of the
setting, the riparian vegetation, the climatic and hydrologic transported pieces of wood are strongly smoothed, without
regime, and the management of river and its margins (Gurnell any branches, roots, and bark. It reveals the important role of
et al., 2002). In relation to channel morphology, wood char- mechanical attrition, including decomposition, abrasion, and
acteristics, and flow regime, Gurnell et al. (2002) defined a fragmentation (Moulin and Piégay, 2004).
conceptual model for evaluating the differing wood dynamics The amount of in-channel wood depends also on a series
in: (1) small rivers where LW remains close to its initial of factors such as the physical-geographical characteristics of
deposition area; (2) medium rivers where in-channel wood is the region, the composition of riparian tree species, and tree
more mobile; and (3) large rivers where a large quantity of LW stand density and age (Evans et al., 1993). Climate plays an
comes from and is stored on riparian zones. It is necessary to important role in growth and mortality of trees. Bioclimatic
stress the contrast between small and large rivers. In moun- features create variation in the size, abundance, and storage
tainous headwater streams, channels are narrow and confined time of in-channel wood. Even if high-elevation streams show
by hillslopes. As a result, the pieces of LW are commonly an in-channel number of wood pieces similar to other climatic
longer than the bank-full channel width, exhibit a random areas, LW is here smaller, less abundant, and, therefore, more
distribution, tend to be oriented perpendicularly to the axis of frequently transported along the channel. The slow tree
water flow and remain suspended above the channel banks for growth rate and frequent disturbances by fire and insects may
a long time (Bilby and Ward, 1989). They require further explain the presence of younger and smaller riparian trees
breakdown from decay and mechanical erosion before enter- (Kreutzweiser et al., 2005; Mao et al., 2008). Moreover, cooler
ing the stream system. water temperatures may limit microbial activity and the decay
Geomorphologic pattern is also an important control on of wood (Hassan et al., 2005). Collapse of trees due to ice
LW input, storage, and transport in large rivers, only partly loading is more frequent in northern latitudes (Keller and
related to the size effect referred above. Due to the greater Swanson, 1979). Windstorms are common in many regions of
capacity for the higher energy, constrained channels with temperate and tropical zones where they provide large inputs
bedrock and boulder substrate contain a lower abundance of wood at irregular intervals.
of wood than unconfined reaches with fine substrate, and Riverside vegetation features, particularly size, shape, and
undergo a more active undercutting of trees on the bank density, also affect wood abundance and transfer (Gurnell
(Murphy and Koski, 1989; Bilby and Bisson, 1998). Wind- et al., 2002). A large piece of dense wood with rootwad and
throws and snowfalls may supply branches and treetops to branches is less mobile than a small near-cylindrical piece of
straight channels with erosion-resistant banks. Bank failure low density wood. The loadings of LW found in streams
caused by lateral erosion appears as the principal input pro- draining unmanaged, maturing, and old-growth forests are
cess on free-meandering streams that can rework their flood- greater than in streams flowing through managed second-
plain (Moulin et al., 2011). Water flow erodes concave banks growth stands (Bilby and Ward, 1991; Evans et al., 1993). In
of meander bends and entrains whole living trees from mature the former streams, not only are the wood inputs higher but
forest stands. In wandering rivers, the wood is preferentially dimensions of pieces are larger. The size of in-channel LW
deposited along the adjacent floodplain where it may form a reflects the size of the surrounding trees that derives from both
more or less continuous debris line, particularly along flow their species and ages. Moreover, irregularly shaped wood
axis concavities (Piégay and Gurnell, 1997). Even if lowland pieces are likely to enlarge a logjam or to lodge against a
rivers provide less amounts of wood than piedmont rivers boulder. Conifers supply more cylindrical LW than deciduous
because they have lower rates of lateral migration of the species. Finally, wood density determines the capacity of LW to
channel across the floodplain, they may support substantial float in freshwater. The buoyancy typically varies with tree
volumes of LW because of a lower hydraulic capacity to move species and decay rates.
them. The wood storage 4100 t ha 1 is then observed in this Besides natural events, human pressure on the river system
kind of rivers (Gippel et al., 1996; Shields and Smith, 1992), and its floodplain also alters input, storage, and transfer of wood
whereas those of piedmont rivers commonly exhibit less than in channels. This influence varies with riparian land use and
10 t ha 1, for instance, in the Drôme River, France, and along maintenance practises. The pruning of riparian trees and the
the Tagliamento River, Italy (Gurnell et al., 2002; Piégay, severe limitation of beaver populations reduce the volume of
2003). In comparison to free-meandering rivers, bank under- wood input. The development of engineering structures such as
cutting delivers lower amounts of LW to braided rivers because embankments limits flooding and channel shifting, thus re-
channel shifting frequently rejuvenates the riparian corridor ducing wood entrance. The systematic removal of in-channel LW
and determines higher turnover rates of alluvial vegetation, for improving navigation and managing hydraulic risks dra-
preventing the riverside trees from attaining maturity matically decreases the abundance of wood in rivers. Transverse
(Lassettre et al., 2008). These river corridors also lack wood hydraulic structures such dams and weirs hinder fluvial transport
accumulations because their channels present a low trapping of wood. Reservoirs may trap all floating debris. On the other
efficiency: they are unable to retain the pieces of wood that hand, land use may also promote LW input. Among the wood
are in transit (Piégay and Gurnell, 1997). LW is commonly trapped in the Génissiat dam’s reservoir on the Rhône River,
deposited in sinuous stream reaches promoting frequent France, Moulin and Piégay (2004) found that 40% of the large
contact between the wood and the channel, whereas LW pieces of wood (diameter412.5 cm) have a human origin
moves easier through straight narrow reaches characterized (Figure 4). Streamside slides and bank erosion may result from
by higher shear stresses, deeper channels, and lower number timber harvesting and road building, which provide large
of mid-channel bars (Nakamura and Swanson, 1994; amounts of LW in headwater streams.
 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and Processes 27

100 100

80 d > 0.125 m 80 0.125 > d > 0.075 m

Frequency (%)
60 60

40 40

20 20

0 0
Broken Cut by saw Broken Cut by Cut by saw
beaver
100 100

80 0.075 > d > 0.04 m 80 0.04 > d > 0.0054 m

Frequency (%)

60 60

40 40

20 20

0 0
Broken Cut by Cut by Broken Cut by Cut by
beaver saw beaver saw
Figure 4 Distribution of the pieces trapped by the Génissiat dam’s reservoir according to their taxonomic genus. Modified from Moulin and
Piégay (2004).

12.3.3 LW Effects on Fluvial Processes, Channel relative to the water depth (Hygelung and Manga, 2003). The
Morphology, and Riparian Features influence of LW on flow resistance decreases with increasing
flows (Gippel, 1995).
This section considers LW after introduction within the Owing to the increased flow resistance (Curran and Wohl,
channel and addresses local-scale effects. Especially in large 2003), LW may alter the stream ability to transport sediment
accumulations, in-channel wood influences physical processes (Chin and Wohl, 2005) and increase sediment storage (May
via diversion and obstruction of water flow. It increases and Gresswell, 2003a) at the reach scale. As LW increases
hydraulic diversity and the structural diversity of forms and roughness, their entrainment requires higher shear stress.
grain size patterns in small as well as large rivers (Montgomery Therefore, logjams reduce the frequency of bed particle move-
et al., 2003). ment and the bedload transport rate (Assani and Petit, 1995).
LW structures may provide local base levels; they form storage
sites at which coarse sediment may remain immobilized for
12.3.3.1 Wood Jam Effects on the Hydraulic Conditions
longer periods than upstream from large boulders (Faustini and
and Transport of Material
Jones, 2003). The total volume of sediment stored behind LW-
Large pieces of wood (and associated morphological features) induced steps in small and medium streams that can be about
overall create and maintain the diversity of hydraulic con- 120% of the mean annual sediment yield (Marston, 1982;
ditions at both the local and reach scales because they act as Andreoli et al., 2007). The disruption of wood accumulations
roughness elements, providing additional flow resistance makes the previously stored sediment available for transport
(Figure 5). Water may flow over and through the large pieces (Mosley, 1981). Finally, the retention of particulate organic
of wood, plunge into the pools below, and promote cascades matter (i.e., leaves, needles, and twigs) is closely related to
(Chin and Wohl, 2005). Although LW covers less than 2% of sediment deposition. LW is not only a source of organic matter
the surface area of the channel of the Cultus River, Oregon, when it decays, but it also provides ‘‘conditions suitable
it provides roughly half the total roughness (Manga and for localized deposition and retention of organic material’’
Kirchner, 2000). Large pieces of wood and their accumulations (Daniels, 2006). In a low-energy meandering river, Daniels
promote the dissipation of energy (Heede, 1972) through (2006) showed that the removal of LW induced a 33% overall
flow-contraction and pool-formation processes (Shields and reduction in the storage of organic matter in bed material.
Gippel, 1995). The dissipation of energy in LW-rich channels
is revealed by decreased average velocity, increased transit time
12.3.3.2 LW Effects on Channel Morphology and Grain
of water, and deflection of water movement into eddies and
Size Pattern
backwaters behind jam dams (Trotter, 1990). The LW contri-
bution to flow resistance depends on both hydraulic con- As wood accumulations play a role in hydraulics and transport
ditions in the river and geometrical characteristics of the LW, of sediment and organic matter, they also exert tremendous
such as the density and spacing of pieces, and their size influence on channel morphology and grain-size patterns at
28 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and Processes

Depth and deposits

Deep/sands

Deep/gravels
Shallow/pebbles
1
Shallow/gravels

Shallow/sands
2
Shallow/silt 4
Flow velocity

Obstacles 5
3
Boulder

Logjam 0 1m

Habitats 3 Habitat for juvenile fish

1 Feeding area for invertebrates
(accumulation of organic matter) 4 Feeding area for adult fish

2 Hydraulic refugia and cover 5 Spawning area

Figure 5 Logjam in a small stream. Modified from Le Lay, Y.-F., Piégay, H., 2007. Le bois mort dans les paysages fluviaux franc- ais: éléments
pour une gestion renouvelée. L’Espace Géographique 36(1), 51–64. p. 56.

the local scale, even in streams with abundant large boulders gravel beds of mountainous streams. They indeed accumulate
(Faustini and Jones, 2003; Gomi et al., 2003). LW has an effect gravel, provide steps, and thus reduce the flow energy. The
on longitudinal as well as cross-sectional profiles of stream- stream adjustment to slope induces additional gravel bars.
beds. A typical longitudinal profile consisting of the following Gravel deposition secures the logs in place so that log steps
iterative association was recognized in small streams of appear as ‘‘inflexible structures if not washed out, reoriented
western Oregon: an upstream sediment deposit, a wood by high flows, or rotted with time’’ (Heede, 1972). In forested
obstruction, and a downstream plunge pool (Heede, 1972; headwater streams, log steps form a series of waterfalls and
Swanson et al., 1976). LW may therefore significantly control cascades that occupy a relatively small percentage of channel
channel’s gradient and mesohabitat sequences. Pool for- length (Keller and Swanson, 1979; Marston, 1982).
mation is one of the most important LW geomorphologic LW also alters cross-sectional profile. It can influence
effects in small to moderately sized gravel-bed channels channel meandering and bank stability, thus affecting channel
(Montgomery et al., 1995). ‘‘Pools are closely associated with width. Large pieces of wood may either facilitate the lateral
LW dams of all types. As a result, channels containing LW mobility of channels and bank undercutting, or armor chan-
accumulations support a greater number of pools than those nel banks (Keller and Swanson, 1979). It changes the width-
which are relatively free of LW’’ (Gurnell and Sweet, 1998). to-depth ratio of channels (Bilby and Ward, 1991). As LW
Bisson et al. (1982, 1987) developed a classification of pools, diverts the flow of water, it locally increases velocity, thus
riffles, and glides that are relevant to fish habitat. Among promoting sediment transport and bank erosion and facili-
10 pool types, six were related to LW or root wad. LW has an tating bank scour and undercutting (Heede, 1972). In their
impact on several well-defined morphological dimensions of study of New Zealand’s streams, Meleason et al. (2005) ob-
pools such as spacing, area, and volume. Although steeper served that LW induced local erosion,‘‘by way of scour holes in
channels present lower pool-to-pool spacing than lower- the sediment bed and bank undercutting.’’ Thus, large pieces
gradient channels, pool spacing appears to be more in- of wood may cause channels to widen and deepen upstream
dependent of LW loading in generally steeper, step-pool and downstream from their location (Thompson, 1995).
channels. Moreover, pool volume is positively correlated with For instance, Keller and Swanson (1979) studied a large log-
log dimensions (Bilby and Ward, 1989). Larger logs create jam in a low gradient meandering stream of North Carolina
larger and deeper pools. Wood jams can create larger pools that had resulted in a 230% increase in channel width due to
than those caused by boulder steps (Mao et al., 2008). LW bank erosion caused by the wood accumulation. The addition
jams may also control abundance and morphology of pools of logs induces stream widening and deepening, whereas
on larger rivers. They are associated with 70% of the pools of stream width and depth decrease when wood is removed from
the Queets River, Washington. LW-formed pools are deeper the channel (Trotter, 1990). Despite stabilization by tree roots
than free-formed pools (Abbe and Montgomery, 1996). In and boulders, wood obstructions may also facilitate the lateral
small streams, logs fallen across the channel stabilize the migration of channels and the development of secondary
 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and Processes 29

channel systems (Bisson et al., 1987; Nakamura and Swanson, streambed contains smaller grain sizes than in reaches without
1993). When a wood accumulation blocks the channel, it may wood (Manga and Kirchner, 2000). If the accumulations of LW
cause the water flow to by-pass the jam and facilitate avulsions force water to flow directly over logs, backwater effects allow the
or meander cutoffs (Swanson and Lienkaemper, 1978). deposition of fine-grained sediment upstream from LW, while
Wood obstructions may, conversely, armor banks. In a high-flow scour provokes channel-bed coarsening downstream
mountain stream of the Chilean Andes, Andreoli et al. (2007) (Chin, 1989; Thompson, 1995). If the stream circumnavigates
suggested that LW that lines the bankfull channel edges ‘‘are wood accumulation, then wood initiates sediment deposition
likely to shelter the banks from the direct action of the main along the bank upstream and downstream from the obstacle
flow, thereby limiting bank erosion as long as they are not where flow stages are more moderate. As large pieces of wood
mobilized.’’ The numerous pieces of large wood that accu- facilitate the accumulation of fine sediment, median bed par-
mulate along the concave outer margins of meander bends ticle size is more variable in channels rich in LW than in
may form stable jams structured by a few key members and channels without LW (Faustini and Jones, 2003).
racked members. They introduce a local flow deflection which
favors the trapping of driftwood. This additional material
12.3.3.3 Relative Influence of LW According to Channel
allows the jam to increase its size, mass, and stability. ‘‘Me-
Size
ander jams establish local hard points within alluvial valleys
that limit channel migration and influence meander curva- The morphological function of wood in streams and rivers
ture’’ (Abbe and Montgomery, 2003). They protect banks varies with channel size and gradient within a catchment
against imposed shear stresses, halting local bank erosion and (Nakamura and Swanson, 1993; Baillie et al., 2008). LW may
maintaining locally narrow channel widths (Buffington and have an important geomorphic role in first- through fourth-
Montgomery, 1999). The banks may remain stable at accu- order streams (Bilby and Bisson, 1998) where channel widths
mulation sites while the channel may continue to migrate are similar or smaller than the length of wood pieces. The
upstream of the jam and increase meander amplitude. influence of wood on the step forming and channel widening
It therefore appears that the large pieces of wood tend to is also greater on low-order streams. Chen et al. (2008)
increase average and standard deviation in channel width and evaluated the relationship between LW characteristics and
depth (Bisson et al., 1987; Trotter, 1990; Bilby and Bisson, channel morphological features in 35 first- to fifth-order
1998). Hydraulic conditions and channel geometry may streams. The authors considered particularly the following key
change with wood removal (Dufour et al., 2005) (Figure 6). morphological functions of LW: step and pool creation, bank
Wood accumulations in rivers change hydraulics and forms stability, sediment and large wood storage, and trapping of
and also associated grain size pattern. They can induce the smaller woody debris. More than half of the LW was thus
deposition and sorting of sediments, promoting textural fining identified as functional LW. The percentage of functional LW
(Buffington and Montgomery, 1999). In LW-rich reaches, the by volume increased with increasing bankfull width.

Sections with logjams Control section

in 1999, removed in 2000 (without wood in 1999 and 2001)

1.2 1.2

0.8 0.8

0.4 0.4

Dea Dt
10 20 30 40 10 20 30 40

Function and 99%

1.2
confidence interval

0.8
c of Geary

1999
0.4
2001
Deb
10 20 30 40 50 60 70 80 Distance (m)

Figure 6 c of Geary autocorrelation function performed on a sample of relative elevation values measured in the channel bed of the Doulon
River (Central Massif, France) in 1999 and 2001 (before and after wood removal on sites Dea and Deb). Modified from Dufour, S., Piégay, H.,
Landon, N., Moulin, B., 2005. Production, répartition et effets hydro-morphologiques du bois mort dans un petit cours d’eau franc- ais de
moyenne montagne. Zeitschrift für Geomorphologie 49(3), 391–409. p. 402.
30 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and Processes

Generally, the influence of LW on step and pool formation the banks, the reversal of flow direction in many tributaries,
and sediment transport tends to decline in the downstream and the diversion of water onto riparian areas, especially at
direction with decreasing channel slope (Keller and Swanson, high flow. A series of nearly permanent lakes formed in the
1979). In large rivers with bars and shallow features relative to inundated lowland forests.
the wood size, preferential sites of deposits are observed Abbe and Montgomery (2003) identified three funda-
within the channel. In the Drôme River, France, LW is mainly mental types of wood accumulation based on possible LW
deposited on the top or edge of unvegetated gravel bars when transport: (1) autochthonous jams that consist of direct bank
the flow decreases. Their time residence is short and their input and log steps where LW are stored at the point where
geomorphic role is negligible (Piégay et al., 1999). Further they entered the channel; (2) allochthonous jams are made of
downstream in lowland rivers and streams, velocity becomes LW that have moved by fluvial transport termed debris-flow/
so slow that wood stays where they fall. In this situation many flood jams, bench jams, bar-apex jams, meander jams, log raft,
snags can be observed in the middle of the channel as well as and unstable debris according to the preferential sites; and
anchored trees along the banks (Piégay, 2003). Likewise, in (3) combination jams consisting of valley jams and flow-
low-gradient coastal rivers of Maine, USA, LW creates very few deflection jams that are composed of in situ key members
pools and only 5–20% of the wood is associated with sedi- along with trapped driftwood. High-order streams display
ment storage (Magilligan et al., 2008). If it is commonly held wood accumulations that can influence bed and bank scour,
that LW has its greatest influence on the morphology of small secondary channel development, bar stability, island for-
streams, the small size of channels must be considered in re- mation, and the subsequent succession of riparian vegetation
gard to wood length. Dimensions of LW resulting from the fall (Abbe and Montgomery, 1996).
of Sequoia sempervirens allow them to cross streams wide of Logjams are roughness elements that play a major role in
several dozens of meters. The nature of LW influence depends the initiation, development, and maintenance of vegetated
on the size of individual pieces relative to channel size. bars and islands (Gurnell et al., 2001; Gurnell and Petts,
2006). Some riparian tree species such as poplar (Populus) can
reproduce vegetatively and support pioneer island develop-
ment. Uprooted black poplar trees (Populus nigra L.) are able
12.3.3.4 Geomorphological Effects of LW on Riparian
to anchor themselves to gravel bars by developing shoots and
Areas
adventitious roots (Edwards et al., 1999; Gurnell et al., 2005).
Human impacts on large rivers have strongly reduced the Abbe and Montgomery (2003) described the geomorphologic
input of LW in order to promote navigation and flood pro- role of bar apex jams structured by key members (such as large
tection. ‘‘In pristine conditions, large rivers of the temperate logs with intact rootwads facing upstream). Bar apex jams alter
zone drained forested corridors and were affected by large the local channel hydraulics, developing a zone of flow de-
quantities of wood and such high hydrological connectivity celeration, a saddle point and a vortex flow upstream of the
between the channel and the floodplain that it was difficult to obstruction, a flow separation, and two zones of flow con-
distinguish the boundaries between these landforms’’ (Gurnell striction and acceleration adjacent to the jam. Bar apex jam
et al., 2002). Historical records demonstrate influences that controls the spatial pattern of scour and deposition. Such a
extensive logjams exerted on adjacent riparian areas. In the jam is typically associated with several alluvial bedforms: (1)
nineteenth century, many North American fluvial systems an arcuate, coarse gravel bar upstream; (2) a crescentic pool
exhibited complex morphologies, which strongly disrupted immediately upstream and adjacent to the rootwad; and (3) a
human navigation. The Great Raft of the Red River that downstream central bar of finer sediments along the axis of
affected 390–480 km of main channel c. 1806 attracted con- the key member (Naiman et al., 2000). Whereas sediment
siderable attention (Triska, 1984; Keller and Macdonald, preferentially accumulates upstream of the large pieces of
1995). Reclus (1859) was fascinated by the gigantic wood wood in high-gradient systems, mid-channel bars downstream
accumulation: ‘‘Nothing may provide a picture for this fabu- of LW obstructions may serve as major storage locations of
lous pile of logs tangled by their roots and branches. sediment in low-energy systems (Gurnell and Gregory, 1995).
Stretching in the sludge of the shore or erecting their fantastic Logjams create sites of sediment aggradation that provide
heads out of the blackish water, they look like the ancient refugia against high velocity flows ‘‘for the establishment of
plesiosaurs which dragged themselves along the muddy chaos pioneer plant species initiating vegetation development’’
ages ago.’’ Triska (1984) considered the landscape derived (Fetherston et al., 1995). Tree size and age decrease down-
from the Great Raft ‘‘as typical of a pristine lowland river.’’ He stream and reflect the progressive colonization of newly de-
identified the key steps of its natural formation: (1) the posited sediments. These wood accumulations may be
clumping of LW during high discharges; (2) the jamming of stable for decades to centuries, where mature riparian forest
the river width at a channel constriction; (3) the impound- may persist in a riverine environment characterized by fre-
ment of water and the slowing of current velocity; (4) the quent disturbances and channel migration. The growth and
filling of interstices by finer debris; (5) the submersion of coalescence of forested islands along the central bar may
floating debris by waterlogging; and (6) the deposition of eventually integrate with the broader floodplain.
sediment. Such a massive accumulation of LW was able to Wood in river facilitates plant colonization downstream
directly influence the river morphology by inducing the of obstructions that reduce flow velocities and increase
channel aggradation and avulsion, the development of ripar- local channel aggradation and accumulation of organic
ian lakes, and floodplain formation. The rise of water ele- matter (Abbe and Montgomery, 1996). Vegetated islands
vation into the main channel promoted sedimentation along reinforce this process by increasing the LW trapping efficiency
 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and Processes 31

of river systems, provided that the river is not deeply incised decomposition by scraping, gouging, and tunneling wood.
and that the level of floods is sufficiently high (Wyzga and Shredders break primarily large pieces of decomposing vas-
Zawiejska, 2005). Similarly, braided rivers with vegetated is- cular plant tissue so that finer particles may be used by gath-
lands store more wood than bar-braided rivers because islands ering type invertebrates. LW also provides a surface of biofilm
increase the contact between the active channel and the that serves as a food source for grazing organisms. Logjams
forested areas (Gurnell et al., 2002). LW facilitates the can serve as retention sites for leaf litter and other particulate
encroachment of vegetation on floodplains, allowing forests organic matter, enhancing nutrient retention within the
to extend on floodplains. Nurse logs (Harmon et al., 1986; channel and its margins. In oligotrophic systems of small
Fetherston et al., 1995) and LW piles may provide moisture streams in the Pacific Coastal Ecoregion, salmon (Oncor-
and nutrients for establishing seeds and plants and protect hynchus spp.) die after spawning. LW retains their carcasses
them from erosion, abrasion, and herbivory (Naiman and that are an important food resource for many aquatic species.
Décamps, 1997). Nitrogen and carbon derived from spawning salmon are in-
Generally, the contribution of wood in island formation corporated in the tissues of invertebrates and juvenile sal-
and riparian development remains poorly understood. Islands monid fishes (Bilby and Bisson, 1998).
may provide favorable local conditions for tree recruitment LW influences the structure and function of benthic com-
and growth. However, other factors, such as bar micro- munities. Following LW additions to a small stream in North
topography and associated patterns of fine sediment de- Carolina, abundance and biomass at debris dam transects
position, may also provide required conditions such that it is were respectively 24-fold and 2.1-fold greater than before log
difficult to determine at the reach scale which processes are the addition (Wallace et al., 1995). LW additions also influenced
most important. Thus, a conundrum may exist as to whether invertebrate community composition. Abundances and bio-
the capacity of LW to create vegetated islands or the vegetated masses of collectors and predators increased; those of scrapers
island capacity to trap wood is foremost. and filterers decreased. As some species cannot tolerate a fre-
quently shifting bottom, the influence of LW on macro-
invetebrates nevertheless varies with respect to the availability
of other stable supports. Though wood acts as the center of
12.3.4 In-Channel Wood and River Management
invertebrate production in low-gradient rivers in which it
often provides all the stable substrates, it may only support an
The interactions between LW and fluvial processes have im-
increased diversity of shredders in streams that also exhibit
portant implications for the management of rivers both in
stable streambed consisting of boulders, cobbles, or gravels
terms of ecological conservation and human safety.
(Benke and Wallace, 2003; Gabriel et al., 2009). Likewise, LW
may increase the abundance and biomass of fish (Fausch and
Northcote, 1992). In the Drôme, Rhône and Loire rivers,
12.3.4.1 Ecological Benefits of LW
France, Thévenet (1998) underlined the influence of wood on
Because LW creates structural complexity, it plays an important fish abundance and diversity: sites with LW exhibited 1.6–7
functional role in aquatic habitat and contributes to the times more fish and 1.3–2 times more species than sites
biomass and biological diversity of aquatic communities. The without wood (Figure 7).
complex physical structure of highly wood-laden river pro- Stream-dwelling organisms using large pieces of wood can
vides a diversity of habitat patches for fish, substrate for also influence fluvial geomorphology. In addition to bio-
macroinvertebrates, and boosts trophic processes. logical processes that contribute to the input and degradation
The life histories of numerous fish species have some as- of LW in channels, beaver can alter aquatic ecosystems in the
sociation with large wood for cover, spawning (egg attach- northern temperate zone. ‘‘Beaver dams measurably affect the
ment, nest materials), and feeding (Dolloff and Warren, rates of groundwater recharge and stream discharge, retain
2003). It may become a necessary element in rivers charac- enough sediment to cause measurable changes in valley floor
terized by predominantly fine substrates. Although riffles re- morphology’’ (Pollock et al., 2003). They also physically add
quire a high metabolic cost of swimming against rapid flow, to the LW load by directly felling timber onto floodplains and
LW-associated pools slow down and deepen water flow, which adjacent stream channels.
attract many stream fishes for rest, food, and cover (Bisson
et al., 1987). LW in streams and rivers can serve as hydraulic
refugia during high flow and nursery habitat for anadromous 12.3.4.2 Risks and Nuisances Associated with LW
fish (Senter and Pasternack, 2010). Moreover, wood accumu-
lations encourage sediment deposition and create habitats In the nineteenth century, river pilots recognized the principal
favored by some benthic invertebrates. Stream-dwelling in- impediments to navigation on the Mississippi River. Besides
vertebrates directly utilize in-channel LW as substrate. LW the instability of the banks and some dangerous currents, they
provides sites for oviposition, pupation, and emergence by feared coarse woody debris and named them. Cramer (1817)
benthic macroinvertebrates. distinguished three types of LW obstacles: (1) planters or snags
Macroinvertebrate functional groups – i.e., grazers, shred- that are tree trunks firmly and perpendicularly embedded in
ders, gatherers, filterers, and predators – play an important the bottom of the river; (2) sawyers are smaller pieces of trees
role in the processes of forested stream ecosystems (Anderson embedded less perpendicularly in the river, ‘‘yielding to the
et al., 1978; Wallace and Webster, 1996). Shredding in- pressure of the current, disappearing and appearing by turns
vertebrates directly feed on wood. They encourage wood above water’’; and (3) wooden-islands or rafts derived from an
32 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and Processes

7 7
*
6 * 6

Specific richness
5 5

Abundance
4 4
3 3
2 2
1 1
0 0
1 2 3 4 1 2 3 4
Figure 7 Fish abundance and species richness in relation to wood cover (wood density). The breakpoints are calculated to obtain the same
number of individuals per cell: (1) 0%; (2) [0–5]; (3) [5–15]; (4) 415%. Modified from Thévenet (1998).

30
Cumulative sum

20

10

0
1880 1900 1920 1940 1960 1980 2000
Years
Figure 8 Negligent maintenance (gray line) and logjams (black line) as factors explaining flooding in France – analysis of 111 floods from
regional daily newspapers (1882–2005). Modified from Le Lay and Rivière-Honegger (2009).

accumulation of driftwood deposited on bars. In large rivers, (Wallerstein and Thorne, 1998). Logjams may cause increased
the logjams are still usually removed to improve navigation. backwater and bank scour at bridges. Large pieces of wood
Since 1950, most European rivers have contained larger with attached rootwads are commonly the key members of
amounts of LW than they did over the first half of the twentieth these wood accumulations (Diehl and Bryan, 1993). When
century. The post-World War II abandonment of agricultural bridges are blocked by LW, streams can flood and damage
lands in the floodplain and the lack of riverside trees mainten- riverside houses and facilities. More generally, LW and chiefly
ance by riverine landowners induced the widespread afforest- logjams can obstruct river flows and raise the elevation of
ation and forest maturation. Since 1945, the number of logjams water surface (Gippel, 1995; Gippel et al., 1996). As a result,
noted by riverine communities has increased in the watershed of LW can locally increase the frequency and duration of inun-
Isere River, France (Figure 8). Such an environmental evolution dation. Floating logs can reduce the efficiency of intake weirs
promotes a series of hazards associated with wood accumu- of constructed flood channels and, likewise, they may clog
lations. That is why governments and managers regularly re- flow-control slits, limiting the efficiency of permeable dams
mind riverine landowners; they have a duty to maintain streams designed to mitigate erosion by controlling sediment dis-
and rivers (Le Lay and Rivière-Honegger, 2009). charge. The degradation of large amounts of LW can impair
LW appears to be a geomorphic hazard in areas where high water quality and visual aesthetics of reservoir ponds.
population densities occur along small forested mountain Today, the evolution of legislative action and social de-
streams (Braudrick and Grant, 2001). In Japan, Ishikawa mand for improved environmental conservation encourages
(1989) developed a classification of several types of damage decision makers to reconsider their management of wastes
caused by LW, including floating logs may directly strike accumulating against dams and hydraulic structures. Floating
houses and structures and LW, transported by debris flow wood not only is a nuisance likely to induce economic costs
and streamflow damage structures such as roads, bridges, and for managers but may also be a resource for the purposes
culverts (Figure 9). The wood accumulating behind bridge of composting and heating. The success of such attempts de-
abutments increases the destructive power of water flow by pends on LW integration into a process of gathering, treating,
increasing resistance. The structure is thus put under increased and covering and on the ability of all the stakeholders to be
stress and may fail due to undermining and/or water pressure involved in the management of catchment.
 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and Processes 33

Figure 9 Wood accumulation against a bridge of the Roanoke River. Photo by B. Moulin, USGS, in 2010.

12.3.4.3 Applications of Knowledge in Terms of River managers may consider finding ‘‘a balance between
Restoration and Sustainable Management woody debris preservation and reintroduction for ecological
purposes, and the need for channel clearing for risk manage-
Particularly in lowland rivers, LW has been controlled in
ment’’ (Moulin and Piégay, 2004). The promotion of a
channels to increase conveyance for flood mitigation and
maintenance strategy should be based on defined objectives
irrigation supply, as well as to improve navigation, reduce
within pre-determined homogeneous reaches. Channel clear-
bank erosion, limit damage to bridges, remove barriers to fish
ing and riparian vegetation pruning may be demanded by a
migration, and improve recreation such as swimming and
public who prefer maintained riverscapes to more natural
boating (Harmon et al., 1986; Bisson et al., 1987). Such
riverscapes (Le Lay et al., 2008). Such practices facilitate access
practices have been common from the nineteenth century in
to the river which may support leisure activities. Artificial
the USA and before that in Europe. For instance, since the
structures may even be conceived to trap floating LW upstream
sixteenth century many French regulations have required the
of developed and urban areas. However, this requires regular
rivers to be kept free of LW in order to maximize the channel
maintenance to remain effective. Given the growing recog-
capacity, notably for the purpose of navigation. The cleared
nition of its ecological benefits, the presence of in-channel LW
LW was then floated toward large cities such as Paris and Lyon
has been recommended since the 1980s, de-snagging oper-
as they expanded on the floodplains (Le Lay and Piégay, 2007)
ations have been criticized, and the replacement of LW has
where it was used for fuel.
been achieved in some previously cleared rivers. The re-
Before 1970, many scientists and managers considered LW
introduction of LW in channels where it is lacking induces
as a cause of oxygen depletion and overall as a hindrance to
neither a significant loss of conveyance nor an increase in
the upstream migration of anadromous salmonids (Bryan,
flood frequency (Gippel et al., 1996). Wood reintroduction is
1983). More recent investigations have shown that logjams
a method increasingly used in restoration projects to improve
naturally present in channels do not block fish access
the hydrological, morphological, and ecological status of
to spawning and nursery areas and by contrast highlighted
degraded streams and rivers. The placement of stable wood
their benefits for fish populations (Harmon et al., 1986). The
structures may contribute to local bank protection and to the
removal of LW was also conducted to limit the instability of
active creation of in-stream habitats on a reach scale. However,
banks and streambeds. Nevertheless, it now appears that log-
Kail et al. (2007) recommended the protection of a buffer
jams are able to protect banks from erosion and to stabilize
zone consisting of riparian forest and the use of non-fixed
gravel bars in many cases. Moreover, the flood control benefits
wood structures that are able to restore habitat forming pro-
of channel clearing may be modest. LW has a significant
cesses on a catchment scale.
effect on flood levels where it forms large logjams or where
it occurs in constrained channels (Young, 1991). The effect
of LW on channel capacity is nevertheless very small in
large rivers at high flows. Moreover, LW may reduce flood References
peaks and increase the timing of flood propagation down-
stream (Gippel, 1995). Abbe, T.B., Montgomery, D.R., 1996. Large woody debris jams, channel hydraulics
and habitat formation in large rivers. Regulated Rivers: Research and
LW removal from streams has many deleterious geomor-
Management 12, 201–221.
phologic and ecological effects. Removing distributes the flow Abbe, T.B., Montgomery, D.R., 2003. Patterns and processes of wood debris
more evenly across the stream, thus contracting eddies and accumulation in the Queets river basin, Washington. Geomorphology 51,
areas of reduced velocity and reducing the maximum velocity 81–107.
(Shields and Smith, 1992; Gippel, 1995). It also removes Anderson, N.H., Sedell, J.R., Roberts, L.M., Triska, F.J., 1978. The role of aquatic
small waterfalls over which stream energy is dissipated (Bilby invertebrates in processing of wood debris in coniferous forest streams.
American Midland Naturalist 100(1), 64–82.
and Likens, 1980). It leads to sediment scour and channel Andreoli, A., Comiti, F., Lenzi, M.A., 2007. Characteristics, distribution and
incision, thus altering stream morphology. All these changes geomorphic role of large woody debris in a mountain stream of the Chilean
have impacts on physical habitats for aquatic communities. Andes. Earth Surface Processes and Landforms 32(11), 1675–1692.
34 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and Processes

Assani, A.A., Petit, F., 1995. Log-jam effects and bed-load mobility from Diehl, T.H., Bryan, B.A., 1993. Supply of large woody debris in a stream channel.
experiments conducted in a small gravel-bed forest ditch. Catena 25, 117–126. In: Shen, H.W., Su, S.T., Feng, W. (Eds.), Proceedings of Hydraulic Engineering
Aumen, N.G., Bottomley, P.J., Ward, G.M., Gregory, S.V., 1983. Microbial ’93 Conference. American Society of Civil Engineers, San Francisco, CA, vol. 1,
decomposition of wood in streams: distribution of microflora and factors pp. 1055–1060.
affecting [14C] lignocelluloses mineralization. Applied and Environmental Dolloff, C.A., Warren, M.L., 2003. Fish relationships with large wood in small
Microbiology 46(6), 1409–1416. streams. In: Gregory, S.V., Boyer, K.L., Gurnell, A.M. (Eds.), The Ecology and
Baillie, B.R., Garrett, L.G., Evanson, A.W., 2008. Spatial distribution and influence of Management of Wood in World Rivers. American Fisheries Society, Bethesda,
large woody debris in an old-growth forest river system, New Zealand. Forest MD, pp. 179–193.
Ecology and Management 256, 20–27. Dufour, S., Piégay, H., Landon, N., Moulin, B., 2005. Production, répartition et
Benda, L.E., Sias, J.C., 2002. A quantitative framework for evaluating the mass effets hydro-morphologiques du bois mort dans un petit cours d’eau franc- ais de
balance of in-stream organic debris. Forest Ecology and Management 5626, moyenne montagne. Zeitschrift für Geomorphologie 49(3), 391–409.
1–16. Edwards, P.J., Kollmann, J., Gurnell, A.M., Petts, G.E., Tockner, K., Ward, J.V.,
Benke, A.C., Wallace, J.B., 1990. Wood dynamics in coastal plain blackwater 1999. A conceptual model of vegetation dynamics on gravel bars of a large
streams. Canadian Journal of Fisheries and Aquatic Sciences 47, 92–99. Alpine river. Wetlands Ecology and Management 7, 141–153.
Benke, A.C., Wallace, J.B., 2003. Influence of wood on invertebrate communities in Evans, B.F., Townsend, C.R., Crowl, T.A., 1993. Distribution and abundance of
streams and rivers. In: Gregory, S.V., Boyer, K.L., Gurnell, A.M. (Eds.), The coarse woody debris in some southern New Zealand streams from contrasting
Ecology and Management of Wood in World Rivers. American Fisheries Society, forest catchments. New Zealand Journal of Marine and Freshwater Research 27,
Bethesda, MD, pp. 149–177. 227–239.
Bilby, R.E., Bisson, P.A., 1998. Function and distribution of large woody debris. Fausch, K.D., Northcote, T.G., 1992. Large woody debris and salmonid habitat in a
In: Naiman, R.J., Bilby, R.E. (Eds.), River Ecology and Management: Lessons small coastal British Columbia Stream. Canadian Journal of Fisheries and
from the Pacific Coastal Ecoregion. Springer, New York, NY, pp. 324–346. Aquatic Sciences 49, 682–693.
Bilby, R.E., Likens, G.E., 1980. Importance of organic debris dams in the structure Faustini, J.M., Jones, J.A., 2003. Influence of large woody debris on channel
and function of stream ecosystems. Ecology 60, 1107–1113. morphology and dynamics in steep, boulder-rich mountain streams, western
Bilby, R.E., Ward, J.W., 1989. Changes in characteristics and function of woody Cascades, Oregon. Geomorphology 51, 187–205.
debris with increasing size of streams in western Washington. Transactions of Fetherston, K.L., Naiman, R.J., Bilby, R.E., 1995. Large woody debris, physical
the American Fisheries Society 118, 368–378. process, and riparian forest development in montane river networks of the
Bilby, R.E., Ward, J.W., 1991. Characteristics and function of large woody debris in Pacific Northwest. Geomorphology 13, 133–144.
streams draining old-growth, clear-cut, and second-growth forests in Fuller, M.L., 1912. The New Madrid Earthquakes. U.S. Geological Survey Bulletin
southwestern Washington. Canadian Journal of Fisheries and Aquatic Sciences 494, 119.
48, 2499–2508. Gabriel, C.M., Clarke, K.L., Campbell, C.E., 2009. Invertebrate communities in
Bisson, P.A., Bilby, R.E., Byrant, M.D., et al., 1987. Large woody debris in forested Compensation Creek, a man-made stream in boreal Newfoundland: the influence
streams in the Pacific Northwest: past, present, and future. In: Salo, E.O., of large woody debris. River Research and Applications, 14. http://dx.doi.org/
Cundy, T.W. (Eds.), Streamside Management and Fishery Interactions. Institute of 10.1002/rra.1295, 14 pp.
Forest Resources, University of Washington, Seattle, WA, pp. 143–190. Gippel, C.J., 1995. Environmental hydraulics of large woody debris in streams and
Bisson, P.A., Nielsen, J.L., Palmason, R.A., Grove, L.E., 1982. A system for naming rivers. Journal of Environmental Engineering 121, 388–395.
habitats types in small streams with examples of habitat utilisation by salmonids Gippel, C.J., Finlayson, B.L., O’Neill, I.C., 1996. Distribution and hydraulic
during low streamflow. In: Armantrout, N.B. (Ed.), Acquisition and Utilisation of significance of large woody debris in a lowland Australian river. Hydrobiologia
Aquatic Habitat Inventory Information. American Fisheries Society, Portland, OR, 318, 179–194.
pp. 62–73. Gomi, T., Sidle, R.C., Woodsmith, R.D., Bryant, M.D., 2003. Characteristics of
Bocchiola, D., Rulli, M.C., Rosso, R., 2006. Transport of large woody debris in the channel steps and reach morphology in headwater streams, southeast Alaska.
presence of obstacles. Geomorphology 76, 166–178. Geomorphology 51, 225–242.
Braudrick, C.A., Grant, G.E., 2001. Transport and deposition of large woody debris Gregory, S.V., Boyer, K.L., Gurnell, A.M. (Eds.), 2003. The Ecology and
in streams: a flume experiment. Geomorphology 41, 263–283. Management of Wood in World Rivers. American Fisheries Society, Bethesda,
Braudrick, C.A., Grant, G.E., Ishikawa, Y., Ikeda, H., 1997. Dynamics of wood MD, 431 pp.
transport in streams: a flume experiment. Earth Surface Processes and Gurnell, A.M., Gregory, K.J., 1995. Interactions between semi-natural vegetation and
Landforms 22(7), 669–683. hydrogeomorphological processes. Geomorphology 13, 49–69.
Brayshaw, D., Hassan, M.A., 2009. Debris flow initiation and sediment recharge in Gurnell, A.M., Petts, G.E., Hannah, D.M., et al., 2001. Riparian vegetation and
gullies. Geomorphology 109, 122–131. island formation along the gravel-bed Fiume Tagliamento, Italy. Earth Surface
Bryant, M.D., 1981. Organic debris in salmonid habitat in southeast Alaska: Processes and Landforms 26(1), 31–62.
measurement and effects. In: Armantrout, N.B. (Ed.), Acquisition and Utilization Gurnell, A.M., Piégay, H., Swanson, F.J., Gregory, S.V., 2002. Large wood and
of Aquatic Habitat Inventory Information. American Fisheries Society, Portland, fluvial processes. Freshwater Biology 47, 601–619.
OR, pp. 259–265. Gurnell, A.M., Sweet, R., 1998. The distribution of large woody debris accumulations
Bryan, M.D., 1983. The role and management of woody debris in West Coast and pools in relation to woodland stream management in a small, low-gradient
salmonid nursery streams. North American Journal of Fisheries Management 3, stream. Earth Surface Processes and Landforms 23, 1101–1121.
322–330. Gurnell, A.M., Tockner, K., Edwards, P., Petts, G., 2005. Effects of deposited wood
Buffington, J.M., Montgomery, D.R., 1999. Effects of hydraulic roughness on surface on biocomplexity of river corridors. Frontiers in Ecology and the Environment
textures of gravel-bed rivers. Water Resources Research 35(11), 3507–3521. 3(7), 377–382.
Cadol, D., Wohl, E., Goode, J.R., Jaeger, K.L., 2009. Wood distribution in Gurnell, E.P., Petts, G.E., 2006. Trees as riparian engineers: the Tagliamento River,
neotropical forested headwater streams of La Selva, Costa Rica. Earth Surface Italy. Earth Surface Processes and Landforms 31, 1558–1574.
Processes and Landforms 34(9), 1198–1215. Harmon, M.E., Franklin, F.J., Swanson, F.J., et al., 1986. Ecology of coarse woody
Chen, X., Wei, X., Scherer, R., Hogan, D., 2008. Effects of large woody debris on debris in temperate ecosystems. In: MacFadayen, A., Ford, E.D. (Eds.), Advances
surface structure and aquatic habitat in forested streams, southern interior of in Ecological Research. Academic Press, London, vol. 15, pp. 133–302.
British Columbia, Canada. River Research and Applications 24, 862–875. Hassan, M.A., Hogan, D.L., Bird, S.A., May, C.L., Gomi, T., Campbell, D., 2005.
Chin, A., 1989. Step pools in stream channels. Progress in Physical Geography 13, Spatial and temporal dynamics of wood in headwater streams of the
391–407. Pacific Northwest. Journal of American Water Resources Association 41,
Chin, A., Wohl, E., 2005. Toward a theory for step pools in stream channels. 899–919.
Progress in Physical Geography 29(3), 275–296. Heede, B.H., 1972. Influences of a forest on the hydraulic geometry of two
Cramer, Z., 1817. The Navigator. Cramer, Spear and Eichbaum, Pittsburg, PA, 370 pp. mountain streams. Water Resources Bulletin 8(3), 523–530.
Curran, J.H., Wohl, E.E., 2003. Large woody debris and flow resistance in step- Hygelung, B., Manga, M., 2003. Field measurements of drag coefficients for model
pool channels, Cascade Range, Washington. Geomorphology 51, 141–157. large woody debris. Geomorphology 51, 175–185.
Daniels, M.D., 2006. Distribution and dynamics of large woody debris Ishikawa, Y., 1989. Studies on Disasters Caused by Debris Flows Carrying Floating
and organic matter in a low-energy meandering stream. Geomorphology 77, Logs Down Mountain Streams. Ph.D. Thesis, Kyoto University, Kyoto, Japan,
286–298. 121 pp.
 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and Processes 35

Kail, J., Hering, D., Muhar, S., Gerhard, M., Preis, S., 2007. The use of large wood Moulin, B., Schenk, E.R., Hupp, C.R., 2011. Distribution and characterization of
in stream restoration: experiences from 50 projects in Germany and Austria. in-channel large wood along the low-gradient Roanoke River, NC. Earth Surface
Journal of Applied Ecology 44, 1145–1155. Processes and Landforms 36. http://dx.doi.org/10.1002/esp.2135.
Keller, E.A., Macdonald, A., 1995. River channel change: the role of large woody Murphy, M.L., Koski, K.V., 1989. Input and depletion of woody debris in Alaska
debris. In: Gurnell, A.M., Petts, G.E. (Eds.), Changing River Channels. Wiley, streams and implications for streamside management. North Journal of Fisheries
Chichester, pp. 217–235. Management 9, 427–436.
Keller, E.A., Swanson, F.J., 1979. Effects of large organic material on channel form Naiman, R.J., Bilby, R.E., Bisson, P.A., 2000. Riparian ecology and management in
and alluvial processes. Earth Surface Processes and Landforms 4, 361–380. the Pacific coastal rain forest. BioScience 50, 996–1011.
Kreutzweiser, D.P., Good, K.P., Sutton, T.M., 2005. Large woody debris Naiman, R.J., Décamps, H., 1997. The ecology of interfaces – riparian zones.
characteristics and contributions to pool formation in forest streams of the Annual Review of Ecology and Systematics 28, 621–658.
Boreal Shield. Canadian Journal of Forest Research 35(5), 1213–1223. Nakamura, F., Swanson, F.J., 1993. Effects of coarse woody debris on morphology
Lancaster, S.T., Hayes, S.K., Grant, G.E., 2003. Effects of wood on debris flow and sediment storage of a mountain stream in western Oregon. Earth Surface
runout in small mountain watersheds. Water Resources Research 39(6), 1168, Processes and Landforms 18, 43–61.
21 pp. Nakamura, F., Swanson, F.J., 1994. Distribution of coarse woody debris in a
Lassettre, N.S., Piégay, H., Dufour, S., Rollet, A.-J., 2008. Decadal changes in mountain stream, western Cascades Range, Oregon. Canadian Journal of Forest
distribution and frequency of wood in a free meandering river, the Ain River, Research 24, 2395–2403.
France. Earth Surface Processes and Landforms 33, 1098–1112. Pettit, N.E., Naiman, R.J., Rogers, K.H., Little, J.E., 2005. Post-flooding distribution
Le Lay, Y.-F., Piégay, H., 2007. Le bois mort dans les paysages fluviaux and characteristics of large woody debris piles along the semi-arid Sabie River,
franc- ais: éléments pour une gestion renouvelée. L’Espace Géographique 36(1), South Africa. River Research and Applications 21(1), 27–38.
51–64. Piégay, H., 2003. Dynamics of wood in large rivers. In: Gregory, S.V., Boyer, K.L.,
Le Lay, Y.-F., Piégay, H., Gregory, K., et al., 2008. Variations in cross-cultural Gurnell, A.M. (Eds.), Ecology and Management of Wood in World Rivers.
perception of riverscapes in relation to in-channel wood. Transactions of the American Fisheries Society, Bethesda, MD, pp. 109–134.
Institute of British Geographers 33, 268–287. Piégay, H., Gurnell, A.M., 1997. Large woody debris and river geomorphological
Le Lay, Y.-F., Rivière-Honegger, A., 2009. Expliquer l’inondation: la presse pattern: examples from S.E. France and S. England. Geomorphology 19, 99–116.
quotidienne régionale dans les Alpes et leur piémont (1882–2005). Géocarrefour Piégay, H., Thévenet, A., Citterio, A., 1999. Input, storage and distribution of large
84(4), 259–270. woody debris along a mountain river continuum, the Drôme River, France.
MacVicar, B.J., Piégay, H., Henderson, A., Comiti, F., Oberlin, C., Pecorari, E., Catena 35, 19–39.
2009. Quantifying the temporal dynamics of wood in large rivers: field trials of Pollock, M.M., Heim, M., Werner, D., 2003. Hydrologic and geomorphic effects of
wood surveying, dating, tracking, and monitoring techniques. Earth Surface beaver dams and their influence on fishes. In: Gregory, S.V., Boyer, K.L.,
Processes and Landforms 34, 2031–2046. Gurnell, A.M. (Eds.), The Ecology and Management of Wood in World Rivers.
Magilligan, F.J., Nislow, K.H., Fisher, G.B., Wright, J., Mackey, G., Laser, M., 2008. American Fisheries Society, Bethesda, MD, pp. 213–233.
The geomorphic function and characteristics of large woody debris in low Reclus, E., 1859. Le Mississipi. Etudes et souvenirs. 2. Le delta et la Nouvelle-
gradient rivers, coastal Maine, USA. Geomorphology 97(3–4), 467–482. Orléans. Revue des Deux Mondes 22, 608–646.
Manga, M., Kirchner, J.W., 2000. Stress partitioning in streams by large woody Reeves, G.H., Burnett, K.M., McGarry, E.V., 2003. Sources of large wood in the
debris. Water Resources Research 36(8), 2373–2379. main stem of a fourth-order watershed in coastal Oregon. Canadian Journal of
Mao, L., Andreoli, A., Comiti, F., Lenzi, M.A., 2008. Geomorphic effects of large Forest Research 33, 1363–1370.
wood jams on a sub-antarctic mountain stream. River Research and Applications Rosenfeld, J.S., Huato, L., 2003. Relationship between large woody debris
24(3), 249–266. characteristics and pool formation in small coastal British Columbia streams.
Marcus, W.A., Marston, R.A., Colvard, C.R., Gray, R.D., 2002. Mapping the spatial North American Journal of Fisheries Management 23, 928–938.
and temporal distributions of woody debris in streams of the Greater Senter, A.E., Pasternack, G.B., 2010. Large wood aids spawning Chinook
Yellowstone ecosystem, USA. Geomorphology 44, 323–335. salmon (Oncorhynchus Tshawytscha) in marginal habitat on a regulated
Marston, R.A., 1982. The geomorphic significance of log steps in forest streams. river in California. River Research and Applications, 16. http://dx.doi.org/
Annals of the Association of American Geographers 72(1), 99–108. 10.1002/rra.1388, 16 pp.?
Martin, D.J., Benda, L.E., 2001. Patterns of instream wood recruitement and Shields, F.D., Gippel, C.J., 1995. Prediction of effects of woody debris removal on
transport at the watershed scale. Transactions of the American Fisheries Society flow resistance. Journal of Hydraulic Engineering 121(4), 341–354.
130, 940–958. Shields, F.D., Smith, R.H., 1992. Effects of large woody debris removal on physical
Martius, C., 1997. Decomposition of wood. In: Junk, W.J. (Ed.), The Central characteristics of a sand-bed river. Aquatic Conservation 2, 145–163.
Amazon Floodplain. Ecology of a Pulsing System. Springer, Berlin, pp. 267–277. Spänhoff, S., Alecke, C., Meyer, E.I., 2001. Simple method for rating the decay
Mathooko, J.M., Otieno, C.O., 2002. Does surface textural complexity of woody stages of submerged woody debris. Journal of the North American Benthological
debris in lotic ecosystems influence their colonization by aquatic invertebrates? Society 20(3), 385–394.
Hydrobiologia 489, 11–20. Swanson, F.J., Gregory, S.V., Sedell, J.R., Campbell, A.G., 1982. Land–water
May, C.L., Gresswell, R.E., 2003a. Processes and rates of sediment and wood interactions: the riparian zone. In: Edmonds, R.L. (Ed.), Analysis of Coniferous
accumulation in headwater streams of the Oregon Coast Range, USA. Earth Forest Ecosystems in the Western United States. Hutchinson Ross, Stroudsburg,
Surface Process and Landforms 28, 409–424. pp. 267–291.
May, C.L., Gresswell, R.E., 2003b. Large wood recruitment and redistribution in Swanson, F.J., Lienkaemper, G.W., 1978. Physical consequences of large organic
headwater streams in the southern Oregon Coast Range, U.S.A. Canadian debris in Pacific Northwest streams. U.S. Forest Service, General Technical
Journal of Forest Research 33, 1352–1362. Report PNW-69, 12 pp.
Meleason, M.A., Davies-Colley, R., Wright-Stow, A., Horrox, J., Costley, K., 2005. Swanson, F.J., Lienkaemper, G.W., Sedell, J.R., 1976. History, physical effects and
Characteristics and geomorphic effect of wood in New Zealand’s native forest management implications of large organic debris in western Oregon streams.
streams. International Review of Hydrobiology 90(5–6), 466–485. U.S. Forest Service, General Technical Report PNW-56, 15 pp.
Montgomery, D.R., Buffington, J.M., Smith, R.D., Schmidt, K.M., Pess, G., 1995. Thévenet, A., 1998. Intérêt des débris ligneux grossiers pour les poisons dans les
Pool spacing in forest channels. Water Resources Research 31(4), grands cours d’eau. Pour une prise en compte de la dimension écologique des
1097–1105. débris ligneux grossiers dans la gestion des cours d’eau. Ph.D. Thesis,
Montgomery, D.R., Collins, B.D., Buffington, J.M., Abbe, T.B., 2003. Geomorphic University of Lyon 1, Lyon, France, 100 pp.
effects of wood in rivers. In: Gregory, S.V., Boyer, K.L., Gurnell, A.M. (Eds.), The Thompson, D.M., 1995. The effects of large organic debris on sediment processes
Ecology and Management of Wood in World Rivers. American Fisheries Society, and stream morphology in Vermont. Geomorphology 11, 235–244.
Bethesda, MD, pp. 21–47. Triska, F.J., 1984. Role of wood debris in modifying channel geometry and riparian
Mosley, P., 1981. The influence of organic debris on channel morphology and areas of large lowland river under pristine conditions: a historical case study.
bedload transport in a New Zealand forest stream. Earth Surface Processes and Verhandlungen – Internationale Vereinigung für Theoretifche und Angewandte
Landforms 6(6), 571–579. Limnologie 22, 1876–1892.
Moulin, B., Piégay, H., 2004. Characteristics and temporal variability of large woody Trotter, E.H., 1990. Woody debris, forest-stream succession, and catchment
debris trapped in a reservoir on the river Rhône (Rhône): implications for river geomorphology. Journal of the North American Benthological Society 9,
basin management. River Research and Applications 20, 79–97. 141–156.
36 Wood Entrance, Deposition, Transfer and Effects on Fluvial Forms and Processes

Wallace, J.B., Webster, J.R., 1996. The role of macroinvertebrates in stream Wyzga, B., Zawiejska, J., 2005. Wood storage in a wide mountain: case study of
ecosystem function. Annual Review of Entomology 41, 115–139. the Czarny Dunajec, Polish Carpathians. Earth Surface Processes and Landforms
Wallace, J.B., Webster, J.R., Meyer, J.L., 1995. Influence of log additions on 30, 1475–1494.
physical and biotic characteristics of a mountain stream. Canadian Journal of Young, W.J., 1991. Flume study of the hydraulic effects of large woody debris in
Fisheries and Aquatic Sciences 52, 2120–2137. lowland rivers. Regulated Rivers: Research and Management 6, 203–211.
Wallerstein, N.P., Thorne, C.R., 1998. Computer model for the prediction of scour at Zimmerman, R.C., Goodlett, J.C., Comer, G.H., 1967. The influence of vegetation on
bridges affected by large woody debris. Water Resources Engineering 1, 157–162. channel form of small streams. In: Proceedings of International Association for
Wohl, E., Ogden, F.L., Goode, J., 2009. Episodic wood loading in a mountainous Scientific Hydrology, Symposium on river morphology. International Association
neotropical watershed. Geomorphology 111, 149–159. for Scientific Hydrology, Wallington, United Kingdom, pp. 255–275.
Wong, M.K.M., Goh, T.K., Hodgkiss, I.J., et al., 1998. Role of fungi in freshwater
ecosystem. Biodiversity and Conservation 7, 1187–1206.

Biographical Sketch

Yves-Franc- ois Le Lay , 35 years old, is an assistant professor of geography, Department of Social Sciences, Ecole
normale supérieure de Lyon (CNRS-UMR 5600 Environnement, Ville, Société). In 2007, he successfully defended
his PhD thesis ‘Man and wood in rivers. Representations, practices and management strategies in the framework
of watercourse maintenance’. In 2008, he was recipient of the Lyon City young researcher prize for his scientific
contribution. He is specialized in aquatic and forest environments, with a special focus on the assessment of
public and expert perception of nature objects (in-channel wood, gravel bars, remarkable trees, and urban parks),
on the evaluation of practices related to watercourses and vegetation in mountainous, Mediterranean and urban
areas, and on the applications to environmental education.

Bertrand Moulin, 39 years old, has obtained PhD in environmental geography from the University of Saint-
Etienne (France). In 2005, he successfully defended his PhD thesis ‘Spatial and temporal variability of wood in the
catchment of Isere River upstream from Grenoble (France)’. He is now a geomorphologist at the Institut national
de recherches archéologiques préventives (INRAP). He is specialized in fluvial geomorphology, geoarcheology,
dendrology, and geomatics and works with the USGS on several contracts since 2005. He is also active in
AEMGEO, an association of environmental evaluation which is implicated in different research programs on
urban stream dynamics.

Hervé Piégay, 45 years old, Research Director at CNRS/UMR 5600. He is a geographer specialized in fluvial
geomorphology with main issues on interactions between vegetation and fluvial forms and development of tools
and management strategies for practitioners (restoration, monitoring, planning, and maintenance). He has
published more than 110 papers in peer-reviewed journals and edited books in the fields of fluvial geomorph-
ology of mountain and piedmont rivers and is strongly involved in interdisciplinary research groups, such as the
Zone Atelier Bassin du Rhône (LTER CNRS structure). He has developed research on wood in large rivers, opening
research on wood movement, temporal variability of wood in rivers, and human perception of wood. He has
coordinated several edited books, notably one on Tools in Fluvial Geomorphology with M.G. Kondolf (2003)
published by J. Wiley.
 12.4 River Processes and Implications for Fluvial Ecogeomorphology: A
European Perspective
M Rinaldi, University of Florence, Florence, Italy
B Wyżga, Institute of Nature Conservation, Polish Academy of Sciences, Krakow, Poland
S Dufour, Université Rennes 2, Rennes, France
W Bertoldi and A Gurnell, Queen Mary, University of London, London, UK
r 2013 Elsevier Inc. All rights reserved.

12.4.1 Introduction 37
12.4.2 The Long-term Perspective: Past, Present, and Future Trends in Channel Adjustments 38
12.4.2.1 Past Trends in Channel Adjustment 38
12.4.2.2 Riparian Vegetation and Channel Change 39
12.4.2.3 Impacts of Climate Change on Channel Dynamics 41
12.4.3 Progress in Understanding and Modeling Channel Processes Related to Fluvial Ecogeomorphology 42
12.4.3.1 Sediment Transport 42
12.4.3.2 Bank Erosion 43
12.4.3.3 Channel Pattern 44
12.4.4 River Processes and Ecogeomorphology 46
12.4.4.1 Ecological and Geomorphic Processes 46
12.4.4.2 Fluvial Geomorphology and River Restoration in Europe 46
12.4.4.3 Hydromorphology and WFD 47
References 48

Abstract

In this chapter, the state of the art of the research of fluvial processes and their linkages with ecology are presented, with a
focus on European physical context. European river systems have experienced a long history of catchment-wide (land-use
changes) and in-channel human disturbances (channelization, dams, sediment mining, etc.). In many rivers, the impacts
on river morphology culminated in the twentieth century, particularly during recent decades, and major alterations to river
functioning are also forecast for the nearby future due to climate changes. Considerable progress in understanding,
quantifying, and modeling channel processes has emerged from recent European research and findings on sediment
transport, bank erosion, and the formation of various channel patterns in rivers are discussed. Major concepts linking the
structure and functioning of river ecosystem with fluvial processes are presented, followed by a discussion of the impacts of
geomorphic processes on the connectivity, habitat heterogeneity, and a scale of flood disturbance in riverine ecosystems.
With the increasing emphasis on reestablishing good ecological quality of European rivers, an improved understanding of
the impacts of hydromorphology on river biodiversity and defining the scientific bases for river management and restor-
ation by the use of process-oriented approach are necessary.

12.4.1 Introduction particularly the installation of dams; and sediment mining (Petts
et al., 1989). These disturbances have induced complex and
Physical river processes are increasingly seen as vital for creating multiple phases of channel adjustments, with a number of
and maintaining physical habitats and aquatic and riparian detrimental environmental, ecological, and societal effects
ecosystems. As a result, their understanding as well as linkages (Bravard et al., 1999). A first step for understanding present
and feedbacks with ecology and hydrology are assuming crucial process–form interactions and predicting future trends of
importance for river management and restoration. channel adjustments is to reconstruct the past history of human
European river systems have experienced a long history of impacts and the long-term trajectory of channel changes.
human impacts and modifications including changes in catch- Physical processes and their linkages with ecosystem
ment and floodplain land use; channelization; flow regulation, quality have become a priority within the context of the
Water Framework Directive (WFD; European Commission,
2000), with ‘hydromorphology’ forming one of the aspects
that need to be considered in the monitoring and ecological
Rinaldi, M., Wyżga, B., Dufour, S., Bertoldi, W., Gurnell, A., 2013. River assessment of rivers. This increasing focus on the physical
processes and implications for fluvial ecogeomorphology: a European
perspective. In: Shroder, J. (Editor in chief), Butler, D.R., Hupp, C.R. (Eds.),
habitats created by geomorphological processes has high-
Treatise on Geomorphology. Academic Press, San Diego, CA, vol. 12, lighted the importance of these processes in sustaining
Ecogeomorphology, pp. 37–52. biodiversity.

Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00321-3 37

38 River Processes and Implications for Fluvial Ecogeomorphology: A European Perspective

Against this background, the aim of this chapter is to most common modifications were embankments, channeli-
synthesize the current state of the art in relation to fluvial zation, and flow diversions, which were undertaken to provide
processes and their linkages with ecology. In particular, the flood protection, support the expansion of agriculture, and,
importance of process-oriented approaches and strategies for lately, to improve navigation along large rivers. From the end
river management and restoration are emphasized with a of the nineteenth century, apart from reafforestation and
focus on the European physical context. interventions for stabilizing hillslopes (Figure 2(a)), a large
number of check dams were built along headwater channels in
many mountain areas (Figure 2(b)) coupled with larger dams
12.4.2 The Long-term Perspective: Past, Present, along downstream river reaches. These structures drastically
and Future Trends in Channel Adjustments reduced the sediment supply to downstream reaches and also
created discontinuities in the fluxes of water and sediment,
Fluvial landscapes follow complex trajectories in time de- impacting morphological processes and their interrelation-
pending on their geographical context, because they result ships with ecological dynamics. However, large impacts on
from the combination and the overlay of three main inter- river morphology became most marked in many rivers during
acting factors: hydrological and morphological processes; the twentieth century, and, particularly, during recent decades,
biological interactions; and human setting (Dufour and in response to increasing urban and industrial development
Piégay, 2009). Rivers continuously adjust to environmental after World War II.
conditions following a trajectory of adjustments as environ- Numerous studies on European rivers have demonstrated
mental conditions change. The conjunction of key drivers is similar trends of channel adjustments. These have included an
continuously variable in time and space (climate, vegetation early, historical period characterized by aggradational pro-
cover, and human activities) and the local conditions are al- cesses affecting different components of the fluvial system
ways new. Reconstructing the fluvial trajectory and the im- (alluvial plain, channel bed, and delta), followed by a reversal
pacting factors are key issues for understanding past evolution of the general aggradational trend in the late nineteenth cen-
and possible future trends of a fluvial system. tury and twentieth century (Figure 3) as a result of various
types of human disturbances (Petts et al., 1989; Łajczak, 1995;
Bravard et al., 1997) and widespread hillslope reafforestation
and upland sediment retention, against a background of cli-
12.4.2.1 Past Trends in Channel Adjustment
mate changes following the end of the Little Ice Age. These
Changes to catchment and riparian vegetation cover were the trends have been observed in many parts of Europe, including
earliest human activities inducing significant adjustments to the piedmont areas of mountain (Liébault and Piégay, 2002)
river channel and valley floor morphology. In Europe, de- and Mediterranean (Hooke, 2006) regions.
forestation was linked with pastoral and then agricultural ac- A significant expansion in the literature on channel ad-
tivities that started in the Neolithic period, ca 4500 BC justments has occurred during the last two decades, with
(Williams, 2000). Since then deforestation across Europe has studies focusing on river channel changes caused by human
progressed with some phases of notably increased intensity, disturbances emanating from many areas of Europe, including
such as the Roman period, culminating in peak levels of de- France (Liébault and Piégay, 2001, 2002), Poland (Wyżga,
forestation in the nineteenth century. At this time, most 1993b, 2001a, 2001b, 2008), Italy (Rinaldi and Simon, 1998;
montane and foothill slopes across Europe were under culti- Rinaldi, 2003; Surian and Rinaldi, 2003; Surian et al., 2009a)
vation or grazing. Because the intensity of slope wash on (Figure 4), Spain (Garcia-Ruiz et al., 1997; Rovira et al.,
cultivated slopes exceeds that on forested slopes by a few 2005), and the United Kingdom (e.g. Winterbottom, 2000). In
orders of magnitude, deforestation was commonly accom- particular, in-channel sediment mining has been widely rec-
panied by aggradation on valley floors (Starkel, 1995). From ognized among the human impacts inducing large effects on
the seventeenth to the nineteenth centuries, the intensity of river morphology during the twentieth century (Rinaldi et al.,
runoff and erosion on hillslopes became so high that wide- 2005; Rovira et al., 2005). Despite the different types of dis-
spread channel aggradation occurred with a downstream- turbances, common channel responses have been observed,
progressing transformation of single-thread to braided chan- with two dominant types of morphological adjustments being
nels in many mountain and piedmont valley reaches (e.g., channel incision and narrowing. In most cases, these re-
Wyżga, 1993a; Gurnell et al., 2009). In contrast with con- sponses have been attributed to alterations in sediment fluxes
tinuing deforestation in other parts of the world, during the (Liébault and Piégay, 2002; Surian and Rinaldi, 2003), but
twentieth century the trend was reversed in many European where upstream progression of channel changes have oc-
catchments, especially montane ones (Figure 1). This in- curred, channelization and a resultant increase in the river’s
creasing forest cover contributed to a reduction in catchment transport capacity have been implicated (Wyżga, 2008;
sediment supply (Liébault et al., 2005) followed by channel Zawiejska and Wyżga, 2010).
narrowing and incision (e.g., Kondolf et al., 2002; Lach and These channel adjustments have had dramatic environ-
Wyżga, 2002; Keesstra et al., 2005) and a decline of multi- mental and societal effects (Bravard et al., 1999), including
thread channel patterns in many rivers (Gurnell et al., 2009). many ecological effects such as: (1) destruction or reduction of
In addition to catchment land-use disturbances, European in-channel alluvial features that are important for habitat di-
rivers have experienced a long history of in-channel human versity; (2) bed coarsening and loss of gravel of a size
modifications stretching back to Roman times (Petts et al., suitable for fish spawning; (3) water-table lowering and con-
1989; Billi et al., 1997). Up to the nineteenth century, the sequent effects on riparian vegetation; and (4) disruption of
 River Processes and Implications for Fluvial Ecogeomorphology: A European Perspective 39

Figure 1 Examples from French Alpine regions of increasing forest cover on hillslopes and fluvial corridors, and associated channel narrowing.
(a) Ubaye River and its tributary, Riou Bourdoux (downstream Barcelonnette, Southern Alps). Reproduced from Piégay, H., Grant, G., Nakamura,
F., Trustrum, N., 2006. Braided river management: from assessment of river behaviour to improved sustainable development. In: Sambrook-
Smith, G.H., Best, J.L., Bristow, C.S., Petts, G.E. (Eds.), Braided Rivers: Process, Deposits, Ecology and Management. Special Publication 36 of
the International Association of Sedimentologists. Blackwell, Oxford, pp. 257–275, with permission from Wiley. (b) Drome River in the Vercheny
Plain, 30 km upstream of Saillans. Reproduced from Kondolf, G.M., Piégay, H., Landon, N., 2002. Channel response to increased and decreased
bedload supply from land use change: contrasts between two catchments. Geomorphology 45, 35–51.

lateral hydraulic connectivity with the floodplain, with the burning, and agricultural land use, all of which favored pi-
latter effectively becoming a terrace, and a resulting loss of wet oneer stages and nonwoody communities.
areas and related habitats. Initially underestimated by fluvial geomorphologists, the
Given all these negative impacts, a capability to predict or influence of riparian vegetation (both standing and dead
generate possible scenarios of future trends of channel ad- trees) on the morphological river pattern is now well estab-
justments is essential to prevent their consequences. An ex- lished (Corenblit et al., 2007; Francis et al., 2009; Gurnell
ample of such a capability is the analysis of how different et al., 2009). Modification to channel morphology can be
sediment management strategies might affect future channel driven not only by changes in sediment supply and/or flood
dynamics (Surian et al., 2009b). discharges but also by changes in local human activities (e.g.,
grazing abandonment). This point is critical in the European
context because most river corridors have experienced strong
human influence over several centuries (Petts et al., 1989) and
12.4.2.2 Riparian Vegetation and Channel Change
even millennia (e.g., Higler, 1993). Thus, the general trend
One particular consequence of changes in morphological over the Holocene has been a decrease in natural habitats and
channel pattern over time is their effect on riparian vegetation. a progressive humanization of riparian areas. However, this
Phases of dynamic channel change, such as those that oc- human impact has varied due to changes in population
curred during the Little Ice Age, were characterized by an open densities and technology with some periods apparently suf-
landscape dominated by pioneer vegetation communities, fering more intense impacts, such as the Roman period or the
whereas phases of more stable channel morphology favored climatic optimum during the Middle Ages. Historical data
the establishment of woodland dominated by hardwood document heavily impacted riparian areas during the nine-
species (Figure 5). At a reach scale, the physiognomy of the teenth century, when they were extensively grazed or cultivated
fluvial landscape was also shaped by direct human impacts on and dominated by a predominantly open landscape and
riparian vegetation, including woodland clearance, grazing, pioneer vegetation communities (e.g., Kondolf et al., 2002).
40 River Processes and Implications for Fluvial Ecogeomorphology: A European Perspective

Figure 2 Examples of works done at the end of the nineteenth century in France for stabilization of hillslopes (a), and longitudinal profile of
mountain streams (b). Reproduced from Piégay, H., Rinaldi, M., 2006. Gestione sostenibile dei sedimenti in fiumi ghiaiosi incisi in Francia. Atti
Giornate di Studio ‘‘Nuovi approcci per la comprensione dei processi fluviali e la gestione dei sedimenti. Applicazioni nel bacino del Magra.’’
Sarzana, 24–25 October 2006. Autorità di Bacino del Fiume Magra, pp. 59–80, with permission from Autorita.

Figure 3 Sedimentary record of the aggradational/degradational tendencies of the middle Raba River, Polish Carpathians, from the last two
centuries. The shallow braid was eroded in the upper part of the sequence of overbank deposits and filled with massive gravel at about the turn
of the twentieth century, with the culmination of the aggradational river tendency. A high position of the braid above the low-water level in the
contemporary channel testifies to the rapid incision of the Raba over the last century. Marks on the rope stretched along the cut bank are spaced
at 1-m intervals. Reproduced from Wyżga, B., 2008. A review on channel incision in the Polish Carpathian rivers during the 20th century. In:
Habersack, H., Piégay, H., Rinaldi, M. (Eds.), Gravel-Bed Rivers VI: From Process Understanding to River Restoration. Developments in Earth
Surface Processes 11. Elsevier, Amsterdam, pp. 525–555.
 River Processes and Implications for Fluvial Ecogeomorphology: A European Perspective 41

Channel types Single-thread Transitional Braided

Increasing incision
Increasing narrowing
(relative to the initial morphology)
Figure 4 Summary of main types of channel adjustments in Italian rivers during the past 100 years. Starting from three initial morphologies,
different channel adjustments were observed according to variable amount of incision and narrowing. Modified from Surian, N., Rinaldi, M.,
2003. Morphological response to river engineering and management in alluvial channels in Italy. Geomorphology 50, 307–326.

The relatively high fluvial activity at that time contributed to the 12.4.2.3 Impacts of Climate Change on Channel Dynamics
maintenance of an open landscape, notably in high-gradient
river reaches. Since about the end of the nineteenth century, Besides human activities, climate change is another important
European river margins have exhibited two different trends. factor inducing long-term channel adjustments due to its
Along upland and piedmont river reaches, a decrease in popu- impacts on the amount and timing of runoff, the vegetation
lation density and changes in management practices have cover, and activation or suppression of sediment sources. For
allowed an increase in the area of riparian forest. At the same instance, increased frequency/magnitude of floods and en-
time, many valley reaches, especially in lowland areas, have re- hanced fluvial activity during the second half of the Little
mained under cultivation and grazing or, where protected from Ice Age (1750–1900) were documented across vast areas of
flooding, have experienced industrial or urban development. Europe (Rumsby and Macklin, 1996). In mountain and
The progressive clearance of riparian vegetation, coupled piedmont areas of Western, Central, and Southern Europe,
with removal of wood from channels, has led to a heavy de- they coincided with the phase of intense agricultural and
crease in wood loading in European rivers. Although this has pastoral activities on hillslopes and led to increased sediment
eliminated a considerable source of hydraulic roughness from delivery to channels, bed aggradation, channel widening, and
channels, the desnagging of European rivers has not resulted river braiding (e.g., Bravard, 1989; Wyżga, 1993a).
in vertical instability of their beds (Brierley et al., 2005), Although climate change during the Little Ice Age and the
probably because the reduction in flow resistance has been rate of subsequent twentieth-century warming are considered
compensated by an increasing sediment supply as a result of exceptional in the Holocene, they are relatively small in
deforestation across the catchment. Only in the twentieth comparison with projected global temperature increases of
century, when catchment sediment supply decreased following 1.4–5.8 1C under various scenarios over the twenty-first cen-
an increase in forest cover, could the lack of hydraulic tury (European Commission, 2005). A moderate temperature
roughness associated with woody debris contribute to the increase in Southern, Central, and Eastern Europe is forecast
rapid channel incision that has been recorded recently in to be accompanied by a reduction in precipitation totals,
many European rivers. This trend contrasts with the situation whereas a much larger increase in temperature associated with
in North America and Australia where rivers and valleys were increased winter and spring precipitation is forecast for
the first landscape features modified after European settlement Northern Europe (European Commission, 2005). Moreover,
and where clearance of riparian vegetation and removal of the incidence of extreme summer precipitation events is pre-
woody debris immediately induced rapid, deep channel dicted to increase over large areas of Europe (e.g., Christensen
incision (Brierley et al., 2005). and Christensen, 2004).
42 River Processes and Implications for Fluvial Ecogeomorphology: A European Perspective

500 m 1936 flash flooding, increase soil erosion (cf. Nearing et al., 2005),
and, because of the nonlinear relationship between catchment
water and sediment discharges (Coulthard et al., 2008),
sediment delivery to channels is likely to increase sub-
stantially. By contrast, in Northern Europe the magnitudes of
the snowmelt floods that are typical of this region are likely to
decrease despite increased winter precipitation. This is because
a larger proportion of winter precipitation is likely to fall as
rain and there may also be an increased occurrence of mid-
winter thaws that will tend to reduce the amount of water
accumulated in the winter snow pack. The opposite tendencies
of precipitation totals in Central and Northern Europe may
1961 cause differing trends in the evolution of glaciers and pro-
glacial rivers in the Alps and Scandinavian Mountains. In the
Alps, rapid retreat of glaciers, coupled with reduced sediment
delivery to proglacial rivers, may result in their progressive
stabilization by developing riparian vegetation (cf. Gurnell
et al., 1999), whereas in Scandinavia, glacier advance linked
with increased sediment production may induce enhanced
river braiding.
While predicting changes to ecogeomorphological river
processes under rapid climate change is subject to great
1984 uncertainty, it is clear that both riverine habitats and bio-
coenoses will substantially alter with future channel adjust-
ments, latitudinal and altitudinal shifts in the density and
composition of riparian vegetation, changes to water tem-
perature, river regime and the persistence of runoff, etc. This
suggests that the historical state of rivers and riverine bio-
coenoses may be unsuitable for defining reference conditions
for evaluating the hydromorphological and biotic quality
of rivers.

2000
12.4.3 Progress in Understanding and Modeling
Channel Processes Related to Fluvial
Ecogeomorphology

12.4.3.1 Sediment Transport

Physical processes, including those of sediment production,
transport, and storage, are increasingly seen as vital for the
ecological functioning of fluvial systems. Among these, sedi-
ment transport is certainly one of the most important in
Figure 5 Corridor evolution from 1937 to 2000 of the Arve River natural river channels, and its measurement and prediction
(tributary of the Rhône River, France). Watershed land-cover continue to receive considerable attention. Progress in
changes, dams, embankments, and gravel-mining-generated channel understanding and modeling sediment transport has been
metamorphosis from a braided to a single-thread channel facilitated by the collection of extensive field data over recent
morphology. Source: IGN. decades under a relatively wide range of flow, bed load
transport, and channel slope conditions, which has supported
These climate changes will have significant impacts the development of new bed load formulae and/or the testing
on physical and ecogeomorphological processes in rivers. of older, well-known equations (Diplas and Shaheen, 2008).
With increased temperature and reduced precipitation in In some cases, these data have also provided the opportunity
catchments, many smaller rivers in Southern Europe may to investigate the behavior of gravel streams and the structure
change from perennial to seasonal ones. A reduction in the of channel beds in the context of watershed processes and
density of vegetation cover in the region will tend to increase characteristics (i.e., Hassan et al., 2008). Collection of new
soil erosion (Nearing et al., 2005) and catchment sediment field data has certainly received a boost from recent techno-
supply, resulting in increased bedload flux in rivers and the logical progress, with traditional methods and sampling being
development of nonarmored channel beds (cf. Reid et al., combined with innovative techniques such as the use of
1999). Across most of the continental area of Europe, the piezoelectric sensors (Rickenmann and McArdell, 2008) or the
increased occurrence of extreme rainfall events will intensify magnetic bedload movement detector (Hassan et al., 2009).
 River Processes and Implications for Fluvial Ecogeomorphology: A European Perspective 43

12.4.3.2 Bank Erosion across a hypothetical drainage basin in response to down-

stream variations in energy conditions, bank material, and
Bank erosion is second key process in fluvial dynamics, which climatic conditions (see, e.g., Lawler, 1992). More recently,
has a large influence on a series of physical, ecological, and Fonstad and Marcus (2003) have hypothesized the existence
socioeconomic issues. It has become progressively recognized of self-organized in riverbank systems, such that they can
that some lateral dynamics associated with bank erosion have change internally (i.e., organize themselves) without any
positive effects in terms of the promotion of riparian vege- change in the magnitude and frequency of external inputs to
tation succession, the establishment and evolution of river the system.
and floodplain morphology, and the creation of dynamic Although a comprehensive theory or model capable of
habitats crucial for aquatic and riparian species (Florsheim predicting the location and amount of lateral mobility at a
et al., 2008). catchment scale is not yet available, significant progress in
The identification and prediction of the spatial distribution modeling various processes has been made at the scale of a
of bank processes, the tendency to lateral channel mobility, single bank profile. A recent comprehensive review on mod-
and its controlling factors collectively form an important issue. eling bank-erosion processes by Rinaldi and Darby (2008)
As a first approximation, the lateral mobility distribution at a emphasized how dynamic interactions and feedbacks between
river network scale can be related to interaction between different processes are the key to understanding riverbank
stream energy (or power) and boundary resistance. It is also dynamics (Figure 6). Developments of a simulation approach
well recognized that bank retreat is the integrated product of based on the interaction of the main physical processes
three interacting groups of processes (subaerial processes, (hydrodynamics, groundwater, and mass failure) are reported
erosion processes, and mass wasting). In combination, these in Rinaldi et al. (2008b). Numerical modeling approaches,
suggest the existence of changing bank process dominance however, need to account for the large parameter

98

97
a
96
Elevation (m asl)

95

94
b
93
c
92

91

90

89
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
(a)

2.5 99
Sediment delivered m3 m−1)

Flow stage 98
2.0
Flow stage (m asl)

Fluvial erosion
97
Cantilever failure
1.5 Slide failure 96

1.0 95
94
0.5
93
0.0 92
0 5 10 15 20 25 30 35 40
(b) Time (h)
Figure 6 Riverbank stability analysis by the application of integrated groundwater flow–fluvial erosion–mass failure models to a bank of the
Sieve River, Italy. (a) Discretization of the riverbank for the groundwater flow model; (b) results of the simulation in terms of sediment delivered
by the different processes. Modified from Rinaldi, M., Darby, S.E., 2008. Modelling river-bank-erosion processes and mass failure mechanisms:
progress towards fully coupled simulations. In: Habersack, H., Piégay, H., Rinaldi, M. (Eds.), Gravel-Bed Rivers 6: From Process Understanding
to River Restoration. Series Developments in Earth Surface Processes 11. Elsevier, Amsterdam, pp. 213–239.
44 River Processes and Implications for Fluvial Ecogeomorphology: A European Perspective

uncertainties, as is illustrated in some first attempts to quantify sensing survey techniques has greatly improved the potential
these effects on bank stability modeling (see, e.g., Samadi to monitor braided channel networks at the high spatial and
et al., 2009). Because a mechanistic, process-based approach temporal resolutions needed to advance scientific under-
may be impractical to implement when the relevant processes standing (Lane et al., 2003; Westaway et al., 2003). The use of
are difficult to parameterize, an empirical approach of bank aerophotogrammetry, terrestrial and airborne light detection
profile evolution may be preferred in some cases (see, e.g., and ranging (LiDAR), and satellite multispectral imagery
Pizzuto, 2009). allows the investigation of channel processes in relation to
An obvious linkage between bank and ecological processes flood events and the influence of vegetation patterns on river
is represented by the role that riparian vegetation may have on morphology.
bank stability. Recently, much progress has been made in this At a large spatial scale, patterns of sediment transport ac-
field, in particular emphasizing understanding and modeling tivity in braided channel networks suggest that just a few
of hydrological effects and mechanical root reinforcement branches are simultaneously active (Ashmore, 2001). As a re-
(Simon and Collison, 2002; Pollen, 2006; Van De Wiel and sult, it is possible to distinguish between a total braiding index
Darby, 2007). Prediction of bank failure including vegetative and an active braiding index, where the latter is generally
elements is very relevant for quantifying wood recruitment in 40–60% of the former. Bertoldi et al. (2009b) proposed di-
the fluvial system (Downs and Simon, 2001) as well as for mensionless relationships between external parameters (dis-
managing and/or mitigating excessive sediment input (Simon charge, longitudinal bed slope, and grain size) and average
et al., 2006). planform parameters, distinguishing between a local scale
related to sediment transport and a single branch or node, and
a global scale that represents the historical legacies of temporal
evolution.
12.4.3.3 Channel Pattern
At the local scale, a peculiarity of multithread rivers is the
Understanding the linkages between ecology and geomorph- presence of confluences and bifurcations that act as links be-
ology in fluvial systems requires a deep knowledge about their tween the various branches. Confluences have been widely
spatial and temporal scales of evolution (for instance Hupp studied (e.g., Best, 1988; Rice et al., 2008) focusing on both
and Rinaldi, 2007). In particular, each major channel pattern their morphological evolution and their fully three-dimen-
(braiding, meandering, and anastomosing) is characterized by sional (3D) flow fields (Rhoads and Sukhodolov, 2004). Re-
specific channel processes that control the system dynamics. cently, Parsons et al. (2007) investigated the 3D flow field and
Recent developments in geomorphological research have fo- mixing processes in a large confluence–diffluence unit using
cused in this direction, taking advantage of both empirical acoustic Doppler current profiling coupled with bed topo-
observations and mathematical modeling to attain a quanti- graphy survey. Recently, bifurcation dynamics have been in-
tative description of river forms and processes, and of their vestigated by a combination of quantitative experimental,
evolution. numerical, analytical, and field-based research (Bolla Pittaluga
Braided rivers are characterized by a multitude of inter- et al., 2003; Zolezzi et al., 2006; Bertoldi and Tubino, 2007;
connected, highly movable channels, separated by bars or Kleinhans et al., 2008). This research has identified the occur-
relatively short-lived islands developing through vegetation rence of stable bifurcations with strongly asymmetric water
encroachment on mid-channel bars. Generally, they are partition in the distributaries. Water and sediment fluxes in
characterized by high values of the stream power and by a particular branches (or braids) depend on local flow conditions
large amount of transported sediment that is widely deposited constrained by gradient, channel curvature, the aspect ratio of
in the braidplain (Ashmore, 1991). A distinct feature of brai- the branches, and they can also be strongly affected by the
ded rivers resulting from their complex sedimentary structure presence of vegetation and by the migration of sediment bars.
and flow patterns is their thermal behavior. Widely varying Numerical modeling of braided systems is a challenging
water temperatures reflect the occurrence of multiple branches task because of the continuous and rapid rearrangement of the
with different physical dimensions affected by surface – sub- branches and nodes. The standard approach that solves the
surface water exchange through the hyporheic zone. This governing equations for the fluid and solid phases (Enggrob
temperature patchiness, in addition to their flow velocity, and Tjerry, 1999) can be used only for small areas and short
sedimentary, morphological, and vegetation complexity, has time intervals, and so alternative models for braiding have
strong ecological importance (Gurnell et al., 2005; Acuna and been developed. Murray and Paola (1994) have shown that a
Tockner, 2009). Braided rivers were once very common in simple cellular model is able to reproduce many features of
European regions, although they have undergone dramatic braided networks. Thomas and Nicholas (2002) and Coul-
changes in the last century, mainly due to human activities thard et al. (2007) improved this methodology, showing that
(see Surian and Rinaldi, 2003; Gurnell et al., 2009). it can be used also to model actual river evolution. Jagers
The occurrence of many branches and nodes of different (2003) has also studied the problem of modeling planform
sizes, continuously interacting to support a highly dynamic changes in braided rivers, by implementing two different
planform evolution, makes braided rivers extremely complex models: a neural network and a branches model, testing their
and the identification of their inherent spatial and temporal accuracy with observed data. The latter approach seems to be
scales challenging, although statistical studies suggest that the most promising in ensuring good temporal predictability
their multichannel geometry is self-affine over a range of from physically based evolution rules. Recently, Perona et al.
spatial scales (Foufoula-Georgiou and Sapozhnikov, 2001). (2009) proposed a stochastic model for sediment and vege-
However, in the last decade, the development of remote- tation dynamics in Alpine braided rivers, where the key
 River Processes and Implications for Fluvial Ecogeomorphology: A European Perspective 45

processes were floodplain erosion driven by floods and col- Advances in modeling have included the use of computa-
onization of vegetation patches (Figure 7). The model re- tional fluid dynamics to predict the formation, development,
produces changes in riparian vegetation cover well. and migration of free-forming meander bends, through a fully
Meandering is the most common planform configuration 3D model (Ruther and Olsen, 2007). The development of a
of single-thread channels. Meanders are more likely to develop cellular scheme has included a novel technique for deter-
in rivers wandering through cohesive alluvial plains, but they mining bend radius of curvature, used to drive meandering
can form also in nonsedimentary and tidal environments and lateral erosion (Coulthard and Van De Wiel, 2006). The
(Seminara, 2006). Channel axis evolution has been widely impact of spatial variations in channel width on wavelength
studied through geomorphological field observation (e.g., selection, mid-channel bar formation, and bank erosion
Hooke, 2008), and theoretical fluid dynamic-based model (Luchi et al., 2009), and the long-term evolution of meanders
(see Camporeale et al. (2007) for a review). In particular, a as a function of their morphological regime (Frascati and
planform evolution equation has been obtained (Zolezzi and Lanzoni, 2009) and their migration speed (Crosato, 2008)
Seminara, 2001) relating the outer bank erosion (and inner have also been investigated through numerical modeling. A
point bar accretion) to the 3D flow field induced by axis key advance based on modeling has been investigation of the
curvature. This allows a good description of bed configuration role of vegetation growth in controlling the long-term evo-
and flow field, both in time and in space. lution of meander planforms (see Perucca et al., 2006, 2007).
Recent contributions add further ingredients to this general Recently, anastomosing rivers have been recognized as a
picture that support better understanding of the spatial scales distinct channel pattern type that was probably quite common
at which eco-geomorphological linkages can take place. In- in the past across the lowlands of Europe. Such rivers consist
creasingly detailed field measurements are revealing much of multiple, low-energy, interconnected channels separated by
about the functioning of meanders, for example, the potential long-lived, vegetated islands, which are large relative to the
influence of the flow field in sharp bends (Blanckaert and channels and originate through floodplain fragmentation
de Vriend, 2005), whereas larger-scale observations of chute (Knighton and Nanson, 1993). The channels are narrow,
cutoffs have investigated their impact on long-term channel deep, and laterally stable due to the low erodibility of vege-
sinuosity (Stolum, 1996; Hooke, 2004; Camporeale et al., tation-reinforced banks (Gradziński et al., 2003). An anasto-
2005) and their association with the chaotic nature of plan- mosing channel pattern develops through manifold avulsions
form meander evolution and the occurrence of self-organized coupled with slow abandonment of old channels under an
behavior. aggradational sedimentary regime linked with a rise of base

1999 2005

2009
Bank erosion

Vegetation growth

1999: Autoritá di Bacino Alto Adriatico, Italy

2005: NERC, UK
2009: University of Padova, Italy

Meters
0 200 400

Figure 7 Example of morphological evolution of a gravel-bed braided river over a time span of 10 years. Location of floodplain bank erosion
and extensive vegetation growth are highlighted.
46 River Processes and Implications for Fluvial Ecogeomorphology: A European Perspective

level (Makaske, 2001), substrate subsidence, and blocking of between the river channel and its floodplain, the frequency of
the channels by growing aquatic vegetation (Gradziński et al., floodplain inundation, and the scale of modification to the
2003) or debris dams (Harwood and Brown, 1993). Anasto- floodplain portion of the riverine ecosystem caused by flood
mosing rivers are now rare in Europe, because they charac- disturbances. Bed degradation, especially if operating in
teristically occur toward the lower reaches of river networks, watercourses underlain by a thin cover of alluvium, weakens
where channel gradients are relatively low and channel banks water/nutrient circulation across the channel bed, reduces the
are composed of fine cohesive alluvial sediments, and where size of the hyporheic zone, and may lead to complete loss of
such multithread channels have historically been simplified to the hyporheos with bedrock exposure on the riverbed.
support navigation. However, they were much more common Although the above-mentioned ecological models mostly
before the era of riparian deforestation and river channeliza- focus on the productivity of the river ecosystem, evaluation of
tion (Brown and Keough, 1991) and a few such multithread the ecological integrity of European watercourses is mainly
systems with narrow, deep, and laterally stable channels still based on the diversity of river biocoenoses. Connectivity,
remain active in valley reaches unaffected by channelization habitat heterogeneity and a moderate scale of flood disturbance
works (e.g., Brown, 1997; Gradziński et al., 2003). are key factors controlled by the dynamics of fluvial processes,
which contribute to the maintenance of high biodiversity of the
riverine ecosystem (Ward et al., 1999). Disequilibrium fluvial
processes disadvantageously modify lateral and vertical river
12.4.4 River Processes and Ecogeomorphology
connectivity, whereas longitudinal connectivity is disrupted or
reduced due to dam and weir construction. Successional phe-
12.4.4.1 Ecological and Geomorphic Processes
nomena and flood-related channel migration, erosion, and
During the last three decades, several scientific concepts have deposition of sediment and organic matter (including large
been developed to link the structure and functioning of river wood) create a complex mosaic of terrestrial and aquatic
ecosystems with fluvial processes. Notably, the concepts were habitats within the river corridor with suitable conditions for
complementary rather than competitive, addressing different diverse organisms and their different life stages. The location of
aspects of the functioning of fluvial systems. The River Con- patches with given habitat conditions changes through time
tinuum Concept (RCC; Vannote et al., 1980) predicted changes (shifting habitat mosaic; Pringle et al., 1988, Stanford et al.,
in the balance between allochthonous inputs and autoch- 2005) but, under dynamic equilibrium conditions, their
thonous production of organic matter as well as in the resultant amounts and proportions remain more or less the same (see,
composition of the macroinvertebrate communities that occur e.g., Van der Nat et al., 2003). Absolute magnitudes of floods
in response to changing physical conditions along the river are determined by hydrological phenomena but the disturb-
continuum. In contrast to the previous concepts of river zon- ances they cause in the floodplain portion of riverine ecosystem
ation, the RCC indicated that a watercourse is an open eco- depend also on the vertical tendency of the river channel
system in which gradual changes in physical conditions (such (Dufour and Piégay, 2008). In aggrading river systems, in-
as the width, depth, flow characteristics, complexity of the flow creased frequency/intensity of flood scouring and deposition
network, and interaction with the bank) are associated with prevents recovery of plant communities before the next dis-
continuous changes in the structure of riverine biocoenoses turbance, whereas in incising river systems the terrestialization
from the headwaters to the lower reaches. The Flood Pulse of aquatic water bodies within the floodplain and the matur-
Concept (FPC; Junk et al., 1989) focused on the lateral ex- ation of plant communities occur with reduced frequency/in-
change of water, nutrients, and organisms between river chan- tensity of flood inundation, scouring, or deposition of the
nel and the connected floodplain. It emphasized strong floodplain surface. In both cases, the tendencies result in re-
interactions between hydrological and ecological processes, duced biodiversity of floodplain communities (Hupp and
indicating that the pulsing of river discharge is the driving force Bornette, 2003). In accordance with the intermediate disturb-
in the river–floodplain system. Floods were considered as dis- ance hypothesis (e.g., Ward and Stanford, 1983), a highly di-
turbances leading to a regular setback of floodplain community verse, shifting mosaic of plant species on a floodplain can be
development, maintaining the system in an immature but maintained only with the intermediate frequency and intensity
highly productive stage. Tockner et al. (2000) extended the FPC, of flooding (Hupp and Bornette, 2003) that typifies the
indicating that expansion–contraction cycles occurring below floodplains bordering vertically stable rivers.
bankfull (flow pulses) also exert a significant influence on the
size and heterogeneity of habitat, connectivity, and functional
processes in river ecosystems. Moreover, recognition of the
12.4.4.2 Fluvial Geomorphology and River Restoration in
hyporheic zone as a habitat for some groups of river biota
Europe
(Stanford and Ward, 1988) as well as water and nutrient ex-
change across the channel bed indicated the significance of River restoration in Europe, as well as in other parts of the
vertical connectivity of river ecosystem. world, is increasingly including consideration of physical
Geomorphic processes operating in particular river reaches processes, such as bank erosion, sediment transport, channel
determine morphological characteristics of the river channel, incision, and water flow patterns, as a necessary condition for
floodplain, and valley sides and thus the arrangement of river enhancing river conditions and promoting channel recovery.
biocoenoses along the river continuum. Vertical tendencies of The first applications of morphological restoration in Europe
the channel bed, which reflect persistence or disruption of were located in northern countries (United Kingdom,
dynamic equilibrium, influence the degree of connectivity Denmark, etc.), and were mainly implemented on short
 River Processes and Implications for Fluvial Ecogeomorphology: A European Perspective 47

reaches of low-energy, generally small streams. Restoration integrate geomorphology and ecology; and (4) the use of
measures included increasing the variability in channel width process-based approaches and the integration of process
and depth, recreation of meanders, reintroduction of gravel to understanding with morphological interventions.
recreate fish spawning habitats, and reconnecting former
channels to the main channel (see for example Brookes,
1990).
12.4.4.3 Hydromorphology and WFD
Since the 1990s increasing scientific debates have occurred
regarding the appropriate spatial scale of interventions (local The European Commission (EC) WFD (European Commis-
scale vs. reach and basin scale), and the consequent need to sion, 2000) introduced the term ‘hydromorphology’, which
consider process-oriented approaches and self-restoration considers physical habitat conditions for aquatic biota as they
strategies (Clarke et al., 2003; Palmer et al., 2005; Wohl et al., are determined by hydrological regime and morphological
2005). A progressive evolution toward process-based restor- pattern. After several decades of concern about pollutants as
ation has occurred, where the aim is to restore natural geo- major impacts on the ecological integrity of watercourses,
morphic processes to promote conditions of self-sustaining identification of major stressors for river ecosystems now
physical diversity, rather than tackling problems through local concentrates on modifications to the flow regime, impacts of
interventions. In addition, applications of morphological artificial barriers on biota migration, water flow and sediment
restoration have shifted toward higher-energy, bedload-trans- transport, and modifications of river channel and floodplain
port-dominated channels (e.g., in the piedmont Alpine areas). forms and processes (Vaughan et al., 2009). Following its
In such environments, successful restoration must include the introduction, hydromorphology has increasingly become a
full spectrum of scales and consideration of the related natural cross-disciplinary topic at the interface between hydrology,
processes and human boundary conditions (Habersack and geomorphology, and ecology, and it is rapidly becoming
Piégay, 2008). Notwithstanding these increasing experiences central to practical applications for sustainable river manage-
and evolving approaches, European river restoration using ment (Newson and Large, 2006).
geomorphic principles is still in its infancy, emphasizing an- The inclusion of hydromorphological quality within the
alysis (i.e., assessment of problems, proposition of strategies WFD creates new perspectives and opportunities to embed
and interventions based on comprehension of processes, etc.) consideration of physical processes in future restoration ac-
rather than implementing specific interventions (see Gumiero tions and strategies. At the same time, the WFD poses a series
et al., 2008). However, there are exceptions to this general of issues, such as the problematic definition of a reference
trend, with a number of restoration projects implemented state, the assessment of deviation from that state, and the
during the last two decades, that have included considerations ecologically driven strategies for restoration, which must be
of morphological forms and processes, including: (1) channel delivered by regulators.
widening and dike relocation further away from the channel Nowadays, there is wide agreement that it is inappropriate
(Jäggi and Zarn, 1999; Habersack and Piégay, 2008); (2) for- to define a static reference state for the morphological forms
mer channel reconstruction and reconnection (Schropp, 1995; and processes in a river subject to restoration. The static ref-
Schropp and Bakker, 1998; Simons et al., 2001; Gurnell et al., erence state concept should be replaced by a guiding image of
2006; Habersack and Piégay, 2008; Hornich and Baumann, a dynamic river ecosystem (e.g., Palmer et al., 2005), which
2008; Muhar et al., 2008); (3) sediment reintroduction or actively reacts to the fluctuations of water discharge and
promotion of bedload supply input from floodplains, tribu- sediment flux and progressively adjusts to catchment en-
taries, and hillslopes (Habersack and Piégay, 2008; Rollet vironmental changes. To satisfy this requirement, reference
et al., 2008); and (4) enabling natural recruitment (Gregory processes and reference process–form interactions need to be
and Davis, 1992) and artificial placement of large wood collectively considered and defined within dynamic river sys-
(Gerhard and Reich, 2001). tems or reaches in a variety of European environments
Besides specific restoration measures and actions, preser- (Bertoldi et al., 2009a; Dufour and Piégay, 2009).
vation or recreation of natural physical features also requires a European Standard EN-14614 (CEN, 2004) requires an
sustainable management of associated processes by modifi- assessment of channel, river banks, riparian zone, and flood-
cations in management policy and related legal recom- plain features, and is an open guidance on which a detailed
mendations. A relevant example is provided by the application methodology for hydromorphological assessment, incorpor-
of the Erodible Corridor Concept (Piégay et al., 2005), which ating consideration of ongoing geomorphic processes, can be
is now a recommended procedure within the French legisla- developed. However, the methodologies used up to now in
tion that incorporates well-defined constraints for bank pro- most European countries tend to reflect river habitat survey
tection and mining authorization policies. Application of procedures. Most of them were not originally developed to fit
the ECC concept is increasing in other European countries with the WFD requirements and provide poor consideration
(e.g., Poland and Italy, see Nieznański et al. (2008) and of processes and channel adjustment trends, because the main
Rinaldi et al. (2009), respectively). focus of a river habitat survey is on the presence and quantity
At the conclusion of the IVth ECRR International Confer- of channel physical features. However, an increasing con-
ence on River Restoration 2008, the following points were sideration of processes rather than forms has been observed in
identified for future developments (Rinaldi et al., 2008a): (1) recent years, with some examples of new developing methods
the importance of a clear spatial and temporal context for in France (Wiederkehr et al., 2010 or SYRAH procedure by
restoration strategies; (2) the need for a range of approaches Chandesris et al., 2008), Spain (Ollero et al., 2007), and Italy
and to involve interdisciplinary groups; (3) a critical need to (Rinaldi et al., 2010).
48 River Processes and Implications for Fluvial Ecogeomorphology: A European Perspective

Interdisciplinary research is still needed in the areas of Camporeale, C., Perona, P., Porporato, A., Ridolfi, L., 2007. Hierarchy of models for
ecology and geomorphology (i.e., ecogeomorphology) to ap- meandering rivers and related morphodynamic processes. Reviews of
Geophysics 45, 1–28. http://dx.doi.org/10.1029/2005RG000185.
propriately implement hydromorphological assessment and
Chandesris, A., Mengin, N., Malavoi, J.R., Souchon, Y., Pella, H., Wasson, J.G.,
monitoring procedures within the scope of the WFD. The 2008. Systeme Relationnel d’Audit de l’Hydromorphologie des Cours d’Eau.
most important contribution needed is interpretation of the Principes et Methodes, v3.1. Cemagref, Lyon, 81 pp.
contribution of physical processes and forms to biodiversity, Christensen, O.B., Christensen, J.H., 2004. Intensification of extreme European
and the hydromorphological component of the ecological summer precipitation in a warmer climate. Global and Planetary Change 44,
107–117.
quality in rivers. Clarke, S.J., Bruc-Burgess, L., Wharton, G., 2003. Linking form and function:
towards an eco-hydromorphic approach to sustainable river restoration. Aquatic
Conservation: Marine and Freshwater Ecosystems 13, 439–450.
Corenblit, D, Tabacchi, E, Steiger, J, Gurnell, A.M., 2007. Reciprocal interactions
References and adjustments between fluvial landforms and vegetation dynamics in river
corridors: a review of complementary approaches. Earth-Science Reviews 84,
Acuna, V., Tockner, K., 2009. Surface–subsurface water exchange rates along 56–86.
alluvial river reaches control the thermal patterns in an Alpine river network. Coulthard, T., Van De Wiel, M.V.D., 2006. A cellular model of river meandering.
Freshwater Biology 54(2), 306–320. Earth Surface Processes and Landforms 31(1), 123–132.
Ashmore, P.E., 1991. How do gravel-bed rivers braid? Canadian Journal of Earth Coulthard, T.J., Hicks, D.M., Van De Wiel, M.J., 2007. Cellular modelling of river
Sciences 28, 326–341. catchments and reaches: advantages, limitations and prospects. Geomorphology
Ashmore, P.E., 2001. Braiding phenomena: statics and kinetics. In: Mosley, M.P. 90, 192–207.
(Ed.), Gravel-Bed Rivers V. New Zealand Hydrologic Society, pp. 95–121. Coulthard, T.J., Lewin, J., Macklin, M.G., 2008. Non-stationarity of basin scale
Bertoldi, W., Gurnell, A., Surian, N., Tockner, K., Zanoni, L., Ziliani, L., Zolezzi, G., sediment delivery in response to climate change. In: Habersack, H., Piégay, H.,
2009a. Understanding reference processes: linkages between river flows, Rinaldi, M. (Eds.), Gravel-Bed Rivers VI: From Process Understanding to River
sediment dynamics and vegetated landforms along the Tagliamento River, Italy. Restoration. Developments in Earth Surface Processes 11. Elsevier, Amsterdam,
River Research and Applications 25(5), 501–516. pp. 315–335.
Bertoldi, W., Tubino, M., 2007. River bifurcations: experimental observations on Crosato, A., 2008. Analysis and modelling of river meandering. Ph.D. thesis, Delft
University of Technology.
equilibrium configurations. Water Resources Research 43, W10437.
Diplas, P., Shaheen, H., 2008. Bed load transport and streambed structure in gravel
http://dx.doi.org/10.1029/2007WR005907.
streams. In: Habersack, H., Piégay, H., Rinaldi, M. (Eds.), Gravel-Bed Rivers VI:
Bertoldi, W., Zanoni, L., Tubino, M., 2009b. Planform dynamics of braided streams.
From Process Understanding to River Restoration. Developments in Earth
Earth Surface Processes and Landforms 34(4), 547–557.
Surface Processes 11. Elsevier, Amsterdam, pp. 291–312.
Best, J.L., 1988. Sediment transport and bed morphology at river channel
Downs, P.W., Simon, A., 2001. Fluvial geomorphological analysis of the recruitment
confluences. Sedimentology 35, 481–498. of large woody debris in the Yalobusha River network, central Mississippi, USA.
Billi, P., Rinaldi, M., Simon, A., 1997. Disturbance and adjustment of the Arno Geomorphology 37, 65–91.
River, Central Italy. I: historical perspective, the last 2000 years. In: Wang, Dufour, S., Piégay, H., 2008. Geomorphological controls of Fraxinus excelsior
S.S.Y., Langendoen, E.J., Shields, Jr. F.D. (Eds.), Management of Landscapes growth and regeneration in floodplain forests. Ecology 89, 205–215.
Disturbed by Channel Incision, Stabilization, Rehabilitation, Restoration. Center Dufour, S., Piégay, H., 2009. From the myth of a lost paradise to targeted river
for the Computational Hydroscience and Engineering. University of Mississippi, restoration: forget natural references and focus on human benefits. River
Oxford, MS, pp. 595–600. Research and Applications 25, 568–581.
Blanckaert, K., de Vriend, H., 2005. Turbulence characteristics in sharp open Enggrob, H., Tjerry, S., 1999. Simulation of morphological characteristics of a
channel bends. Physics of Fluids 17(5), 055102. http://dx.doi.org/10.1063/ braided river. Proceedings of RCEM Symposium. Genova, Italy, 6–11 September
1.1886726. 1999. Arti grafiche Lux, Genova, pp. 585–594.
Bolla Pittaluga, M., Repetto, R., Tubino, M., 2003. Channel bifurcation in braided European Commission, 2000. Directive 2000/60 EC of the European Parliament
rivers: equilibrium configurations and stability. Water Resources Research 39(3), and of the Council of 23 October 2000 establishing a framework for
1046–1059. Community action in the field of water policy. Official Journal L 327,
Bravard, J.P., 1989. La métamorphose des rivières des Alpes franc- aises à la fin du 22/12/2000, 73 pp.
Moyen-âge et à l’époque moderne. Bulletin de la Société Géographique de European Commission, 2005. Climate change and the European water dimension.
Liège, 25. EU Report No. 21553, European Commission – Joint Research Centre, Ispra,
Bravard, J.P., Amoros, C., Pautou, G., et al., 1997. River incision in south-east Italy.
France: morphological phenomena and ecological effects. Regulated Rivers: Florsheim, J.L., Mount, J.F., Chin, A., 2008. Bank erosion as a desirable attribute
Research and Management 13, 75–90. of rivers. BioScience 58(6), 519–529. doi:10.1641/B580608.
Bravard, J.P., Kondolf, G.M., Piégay, H., 1999. Environmental and societal effects of Fonstad, M., Marcus, W.A., 2003. Self-organised criticality in riverbank systems.
channel incision and remedial strategies. In: Darby, S.E., Simon, A. (Eds.), Annals of Association of American Geographers 93(2), 281–296.
Incised River Channels: Processes, Forms, Engineering and Management. Wiley, Foufoula-Georgiou, E., Sapozhnikov, V., 2001. Scale invariances in the morphology
and evolution of braided rivers. Mathematical Geology 33(3), 273–291.
Chichester, pp. 303–341.
Francis, R.A., Corenblit, D., Edwards, P.J., 2009. Perspectives on
Brierley, G.J., Brooks, A.P., Fryirs, K., Taylor, M.P., 2005. Did humid-temperate
biogeomorphology, ecosystem engineering and self-organisation in island-
rivers in the Old and New Worlds respond differently to clearance of riparian
braided fluvial ecosystems. Aquatic Sciences 71, 290–304.
vegetation and removal of woody debris? Progress in Physical Geography 29,
Frascati, A., Lanzoni, S., 2009. Morphodynamic regime and long-term evolution of
27–49. meandering rivers. Journal of Geophysical Research 114, F02002.
Brookes, A., 1990. Restoration and enhancement of engineered river channels:
http://dx.doi.org/10.1029/2008JF001101.
some European experiences. Regulated Rivers: Research and Management 6, Garcia-Ruiz, J.M., White, S.M., Lasanta, T., et al., 1997. Assessing the effects of
45–56. land-use changes on sediment yield and channel dynamics in the central
Brown, A.G., 1997. Biogeomorphology and diversity in multiple-channel river Spanish Pyrenees. In: Walling, D.E., Prost, J.L. (Eds.), Human Impact on
systems. Global Ecology and Biogeography Letters 6, 179–185. Erosion and Sedimentation. Proceedings of Rabat Symposium S6. Institute of
Brown, A.G., Keough, M., 1991. Palaeochannels, palaeolandsurfaces and the three- Hydrology Publication No. 245. IAHS Press, Wallingford, pp. 151–158.
dimensional reconstruction of floodplain environmental change. In: Carling, P.A., Gerhard, M., Reich, M., 2001. Totholz in Fliessgewässern – Empfehlungen zur
Petts, G.E. (Eds.), Lowland Floodplain Rivers: Geomorphological Perspectives. Gewässerentwicklung. Gemeinnützige Fortbildungsgesellschaft für
Wiley, Chichester, pp. 185–202. Wasserwirtschaft und Landschaftsentwicklung, Mainz.
Camporeale, C., Perona, P., Porporato, A., Ridolfi, L., 2005. On the long term Gradziński, R., Baryła, J., Doktor, M., et al., 2003. Vegetation-controlled modern
behavior of meandering rivers. Water Resources Research 41, W12043. anastomosing system of the upper Narew River (NE Poland) and its sediments.
http://dx.doi.org/10.1029/2005WR004109. Sedimentary Geology 157, 253–276.
 River Processes and Implications for Fluvial Ecogeomorphology: A European Perspective 49

Gregory, K.J., Davis, R.J., 1992. Coarse woody debris in stream channels in Lach, J., Wyżga, B., 2002. Channel incision and flow increase of the upper Wisłoka
relation to river channel management in woodland areas. Regulated Rivers: River, southern Poland, subsequent to the reafforestation of its catchment. Earth
Research and Management 7, 117–136. Surface Processes and Landforms 27, 445–462.
Gumiero, B., Rinaldi, M., Fokkens, B. (Eds.), 2008. Proceedings of the IVth ECRR Łajczak, A., 1995. The impact of river regulation, 1850–1990, on the channel and
International Conference on River Restoration 2008. 16–22 June 2008, ECRR, floodplain of the Upper Vistula River, Southern Poland. In: Hickin, E.J. (Ed.),
Venice, 1058 pp. River Geomorphology. Wiley, Chichester, pp. 209–233.
Gurnell, A.M., Edwards, P.J., Petts, G.E., Ward, J.V., 1999. A conceptual model for Lane, S.N., Westaway, R.M., Hicks, D.M., 2003. Estimation of erosion and
alpine proglacial river channel evolution under changing climatic conditions. deposition volume in a large, gravel-bed, braided river using synoptic remote
Catena 38, 223–242. sensing. Earth Surface Processes and Landforms 28, 249–271.
Gurnell, A.M., Morrissey, I.P., Boitsidis, A.J., et al., 2006. Initial adjustments within a Lawler, D.M., 1992. Process dominance in bank erosion systems. In: Carling, P.A.,
New River Channel: interactions between fluvial processes, colonising vegetation Petts, G.E. (Eds.), Lowland Floodplain Rivers: Geomorphological Perspectives.
and bank profile development. Environmental Management 38, 580–596. Wiley, Chichester, pp. 117–143.
Gurnell, A.M., Surian, N., Zanoni, L., 2009. Multi-thread river channels: a Liébault, F., Gomez, B., Page, M., et al., 2005. Land-use change, sediment
perspective on changing European alpine river systems. Aquatic Sciences 71, production and channel response in upland regions. River Research and
253–265. Applications 21, 739–756.
Gurnell, A.M., Tockner, K., Edwards, P.J., Petts, G.E., 2005. Effects of deposited Liébault, F., Piégay, H., 2001. Assessment of channel changes due to long-term
wood on biocomplexity of river corridors. Frontiers in Ecology and Environment bedload supply decrease, Roubion River, France. Geomorphology 36, 167–186.
3, 377–382. Liébault, F., Piégay, H., 2002. Causes of 20th century channel narrowing in
Habersack, H., Piégay, H., 2008. River restoration in the Alps and their mountain and piedmont rivers of southeastern France. Earth Surface Processes
surroundings: past experience and future challenges. In: Habersack, H., Piégay, and Landforms 27, 425–444.
H., Rinaldi, M. (Eds.), Gravel-Bed Rivers VI: From Process Understanding to Luchi, R., Zolezzi, G., Tubino, M., 2009. Modelling mid-channel bars in meandering
River Restoration. Developments in Earth Surface Processes 11. Elsevier, channels. Earth Surface Processes and Landforms 35, 902–917.
Amsterdam, pp. 703–737. Makaske, B., 2001. Anastomosing rivers: a review of their classification, origin and
Harwood, K., Brown, A.G., 1993. Fluvial processes in a forested anastomosing sedimentary products. Earth-Science Review 53, 149–196.
river: flood partitioning and changing flow patterns. Earth Surface Processes and Muhar, S., Jungwirth, M., Unfer, G., et al., 2008. Restoring riverine landscapes at
Landforms 18, 741–748. the Drau River: successes and deficits in the context of ecological integrity.
Hassan, M.A., Church, M., Rempel, J., Enkin, R.J., 2009. Promise, performance and In: Habersack, H., Piégay, H., Rinaldi, M. (Eds.), Gravel-Bed Rivers VI: From
current limitations of a magnetic Bedload Movement Detector. Earth Surface Process Understanding to River Restoration. Developments in Earth Surface
Processes and Landforms 34, 1022–1032. Processes 11. Elsevier, Amsterdam, pp. 779–807.
Hassan, M.A., Smith, B.J., Hogan, D.L., Luzi, D.S., Zimmermann, A.E., Eaton, B.C., Murray, B., Paola, C., 1994. A cellular model of braided rivers. Nature 371,
2008. Sediment storage and transport in coarse bed streams: scale 54–57.
considerations. In: Habersack, H., Piégay, H., Rinaldi, M. (Eds.), Gravel-Bed Nearing, M.A., Jetten, V., Baffaut, C., et al., 2005. Modeling response of soil
Rivers VI: From Process Understanding to River Restoration. Developments in erosion and runoff to changes in precipitation and cover. Catena 61, 131–154.
Earth Surface Processes 11. Elsevier, Amsterdam, pp. 473–497. Newson, M.D., Large, A.R.G., 2006. Natural rivers, hydromorphological quality and
Higler, L.W.G., 1993. The riparian community of north-west European lowland river restoration: a challenging new agenda for applied fluvial geomorphology.
streams. Freshwater Biology 29, 229–241. Earth Surface Processes and Landforms 31, 1606–1624.
Hooke, J., 2004. Cutoffs galore!: occurrence and causes of multiple cutoffs on a Nieznański, P., Wyżga, B., Obrdlik, P., 2008. Oder border meanders: a concept of
meandering river. Geomorphology 61, 225–238. the erodible river corridor and its implementation. In: Gumiero, B., Rinaldi, M.,
Hooke, J.M., 2006. Human impacts on fluvial systems in the Mediterranean region. Fokkens, B. (Eds.), 4th ECRR International Conference on River Restoration.
Geomorphology 79, 311–335. ECRR, Venice, pp. 479–486.
Hooke, J.M., 2008. Temporal variations in fluvial processes on an active Ollero, O.A., Balları́n, F.D., Dı́az, B.E., et al., 2007. Un indice hydrogeomorfologico
meandering river over a 20-year period. Geomorphology 100, 3–13. (IHG) para la evaluacion del estado ecologico de sistemas fluviales.
Hornich, R., Baumann, N., 2008. River restoration of the River Mur along the Geographicalia 52, 113–141.
border between Austria and Slovenia. In: Gumiero, B., Rinaldi, M., Fokkens, B. Palmer, M.A., Bernhardt, E.S., Allan, J.D., et al., 2005. Standards for ecologically
(Eds.), 4th ECRR International Conference on River Restoration. ECRR, Venice, successful river restoration. Journal of Applied Ecology 42, 208–217.
pp. 487–496. Parsons, D.R., Best, J.L., Lane, S.N., Orfeo, O., Hardy, R.J., Kostaschuk, R., 2007.
Hupp, C.R., Bornette, G., 2003. Vegetation as a tool in the interpretation of fluvial Form roughness and the absence of secondary flow in a large
geomorphic processes and landforms in humid temperate areas. In: Kondolf, confluence–diffluence, Rio Parana, Argentina. Earth Surface Processes and
G.M., Piégay, H. (Eds.), Tools in Fluvial Geomorphology. Wiley, Chichester, Landforms 32, 155–162.
pp. 269–288. Perona, P., Molnar, P., Savina, M., Burlando, P., 2009. An observation-based
Hupp, C.R., Rinaldi, M., 2007. Riparian vegetation patterns in relation to fluvial stochastic model for sediment and vegetation dynamics in the floodplain of an
landforms and channel evolution along selected rivers of Tuscany (Central Italy). Alpine braided river. Water Resources Research 45, W09418.
Annals of the Association of American Geographers 97(1), 12–30. Perucca, E., Camporeale, C., Ridolfi, L., 2006. Influence of river meandering
Jagers, H.R.A., 2003. Modelling planform changes of braided rivers. Ph.D. thesis, dynamics on riparian vegetation pattern formation. Journal of Geophysical
University of Twente, The Netherlands. Research 111, G01001. http://dx.doi.org/10.1029/2005JG000073.
Jäggi, M., Zarn, B., 1999. Stream channel restoration and erosion control for Perucca, E., Camporeale, C., Ridolfi, L., 2007. Significance of the riparian
incised channels in alpine environments. In: Darby, S., Simon, A. (Eds.), Incised vegetation dynamics on meandering river morphodynamics. Water Resources
River Channels. Wiley, Chichester, pp. 343–370. Research 43, W03430. http://dx.doi.org/10.1029/2006WR005234.
Junk, J.W., Bayley, P.B., Sparks, R.E., 1989. The flood pulse concept in river flood Petts, G.E., Möller, H., Roux, A.L. (Eds.), 1989. Historical Change of Large Alluvial
plain systems. Canadian Special Publications of Fisheries and Aquatic Sciences Rivers: Western Europe. Wiley, Chichester.
106, 110–121. Piégay, H., Darby, S.E., Mosselman, E., Surian, N., 2005. A review of techniques
Keesstra, S.D., van Huissteden, J., Vanderberghe, J., Van Dam, O., de Gier, J., available for delimiting the erodible river corridor: a sustainable approach to
Pleizer, I.D., 2005. Evolution of the morphology of the river Dragonja (SW managing bank erosion. River Research and Applications 21, 773–789.
Slovenia) due to land-use changes. Geomorphology 69, 191–207. Pizzuto, J., 2009. An empirical model of event scale cohesive bank profile
Kleinhans, M.G., Jagers, H.R.A., Mosselman, E., Sloff, C.J., 2008. Bifurcation evolution. Earth Surface Processes and Landforms 34, 1234–1244.
dynamics and avulsion duration in meandering rivers by one-dimensional Pollen, N., 2006. Temporal and spatial variability of root reinforcement of
and three-dimensional models. Water Resources Research 44, W08454. streambanks: accounting for soil shear strength and moisture. Catena 69,
http://dx.doi.org/10.1029/2007WR005912. 197–205. http://dx.doi.org/10.1016/j.catena.2006.05.004.
Knighton, D., Nanson, G.C., 1993. Anastomosis and the continuum of channel Pringle, C.M., Naiman, R.J., Bretschko, G., et al., 1988. Patch dynamics in lotic
pattern. Earth Surface Processes and Landforms 18, 613–625. systems: the stream as a mosaic. Journal of the North American Benthological
Kondolf, G.M., Piégay, H., Landon, N., 2002. Channel response to increased and Society 7, 503–524.
decreased bedload supply from land use change: contrasts between two Reid, I., Laronne, J.B., Powell, D.M., 1999. Impact of major climatic change on
catchments. Geomorphology 45, 35–51. coarse-grained river sedimentation: a speculative assessment based on
50 River Processes and Implications for Fluvial Ecogeomorphology: A European Perspective

measured flux. In: Brown, A.G., Quine, T.A. (Eds.), Fluvial Processes and results of post-project monitoring. Regulated Rivers: Research and Management
Environmental Change.. Wiley, Chichester, pp. 105–115. 17, 473–491.
Rhoads, B.L., Sukhodolov, A., 2004. Spatial and temporal structure of shear-layer Stanford, A., Ward, J.V., 1988. The hyporheic habitat of river ecosystems. Nature
turbulence at a stream confluence. Water Resources Research 40, W06304. 335, 64–66.
Rice, S., Roy, A.G., Rhoads, B.L. (Eds.), 2008. Tributaries and the Fluvial Network. Stanford, J.A., Lorang, M.S., Hauer, F.R., 2005. The shifting habitat mosaic of river
Wiley, Hoboken, NJ. ecosystems. Verhandlungen Internationale Vereinigung für Theoretische und
Rickenmann, D., McArdell, B.W., 2008. Calibration of piezoelectric bedload impact Angewandte Limnologie 29, 123–136.
sensors in the Pitzbach mountain stream. Geodinamica Acta 21, 35–52. Starkel, L., 1995. Changes of river channels in Europe during the Holocene.
Rinaldi, M., 2003. Recent channel adjustments in alluvial rivers of Tuscany. Central In: Gurnell, A.M., Petts, G.E. (Eds.), Changing River Channels. Wiley, Chichester,
Italy. Earth Surface Processes and Landforms 28, 587–608. pp. 27–42.
Rinaldi, M., Darby, S.E., 2008. Modelling river-bank-erosion processes and mass Stolum, H., 1996. River meandering as a self-organization process. Science 271,
failure mechanisms: progress towards fully coupled simulations. In: Habersack, 1710–1713.
H., Piégay, H., Rinaldi, M. (Eds.), Gravel-Bed Rivers 6: From Process Surian, N., Rinaldi, M., 2003. Morphological response to river engineering and
Understanding to River Restoration. Series Developments in Earth Surface management in alluvial channels in Italy. Geomorphology 50, 307–326.
Processes 11. Elsevier, Amsterdam, pp. 213–239. Surian, N., Rinaldi, M., Pellegrini, L., et al., 2009a. Channel adjustments in Italian
Rinaldi, M., Grant, G., Kondolf, G.M., Piégay, H., 2008a. Restoration and rivers: evolutionary trends, causes and management implications. In: James,
management of physical processes and sediments. Introduction. In: Gumiero, B., L.A., Rathburn, S.L., Whittecar, G.R. (Eds.), Management and Restoration of
Rinaldi, M., Fokkens B. (Eds.), Proceedings of the IVth ECRR International Fluvial Systems with Broad Historical Changes and Human Impacts. Geological
Conference on River Restoration 2008. 16–22 June 2008. ECRR, Venice, Society of America Special Paper 451. Geological Society of America, Boulder,
pp. 377–380. CO, pp. 83–95.
Rinaldi, M., Mengoni, B., Luppi, L., Darby, S.E., Mosselman, E., 2008b. Numerical Surian, N., Ziliani, L., Comiti, F., Lenzi, M.A., Mao, L., 2009b. Channel adjustments
simulation of hydrodynamics and bank erosion in a river bend. Water Resources and alteration of sediment fluxes in gravel-bed rivers of north-eastern Italy:
Research 44, W09429. http://dx.doi.org/10.1029/2008WR007008. potentials and limitations for channel recovery. River Research and Applications
Rinaldi, M., Simon, A., 1998. Bed-level adjustments in the Arno River. Central Italy. 25, 551–567.
Geomorphology 22, 57–71. Thomas, R., Nicholas, A.P., 2002. Simulation of braided river flow using a new
Rinaldi, M., Simoncini, C., Piégay, H., 2009. Scientific strategy design for cellular routing scheme. Geomorphology 43, 179–195.
promoting a sustainable sediment management: the case of the Magra River Tockner, K., Malard, F., Ward, J.V., 2000. An extension of the flood pulse concept.
(Central-Northern Italy). River Research and Applications 25, 607–625. Hydrological Processes 14, 2861–2883.
http://dx.doi.org/10/1002/rra.1243. Van Wiel, M.J., Darby, S.E., 2007. A new model to analyse the impact of
Rinaldi, M., Surian, N., Comiti, F., Bussettini, M., 2010. The morphological woody riparian vegetation on the geotechnical stability of riverbanks.
quality index (IQM) for stream evaluation and hydromorphological classification. Earth Surface Processes and Landforms 32, 2185–2198. http://dx.doi.org/
Italian Journal of Engineering Geology and Environment, Special Issue 1, doi: 10.1002/esp.1522.
10.4408/IJEGE.2011-01.O-02. Van der Nat, D., Tockner, K., Edwards, P.J., Ward, J.V., Gurnell, A.M., 2003. Habitat
Rinaldi, M., Wyżga, B., Surian, N., 2005. Sediment mining in alluvial channels: change in braided flood plains (Tagliamento, NE-Italy). Freshwater Biology 48,
physical effects and management perspectives. River Research and Application 1799–1812.
21, 805–828. Vannote, R.L., Minshall, G.W., Cummins, K.W., Sedell, J.R., Cushing, C.E., 1980.
Rollet, A.J., Piégay, H., Bornette, G., Persat, H., 2008. Sediment dynamics, channel The river continuum concept. Canadian Journal of Fisheries and Aquatic
morphology and ecological restoration downstream a dam: the case of the Ain Sciences 37, 130–137.
River. In: Gumiero, B., Rinaldi, M., Fokkens, B. (Eds.), 4th ECRR International Vaughan, I.P., Diamond, M., Gurnell, A.M., et al., 2009. Integrating ecology with
Conference on River Restoration, pp. 497–504. ECRR. hydromorphology: a priority for river science and management. Aquatic
Rovira, A., Batalla, R.J., Sala, M., 2005. Response of a river sediment budget after Conservation: Marine and Freshwater Ecosystems 19, 113–125.
historical gravel mining (the lower Tordera, NE Spain). River Research and Ward, J.V., Stanford, J.A., 1983. The intermediate disturbance hypothesis: an
Applications 21, 829–847. explanation for biotic diversity patterns in lotic ecosystems. In: Fontaine, T.D.,
Rumsby, B.T., Macklin, M.G., 1996. River response to the last neoglacial (the ‘Little Bartell, S.M. (Eds.), Dynamics of Lotic Ecosystems. Ann Arbor Science
Ice Age’) in northern, western and central Europe. In: Branson, J., Brown, A.G., Publishers, Ann Arbor, MI, pp. 347–356.
Gregory, K.J. (Eds.), Global Continental Changes: The Context of Ward, J.V., Tockner, K., Schiemer, F., 1999. Biodiversity of floodplain river
Palaeohydrology. Geological Society Special Publication 115. Geological Society, ecosystems: ecotones and connectivity. Regulated Rivers: Research and
London, pp. 217–233. Management 15, 125–139.
Ruther, N., Olsen, N., 2007. Modelling free-forming meander evolution in a Westaway, R.M., Lane, S.N., Hicks, D.M., 2003. Remote survey of large-scale
laboratory channel using three-dimensional computational fluid dynamics. braided, gravel-bed rivers using digital photogrammetry and image analysis.
Geomorphology 89, 308–319. http://dx.doi.org/10.1016/j.geomorph.2006. International Journal of Remote Sensing 24, 795–815.
12.009. Wiederkehr, E., Dufour, S., Piégay, H., 2010. Location and semi-automatic
Samadi, A., Amiri-Tokaldany, E., Darby, S.E., 2009. Identifying the effects of characterisation of potential fluvial geomorphosites. Examples of applications
parameter uncertainty on the reliability of riverbank stability modelling. from geomatic tools in the Drôme River basin (France). Géomorphologie: relief,
Geomorphology 106, 219–230. processus, environnement 2, 175–188.
Schropp, M.H.I., 1995. Principles of designing secondary channels along the River Williams, M., 2000. Dark ages and dark areas: global deforestation in the deep
Rhine for the benefit of ecological restoration. Water Science and Technology past. Journal of Historical Geography 26, 28–46.
31, 379–382. Winterbottom, S.J., 2000. Medium and short-term channel planform changes on the
Schropp, M.H.I., Bakker, C., 1998. Secondary channels as a basis for the ecological Rivers Tay and Tummel, Scotland. Geomorphology 34, 195–208.
rehabilitation of Dutch rivers. Aquatic Conservation: Marine and Freshwater Wohl, E., Angermeier, P.L., Bledsoe, B., et al., 2005. River Restoration. Water
Ecosystems 8, 53–59. Resources Research 41, W10301. http://dx.doi.org/10.1029/2005WR003985.
Seminara, G., 2006. Meanders. Journal of Fluid Mechanics 554, 271–297. Wyżga, B., 1993a. Present-day changes in the hydrologic regime of the Raba River
http://dx.doi.org/10.1017/S0022112006008925. (Carpathians, Poland) as inferred from facies pattern and channel geometry.
Simon, A., Collison, A.J., 2002. Quantifying the mechanical and hydrological effects In: Marzo, M., Puigdefábregas, C. (Eds.), Alluvial Sedimentation. International
of vegetation on streambank stability. Earth Surface Processes and Landforms Association of Sedimentologists Special Publication 17. Blackwell, Oxford,
27, 527–546. pp. 305–316.
Simon, A., Pollen, N., Langendoen, E., 2006. Influence of two woody riparian Wyżga, B., 1993b. River response to channel regulation: case study of the Raba
species on critical conditions for streambank stability: Upper Truckee River, Carpathians, Poland. Earth Surface Processes and Landforms 18,
River, California. Journal of the American Water Resources Association 42, 541–556.
99–113. Wyżga, B., 2001a. A geomorphologist’s criticism of the engineering approach to
Simons, J.H.E.J., Bakker, C., Schropp, M.H.I., Jans, L.H., Kok, F.R., Grift, R.E., channelization of gravel-bed rivers: case study of the Raba River, Polish
2001. Man-made secondary channels along the river Rhine (The Netherlands): Carpathians. Environmental Management 28, 341–358.
 River Processes and Implications for Fluvial Ecogeomorphology: A European Perspective 51

Wyżga, B., 2001b. Impact of the channelization-induced incision of the Skawa and Zolezzi, G., Bertoldi, W., Tubino, M., 2006. Morphological analysis and prediction
Wisłoka Rivers, southern Poland, on the conditions of overbank deposition. of channel bifurcations. In: Sambrook-Smith, G., Best, J.L., Bristow, C.S., Petts,
Regulated Rivers: Research and Management 17, 85–100. G.E. (Eds.), Braided Rivers: Process, Deposits, Ecology and Management.
Wyżga, B., 2008. A review on channel incision in the Polish Carpathian rivers Special Publication of International Association of Sedimentologists 36.
during the 20th century. In: Habersack, H., Piégay, H., Rinaldi, M. (Eds.), Blackwell, Oxford, pp. 227–250.
Gravel-Bed Rivers VI: From Process Understanding to River Restoration. Zolezzi, G., Seminara, G., 2001. Downstream and upstream influence in river
Developments in Earth Surface Processes 11. Elsevier, Amsterdam, meandering. Part 1. General theory and application to overdeepening. Journal of
pp. 525–555. Fluid Mechanics 438, 183–211.
Zawiejska, J., Wyżga, B., 2010. Twentieth-century channel change on the Dunajec
River, southern Poland: patterns, causes and controls. Geomorphology,
doi:10.1016/j.geomorph.2009.01.014.

Biographical Sketch

Professor Massimo Rinaldi obtained a BSc and PhD in engineering geology at the University of Florence, Italy. He
is currently a professor of engineering geology in the Department of Civil and Environmental Engineering,
University of Florence. His main research topics are channel adjustments, modeling riverbank processes,
hydromorphology, and their applications to river management and restoration.

Bart"omiej Wyżga obtained an MSc in geology, as well as a habilitation diploma in earth sciences from the
Jagiellonian University in Kraków and a PhD in geography from the Adam Mickiewicz University in Poznań,
Poland. He is currently an associate professor in the Institute of Nature Conservation, Polish Academy of Sciences.
His research fields encompass fluvial geomorphology, hydrology, and alluvial sedimentology, with a special focus
on human impacts on gravel-bed rivers as well as biogeomorphology and river restoration.

Simon Dufour obtained a BSc and PhD in geography from the University of Lyon, France. He is currently an
assistant professor of geography at Rennes University, where he specializes in research on fluvial landscapes,
especially on spatial pattern of riparian buffers at large scale, interactions between vegetation and hydro-geo-
morphic processes, fluvial landscape evolution, river and floodplain management and restoration, remote-sensing
uses, and ecosystem services.

Walter Bertoldi, a PhD in environmental engineering, the University of Trento, Italy, is currently a research
associate in physical geography at Queen Mary University of London. His research focuses on braided river
geomorphology, with a particular interest in understanding linkages between water, sediments, and vegetation.
52 River Processes and Implications for Fluvial Ecogeomorphology: A European Perspective

Professor Angela Gurnell obtained a BSc, PhD, and DSc in geography from the University of Exeter, UK. She is
currently a professor of physical geography in the Geography Department at Queen Mary, University of London,
where she specializes in research on hydrogeomorphology, biogeomorphology, and their applications to river
management.
 12.5 Riparian Vegetation and the Fluvial Environment: A Biogeographic
Perspective
J Bendix, Syracuse University, Syracuse, NY, USA
JC Stella, State University of New York College of Environmental Science and Forestry, Syracuse, NY, USA
r 2013 Elsevier Inc. All rights reserved.

12.5.1 Introduction 53
12.5.2 Early History: Pattern and Process in Riparian Zones 54
12.5.3 Influence of Hydrogeomorphology on Vegetation: Evolution from Descriptive to Quantitative Studies 55
12.5.4 Specific Mechanisms of Hydrogeomorphic Impact 56
12.5.4.1 Flood Energy 56
12.5.4.2 Sedimentation 56
12.5.4.3 Prolonged Inundation 57
12.5.4.4 Water-Table Depth and Dynamics 57
12.5.4.5 Soil Chemistry 58
12.5.4.6 Propagule Dispersal 58
12.5.5 Influence of Vegetation on Geomorphology 59
12.5.6 Feedbacks between Vegetation and Hydrogeomorphology 60
12.5.7 Patterns in Published Literature 62
12.5.7.1 The Sample and Coding 63
12.5.7.2 General Characteristics of the Sampled Literature 64
12.5.7.3 Differences among Biomes 65
12.5.7.4 Scale-Related Differences 66
12.5.7.5 Hydrogeomorphic Mechanisms in the Context of Scale and Biogeography 67
12.5.8 Patterns and Perceptions Revealed in the Literature 68
References 69

Abbreviations LWD Large woody debris

CHILD Channel–hillslope integrated landscape TN Total nitrogen
development TOC Total organic carbon

Abstract

In this chapter, we review the historical arc of research on biogeomorphic interactions between fluvial geomorphology and
riparian vegetation. We then report on an examination of the past 20 years of published research on this topic. Having
classified studies according to the key relationships they have identified, we map those relationships to seek spatial patterns
that emerge in terms of either physiographic environment or actual geographic location. We also consider the varied
patterns of causal interactions that emerge at different spatial scales.

12.5.1 Introduction understanding of the linked physical and biotic processes that
sustain them (Osterkamp and Hupp, 1984; Gregory et al.,
The recognition of riparian zones as biogeomorphic units in 1991; Naiman and Decamps, 1997; Naiman et al., 2005). For
the landscape is critical both for theoretical understanding and much of this time, emphasis was placed on the importance of
for conservation of these high-value habitats. Since the middle understanding hydrogeomorphic influences on riparian vege-
of the twentieth century, study of the systematic patterns of tation, because fluvial forces set the physical template for
vegetation communities in riparian zones has led to a greater ecological communities. Recently, there has been greater study
of the reciprocal influence of biological organisms and pro-
cesses on geomorphic processes, leading to a biogeomorphic
Bendix, J., Stella, J.C., 2013. Riparian vegetation and the fluvial
understanding of fluvial/riparian ecosystems (Corenblit et al.,
environment: a biogeographic perspective. In: Shroder, J. (Editor in Chief),
Butler, D.R., Hupp, C.R. (Eds.), Treatise on Geomorphology. Academic 2007). This view recognizes the substantial influence
Press, San Diego, CA, vol. 12, Ecogeomorphology, pp. 53–74. that vegetation pattern and structure can have in altering the

Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00322-5 53

54 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective

distribution of physical forces of flow and sediment regimes, conclusions about drivers, responses, and landscape patterns
which ultimately create feedbacks to biological communities in riparian vegetation communities? In which biomes and at
and ecosystems (Bendix and Hupp, 2000). what scales is the direction of influence primarily from the
Early work relating riparian ecology to fluvial processes was physical to the biotic, the reverse, or one of strongly linked
primarily descriptive in nature, with a common thread being feedbacks? How does the biome or geographic setting influ-
the classification of riparian vegetation communities and their ence the relative strength of ecosystem drivers, including dis-
association with particular fluvial landforms, or even simply turbance regimes (flooding, scour, and sediment deposition),
particular vertical locations relative to stream channels. These abiotic stress (e.g., seasonal drought), soil chemistry, and/or
studies were generally place- and taxa-specific. The develop- limitations on life-history processes such as propagule
ment of geomorphology as a more quantitative science since dispersal?
the middle of the last century (e.g., Leopold and Maddock, We believe that a biogeomorphic approach is necessary for
1953; Leopold et al., 1964) laid the basis for more generality a legitimate understanding of either fluvial processes or the
and a common biogeomorphic process approach to the study ecology of the riparian zone. The large volume of relevant
of rivers and riparian zones (Osterkamp and Hupp, 2010). empirical research since several landmark papers in the 1980s
Examples of this work include the hydraulic geometry ap- and early 1990s (e.g., Osterkamp and Hupp, 1984; Hupp and
proach (Leopold and Maddock, 1953), process-based classi- Osterkamp, 1985; Gregory et al., 1991) suggests the need to go
fication of river environments (Montgomery and Buffington, beyond a catalog-style review to a more analytical approach to
1997), and quantifications of interactions between sediment understanding the complex relationships between the eco-
transport capacity and supply across a wide range of spatial logical and geomorphic elements of the riparian environment.
and temporal scales (e.g., Howard et al., 1994). We therefore seek to examine the literature on the interactions
However, much of the early work was conducted on North between riparian plant communities and fluvial landforms/
American streams, with particular ecological communities and processes, within a bipartite, spatially oriented framework that
physical regimes influencing a great deal of the development emphasizes how biogeomorphic interactions between riparian
in early theories and empirical examples. The histories of both vegetation and fluvial geomorphology vary with both geog-
geomorphology and ecology include reminders of the danger raphy and the scale of observation.
of generalizing theories across dissimilar environments. Cri- In this chapter, we begin with a condensed review of the
tiques of the roles of William Morris Davis in the former historical arc of research on the varied fluvial impacts on
discipline and of Frederick C Clements in the latter have noted vegetation, the means whereby vegetation influences fluvial
that theoretical models developed in distinctive geographic geomorphology, and the feedbacks between the two. The bulk
settings did not translate appropriately when imposed else- of that research has been on the specific mechanisms of
where. In a recent example, studies by Walter and Merritts hydrogeomorphic influence on riparian plant communities;
(2008) in eastern North America suggest that the geomorphic therefore, these mechanisms receive particular emphasis in
context in the region where much of the early hydraulic our review. We then report on an examination of the past 20
geometry research (e.g., Leopold and Maddock, 1953) was years of published research on this topic. Having classified
conducted may have been in disequilibrium due to historical studies according to the key relationships they have identified,
human influences (i.e., mill dams). These well-known ex- we map both the causal directions and the mechanisms to
amples serve as reminders that processes and relationships seek spatial patterns that emerge in terms of either physio-
vary spatially. In particular, the very different environmental graphic environment or actual geographic location. We also
conditions and vegetation structure occurring in different consider the varied patterns of causal interactions that may
biomes suggest that plant adaptations to hydrogeomorphic emerge at different spatial scales.
constraints, and the subsequent influence of vegetation on
fluvial processes, may vary widely from one setting to another.
Another spatial complication arises when we consider the 12.5.2 Early History: Pattern and Process in
scale at which biogeomorphic relationships are observed. A Riparian Zones
variety of studies have indicated that the causal relationships
observed between riparian vegetation and hydrogeomorphic The current understanding of the physical/biological linkages in
processes may differ depending upon the scale of observation river ecosystems is inherited from a long line of conceptual and
(Baker, 1989; Dixon et al., 2002; Petty and Douglas, 2010). In empirical work in stream ecology and river science. The issues
some, but not all, instances, the relationships at smaller scales of biogeography and scale in studies of riparian ecology date to
may be hierarchically constrained by those operating at larger early work relating geomorphic landforms to vegetation com-
scales (Bendix, 1994a). Although there have, by now, been munities. For almost a century, scholars have acknowledged
several case studies in varied environments of the impact of and described associations between geomorphic landforms and
scale on riparian biogeomorphic environments, we lack a the distribution of vegetation communities in river and flood-
systematic review of the empirical findings regarding scale plain systems. More recently, researchers have investigated the
impacts – both from studies that explicitly explored scale and specific processes that drive these vegetation associations with
from those that did not but contain relevant information. landform. Early studies drew on the disciplinary traditions of
These issues raise the questions of how geography and geography, geomorphology, and ecology, establishing strands
scale influence our understanding of riparian zones and that have persisted through time. Beginning in the 1930s, a
their linked physical/biological processes. For example, how primary focus of this work was the description and classifi-
do the geographic setting and scale of observation affect our cation of riparian vegetation communities, but, even in these
 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective 55

early efforts, there were attempts to systematically relate vege- occur). Sigafoos identified distinctive vegetation bands
tation communities to fluvial landforms and processes. The along the Potomac River related to inundation frequency;
distinct zones of vegetation described by Hefley (1937) on the these observations set the stage for his groundbreaking work
low terraces of the Canadian River in Oklahoma were sub- reconstructing past floods from dendrogeomorphological
sequently attributed by Ware and Penfound (1949) to a com- records (Sigafoos, 1964). Although much of the research in
bination of flood damage and drought. this period was in the middle and eastern parts of North
Such studies were the precursors of later work that would America, valuable research emerged in the West as well. Everitt
use more rigorous statistical tools to document similar rela- (1968) related cottonwood regeneration on Utah’s Fremont
tionships (Hupp and Osterkamp, 1985; Harris, 1987; Shin River to flood dynamics. Teversham and Slaymaker (1976)
and Nakamura, 2005). Our understanding of process and studied riparian vegetation composition in Lillooet River val-
quantitative tools for testing this have evolved over time to the ley, British Columbia, documenting species groupings that
point that some researchers have developed and tested pre- were related to both sediment size and flood frequency. These
dictive models of vegetation response to hydrogeomorphic examples, by no means exhaustive of the studies during this
factors such as flood inundation (Auble et al., 1994; Dixon period, give an indication of the range of contemporary
and Turner, 2006), flood energy (Bendix, 1999; Sandercock perspectives.
and Hooke, 2010; Stromberg et al., 2010), ice scour (Auble and Beginning in the 1980s, interest in riparian ecology
Scott, 1998), channel meandering, and floodplain sedimen- broadened greatly, leading to classification systems of vege-
tation (Harper et al., 2011). Some of these models are phe- tation communities with landforms and quantification of the
nomenistic (e.g., Auble et al., 1994; Shafroth et al., 1998), physical processes that sustained them. Osterkamp and Hupp
whereas others are mechanistic in nature (Lytle and Merritt, (1984) (also see Hupp and Osterkamp, 1985) investigated
2004; Stella, 2005; Dixon and Turner, 2006; Harper et al., these relationships along northern Virginia streams and
2011). documented assemblages of woody plants on geomorphic
surfaces that are distinguished by inundation frequency and
duration. Similar studies using categorical classifications of
12.5.3 Influence of Hydrogeomorphology on landforms that are associated with different flood inundation
Vegetation: Evolution from Descriptive to frequencies include examples from Northern California
Quantitative Studies (Harris, 1987), Central Oregon (Kovalchik and Chitwood,
1990), and Japan (Shin and Nakamura, 2005). In this ap-
Hydrogeomorphic processes generally influence riparian proach, vegetation species presence and/or cover was recorded
vegetation through their contribution to the physical dis- and associated with elevation above the active channel, and by
turbance regime, which affects plant demography (e.g., dis- association with flood frequency and duration. Often, some
persal to safe sites and flood mortality), and through their type of multivariate analysis was used to classify species into
impacts on the resource environment for plant growth. These groups that sort along hydrologic gradients (Harris, 1987;
drivers interact with plant life-history processes, traits, and Auble et al., 1994; Bendix, 1994b; Rodrı́guez-González et al.,
physiological tolerances to influence population demo- 2010).
graphics and community dynamics in riparian communities. Beginning in the 1990s, researchers began to develop
As early as the 1930s, when plant zonation patterns on rivers quantitative models linking vegetation response to mech-
were first documented, researchers had recognized that anistic, hydrological, and hydraulic drivers. In an innovative
flooding regimes influence the composition, distribution, and approach combining standard numerical modeling in plant
structure of riparian vegetation. Illichevsky (1933) docu- ecology and river engineering, Auble et al. (1994) conducted a
mented distinct cross-sectional compositional belts in the plant community analysis of the Gunnison River (CO), using
floodplain of the Dnieper River, noting differences in plant cluster analysis in conjunction with hydraulic modeling to
assemblages between the inundated and dry zones. Examples quantify the flood duration of distinct plant communities and
of similar work in the years that followed include Shelford’s model vegetation change under proposed regulated flow re-
(1954) documentation of biotic communities in the lower gimes. In California’s Transverse Ranges, Bendix (1999) used
Mississippi valley floodplain and their age and elevation hydraulic modeling, ordination, and regression to relate plant
relative to the mean low water level, and Fanshawe’s (1954) community composition to computed values of unit stream
use of topographic position on bars and banks as a criterion power, as opposed to earlier studies that relied on assump-
for distinguishing plant communities on river fringes in Brit- tions of relative flood energy inferred from landform position.
ish Guiana. His research demonstrated that variation in hydraulic par-
Studies during this early period generally described pat- ameters imposes spatial variation in the ecological impact of
terns more than the processes that influenced them, although floods within watersheds (Bendix, 1997, 1998). In another
there are some notable exceptions (Hefley, 1937; Ware and application of hydraulic modeling, Dixon et al. (2002) inte-
Penfound, 1949), in particular, investigations relating vege- grated calculations of energy slope from a one-dimensional
tation to flood frequency and duration, and linking flood hydraulic model with quantitative landscape analysis to in-
disturbance to successional zonation (Wistendahl, 1958; vestigate the influence of physical characteristics at the local
Lindsey et al., 1961). Sigafoos (1961) took issue with succes- and landscape scales on the distribution of pioneer tree
sional interpretations, arguing that zonation instead reflected seedlings along the Wisconsin River. This was an extension of
tolerance to flooding (see Hupp (1988) for a useful dis- the approach WC Johnson used to infer maximum streamflow
cussion of where and why successional zonation might thresholds that cause seedling mortality from flood removal
56 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective

and ice scour on both the Missouri River (Johnson et al., 1976; vulnerable when plants are small relative to the magnitude of
Johnson, 1992) and the Platte River, Nebraska (Johnson, flood energy, such as during seedling establishment. This is
1994, 1998). also the case for plant removal by scour, which will generally
With the increasing emphasis on sophisticated quantitative occur when scour depth approaches the depth of coarse roots.
analyses, the mechanisms of plant recruitment and mortality Root or stem breakage not only is common for herbaceous
were studied in more detail, and species-level investigations as- vegetation during high flow but also can occur for woody
sumed greater prominence to complement studies conducted at plants in large floods (Chambers et al., 1991; Groeneveld and
the whole-community level (Shafroth et al., 1998; Stella, 2005; French, 1995; Gurnell et al., 2002). The energy required to
Dixon and Turner, 2006; Shafroth et al., 2010). Riparian species remove or break plants varies with their size and flexibility,
differ in key life-history traits that are linked to recruitment and their root characteristics, and the nature of the substrate in
survival in dynamic fluvial environments, including inundation which they are rooted (Bendix, 1999).
duration (Huffman, 1980; Toner and Keddy, 1997; Robertson At the vegetation patch scale, plant removal occurs when
et al., 2001; Kozlowski, 2002; van Eck et al., 2004), seed dis- scour depth exceeds a critical depth that is related to the size
persal and characteristics of fecundity, release timing, seed lon- and structure of the vegetation roots. The patchy nature of
gevity and buoyancy (Kubitzki and Ziburski, 1994; Danvind and riparian vegetation introduces complexities and feedbacks to
Nilsson, 1997; Lopez, 2001; Nilsson et al., 2002; Stella et al., flood energy because vegetation often reduces near-bed vel-
2006; Gurnell et al., 2008), and those related to survival of scour ocities and scour depths deep within a patch, but increases
and burial (Bornette et al., 2008), particularly hydraulic resist- velocities at the upstream and outside edges of vegetation
ance and stem flexibility, rooting depth, and resprouting ability patches through flow contraction. This can create wider and
(Brewer et al., 1998; Levine and Stromberg, 2001; Karrenberg deeper scour patterns than those produced around individual
et al., 2002; Lytle and Poff, 2004; Renofalt and Nilsson, 2008; stems and increased susceptibility to flood mortality for ex-
Rodrı́guez-González et al., 2010). Differences in reproductive posed plants (Lightbody et al., 2008; Yager and Schmeeckle,
and survival traits have been used successfully in several mech- 2007).
anistic, numerical approaches to predict relative abundance of The impacts of flood energy on the species composition of
species within riparian communities (Shafroth et al., 1998; riparian plant communities may be seen both through dif-
Stella, 2005; Dixon and Turner, 2006; Bornette et al., 2008). ferential survival of the potential damage described above and
through colonization by pioneers replacing the species that do
not survive (Bendix and Hupp, 2000). More subtly, even
12.5.4 Specific Mechanisms of Hydrogeomorphic floods that do not cause mortality may clear leaf litter, pro-
Impact viding an advantage for species that require bare mineral soil
for colonization (Yanosky, 1982).
Hydrogeomorphic processes interact with riparian vegetation
through life-history characteristics, plant traits, and physio-
12.5.4.2 Sedimentation
logical tolerances (Rood et al., 2003). There are many mech-
anisms that influence vegetation composition, distribution, Sedimentation can act on plants in either positive or negative
and density; however, for ease of analysis we have grouped ways, and includes effects of landform construction (Cooper
them into six main categories, all of which are consequences et al., 2003; Latterell et al., 2006), mortality due to burial
to some degree of periodic floods and the dynamic hydrology (Hack and Goodlett, 1960), and sediment texture controls on
characteristic of near-channel environments: water availability (McBride and Strahan, 1984a). At the
landscape scale, the sediment regime has a controlling influ-
1. flood energy, which exerts a limiting effect on vegetation
ence on plant establishment, community composition, bio-
distribution due to scour and stem breakage;
mass, and vegetation structure (Everitt, 1968; Hickin, 1984;
2. sedimentation, which can exert influence through creation
Bechtold and Naiman, 2006; Naiman et al., 2010). Channel
of new habitat for plant colonization and reduction of
migration processes drive formation and development of
competition, burial of existing vegetation, and/or sediment
geomorphic surfaces to control the establishment and spatial
texture effects on water availability;
extent of pioneer communities (McKenney et al., 1995; Lytle
3. prolonged inundation, which generally reduces physio-
and Merritt, 2004; Harper et al., 2011). Increasingly, studies
logical function and survival;
have focused on description and quantification of the relative
4. water-table depth and dynamics, which regulate soil
importance of different establishment pathways for riparian
moisture;
forest stands on alluvial surfaces, including point bars, aban-
5. soil chemistry influences, including mineral nutrition, sal-
doned channels, and mid-channel bars (Baker and Walford,
inity, and pollutants; and
1995; Cooper et al., 2003, 2006; Latterell et al., 2006; Van Pelt
6. fluvial controls on propagule dispersal.
et al., 2006; Stella et al., 2011). One of the primary influences
of new landforms on colonizing vegetation is through sedi-
ment texture effects on moisture-holding capacity, which in
12.5.4.1 Flood Energy turn controls plant survival and growth (Mahoney and Rood,
The hydraulic energy associated with flooding in dynamic 1992; Hughes et al., 2000; Hupp and Rinaldi, 2007).
river systems causes damage or mortality of plants through At smaller spatial scales, sedimentation affects plants
root zone scour and stem breakage (Hughes, 1997; Polzin and through mortality by burial. The severity of this effect depends
Rood, 2006; Perucca et al., 2007). Vegetation is particularly on the size of the plant relative to sediment deposition rate,
 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective 57

which is related to the transport capacity of individual flood Mitsch et al., 1991; Megonigal et al., 1997; Rodrı́guez-Gon-
events, basin-scale sediment supply (Buffington and Mont- zález et al., 2010), whereas others have found them to be
gomery, 1999), meso-scale influences such as bar topography positive (Burke et al., 1999; Clawson et al., 2001; Hanson
(Dietrich and Whiting, 1989; Francis and Gurnell, 2006), and et al., 2001). Some of these contradictions may reflect geo-
local roughness fields from live vegetation or woody debris graphic variation, as environments typified by prolonged
(Gurnell and Petts, 2006; Yager and Schmeeckle 2007). hydroperiod are likely to support species that show a positive
Many plant species can survive burial and resprout from response to inundation (Hupp, 2000).
epicormic buds (Bond and Midgley, 2001). However, the se-
verity of impact (i.e., plant mortality rate) depends on the
depth of sediment deposited during an event, as well as the
12.5.4.4 Water-Table Depth and Dynamics
plant size, season (dormant vs. active), and taxon-specific
physiology. Ewing (1996) found decreased growth and Although many riparian plant species are well adapted to their
physiological function by alder and sedge plants after partial dynamic river systems through traits such as abundant seed
sediment burial. Levine and Stromberg (2001) found in- production, wind dispersal, and fast growth (Karrenberg et al.,
creased resistance by cottonwood over tamarisk to sediment 2002; Lytle and Poff, 2004), there are generally life-history
burial, and other work along the Bill Williams River, AZ, has tradeoffs that include demand for abundant soil moisture and
documented substantially greater mortality of tamarisk com- intolerance to drought. In particular, survival of seedlings in
pared to willow seedlings in response to two dam-controlled arid and semi-arid climates is a challenge because seasonally
flood releases (Shafroth et al., 2010). Mortality occurred by fluctuating water tables and severe vapor pressure deficits can
both scour and burial and was likely nonproportional among dramatically reduce water availability during the critical es-
species because of the substantially greater first-year height tablishment stage (Donovan and Ehleringer, 1991; Horton
and diameter growth of willow relative to tamarisk (Sher et al., and Clark, 2001; Hughes et al., 2001; Rood et al., 2003). On
2002; Shafroth et al., 2010). snowmelt-dominated rivers, extended flow and groundwater
declines typically occur in late spring and early summer, as the
snowpack melts (Peterson et al., 2000). Seedlings and other
12.5.4.3 Prolonged Inundation
shallowly rooted plants are particularly vulnerable, and sea-
Although temporary flooding generally increases the supply of sonal water shortage can act with other stressors in the ripar-
water and nutrients to terrestrial plants (Burke et al., 1999; ian zone (e.g., scour, herbivory, and competition with
Clawson et al., 2001; Kozlowski and Pallardy, 2002), long- herbaceous species) to limit population dynamics among ri-
term soil saturation induces soil anoxia, which limits nutrient parian plants (Lytle and Merritt, 2004; Scott et al., 1996;
availability and gas exchange for plants (Mitsch and Gosse- Stromberg et al., 1991).
link, 2007), and soil toxicity due to reducing conditions. These In arid and semi-arid regions, the connection of stream to
negative effects are mitigated to some degree by plants’ riparian water table plays a critical role in resource supply and
adaptive responses to prolonged flooding, including tissues to plant survival (Zimmerman, 1969; Rood et al., 2003), and can
facilitate oxygen exchange such as aerenchyma and lenticel be the limiting factor for seedling survival and the population
development (Blom and Voesenek, 1996; Crawford, 1996; structure of pioneer plants (Lytle and Merritt, 2004). Long-
Rood et al., 2003; Walls et al., 2005). In riparian and other term alterations in flow magnitude, timing, and recession rate
ecosystems where stressful abiotic conditions may limit plants’ have exacerbated the effects of seasonal water limitation in
size, life span, and/or recruitment opportunities, increased these systems and threaten to further reduce the extent of ri-
sprouting has been associated with waterlogged soils parian woodlands (Fenner et al., 1985; Rood and Mahoney,
(Rodrı́guez-González et al., 2010), and may be an important 1990; Rood et al., 1999; Stella, 2005; Braatne et al., 2007); nor
strategy in particular for species that do not maintain a seed is the importance of water-table proximity limited to dry en-
bank (Bond and Midgley, 2001; Nzunda et al., 2007). vironments. On the floodplain of New Jersey’s Raritan River,
Much of the research on riparian plants’ physiological re- Frye and Quinn (1979) interpreted species distributions as a
sponses to inundation is based on laboratory experiments function of soil texture and depth to water table. In this set-
(e.g., Pereira and Kozlowski, 1977; Conner et al., 1997; Li ting, they considered a shallow water table to be a limiting
et al., 2005; Day et al., 2006), with a somewhat lesser em- factor, arguing that it excluded deep-rooted species. Numerous
phasis on field studies (Megonigal et al., 1997; Eschenbach studies in lowland riparian environments highlight these
and Kappen, 1999; Tardif and Bergeron, 1999). In the field, species-specific differences in soil saturation and toleration of
variation in vegetation across slight topographic gradients root anoxia (Niiyama, 1990; Conner et al., 1997; Rodrı́guez-
has often been attributed to differential tolerance of inun- González et al., 2010).
dation (Bell, 1974; Nixon et al., 1977; Robertson et al., 1978). Despite their vulnerability to drought, riparian plants
But in observational field studies, inundation duration is often demonstrate some mitigating morphological and physiological
correlated with other mechanisms such as maximum flood traits. Annual species avoid drought by completing their life
energy, sediment texture, nutrient availability, and redox po- cycle during wet seasons. For perennial plants, rapid root ex-
tential (Rodrı́guez-González et al., 2010). Therefore, teasing tension and small shoot:root biomass ratios are adaptations
out specific contributions of inundation alone is difficult, es- that potentially reduce stresses related to seasonally variable
pecially as these effects are nonlinear and in some cases con- water tables (Amlin and Rood, 2002; Horton and Clark, 2001;
tradictory. For example, various studies on the influence of Hughes, et al., 1997; Kranjcec et al., 1998; Segelquist et al.,
inundation on tree growth have found negative effects (e.g., 1993). Other morphological responses to water stress include
58 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective

reduction in leaf size (Stella and Battles, 2010), specific leaf area Gasith and Resh, 1999; Cramer and Hobbs, 2002). Glenn
(Busch and Smith, 1995), crown dieback (Scott et al., 1999), et al. (1998) found higher salinity tolerance by saltcedar
branch abscission (Rood et al., 2000), and reduced diameter (Tamarix spp.), a non-native shrub in the US Southwest,
growth (Stromberg and Patten, 1996). Studies of physiological compared to native riparian species, and concluded that this
function in adults indicate that these species are generally in- abiotic factor was important in facilitating invasion by the
tolerant of drought and have low xylem cavitation thresholds saltcedar throughout the region’s arid and saline riparian soils.
(Amlin and Rood, 2003; Cooper et al., 2003; Leffler et al., Sher et al. (2002), however, found that salinity and other soil
2000; Tyree et al., 1994). Higher water-use efficiency as indi- abiotic variables were much less important in predicting the
cated by higher d13C values is a common response of mortality rate of first-year tamarisk seedlings than the density
water-stressed plants (Smedley et al., 1991) and has been ob- of native competitors (cottonwood and willow), which sur-
served for riparian seedlings under experimental drought vived across a range of substrates.
(Zhang et al., 2004) and adult natural populations between
wet and dry years (Leffler and Evans, 1999). Water-
12.5.4.6 Propagule Dispersal
table manipulations in controlled mesocosms have been used
to simulate dynamic riverine environments in order to Hydrogeomorphic influences on plant dispersal constitute
study the effects of seasonal moisture stress on plant survival another important mechanism affecting riparian vegetation, in
and growth (e.g., Cordes et al., 1997; Horton and Clark, part, because of co-evolution of propagule traits with dis-
2001; Mahoney and Rood, 1992; Segelquist et al., 1993; Stella turbance regimes (Lytle and Poff, 2004). Unlike in most ter-
et al., 2010). In one study, Stella and Battles (2010) found restrial environments, many plants in riparian ecosystems
that seedlings of related species grown under identical con- have little dependence on a persistent seedbank, particularly
ditions of water stress displayed different adaptations, with where physical disturbance from periodic flooding precludes
cottonwood minimizing specific leaf area to a greater degree seed storage (Pettit and Froend, 2001). Lacking a mechanism
than willow, which was more effective at reducing stomatal for multiyear seed storage, riparian plant recruitment typically
conductance and leaf size. results from dispersal of short-lived seed directly from parent
plants (Young and Clements, 2003a, 2003b; Stella et al.,
2006), or else transport of plant fragments as propagules
12.5.4.5 Soil Chemistry
(Gurnell et al., 2008). For seeds, dispersal is primarily by wind
Nonhydrologic factors also contribute to riparian plant dis- (anemochory) and water (hydrochory). The latter frequently
tributions, including resource competition, lack of soil fer- depends upon synchronization of reproduction and seed
tility, and salinity (Shafroth et al., 1995; Auble and Scott, dispersal with hydrological regimes that create favorable sub-
1998; Sher et al., 2000). Detailed work on establishment strate and moisture conditions for seed dispersal (Hupp,
processes on fluvial landforms, including point bars, banks, 1992), and rapid germination and recruitment (Siegel and
and abandoned channels, has highlighted the importance of Brock, 1990; Scott et al., 1996; Mahoney and Rood, 1998;
the soil environment, particularly nutrients and organic mat- Stella et al., 2006). Vegetative reproduction via flood dispersal
ter (Van Cleve et al., 1996; Bechtold and Naiman, 2006, of plant fragments has also been observed for a range of taxa
2009). In a study by Kalliola et al. (1991) on riparian forest (Gurnell et al., 2008), from willows (Douhovnikoff et al.,
colonization in western Amazonia, distinct vegetation com- 2005) to columnar cacti (Parker and Hamrick, 1992).
munities colonized four common fluvial landforms: channel Hydrochory may be most important during overbank flows
bars, swales, abandoned channels, and riverbanks. These when propagules are dispersed across floodplains (Schneider
newly deposited fluvial sediments are poor in organic carbon and Sharitz, 1988; Gurnell et al., 2006, 2008). Wind dispersal
and nitrogen, are affected by seasonal fluctuations in the riv- also occurs preferentially along longitudinal stream corridors
ers, and support vegetation patches that tend to be narrow, because local channel topography and the morphology of ri-
curved, or linear patches. Local site and colonizing vegetation parian canopies often serve to guide prevailing winds (Devitt
characteristics vary considerably between the different river et al., 1998).
types (e.g., meandering or braided, rich or poor in suspended As a result of these patterns, hydrogeomorphic forces are
sediment). highly effective at accomplishing dispersal in both longi-
Bechtold and Naiman (2006) studied nutrient storage and tudinal and lateral dimensions along stream channels (Merritt
mineralization along a toposequence on the Phugwane River and Wohl, 2002). Propagule deposition is not uniform,
in South Africa, and found that total organic carbon (TOC), however, because of variation in propagule source density
total nitrogen (TN), and potential N mineralization were (Clark et al., 1999), substrate exposure (Johnson, 1994), and
strongly linked to particle size distributions, with TOC and TN local variations in water velocity, hydraulic roughness and
positively correlated with silt and clay concentration. After form drag due to channel morphology, existing vegetation
accounting for the effect of particle size, landform differences structure and density, bed and bank substrate, and large
were not predictive of nutrient dynamics; therefore, they woody debris (Johansson et al., 1996; Merritt and Wohl, 2002;
concluded that a collinear gradient of soil texture across al- Pettit and Naiman, 2006). Field studies (e.g., Gurnell et al.,
luvial landforms constitutes the primary nutrient influence on 2008), flume experiments (e.g., Merritt and Wohl, 2002), and
riparian soil and plant community succession. models (Levine, 2003) have attempted to quantify and predict
Soil salinity is commonly cited in dryland riparian systems patterns of riparian propagule dispersal, and the effects of
as a driving factor determining plant distributions, product- dams on riparian plant distributions through interruption of
ivity, and species’ relative competitiveness (Di Tomaso, 1998; dispersal patterns have been documented (Nilsson and
 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective 59

Jansson, 1995; Jansson et al., 2000). Although these studies estimates of n vary based on vegetation community type and
have provided details from a variety of settings, the complex time of year, and it is important to note that this method of
nature of fluvial hydrodynamics, heterogeneous riparian en- quantifying vegetation effects is not mechanistic, nor does it
vironments, and variation in seed sources and morphologies take into account interactions between vegetation drag and
currently preclude a comprehensive, predictive understanding hydrodynamic forces (Corenblit et al., 2007).
of dispersal. More recently, researchers have begun to quantify the ef-
fects of vegetation on flow velocities, both as one-way pro-
cesses (involving rigid stems) and interactive ones (involving
12.5.5 Influence of Vegetation on Geomorphology flexible vegetation). Experimental and theoretical work in-
volving rigid, nonsubmerged stems approximate the effects of
In contrast to the study of physical factors on vegetation, woody plants on overbank floods, and show that the diameter,
which have been conducted primarily by ecologists and bio- density, and clustering of stems affect both flow velocity
geographers, research on the influence of vegetation on fluvial and stage. Stone and Shen (2002) conducted flume experi-
geomorphic processes and landforms has been conducted ments with cylinders to represent rigid plant stems of various
primarily in the fields of geomorphology (typically from sizes and densities. They showed that flow resistance varies
physical geography and earth sciences) and civil and en- with density, length, and diameter of stems, as well as flow
vironmental engineering (Darby, 1999; Corenblit et al., 2007; depth, and developed physically based formulas for resulting
Sandercock et al., 2007). A long history of research exists on flow resistance and velocity. Nepf (1999) extended the
vegetation influence over the hydrological cycle (e.g., Tabacchi work on rigid vegetation to emergent stems with feedbacks
et al., 2000), on ecophysiological fluxes (e.g., Roberts, 2000), to vegetative drag and flow turbulence, and developed a
and on water quality (e.g., Dosskey et al., 2010). However, model to describe the drag, turbulence, and diffusion for flow
until relatively recently, quantitative research on vegetation through emergent vegetation with varying properties of stem
controls over hydrogeomorphic processes was generally de- density.
scriptive (e.g., Nanson and Beach, 1977), or fairly limited in The role of large woody debris (LWD) in the initiation and
scope and in the complexity of processes observed (Murray evolution of geomorphic landforms, first described by Keller
et al., 2008). and Swanson (1979), has received substantial attention over
During flooding, woody plant stems and canopies add the last two decades (e.g., Bilby and Ward, 1991; Fetherston
drag to a fluvial system, influencing both hydraulics and et al., 1995; Abbe and Montgomery, 1996; Gurnell et al., 2002;
sediment dynamics (Nepf, 1999; Lightbody and Nepf, 2006). Jeffries et al., 2003; Lassettre et al., 2008; Opperman et al.,
In addition, vegetation roots increase erosion resistance of 2008). Woody debris loading and transport have been cor-
banks, and trapping sediment adds cohesion to the substrate related with in-channel sediment storage (e.g., Bilby and
(Knighton, 1984). Corridor-wide fluvial characteristics and Ward, 1991), pool spacing (e.g., Montgomery et al., 1995),
processes affected by vegetation include velocity, flood stage, channel island formation (e.g., Gurnell et al., 2002), flood-
flow resistance, and sediment transport (Murray et al., 2008). plain accretion (e.g., Jeffries et al., 2003), and increased
Local geomorphic influences include bank erosion resistance, channel complexity (e.g., Pettit et al., 2006; Abbe and
bar sedimentation, formation of logjams, and floodplain Montgomery, 1996), among other features. Abbe and Mont-
sedimentation rates (Hickin, 1984). gomery (1996) documented the initiation of channel-altering
Efforts to predict the contribution of vegetation to velocity jams when key-member logs become lodged in the channel.
date back to the nineteenth century, when Manning integrated Over time, these jams collect more LWD and sediment, and
empirical research by Darcy, Weisbach, St. Venant, Ganguillet, form stable structures over 101–102 years that control local
Kutter, and others to develop his equation for flow resistance channel hydraulics and ultimately assist in riparian forest
in vegetated channels (Manning, 1891). The roughness co- development. Jeffries et al. (2003) documented that LWD on
efficient n in the Manning equation represents the collective floodplains increased the rate and the variability of overbank
drag exerted on the flow by surface roughness (including flooding and deposition depths.
vegetation) and channel sinuosity. Subsequent research has LWD interacts with sediment to exert strong effects on al-
disaggregated this term into the sum of various environmental luvial channel morphology, particularly where longitudinal
components, including vegetation (e.g., Cowan, 1956). The slopes are gentle enough to trap bedload behind logjams. In a
widespread use of Manning’s equation across a range of study of LWD effects on channel habitat in the Pacific
hydrological, geomorphological, and engineering applications Northwest (US), Montgomery et al. (1995) found that pool
makes the determination of Manning’s n profoundly import- spacing depended on LWD loading and channel type, slope,
ant. Of the components that contribute to n, vegetation is and width. Mean pool spacing in channels with gentle slopes
among the most variable and, hence, critical (Arcement and decreased 13-fold with increased LWD loading, whereas pool
Schneider, 1989). Furthermore, the changes in resistance that spacing in steeper channels was not affected. From their study
accompany plant growth (or loss) mean that the roughness of sediment processes on the Tagliamento River (Italy),
coefficient may be quite variable through time (Chow, 1959). Gurnell et al. (2001) developed a conceptual model of island
Darby (1999) modified existing hydraulic models to pre- development that integrates interactions between LWD and
dict stage-discharge curves for channels with nonuniform vegetation, geomorphic features, sediment texture, and
cross sections, sand and gravel-bed materials, and flexible or hydrological regime. Depositional and erosional processes
nonflexible riparian vegetation. Because roughness depends resulted in different island types and floodplain develop-
on vegetation stature, density, and flexibility, empirical mental stages.
60 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective

The potential for LWD to resprout or for live trees to form narrowing. He suggested that within 25 years after the species’
logjams in stream channels is another mechanism by which introduction, it had driven the system into a new equilibrium
vegetation affects geomorphic processes, often with longer- condition. Numerous subsequent studies have also shown
term effects than for dead wood (Opperman et al., 2008). vegetation colonization of former active channel bars and
Following a 100-year flood on the Sabie River in South Africa, banks to reduce channel planform width and width/depth
Pettit et al. (2006) found that patterns of LWD accumulation ratios, as well as change channel cross-sectional geometry (e.g.,
were dominated by characteristics of the trapping sites, espe- Friedman et al., 1996a, 1996b; Johnson, 1994, 1998; Trush
cially trees that fell but remained rooted in place and et al., 2000; Liébault and Piégay, 2001). Although these devel-
resprouted (36% of piles surveyed). These resprouted piles opments can be induced by natural processes such as tempor-
supported tree seedlings (28% of piles) and created hetero- ary increases in precipitation (e.g., Martin and Johnson, 1987)
geneous topography in the main channel and floodplain. and recovery from large floods (Liébault and Piégay, 2002),
Opperman and Merenlender (2007) found that living trees in many are the result of dams and other human modifications to
Northern California streams represented a major portion of physical processes that reduce sediment supply (e.g., Kondolf
the functional in-stream large wood, was 28% more persistent et al., 2007) or magnitude and frequency of high-flow events
than LWD, and was likely more stable than dead wood of (e.g., Johnson, 1998; Michalková et al., 2011).
similar size. They concluded that live wood may have a dis- Recent experimental work highlights the importance of
proportionate influence on channel morphology, particularly vegetation in transforming braided streams into single-thread
in riparian ecosystems lacking many large diameter trees. channels in coupled stream–floodplain systems (Tal and
Bank cohesion is another important geomorphic charac- Paola, 2007, 2010; Murray et al., 2008; Braudrick et al., 2009).
teristic that is at least in part a function of vegetation attri- Gran and Paola (2001) and Tal and Paola (2007) have
butes, including erosion resistance conferred by roots and shown in flume settings that floodplains supporting vege-
increased cohesion from fine sediment deposition induced by tation (alfalfa sprouts) produced more single-thread river
plant hydraulic roughness (Knighton, 1984; Murray et al., planforms than braided ones, and that the resulting channels
2008). In an early example, Clifton (1989) showed that were narrower, deeper, and less mobile. Furthermore, the ex-
morphology of mountain streams varied with vegetation perimental addition of vegetation to previously braided
structure, in addition to physiography and land use. Abernathy channels has induced self-organization of the channel net-
and Rutherford (1998) developed a classification scheme for work, including formation of a single thread, control of bend
assessing the role of vegetation in stream bank erosion at and bar migration rates, and formation of channel cutoffs
different points throughout a catchment. Their experimental (Braudrick et al., 2009; Tal and Paola, 2010).
work focused on the enhancement of bank strength by re- However, the magnitude of vegetation effects is highly
ducing pore-water pressures and by directly reinforcing bank context dependent. Labbe et al. (2011) found that riparian
material by plant roots (Abernathy and Rutherford, 2001). vegetation was not a significant control on channel cross-
Experimental work by Simon and Collison (2002) quantified section form on the Tualatin River, Oregon. Sandercock et al.
bank strength effects on constituent hydrologic and mechan- (2007) noted that in contrast to humid climates, semi-arid
ical processes, including soil moisture modification and root and arid climate riparian zones are characterized by locally
reinforcement, as well as the potentially destabilizing influ- patchy vegetation and channel morphology influenced by
ence of vegetation weight, and tested their effects on the extreme flood events. Therefore, the role of vegetation in bank
probability of bank failure. Tree and grass roots exerted spe- stability in these environments is expected to be lower overall
cies-specific increases on soil strength, and mechanical effects than in humid climates, and the degree of influence highly
of plant cover increased stability by 32–71%. Martin and dependent on local distribution, composition, and density
Church (2000) found that the addition of roots to riverbanks (Sandercock et al., 2007).
improved stability even under worst-case hydrological con-
ditions, and was apparent over a range of bank geometries,
12.5.6 Feedbacks between Vegetation and
with the best protection offered by trees closest to the poten-
Hydrogeomorphology
tial bank failure locations. Much of the work on root re-
inforcement has been based on models derived from materials
If, as discussed above, hydrogeomorphic factors influence
science; however, Pollen and Simon (2005) noted that the
vegetation, and vegetation influences hydrogeomorphic pro-
results derived can vary widely depending on the sophisti-
cesses, then it follows that there are likely to be feedbacks
cation of the models used.
between the two. Indeed, not only are there interactions be-
At a landscape scale, increased bank cohesion due to
tween vegetation and the hydrogeomorphic mechanisms de-
belowground vegetation effects is instrumental in controlling
scribed in Sections 12.5.4.1–12.5.4.6, but many of those
river meander rates (Micheli et al., 2004) and frequency of
mechanisms also interact with each other. A diagram of
channel cutoffs (Micheli and Larsen, 2011). Hupp (1992) cited
these interactions serves both to summarize many of the
the combined impact of stabilization by roots and sedimen-
relationships described in this chapter and to illustrate
tation due to the flow resistance from plant stems in recovery
the complexities of riparian biogeomorphic feedbacks (Fig-
and meander initiation along formerly channelized streams.
ure 1). The relationships shown may be briefly described as
Graf (1978) found that following the introduction of tamarisk
follows:
to upper reaches of the Colorado and Green Rivers, the com-
bination of bank stabilization and overbank sedimentation led a. Flood energy affects riparian vegetation directly, through
to the development of stable islands and dramatic channel mechanical damage or scour, as discussed in Section 12.5.4.1.
 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective 61

Flood
energy
c o

a
Sedimentation b
d Propagule
dispersal
p
Vegetation
e
g m
f n
k
j
Soil
h chemistry
Inundation

Water
l
i table
Figure 1 The network of potential biogeomorphic interactions and feedbacks that may link riparian vegetation and fluvial geomorphology.
Influences represented by the arrows are discussed in the text.

b. Riparian vegetation affects flood energy through hydraulic vegetation, through its impact on flood energy (b) has a very
roughness (Section 12.5.5). strong influence on sedimentation (c). Overall, Figure 1 serves
c. Flood energy affects sedimentation, controlling the size to illustrate the centrality of feedbacks in the riparian zone.
distribution of mobilized sediment, its sorting, and the Some or all of these feedbacks have been noted in earlier
spatio-temporal patterns of deposition. reviews, including those by Parker and Bendix (1996), Bendix
d. Sedimentation affects riparian vegetation directly, through and Hupp (2000), Hupp and Bornette (2003), Steiger et al.
burial of existing plants and through creation of alluvial (2005), and Corenblit et al. (2007). A limited number of
surfaces for colonization (Section 12.5.4.2). empirical and review studies have specifically addressed the
e. Sedimentation affects soil chemistry, through the com- issue of feedbacks. Although these feedbacks were addressed
position of the alluvium deposited and particle size con- in a few early studies (see below), most such work has been
trols on nutrient dynamics and mineralization rate. published since the mid-1990’s. Murray et al. (2008) referred
f. Sedimentation affects water-table depth through the to the feedbacks between organisms, physical processes, and
height of depositional landforms and the texture of the morphology as ‘biomorphodynamics.’ They attributed the in-
deposits (i.e., capillarity). creased attention to such interactions in a variety of environ-
g. Sedimentation affects inundation through the height of ments (not limited to fluvial/riparian settings) to the increased
depositional landforms. popularity of interdisciplinary research, the challenges of
h. Inundation affects riparian vegetation directly, through realistically analyzing responses to environmental change, and
anoxia (Section 12.5.4.3). to the advances in modeling capabilities at a variety of scales.
i. Inundation affects water-table depth and dynamics In his argument for application of complexity theory to
through contributions to groundwater. biogeomorphology, Stallins (2006) argued for a focus on feed-
j. Water-table depth and dynamics affect riparian vegetation backs, stating a need to ‘‘move beyond listbound descriptions of
directly, through provision of water or through limiting unidirectional interactions of geomorphic and ecological com-
the aerated rooting zone (Section 12.5.4.4). ponents.’’ He emphasized the importance of recursivity, whereby
k. Riparian vegetation affects the water table through the interactions between floods and vegetation constrain future
transpiration. development of the system, so that it becomes self-organizing.
l. The water table affects soil chemistry through its impact This notion is inherent in the view of Corenblit et al. (2009)
on redox potential. who argued that the self-organizing properties of riparian sys-
m. Soil chemistry affects riparian vegetation directly, by tems reflect both contemporary feedbacks that allow for niche
serving as the primary control of nutrient availability construction by plants and longer-term natural selection as
(Section 12.5.4.5). plants evolve for specific roles within the system.
n. Vegetation affects soil chemistry through plant litter and Empirical studies of feedbacks have focused primarily on
mineralization rate in the root zone. pathways whereby riparian vegetation promotes sedimentation
o. Flood energy provides transport for propagules through increased hydraulic roughness by live, rooted stems, or
(Section 12.5.4.6). through the role of LWD; in turn, the increased sedimentation
p. Dispersal of propagules directly affects vegetation, allow- affects the vegetation able to grow at a site. In some instances,
ing its establishment at new sites. these feedbacks become part of the mechanism whereby suc-
cession occurs.
It is important to note that some of the most important In an early example of feedbacks centered on roughness,
relationships are actually indirect. For example, riparian Nanson and Beach (1977) related overbank sedimentation
62 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective

and forest succession on the Beatton River floodplain in often recursive model structures. An analog is provided by
British Columbia. They found that dense balsam poplar col- efforts in the field of hillslope modeling, in which represen-
onizing new alluvial surfaces promoted increased sedimen- tations of vegetation growth are coupled with surface runoff
tation, for which the poplar had a high tolerance. Eventually, and landscape evolution models (e.g., channel–hillslope in-
however, aggradation reduced sedimentation sufficiently for tegrated landscape development (CHILD) model, Tucker and
shade-tolerant white spruce to replace the intolerant poplar. Bras, 1999; Tucker et al., 2001; Istanbulluoglu and Bras,
As the spruce matured, sedimentation declined even further 2005). An early numerical modeling effort for river channels,
because the lower-stem density of the mature spruce forest about which much less work has been conducted than for
reduced hydraulic roughness. On point bars of Dry Creek in hillslopes, consists of work by Murray and Paola (1994,
California, McBride and Strahan (1984b) noted that colon- 1997), which explored channel braiding through simple nu-
ization of gravel substrate by willows and cottonwoods re- merical feedbacks between bedload transport and channels
sulted in deposition of finer sediments. The finer substrate with noncohesive banks. Over a series of modeling exercises,
allowed for subsequent establishment of alder, which even- they found that braiding occurred under conditions of sparse
tually formed a canopy that ultimately inhibited the shade- or slowly growing vegetation, and high sediment fluxes were
intolerant willows and cottonwoods. Along rivers in the Oz- caused by high discharge and/or steep regional slopes. In their
arks, McKenney et al. (1995) described models whereby studies, the channels reorganized on time frames shorter than
channel migration was driven by the age of the vegetation on the riparian plants’ life spans.
gravel bars. Young, dense stands or patches of riparian vege- An alternative feedback quantification approach is
tation created high roughness and favored deposition and provided by Perucca et al. (2007), who coupled a fluid dy-
stabilization of the surface. As the vegetation aged and self- namic model of meandering rivers with a process-based
thinning led to reduced stem densities, roughness decreased, model of riparian biomass dynamics. In this approach, ri-
and the surfaces actually became less geomorphically stable. parian vegetation influences channel morphology via a rela-
Pettit and Naiman (2006) provided an instance in which tionship between biomass, density, and bank erodibility. Their
LWD is central to riparian feedback. A large flood on the Sabie numerical results show a strong effect of vegetation growth on
River (geomorphic process) not only devastated the existing meander evolution and, like Murray and Paola (1994, 1997),
riparian vegetation, but also interacted with woody debris they emphasized the sensitivity of their results to the relative
(vegetation influence) to create numerous LWD jams. Those temporal scales of plant growth versus morphodynamic
LWD accumulations, in turn, mediated the geomorphic processes.
mechanisms of flood energy, water table, and soil chemistry by
providing seedling germination sites that were protected from
floods and had high moisture and nutrient availability.
Certain feedback scenarios may be constrained to specific 12.5.7 Patterns in Published Literature
biogeographic locations. In Washington’s Olympic Moun-
tains, Latterell et al. (2006) described a dynamic patch mosaic The forgoing review illustrates that a large volume of research
of fluvial landforms, including a variety of channels, bars, has brought increasingly complex perspectives on ecogeo-
floodplains, and terraces, each with a characteristic vegetation morphic interactions to the study of fluvial/riparian environ-
assemblage. The distribution of these landform/patches was ments. That large volume suggests that a systematic analysis of
controlled by lateral channel migration, but channel behavior published work might reveal underlying patterns in what has
is influenced by LWD accumulations. Woody debris accumu- been learned to date. Accordingly, we undertook an analysis of
lations in turn require key-member logs, contributed by ero- the past 20 years of published literature relating riparian
sion of mature terraces (classified based on stand age and vegetation to hydrogeomorphic processes and/or specific flu-
composition), along with smaller woody debris from other vial landforms. This content analysis was designed to quantify
landforms. They cautioned, however, that their model of the incidence of published content according to a scheme that
patch structure and dynamics would only be applicable in was systematic and objective. Our particular interest was to
comparable environments, and should not be applied to discern whether certain relationships tend to be revealed in
dissimilar settings, such as those where vegetation distribution particular environments, or when studied at particular spatial
is constrained by water availability or where trees do not reach scales. Parker and Bendix (1996) called for research examining
sufficient size for woody debris to play an important a role. In the extent to which vegetation–landform relationships vary
the Transverse Ranges of California, Bendix and Cowell (2010) geographically, suggesting that some might be generalized for
found that the distribution of flood depth across valley floors a variety of environments while others might be unique to
when integrated with species composition allowed for the certain regions. This analysis of the literature offers one ap-
determination of the distribution and rate at which burned proach to that challenge.
snags fell after wildfire. This, in turn, governs the supply and There is a danger in such a study of detecting the patterns
distribution of woody debris, with its attendant geomorphic of researchers’ interests, rather than actual variation in natural
and ecological roles. The exogenous role of fire as a catalyst for processes. Nonetheless, the studies being published pre-
this set of feedbacks limits the applicability of these findings to sumably have some relationship to the predominant processes
environments where riparian forests do in fact burn (Dwire in the locales where they are conducted, so that patterns in the
and Kauffman, 2003). types of interaction in the literature should offer some clues as
Attempts to quantitatively model riparian biogeomorphic to the relative import of those interactions in different lo-
feedbacks are relatively new and typically require complex, cations and at different scales.
 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective 63

12.5.7.1 The Sample and Coding Coding of biome for each study was based on the de-
scription provided in the paper’s text. If such a description was
As an initial step, we searched the Scopus and Web of Science lacking, we located the study site on Figure 1 to determine the
databases for all publications from 1990 through 2009 with appropriate biome. Scale was coded as ‘site’ if the data had been
the term ‘riparian vegetation’ in the title, abstract, or keywords. collected at a single cross section involving o500 m length of
This search yielded 43700 references. We examined the ab- channel, as ‘reach’ if data were collected from a greater length of
stracts of these papers and retained those that met the fol- the river, up to 1 km, and ‘watershed’ if the data were
lowing criteria: (1) the paper included substantive analysis of from multiple sites or from 41 km length of the river.
hydrogeomorphic interaction with the riparian vegetation; (2) Causal direction reflected the emphasis of the study:
the study was data based, rather than a review or solely the- if it focused on hydrological and geomorphic influences on
oretical discussion; and (3) the data were collected in the field, vegetation, it was coded as ‘geomorphic’; if it dealt primarily
rather than laboratory, because the latter would not be in- with the impact of vegetation on geomorphology and/or hy-
herently reflective of geographic or scalar variation. drology, it was coded ‘vegetation’; and if the primary emphasis
For each of the papers that passed this filter, we recorded was on feedbacks between the two, it was recorded as ‘feedback’.
the year of publication, the journal (or conference proceed- Because so much research has been directed at dis-
ing), the continent and country on which the data were col- entangling the specific mechanisms of hydrogeomorphic in-
lected, the study region’s ecological biome, the scale of fluence on riparian vegetation, we further coded those studies
observation of the study, the causal direction of the environ- for which the causal direction was geomorphic, focusing on
mental relationships studied, and the causal mechanism the six categories of causal mechanisms described above
found (for hydrogeomorphic influence only; see below). The (Table 1). In each instance, we coded for the primary influence
first four of these are rather straightforward, although country identified in the paper. If the study did not find that a single
was occasionally complicated by cross-boundary studies. For influence predominated, but rather that multiple factors were
biome, we classified each study as being in one of the 14 exerting comparable degrees of influence, we coded it as
biomes identified by the Millennium Ecosystem Assessment ‘multiple’. We tried to be parsimonious in the use of this
(2005; Figure 2). classification, and although many papers mentioned potential

Biomes
Tropical and subtropical moist broadleaf forest Tropical and subtropical grassland, savanna, and shrubland
Tropical and subtropical dry broadleaf forest Temperate grassland, savanna, and shrubland
Tropical and subtropical coniferous forest Montane grassland and shrubland
Temperate broadleaf and mixed forest Flooded grassland and savanna
Temperate coniferous forest Mangrove
Boreal forest / taiga Desert and xeric shrubland
Tundra Rock and ice
Mediterranean forest, woodland, and scrub

Figure 2 Biome boundaries used in the study. Based on the Millennium Ecosystem Assessment, 2005. Ecosystems and Human Well-Being:
Biodiversity Synthesis. World Resources Institute, Washington, DC, 86 pp.
64 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective

covariables in passing, we did not code a study as multiple if journals with multiple papers accounted for almost half
its focus seemed clearly to be on one mechanism. (136) of the total (Table 2). In general, the number of papers
on the topic has increased (Figure 3), albeit with some fluc-
tuation. Over the past 5 years, more than 20 papers per year
12.5.7.2 General Characteristics of the Sampled Literature
have dealt with riparian vegetation–landform relationships.
There were 274 studies that met our criteria for inclusion in The majority of the papers (189) examined geomorphic in-
the sample. The papers appeared in 101 venues, but 12 fluences, with 62 being coded as vegetation, and just 23

Table 1 Causal mechanisms coded for studies identified with the primary causal direction of geomorphic influence over vegetation

Causal mechanism Criteria for classification

code

Energy Hydraulic energy of floodwaters, whether destroying vegetation or eroding its substrate (Section 12.5.4.1)
Sedimentation Sediment deposition, whether by its impact on existing vegetation or by creation of new substrate (Section 12.5.4.2)
Inundation Extended inundation by floodwaters (Section 12.5.4.3)
Water table Depth to the water table under the landform(s) on which the vegetation was growing, and variation in water-table depth.
(Section 12.5.4.4)
Soil chemistry Soil chemistry of the landform(s) on which the vegetation was growing (Section 12.5.4.5)
Dispersal Transport of plant propagules by floodwaters (Section 12.5.4.6)
Multiple Multiple factors were exerting comparable degrees of influence

Table 2 Journals in which five or more of the sampled papers were published, by number and percent of the total sample, with number within
each causal direction

Journal Number (%) of papers Geomorphic Vegetation Feedback

River Research and Applications (and Regulated rivers) 31 (11.3) 28 3 2

Geomorphology 19 (6.9) 3 10 6
Earth Surface Processes and Landforms 17 (6.2) 5 10 2
Wetlands 14 (5.2) 14 0 0
Ecological Applications 10 (3.6) 8 1 1
Plant Ecology 8 (2.9) 8 0 0
Forest Ecology and Management 7 (2.6) 7 0 0
Journal of the American Water Resources Association 7 (2.6) 2 3 2
Journal of Vegetation Science 7 (2.6) 7 0 0
Physical Geography 6 (2.2) 4 1 1
Annals of the Association of American Geographers 5 (1.8) 5 0 0
Hydrological Processes 5 (1.8) 1 3 1

35

30

Geomorphic driver
25 Vegetation driver
Number of studies

Feedbacks
20

15

10

0
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009

Year
Figure 3 Year of publication of the studies included in the sample, and the causal direction examined.
 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective 65

160

Xeric shrub
140
Temperate broad forest
Temperate conif forest
120
Mediterranean
Temperate grassland

Number of studies
100
Tropical grassland
Boreal forest
80 Tropical moist forest
Tropical dry forest
60

40

20

0
Africa Asia Australia Europe North South
America America
Figure 4 Number of studies conducted in each biome, on each continent.

feedback. Indeed, the interest in vegetation as driver is rela- and western North America). The relative scarcity of papers in
tively recent, with few papers appearing before 1997. The the data set from South America, Africa, and Asia contri-
disparity in the causal direction studied is reflected in Table 2, butes to the underrepresentation of tropical biomes in our
as only four of the journals with five or more papers in the survey.
data set had a majority of their papers on vegetation’s influ- Across all of the biomes, the studies tended toward larger
ence: Geomorphology, Earth Surface Processes and Landforms, scales of observation, with watershed scale being the most
Journal of the American Water Resources Association, and Hydro- common and site scale being the least common (Figure 5).
logical Processes. It is logical that these journals, with their Overall, flood energy was the most common mechanism
emphasis on landforms and hydrology, tended to publish (28%), followed by water table and multiple (both 21%), and
studies in which the influences on those topics are examined. sedimentation (12%). These were also the mechanisms that
However, the extent to which the rest of the literature em- were found to operate in most biomes. There were, however,
phasizes vegetation as the dependent variable is rather strik- distinct differences in the spatial distribution of these mech-
ing. The smaller number of studies addressing feedbacks may anisms, as discussed below.
well have a logistical explanation. Although most scientists
working in riparian environments would probably acknow-
ledge the importance of feedbacks, to actually study them in 12.5.7.3 Differences among Biomes
the field poses the challenge of incorporating two field studies
(vegetation driver and geomorphic driver) into one, in order The research within most of the biomes reflects the general
to discern their interactions. In addition, feedbacks take time tendency of the literature to focus on the geomorphic causal
to observe in the field, as there must be time for an initial direction (Table 3). The two tropical and subtropical forest
process to occur, and then for a subsequent process to occur in biomes are exceptions to this generalization, although with
response. just 10 studies between them, they can hardly be said to
There was also a distinct geographic bias to this literature. constitute a regional research trend. Our discussion, then,
The great majority of the studies was carried out in North deals primarily with the hydrogeomorphic influences on
America, with a secondary peak in Europe (Figure 4). Aus- vegetation.
tralia, Asia, and Africa were the setting for substantially fewer The number of studies within each biome that focused on
studies, with South America accounting for the least of all. each mechanism of hydrogeomorphic influence is shown in
This pattern presumably reflects the fact that the majority of Figure 6. A striking, and intuitive, feature of this graph is the
the scholars publishing the studies are based at institutions in prominence of depth to water table in desert and xeric shrub-
North America, Europe and, to a lesser extent, Australia. In land (44% of the studies in the biome). Where water is the
turn, with more than 80% of the studies being conducted in overriding limiting factor for vegetation, the proximity of
Europe (primarily Western Europe), North America, and subsurface water becomes all the more important. The flashy
Australia, it is unsurprising that there is also an imbalance in hydrology that characterizes many desert environments pre-
the biomes studied. Most study areas were located in tempe- sumably accounts for the prominence of flood energy as a
rate broadleaf and mixed forest (fairly widely distributed), mechanism in this biome, as high-energy floods are to be ex-
desert and xeric shrubland (mostly in North America), tem- pected. Conversely, flood duration in most deserts tends to be
perate coniferous forest (mostly in North America), and short, so that the 11% coded as inundation are more puzzling.
Mediterranean forest, woodland and scrub (mostly in Europe Some authors did simply discuss inundation as an important
66 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective

Table 3 Number of papers by study area biome and causal direction

Biome Number of papers Geomorphic Vegetation Feedback

Temperate broadleaf and mixed forest 77 44 28 5

Desert and xeric shrubland 58 45 6 7
Temperate coniferous forest 49 38 9 2
Mediterranean forest, woodland, and scrub 38 28 5 5
Tropical and sub-tropical grassland, savanna, and shrubland 21 15 4 2
Temperate grassland, savanna, and shrubland 14 11 2 1
Tropical moist forest 8 1 7 0
Boreal forest 7 6 1 0
Tropical and sub-tropical dry broadleaf forest 2 1 0 1

variable without specifying whether the impact of the inun- table, sedimentation, dispersal, and inundation. The tempe-
dation was through the submergence or through mechanical rate broadleaf forest and the temperate grassland are the only
damage sustained during inundation, and this vagueness may biomes in which soil chemistry appears as an influence.
account for the high numbers of inundation references. The paucity of studies from tropical forests makes it dif-
Mediterranean forest, woodland, and scrub have much in ficult to infer much about these biomes. It is perhaps un-
common with desert and xeric shrubland, with the pro- surprising that water-table depth was important to riparian
nounced dry season making water a critical limiting resource, vegetation in tropical and subtropical dry broadleaf forest,
and characteristically with similarly flashy hydrologic regimes. but with only one study conducted there is not enough to
In Mediterranean environments, flood energy predominated establish a pattern. Again, with only one study from a low-
as in xeric regions, but water table was much less prominent. order mountain stream in tropical and subtropical moist
Many of the studies coded as multiple in this biome did in- broadleaf forest, citing multiple mechanisms, it is similarly
clude water table as one of the variables. Nonetheless, given not informative.
the prominence of obligate riparian species in Mediterranean Finally, the boreal forest/taiga, although also indicating a
biomes, presumably present only due to the presence role for flood energy, is notable for the prominence of dis-
of a (relatively) shallow water table, it is surprising that this persal. In this case, the pattern is clearly a result of the sample
variable did not emerge in more studies. Sedimentation characteristics, rather than an anomalously important role for
was important here, reflecting the high sediment load and hydrochory in boreal forests. The sample for this biome was
sediment mobility typical of many Mediterranean streams. small (Figure 6), and happened to be dominated by a group
Sedimentation may also be important because of the impact of scholars in Sweden who published several papers on dis-
of sediment texture on capillarity, which mediates soil mois- persal during the time period covered by our study (e.g.,
ture availability. Both the mobility of sediment and its im- Andersson et al., 2000). Indeed, this example offers a useful
portance for water availability are characteristic of streams in reminder that the interests of individual researchers have a
most dry environments, so that the less frequent role reported distinct potential to skew results for any biome (or scale) in
for sedimentation in desert studies is rather anomalous. which the number of papers is limited.
The temperate and the tropical/subtropical grassland, sa-
vanna, and shrubland had much in common. Each had flood
12.5.7.4 Scale-Related Differences
energy as the most prominent mechanism, with water table
and sedimentation also being important. Water table was, The contrasts in mechanisms predominating in studies with
however, distinctly more important in the tropical and sub- different scales of observation (Figure 7) are less dramatic
tropical grasslands than those in temperate zones. This likely than among the varied biomes, but some differences are ap-
reflects the greater heat stress and consequent water demand parent. In keeping with their overall importance, flood energy
in the lower-latitude biome than in temperate areas. As in and water table are important across all scales. But both are
other dry environments, high sediment mobility and the ca- more prominent in studies at the site scale (o–0.5 km of
pillary influence of substrate texture are reflected in the im- stream bank length) than at reach (0.5–1 km) or watershed
portance of sedimentation. (41 km) scales. This is logical, as both have greater potential
The temperate broadleaf and mixed forest has the most variance at a given cross-section than along a gradient up or
heterogeneous mix of mechanisms examined. This reflects two downstream. Along a cross-section, they will vary from a
factors acting in combination. One is simply that with so presumed maximum energy and minimum distance to water
many studies having been conducted in the biome, there is an table at or near the thalweg to zero energy and maximal water-
increased likelihood that most conceivable mechanisms will table depth at some distance from the channel. This may also
have been examined. It is a type of environment in which be why inundation is so much more prominent at the site
there is no overriding stressor (such as drought), allowing for scale. It is similarly logical that dispersal is irrelevant at the
a range of individual mechanisms to prevail, depending on individual site scale, where there is no space for transport to
the specific conditions in a given study area. Much the same is occur, but increases in prominence with increasing scale. It is
true of the temperate coniferous forest. In each biome, flood less clear why no studies at the site scale focused on multiple
energy is the most notable mechanism, but is joined by water mechanisms.
 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective 67

No. of
studies
20

18

16
14
12
10
8
6
4
2 Multiple
0 Flood energy
b Water table
ru t
cs
h es Sedimentation
r i for st
Xe d re
oa fo an Inundation
br nif ne d
te co rra lan
ra e ite ss nd Dispersal
pe at ed gr
a sla st
m per M e s re
Te m rat gra fo t Chemistry
Te pe l l es
m ica rea for st
Te o p o ry re
Tr B d t fo
l s
ca oi
o pi l m
Tr ica
op
Tr
Figure 6 Mechanisms of hydrogeomorphic influence in studies within each biome.

60

50 Watershed
Reach
Number of studies

40 Site

30

20

10

0
gy

al
try

n
le
n

e
io

rs
tio

ab

pl
er
is

at

pe

ti
da
en
em

rt

ul
t
en

is
e

M
un
d
ch

D
at

m
oo

In

di
il

Fl
So

Se

Figure 7 Number of studies emphasizing each mechanism at each scale.

12.5.7.5 Hydrogeomorphic Mechanisms in the Context of common hydrogeomorphic driver affecting riparian vegetation.
Scale and Biogeography However, it also reveals exceptions that indicate the importance
of both the setting and the scale of observation in determining
The broad pattern of the mechanisms acting at varied scales in which drivers will predominate. In desert and xeric shrub, al-
different biomes can be seen in a plot of the modal mechanism though flood energy is important (Figure 6), the overriding
for each combination of scale and biome (Figure 8). This plot importance of distance to the water table is present across all
serves to reinforce the primacy of flood energy as the most scales.
68 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective

Temperate conif forest Flood Flood Flood

energy energy energy

Water Flood Flood

Temperate broad forest
table energy energy

Flood Flood
Flood energy and energy and
Temperate grassland
energy sedimentation sedimentation

Flood
Mediterranean Sedimentation Sedimentation
energy

Water
Xeric shrub Water table Water table
table

Flood energy,
Tropical grassland inundation, Water table
Flood
sedimentation energy

Site Reach Watershed

Figure 8 Modal mechanisms cited for each combination of scale and biome. Multiple mechanisms reflect scale–biome combinations for which
equal numbers of studies cited the mechanisms shown. Biomes in which fewer than 10 studies had been conducted are omitted from the figure.

In other biomes, the primary mechanism shifts with the the longstanding realization that vegetation also affects fluvial
scale of the study. Flood energy appears most consistently at processes and landforms. This disparity is in part an artifact of
the watershed scale. Clearly, the downstream changes in valley our sampling strategy: we included only field studies, and much
morphology that can be found in many river systems across a of the research on vegetation as a driver has been experimental,
range of environments allow for sufficient variance in flood typically using flumes. Furthermore, the use of riparian vege-
energy for marked patterns of vegetation response to emerge tation as our search term may have served to exclude some of
(Hupp, 1982; Bendix, 1997). At the site scale, however, water the studies examining the impact of plants on fluvial processes.
table assumed the greatest importance for temperate broadleaf Whereas the modifier riparian is logical for a paper with an
and mixed forest. In this humid environment, a high water ecological or biogeographic focus, it becomes redundant as a
table may prove to be a limiting factor, rather than a resource. description of vegetation in a paper on fluvial geomorphology,
Thus, when viewed in detail across a given site, species with for which all vegetation is likely to be riparian. The scarcity of
specialized adaptations occur where the water table is highest papers in the sample that address feedback is probably more
(e.g., Nakamura et al., 2002; Sharitz and Mitsch, 1993). representative of the overall literature, as this is a topic that
Sedimentation is the principal factor at the watershed scale received little attention until recently. The recent publication of
only for temperate grassland, but at decreasing scales appears major theoretical articles specifically advocating recognition
elsewhere, and is the predominant driver for the Mediterra- and study of feedbacks (e.g., Stallins, 2006; Corenblit et al.,
nean environment at both reach and site scale. The deposition 2009) may well accelerate the increase in such work that was
of sediment, whether burying plants (e.g., Wang et al., 1994) noted by Murray et al. (2008).
or creating substrate (e.g., Greco et al., 2007), is in many in- The overall body of literature is dominated by empirical
stances a localized phenomenon, and its variance is likely to studies detailing the impacts of flood energy and water-
be more evident at the site or reach scale than across a table distance on riparian vegetation. But both their relative
watershed. importance and the importance of other mechanisms do vary
with both the biome in which studies are set and the scale of
observation. This variance confirms Parker and Bendix’s
12.5.8 Patterns and Perceptions Revealed in the (1996) speculation that there is a geography of biogeo-
Literature morphic interactions. Indeed, the nature of our sample may
well obscure that geography. Many of the world’s biomes
Our review of the past 20 years of research shows that there has (Figure 2) are either underrepresented or entirely absent from
been a steady and increasing volume of research on riparian our sample.
biogeomorphic relationships. Most of the studies in our sample The underrepresentation of key biomes is one of several
examined hydrogeomorphic influences on vegetation, despite reasons for caution in interpreting our findings. Our emphasis
 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective 69

on identifying a single key factor in as many studies as pos- Bendix, J., Hupp, C.R., 2000. Hydrological and geomorphological impacts on
sible inevitably required simplification of what were often riparian plant communities. Hydrological Processes 14, 2977–2990.
Bilby, R.E., Ward, J.W., 1991. Characteristics and function of large woody debris in
complex analyses. Once a key mechanism is identified from a
streams draining old-growth, clear-cut, and second-growth forests in
study, the question remains of whether that mechanism was southwestern Washington. Canadian Journal of Fisheries and Aquatic Sciences
dominant in the environment (or at the scale) in which it was 48, 2499–2508.
studied, or was simply the mechanism most easily measured Blom, C., Voesenek, L., 1996. Flooding: the survival strategies of plants. Trends in
or of most interest to the author(s). Ecology and Evolution 11, 290–295.
Bond, W.J., Midgley, J.J., 2001. Ecology of sprouting in woody plants: the
Notwithstanding these caveats, we believe that there is persistence niche. Trends in Ecology and Evolution 16, 45–51.
much to learn from this survey. Clearly, some spatial patterns Bornette, G., Tabacchi, E., Hupp, C.R., Puijalon, S., Rostan, J.C., 2008. A model of
do exist in terms of spatial variation and the importance of plant strategies in fluvial hydrosystems. Freshwater Biology 53, 1692–1705.
different types of biogeomorphic interaction. The empty Braatne, J.H., Jamieson, R., Gill, K.M., Rood, S.B., 2007. Instream flows and the
decline of riparian cottonwoods along the Yakima River, Washington, USA. River
spaces in our data may be the most interesting of all: the
Research and Applications 23, 247–267.
biomes that have not been studied, and the types of inter- Braudrick, C.A., Dietrich, W.E., Leverich, G.T., Sklar, L.S., 2009. Experimental
actions that have not been studied within certain biomes, evidence for the conditions necessary to sustain meandering in coarse-bedded
point the way for future research. rivers. Proceedings of the National Academy of Sciences of the United States of
America 106, 16936–16941.
Brewer, J.S., Levine, J.M., Bertness, M.D., 1998. Interactive effects of elevation and
burial with wrack on plant community structure in some Rhode Island salt
References marshes. Journal of Ecology 86, 125–136.
Buffington, J.M., Montgomery, D.R., 1999. Effects of sediment supply on surface
Abbe, T.B., Montgomery, D.R., 1996. Large woody debris jams, channel hydraulics textures of gravel-bed rivers. Water Resources Research 35, 3523–3530.
and habitat formation in large rivers. Regulated Rivers: Research and Burke, M.K., Lockaby, B.G., Conner, W.H., 1999. Aboveground production and
Management 12, 201–221. nutrient circulation along a flooding gradient in a South Carolina coastal plain
Abernathy, B., Rutherford, I.D., 1998. Where along a river’s length will vegetation forest. Canadian Journal of Forest Research 29, 1402–1418.
most effectively stabilise stream banks? Geomorphology 23, 55–75. Busch, D.E., Smith, S.D., 1995. Mechanisms associated with decline of woody
Abernathy, B., Rutherford, I.D., 2001. The distribution and strength of riparian tree species in riparian ecosystems of the southwestern U. S. Ecological
roots in relation to riverbank reinforcement. Hydrological Processes 15, 63–79. Monographs 65, 347–370.
Amlin, N.M., Rood, S.B., 2002. Comparative tolerances of riparian willows and Chambers, J.C., Macmahon, J.A., Haefner, J.H., 1991. Seed entrapment in alpine
ecosystems – effects of soil particle-size and diaspore morphology. Ecology 72,
cottonwoods to water-table decline. Wetlands 22, 338–346.
1668–1677.
Amlin, N.M., Rood, S.B., 2003. Drought stress and recovery of riparian
Chow, V.T., 1959. Open-Channel Hydraulics. McGraw-Hill, New York, NY, 698 pp.
cottonwoods due to water table alteration along Willow Creek, Alberta. Trees-
Clark, J.S., Silman, M., Kern, R., Macklin, E., HilleRisLambers, J., 1999. Seed
ÙStructure and Function 17, 351–358.
dispersal near and far: patterns across temperate and tropical forests. Ecology
Andersson, E., Nilsson, C., Johansson, M.E., 2000. Plant dispersal in boreal rivers
80, 1475–1494.
and its relation to the diversity of riparian flora. Journal of Biogeography 27,
Clawson, R.G., Lockaby, B.G., Rummer, B., 2001. Changes in production and
1095–1106.
nutrient cycling across a wetness gradient within a floodplain forest. Ecosystems
Arcement, G.J., Schneider, V., 1989. Guide for selecting Manning’s roughness
4, 126–138.
coefficients for natural channels and flood plains. U.S. Geological Survey Water
Clifton, C., 1989. Effects of vegetation and land use change on channel
Supply Paper 2339, Washington, DC.
morphology. In: Gresswell, R.A., Barton, B.A., Kershner, J.L. (Eds.), Practical
Auble, G.T., Friedman, J.M., Scott, M.L., 1994. Relating riparian vegetation to
Approaches to Riparian Resource Management. BLM-MT-PT-89-001-4351.
present and future streamflows. Ecological Applications 4, 544–554. Bureau of Land Management, Billings, MT, pp. 121–129.
Auble, G.T., Scott, M.L., 1998. Fluvial disturbance patches and cottonwood Conner, W.H., McLeod, K.W., McCarron, J.K., 1997. Flooding and salinity effects on
recruitment along the upper Missouri River, Montana. Wetlands 18, growth and survival of four common forested wetland species. Wetlands Ecology
546–556. and Management 5, 99–109.
Baker, W.L., 1989. Macro- and micro-scale influences on riparian vegetation in Cooper, D.J., Andersen, D.C., Chimner, R.A., 2003. Multiple pathways for woody
western Colorado. Annals of the Association of American Geographers 79, plant establishment on floodplains at local to regional scales. Journal of
65–78. Ecology 91, 182–196.
Baker, W.L., Walford, G.M., 1995. Multiple stable states and models of riparian Cooper, D.J., Dickens, J., Hobbs, N.T., Christensen, L., Landrum, L., 2006.
vegetation succession on the Animas River, Colorado. Annals of the Association Hydrologic, geomorphic and climatic processes controlling willow establishment
of American Geographers 85, 320–338. in a montane ecosystem. Hydrological Processes 20, 1845–1864.
Bechtold, S.J., Naiman, R.J., 2006. Soil texture and nitrogen mineralization potential Cordes, L.D., Hughes, F.M.R., Getty, M., 1997. Factors Affecting the regeneration
across a riparian toposequence in a semi-arid savanna. Soil Biology and and distribution of Riparian Woodlands along a Northern Prairie River:
Biochemistry 38, 1325–1333. The Red Deer River, Alberta, Canada. Journal of Biogeography 24, 675–695.
Bechtold, J.S., Naiman, R.J., 2009. A quantitative model of soil organic matter Corenblit, D., Steiger, J., Gurnell, A.M., Naiman, R.J., 2009. Plants intertwine fluvial
accumulation during floodplain primary succession. Ecosystems 12, 1352–1368. landform dynamics with ecological succession and natural selection: a niche
Bell, D.T., 1974. Tree stratum composition and distribution in the streamside forest. construction perspective for riparian systems. Global Ecology and Biogeography
American Midland Naturalist 92, 35–46. 18, 507–520.
Bendix, J., 1994a. Scale, direction, and pattern in riparian vegetation–environ- Corenblit, D., Tabacchi, E., Steiger, J., Gurnell, A.M., 2007. Reciprocal interactions and
ment relationships. Annals of the Association of American Geographers 84, adjustments between fluvial landforms and vegetation dynamics in river corridors: a
652–665. review of complementary approaches. Earth-Science Reviews 84, 56–86.
Bendix, J., 1994b. Among-site variation in riparian vegetation of the southern Cowan, W.L., 1956. Estimating hydraulic roughness coefficients. Agricultural
California transverse ranges. American Midland Naturalist 132, 136–151. Engineering 37, 473–475.
Bendix, J., 1997. Flood disturbance and the distribution of riparian species Cramer, V.A., Hobbs, R.J., 2002. Ecological consequences of altered hydrological
diversity. Geographical Review 87, 468–483. regimes in fragmented ecosystems in southern Australia: impacts and possible
Bendix, J., 1998. Impact of a flood on southern Californian riparian vegetation. management responses. Austral Ecology 27, 546–564.
Physical Geography 19, 162–174. Crawford, R.M.M., 1996. Whole plant adaptations to fluctuating water tables. Folia
Bendix, J., 1999. Stream power influence on southern Californian riparian Geobotanica and Phytotaxonomica 31, 7–24.
vegetation. Journal of Vegetation Science 10, 243–252. Danvind, M., Nilsson, C., 1997. Seed floating ability and distribution of alpine
Bendix, J., Cowell, C.M., 2010. Fire, floods and woody debris: interactions between plants along a northern Swedish river. Journal of Vegetation Science 8,
biotic and geomorphic processes. Geomorphology 116, 297–304. 271–276.
70 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective

Darby, S.E., 1999. Effect of riparian vegetation on flow resistance and flood Gregory, S.V., Swanson, F.J., McKee, W.A., Cummins, K.W., 1991. An ecosystem
potential. Journal of Hydraulic Engineering 125, 443–454. perspective of riparian zones. Bioscience 41, 540–551.
Day, R.H., Doyle, T.W., Draugelis-Dale, R.O., 2006. Interactive effects of substrate, Groeneveld, D.P., French, R.H., 1995. Hydrodynamic control of an emergent
hydroperiod, and nutrients on seedling growth of Salix nigra and Taxodium aquatic plant (Scirpus acutus) in open channels. Water Resources Bulletin 31,
distichum. Environmental and Experimental Botany 55, 163–174. 505–514.
Devitt, D.A., Sala, A., Smith, S.D., Cleverly, J., Shaulis, L.K., Hammett, R., 1998. Gurnell, A., Thompson, K., Goodson, J., Moggridge, H., 2008. Propagule
Bowen ratio estimates of evapotranspiration for Tamarix ramosissima stands deposition along river margins: linking hydrology and ecology. Journal of
on the Virgin River in southern Nevada. Water Resources Research 34, Ecology 96, 553–565.
2407–2414. Gurnell, A.M., Boitsidis, A.J., Thompson, K., Clifford, N.J., 2006. Seed bank, seed
Di Tomaso, J.M., 1998. Impact, biology, and ecology of saltcedar (Tamarix spp.) in dispersal and vegetation cover: colonization along a newly-created river channel.
the southwestern United States. Weed Technology 12, 326–336. Journal of Vegetation Science 17, 665–674.
Dietrich, W.E., Whiting, P.J., 1989. Boundary shear stress and sediment transport in Gurnell, A.M., Petts, G.E., 2006. Trees as riparian engineers: the Tagliamento River,
river meanders of sand and gravel. In: Ikeda, S., Parker, G. (Eds.), River Italy. Earth Surface Processes and Landforms 31, 1558–1574.
Meandering, American Geophysical Union Water Resources Monograph, vol. 12, Gurnell, A.M., Petts, G.E., Hannah, D.M., et al., 2001. Riparian vegetation and
pp. 1–50. island formation along the gravel-bed Fiume Tagliamento, Italy. Earth Surface
Dixon, M.D., Turner, M.G., 2006. Simulated recruitment of riparian trees and shrubs Processes and Landforms 26, 31–62.
under natural and regulated flow regimes on the Wisconsin River, USA. River Gurnell, A.M., Piegay, H., Swanson, F.J., Gregory, S.V., 2002. Large wood and
Research and Applications 22, 1057–1083. fluvial processes. Freshwater Biology 47, 601–619.
Dixon, M.D., Turner, M.G., Jin, C., 2002. Riparian tree seedling distribution on Hack, J.T., Goodlett, J.C., 1960. Geomorphology and forest ecology of a mountain
Wisconsin river sandbars: controls at different spatial scales. Ecological region in the Central Appalachians. U.S. Geological Survey Professional Paper
Monographs 72, 465–485. 347, Washington, DC.
Donovan, L.A., Ehleringer, J.R., 1991. Ecophysiological differences among juvenile Hanson, P.J., Todd, Jr. D.E., Amthor, J.A., 2001. A six-year study of sapling and
and reproductive plants of several woody species. Oecologia 86, 594–597. large-tree growth and mortality responses to natural and induced variability in
Dosskey, M.G., Vidon, P., Gurwick, N.P., Allan, C.J., Duval, T.P., Lowrance, R., precipitation and throughfall. Tree Physiology 21, 345–358.
2010. The role of riparian vegetation in protecting and improving chemical water Harper, E.B., Stella, J.C., Fremier, A.K., 2011. Global sensitivity analysis for
quality in streams. Journal of the American Water Resources Association 46, complex ecological models: a case study of riparian cottonwood population
261–277. dynamics. Ecological Applications 21, 1225–1240.
Douhovnikoff, V., McBride, J.R., Dodd, R.S., 2005. Salix exigua clonal growth and Harris, R.R., 1987. Occurrence of vegetation on geomorphic surfaces in the active
population dynamics in relation to disturbance regime variation. Ecology 86, floodplain of a California alluvial stream. American Midland Naturalist 118,
446–452. 393–405.
Dwire, K.A., Kauffman, J.B., 2003. Fire and riparian ecosystems in landscapes of Hefley, H.M., 1937. Ecological studies on the Canadian River floodplain in
the western USA. Forest Ecology and Management 178, 61–74. Cleveland County, Oklahoma. Ecological Monographs 7, 346–402.
Eschenbach, C., Kappen, L., 1999. Leaf water relations of black alder [Alnus Hickin, E.J., 1984. Vegetation and river channel dynamics. Canadian Geographer
glutinosa (L.) Gaertn.]. Trees 14, 28–38. 28, 111–126.
Everitt, B.L., 1968. Use of the cottonwood in an investigation of the recent history Horton, J.L., Clark, J.L., 2001. Water table decline alters growth and survival of
of a flood plain. American Journal of Science 266, 417–539. Salix gooddingii and Tamarix chinensis seedlings. Forest Ecology and
Ewing, K., 1996. Tolerance of four wetland plant species to flooding and sediment Management 140, 239–247.
deposition. Environmental and Experimental Botany 36, 131–146. Howard, A.D., Dietrich, W.E., Seidl, M.A., 1994. Modeling fluvial erosion on
Fanshawe, D.B., 1954. Riparian vegetation in British Guiana. Journal of Ecology 42, regional to continental scales. Journal of Geophysical Research 99,
289–295. 13,971–13,986.
Fenner, P., Brady, W.W., Patton, D.R., 1985. Effects of regulated water flows on Huffman, R.T., 1980. The relation of flood timing and duration to variation in
regeneration of Fremont cottonwood. Journal of Range Management 38, selected bottomland hardwood communities of southern Arkansas.
135–138. Miscellaneous Paper EL-80-4, U.S. Army Engineer Waterways Experiment
Fetherston, K.L., Naiman, R.J., Bilby, R.E., 1995. Large woody debris, physical Station, Vicksburg, Mississippi.
process, and riparian forest development in montane river networks of the Hughes, F.M.R., 1997. Floodplain biogeomorphology. Progress in Physical
Pacific Northwest. Geomorphology 13, 133–144. Geography 21, 501–529.
Francis, R.A., Gurnell, A.M., 2006. Initial establishment of vegetative fragments Hughes, F.M.R., Adams, W.M., Muller, E., et al., 2001. The importance of different
within the active zone of a braided gravel-bed river (River Tagliamento, NE, scale processes for the restoration of floodplain woodlands. Regulated Rivers –
Italy). Wetlands 26, 641–648. Research and Management 17, 325–345.
Friedman, J.M., Osterkamp, W.R., Lewis, W.M., 1996a. Channel narrowing Hughes, F.M.R., Barsoum, N., Richards, K.S., Winfield, M., Hayes, A., 2000. The
and vegetation development following a Great Plains flood. Ecology 77, response of male and female black poplar (Populus nigra L. subspecies
2167–2181. betulifolia (Pursh) W. Wettst) cuttings to different water table depths and
Friedman, J.M., Osterkamp, W.R., Lewis, Jr. W.M., 1996b. The role of vegetation sediment types: implications for flow management and river corridor biodiversity.
and bed-level fluctuations in the process of channel narrowing. Geomorphology Hydrological Processes 14, 3075–3098.
14, 341–351. Hughes, F.M.R., Harris, T., Richards, K., et al., 1997. Woody riparian species
Frye, II R.J., Quinn, J.A., 1979. Forest development in relation to topography and response to different soil moisture conditions: laboratory experiments on
soils on a floodplain of the Raritan River, New Jersey. Bulletin of the Torrey Alnus incana (L.) Moench. Global Ecology and Biogeography Letters 6,
Botanical Club 106, 334–345. 247–256.
Gasith, A., Resh, V.H., 1999. Streams in Mediterranean climate regions: abiotic Hupp, C.R., 1982. Stream-grade variation and riparian-forest ecology along Passage
influences and biotic responses to predictable seasonal events. Annual Review Creek, Virginia. Bulletin of the Torrey Botanical Club 109, 488–499.
of Ecology and Systematics 30, 51–81. Hupp, C.R., 1988. Plant ecological aspects of flood geomorphology and paleoflood
Glenn, E., Tanner, R., Mendez, S., Kehret, T., Moore, D., Garcia, J., Valdes, C., history. In: Baker, V.R., Kochel, R.C., Patton, P.C. (Eds.), Flood Geomorphology.
1998. Growth rates, salt tolerance and water use characteristics of native and Wiley, New York, NY, pp. 335–356.
invasive riparian plants from the delta of the Colorado River, Mexico. Journal of Hupp, C.R., 1992. Riparian vegetation recovery patterns following stream
Arid Environments 40, 281–294. channelization: a geomorphic perspective. Ecology 73, 1209–1226.
Graf, W.L., 1978. Fluvial adjustments to the spread of tamarisk in the Colorado Hupp, C.R., 2000. Hydrology, geomorphology and vegetation of Coastal Plain rivers
Plateau region. Geological Society of America Bulletin 89, 1491–1501. in the south-easthern USA. Hydrological Processes 14, 2991–3010.
Gran, K., Paola, C., 2001. Riparian vegetation controls on braided stream dynamics. Hupp, C.R., Bornette, G., 2003. Vegetation, fluvial processes and landforms in
Water Resources Research 37, 3275–3283. temperate areas. In: Piegay, H., Kondolf, M. (Eds.), Tools in Geomorphology.
Greco, S.E., Fremier, A.K., Larsen, E.W., Plant, R.E., 2007. A tool for tracking Wiley, Chichester, pp. 269–288.
floodplain age land surface patterns on a large meandering river with Hupp, C.R., Osterkamp, W.R., 1985. Bottomland vegetation distribution
applications for ecological planning and restoration design. Landscape and along Passage Creek, Virginia, in relation to fluvial landforms. Ecology 66,
Urban Planning 81, 354–373. 670–681.
 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective 71

Hupp, C.R., Rinaldi, M., 2007. Riparian vegetation patterns and diversity in relation manipulating water and sediment flows. Journal of Arid Environments 49,
to fluvial landforms and channel evolution along selected rivers of Tuscany 111–131.
(central Italy). Annals of the American Association of Geographers 97, 12–30. Levine, J.M., 2003. A patch modeling approach to the community-level
Illichevsky, S., 1933. The river as a factor of plant distribution. Journal of Ecology consequences of directional dispersal. Ecology 84, 1215–1224.
21, 436–441. Li, S., Martin, L.T., Pezeshki, S.R., Shields, F.D., 2005. Responses of black willow
Istanbulluoglu, E., Bras, R.L., 2005. Vegetation-modulated landscape evolution: (Salix nigra) cuttings to simulated herbivory and flooding. Acta Oecologica 28,
effects of vegetation on landscape processes, drainage density, and topography. 173–180.
Journal of Geophysical Research 110, F02012. http://dx.doi.org/10.1029/ Liébault, F., Piégay, H., 2001. Assessment of channel changes due to long
2004JF000249. term bedload supply decrease, Roubion River, France. Geomorphology 36,
Jansson, R., Nilsson, C., Dynesius, M., Andersson, E., 2000. Effects of river 167–186.
regulation on river-margin vegetation: a comparison of eight boreal rivers. Liébault, F., Piégay, H., 2002. Causes of 20th century channel narrowing in
Ecological Applications 10, 203–224. mountain and piedmont rivers of southeastern France. Earth Surface Processes
Jeffries, R., Darby, S.E., Sear, D.A., 2003. The influence of vegetation and organic and Landforms 27, 425–444.
debris on flood-plain sediment dynamics: case study of a low-order stream in Lightbody, A., Rominger, J.T., Nepf, H.M., Paola, C., 2008. Effect of emergent
the New Forest, England. Geomorphology 51, 61–80. aquatic vegetation on sediment transport within the St. Anthony Falls Laboratory
Johansson, M.E., Nilsson, C., Nilsson, E., 1996. Do rivers function as corridors for Outdoor StreamLab. Eos Transactions of AGU 89(53), Fall Meeting Supplement,
plant dispersal? Journal of Vegetation Science 7, 593–598. Abstract H32C-01.
Johnson, W.C., 1992. Dams and riparian forests: case study from the upper Lightbody, A.F., Nepf, H.M., 2006. Prediction of velocity profiles and longitudinal
Missouri River. Rivers 3, 229–242. dispersion in emergent salt marsh vegetation. Limnology Oceanography 51,
Johnson, W.C., 1994. Woodland expansion in the Platte River, Nebraska: patterns 218–228.
and causes. Ecological Monographs 64, 45–84. Lindsey, A.A., Petty, R.O., Sterling, D.K., Van Asdall, W., 1961. Vegetation and
Johnson, W.C., 1998. Adjustment of riparian vegetation to river regulation in the environment along the Wabash and Tippecanoe rivers. Ecological Monographs
Great Plains, USA. Wetlands 18, 608–618. 31, 105–156.
Johnson, W.C., Burgess, R.L., Keammerer, W.R., 1976. Forest overstory vegetation Lopez, O.R., 2001. Seed flotation and postflooding germination in tropical terra
and environment on the Missouri River floodplain in North Dakota. Ecological firme and seasonally flooded forest species. Functional Ecology 15, 763–771.
Monographs 46, 59–84. Lytle, D.A., Merritt, D.M., 2004. Hydrologic regimes and riparian forests: a
Kalliola, R., Salo, J., Puhakka, M., Rajasilta, M., 1991. New site formation and structured population model for cottonwood. Ecology 85, 2493–2503.
colonizing vegetation in primary succession on the western Amazon floodplains. Lytle, D.A., Poff, N.L., 2004. Adaptation to natural flow regimes. Trends in Ecology
Journal of Ecology 79, 877–901. and Evolution 19, 94–100.
Karrenberg, S., Edwards, P.J., Kollmann, J., 2002. The life history of Salicaceae Mahoney, J.M., Rood, S.B., 1992. Response of a hybrid poplar to water-table decline
living in the active zone of floodplains. Freshwater Biology 47, 733–748. in different substrates. Forest Ecology and Management 54, 141–156.
Keller, E.A., Swanson, F.J., 1979. Effects of large organic material on channel form Mahoney, J.M., Rood, S.B., 1998. Streamflow requirements for cottonwood seedling
and fluvial processes. Earth Surface Processes 4, 361–380. recruitment – an integrative model. Wetlands 18, 634–645.
Knighton, D., 1984. Fluvial Forms and Processes. Edward Arnold, London, Manning, R., 1891. On the flow of water in open channels and pipes. Transactions
219 pp. of Institution of Civil Engineers of Ireland 20, 161–207.
Kondolf, G.M., Piegay, H., Landon, N., 2007. Changes in the riparian zone of Martin, C.W., Johnson, W.C., 1987. Historical channel narrowing and riparian
the lower Eygues River, France, since 1830. Landscape Ecology 22, vegetation expansion in the Medicine Lodge River Basin, Kansas, 1871–1983.
367–384. Annals of the Association of American Geographers 77, 436–449.
Kovalchik, B.L., Chitwood, L.A., 1990. Use of geomorphology in the classification Martin, Y., Church, M., 2000. The effect of riparian tree roots on the mass-stability
of riparian plant associations in mountainous landscapes of central Oregon. of riverbanks. Earth Surface Processes and Landforms 25, 921–937.
U.S.A. Forest Ecology and Management 33, 405–418. McBride, J.R., Strahan, J., 1984a. Establishment and survival of woody riparian
Kozlowski, T.T., 2002. Physiological–ecological impacts of flooding on riparian species on gravel bars of an intermittent stream. American Midland Naturalist
forest ecosystems. Wetlands 22, 550–561. 112, 235–245.
Kozlowski, T.T., Pallardy, S.G., 2002. Acclimation and adaptive responses of woody McBride, J.R., Strahan, J., 1984b. Fluvial processes and woodland succession
plants to environmental stresses. Botanical Review 68, 270–334. along Dry Creek, Sonoma County, California. In: Warner, R.E., Hendrix, K.M.
Kranjcec, J., Mahoney, J.M., Rood, S.B., 1998. The responses of three riparian (Eds.), California Riparian Systems. University of California Press, Berkeley, CA,
cottonwood species to water table decline. Forest Ecology and Management 110, pp. 110–119.
77–87. McKenney, R., Jacobson, R.B., Wertheimer, R.C., 1995. Woody vegetation and
Kubitzki, K., Ziburski, A., 1994. Seed Dispersal in flood-plain forests of Amazonia. channel morphogenesis in low-gradient, gravel-bed streams in the Ozark
Biotropica 26, 30–43. Plateaus, Missouri and Arkansas. Geomorphology 13, 175–198.
Labbe, J.M., Hadley, K.S., Schipper, A.M., Leuven, R.S.E.W., Gardiner, C.P., 2011. Megonigal, J.P., Conner, W.H., Kroeger, S., Sharitz, R.R., 1997. Aboveground
Influence of bank materials, bed sediment, and riparian vegetation on production in Southeastern floodplain forests: a test of the subsidy-stress
channel form along a gravel-to-sand transition reach of the Upper Tualatin River, hypothesis. Ecology 78, 370–384.
Oregon, USA. Geomorphology 125, 374–382. Merritt, D.M., Wohl, E.E., 2002. Processes governing hydrochory along rivers:
Lassettre, N.S., Piégay, H., Dufour, S., Rollet, A.J., 2008. Decadal changes in hydraulics, hydrology, and dispersal phenology. Ecological Applications 12,
distribution and frequency of wood in a free meandering river, the Ain River, 1071–1087.
France. Earth Surface Processes and Landforms 33, 1098–1112. Michalková, M., Piégay, H., Kondolf, G.M., Greco, S.E., 2011. Lateral erosion of the
Latterell, J.J., Bechtold, S., O’Keefe, T.C., Van Pelt, R., Naiman, R.J., 2006. Dynamic Sacramento River, California (1942–1999), and responses of channel and
patch mosaics and channel movement in an unconfined river valley of the floodplain lake to human influences. Earth Surface Processes and Landforms 36,
Olympic Mountains. Freshwater Biology 51, 523–544. 257–272.
Leffler, A.J., England, L.E., Naito, J., 2000. Vulnerability of fremont cottonwood Micheli, E.R., Kirchner, J.W., Larsen, E.W., 2004. Quantifying the effect of riparian
(Populus fremontii Wats.) individuals to xylem cavitation. Western North forest versus agricultural vegetation on river meander migration rates, central
American Naturalist 60, 204–210. Sacramento River, California, USA. River Research and Applications 20,
Leffler, A.J., Evans, A.S., 1999. Variation in carbon isotope composition 537–548.
among years in the riparian tree Populus fremontii. Oecologia 119, Micheli, E.R., Larsen, E.W., 2011. River channel cutoff dynamics, Sacramento River,
311–319. California, USA. River Research and Applications 27, 328–344.
Leopold, L.B., Maddock, T., 1953. The hydraulic geometry of stream channels and Millennium Ecosystem Assessment, 2005. Ecosystems and Human Well-Being:
some physiographic implications. U.S. Geological Survey Professional Paper Biodiversity Synthesis. World Resources Institute, Washington DC, 86 pp.
252, Washington, DC, pp. 39–85. Mitsch, W.J., Gosselink, J.G., 2007. Wetlands, Fourth ed. Wiley, New York, NY, 582
Leopold, L.B., Wolman, M.G., Miller, J.P., 1964. Fluvial Processes in pp.
Geomorphology. W.H. Freeman and Company, San Francisco, 522 pp. Mitsch, W.J., Taylor, J.R., Benson, K.B., 1991. Estimating primary productivity of
Levine, C.M., Stromberg, J.C., 2001. Effects of flooding on native and exotic plant forested wetland communities in different hydrologic landscapes. Landscape
seedlings: implications for restoring south-western riparian forests by Ecology 5, 75–92.
72 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective

Montgomery, D.R., Buffington, J.M., 1997. Channel–reach morphology in mountain Petty, A.M., Douglas, M.M., 2010. Scale relationships and linkages between woody
drainage basins. Geological Society of America Bulletin 109, 596–611. vegetation communities along a large tropical floodplain river, north Australia.
Montgomery, D.R., Buffington, J.M., Smith, R.D., Schmidt, K.M., Pess, G., 1995. Journal of Tropical Ecology 26, 79–92.
Pool spacing in forest channels. Water Resources Research 31, 1097–1105. Pollen, N., Simon, A., 2005. Estimating the mechanical effects of riparian vegetation
Murray, A.B., Knaapen, M.A.F., Tal, M., Kirwan, M.L., 2008. Biomorphodynamics: on stream bank stability using a fiber bundle model. Water Resources Research
physical–biological feedbacks that shape landscapes. Water Resources Research 41, 1–11.
44, W11301. Polzin, M.L., Rood, S.B., 2006. Effective disturbance: seedling safe sites and patch
Murray, A.B., Paola, C., 1994. A cellular model of braided rivers. Nature 371, recruitment of riparian cottonwoods after a major flood of a mountain river.
54–57. Wetlands 26, 965–980.
Murray, A.B., Paola, C., 1997. Properties of a cellular braided stream model. Earth Renofalt, B.M., Nilsson, C., 2008. Landscape scale effects of disturbance on
Surface Processes and Landforms 22, 1001–1025. riparian vegetation. Freshwater Biology 53, 2244–2255.
Naiman, R.J., Bechtold, J.S., Beechie, T.J., Latterell, J.J., van Pelt, R., 2010. A Roberts, J., 2000. The influence of physical and physiological characteristics of
process-based view of floodplain forest patterns in coastal river valleys of the vegetation on their hydrological response. Hydrological Processes 14,
Pacific Northwest. Ecosystems 13, 1–31. 2885–2901.
Naiman, R.J., Decamps, H., 1997. The ecology of interfaces: riparian zones. Annual Robertson, A.I., Bacon, P., Heagney, G., 2001. The responses of floodplain primary
Review of Ecology and Systematics 28, 621–658. production to flood frequency and timing. Journal of Applied Ecology 38, 126–136.
Naiman, R.J., Decamps, H., McClain, M.E., 2005. Riparia: Ecology, Conservation, Robertson, P.A., Weaver, G.T., Cavanaugh, J.A., 1978. Vegetation and tree species
and Management of Streamside Communities. Elsevier, Amsterdam. near the northern terminus of the southern floodplain forest. Ecological
Nakamura, F., Jitsu, M., Kameyama, S., Mizugaki, S., 2002. Changes in riparian Monographs 48, 249–267.
forests in the Kushiro Mire, Japan, associated with stream channelization. River Rodrı́guez-González, P.M., Stella, J.C., Campelo, F., Ferreira, M.T., Albuquerque, A.,
Research and Applications 18, 65–79. 2010. Subsidy or stress? Tree structure and growth in wetland forests along a
Nanson, G.C., Beach, H.F., 1977. Forest succession and sedimentation on a hydrological gradient in Southern Europe. Forest Ecology and Management 259,
meandering-river floodplain, northeast British Columbia, Canada. Journal of 2015–2025.
Biogeography 4, 229–251. Rood, S.B., Braatne, J.H., Hughes, F.M.R., 2003. Ecophysiology of riparian
Nepf, H.M., 1999. Drag, turbulence, and diffusion in flow through emergent cottonwoods: stream flow dependency, water relations and restoration. Tree
vegetation. Water Resources Research 35, 479–489. Physiology 23, 1113–1124.
Niiyama, K., 1990. The role of seed dispersal and seedling traits in colonization Rood, S.B., Mahoney, J.M., 1990. Collapse of riparian poplar forests downstream
and coexistence of Salix species in seasonally flooded habitat. Ecological from dams in western prairies: probable causes and prospects for mitigation.
Research 5, 317–331. Environmental Management 14, 451–464.
Nilsson, C., Andersson, E., Merritt, D.M., Johansson, M.E., 2002. Differences in Rood, S.B., Taboulchanas, K., Bradley, C.E., Kalischuk, A.R., 1999. Influence of flow
riparian flora between riverbanks and river lakeshores explained by dispersal regulation on channel dynamics and riparian cottonwoods along the Bow River,
traits. Ecology 83, 2878–2887. Alberta. Rivers 7, 33–48.
Nilsson, C., Jansson, R., 1995. Floristic differences between riparian corridors of Rood, S.B., Zanewich, K., Stefura, C., Mahoney, J.M., 2000. Influence of water
regulated and free-flowing boreal rivers. Regulated Rivers: Research and table decline on growth allocation and endogenous gibberellins in black
Management 11, 55–66. cottonwood. Tree Physiology 20, 831–836.
Nixon, E.S., Willett, L., Cox, P.W., 1977. Woody vegetation of a virgin forest in an Sandercock, P.J., Hooke, J.M., 2010. Assessment of vegetation effects on hydraulics
eastern Texas river bottom. Castanea 42, 227–236. and of feedbacks on plant survival and zonation in ephemeral channels.
Nzunda, E.F., Griffiths, M.E., Lawes, M.J., 2007. Multi-stemmed trees in subtropical Hydrological Processes 24, 695–713.
coastal dune forest: survival strategy in response to chronic disturbance. Journal Sandercock, P.J., Hooke, J.M., Mant, J.M., 2007. Vegetation in dryland river
of Vegetation Science 18, 693–700. channels and its interaction with fluvial processes. Progress in Physical
Opperman, J.J., Meleason, M., Francis, R.A., Davies-Colley, R., 2008. "Livewood": Geography 31, 107–129.
geomorphic and ecological functions of living trees in river channels. Bioscience Schneider, R.L., Sharitz, R.R., 1988. Hydrochory and regeneration in a bald
58, 1069–1078. cypress-water tupelo swamp forest. Ecology 69, 1055–1063.
Opperman, J.J., Merenlender, A.M., 2007. Living trees provide stable large wood in Scott, M.L., Friedman, J.M., Auble, G.T., 1996. Fluvial processes and the
streams. Earth Surface Processes and Landforms 32, 1229–1238. establishment of bottomland trees. Geomorphology 14, 327–339.
Osterkamp, W.R., Hupp, C.R., 1984. Geomorphic and vegetative characteristics Scott, M.L., Shafroth, P.B., Auble, G.T., 1999. Responses of riparian cottonwoods
along three northern Virginia streams. Geological Society of America Bulletin 95, to alluvial water table declines. Environmental Management 23, 347–358.
1093–1101. Segelquist, C.A., Scott, M.L., Auble, G.T., 1993. Establishment of Populus deltoides
Osterkamp, W.R., Hupp, C.R., 2010. Fluvial processes and vegetation – glimpses under simulated alluvial groundwater decline. American Midland Naturalist 130,
of the past, the present, and perhaps the future. Geomorphology 116, 274–285.
274–285. Shafroth, P.B., Auble, G.T., Stromberg, J.C., Patten, D.T., 1998. Establishment of
Parker, K.C., Bendix, J., 1996. Landscape-scale geomorphic influences on woody riparian vegetation in relation to annual patterns of streamflow, Bill
vegetation patterns in four environments. Physical Geography 17, 113–141. Williams River, Arizona. Wetlands 18, 577–590.
Parker, K.C., Hamrick, J.L., 1992. Genetic diversity and clonal structure in a Shafroth, P.B., Friedman, J.M., Ischinger, L.S., 1995. Effects of salinity on
columnar cactus, Lophocereus schottii. American Journal of Botany 79, 86–96. establishment of Populus fremontii (cottonwood) and Tamarix ramosissima
Pereira, J.S., Kozlowski, T.T., 1977. Variations among woody angiosperms in (saltcedar) in Southwestern United States. Great Basin Naturalist 55, 58–65.
response to flooding. Physiologia Plantarum 41, 184–192. Shafroth, P.B., Wilcox, A.C., Lytle, D.A., et al., 2010. Ecosystem effects of
Perucca, E., Camporeale, C., Ridolfi, L., 2007. Significance of the riparian environmental flows: modelling and experimental floods in a dryland river.
vegetation dynamics on meandering river morphodynamics. Water Resources Freshwater Biology 55, 68–85.
Research 43, W03430. Sharitz, R.R., Mitsch, W.J., 1993. Southern floodplain forests. In: Martin, W.H.,
Peterson, D.H., Smith, R.E., Dettinger, M.D., Cayan, D.R., Riddle, L., 2000. An Boyce, S.G., Echtemacht, A.C.E. (Eds.), Biodiversity of the Southeastern
organized signal in snowmelt runoff over the western United States. Journal of United States: Lowland Terrestrial Communities. Wiley, New York, NY, pp.
the American Water Resources Association 36, 421–432. 311–372.
Pettit, N.E., Froend, R.H., 2001. Availability of seed for recruitment of riparian Shelford, V.E., 1954. Some lower Mississippi Valley flood plain biotic communities:
vegetation: a comparison of a tropical and a temperate river ecosystem in their age and elevation. Ecology 35, 126–142.
Australia. Australian Journal of Botany 49, 515–528. Sher, A.A., Marshall, D.L., Gilbert, S.A., 2000. Competition between native
Pettit, N.E., Latterell, J.J., Naiman, R.J., 2006. Formation, distribution and Populus deltoides and invasive Tamarix ramosissima and the implications
ecological consequences of flood-related wood debris piles in a bedrock for reestablishing flooding disturbance. Conservation Biology 14, 1744–1754.
confined river in semi-arid South Africa. River Research and Applications 22, Sher, A.A., Marshall, D.L., Taylor, J.P., 2002. Establishment patterns of native
1097–1110. Populus and Salix in the presence of invasive nonnative Tamarix. Ecological
Pettit, N.E., Naiman, R.J., 2006. Flood-deposited wood creates regeneration niches Applications 12, 760–772.
for riparian vegetation on a semi-arid South African river. Journal of Vegetation Shin, N., Nakamura, F., 2005. Effects of fluvial geomorphology on riparian tree
Science 17, 615–624. species in Rekifune River, northern Japan. Plant Ecology 178, 15–28.
 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective 73

Siegel, R.S., Brock, J.H., 1990. Germination requirements of key southwestern Toner, M., Keddy, P., 1997. River hydrology and riparian wetlands: a
woody riparian species. Desert Plants 10, 3–8. predictive model for ecological assembly. Ecological Applications 7,
Sigafoos, R.S., 1961. Vegetation in relation to flood frequency near Washington, 236–246.
D.C. U.S. Geological Survey Professional Paper 424-C, Washington, DC, pp. Trush, W.J., McBain, S.M., Leopold, L.B., 2000. Attributes of an alluvial river and
C248–C250. their relation to water policy and management. PNAS 97, 11858–11863.
Sigafoos, R.S., 1964. Botanical evidence of floods and flood-plain deposition. U.S. Tucker, G.E., Bras, R.L., 1999, Dynamics of vegetation and runoff erosion, U. S.
Geological Survey Professional Paper 485-A, Washington, DC. Army Corps of Engineers. Construction Engineering Research Laboratory,
Simon, A., Collison, A.J.C., 2002. Quantifying the mechanical and hydrologic Champaign, Ill.
effects of riparian vegetation on streambank stability. Earth Surface Processes Tucker, G.E., Lancaster, S.T., Gasparini, N.M., Bras, R.L., 2001. The
and Landforms 27, 527–546. channel–hillslope integrated landscape development model. In: Harmon, R.S.,
Smedley, M.P., Dawson, T.E., Comstock, J.P., Donovan, L.A., Sherrill, D.E., Cook, Dow, W.W. (Eds.), Landscape Erosion and Evolution Modeling. Kluwer, Norwell,
C.S., Ehleringer, J.R., 1991. Seasonal carbon isotope discrimination in a MA, pp. 349–384.
grassland community. Oecologia (Berlin) 85, 314–320. Tyree, M.T., Kolb, K.J., Rood, S.B., Patino, S., 1994. Vulnerability to drought-
Stallins, J.A., 2006. Geomorphology and ecology: unifying themes for complex induced cavitation of riparian cottonwoods in Alberta – a possible factor in the
systems in biogeomorphology. Geomorphology 77, 207–216. decline of the ecosystem. Tree Physiology 14, 455–466.
Steiger, J., Tabacchi, E., Dufour, S., Corenblit, D., Peiry, J.L., 2005. Van Cleve, K., Viereck, L.A., Dyrness, C.T., 1996. State factor control of soils and
Hydrogeomorphic processes affecting riparian habitat within alluvial forest succession along the Tanana River in interior Alaska, USA. Arctic and
channel–floodplain river systems: a review for the temperate zone. River Alpine Research 28, 388–400.
Research and Applications 21, 719–737. van Eck, W.H.J.M., van de Steeg, H.M., Blom, C.W.P.M., de Kroon, H., 2004. Is
Stella, J., Hayden, M., Battles, J., Piégay, H., Dufour, S., Fremier, A., 2011. The role tolerance to summer flooding correlated with distribution patterns in river
of abandoned channels as refugia for sustaining pioneer riparian forest floodplains? A comparative study of 20 terrestrial grassland species. Oikos 107,
ecosystems. Ecosystems 14, 776–790. 393–405.
Stella, J.C., 2005. A field-calibrated model of pioneer riparian tree recruitment for Van Pelt, R., O’Keefe, T.C., Latterell, J.J., Naiman, R.J., 2006. Riparian forest stand
the San Joaquin Basin, CA. PhD dissertation. University of California, Berkeley.
development along the Queets River in Olympic National Park, Washington.
Stella, J.C., Battles, J.J., 2010. How do riparian woody seedlings survive seasonal
Ecological Monographs 76, 277–298.
drought? Oecologia 164, 579–590.
Walls, R.L., Wardrop, D.H., Brooks, R.P., 2005. The impact of experimental
Stella, J.C., Battles, J.J., McBride, J.R., Orr, B.K., 2010. Riparian seedling mortality
sedimentation and flooding on the growth and germination of floodplain trees.
from simulated water table recession, and the design of sustainable flow
Plant Ecology 176, 203–213.
regimes on regulated rivers. Restoration Ecology 18, 284–294.
Walter, R.C., Merritts, D.J., 2008. Natural streams and the legacy of water-powered
Stella, J.C., Battles, J.J., Orr, B.K., McBride, J.R., 2006. Synchrony of seed
mills. Science 319, 299–304.
dispersal, hydrology and local climate in a semi-arid river reach in California.
Wang, S.-C., Jurik, T.W., van der Valk, A.G., 1994. Effects of sediment load on
Ecosystems 9, 1200–1214.
various stages in the life and death of cattail (Typha X glauca). Wetlands 14,
Stone, B.M., Shen, H.T., 2002. Hydraulic resistance of flow in channels with
cylindrical roughness. Journal of Hydraulic Engineering 128, 500–506. 166–173.
Stromberg, J.C., Lite, S.J., Dixon, M.D., 2010. Effects of stream flow patterns on Ware, G.H., Penfound, W.T., 1949. The vegetation of the lower levels of the
riparian vegetation of a semiarid river: implications for a changing climate. River floodplain of the South Canadian River in Central Oklahoma. Ecology 30,
Research and Applications 26, 712–729. 478–484.
Stromberg, J.C., Patten, D.T., 1996. Instream flow and cottonwood growth in the Wistendahl, W.A., 1958. The flood plain of the Raritan River, New Jersey. Ecological
eastern Sierra Nevada of California. U. S. A. Regulated Rivers: Research and Monographs 28, 129–153.
Management 12, 1–12. Yager, E.M., Schmeeckle, M.W., 2007. The influence of emergent vegetation on
Stromberg, J.C., Patten, D.T., Richter, B.D., 1991. Flood flows and dynamics of turbulence and sediment transport in rivers. In: Boyer, D., Alexandrova, O.
Sonoran riparian forests. Rivers 2, 221–235. (Eds.), Proceedings of the 5th IAHR International Symposium on Environmental
Tabacchi, E., Lambs, L., Guilloy, H., Planty-Tabacchi, A.M., Muller, E., Decamps, H., Hydraulics, pp. 69–74.
2000. Impacts of riparian vegetation on hydrological processes. Hydrological Yanosky, T.M., 1982. Effects of flooding upon woody vegetation along parts of the
Processes 14, 2959–2976. Potomac River flood plain. U.S. Geological Survey Professional Paper 1206,
Tal, M., Paola, C., 2007. Dynamic single-thread channels maintained by the Washington, DC.
interaction of flow and vegetation. Geology 35, 347–350. Young, J.A., Clements, C.D., 2003a. Seed germination of willow species from a
Tal, M., Paola, C., 2010. Effects of vegetation on channel morphodynamics: results desert riparian ecosystem. Journal of Range Management 56, 496–500.
and insights from laboratory experiments. Earth Surface Processes and Young, J.A., Clements, C.D., 2003b. Germination of seeds of Fremont cottonwood.
Landforms 35, 1014–1028. Journal of Range Management 56, 660–664.
Tardif, J., Bergeron, Y., 1999. Population dynamics of Fraxinus nigra in response to Zhang, X.L., Zang, R.G., Li, C.Y., 2004. Population differences in physiological and
flood-level variations, in Northwestern Quebec. Ecological Monographs 69, morphological adaptations of Populus davidiana seedlings in response to
107–125. progressive drought stress. Plant Science 166, 791–797.
Teversham, J.M., Slaymaker, O., 1976. Vegetation composition in relation to Zimmerman, R.C., 1969. Plant ecology of an arid basin, Tres Alamos-Redington
flood frequency in Lillooet River valley, British Columbia. Catena 3, area southeastern Arizona. U.S. Geological Survey Professional Paper 485-D,
191–201. Washington, DC, pp. 39–85.

Biographical Sketch

Jacob Bendix is associate professor of geography in the Maxwell School at Syracuse University, and adjunct
associate professor of environmental forest biology at the State University of New York College of Environmental
Science and Forestry (SUNY-ESF). His research interests focus on fluvial geomorphology, disturbance ecology, and
riparian environments. He earned his BA from the University of California, Berkeley (geography), his MS from the
University of Wisconsin–Madison (geography, fluvial geomorphology emphasis), and his PhD from the Uni-
versity of Georgia (geography, biogeography emphasis).
74 Riparian Vegetation and the Fluvial Environment: A Biogeographic Perspective

John C Stella is assistant professor of forest and natural resources management at the State University of New York
College of Environmental Science and Forestry (SUNY-ESF), and holds an adjunct appointment (geography) in
the Maxwell School at Syracuse University. His research interests focus on riparian and stream ecology, den-
droecology, plant ecohydrology and stable isotope biogeochemistry, and restoration ecology. He earned his BA
from Yale University (architecture), and his MS and PhD from the University of California, Berkeley (environ-
mental science, policy and management). His research sites are located in semi-arid regions of California and the
US Southwest, Mediterranean Europe, and the Adirondack mountains of New York.
 12.6 The Impacts of Vegetation on Roughness in Fluvial Systems
WC Hession, Virginia Tech, Blacksburg, VA, USA
JC Curran, University of Virginia, Charlottesville, VA, USA
r 2013 Elsevier Inc. All rights reserved.

12.6.1 Introduction 76
12.6.2 In-Stream Emergent Vegetation 77
12.6.2.1 Reach-Scale Impacts of Emergent Vegetation 78
12.6.2.2 Hydraulics and Turbulence 78
12.6.2.3 Emergent Vegetation and Sediment Transport 79
12.6.3 In-Stream Submerged Vegetation 80
12.6.3.1 Reach-Scale Impacts of Submerged Vegetation 80
12.6.3.2 Hydraulics and Turbulence 82
12.6.3.3 Submerged Vegetation and Sediment Transport 83
12.6.4 Streambank Vegetation 83
12.6.4.1 Reach-Scale Impacts of Streambank Vegetation 84
12.6.4.2 Hydraulics and Turbulence 84
12.6.4.3 Streambank Vegetation and Sediment Transport 85
12.6.5 Floodplain Vegetation 85
12.6.5.1 Reach-Scale Impacts of Floodplain Vegetation 85
12.6.5.2 Hydraulics and Turbulence 86
12.6.5.3 Floodplain Vegetation and Sediment Transport 88
12.6.6 Future Directions 88
References 89

Glossary measuring the Doppler shift of laser light scattered by the

Effective plant height The mean plant height after flexible particles.
plants have been deflected by the flow. Log law equation The equation describing the
Einstein transport equation TA stochastic equation of logarithmic form of the velocity profile. The exact form of
sediment transport derived from probabilities of sediment the equation is dependent on the characteristics of the flow.
movement in a channel. The equation predicts the total Meyer-Peter and Mueller transport equation A widely
sediment load. used sediment transport equation that predicts the rate of
Fronds The large, divided leaf of the plant that extended sediment transport through a channel as a function of the
from the plant stem. amount of shear stress acting on the channel boundaries
Horseshoe vortex A vortex characterized by a horseshoe that is in excess of the minimum shear stress required to
morphology. The vortex consists of a core vortex and two initiate sediment movement.
vortices trailing in the stream wise direction at a 45 angle to Particle image velocimetry (PIV) A tool used to quantify
the bed. the velocity field in a fluid. Application of PIV involves
j  e turbulence model A two equation turbulence pulsing a laser light sheet in a flow that has been seeded
model that accounts for both the turbulent kinetic with micron-size particles. The light sheet is recorded by
energy and the turbulent dissipation in the flow. The video camera or photographed and the displacement of
turbulent dissipation measures the scale of the turbulence seeded particles between sequential images is measured to
in the flow. determine the movement of the particles with the flow.
Kelvin-Helmholtz instability Instability between two With a known time between sequential images, the velocity
fluid masses as a result of velocity shear between the fluids. of the particles can be calculated.
Laser doppler velocimetry (LDV) An optical technique Reynolds stress The mean forces per unit area
to measure the local velocity field in a fluid. Application imposed on fluid flow by fluctuations in turbulent
of LDV involves shining a laser light sheet in a flow velocities. The Reynolds stresses are expressed
that has been seeded with micron-size particles and mathematically by a tensor.

Hession, W.C., Curran, J.C., 2013. The impacts of vegetation on roughness

in fluvial systems. In: Shroder, J. (Editor in Chief), Butler, D.R., Hupp, C.R.
(Eds.), Treatise on Geomorphology. Academic Press, San Diego, CA, vol. 12,
Ecogeomorphology, pp. 75–93.

Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00323-7 75

76 The Impacts of Vegetation on Roughness in Fluvial Systems

Shear velocity The shear stress created by the flow is the sum of the root mean squares of the fluctuating
expressed in terms of velocity. It is found by taking the components of the three-dimensional flow. Units are of
square root of the shear stress divided by density. velocity squared [L2T-2].
Turbulent kinetic energy (TKE) Measure of the energy in Von Karman vortex street A series of alternating vortices
the flow, specifically the energy associated with the in the streamwise direction that dissipate with distance
turbulent eddies in a flow. TKE is expressed per unit mass. It downstream.

Abstract

Interest in the interactions and feedback loops between vegetation and geomorphology has grown largely in recent years.
This interest is partially driven by the popularity of stream restoration activities worldwide. Plants create a complicated
system of feedbacks and linkages between channel flow and morphology, sediment deposition and erosion, and plant
morphology, density, and spatial extent. Here, we focus on what we feel is the first step in understanding the complex
processes involved – how vegetation impacts roughness in fluvial systems. We frame our discussion through the location of
vegetation in relation to open channels and the flows they encounter: in-channel emergent; in-channel submerged;
streambank; and floodplain. For each section, we begin with a focused discussion of how vegetation influences roughness at
the reach scale, then concentrate specifically on hydraulics and turbulence, and conclude with a discussion on how the
vegetation and associated roughness influence sediment dynamics. The chapter ends with a discussion on the complexities
related to vegetation and fluvial processes, and some of the research opportunities and challenges.

12.6.1 Introduction 2003), sediment transport (Sand-Jensen, 1998; Cotton et al.,

2006) and channel morphology (Tal and Paola, 2007; Hession
Research in linkages between vegetation and geomorphology et al., 2010), and aquatic habitat (Muhar, 1996; Downes et al.,
continues to grow (e.g., N.R.C., 2002, 2007; Bennett and 1998) and biodiversity (Beisel et al., 2000; Sullivan et al.,
Simon, 2004; Hession et al., 2010). For example, interest in 2006; Klaar et al., 2009). In general, the roughness of
river and stream restoration has increased dramatically over streams/rivers and their floodplains can be partitioned
the last two decades. Conservative estimates place river into components, mainly the roughness due to surface
restoration costs for the continental US in excess of $14 billion material, vegetation, and morphology irregularity (Cowen,
since 1990 with more than $400 million spent on restoration 1956; Arcement and Schneider, 1989; Defra/EA, 2003). This
projects in the Chesapeake Bay Watershed alone (Bernhardt chapter is focused on the impact of vegetation on roughness
et al., 2005; Hassett et al., 2005). Restoration of riparian or in fluvial systems. We have divided our discussion into
streamside forests is a major focus of many stream restoration broad category distinctions: (1) in-stream emergent vege-
activities throughout the US (N.R.C., 1992; U.S.E.P.A., 1999; tation; (2) in-stream submerged vegetation; (3) streambank
Bernhardt et al., 2005). In fact, the Chesapeake Bay Program vegetation; and (4) floodplain vegetation. The discussion in
exceeded a goal of establishing at least 16 090 km (10 000 mi) each category includes field studies, flume studies, and
of riparian forest buffers by the year 2010, and have expanded modeling activities. Each section begins with a discussion of
their goal to reforesting 70% of all streams and shorelines in reach-scale effects, changes to local channel hydraulics, and
the basin (Chesapeake Bay Executive Council, 2003). Add- closes by reviewing implications for sediment transport and
itional riparian forest programs include the Federal Conser- deposition.
vation Reserve Enhancement Program (CREP; N.R.C., 2002) There is no standard, universal definition of where a
and a recently launched initiative of the Conservation Reserve stream ends and the actual streambank begins, much less
Program to reforest 2025 km2 of river floodplains in the US a simple technique to define the end of the streambank
(Johnson, 2004). In addition to these government-sponsored and beginning of floodplain. Floodplains can be identified
reforestation efforts, large areas of bottomland are naturally based on frequency of inundation (Moody et al., 1999),
reverting from crop and pasture (commonly abandoned) to morphology (Leopold, 1994; Rosgen, 1996), or change in
woodland (Trimble, 2004). These reforestation efforts are not vegetation type (Osterkamp and Hupp, 1984; Richard et al.,
limited to the US, but are a worldwide trend (Gippel, 1999; 2005), and are dynamic by nature (Leopold et al., 1964;
Anderson, 2006). The impacts of activities such as riparian Hughes et al., 2008). In this chapter, we utilize Leopold’s
reforestation and conversion from agricultural to forest up- (1994) definition of the floodplain as ‘‘a level area near a river
lands will have a significant impact on fluvial geomorphology, channel, constructed by the river in the present climate and
with a major effect coming from the changes to how rough- overflowed during moderate flow events’’ and that it coincides
ness is characterized across a landscape. Here, we focus on with the elevation of bankfull stage. Osterkamp and Hupp
how vegetation influences roughness in fluvial systems. (1984, 2010) provided definitions of alluvial surfaces and a
Roughness in fluvial systems is a critical characteristic block diagram showing their positions (Figure 1) that visually
influencing water-surface elevations and flow (Defra/EA, organizes our sections.
 The Impacts of Vegetation on Roughness in Fluvial Systems 77

AS

AB
HL
Tu Tp FP DB FB FP Tp Tu
CB
Alluvium
Not to scale Consolidated rocks
Figure 1 Block diagram showing geomorphic features. Reproduced from Osterkamp, W.R., Hupp, C.R., 1984. Geomorphic and vegetative
characteristics along three northern Virginia streams. GSA Bulletin 95, 1093–1101, with permission from GSA.

The most widely used methods for describing roughness in

streams are based on the semi-empirical formulas of Chezy
Veg
and Manning (1890). Most commonly used in general en-
gineering practice is the Manning equation

R2=3 S1=2 Unveg

V¼
n b
0.05 0.1 0.15 0.2
where V is the mean flow velocity (m s1), R the hydraulic Figure 2 Boxplot showing the range of flow resistance values
radius (m), S the friction slope (m/m), and n the Manning’s recommended for stream channels with (veg) and without vegetation
roughness coefficient. Values of n (or flow resistance) can be (unveg) as compiled from Chow (1959). Reproduced from Anderson,
calculated using measurements of velocity, depth, and slope B.G., 2006. Quantifying the interaction between riparian vegetation
(Leopold et al., 1964; Limerinos, 1970). Calculations are and flooding: from cross-section to catchment scale. Dissertation in
commonly performed at gauging stations to provide estimates support of doctoral degree. School of Anthropology, Geography and
of n over various flow conditions (Barnes, 1967; Hicks and Environmental Studies, University of Melbourne, 529 pp, with
permisssion from Geography and Environmental Studies.
Mason, 1998), but this process provides a roughness value that
lumps together the effects of all types of roughness (Leopold
et al., 1964; Defra/EA, 2003) and can have high levels of un- 12.6.2 In-Stream Emergent Vegetation
certainty (Kim et al., 2010). Despite these drawbacks, this
method has been used to estimate the incremental increase in Emergent vegetation refers to plants rooted below the normal
n value due to the presence of vegetation by calculating the water surface whose stems extend above this surface. The stems
n values for sites with a range of bank vegetation (Coon, 1995, of these plants alter the velocity profile, channel stresses, and
1998). sediment transport rates. This section explores recent research
The influence of vegetation on overall stream roughness concerning the effects of emergent vegetation by reviewing the
has traditionally been incorporated into the Manning’s methods used to measure flow resistance, reach-scale effects,
roughness. To provide a general sense of the range of rough- changes to local channel hydraulics, and rates of sediment
ness values attributed to vegetation, we summarized some transport and deposition within areas of emergent vegetation.
Manning’s n values or incremental factors from previous The complex flow dynamics around emergent vegetation
publications. For instance, Cowen’s (1956) procedure in- has led to a large amount of research being conducted in la-
creases Manning’s n by 0.005–0.10 for vegetation influences boratory flumes, where experimental conditions can be
on flow. Given that basic n0 (for a straight, uniform, and somewhat controlled and hydraulics measured. The plants
smooth channel in natural materials) ranges from 0.020 to used in flume experiments range from simulated plant stems
0.028, this represents as much as a 500% increase in rough- using cylindrical objects such as wooden dowels (Bennett et al.,
ness due to vegetation alone. Chow (1959) recommended that 2002) to natural grasses and willows (Jarvela, 2002). When
flow resistance values for stream channels with and without using real plants, research has focused on how changes to
vegetation are summarized in Figure 2 (Anderson, 2006). As plant morphology affects flow profiles. Simulated vegetation
with Cowen’s (1956) values, roughness values change dras- remains popular in flume experiments, as it has the advantage
tically in the presence of vegetation. of control over spatial density (Bennett et al., 2002), and does
78 The Impacts of Vegetation on Roughness in Fluvial Systems

not require a natural substrate. Recent advances in instru- 12.6.2.2 Hydraulics and Turbulence
mentation involve the application of laser technologies to
hydraulic studies. The use of laser Doppler velocimetry, particle The stems of emergent vegetation have the potential to alter
imaging, and the combination of dyes and laser-induced local and reach-scale hydraulics through their influence on
fluorescence with imaging all require clear water flow, making bed roughness, Reynolds stresses, and flow profiles. Visual-
growing vegetation in a natural substrate difficult as fine sedi- ization and quantification of flow profiles is a burgeoning area
ment would be subject to entrainment. of research with the growth of the use of Particle Image
Velocimetry (PIV) in flume studies. As a brief description,
micron-size particles are added to the flow and a laser is used
to illuminate these particles. As the area is illuminated, a
12.6.2.1 Reach-Scale Impacts of Emergent Vegetation
camera records images of the illuminated area at a minimum
Reach-scale hydraulics are altered where emergent vegetation rate of 30 images per second. Because the images are taken
growth occurs as discrete patches (Jarvela, 2002; Wilson, quickly, individual seeding particles can be tracked over time.
2007). Sedges and willows were planted in a flume using a Image analysis allows the user to measure the direction and
variety of arrangements to measure the effects of plant rigidity speed of individual flow particles, which enables calculation
and spatial distribution on the reach-averaged friction factor. of turbulent flow properties. The reader is referred to works by
Results indicated a negative correlation between Reynolds Hart (1998), Fox and Belcher (2009), and Hurther et al.
number and friction factor, but did not completely explain (2009) for a complete description of the principles and ap-
measured changes in friction factor. The arrangement and plication of PIV.
density of the plants exerted a significant influence over fric- At low Reynolds number flows, mixing processes are
tion factor, as did the presence or absence of leaves on willows. dominated by diffusion. Laser-enabled imaging was used in
For example, where the density of willow plantings was dou- fluorescein dye studies of diffusive mixing processes through
bled, the friction factor also doubled, and where leaves were large areas of emergent simulated grasses to elucidate the ef-
present on the willows, the friction factor tripled. fects of vegetation on diffusive mixing (Nepf et al., 1997;
The drag forces acting on emergent vegetation, together Nepf, 1999; Serra et al., 2004). Using imaging to trace the path
with surface friction acting on the bed and walls (or banks) of of the dye over time, lateral diffusion and drag coefficients
a flume (or river), balance the gravitational forces driving flow were calculated over a range of Reynolds numbers, relative
through a vegetated channel. Building on the work of Petryk depths, and percent channel area filled with plant stems. At
and Bosmajian (1975), Wu et al. (1999) introduced a par- low Reynolds number flow, the inertial-viscous flow regime
ameter, lAL, to account for the influence of vegetation in their was expected to be dominated by wake effects around indi-
derivation of the drag coefficient, where l is the vegetal area vidual stems. The diffusion coefficient did show a dependence
coefficient defined by the area fraction per unit length of on Reynolds number, with the largest diffusion coefficients at
channel and dependent on vegetation type, density, and low Reynolds numbers. Dye measurements indicated that flow
configuration; A is the area; and L is the channel reach length. patterns depended on plant density; a finding confirmed by a
When the type of vegetation is specified, this parameter is measured correlation between drag coefficient and the percent
converted to a Manning’s n, making its application to reach- channel area filled with plant stems. Drag coefficients were
scale resistance estimates more generally accessible. The re- higher where plants were sparse because each stem exerted an
sulting roughness coefficient decreased with increased flow individual influence on the flow. As the Reynolds number
velocity, a consequence of a uniform flow velocity within increased beyond 200, the flow regime shifted to the fully
emergent vegetation. inertial regime, dominated by vortex shedding and turbulent
Field studies of emergent vegetation are limited and have flows. The drag coefficient became dependent on both Rey-
focused on general patterns associated with plant growth. nolds number and percent area of stems. These experimental
Gurnell et al. (2006, 2010) described the effects of seasonality results provided verification of a physically based model de-
and site characteristics on vegetation growth and flow patterns veloped to describe the diffusive processes in emergent vege-
in the UK. By conducting field research, they were able to tation (Nepf et al., 1997). Similar experimental techniques
address both spatial and temporal variability of vegetation were applied to a recent study of the mixing processes through
growth and the associated effects on channel flows. When natural vegetation, in this case, reeds (Shucksmith et al., 2010).
vegetation was abundant in the spring and summer, flow re- This recent study also found a greater uniformity in velocity
sistance and depth increased. During the subsequent dieback profiles and a concurrent reduction in shear dispersion, and
during fall and winter, flow depth lowered. The authors hence longitudinal mixing, in emergent vegetation. The results
speculated on a connection between seasonal overbank verify those of earlier studies and extend their applicability to
flooding and elevated flows due to plant growth. Effects of the natural vegetation.
seasonal growth and dieback of emergent vegetation were Nepf (1999) developed a physical model where turbulent
verified by a recent 3-year study of a river in Japan (Asaeda kinetic energy was parametrized as the sum of bed shear and
et al., 2010). As the vegetation in the channel grew during stem-generated wakes around simulated grasses. Diffusivity
spring and summer, flow velocities slowed and water depth was modeled as a function of turbulent and mechanical dif-
increased, similar to what was observed in the UK. When the fusion, with mechanical diffusion dependent on the presence
plant shoots collapsed during dieback, overall channel of vegetation. Building off experimental findings (Gambi
roughness increased by almost 50% before the plants et al., 1990), the model showed reduced velocity and turbu-
degraded. lence intensities within the vegetated area. Thus, as plant
 The Impacts of Vegetation on Roughness in Fluvial Systems 79

density increased, turbulence intensity increased, and mech- dowel area created a von Karman vortex street, a series of
anical diffusion contributed a greater proportion of the total alternating vortices in the streamwise direction that dissipated
diffusion. Total diffusivity was reduced in vegetated channels with distance downstream. The combined results from Nepf
due to a decrease in the scale of the vortices created by the and Vivoni (2000) and Liu et al. (2008) provided necessary
vegetation. Changes in flow velocity and turbulence pro- details about flow hydraulics and a baseline of the wake-
duction created a nonlinear response in the flow such that generated shear stress contribution needed to improve vege-
turbulent kinetic energy increased with increasing stem dens- tated flow modeling.
ity until reaching a point above which the energy decreased as Recent flume experiments investigated the lateral two-
more stems were added. dimensional (2D) structure of flow at the interface of channel
Strong vortex formation occurs at the interface between flow and emergent vegetation within a channel width (White
vegetated and nonvegetated channel areas, regardless of whe- and Nepf, 2008; Zong and Nepf, 2010). These experiments
ther the interface is between the channel and floodplain simulated vegetation using dowels so that the 2D flows could
(Shiono and Knight, 1991) or within a partially vegetated be visualized and current models expanded to include sec-
channel (Tsujimoto, 1999). Vortices are an important means ondary flow patterns. Experimental results showed the pres-
of energy dissipation, making the interface between vegetated ence of a second inflection point at the transition between the
and nonvegetated regions of interest for correctly estimating a vegetated and nonvegetated channel areas, characterized by a
reach-scale friction factor in a partially vegetated channel. similar sharp spike in Reynolds stresses to that measured near
Research into this topic has focused on quantifying the in- the bed of a vegetated region. The vegetated region affected
fluence of spatial distribution of vegetated patches on channel flow velocities beyond the patch edge. As flow approached
hydraulics, which necessarily influences the total plant inter- dense vegetation, it began to slow at a distance upstream equal
face area in the channel (Helmio, 2002, 2004). Two basic to the effective width of the vegetated region. Once within the
types of models of the turbulent flow through vegetation had vegetation patch, flow velocity decreased rapidly. The same
evolved by 2005. One considered the flow profile as a single trend was measured for sparse vegetation patches, but velocity
layer and modeled vegetation by modifying the k  e turbu- reduction occurred closer to the plants and decreased less
lence model. The other was distinct to submerged vegetation within the patch (Zong and Nepf, 2010). Velocity and shear
as it separated the flow into two separate layers: one for flow stress decreased gradually with lateral distance from the in-
within the vegetated area and a second for flow above the flection point and into open channel flow. At the vegetation
canopy. Both types of models reproduced the general shape of interface, a shear layer formed, characterized by coherent
the velocity profile, shear stress, and eddy viscosity within the vortices that generated a regular pattern of momentum sweeps
vegetation as measured experimentally (Defina and Bixio, and ejections across the vegetation interface.
2005), but neither reproduced the profile shape in the region White and Nepf (2008) subsequently developed a 2D
immediately adjacent to the bed. When comparing quantita- model describing the lateral flow profile in a partially vege-
tive turbulent values predicted and measured in the vegetated tated channel. Flows were separated into four distinct zones
region, agreement was limited to 10% when using a two-layer (Figure 3). Zones I and IV were defined below and above the
model, and was worse with a k  e model. transition flows, respectively, and velocity scaled with a rela-
Laser Doppler velocimetry (LDV) systems measure turbu- tionship between drag and gradient. Zone II marked the
lent flow characteristics, and the application of these systems transition from channel flow to slow-moving flow within the
to flows around emergent and submerged vegetation has ad- vegetated area and was separated into an inner layer and a
vanced understanding of small-scale hydraulics. The turbulent mixing layer. The inner layer width was dependent on drag
exchange of fluid and momentum through simulated emer- forces generated by the vegetation and defined the mixing
gent grassy vegetation was measured using an LDV system length into the vegetation field. The shape of the velocity
(Nepf and Vivoni, 2000; Liu et al., 2008). Significant shear profile in zone II fit a hyperbolic tangent, which reproduced
stresses were measured and turbulence intensity remained the sharp decline in flow velocity with distance into the
uniform with depth. A velocity gradient formed around each vegetation. Zone III represented the outer layer flow where
dowel between the slower flow in line behind the dowel and hydraulics were independent of the forces acting on the lower
the higher velocity flow on either side of the dowel, creating a flows. In this zone, the shear stresses were balanced by the
horseshoe vortex. A horseshoe vortex is identified by its pressure gradient, creating a gradual reduction in flow velocity
characteristic morphology defined by a core vortex with two with distance away from the vegetation. To connect the hy-
vortices trailing in the streamwise direction at a 451 angle to draulics across zones II and III, White and Nepf (2008) de-
the bed. The horseshoe vortex brought high-velocity flows fined a slip velocity and momentum exchange at the inflection
from the region near the dowel into the immediate region at point. The model generated from this research represented an
the base, creating a localized area of increased turbulence and advance in the application of technology to the study of
a counterclockwise mixing pattern. Directly above the spike in vegetated channels, and in the understanding and modeling of
flow velocity and turbulence, the flow slowed and turbulence flows in channels with emergent vegetation.
reduced, marking the area near the bed as an inflection point.
Both turbulent and velocity profiles are uniform over depth
above the inflection. Fluid exchange occurred primarily
12.6.2.3 Emergent Vegetation and Sediment Transport
through longitudinal advection driven by turbulent wakes
generated around plant stems (Nepf and Vivoni, 2000). In the Fine sediment deposition within emergent vegetation is not a
flow regions between dowels, turbulent eddies shed from the well-studied process to date. This lack of research attention is
80 The Impacts of Vegetation on Roughness in Fluvial Systems

L IV

III

U(y)
} ym
II } I
Vegetation I

Figure 3 Streamlines for a typical vortex structure in a frame moving with the vortex. Reproduced from White, B.L., Nepf, H.M., 2008. A
vortex-based model of velocity and shear stress in a partially vegetated shallow channel. Water Resources Research 44, W10412, with
permission from AGU.

more likely a reflection of the difficulties associated with the The influence of seasonal phases of emergent grassy vege-
research and not a consequence of overlooking an important tation growth on fine-sediment deposition was recently
phenomenon. Recognition of the importance of sediment measured in the field by Asaeda et al. (2010). When plants first
deposition within vegetation has proved to be much easier emerged in spring, deposition rates were high, but as the
than quantification of rates and volumes. Field observations plants grew to full size in the summer, the accumulated
document thick vegetative growth in near-bank areas where sediment eroded. This occurred despite a significant reduction
vegetation acts as a very effective sediment trap (Gurnell et al., in flow velocity through the plants. Sediments accumulated
2006). Sediment accumulation near banks may reduce the again during the shoot collapse phase when flow resistance in
effective channel width over time, essentially altering the flow the channel increased. The sediment did not erode again until
structure of the channel. after the plants had fully decomposed. This cycle was observed
Theoretical models and flume experiments have attempted to repeat over the 3 years of study observations showing that
to predict and quantify sediment transport and deposition the range in plant morphologies exerted a greater influence
rates around and within plant areas. Agricultural concerns over sediment deposition and erosion than did the flow rate
about sedimentation around row crops and stream buffers (Asaeda et al., 2010). The results of these experiments con-
motivated many early flume experiments (e.g., Abt et al., firmed the importance of vegetation parameters when con-
1994). Deposition rates onto removable plots of corn and sidering how sediment transport and flow patterns are altered
grass were measured in a flume. The use of natural plants by emergent vegetation.
demonstrated the importance of plant flexibility and blade
length on sediment deposition volume. The longer and more
flexible plants bent forward onto the channel bed, which re-
12.6.3 In-Stream Submerged Vegetation
duced sediment deposition in those areas. However, once
sediment was deposited, it was protected from erosion by the
Submerged vegetation grows in channels where the water
plant stems (Abt et al., 1994). To parametrize the effects of
depth is sufficient to form a velocity profile above the vege-
vegetation, the Meyer-Peter and Mueller transport equation
tation canopy. As a category, there has been much more re-
was adjusted for the presence of emergent vegetation using the
search into the effects of submerged vegetation on channel
results of flume experiments where uniform sediment was fed
form and processes than for emergent plants. Because the
into a field of rigid, fixed rods arranged in a predetermined
whole plant is submerged and not just the stem, the entire
spatial pattern and density until an equilibrium transport rate
plant morphology alters the flow profile, turbulence, and
was reached (Jordanova and James, 2003). The results were
sediment transport rates. The same methods and technologies
mixed as all the complexities of plant morphology and re-
are used to research submerged vegetation as were discussed in
sponse to flow were condensed into a single parameter.
the previous section. Thus, this section explores the recent
Depositional patterns within vegetated patches differ de-
research concerning the effects of submerged vegetation by
pending on the length and density of the simulated vegetation
reviewing reach-scale effects, changes to local channel hy-
(dowels) and the size of the sediment in suspension (Sharpe
draulics, and sediment deposition rates and volumes.
and James, 2006). Large grain sizes deposited near patch
edges, whereas finer sediments were transported further into
patches before depositing. Dowel density had a similar effect
12.6.3.1 Reach-Scale Impacts of Submerged Vegetation
on deposition volumes, with larger volumes depositing near
the edge when stem density was high. The control exerted by Submerged vegetation creates a drag force that reduces flow
hydraulic variables over sediment deposition rates and vol- rates and, over the long term, alters channel shape. The loss of
umes was verified by observations in salt marshes (Mudd flow capacity due to submerged vegetation was widely recog-
et al., 2010). Where velocities through marsh vegetation in- nized by the mid-twentieth century (e.g., Kouwen et al., 1969;
creased, the amount of suspended sediment that entered and Phelps, 1970), although field studies quantifying the effects
deposited within vegetated patches increased. of submerged vegetation were few due to the difficulties
 The Impacts of Vegetation on Roughness in Fluvial Systems 81

associated with making direct measurements. The influence of parameter for roughness height and its application tested
vegetation in establishing and altering channel morphologies through a series of flume experiments using flexible plastic
over long reaches has also been recognized (Gurnell et al., strips to simulate river grasses (Kouwen and Unny, 1973).
2010). In a study of rivers in the UK, macrophyte patches were Experimental results verified use of the modified velocity
shown to create distinctly different large-scale flow patterns profile for erect and waving strips where plant roughness
depending on the density of the patch and the total amount of heights were similar. Prone vegetation reduced friction factors
channel area covered by plants (Cotton et al., 2006). Where by a factor of 5 and did not produce a good fit to the modified
vegetation grew in individual stands, limited flow continued profile. The results demonstrated a need to consider and fur-
through the plants, whereas the majority passed through the ther quantify vegetation flexibility and the feedback between
narrow channels between plants. Flow velocities in the narrow flexibility and channel discharge when estimating the rough-
channels were consistently higher due to flow constriction ness height over vegetated channel beds.
than within planted patches, where flow velocities were re- Later research continued to focus on adjustments to the
duced by 48%. Where vegetated patches were thick and ex- roughness factor (Petryk and Bosmajian, 1975; Kouwen and
tensive across the channel, velocity reductions were general Li, 1980; Jarvela, 2002; Wilson, 2007). The body of literature
over the channel reach. Over time, the flow patterns in and for which vegetated roughness is parametrized through Man-
around vegetation patches contributed to changes in channel ning’s n is detailed by Green (2004), and a history of the early
bed topography and then general channel morphology. modifications to the log-law is given by Stephan and Gut-
The general velocity profile through a channel with sub- knecht (2002). Jarvela’s (2002) experiments were described in
merged vegetation is characterized as consisting of two parts: a the emergent vegetation section but deserve another mention
uniform low-flow profile within the vegetated area (Kouwen here as they also measured friction factors associated with
and Unny, 1973) and a logarithmic-shaped profile in the flow different densities and arrangements of submerged grasses.
above the plant canopy (Figure 4). The universal log-law The friction factor correlated to relative roughness and
equation can be applied over the entire velocity profile, but Reynolds number, decreasing with higher Reynolds numbers
modifications to the roughness height and zero plane dis- and increasing with relative roughness values. More recently,
placement are required to account for flow over vegetation natural vegetation was studied to verify and extend the use of
(Shi and Hughes, 2002). An added complexity of submerged Kouwen’s logarithmic profile method (Kouwen and Li, 1980;
vegetation is plant flexibility, the ability of submerged plants Carollo et al., 2002, 2005). Experimental results demonstrated
to bend and wave in the flow. An early and significant study on a propensity for the equations of Kouwen to overestimate flow
the change in flow profile in the presence of flexible, sub- resistance and a need to calibrate to measured vegetation
merged vegetation parametrized the bending of the plants as a concentration and flexural rigidity. With these adjustments,
function of the elasticity of the plant and the channel flow the method applied well to natural submerged plants, either
(Kouwen et al., 1969). Combining these variables into a single in patches or covering extensive amounts of channel area
parameter, the flexural rigidity of the vegetation (J), Kouwen (Nikora et al., 2008).
et al. defined J ¼ EI, where E is the longitudinal modulus of The value assigned to the vegetation coefficient has a
elasticity of the plant and I the second increment of the plant dominant influence over predicted channel roughness, and
cross-section. For low values of EI, the plant bends over research has focused on adjusting this coefficient according to
completely and lies on the bed surface. For high rigidity val- the amount of plant frontal area facing the flow (Green, 2004;
ues, plants remain erect. When this parameter is multiplied by Luhar et al., 2008). The derivation of a friction coefficient for
the number of plants in a channel cross-section, it represents vegetated regions by Wu et al. (1999) described for the case of
the total plant resistance to flow. The logarithmic velocity emergent vegetation, also considered submerged vegetation
profile equation was adjusted to incorporate the new when they incorporated information on plant stiffness and
density into their coefficient, lAL. Using an accurate value of
the vegetation coefficient, the authors showed good agreement
y Measured
between predicted and experimental values of roughness.
velocity profile
With the growing interest in vegetative resistance and number
of adjustments to the friction factor, Baptist et al. (2006)
compared the results of four methods to existing experimental
Logarithmic
velocity profile data. The first was a theoretical derivation of the friction factor
that divided the flow profile into segments and applied the
u(y)
effective water depth and velocity profile equations to calcu-
h late resistance separately in segments below and above the
H
canopy. The second method was an analytical derivation of
the velocity through the vegetated region using the conser-
vation of momentum. The third employed numerical simu-
0 y′ lations using a 1D k  e turbulence model. The fourth
y″ yP
method, which employed a genetic programming technique,
reproduced the experimental results with the greatest accuracy.
Figure 4 Vertical velocity profile above aquatic vegetation. Reproduced The genetic programming method has not been employed
from Stephan, U., Gutknecht, D., 2002. Hydraulic resistance of extensively, as recent research has focused on developing
submerged flexible vegetation. Journal of Hydrology 269, 27–43. process-based methods for calculating roughness, and the
82 The Impacts of Vegetation on Roughness in Fluvial Systems

reader is referred to the work by Baptist et al. (2006) for a layer and the inflection point just above the bed surface (de-
complete description of the technique. scribed in the section on turbulence in emergent vegetation).
The use of Acoustic Doppler Velocimeters (ADVs) in flume The shear layer at the canopy was characterized as a second
experiments has enabled friction factor adjustments based on inflection point in the flow that moderated vertical mo-
the measured turbulent flow field around flexible grasses mentum exchange across the canopy (Murphy et al., 2007). As
(Stephan and Gutknecht, 2002). In flows without vegetation, flow moved upward through the canopy layer, it mixed with
turbulence intensity is reduced as relative submergence de- higher velocity flows above the canopy. The mixing created
creases because coherent flow structures, which create pressure Kelvin–Helmholtz instability and a strong vorticity that pulled
fluctuation on the bed surface, were unable to form com- flow from above the canopy into the vegetated area and dis-
pletely. In vegetated channels, this trend was not observed. placed slower moving flow upward in a clockwise pattern (Liu
Stephan and Gutknecht (2002) hypothesized that flexible et al., 2008). As total turbulent kinetic energy increased, the
vegetation responded to pressure fluctuations by bending. inflection point shifted vertically upward and further from the
Pressure fluctuations formed and were dissipated by the canopy (Wilson et al., 2003). When flow in the shear transi-
waving of flexible plants. Thus, the flow profile over a bed of tion layer was measured over three dimensions, vortex growth
submerged vegetation with low relative roughness could be fit was shown to continue until balanced by drag dissipation
by a logarithmic profile modified for effective vegetation around plant stems and fronds, and an equilibrium condition
height. Defining effective plant height as the mean plant developed (Ghisalberti and Nepf, 2004). In fully developed
height after deflection by the flow, and reducing total flow flows at equilibrium over submerged vegetation, the ratio of
depth by the effective plant height, caused the zero plane turbulent production to dissipation attained a constant value.
displacement to scale directly, and allowed for calculation of The description of the inflection point is similar for flow
the shear velocity from the logarithmic profile. Building on processes operating at the lateral interface of submerged
this work, Jarvela (2005) measured the effective height and vegetation and clear channel flow (White and Nepf, 2008).
shear velocity over wheat. The experiments verified applic- A mixing layer characterized by formation of coherent vortices
ability of the shear velocity equation with depth adjusted for formed along the interface, creating velocity and turbulent
mean plant height to channels with submerged, flexible flow profiles with similar spikes as were observed across the
vegetation. The method provided a way to estimate shear plant canopy. The similarities indicate that flow across a
velocity that did not require characterizing turbulence or cal- vegetation interface develops a similar layer of coherent vor-
culating Reynolds stresses. This work extended the general tices and peak in Reynolds stresses regardless of whether that
applicability of the logarithmic velocity profile to the flow over transition is lateral or vertical. Mixing layers cannot be as-
submerged vegetation in general and made the calculation of sumed for the interface of every submerged vegetation patch.
the velocity profile generally accessible. They formed only where there was sufficient absorption of
momentum by the plant frontal area of the canopy to trigger
the Kelvin–Helmholtz instability (Nepf and Ghisalberti,
2008). Thus, only flows across densely vegetated interfaces
12.6.3.2 Hydraulics and Turbulence
were characterized by an inflection point.
Plant morphology, particularly the presence of leaves and the Momentum exchange was limited by dense canopies where
frond shape, has a demonstrated effect on the drag coefficient the frontal area of the plants consumed much of the turbulent
(Nepf, 1999; Baptist, 2003; Wilson et al., 2003). Aquatic plant energy, preventing it from penetrating into the canopy. Dye-
growth has been shown to reduce bed shear stress by up to tracing studies by Murphy et al. (2007) measured the limits of
80% (Baptist, 2003), and where the frontal area was dense, vortex-driven and diffusion exchange along dense plant can-
flow velocity into the vegetated reach reduced more quickly opies, simulated with wooden dowels. These results were used
than for sparsely vegetated patches. Dense vegetation also re- to develop a two-zone model of dispersion and diffusion
duced the longitudinal extent of momentum exchange across across a submerged vegetation interface. Below the canopy,
the plant interface. This effect was demonstrated by comparing mixing was controlled by dispersion around individual plant
the stems of plants to plants with fronds. Fronds increased the stems and related to the density of the stems. Above the plant
surface area of the plant region and created an additional drag canopy, shear dispersion controlled mixing. Exchange across
element, which in turn increased the amount of turbulence in the canopy was by either vortices formed along the interface as
the flow (Wilson et al., 2003). part of a mixing layer characterized by Kelvin–Helmholtz in-
Reductions in flow velocity and turbulence intensity within stability, or diffusion processes operating within the canopy
a vegetated patch have been demonstrated over a range of flow but near the interface. Recent dye-tracing experiments verified
velocities, vegetation densities, and locations within the the increase in vertical shear and longitudinal mixing across
vegetated area (Gambi et al., 1990). More recent detailed flow natural canopies (Shucksmith et al., 2010).
measurements across the full channel depth characterized the Vortex-driven dispersion was dominant across sparsely
velocity profile by distinct flow regions within and above the vegetated regions where turbulent mixing penetrated into the
vegetation, with a sharp spike in flow velocity and Reynolds vegetation, creating flow profiles with a large degree of uni-
stress at the transition (Nepf and Vivoni, 2000; Jarvela, 2005; formity (Righetti, 2008). The extent of momentum exchange
Liu et al., 2008). The spike in Reynolds stresses just above the across a canopy is a function of how far the Kelvin–Helmholtz
top of the canopy was independently verified using experi- vortices penetrate into the vegetated area, which is limited
ments where dowels simulated vegetation (Liu et al., 2008, across dense canopies (Luhar et al., 2008). The distance over
2010; Righetti, 2008), as were the similarities between this which momentum exchange occurs was defined by Nepf and
 The Impacts of Vegetation on Roughness in Fluvial Systems 83

Vivoni (2000) as de. Beyond this length, turbulence is gener- It was not until the 2000s that quantifying sediment de-
ated by individual plant stems. Significant wake zones around position within aquatic plants became a significant research
plant stems resulted in a dominance of relative plant sub- topic. The study by Baptist (2003), already described for its
mergence over mixing rates below the canopy (Shucksmith finding that submerged vegetation reduced bed shear stress by
et al., 2010). Where vortex translation speed was large com- up to 80%, also measured sediment transport rates out of a
pared to rotation speed, and the submerged vegetation was vegetated area. During the experimental setup, a layer of
flexible, the canopy developed a wave pattern with a frequency uniform sand was placed within the vegetated area and
equal to the Kelvin–Helmholtz instability. The waving pattern changes in the bed profile during the experiment were meas-
enabled turbulent exchange closer to the bed, although not ured using electronic conductivity bed profilers. As expected,
further below the canopy (Nepf and Ghisalberti, 2008). without any sediment influx, the bed experienced net erosion.
Further review of the analytical methods used to calculate The unexpected result came from a comparison of shear
momentum and mass transport across submerged vegetation stresses and associated transport rates for runs with and
is presented in the work of Nepf and Ghisalberti (2008), without vegetation. The same rate of transport could be
where the turbulent flow characteristics around real vegetation achieved at a lower shear stress in a vegetated channel when
planted in a flume were measured using PIV. An analytical compared to the nonvegetated state. Baptist hypothesized that
treatment of the experimental results using the double-aver- higher turbulent flow fluctuations within the vegetation led to
aged Navier–Stokes (DANS) method is provided in Righetti increased suspension rates of the fine sediment.
(2008). The volume of sediment deposition within a vegetated
patch is necessarily a function of the extent to which flow can
transport sediment into that patch. Therefore, models pre-
dicting deposition rates have been derived from analytical
12.6.3.3 Submerged Vegetation and Sediment Transport
models of fluid exchange across a canopy (Luhar et al., 2008).
Quantification of sediment deposition rates within regions of Sediment deposition into a vegetated patch was simulated
submerged vegetation has suffered from the difficulties in- over a distance set by the diverging flow pattern created by
herent in making measurements and the complications of shear vortices along the patch edge, regardless of whether the
flexible vegetation. Early approaches avoided the issue of dir- patch was densely or sparsely vegetated. Direct measurements
ect measurements and derived analytical methods to estimate of sedimentation rates were obtained from flume experiments
sediment deposition. One early contribution computed the recirculating water seeded with glass beads designed with
deposition around submerged vegetation of different spatial settling velocity that promoted deposition (Zong and Nepf,
density and arrangement (Li and Shen, 1973). This work 2010). Depositional patterns varied depending on the length
built on the derivation of wake flow around a cylinder of the vegetated path relative to the deposition length scale, xe,
using linear supposition (Petryk, 1969) to extend the deriv- defined by Zong and Nepf (2010) as the ratio of inertial force
ation to multiple cylinders in a variety of arrangements. The to settling velocity. Where the vegetation extent was less than
Shield’s equation was combined with the drag model to esti- xe, sediment deposited uniformly throughout the vegetated
mate the sediment transport capacity of flow through the cy- area. Where vegetation extended beyond the length set by xe,
linders. Results showed a reduction in flow velocity and deposition was greater near the edge and declined with dis-
increase in sediment deposition when cylinder spacing was in tance into the vegetated patch. A probabilistic model was
a staggered as opposed to straight line arrangement (Li and developed from the experimental results to attempt to predict
Shen, 1973). sediment deposition volumes in submerged vegetation.
Interest in reducing the downstream transport of fine However, the measured extent of deposition was greater than
sediment in fluvial systems led to flume studies using simu- that predicted by the model, indicating that increased turbu-
lated sediment (glass beads), cylinder arrays (Barfield et al., lence intensities in the patch were not accounted for in the
1979; Tollner et al., 1982), and existing sediment transport model.
models. The space between adjacent stems was modeled as a
narrow channel, enabling the substitution of plant spacing for
channel width. Transport rates in individual channels were 12.6.4 Streambank Vegetation
predicted using the Einstein transport equation, and total
transport was the sum of transport through all the individual We define streambank vegetation as those plants that are be-
channels. This approach led to an unrealistically uniform rate tween the active-channel bank (AB) and shelf (AS), and we
of sediment movement across the channel. Field observations include the floodplain bank (FB) as shown in Figure 1. For
showed that the distribution of vegetated patches created re- our purposes, vegetation on depositional bars is classified as
gions of high and low flow velocities, which created distinct in-stream vegetation and discussed in previous sections. There
areas of local sediment erosion and deposition (Gurnell et al., are obvious uncertainties around these boundaries and they
2006). Deposition was predominantly within stands of ex- are utilized here only to help organize the discussion. Typi-
tensive vegetation, and erosion occurred along flow paths cally, the streambank is a sloping surface with or without
between individual stands of sparsely vegetated areas. vegetation connecting the active channel to the floodplain.
Monthly variations in depositional volumes indicated an ac- Streambank vegetation is generally acknowledged to have
tive transport regime through the reach and the importance of an influence on bank stability (Simon and Collison, 2002)
changing plant morphology on sediment trapping efficiency and streambank retreat rates (Pizzuto et al., 2010). However,
(Cotton et al., 2006). there is still much research needed to understand the complex
84 The Impacts of Vegetation on Roughness in Fluvial Systems

processes and feedbacks involved (Wynn and Mostaghimi, 12.6.4.2 Hydraulics and Turbulence
2006a), especially as streambank vegetation is nearly always
included in stream-restoration designs (Shields et al., 1995; Several field efforts have been conducted to quantify rough-
Jennings et al., 1999). Interaction between vegetation growth ness due to streambank vegetation based on fine-scale, at-
on channel banks and sediment accretion on the banks is a-site field measurements (Thorne and Furbish, 1995; Wang
necessary to channel stability, and the rate of recovery of and Wang, 2007). Utilizing a coarsely vegetated bend in the
channel morphology in a post-channelization environment Ocklawaha Creek in Florida, USA, they measured velocity and
has been shown to be a function of the ability of vegetation to water-surface topography during high flows before and after
grow and stabilize the bank (Hupp and Simon, 1991). removing the rough bank vegetation (Thorne and Furbish,
Streambank vegetation influences bank retreat rates by altering 1995). Vegetation roughness on the curved streambank pro-
soil moisture and temperature in streambanks (Wynn and duced a backwater effect that inhibited flow directed toward
Mostaghimi, 2006a), through mechanical reinforcement due the bank, shifting the highest velocity to near the center of the
to root structure (Simon et al., 2004; Wynn and Mostaghimi, channel. After removing vegetation, higher velocity flow was
2006b), and influencing the flow turbulence and boundary displaced toward the outer bank. In addition, in the presence
shear stress (Hopkinson and Wynn, 2009). Although much of roughness, the high-velocity core was shifted downstream,
research has been done related to the roughness of in-stream which they suggested contributed to asymmetrical bend mi-
vegetation and vegetation on the floodplain, less has been gration (Thorne and Furbish, 1995). Wang and Wang (2007),
focused on the actual sloping streambanks (Hopkinson and using sections of the Chaodongweigang River in China with
Wynn, 2009). different densities of bank vegetation, measured velocities and
water surfaces to determine the effects of density, stem diam-
eter, and rigidity of reeds. The water-surface slope in the river
reach with reeds was larger than without, and the cross-sec-
12.6.4.1 Reach-Scale Impacts of Streambank Vegetation
tional elevation was concave with the higher elevations within
Most of the research related to streambank vegetation’s influ- the vegetation than in the middle of the channel. They con-
ence on reach-scale roughness has been to inform the selec- cluded that the vegetation resulted in three velocity zones in
tion of roughness coefficients (e.g., Manning’s n) for use in rivers with streambank reeds: slack water zone in the vege-
hydraulic engineering for channel construction and flood tation, a transition zone, and the main flow zone (Wang and
modeling (Anderson et al., 2006); however, much more re- Wang, 2007).
search is needed to fully establish the functional relationships Wang et al. (2009) followed up their field-based investi-
(Yen, 2002). Yen (2002) provided an excellent review of gation with flume experiments using real vegetation (eel
studies and computational methods used to estimate the re- grass). In this study, they vegetated half the flume area and the
sistance coefficient attributed to vegetation in composite results could realistically be applied to in-stream vegetation or
channels. He warned against ‘‘selecting the energy and mo- streambank vegetation. Flow rate in the nonvegetation zone
mentum concept as the effective means for analysis’’ given the increased with increased vegetation density, and differences in
complex and variable interacting forces involved and in- streamwise velocities between the vegetated and nonvegetated
creased energy loss due to wakes behind vegetation. zones generated high shear layers and a transverse vortex
Many scientists and practitioners have utilized long-term (Wang et al., 2009). These transverse vortices near the free
records available from gaging stations to estimate or solve for surface increased with increased vegetation density and re-
Manning’s n based on average velocity, depth, and slope sulted in increased secondary currents and turbulence energy
(Leopold et al., 1964; Limerinos, 1970). Numerous studies with increased Froude numbers. They cited the need for more
have also provided tables with and without actual pictures of research to investigate the interactions between the coherent
streams to help practitioners estimate roughness coefficients vortices and the secondary currents in partially vegetated flows
for natural streams with variable types of vegetation (Cowen, (Wang et al., 2009). Afzalimehr and Dey (2009) also per-
1956; Chow, 1959; Barnes, 1967; Arcement and Schneider, formed flume studies, but focused on the effects of the inter-
1989; Hicks and Mason, 1998), but few have tried to parse out action between bank vegetation and gravel beds on velocities
the roughness attributed specifically to streambank vegetation. and Reynolds stress distributions. Using the natural vegetation
Coon (1995, 1998) did set out to identify the impact of Os (similar to wheat), they found that vegetation on the banks
vegetated banks using 21 stream sites ranging in channel size caused the maximum velocity to be considerably different
and level of vegetation on their banks in New York, US, at or than the free-water-surface velocity, unlike flow over a gravel
near United States Geological Survey (USGS) gauging stations. bed with no vegetation. The vegetation changed uniform flow
Using percentage of vegetated wetted perimeter as a variable to nonuniform flow due to nonlinear Reynolds stress distri-
and analyzing a range of flows at each site, he estimated the butions, and its shape varied with distance from the wall
amount of roughness attributed to vegetation. For the most (Afzalimehr and Dey, 2009).
part, vegetation had little or no effect on roughness of streams Much of the effort to understand the distribution of the
wider than 33 m (100 ft); however, vegetated banks did in- Reynolds shear stresses was to obtain improved estimates of
crease Manning’s roughness by 0.005–0.012 in narrower shear velocities for determining bed resistance (Afzalimehr
channels (Coon, 1995, 1998). Results depended on season and Dey, 2009). For uniform flow over a gravel bed (no
and type, density, and percent submergence of vegetation, vegetation), the Reynolds stress was linear, with maximum
which is also supported by Fischenich (2000) and Anderson shear stress at the bed and minimum near the free surface.
et al. (2006). However, with vegetation on the banks (the flume walls in
 The Impacts of Vegetation on Roughness in Fluvial Systems 85

this case) the Reynolds stress distribution was nonlinear or 12.6.5 Floodplain Vegetation
convex in shape with the apex at a higher location. A sub-
stantial deceleration near the vegetated walls created a strong As previously mentioned, we assume that the floodplain does
lateral shear layer in the interfacial region (Afzalimehr and not start until we reach a level area near a river channel
Dey, 2009). Hopkinson and Wynn (2009) conducted flume (Leopold, 1994) and that this is typically equivalent to the
studies to evaluate the three-dimensional velocity structure, bankfull stage (Rosgen, 1996). Referring to Figure 1, we in-
turbulence, and near-bank shear stresses across three vege- clude only the floodplain (FP) in this section as we have al-
tation treatments (tree, shrub, and grass) and a sloping ready discussed the FB as part of the streambank, and we will
streambank. Similar to Thorne and Furbish (1995), Wang and not extend to the terraces (T). This distinction is important
Wang (2007), and Wang et al. (2009), the free streamwise because where the vegetation is located can drastically change
velocity was increased in the presence of bank vegetation and how it influences flow, velocity, turbulence, and shear stresses
near-bank velocities were reduced for all three vegetation and at what stage these influences become important.
treatments. Hopkinson and Wynn (2009) did find differences Roughness due to vegetation on floodplains has long been
in turbulence intensity and Reynolds stress distributions of interest to engineers due to the role it plays in flooding and
across the different vegetation treatments. Turbulent kinetic flood prediction (Lang et al., 2004). Floodplain vegetation was
energy and Reynolds stresses near the streambank were in- (and, in some cases, still is) considered a nuisance (Darby and
creased for the upright shrub treatment, particularly at the toe, Thorne, 1995; Jarvela, 2004) and was actively managed or
but grass folded to a prone position and protected the removed to reduce roughness and, therefore, reduce flooding
streambank by reducing shear stresses near the boundary (see Figure 5, from the cover of U.S. Army Corps of Engineers,
(Hopkinson and Wynn, 2009). This finding supports a pre- 1997). Although some floodplain vegetation is still managed
vious modeling study by Kean and Smith (2004) that found or removed for flood reduction purposes (Leu et al., 2008),
woody bank vegetation reduced perimeter-averaged boundary the general trend worldwide is to encourage more natural
shear stress, as well as the boundary shear stress below the floodplain vegetation to improve streambank stability and
vegetation and velocity within the vegetation. Hopkinson and ecological health (N.R.C., 2002). Re-vegetation of stream-
Wynn (2009) did find that at low tree densities, the turbulence banks and floodplains is a widely practiced technique in river
intensity and Reynolds stress distributions were similar to restoration activities worldwide (Gippel, 1999; Bernhardt
those of the bare bank experiments. Similarly, Kean and Smith et al., 2005; Anderson, 2006). Although there is concern that
(2004) found reductions in near-bank flow and boundary re-vegetation of river and stream corridors will increase
shear stress due to both drag on vegetation and stress on the flooding, the issue is complex, variable, and much uncertainty
banks for sparse vegetation, but determined that drag on exists (Defra/EA, 2003; Anderson et al., 2006). Anderson et al.
vegetation was the dominant effect in channels with dense (2007) employed a newly developed model called roughness
vegetation. of vegetation in rivers (ROVER; Anderson et al., 2007) to
Streambank vegetation type and density influence rough- model flood waves as they propagated down re-vegetated river
ness (Coon, 1995, 1998), velocity distributions (Wang et al., reaches and found that the peak discharge was reduced, but
2009), and turbulence and shear stresses (Hopkinson and the duration of the flood was increased.
Wynn, 2009). In the discussion of floodplain vegetation, we
see that the actual location of vegetation (streambank vs.
floodplain) can also have a profound influence, particularly
on the exact locations of turbulence intensity and shear
12.6.5.1 Reach-Scale Impacts of Floodplain Vegetation
stresses along the cross-sectional profile of the stream
(McBride et al., 2007). Natural floodplains tend to have diverse vegetation com-
munities due to frequent disturbance by flooding (Gregory,
1992; Hupp and Osterkamp, 1996; Osterkamp and Hupp,
12.6.4.3 Streambank Vegetation and Sediment Transport
2010). This diversity in vegetation has led to spatially hetero-
Vegetation is used in stream restoration projects in an effort to geneous roughness values (Freeman et al., 2000; Forzieri et al.,
stabilize banks and reduce streambank retreat (Bernhardt 2010; Girard et al., 2010) over time (Geerling et al., 2007).
et al., 2005). Although it is generally accepted that vegetation Along the Amazon River in Brazil, Mertes et al. (1995) utilized
that grows on the banks influences flow, sediment transport, remote sensing to evaluate the spatial heterogeneity of flood
and river morphology (Hupp and Simon, 1991; Afzalimehr hydrology and vegetation; they found that vegetation com-
and Dey, 2009; Wang et al., 2009; Li and Millar, 2010), most munity diversity was influenced by the manner in which the
research related to streambank vegetation and sediment has flood waters physically reach the floodplain (whether from
been related to streambank retreat and bank stability (Wynn local sources such as hillslopes or from exchange with the
and Mostaghimi, 2006a; Lawler, 2008; Pollen-Bankhead and river). Several researchers have attempted to quantify this
Simon, 2010) or the contribution of sediment loads from reach-scale variability in floodplain roughness using airborne
eroding banks at the watershed level (Lawler, 2008). Research laser scanning altimetry or airborne light detection and ran-
focused specifically on how bank vegetation influences sedi- ging (LIDAR; Coby et al., 2002; Forzieri et al., 2010), as well as
ment transport has been limited. Those sections describing in- terrestrial laser scanning (TLS; Antonarakis and Richards,
stream and floodplain vegetation influences on sediment 2010). However, there is some indication that such detailed
transport pertain to streambank vegetation; therefore, the quantification of complex roughness distributions does not
reader is referred to those sections for further information. significantly improve floodplain flow modeling over the more
86 The Impacts of Vegetation on Roughness in Fluvial Systems

Existing conditions

Elev. 100′
Elev. 96.0′
Not to scale

Improved conditions
Figure 5 Cover of U.S. Army Corps of Engineers (1997) showing ‘‘improved’’ conditions as removing floodplain vegetation to reduce roughness
and, therefore, flooding. Reproduced from U.S. Army Corps of Engineers, 1997. Channel stability assessment for flood control projects (Technical
engineering and design guides as adapted from the U.S. Army Corps of Engineers, No. 20). American Society of Civil Engineers 978-0-7844-
0201-6. ASCE Press, New York, NY, 108 pp, with permisssion from ASCE press.

traditional use of roughness values based on simple land use dispersion in vegetated rivers was as much as 70–100% larger
characteristics (Werner et al., 2005). than for rivers without vegetation.
In general practice, roughness due to floodplain vegetation
has been estimated using look-up tables or visual references
(Cowen, 1956; Chow, 1959; Barnes, 1967; Arcement and 12.6.5.2 Hydraulics and Turbulence
Schneider, 1989; Hicks and Mason, 1998), as described pre-
viously for streambank vegetation. Floodplain vegetation af- Compound channels, even without floodplain vegetation,
fects flow structure, flow resistance, and turbulent intensities induce complex flow conditions (Figure 7; Shiono and
(Yang et al., 2007). Resistance and flow structure depend on Knight, 1991; Fischenich, 2000; Thornton et al., 2000; Harris
numerous variables, including: (1) type and physical charac- et al., 2003). Where floodplain vegetation is present (Kadlec,
teristics of the vegetation (Freeman et al., 2000; Antonarakis 1990; Fischenich, 2000), drag is generated by velocity gradi-
and Richards, 2010); (2) time of year or seasonal changes ents and eddies are formed during inundation flows, resulting
(Coon, 1998); (3) vegetation succession (Geerling et al., in momentum losses. These losses are typically not in-
2007); (4) density of vegetation (Wang and Wang, 2007); (5) corporated into the existing techniques used to predict
depth of flow (Fischenich, 2000; Anderson, 2006); and (6) roughness and resistance in natural rivers (Fischenich, 2000;
geomorphic setting (Darby, 1999). In equation form, Yen Thornton et al., 2000). Harris et al. (2003) applied genetic
(2002) defined a symbolic roughness parameter as follows: programming to flume data and described how the complexity
  in flow is due to streamlines becoming circuitous as they bend
ks and branch around surface-piercing vegetation elements (see
f ¼ F Re; Fr; Sw ; S0 ; ; Lv ; J; D; M
Re Figure 3). During overbank flows, the resistance of the
floodplain vegetation acts similarly to streambank vegetation
where f is the resistance, Re the Reynolds number, Fr the and tends to reduce velocities in the floodplain and increase
Froude number, Sw the water surface slope, S0 the channel bed velocities in the main channel, which in turn influences tur-
slope, ks the relative roughness, Lv a nondimensional vege- bulence intensities and shear stress distributions (Wang and
tation parameter representing geometry, J represents flexibility, Wang, 2007).
D the relative submergence of the vegetation, and M is the The velocity differential between the floodplain and the
density of vegetation. Fischenich (2000) developed equations main channel can create a shear layer (Shiono and Knight,
for estimating Manning’s n for unsubmerged and submerged 1991; Wormleaton, 1998). This channel–floodplain interface
floodplain vegetation, noting that the velocity distributions has been modeled numerically as an imaginary vertical wall by
(Figure 6) and, therefore, resistance varied greatly due to several researchers (Pasche and Rouve, 1985; Naot et al., 1996).
depth of flow and vegetation type. Perucca et al. (2009) esti- Intensive vortex shedding occurs at the channel–floodplain
mated the dispersion coefficient for rivers with and without interface due to an intensive momentum exchange between two
floodplain vegetation and concluded that the longitudinal distinct regions of varying velocity (Pasche and Rouve, 1985).
 The Impacts of Vegetation on Roughness in Fluvial Systems 87

2 3

Figure 6 Velocity distribution for submerged and unsubmerged vegetation. Velocity distribution represents vegetation condition to the left.
Reproduced from Fischenich, C., 2000. Resistance due to vegetation. ERDC TN-EMRRP-SR-07. U.S. Army Engineering Research and
Development Center, Vicksburg, MS, with permission from U.S Army.

Local velocities Shear layer

Ud

Momentum transfer Depth-mean

velocities
U
Interface
vortices

Ud
w

b

Flood plain

River channel
w

Direction of flow
b
Secondary
flows

Boundary
shear stresses

Figure 7 Hydraulic parameters associated with overbank flow in a two-stage channel. Reproduced from Shiono, K., Knight, D.W., 1991. Turbulent
open-channel flows with variable depth across the channel. Journal of Fluid Mechanics 222, 617–646, with permission from Cambridge.

McBride et al. (2007) conducted a flume study of forested (TKE) was 2 times greater than during overbank flows with
floodplain vegetation and found that, during overbank flows, grassy vegetation on the floodplain. The point of their study
there was a narrow band of high turbulence between the was to determine how and why channels started out as narrow
floodplain and main channel, and turbulent kinetic energy grass-lined channels (Zimmerman et al., 1967), but eventually
88 The Impacts of Vegetation on Roughness in Fluvial Systems

widened during natural reforestation (McBride et al., 2008). influenced the residence time of water on the floodplain,
They built on these results with field measurements of a thereby increasing settling time. Noe and Hupp (2009) meas-
prototype stream during five peak flows and found significantly ured floodplain sediment and nutrient retention in the Coastal
greater TKE in the reforested reach than in the mature forest or Plain floodplains of the Chesapeake Bay watershed, USA, and
grass-lined reaches. They used this information to develop a found that these floodplains stored large proportions of the
nonlinear conceptual model of incision, widening, and re- river’s annual loads of sediment and nutrients. They estimated
covery of a stream during its transition from grass-lined channel median sediment retention rates of 119% for sediment, 59%
to mature forest (McBride et al., 2008, 2010). for phosphorus, and 22% for nitrogen on floodplains of seven
Yang et al. (2007) found that streamwise velocities have rivers over 1–6 years (Noe and Hupp, 2009). It is important to
logarithmic distributions in nonvegetated floodplains, but that note that values above 100% were possible due to uncertainty
vegetated floodplains followed an S-shaped profile with three in several of the estimation variables. Hupp et al. (2009)
zones (which varied depending on whether woody or non- evaluated the effect of human alterations such as dam, stream
woody vegetation was simulated). The lateral velocity gradient channelization, and levee construction on floodplain geo-
increased with the addition of vegetation, with long grasses morphic processes and concluded that ‘‘human alterations
retarding flow the most. Floodplain vegetation increased ap- typically shift affected streams away from natural dynamic
parent shear stress at the vertical interface between main equilibrium’’ and alter the deposition/erosion processes.
channel and floodplain, and Reynolds stresses became very Li and Millar (2010) modified an existing 2D morphody-
complex in the main channel side-slope zone, particularly at namic model of a gravel-bed river to include floodplain
the floodplain-main channel boundary (Yang et al., 2007). vegetation in order to predict changes in bedload transport
Thornton et al. (2000) set out to quantify this apparent shear and channel morphology due to sediment deposition and
stress at the interface between the main channel and both a erosion. They found that vegetation reduced near-bank and
vegetated and nonvegetated floodplain as a function of fluc- floodplain velocities and helped stabilize bank sediments,
tuations in channel velocities. They noted that most researchers which in turn influenced bedload transport and channel
focused on boundary shear force and gravitational force when morphology (Li and Millar, 2010). Woody floodplain vege-
estimating flow resistance, but that the apparent shear stress tation often formed depositional sites during bedload trans-
must be included to account for turbulence and momentum port events (McKenney et al., 1995).
transfer (Thornton et al., 2000). Their experimental results McBride et al. (2008, 2010) studied the process of channel
showed that the apparent shear stress was greater at the inter- widening on small streams in Vermont, USA, during the pro-
face than within the main channel or floodplain, and that as cess of riparian vegetation succession from grass-lined banks to
vegetation density increased, the apparent shear increased. forest through passive reforestation. They discussed the impli-
Darby (1999) modified a hydraulic model to evaluate flow cations of their findings on the ever-popular stream restoration
resistance and flood potential for channels with nonuniform activities in the USA that typically include riparian reforestation
cross-sections, varying bed material, and riparian vegetation (McBride et al., 2010). Allmendinger et al. (2005) found higher
with and without flexibility. The resultant model accounted rates of deposition and lateral migration in reaches with non-
for lateral shear, but ignored secondary flows. Comparison of forested floodplains versus those with forested floodplains.
model results with field data suggested it could be a useful tool They suggested that differences in width between forested and
for design hydraulic analysis in stream restoration. nonforested reaches were related to a balance between rates of
cut-bank erosion and rates of deposition on convex floodplains
(Allmendinger et al., 2005). In discussing the sediment impacts
of potential channel widening due to riparian reforestation,
12.6.5.3 Floodplain Vegetation and Sediment Transport
Hession et al. (2008) noted that much more research is needed
Floodplains are well known to provide sediment storage in to quantify sediment supply and storage dynamics, particularly
fluvial systems (Phillips, 1989; Steiger et al., 2003; Hupp et al., given the incredible amount of re-vegetation taking place in the
2008, 2009). However, even though numerous publications US (Bernhardt et al., 2005; Hassett et al., 2005).
(e.g., N.R.C., 2002) and government policies (Chesapeake Bay
Executive Council, 2003) advocate riparian or floodplain for-
ests as efficient traps for sediments and other pollutants, very 12.6.6 Future Directions
few studies have quantified actual storage rates (Noe and
Hupp, 2009). Geerling et al. (2007) studied plant succession, This chapter has highlighted the trends and recent research
roughness, and aggradation of an excavated channel over 16 advances in the complexities caused by vegetation in and near
years as vegetation grew back on the floodplain. Softwood channels. Plants create a complicated system of feedbacks and
forest establishment increased sedimentation and an equal linkages between channel flow and morphology, sediment
amount of sediment was excavated and re-deposited over the deposition and erosion, plant morphology, density, and spa-
16 years. The establishment of vegetation reduced overall mean tial extent. As research advances our knowledge in one dir-
flow velocity and had a positive effect on overall sedimen- ection, it opens more questions in another. Technological
tation, thereby creating more diverse flow velocity and sedi- advances have also furthered understanding while reminding
mentation patterns (Geerling et al., 2007). This corresponds to us of the many unknowns still needing investigation.
the findings of Dunne et al. (1998), who conducted a sediment Studies that characterize flow profiles have formed a con-
budget along reaches of the Amazon River and noted that the sensus on the shape of the velocity profile through vegetation.
hydraulic roughness of floodplain (due to vegetation) Where the flow is within a vegetated area, the velocity is
 The Impacts of Vegetation on Roughness in Fluvial Systems 89

uniform following an inflection point near the bed, and vel- in Earth Science (MYRES III) focused on Dynamic Interactions
ocities are much reduced when compared to open water flows. of Life and its Landscape (Reinhardt et al., 2010). They iden-
Profiles through submerged vegetation consist of a uniform tified two broad themes, one of which is relevant to our dis-
profile within the plants that transitions to a logarithmic cussion here – the co-evolution of landforms and biological
profile in the free channel flow by an inflection point just communities. This meeting and two manuscripts by the 2009
above a plant canopy (refer to Figure 6). This profile de- BGS keynote speakers (Marston, 2010; Osterkamp and Hupp,
scription holds in general but adjustments must be made for 2010) are excellent source for identifying future directions for
each specific channel as a result of feedbacks between the research. Reviewing these manuscripts provided some inspir-
vegetation flexibility, density, and spatial extent in the channel. ation for future research needs and opportunities related to
Similarly, frictional resistance due to vegetation can be par- vegetation and roughness:
ametrized on a general scale, but the complications due to
1. Use the ongoing river and stream restoration activities to
plant flexibility and morphology make quantification on a
our advantage. Conduct post-project monitoring over large
local scale difficult. Given the number of variables interacting
spatial and temporal scales to help understand vegetation
to create roughness over a channel and floodplain, Gurnell
and geomorphology more fully. In the case of roughness,
et al. (2010) speculated that a multivariate methodology needs
many of these restoration projects remove existing vege-
to be developed. Recent research has also begun to investigate
tation, reshape or re-engineer the stream morphology, and
the interaction between sediment and vegetation, including
plant native vegetation. This vegetation will take years to
sediment transport, erosion, and deposition in vegetated
mature, all the while changing its structure and density,
areas, but these interactions remain poorly quantified.
which will influence roughness, hydraulics, and sediment
Research has followed two basic approaches. One ap-
transport and deposition.
proach focuses on a single aspect of the vegetated channel
2. Develop restoration plans through collaborations between
system, working to explain processes occurring at that spatial
engineers, ecologists, and geomorphologists. Also, where
scale. For example, studies of fluid exchange across the vege-
possible, develop the designs to test various hypotheses
tation interface. The second approach studies over a larger
about how the stream morphology, flow hydraulics, and
spatial scale, working to establish the feedbacks operating
vegetation will respond. This could be done by developing
between vegetation and the channel system. For example, the
restoration plans that vary longitudinally, or employing
feedbacks between processes at work along a vegetated chan-
paired reach studies across similar watersheds.
nel bank and the evolution of channel width and morphology.
Progress employing both methods is needed to develop a full The practical need for improved understanding of and
understanding of how vegetation influences a channel system. prediction of the processes at work in vegetated systems comes
Small scale research is necessary to produce a process-based from the use of a river as an ecological engineer. Gurnell et al.
understanding of the connections between flow hydraulics (2010) mentioned the need to consider river management
and sediment and plant mechanics. Larger scale research is schemes that include mowing of aquatic vegetation and
needed to understand how the flow, vegetation, and sediment channel dredging. By imposing a control over vegetation
interact over a range of spatial and temporal scales. growth, management may be preventing the evolution of the
As described in Tsujimoto (1999), vegetated channels are channel form. It is possible that, if left to grow, the vegetation
an inter-connected system of feedbacks between the flow, would form dense patches. These patches could trap large
sediment transport, geomorphology, and vegetation. Advances enough volumes of fine sediment to induce landform devel-
will be made both through individual research and from the opment around the plants and eventual island formation,
gathering and exchange of ideas between those working at the which would alter the channel morphology away from that of
various research scales and methods. The 2009 Binghamton a straight channel and toward a braided morphology (Walter
Geomorphology Symposium (BGS) serves as an example of and Merritts, 2008). However, it is not known what flow re-
the type of exchange needed to recur for progress to continue gimes would be required to induce some deposition but not
(Hession et al., 2010). The 2009 BGS was similar in theme to choke the channel with sediment. This is analogous to the
the 1995 BGS focused on biogeomorphology (Hupp et al., current use of vegetation in river restoration. Much is being
1995) and represented a gathering of new researchers as well done, both in terms of growing vegetation on the banks and
as veterans from the 1995 symposium. The 2009 symposium removing it from the channel, without a clear understanding
focused on the interactions, dependencies, and feedback loops of how the channel responds to changes in one part of the
at work in vegetated systems, and brought together a diverse feedback loop between vegetation, flow, sediment, and chan-
group of researchers from a range of disciplines to advance the nel form.
fundamental and applied thinking on vegetated fluvial sys-
tems. A common theme throughout the symposium was the
difficulty of inferring processes in these inherently complex
systems due to the feedbacks between vegetation and the References
landscape. Cause and effect relationships were rarely clear and
the general feeling of the attendees was that we must use Abt, S.R., Clary, W.P., Thornton, C.I., 1994. Sediment deposition and entrapment in
multiple research methods across disciplines to begin to vegetated streambeds. Journal of Irrigation and Drainage Engineering 120(6),
1098–1111.
understand these complex systems (Hession et al., 2010). Afzalimehr, H., Dey, S., 2009. Influence of bank vegetation and gravel bed on
The lack of and need for collaborations across ecology and velocity and Reynolds stress distributions. International Journal of Sediment
geomorphology led to the 2008 Meeting of Young Researchers Research 24(2), 236–246.
90 The Impacts of Vegetation on Roughness in Fluvial Systems

Allmendinger, N.E., Pizzuto, J., Potter, Jr. N., Johnson, T.E., Hession, W.C., 2005. Defina, A., Bixio, A.C., 2005. Mean flow and turbulence in vegetated open channel
The influence of riparian vegetation on stream width, eastern Pennsylvania, USA. flow. Water Resources Research 41, W07006.
GSA Bulletin 117(1/2), 229–243. Defra/EA, 2003. Uncertainty in River Flood Conveyance, Phase 2: Conveyance
Anderson, B.G., 2006. Quantifying the interaction between riparian vegetation and Manual, , 2003. Flood and Coastal Defence R&D Programme. Defra/
flooding: from cross-section to catchment scale. Dissertation in support of Environment Agency, Wallingford, UK.
doctoral degree. School of Anthropology, Geography and Environmental Studies, Downes, B.J., Lake, P.S., Glaister, A., Webb, J.A., 1998. Scales and frequencies of
University of Melbourne, 529 pp. disturbances: rock size, bed packing and variation among upland streams.
Anderson, B.G., Andersen, S., Bishop. W.A., 2007. Hydraulic analysis report. Report Freshwater Biology 40, 625–639.
by Water Technology to Price Merrett Consulting for the North Central CMA. Dunne, T., Mertes, L.A.K., Meade, R.H., Richey, J.E., Forsberg, B.R., 1998.
Huntly, Victoria, 39 pp. Exchanges of sediment between the flood plain and channel of the Amazon
Anderson, B.G., Rutherfurd, I.D., Western, A.W., 2006. An analysis of the influence River in Brazil. GSA Bulletin 110, 450–467.
of riparian vegetation on the propagation of flood waves. Environmental Fischenich, C., 2000. Resistance due to vegetation. ERDC TN-EMRRP-SR-07. U.S.
Modelling and Software 21(9), 1290–1296. Army Engineering Research and Development Center, Vicksburg, MS.
Antonarakis, A.S., Richards, K.S., 2010. Determining leaf area index and leafy tree Forzieri, G., Moser, G., Vivoni, E.R., Castelli, F., Canovaro, F., 2010. Riparian
roughness using terrestrial laser scanning. Water Resources Research 46(6), vegetation mapping for hydraulic roughness estimation using very high
W06510. resolution remote sensing data fusion. Journal of Hydraulic Engineering
Arcement, Jr. G.J., Schneider, V.R., 1989. Guide for selecting Manning’s roughness 136(11), 855–867.
coefficients for natural channels and floodplains. USGS Water Supply Paper Fox, J.F., Belcher, B.J., 2009. Comparison of LSPIV, ADV, and PIV Data that is
2339, 38 pp. Decomposed to Measure the Structure of Turbulence Over a Gravel-Bed. 33rd
Asaeda, T., Rajapakse, L., Kanoh, M., 2010. Fine sediment retention as affected by IAHR Congress: Water Engineering for a Sustainable Environment. International
annual shoot collapse: Sparganium erectum as an ecosystem engineer in a Association of Hydraulic Engineering and Research (IAHR), Vancouver, Canada,
lowland stream. River Research and Applications 26(9), 1153–1169. pp. 2288–2295.
Baptist, M.J., 2003. A flume experiment on sediment transport with flexible, Freeman, G.E., Rahmeyer, W.H., Copeland, R.R., 2000. Determination of resistance
submerged vegetation. International workshop on Riparian Forest Vegetated due to shrubs and woody vegetation. ERDC/CHL TR-00-25, Coastal and
Channels: Hydraulic, Morphological, and Ecological Aspects, 20–22 February, Hydraulics Laboratory, U.S. Army Engineer Research and Development Center,
2003. International Association for Hydro-Environment Engineering and Research Vicksburg, MS, 63 pp.
(IAHR), Trento, Italy, 12 pp. Gambi, M.C., Nowell, A.R.M., Jumars, P.A., 1990. Flume observations on flow
Baptist, M.J., Babovic, V., Rodriguez Uthurburu, J., Keijzer, M., Uittenbogaard, R.E., dynamics in Zostera marina (eel grass) beds. Marine Ecology Progress Series
Mynett, A. et al., 2006. On inducing equations for vegetation resistance. Journal 61, 159–169.
of Hydraulic Research 45(4), 435–450. Geerling, G.W., Labrador-Garcia, M., Clevers, J.G.P.W., Ragas, A.M.J., Smits, A.J.M.,
Barfield, B.J., Tollner, E.W., Haynes, J.C., 1979. Filtration of sediment by simulated 2007. Classification of floodplain vegetation by data fusion of spectral (CASI)
vegetation I. Steady-state flow with homogeneous sediment. Transaction of the and LiDAR data. International Journal of Remote Sensing 28(19), 4263–4284.
ASAE 22(3), 540–548. Ghisalberti, M., Nepf, H.M., 2004. The limited growth of vegetated shear layers.
Barnes, Jr. H.H., 1967. Roughness characteristics of natural channels. USGS Water Water Resources Research 40(W07502), 1–12.
Supply Paper 1849, 213 pp. Gippel, C.J., 1999. Environmental hydraulics of large woody debris in streams and
Beisel, J., Usseglio-Polatera, P., Moreteau, J., 2000. The spatial heterogeneity of a rivers. Journal of Environmental Engineering 5, 388–395.
river bottom: a key factor determining macroinvertebrate communities. Girard, P., Fantin-Cruz, I., Loverde de Oliveira, S.M., Hamilton, S.K., 2010. Small-
Hydrobiologia 422(0), 163–171. scale spatial variation of inundation dynamics in a floodplain of the Pantanal
Bennett, S.J., Pirim, T., Barkdoll, B.D., 2002. Using simulated emergent vegetation (Brazil). Hydrobiologia 638(1), 223–233.
to alter stream flow direction within a straight experimental channel. Green, J.C., 2004. Modelling flow resistance in vegetated stream: review and
Geomorphology 44, 115–126. development of new theory. Hydrological Processes 19(6), 1245–1259.
Bennett, S.J., Simon, A., 2004. Riparian Vegetation and Fluvial Geomorphology. Gregory, K.J., 1992. Vegetation and river channel processes. In: Boon, P.J., Calow,
Water Science and Application, American Geophysical Union, Washington, DC, P., Petts, G.E. (Eds.), River Conservation and Management. Wiley, Chichester,
2004. pp. 255–269.
Bernhardt, E.S., Palmer, M.A., Allan, J.D., et al., 2005. Synthesizing U.S. river Gurnell, A.M., O’Hare, J.M., O’Hare, M.T., Dunbar, M.J., Scarlett, P.M., 2010. An
restoration efforts. Science 308, 636–637. exploration of associations between assemblages of aquatic plant morphotypes
Carollo, F.G., Ferro, V., Termini, D., 2002. Flow velocity measurements in vegetated and channel geomorphological properties within British rivers. Geomorphology
channels. Journal of Hydraulic Engineering 128(7), 664–673. 116(1–2), 135–144.
Carollo, F.G., Ferro, V., Termini, D., 2005. Flow resistance law in channels with Gurnell, A.M., van Oosterhout, M.P., de Vlieger, B., Goodson, J.M., 2006. Reach-
flexible submerged vegetation. Journal of Hydraulic Engineering 131(7), scale interactions between aquatic plants and physical habitat: River Frome,
554–564. Dorset. River Research and Application 22(6), 1535–1467.
Chesapeake Bay Executive Council, 2003. Expanded Riparian Forest Buffer Goals. Harris, E.L., Babovic, V., Falconer, R.A., 2003. Velocity prediction in compound
Directive 03-01. Annapolis, MD, 1 pp. channels with vegetated floodplains using genetic programming. International
Chow, V.T., 1959. Open-Channel Hydraulics. McGraw-Hill, New York, NY, 680 pp. Journal of River Basin Management 1(2), 117–123.
Coby, D.M., Mason, D.C., Horritt, M.S., Bates, P.D., 2002. Two dimensional Hart, D.P., 1998. High-speed PIV analysis using compressed image correlation.
hydraulic flood modeling using a finite element mesh decomposed according to Journal of Fluids Engineering 120, 463–470.
vegetation and topographic features derived from airborne scanning laser Hassett, B., Palmer, M., Bernhardt, E., Smith, S., Carr, J., Hart, D., 2005. Restoring
altimetry. Hydrological Processes 17(10), 1977–2000. watersheds project by project: trends in Chesapeake Bay tributary restoration.
Coon, W.F., 1995. Estimates of Roughness Coefficients for Selected Natural Stream Frontiers in Ecology and the Environment 3(5), 259–267.
Channels with Vegetated Banks in New York. United States Geological Survey, Helmio, T., 2002. Unsteady 1D flow model of compound channel with vegetated
Ithaca, NY, 127 pp. floodplains. Journal of Hydrology 269(1–2), 89–99.
Coon, W.F., 1998. Estimation of Roughness Coefficients for Natural Stream Helmio, T., 2004. Flow resistance due to lateral momentum transfer in partially
Channels with Vegetated Banks. United States Geological Survey, Washington, vegetated rivers. Water Resources Research 40(W05206), 1–10.
DC. Hession, W.C., Curran, J.C., Resler, L.M., Wynn, T.M., 2010. Preface:
Cotton, J.A., Wharton, G., Bass, J.A.B., Heppell, C.M., Wotton, R.S., 2006. The geomorphology and vegetation: interactions, dependencies, and feedback loops.
effects of seasonal changes to in-stream vegetation cover on patterns of flow Geomorphology 116(3–4), 203–205.
and accumulation of sediment. Geomorphology 77(3–4), 320–334. Hession, W.C., McBride, M., Pizzuto, J.E., 2008. Riparian vegetation influence on
Cowen, W.L., 1956. Estimating hydraulic roughness coefficients. Agricultural channel morphology. AWRA Summer Specialty Conference, Riparian Ecosystems
Engineering 37, 473–475. and Buffers: Working at the Water’s Edge. Virginia Beach, VA.
Darby, S.E., 1999. Effect of riparian vegetation on flow resistance and flood Hicks, D.M., Mason, P.D., 1998. Roughness Characteristics of New Zealand Rivers.
potential. Journal of Hydraulic Engineering 125(5), 443–454. National Institute of Water and Atmospheric Research. Christchurch, NZ.
Darby, S.E., Thorne, C.R., 1995. Fluvial maintenance operations in managed alluvial Hopkinson, L.C., Wynn, T.M., 2009. Vegetation impacts on near bank flow.
rivers. Aquatic Conservation: Marine and Freshwater Ecosystems 5(1), 37–54. Ecohydrology 2, 404–418.
 The Impacts of Vegetation on Roughness in Fluvial Systems 91

Hughes, F.M.R., Moss, T.M., Richards, K.S., 2008. Uncertainty in Riparian and Limerinos, J.T., 1970. Determination of the Manning Coefficient from Measured Bed
Floodplain Restoration. In: Darby, S.E., Sear, D.A. (Eds.), River Restoration - Roughness in Natural Channels. United States Geological Survey, Washington,
Managing the Uncertainty in Restoring Physical Habitat. Wiley, West Sussex, DC.
pp. 79–104. Liu, D., Diplas, P., Faribanks, J.D., Hodges, C.C., 2008. An experimental study of
Hupp, C.R., Demas, C.R., Kroes, D.E., Day, R.H., Doyle, T.W., 2008. Recent flow through rigid vegetation. JGR Earth Surface 113, F04015.
sedimentation patterns within the central Atchafalaya Basin, Louisiana. Wetlands Liu, D., Diplas, P., Hodges, C.C., Fairbanks, J.D., 2010. Hydrodynamics of
28, 125–140. flow through double layer rigid vegetation. Geomorphology 116(3–4),
Hupp, C.R., Osterkamp, W.R., 1996. Riparian vegetation and fluvial geomorphic 286–296.
processes. Geomorphology 14, 277–295. Luhar, M., Rominger, J., Nepf, H.M., 2008. Interaction between flow, transport and
Hupp, C.R., Osterkamp, W.R., Howard, A.D. (Eds.), 1995. Biogeomorphology – vegetation spatial structure. Environmental Fluid Mechanics 8, 423–439.
Terrestrial and Freshwater Systems. Geomorphology (special issue) 13, 347 pp. Manning, R., 1890. On the flow of water in open channels and pipes. Transactions
Hupp, C.R., Pierce, A.R., Noe, G.B., 2009. Floodplain geomorphic processes and of the Institute of Civil Engineers of Ireland 20, 161–207.
environmental impacts of human alteration along Coastal Plain rivers, USA. Marston, R.A., 2010. Geomorphology and vegetation on hillslopes: interactions,
Wetlands 29, 413–429. dependencies, and feedback loops. Geomorphology 116(3–4), 206–217.
Hupp, C.R., Simon, A., 1991. Bank accretion and the development of vegetated McBride, M., Hession, W.C., Rizzo, D.M., 2008. Riparian reforestation and channel
deposition surfaces along modified alluvial channels. Geomorphology 4, 111–124. change: a case study of two small tributaries to Sleepers River, northeastern
Hurther, D., Mignot, E., Barthelemy, E., 2009. On the structure of turbulent and Vermont, USA. Geomorphology 102(3–4), 445–459.
dispersive shear and associated Turbulent Kinetic Energy (TKE) flux across the McBride, M., Hession, W.C., Rizzo, D.M., 2010. Riparian reforestation and channel
roughness sublayer of a gravel-bed open-channel flow. 33rd IAHR Congress: change: how long does it take? Geomorphology 116(3–4), 330–340.
Water Engineering for a Sustainable Environment, International Association for McBride, M., Hession, W.C., Rizzo, D.M., Thompson, D.M., 2007. The influence of
Hydro-Environment Engineering and Research (IAHR), Vancouver, Canada, riparian vegetation on near-bank turbulence: a flume experiment. Earth Surface
pp. 837–844. Processes and Landforms 32(13), 2019–2037.
Jarvela, J., 2002. Flow resistance of flexible and stiff vegetation: a flume study with McKenney, R., Jacobson, R.B., Wertheimer, R.C., 1995. Woody vegetation and
natural plants. Journal of Hydrology 269, 44–54. channel morphogenesis in low-gradient, gravel-bed streams in the Ozark
Jarvela, J., 2004. Flow resistance in environmental channels: focus on vegetation. Plateaus, Missouri and Arkansas. Geomorphology 13(1–4), 175–198.
Dissertation in support of Doctor of Science in Technology. Helsinki University Mertes, L.A.K., Daniel, D.L., Melack, J.M., Nelson, B., Martinelli, L.A., Forsberg,
of Technology, Espoo, Finland. B.R., 1995. Spatial patterns of hydrology, geomorphology, and vegetation on the
Jarvela, J., 2005. Effect of submerged flexible vegetation on flow structure and floodplain of the Amazon River in Brazil from remote sensing perspective.
resistance. Journal of Hydrology 307(1–4), 233–241. Geomorphology 13, 215–232.
Jennings, G.D., Harman, W.A., Clinton, D.R., Patterson, J.L., 1999. Stream Moody, J.A., Pizzuto, J.E., Meade, R.H., 1999. Ontology of a floodplain. GSA
Restoration Design Experiences in North Carolina. ASAE Paper 99-2022. ASAE, Bulletin 111, 291–303.
St. Joseph, MI. Mudd, S.M., D’Alpaos, A., Morris, J.T., 2010. How does vegetation affect
Johnson, S.L., 2004. Factors influencing stream temperatures in small streams: sedimentation on tidal marshes? Investigating particle capture and hydrodynamic
substrate effects and a shading experiment. Canadian Journal of Fisheries and controls on biologically mediated sedimentation. Journal of Geophysical
Aquatic Sciences 61, 913–923. Research 110(F03029), 14.
Jordanova, A.A., James, C.S., 2003. Experimental study of bed load transport Muhar, S., 1996. Habitat improvement of Austrian rivers with regard to different
through emergent vegetation. Journal of Hydraulic Engineering 129(6), scales. Regulated Rivers: Research and Management 12(4–5), 471–482.
474–478. Murphy, E., Ghisalberti, M., Nepf, H.M., 2007. Model and laboratory study of
Kadlec, R.H., 1990. Overland flow in wetlands: vegetation resistance. Journal of dispersion in flows with submerged vegetation. Water Resources Research 43,
Hydraulic Engineering 116(5), 691–706. W05438.
Kean, J.W., Smith, J.D., 2004. Flow and boundary shear stress in channels with Naot, D., Nezu, I., Nakagawa, H., 1996. Hydrodynamic behavior of partly vegetated
woody bank vegetation. In: Bennett, S.J., Simon, A. (Eds.), Riparian Vegetation open channels. Journal of Hydraulic Engineering 122, 625–633.
and Fluvial Geomorphology. American Geophysical Union, Washington, DC, pp. Nepf, H.M., 1999. Drag, turbulence, and diffusion in flow through emergent
237–252. vegetation. Water Resources Research 35(2), 479–489.
Kim, J., Lee, C., Kim, W., Kim, Y., 2010. Roughness coefficient and its uncertainty Nepf, H.M., Ghisalberti, M., 2008. Flow and transport in channels with submerged
in gravel-bed river. Water Science and Engineering 3(2), 217–232. vegetation. Acta Geophysica 56(3), 753–777.
Klaar, M.J., Maddock, I., Milner, A.M., 2009. The development of hydraulic and Nepf, H.M., Sullivan, J.A., Zavistoski, R.A., 1997. A Model for Diffusion Within
geomorphic complexity in recently formed streams in Glacier Bay National Park, Emergent Vegetation. Limnology and Oceanography 42(8), 1735–1745.
Alaska. River Research and Applications 25(10), 1331–1338. Nepf, H.M., Vivoni, E.R., 2000. Flow structure in depth-limited, vegetated flow.
Kouwen, N., Li, R., 1980. Biomechanics of vegetative channel linings. Journal of Journal of Geophysical Research 105(C12), 28547–28557.
the Hydraulics Division 106(6), 1085–1103. Nikora, V.I., Larned, S., Nikora, N., Debnath, K., Cooper, G., Reid, M., 2008.
Kouwen, N., Unny, T.E., 1973. Flexible roughness in open channels. ASCE Journal Hydraulic resistance due to aquatic vegetation in small streams: field study.
of the Hydraulics Division 99(HY5), 713–728. Journal of Hydraulic Engineering 134(9), 1326–1332.
Kouwen, N., Unny, T.E., Hill, H.M., 1969. Flow retardance in vegetated channels. Noe, G.B., Hupp, C.R., 2009. Retention of riverine sediment and nutrient loads by
Journal of the Irrigation and Drainage Division 95. IR2 Proceedings Paper 6633, coastal plain floodplains. Ecosystems 12, 728–746.
329–342. N.R.C., 1992. Restoration of Aquatic Ecosystems: Science, Technology and Public
Lang, S., Ladson, T., Brett, A., 2004. A review of empirical equations for estimating Policy. National Academy Press, Washington, DC.
stream roughness and their application to four streams in Victoria. Australian N.R.C., 2002. Riparian Areas: Functions and Strategies for Management. National
Journal of Water Resources 8(1), 69–82. Academy Press, Washington, DC.
Lawler, D.M., 2008. Advances in the continuous monitoring of erosion and N.R.C., 2007. River Science at the U.S. Geological Survey. National Academy Press,
deposition dynamics: developments and applications of the new PEEP-3T Washington, DC.
system. Geomorphology 93(1–2), 17–39. Osterkamp, W.R., Hupp, C.R., 1984. Geomorphic and vegetative characteristics
Leopold, L.B., 1994. A View of the River. Harvard University Press. Cambridge, MA. along three northern Virginia streams. GSA Bulletin 95, 1093–1101.
Leopold, L.B., Wolman, M.G., Miller, J.P., 1964. Fluvial Processes in Osterkamp, W.R., Hupp, C.R., 2010. Fluvial processes and vegetation – Glimpses of
Geomorphology. Dover Publications, New York, NY. the past, the present, and perhaps the future. Geomorphology 116(3–4),
Leu, J.M., Chan, H.C., Jia, Y., He, Z., Wang, S.S.Y., 2008. Cutting management of 274–285.
riparian vegetation by using hydrodynamic model simulations. Advances in Pasche, E., Rouve, G., 1985. Overbank flow with vegetatively roughened flood
Water Resources 31(10), 1299–1308. plains. Journal of Hydraulic Engineering 111(9), 1262–1278.
Li, R., Shen, H.W., 1973. Effect of tall vegetations on flow and sediment. Journal of Perucca, E., Camporeale, C., Ridolfi, L., 2009. Estimation of the dispersion
the Hydraulics Division HY5, 793–814. coefficient in rivers with riparian vegetation. Advances in Water Resources 32(1),
Li, S.S., Millar, R.G., 2010. A two-dimensional morphodynamic model of gravel-bed 78–87.
river with floodplain vegetation. Earth Surface Processes and Landforms 36, Petryk, S., 1969. Drag on cylinders in open channel flow. Dissertation in support of
190–202. a Doctoral Degree. Colorado State University, Fort Collins, CO.
92 The Impacts of Vegetation on Roughness in Fluvial Systems

Petryk, S., Bosmajian, G., 1975. Analysis of flow through vegetation. Journal Thornton, C.I., Abt, S.R., Morris, C.E., Fischenich, J.C., 2000. Calculating shear
Hydraulics Division 101(HY7), 871–884. stress at channel-overbank interfaces in straight channels with vegetated
Phelps, H.O., 1970. The friction coefficient for shallow flows over a simulated turf floodplains. Journal of Hydraulic Engineering 126(12), 929–936.
surface. Water Resources Research 6(4), 1220–1226. Tollner, E.W., Barfield, B.J., Hayes, J.C., 1982. Sedimentology of erect vegetal
Phillips, J.D., 1989. Fluvial sediment storage in wetlands. Water Resources Bulletin filters. Journal of the Hydraulics Division 108(12), 1518–1531.
25, 867–873. Trimble, S.W., 2004. Effects of riparian vegetation on stream channel stability and
Pizzuto, J., O’Neal, M., Stotts, S., 2010. On the retreat of forested, cohesive sediment budgets. In: Bennett, S.J., Simon, A. (Eds.), Riparian Vegetation and
riverbanks. Geomorphology 116(3–4), 341–352. Fluvial Geomorphology. American Geophysical Union, Washington, DC, pp.
Pollen-Bankhead, N., Simon, A., 2010. Hydrologic and hydraulic effects of riparian 153–169.
root networks on streambank stability: is mechanical root-reinforcement the Tsujimoto, T., 1999. Fluvial processes in streams with vegetation. Journal of
whole story? Geomorphology 116(3–4), 353–362. Hydraulic Research 37(6), 789–803.
Reinhardt, L., Jerolmack, D., Cardinale, B.J., Vanacker, V., Wright, J., 2010. U.S. Army Corps of Engineers, 1997. Channel stability assessment for flood control
Dynamic interactions of life and its landscape: feedbacks at the interface of projects (Technical engineering and design guides as adapted from the U.S.
geomorphology and ecology. Earth Surface Processes and Landforms 35, Army Corps of Engineers, No. 20). American Society of Civil Engineers 978-0-
78–101. 7844-0201-6. ASCE Press, New York, NY, 108 pp.
Richard, G.A., Julien, P.Y., Baird, D.C., 2005. Statistical analysis of lateral migration U.S.E.P.A., 1999. Riparian Forest Buffers: Linking Land and Water. Chesapeake Bay
of the Rio Grande, New Mexico. Geomorphology 71, 139–155. Program. U.S. Environmental Protection Agency, Annapolis, MD, 16 pp.
Righetti, M., 2008. Flow analysis in a channel with flexible vegetation using Walter, R.C., Merritts, D.J., 2008. Natural streams and the legacy of water-powered
double-averaging method. Acta Geophysica 56(3), 801–823. mills. Science 319(5861), 299–304.
Rosgen, D., 1996. Applied River Morphology. Wildland Hydrology, Pagosa Springs, Wang, C., Wang, P., 2007. Hydraulic resistance characteristics of riparian reed zone
CO. in river. Journal of Hydraulic Engineering 12(3), 267–272.
Sand-Jensen, K., 1998. Influence of submerged macrophytes on sediment Wang, C., Yu, J., Wang, P., Guo, P., 2009. Flow structure of partly vegetated open-
composition and near-bed flow in lowland streams. Freshwater Biology 39, channel flows with eelgrass. Journal of Hydrodynamics, Series B 21(3),
663–667. 301–307.
Serra, T., Fernando, H.J.S., Rodriguez, R.V., 2004. Effects of emergent vegetation on Werner, M.G.F., Hunter, N.M., Bates, P.D., 2005. Identifiability of distributed
lateral diffusion in wetlands. Water Research 38, 139–147. floodplain roughness values in flood extent estimation. Journal of Hydrology
Sharpe, R.G., James, C.S., 2006. Deposition of sediment from suspension in 314(1–4), 139–157.
emergent vegetation. Water SA 32(2), 211–218. White, B.L., Nepf, H.M., 2008. A vortex-based model of velocity and shear stress in
Shi, Z., Hughes, J.M.R., 2002. Laboratory flume studies of microflow environments a partially vegetated shallow channel. Water Resources Research 44, W10412.
of aquatic plants. Hydrological Processes 16, 3279–3289. Wilson, C.A.M.E., 2007. Flow resistance models for flexible submerged vegetation.
Shields, F.D., Bowie, A.J., Cooper, C.M., 1995. Control of stream bank erosion due Journal of Hydrology 342, 213–222.
to bed degradation with vegetation and structure. JAWRA 31(3), 475–489. Wilson, C.A.M.E., Stoesser, T., Bates, P.D., Batemann Pinzen, A., 2003. Open
Shiono, K., Knight, D.W., 1991. Turbulent open-channel flows with variable depth channel flow through different forms of submerged flexible vegetation. Journal
across the channel. Journal of Fluid Mechanics 222, 617–646. of Hydraulic Engineering 129(11), 847–853.
Shucksmith, J.D., Boxall, J.B., Guymer, I., 2010. Effects of emergent and submerged Wormleaton, P.R., 1998. Floodplain secondary circulation as a mechanism for flow
natural vegetation on longitudinal mixing in open channel flow. Water Resources and shear stress redistribution in straight and compound channels.
Research 46, W04504. In: Ashworth, P., Bennett, S., Best, J.L., McLelland, S. (Eds.), Coherent Flow
Simon, A., Collison, A.J.C., 2002. Quantifying the mechanical and hydrologic Structures in Open Channels. Wiley, Chichester, pp. 581–608.
effects of riparian vegetation on streambank stability. Earth Surface Processes Wu, F.-C., Shen, H.W., Chou, Y.-J., 1999. Variation of roughness coefficients for
and Landforms 27, 527–546. unsubmerged and submerged vegetation. Journal of Hydraulic Engineering
Simon, A., Dickerson, W., Heins, A., 2004. Suspended-sediment transport rates 125(9), 934–942.
at the 1.5-year recurrence interval for ecoregionson the United States: Wynn, T.M., Mostaghimi, S., 2006a. Effects of riparian vegetation on stream bank
transport conditions at the bankfull and effective discharge? Geomorphology 58, subaerial processes in southwestern Virginia, USA. Earth Surface Processes and
243–262. Landforms 31, 399–413.
Steiger, J., Gurnell, A.M., Goodson, J.M., 2003. Quantifying and characterizing Wynn, T.S., Mostaghimi, S., 2006b. The effects of vegetation and soil type on
contemporary riparian sedimentation. River Research and Applications 19, streambank erosion, southwestern Virginia, USA. Journal of the American Water
335–352. Resources Association 42(1), 69–82.
Stephan, U., Gutknecht, D., 2002. Hydraulic resistance of submerged flexible Yang, K., Cao, S., Knight, D.W., 2007. Flow patterns in compound channels with
vegetation. Journal of Hydrology 269, 27–43. vegetated floodplains. Journal of Hydraulic Engineering 133, 148–159.
Sullivan, S.M.P., Watzin, M.C., Hession, W.C., 2006. Influence of stream Yen, B.C., 2002. Open channel flow resistance. Journal of Hydraulic Engineering
geomorphic condition on fish communities in Vermont, U.S.A. Freshwater 128(1), 20–39.
Biology 51(10), 1811–1826. Zimmerman, R.C., Goodlett, J.C., Comer, G.H., 1967. The Influence of Vegetation
Tal, M., Paola, C., 2007. Dynamic single-thread channels maintained by the on Channel Form of Small Streams. International Association of Scientific
interaction of flow and vegetation. Geology 35(4), 347–350. Hydrology, Gentbrugge, Belgium.
Thorne, S.D., Furbish, D.J., 1995. Influences of coarse bank roughness on flow Zong, L., Nepf, H.M., 2010. Flow and deposition in and around a finite patch of
within a sharply curved river bend. Geomorphology 12(3), 241–257. vegetation. Geomorphology 116(3–4), 363–372.
 The Impacts of Vegetation on Roughness in Fluvial Systems 93

Biographical Sketch
Dr. Cully Hession is an associate professor in the Center for Watershed Studies within the department of Bio-
logical Systems Engineering at Virginia Tech. He is a professional engineer in Virginia with over 25 years of
experience in watershed hydrology/modeling and more than 15 years experience in stream morphology and
aquatic ecosystem research, education, assessment, and design. Dr. Hession’s main research involves cross-dis-
ciplinary efforts to understand the influence of watershed and riparian disturbance on channel structure, and
linking physical channel conditions to aquatic ecosystem structure and function.

Joanna Crowe Curran is a fluvial geomorphologist and engineer. Her research focuses on the mechanics of
sediment transport in open channel hydraulics, particularly on how changes in channel roughness and sediment
supply effect the channel reach, whether these changes are due to dam removals or bed armoring. Her research
has spans channels of both steep and low gradient, from step pool bedform formation processes to large woody
debris transport in coastal areas.
 12.7 Vegetation Ecogeomorphology, Dynamic Equilibrium, and Disturbance
CR Hupp, US Geological Survey, Reston, VA, USA
WR Osterkamp, US Geological Survey, Tucson, AZ, USA
Published by Elsevier Inc.

12.7.1 Introduction 94
12.7.2 Vegetation Patterns 95
12.7.3 Hillslopes 96
12.7.4 Riparian Vegetation, Fluvial Processes, and Landforms 96
12.7.4.1 Systems in Dynamic Equilibrium 98
12.7.4.2 Systems in Nonequilibrium States 99
12.7.5 Dynamic Equilibrium and the Erosional–Depositional Environment 102
12.7.6 Summary 103
Acknowledgments 104
References 104

Glossary Ruderal Growing where natural vergetational cover has

Propagules A cutting, seed or spore that propagates a been disturbed by humans.
plant.
Regime state Forming of channels in material carried by
streams.

Abstract

Early ecologists understood the need to document geomorphic form and process to explain plant species distributions.
Although this relationship has been acknowledged for over a century, with the exception of a few landmark papers, only the
past few decades have experienced intensive research on this interdisciplinary topic. Here the authors provide a summary of
the intimate relations between vegetation and geomorphic/process on hillslopes and fluvial systems. These relations are
separated into systems (primarily fluvial) in dynamic equilibrium and those that are in nonequilibrium conditions in-
cluding the impacts of various human disturbances affecting landforms, geomorphic processes, and interrelated, attendant
vegetation patterns and processes. The authors conclude with a conceptual model of stream regime focusing on sediment
deposition, erosion, and equilibrium that can be expanded to organize and predict vegetation patterns and life history
strategies.

12.7.1 Introduction between vegetation and geomorphic processes has been ac-
knowledged for over a century. Reviews on the subject
Biota may interact with geomorphic processes in a variety of of biogeomorphology include Viles (1988), Thorne (1990),
ways. Appreciation for this interaction has led to the devel- Hupp et al. (1995), Gurnell and Gregory (1995), Steiger et al.
opment of a field of endeavor termed ‘biogeomorphology’ (2005), and Hession et al. (2010). The field, including efforts
(Viles, 1988). Rooted plants for obvious reasons, bacteria and described as eco-geomorphology, has expanded at a rapid rate
their effects on weathering, beaver (Castor) and their effects on in the past 25 years. Because it would be impossible to treat
sedimentation and channel dynamics, and bioturbation from fairly all the forms of biota in various physiographic and
other organisms are examples of how biota play a profound climatic settings within the confines of a single book chapter,
role in geomorphic processes (Rostan et al., 1987; Butler and we will focus on woody vegetation and geomorphology.
Malanson, 1995; Viles, 1995; Schwarz et al., 1996; Hupp and The present chapter largely limits discussions to vegetation
Osterkamp, 1996). Early ecologists understood the need to and fluvial geomorphic relations in humid, temperate regions
document geomorphic form and process to explain plant of eastern North America and Western Europe, except for a
species distributions (Cowles, 1901). Thus, the relationship small section on important hillslope interactions. Vegetation
can be used as a tool for geomorphic interpretation (Hupp
and Bornette, 2003) in at least two major ways: (1) through
Hupp, C.R., Osterkamp, W.R., 2013. Vegetation ecogeomorphology,
dendrogeomorphic analyses (tree rings) to estimate the
dynamic equilibrium, and disturbance. In: Shroder, J. (Editor in Chief),
Butler, D.R., Hupp, C.R. (Eds.), Treatise on Geomorphology. Academic timing of important geomorphic events including floods,
Press, San Diego, CA, vol. 12, Ecogeomorphology, pp. 94–106. mass wasting, and rates of erosion and sedimentation; and

94 Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00324-9

 Vegetation Ecogeomorphology, Dynamic Equilibrium, and Disturbance 95

(2) through the documentation and interpretation of species ecology that found common ground between the opposing
distribution patterns that are responses to prevailing hydro- schools of Darwinism and equilibrium.
geomorphic conditions. Tree-ring studies of geomorphic pro-
cesses are not included in this chapter; however, reviews
can be found in Shroder (1980), Hupp (1988), Hupp and 12.7.2 Vegetation Patterns
Bornette (2003), and Stoffel and Bollschweiler (2008).
Plant ecology developed as a discipline in the second half Although many species grow in a wide range of environments,
of the nineteenth century, but until c. 1900 was largely re- some grow only in specific habitats that may be broadly
stricted to plant classifications or the compilation of species defined in terms of topography and interpreted in terms of
lists (Joyce, 1993; Osterkamp and Hupp, 1996, 2010). Plant geomorphic process. The local distribution of species may be
communities were considered static functions of climate, and understood as a physiological and anatomical response by
interpretation or recognition of the effects of fluvial process individual plants to the physical conditions within an area
was not considered. The period of emphasis only on data (Figure 1, Hack and Goodlett, 1960). The presence of repro-
collection ended when botany and plant ecology became in- ductive plants is assumed to be evidence of propriety of those
fluenced by Darwinian thought, and later by equilibrium plants to a particular habitat; thus, there typically is no ran-
theory (Osterkamp and Hupp, 1996, 2010). Darwinists H.C. dom presence of full-grown plants (Zimmermann and Thom,
Cowles (1901) and F. E. Clements (1909, 1916) revolutionized 1982). Further, numerically insignificant occurrences of plants
concepts of interactions between Earth-surface processes and are often ecologically significant.
vegetation by introducing the model of plant succession, the Discrete patterns of species distribution have been correl-
hypothesis that plant communities are explained in terms of ated with discrete breaks in several geomorphic parameters
orderly change within the communities, not generally as (e.g., slope, aspect, landform, bedrock, and hydrology). The
functions of environment. An alternative explanation was surficial geomorphic processes acting in humid regions are
offered by equilibrist and champion of the individualistic divided into two major categories: (1) those operating with
concept (rather than the community concept of Clements), H. wash and creep processes (interfluvial processes) and (2) those
A. Gleason (1925, 1926), who stressed that, for plants to operating in conjunction with the direct action of streamflow
survive in space and time, they must be adapted to local and (fluvial processes). Although early attempts to relate plant dis-
current environmental requirements. Robert McIntosh (1958, tribution patterns to geomorphic form and process were mostly
1960) and other integrationists proposed approaches to plant concerned with interfluvial areas (Cowles, 1901; Hack and

Approximate crest of ridge Explanation

N
3000
Nose
Contours convex outward (away
from mountain). Runoff proportional
to a function of the radius of
curvature of the contours.

Side Slope
? Contours straight. Runoff proportional
? to a linear function of slope length.
2900
Road
2800 Hollow
Contours concave outward. Runoff
proportional to a power function of
slope length.
2600 2700
Trail to
Crawford Mountain
Channelway
Trail to Dry
Contours sharply concave outward. Run-
Branch Gap off proportional to a power function of
channel length.

− 200 0 200 400 Feet

Footslope
Contour interval 20 feet
datum is mean sea level Transitional area between side slope and
channelway (not present in all valleys).
Figure 1 Portion of a plan table map by Hack and Goodlett (1960) showing typical slope declivity, micro-topography, and potential forest types.
Adapted from Hack, J.T., Goodlett, J.C., 1960. Geomorphology and forest ecology of a mountain region in the central Appalachians. U.S.
Geological Survey Professional Paper 347, 66 pp.
96 Vegetation Ecogeomorphology, Dynamic Equilibrium, and Disturbance

Goodlett, 1960), much recent research (since the early Goodlett, 1956; Mills, 1984; Schaetzl et al., 1989a; Norman
1980s) has focused on fluvial form and dynamics (Hupp and et al., 1995; Gabet et al., 2003). In addition to geomorphic
Osterkamp, 1996; Osterkamp and Hupp, 2010). impacts, tree throw and the creation of canopy gaps may play
Vegetation, geomorphic features, and hydrologic conditions a major role in forest ecology/community composition
(in humid regions and in many arid and semi-arid regions) are (Goodlett, 1954; White and Pickett, 1985; Schaetzl et al.,
interrelated in such a way that any change in one affects the 1989b). Soil disturbance by tree throw creates a pit and
others; all may be affected by geologic underpinning. Ground- mound topography (Figure 2) that can affect as much as 40%
moisture regimes are largely determined by the type of land- of the land surface in northern temperate forests (Denny and
form that exists under present climatic and geomorphic pro- Goodlett, 1956). Because there is a strong relation between
cesses, whereas the type of landform is determined by the tree diameter and mound volume, the species composition of
moisture regime present. The hydrogeomorphic processes may the forest may substantially affect sediment flux rates (Gabet
be strongly interactive with the vegetation, which stabilizes et al., 2003). In regions with relatively high relief, tree throw
landforms, moves large quantities of water, and forms an may be among the most important agents of sediment trans-
above- and below-ground fabric that affects in obvious and port (Norman et al., 1995). Some studies have estimated rates
nuanced ways most hydrogeomorphic dynamics. Fluvial pro- of soil transport (Mills, 1984; Lenart et al., 2010). Gabet et al.
cesses are influenced by basin geomorphology, erodibility of (2003) produced an equation that predicts sediment flux as a
surficial material/sediment dynamics, hydrologic regime, and function of tree throw and important hillslope properties (e.g.,
vegetation within the basin. Thus, both geomorphic and vege- slope, direction of throw, and soil density) and averaged over
tation patterns and processes, in fluvial and interfluvial areas, space and time. Their equation predicts an average horizontal
may be primarily driven by water, which in turn, is affected by sediment flux of 8 m3 ha–1 yr–1 for a variety of forest types on
both through straightforward and nuanced feedback loops. a 101 slope, and using data from Denny and Goodlett (1956)
predicts a flux of 9.8 m3 ha–1 yr–1 in a northern hardwood
forest. Thus, tree throw represents a substantial mechanism for
12.7.3 Hillslopes slope denudation. Further, Osterkamp et al. (2006) have
shown that tree throw may account for much of the fluvial
Vegetation growing on hillslopes was historically considered sediment transported by high-gradient streams even in semi-
to have minor or indirect effects on slope processes; this idea arid parts of western North America.
was cogently challenged by Greenway (1987). Recent study
has shown that vegetation, particularly woody plants, may
play important roles in slope stability by reinforcing soil
strength, and conversely by facilitating slope disturbance. 12.7.4 Riparian Vegetation, Fluvial Processes, and
Hillslope vegetation and landforms are intimately linked and Landforms
co-evolve such that they singularly and simultaneously affect
both the form and process of each other through dynamic Riparian vegetation patterns and fluvial geomorphic forms
feedback mechanisms (Marston, 2010). Two somewhat related and processes are closely integrated environmental phenom-
interfluvial areas of geomorphic interest where vegetation may ena occurring along most perennial streams. In temperate
play an important role are: (1) the variable-source- or partial- fluvial systems, water, either through streamflow conditions or
area concept in relation to recharge and runoff; and (2) the groundwater availability, is the most proximal control on the
impact of tree uprooting (tree-throw) on slopes. Regarding the distributional patterns of perennial riparian plants (Figure 3).
former, not all areas in a catchment contribute equally to Riparian vegetation may also strongly affect the rates of ero-
streamflow (Dunne and Black, 1970). In a classic paper on the sion and deposition of sediment, and may be integral in the
relations between plant ecology and geomorphic form and overall stability of fluvial surfaces. There are at least five major
process, Hack and Goodlett (1960) showed that forest types ways riparian vegetation may affect fluvial geomorphic pro-
could be related to slope types where convex-upward slopes cesses (Hickin, 1984):
contributed mostly to runoff; concave slopes contributed
1. by providing flow resistance;
mostly to groundwater recharge; and linear slopes were
2. by increasing bank and other surface strength (e.g., roots);
intermediate. They produced a map of vegetation types using
3. by facilitating channel-bar formation;
key species that clearly delimited slope declivity (Figure 1)
4. by providing woody debris; and
and laid the groundwork for future studies. The study area of
5. by facilitating deposition on bank-bench surfaces.
Hack and Goodlett (1960) in central Appalachia was revisited
and investigated 35 years later by Osterkamp et al. (1995), In general, these effects are all related to the role vegetation
who confirmed the key role landform/water relations play in plays in sediment erosion and/or deposition. The integral
vegetation distribution across both fluvial and interfluvial nature of vegetation is particularly evident in streams that have
surfaces. The possible importance of vegetation mapping been disturbed by human alteration where specific species
in delineating variable-source areas was explicitly stated in are indicative of specific landforms and process that develop
Dunne and Black (1970). In the UK, especially, upland vege- in response to disturbance (Simon and Hupp, 1992; Hupp
tation types have been successfully related to runoff in studies et al., 2009). This disturbance commonly comes in the form
by Gurnell and Gregory (1987, 1995) and Thorne (1990). of channel incision, channel narrowing, and/or widening.
Downslope movement of sediment may be greatly facili- Riparian vegetation patterns, even along highly altered
tated by tree throw – the uprooting of trees (Denny and streams, are indicative of present and ongoing fluvial forms
 Vegetation Ecogeomorphology, Dynamic Equilibrium, and Disturbance 97

Most soil material falls

downslope of pit, forming
a large pit and a large
Root plate mound.

Mound
Pit

Tree-throw pit
Distrubed soil

(a) Tree on steep slope (b) Uprooted tree (c) Pit/Mound pair

More material slumps back

into pit as root plate
Root plate decomposes, forming a
smaller pit/mound pair.

Mound
Pit

Tree-throw pit Disturbed soil

(a) Tree on gentle slope (b) Uprooted tree (c) Pit/Mound pair

Figure 2 Tip up mound generation and the development of pit and mound topography on steep forested slopes (top row) and on gentle slopes
(bottom row). Adapted from Norman, S.A., Schaetzl, R.J., Small, T.W., 1995. Effects of slope angle on mass movement by tree uprooting.
Geomorphology 14, 19–27.

AS

AB
FP HL
Tu T DB
l
FB FP Tl Tu
CB
Alluvium
Not to scale Consolidated rocks

Figure 3 Typical fluvial landforms along alluvial streams in eastern United States. AS, channel shelf; AB, channel shelf bank; DB, depositional
bar; CB, channel bed; FB, floodplain bank, FP, floodplain; Tl, lower terrace; Tu, upper terrace; HL, hillslope. Adapted from Osterkamp, W.R., Hupp,
C.R., 1984. Geomorphic and vegetative characteristics along three northern Virginia streams. Bulletin of the Geological Society of America 95,
501–513.
98 Vegetation Ecogeomorphology, Dynamic Equilibrium, and Disturbance

and processes, while simultaneously reflecting stages of or other hydrologic information is lacking. As examples, Hupp
channel evolution following incision, channel widening, and/ and Osterkamp (1996) showed that two shrubs common on
or narrowing (Hupp, 1992; Marston et al., 1995; Friedman the channel shelf (a bank feature) are relatively resistant to
and Lee, 2002; Hupp and Rinaldi, 2007). destructive flooding because of small, highly resilient stems
In both equilibrated and nonequilibrium systems, plant and the ability to sprout rapidly from flood-damaged stumps.
distributions across landforms may be driven largely by the Conversely, common terrace species that are rare on flood-
tolerance of species to specific geomorphic processes at the plains may be intolerant of repeated flood damage or inun-
severe end of a stress/equilibrium gradient and by competition dation. Floodplain species are less tolerant of destructive
with other riparian vegetation at the mild physical condition floods than channel-shelf species but are more tolerant
end (Hupp and Osterkamp, 1996; Bornette et al., 2008). of periodic inundation than are terrace species. Riparian
Variation in environmental conditions and plant-adaptive plant communities consist of species with variable tolerances
strategies across North America and Europe largely explain the to the ambient fluvial process regime (Figure 4). The vegeta-
complex patterns of fluvial-species distributions. tion–landform relation may be useful also in determining
Periodic floods of varying magnitude, variation in flow stage of riparian-zone recovery following disturbance, in-
duration, and sediment erosion/deposition dynamics affect cluding gravel mining (Hupp and Rinaldi, 2007) and chan-
vegetation patterns in several ways. Some of the most in- nelization (Hupp, 1992).
fluential effects include the creation of new areas, such Lateral accretion is an episodic process that occurs during
as point bars and depositional islands, regime-related vari- high flows, building the point bar into a commonly crescent-
ation in bank stability, formation of a gradient of flood shaped ridge. Over time, a series of high-flow events produce
intensities along channels, sediment-size variation, and for- the ridge and swale topography associated with meander
mation of water-availability gradients across fluvial surfaces. scrolls. The establishment of ruderal woody vegetation during
Separating these and other factors that influence plant patterns intervening low-flow periods on fresh scroll surfaces creates
is difficult because most factors are distinctly interdependent. bands of increasingly younger vegetation toward the main
Although there have been advances in the past decade channel (McKenney et al., 1995). These bands of vegetation
or more, there is still a general lack of accepted landform may accentuate the ridge and swale topography by creating
and process definitions in the geomorphic sciences and, contrasting depositional environments during subsequent
to a greater extent between the sciences of geomorphology and high flows; the hydraulics necessary to produce meander-scroll
plant ecology. topography and the role of vegetation in its development are
poorly understood. As ridge and swale topography is inte-
grated into the greater floodplain, the ridges support bot-
12.7.4.1 Systems in Dynamic Equilibrium
tomland forest species that reflect relatively dry conditions,
Dynamic equilibrium in geomorphology refers to the mutual whereas swales (commonly called sloughs), remain relatively
adjustment of a catchment with its geologic underpinning and wet and support hydric bottomland plants.
the streams that drain it such that a stream entrains, trans- Riparian areas along low-gradient systems, such as the
ports, and stores the sediment delivered to fluvial landforms Coastal Plain of the southeastern United States, have many of
in a balanced fashion (typically no excessive net fluvial ero- the same geomorphic features discussed above; however, in-
sion or deposition) (Hack, 1960). Fluvial geomorphic pro- undation length and frequency (generally annual) may be
cesses, via the action of flowing water, create a number of more of a driving force on fluvial landforms and vegetation
widely recognized landforms from small-channel bedforms patterns than periodic intensive flooding. Fluvial surfaces on
to extensive floodplains. The latter typically support diverse lowland alluvial systems tend to have net sediment storage
forested ecosystems in humid temperate regions of the world. over millennial timescales owing to typically high sediment
Floodplains, like most fluvial landforms, are dynamic features, loads and low channel gradients. Here, floodplains may
almost constantly eroding in places while aggrading in others. aggrade in two ways. First, by lateral accretion or point-bar
Common fluvial landforms, as defined by Osterkamp and extension where coarse material is deposited on the inside
Hupp (1984), occur as terraces high in the valley section (flow bank of channel bends (a corresponding volume is typically
return interval 43 years) and, in descending order, proceed eroded on the opposite cut bank) and second, by vertical
through the floodplain (flow return interval every 1–3 years), accretion of suspended sediment (typically fines) over the
various riparian features on banks (flow return interval o1 floodplain during overbank flows.
year, generally measured by annual flow duration), and The generally drastic and sudden reduction in flow velocity
channel bars to the channel bed (Figure 3). Studies in North after leaving the main channel and entering the hydraulically
America, Europe, and Japan have shown that there are rough floodplain environment facilitates sediment deposition.
characteristic plant-species distributional patterns for specific As rising floodwaters overtop the bank, the coarser (or heav-
fluvial landforms and processes (Osterkamp and Hupp, 1984; ier) sediment is deposited first, creating natural levees along
Décamps et al., 1988; Tabacchi et al., 1990; Gregory, 1992; the floodplain margin. Levees tend to be most pronounced
Naiman et al., 1993; Pautou and Arens, 1994; Marston along relatively straight reaches between meanders and are
et al., 1995; Hupp and Osterkamp, 1996; Bravard et al., 1997; commonly the highest ground on the floodplain. Levees are
Hughes, 1997; Tabacchi et al., 1998; Bendix and Hupp, 2000; sometimes breached by streamflow, resulting in a crevasse
Nakamura and Shin, 2001; Gurnell and Petts, 2003; Hupp and splay that may insert coarse material deep into the otherwise
Rinaldi, 2007). These relations are significant and may be used fine-grained bottom. Levee development and the breaches that
to infer hydrogeomorphic conditions where gauging-station form are poorly documented in the literature, yet they are
 Vegetation Ecogeomorphology, Dynamic Equilibrium, and Disturbance 99

Terrace
Floodplain assemblage
forest

Riparian
shrubs

Herbaceous
Mean bar plants
low water
Terrace
Floodplain
Channel
Channel bed Bar shelf
A

D C

F E

G
H

Figure 4 Fluvial landforms and associated bottomland vegetation types. Individual species ranges (A through H) showing variation in specificity
for different landforms. Note, some species are broadly distributed (B) and others are limited to one fluvial surface (G, H). Adapted from Hupp,
C.R., 1983. Vegetation patterns on channel features in the Passage Creek Gorge, Virginia. Castanea 48(2), 62–72.

critical in the understanding of the surface-water hydrology of canal and levee construction may lead to channel incision
most bottomlands. Levee height and breaches strongly affect the or filling and large changes in sediment-supply conditions,
hydroperiod (length of annual inundation) in systems domin- depending on the geomorphic setting worldwide (Marston
ated by surface-water flow (Patterson et al., 1985). Floodplain et al., 1995; Gurnell and Petts, 2003; Rinaldi, 2003; Hupp
surface elevations typically decrease from the levee, through et al., 2009). Most of the world’s largest rivers have been
transitional areas, to relatively wet backswamps, although the dammed; the downstream impacts that most affect floodplain
overall relief may be small. Small differences in elevation areas are typically severe reductions in peak stages, frequency
(Figure 5), commonly measured in centimeters, may lead to and duration of over bank flows, and sediment transport
pronounced differences in hydroperiod and, thus, to variation (Williams and Wolman, 1984). Simultaneously, dams may
in riparian community composition (Mitsch and Gosselink, concentrate flows on mid- and low-bank slopes. Land clear-
1993). Vegetation patterns may result from anaerobic con- ance with upland erosion and downstream aggradation
ditions, also associated with prolonged inundation, that lead to (legacy sedimentation dynamics) (Jacobson and Coleman,
production of toxic respiration byproducts and limitation of 1986; Pierce and King, 2007) have led to channel and valley
nutrients and water (Wharton et al., 1982). The degree to which filling and sometimes, subsequent channelization. Addition-
individual species have adapted to anaerobic-related stresses ally, stream channelization has been a common, albeit con-
controls the distinct changes in community composition across troversial, practice along many rivers in parts of the Coastal
short lateral distances on these lowland floodplains. Plain (Simon and Hupp, 1992; Hupp and Bazemore, 1993)
to reduce flooding and facilitate row-crop agriculture on
floodplains; the impact on the riparian zone is a severe
12.7.4.2 Systems in Nonequilibrium States reduction in overbank flow. Levee construction, particularly
along the Mississippi River and its large tributaries and dis-
Human alterations to the landscape, such as flow regulation tributaries, has occurred over a long period and has had
through dam construction, land clearance with upland ero- profound impacts (Mossa, 1996; Biedenharn et al., 2000) on
sion and downstream aggradation, stream channelization, and streamflow and sedimentation dynamics.
100 Vegetation Ecogeomorphology, Dynamic Equilibrium, and Disturbance

Bald cypress, Water tupelo

Water elm, Buttonbush
River birch
Black willow
Overcup oak,
Red maple
Willow and pin
oaks, Sweetgum
Swamp
chestnut oak
Figure 5 Lowland (swamp) vegetation distributions associated with varying length of annual inundation as related to elevation. Adapted from
Sharitz, R.R., Mitsch, W.J., 1993. Southern floodplain forests. In: Martin, W.H., Boyce, S.G., Echternacht, A.C. (Eds.), Biodiversity of the
Southeastern United States, Lowland Terrestrial Communities. Wiley, New York, pp. 311–372.

Vegetation may strongly alter rates of erosion and de- upstream or downstream, or both, from the reach of greatest
position of sediment, and, in large part, may be integral in the impact/alteration and move as a knick point in time and space.
overall stability of fluvial surfaces (Hupp, 1999), particularly as The vegetation–landform relation may be useful also
an ecogeomorphic response to human alteration. This is par- in determining stage of riparian-zone recovery following dis-
ticularly evident on fluvial landforms along nonequilibrium turbance including dam construction, channelization, and
streams during and following recovery due to channel incision gravel mining. The study of vegetation patterns in specific
(Hupp, 1999) and channel narrowing (Marston et al., 1995; relation to dynamic, progressively adjusting fluvial landforms
Friedman et al., 1996; Garcia-Ruiz et al., 1997; Liébault and is generally lacking in the literature. However, substantial
Piégay, 2001, 2002; Friedman and Lee, 2002; Rinaldi, 2003). progress has been made in the interpretation of channel
The community organization and dynamics of vegetation on evolution through the use of conceptual models (Schumm
the bottomlands of rivers are strongly governed by fluvial et al., 1984; Simon and Hupp, 1992; Rinaldi, 2003; Surian
geomorphic processes and landforms, which largely are cre- and Rinaldi, 2003). As an example, the Simon and Hupp
ated and maintained by fluctuations of water discharge. The (1992) model of channel evolution following alteration
fluvial geomorphic effects of various human disturbances vary (channelization) describes six stages of channel and riparian
according to: (1) multiple possible combinations; (2) mutual vegetation conditions beginning with the pre-modified (ref-
adjustments of the fluvial variables; and (3) the physiograpic erence) conditions, through stages of incision, bank failure,
context (Steiger et al., 2005). and bed and bank recovery (aggradation) toward a
Streams that have been affected by substantial human alter- stable quasi-equilibrium stage (Figure 6). Patterns of vege-
ation generally exhibit an unstable regime shift (e.g., channel tation recovery along altered alluvial channels develop in re-
incision or pronounced aggradation). This instability normally sponse to and, in turn, affect patterns of fluvial geomorphic
initiates a complex but predictable sequence of process responses recovery following human alteration. Channel incision and
(Schumm and Parker, 1973) toward the original equilibrated widening by bank erosion and subsequent channel narrowing
state. Thus, an impetus for change following alteration (Hey, owing to sediment deposition are important geomorphic
1979) may ensue at a site; this involves channel widening and processes that limit and affect vegetation patterns through the
filling following channel incision (Simon and Hupp, 1992) or course of geomorphic recovery toward equilibrium. A similar
channel incision may follow severe floodplain aggradation eco-geomorphic response was observed along Italian streams
(Pierce and King, 2007). The magnitude of this impetus for affected by gravel mining and long-term human alteration by
change is governed likely by the extent of the alteration. How- Hupp and Rinaldi (2007). Using a channel-evolution model
ever, like alteration, the impetus for change may attenuate either (Rinaldi, 2003) and incorporating vegetation dynamics,
 Vegetation Ecogeomorphology, Dynamic Equilibrium, and Disturbance 101

Stage I
Premodified

Stage II
Constructed

Stage III

Degradation

Stage IV
Degradation
and widening

Stage V
Aggradation
and widening

Stage Vi
Quasi equilibrium

Direction of bank or bed movement

(Not to scale)
Explanation

Water Slumped material Accreted material

Figure 6 Six-stage model of alluvial channel evolution following channelization with associated vegetation dynamics. Species composition
generally changes with each stage of evolution. Adapted from Hupp, C.R., 1992. Riparian vegetation recovery patterns following stream
channelization: a geomorphic perspective. Ecology 73, 1209–1226.
102 Vegetation Ecogeomorphology, Dynamic Equilibrium, and Disturbance

Hupp and Rinaldi (2007) showed that geomorphic processes Size of patches
in response to alteration and associated vegetation dynamics Depositional Erosional
are intertwined environmental processes that are largely re- (+) (−) (+)
sponsible for channel recovery toward an equilibrated state.
Fine Grain size Coarse

Low Stream gradient High

12.7.5 Dynamic Equilibrium and the
Erosional–Depositional Environment Storage

Deposition
The concept of dynamic equilibrium offers a process-oriented
explanation for dramatic responses to human alterations and
may also allow for prediction of the direction (erosion vs.
deposition) and magnitude of the change (Hack, 1960; Equilibrium
Leopold et al., 1964). Equilibrium states with persistent charac- transport
teristic fluvial landforms may occur on streams that maintain an

Erosion
erosional regime, such as steep mountain streams, or on streams
that maintain a sediment-storage regime; this equilibrium, by
definition, is not static but dynamic. Streams that are not in Entrainment
dynamic equilibrium typically exhibit dramatic, generally rapid,
regime shifts; this may happen naturally (e.g., earthquakes) or
through human alteration. Meandering, Cascading, straight
aggrading degrading
Most bottomland substrate available for vegetation estab-
systems systems
lishment is alluvial; thus, hydraulic scour and size-clast sorting (Coastal Plain) (mountains)
may be important limiting factors. Erosion and channel cut-
offs may lead to almost permanently inundated pools that Figure 7 Model of stream regime across a deposition/erosion
gradient. Adapted from Hupp, C.R., Bornette, G., 2003. Vegetation,
support aquatic vegetation. Terrestrial wetland vegetation may
fluvial processes and landforms in temperate areas. In: Piegay, H.,
be restricted to narrow ranges of sediment size. Most woody
Kondolf, M. (Eds.), Tools in Geomorphology. Wiley & Co., Chichester,
plants maintain their actively absorbing root zones in the pp. 269–288.
upper 15 cm of soil (Kramer and Kozlowski, 1979). Thus, only
species capable of rapid root growth along newly buried stems
(e.g., species of Salix and Populus) may occupy bottomland connected to its floodplain. Such conditions are common in
areas periodically affected by frequent or large amounts of rivers subjected to incision, for example as a consequence of
sediment deposition (Figure 7) or active channel migration flow regulation (Galay, 1983; Babinsky, 1992; Bravard, 1994;
such as along point bars (Hupp, 1988). Obviously, where Hupp, 1999; Hupp et al., 2009).
erosion removes all or part of the soil in the root zone, trees It is possible to plot each floodplain or floodplain reach on
will be killed or severely damaged (Figure 7). Flooding regime a curve (Hupp and Bornette, 2003; Bornette et al., 2008) that
is intimately associated with the depositional environment represents erosion and deposition processes (Figure 7). The
along most streams with adequate sediment supply. position along this curve or model depends on the dominant
Aggradation along stream reaches, both natural and process (deposition vs. erosion, vertical axis) and on the
altered, provides new sites for vegetation colonization and magnitude of hydrogeomorphic processes and size of the
may initiate a succession of vegetation stages indicative of patches (horizontal axis). For each river system or river reach,
progressively changing hydrogeomorphic conditions. Hupp the amount of net aggradation versus erosion determines the
(1992) and Johnson (1994) found that dominance by pioneer position on the hypothetical curve simulating these two pro-
species can be explained by life-history characteristics in- cesses. At the ends of the gradient, the largest patch sizes occur.
cluding: a large, dependable seed crop, seeds effectively dis- Near the center (equilibrium) of the gradient (Figure 7), the
persed by wind or water to suitable riparian sites, rapid number and size of eroded versus deposited patches tend to
germination, and rapid shoot and root development (to be equal, and the turnover rate of these patches remains the
withstand flooding and desiccation). Initial vegetation estab- same along the curve. Most temperate river floodplains can
lishment increases hydraulic roughness facilitating further be plotted between these two polar situations and the size of
sedimentation and ultimately modifying initially un- the deposition patches versus erosion patches is related to the
stable surfaces into relatively stable surfaces favorable for the location of the river floodplain along the curve (Bornette et al.,
recruitment of later, stable-site species (Hupp, 1992; Johnson, 2008). A central equilibrium area corresponds to the situation
1994; Friedman et al., 1996). where silted patches are as large as eroded patches (Figure 7).
Along streams dominated by aggradation (Figure 7), This area reflects the highest habitat heterogeneity where a
extensive depositional areas may be generated during large dynamic equilibrium is reached that corresponds to a shifting
floods. Such floodplains are common along many south- mosaic of habitats at the floodplain scale. Such situations are
eastern US streams where upstream agriculture practices and increasingly less common in temperate areas, due to the fre-
channelization generate large amounts of suspended fines quent and heavy impacts of human activities that disrupt and
(Hupp and Bazemore, 1993). In the degradation situation, drive fluvial processes toward one end or the other of the
large eroded zones are generated, and the river is decreasingly gradient (Hupp and Bornette, 2003). At the scale of an entire
 Vegetation Ecogeomorphology, Dynamic Equilibrium, and Disturbance 103

river, some reaches may experience migrating aggradation, Frequent, relatively high-intensity disturbance should se-
others, degradation. Thus, it is at the reach scale that the lect species according to their tolerance for unstable environ-
disruption of equilibrium is most easily observed. ments (rapid turnover rates). The front or proximal end of this
Stream gradient (or power) also varies systematically along axis (Figure 8) indicates low biological interactions among
this conceptual gradient (Figure 7). Where stream gradient is species; here, physical interactions dominate and select for
high, erosional processes dominate (e.g., low-order high ruderal, disturbance-tolerant species (Bornette et al., 2008). As
mountain cascades); conversely, along low-gradient reaches, disturbance frequency decreases along the third axis, bio-
depositional processes usually dominate. Thus, depending logical interactions increase (competition and herbivory),
on the flow regime, equilibrium conditions and highest which should increasingly select for competitive, stable site
biodiversity may shift along this conceptual gradient. Sedi- species. Also, as disturbance frequency lessens, other limiting
ment sizes, stream gradient, and channel pattern (meandering, factors, such as nutrient availability, fluxes of propagules,
cascading, and straight) may change along the conceptual and other biological interactions become more important
gradient to maintain adjusted conditions (Figure 7), and near- in determining riparian vegetation patterns (Bornette et al.,
equilibrium conditions may occur in a zone around the 2008).
boundary between net erosion and deposition. For instance,
streams in mountainous areas are likely to be naturally
erosional, whereas streams on the Coastal Plain are likely to 12.7.6 Summary
have a naturally net depositional (storage) stream regime
(Hupp, 2000). Streams between these two geographic settings, Separating factors that simultaneously influence vegetation
such as the Piedmont, may have a sediment-transport dom- patterns and geomorphic processes is difficult because most
inated regime. are interdependent. Moreover, consistent definitions of land-
Habitat patch dynamics is a well-known concept in ecology form and process commonly are lacking within geomorph-
and refers to areas (patches) of ground surface that are created ology and particularly between the geomorphic and plant
through various events, such as, in this case, floods. Floods ecological sciences. Where conformity occurs between sci-
can create erosional patches through floodplain scour or ences, it is generally due to the common belief that hydrologic
depositional patches, such as a crevasse splay. The size of these processes control most aspects of fluvial-bottomland eco-
patches depends on the intensity of the event and can be as- systems. Indeed, only hydrologic characteristics provide in-
sociated with the model shown in Figure 7. The turnover rate dependent parameters consistent on all perennial streams.
of the floodplain patches depends on the flow regime and Hydrologic processes, particularly those that relate to water
equilibrium conditions of the river or river reach and may vary availability, similarly affect vegetation patterns on hillslopes.
greatly from one site to another. Bornette et al. (2008) added a Additionally as a feedback mechanism, vegetation may affect
third dimension to this model to depict a gradient of patch geomorphic processes. Sediment entrainment on hillslopes in
turnover rate (axis c, Figure 8). This three-axis model in- temperate regions may be greatly affected by tree throw, for
corporates the major physical limiting factors on riparian example, erosion of soil mounds created by uprooted trees.
vegetation and provides a comprehensive explanation of plant Riparian vegetation patterns and fluvial landforms and
life-history strategies (Bornette et al., 2008), whereas the two- processes are closely integrated along most perennial streams.
axis model simply depicts processes for a given frequency of In temperate fluvial systems, water, either through streamflow
flood disturbance or patch turnover rate. Given that flood conditions or groundwater availability, is the most proximal
frequency plays a major role in the rate of habitat reworking in control of distributional patterns of perennial riparian plants.
riparian areas, the third axis is necessary (Figure 8, Bornette Riparian-zone vegetation also may strongly affect rates of
et al., 2008). The third axis also provides for a temporal scale erosion and of sediment deposition, and may be integral in
of disturbance. the stability of fluvial surfaces. This is particularly evident in

Low Patch size,

disturbance
Water frequency
dispersal High Many
Biological 0
Few
interactions O
Deposition
100% (c)
+

(a)

0 100%
(b) Erosion
Many Few Refugia
Many Few regeneration niches
Figure 8 Three-dimensional model of stream regime and vegetation life-history strategies. Adapted from Bornette, G., Tabacchi, E., Hupp, C.R.,
Puijalon, S., Rostan, J.C., 2008. A model of plant strategies in fluvial hydrosystems. Freshwater Biology 53, 1692–1705, with permission from Wiley.
104 Vegetation Ecogeomorphology, Dynamic Equilibrium, and Disturbance

streams disturbed by human alteration, which can lead to Biedenharn, D.S., Thorne, C.R., Watson, C.C., 2000. Recent morphological
channel incision and narrowing. Studies presented in this evolution of the lower Mississippi River. Geomorphology 34, 227–249.
Bornette, G., Tabacchi, E., Hupp, C.R., Puijalon, S., Rostan, J.C., 2008. A model
chapter show that riparian vegetation patterns, even along
of plant strategies in fluvial hydrosystems. Freshwater Biology 53, 1692–1705.
highly altered streams, identify present and ongoing fluvial Bravard, J.P., 1994. L’incision des lits fluviaux: du phénomène morphodynamique
forms and processes and simultaneously reflect stages of naturel et réversible aux impacts irréversibles. Revue de geographie de Lyon 69,
channel evolution following alteration. 5–10.
Fluvial geomorphic systems, by nature, tend to maintain a Bravard, J.P., Amoros, C., Pautou, G., Bornette, G., Bournaud, M., Des Chatelliers,
M.C., Gibert, J., Peiry, J.L., Perrin, J.F., Tachet, H., 1997. River incision in
dynamic equilibrium among ambient sediment load, water south-east France: morphological phenomena and ecological effects. Regulated
discharge, and channel geometry. Streams or reaches of streams Rivers: Research and Management 13, 75–90.
are typically deemed ‘‘in equilibrium’’ when the stream and its Butler, D.R., Malanson, G.P., 1995. Sedimentation rates and patterns in beaver
hydrogeomorphic form and process are sufficiently (but not ponds in a mountain environment. Geomorphology 13, 225–269.
Clements, F.E., 1909. Darwin’s influence upon plant geography and ecology.
overly) competent to entrain, transport, and store the sediment
American Naturalist 43, 143–151.
provided by the associated catchment in a balanced fashion. Clements, F.E., 1916. Plant Succession: An Analysis of the Development of Vegetation.
Equilibrium conditions may occur in a zone around the Carnegie Institution of Washington, Washington, DC, Publication No. 242.
boundary between net erosion and deposition. For instance, Cowles, H.C., 1901. The physiographic ecology of Chicago and vicinity. Botanical
streams in mountainous areas may have a naturally net erosion Gazette 31, 73–108.
Décamps, H., Fortuné, M., Gazelle, F., Pautou, G.C., 1988. Historical influence of
(entrainment) stream regime, whereas streams in the Coastal
man on the riparian dynamics of a fluvial landscape. Landscape Ecology 1,
Plain tend to have a naturally net depositional (storage) stream 163–173.
regime. Streams in between these two geographic settings, such Denny, C.S., Goodlett, J.C., 1956. Microrelief resulting from fallen trees. In:
as those in the Piedmont, may have a sediment transport- Surficial Geology and Geomorphology of Potter County, Pennsylvania. U.S.
dominated regime as shown in the conceptual gradient of Geological Survey. Professional Paper 288, 59–68.
Dunne, T., Black, R.D., 1970. Partial area contributions to storm runoff in a small
stream conditions. Sediment grain size, stream gradient, and New England watershed. Water Resources Research 6, 1296–1311.
channel pattern (meandering, cascading, and straight) may Friedman, J.M., Lee, V.J., 2002. Extreme floods, channel change, and riparian
adjust along the conceptual gradient to maintain near-equi- forests along ephemeral streams. Ecological Monographs 72, 409–425.
librium conditions. Human alteration within the catchment or Friedman, J.M., Osterkamp, W.R., Lewis, Jr. W.M., 1996. The role of vegetation and
bed-level fluctuations in the process of channel narrowing. Geomorphology 14,
along the stream that substantially alters one or more of these
341–351.
parameters may lead to disequilibrium conditions, in particu- Gabet, E.J., Reichman, O.J., Seabloom, E.W., 2003. The effects of bioturbation on
lar, stream gradient, which may initiate a period of pronounced soil processes and sediment transport. Annual Review of Earth and Planetary
adjustment and excessive erosion or deposition. Sciences 31, 249–273.
The response of riparian vegetation to hydrogeomorphic Galay, V.J., 1983. Causes of riverbed degradation. Water Resources Research 19,
1057–1090. http://dx.doi.org/10.1029/WR019i005p01057.
processes (sediment erosion and deposition) is complex, re- Garcia-Ruiz, J.M., White, S.M., Lasanta, T., Gonzales, C., Errea, M.P., Valero, B.,
sulting from the interaction of diverse physical and biological 1997. Assessing the effects of land-use changes on sediment yield and channel
gradients. These gradients are further complicated, depending dynamics in the central Spanish Pyrenees. In: Walling, D.E., Probst, J.L. (Eds.),
on the degree to which a stream approaches dynamic equi- Human Impact on Erosion and Sedimentation. International Association of
Hydrological Sciences Publication 245. Boulder, CO, pp. 151–158.
librium. A model presented here asserts that two types of
Gleason, H.A., 1925. Species and area. Ecology 6, 66–74.
disturbance could select for contrasting adaptive strategies, Gleason, H.A., 1926. The individualistic concept of the plant association. Bulletin of
based mainly on the ability of plant to resist the disturbing the Torrey Botanical Club 4, 41–49.
event and to colonize new patches. Goodlett, J.C., 1954. Vegetation adjacent to the border of the Wisconsin drift in
Potter County, Pennsylvania. Harvard Forest Bulletin 25.
Greenway, D.R., 1987. Slope Stability: Geotechnical Engineering and
Geomorphology. Wiley, New York, pp. 187–230.
Acknowledgments Gregory, K.J., 1992. Vegetation and river channel processes. In: Boon, P.J., Calow,
P., Petts, G.E. (Eds.), River Conservation and Management. Wiley & Co,
Two late colleagues and friends, John Hack and Bob Sigafoos, Chichester, pp. 255–269.
Gurnell, A.M., Gregory, K.J., 1987. Vegetation characteristics and the prediction of
provided considerable inspiration and insight to both of us,
runoff: analysis of an experiment in the New Forest, Hampshire. Hydrological
their seminal works provide us all with a firm scientific Processes 1, 125–142.
foundation and a deep appreciation for the intimate linkage Gurnell, A.M., Gregory, K.J., 1995. Interactions between semi-natural vegetation and
between vegetation and geomorphology. Another friend and hydrogeomorphological processes. Geomorphology 13, 49–69.
colleague, Gudrun Bornette is sincerely appreciated for her Gurnell, A.M., Petts, G.E., 2003. Island dominated landscapes of large floodplain
rivers: a European perspective. Freshwater Biology 47, 581–600.
efforts to understand the detailed plant ecological linkages to Hack, J.T., 1960. Interpretation of erosional topography in humid temperate regions.
variation in geomorphic processes. We thank Sammy King, American Journal of Science 258-A, 80–97.
Massimo Rinaldi, and Jack Shroder, treatise editor-in-chief, for Hack, J.T., Goodlett, J.C., 1960. Geomorphology and forest ecology of a mountain
providing valuable reviews of the manuscript. region in the central Appalachians. U.S. Geological Survey Professional Paper
347, 66 pp.
Hession, W.C, Wynn, T., Resler, L., Curran, J., 2010. Geomorphology and
vegetation: interactions, dependencies, and feedback loops. Geomorphology,
References Special Issue 116(3–4), 274–285.
Hey, R.D., 1979. Flow resistance in gravel-bed rivers. Journal of the Hydraulics
Division 105, 365–379.
Babinsky, Z., 1992. Hydromorphological consequences of regulating the Lower Hickin, E.J., 1984. Vegetation and river channel dynamics. Canadian. Geographer
Vistula, Poland. Regulated Rivers 7, 337–348. 28, 111–126.
Bendix, J., Hupp, C.R., 2000. Hydrologic and geomorphic impacts on riparian plant Hughes, F.M.R., 1997. Floodplain biogeomorphology. Progress in Physical
communities. Hydrological Processes 14, 2977–2990. Geography 21, 501–529.
 Vegetation Ecogeomorphology, Dynamic Equilibrium, and Disturbance 105

Hupp, C.R., 1988. Plant ecological aspects of flood geomorphology and paleoflood Osterkamp, W.R., Hupp, C.R., 2010. Fluvial processes and vegetation – glimpses of
history. In: Baker, V.R., Kochel, R.C., Patton, P.C. (Eds.), Flood Geomorphology. the past, present, and future. Geomorphology 116, 274–285.
Wiley, New York, pp. 335–356. Osterkamp, W.R., Hupp, C.R., Schening, M.R., 1995. Little River revisited –thirty-
Hupp, C.R., 1992. Riparian vegetation recovery patterns following stream five years after Hack and Goodlett. Geomorphology 13, 1–20.
channelization: a geomorphic perspective. Ecology 73, 1209–1226. Osterkamp, W.R., Toy, T.J., Lenart, M.T., 2006. Development of partial rock veneers
Hupp, C.R., 1999. Relations among riparian vegetation, channel incision processes by root throw in a subalpine setting. Earth Surface Processes Landforms 31,
and forms, and large woody debris. In: Darby, S.E., Simon, A. (Eds.), Incised 1–14.
River Channels. Wiley & Co., Chichester, pp. 219–245. Patterson, G.G., Speiran, G.K., Whetstone, B.H., 1985. Hydrology and its effects
Hupp, C.R., 2000. Hydrology, geomorphology, and vegetation of Coastal Plain on distribution of vegetation in Congaree National Monument, South
rivers in the southeastern United States. Hydrological Processes 14, 2991–3010. Carolina. U.S. Geological Survey, Water-Resources Investigations Report
Hupp, C.R., Bazemore, D.E., 1993. Temporal and spatial patterns of wetland 85-4256.
sedimentation, West Tennessee. Journal of Hydrology 141, 179–196. Pautou, G., Arens, M.F., 1994. Theoretical habitat templates, species traits, and
Hupp, C.R., Bornette, G., 2003. Vegetation, fluvial processes and landforms in species richness: floodplain vegetation in the Upper Rhône River. Freshwater
temperate areas. In: Piegay, H., Kondolf, M. (Eds.), Tools in Geomorphology. Biology 31, 507–522.
Wiley & Co., Chichester, pp. 269–288. Pierce, A.R., King, S.L., 2007. The effects of flooding and sedimentation on
Hupp, C.R., Osterkamp, W.R., 1996. Riparian vegetation and fluvial geomorphic seed germination of two bottomland hardwood tree species. Wetlands 27,
processes. Geomorphology 14, 277–295. 588–594.
Hupp, C.R., Osterkamp, W.R., Howard, A.D., 1995. Biogeomorphology – Terrestrial Rinaldi, M., 2003. Recent channel adjustments in alluvial rivers of Tuscany, central
and Freshwater Systems. Elsevier, Amsterdam, 347 pp. Italy. Earth Surface Processes and Landforms 28, 587–608.
Hupp, C.R., Rinaldi, M., 2007. Riparian vegetation patterns and diversity in relation Rostan, J.C., Amoros, C., Juget, J., 1987. The organic content of the surficial
to fluvial landforms and channel evolution along selected rivers of Tuscany sediment: a method for the study of ecosystems development in abandoned river
(central Italy). Annals of the Association of American Geographers 97, 12–30. channels. Hydrobiologia 148, 45–62.
Hupp, C.R., Schenk, E.R., Richter, J.M., Peet, R.K., Townsend, P.A., 2009. Bank Schaetzl, R.J., Burns, S.F., Johnson, D.L., Small, T.W., 1989a. Tree uprooting:
erosion along the dam regulated lower Roanoke River, North Carolina. review of impacts on forest ecology. Vegetatio 79, 165–176.
Geological Society of America, Special Publication 451, 97–108. Schaetzl, R.J., Johnson, D.L., Burns, S.F., Small, T.W., 1989b. Tree uprooting:
Jacobson, R.B., Coleman., D.J., 1986. Stratigraphy and recent evolution of review of terminology, process, and environmental implications. Canadian
Maryland Piedmont flood plains. American Journal of Science 286, 617–637. Journal of Forest Research 18, 1–11.
Johnson, W.C., 1994. Woodland expansion in the Platte River, Nebraska: patterns Schumm, S.A., Harvey, M.D., Watson, C.C., 1984. Incised Channels: Morphology,
and Causes. Ecological Monographs 64, 45–84. dynamics, and control. Water Resources Publications, Littleton, CO.
Joyce, L.A., 1993. The life cycle of the range condition concept. Journal of Range Schumm, S.A., Parker, R.S., 1973. Implications of complex response of drainage
Management 46, 132–138. systems for Quaternary alluvial stratigraphy. Nature (Physical Science) 243,
Kramer, P.J., Kozlowski, T.T., 1979. Physiology of woody plants. Academic Press, 99–100.
New York, 642 pp. Schwarz, W.L., Malanson, G.P., Weirich, F.H., 1996. Effect of landscape position on
Lenart, M.T., Falk, D.A., Scatena, F.N., Osterkamp, W.R., 2010. Estimating soil the sediment chemistry of abandoned-channel wetlands. Landscape Ecology 11,
turnover rate from tree uprooting during hurricanes in Puerto Rico. Forest 27–38.
Ecology and Management 259, 1076–1084. Shroder, J.F., 1980. Dendrogeomorphology: review and new techniques of tree-ring
Leopold, L.B., Wolman, M.G., Miller, J.P., 1964. Fluvial Processes in dating. Progress in Physical Geography 4, 161–188.
Geomorphology. W. H. Freeman, San Francisco, 522 pp. Simon, A., Hupp, C.R., 1992. Geomorphic and vegetative recovery processes along
Liébault, F., Piégay, H., 2001. Assessment of channel changes due to long-term modified stream channels of West Tennessee. U.S. Geological Survey, Open File
bedload supply decrease, Roubion River, France. Geomorphology 36, 167–186. Report 91-502, pp. 199–214.
Liébault, F., Piégay, H., 2002. Causes of 20th century channel narrowing in Steiger, J., Tabacchi, S.E., Dufour, D.S., Corenblit, D., Peiry, J.-L., 2005.
mountain and piedmont rivers of southeastern France. Earth Surface Processes Hydrogeomorphic processes affecting riparian habitat within alluvial channel-
and Landforms 27, 425–444. floodplain river systems: a review for the temperate zone. River Research and
Marston, R.A., 2010. Geomorphology and vegetation on hillslopes: Interactions, Applications 21, 719–737.
dependencies, and feedback loops. Geomorphology 116, 206–217. Stoffel, M., Bollschweiler, M., 2008. Tree-ring analysis in natural hazards
Marston, R.A., Girel, J., Pautou, G.C., Piégay, H., Bravard, J.-P., Arneson, C., 1995. research? An overview. Natural Hazards and Earth System Science 8(2),
Channel metoamorphosis, floodplain disturbance, and vegetation development, 187–202.
Ain River, France. Geomorphology 13, 121–131. Surian, N., Rinaldi, M., 2003. Morphological response to river engineering and
McIntosh, R.P., 1958. Plant communities. Science 128, 115–120. management in alluvial channels in Italy. Geomorphology 50, 307–326.
McIntosh, R.P., 1960. Natural order and communities. Biologist 42, 55–62. Tabacchi, E., Correll, R., Hauer, D.L., Pinay, G., Planty-Tabacchi, A.-M., Wissmar.,
McKenney, R., Jacobson, R.B., Wertheimer, R.C., 1995. Woody vegetation and R.C., 1998. Development, maintenance, and the role of riparian vegetation in the
channel morphogenesis in low-gradient, gravel-bed streams in the Ozark river landscape. Freshwater Biology 40, 497–516.
Plateaus, Missouri and Arkansas. Geomorphology 13, 175–198. Tabacchi, E., Planty-Tabacchi, A.-M., Décamps, O., 1990. Continuity and
Mills, H.H., 1984. Effect of hillslope angle and substrate on tree tilt, and denudation discontinuity of the riparian vegetation along a fluvial corridor. Landscape
of hillslopes by treefall. Physical Geography 5, 253–261. Ecology 5, 9–20.
Mitsch, W.J., Gosselink, J.G., 1993. Wetlands, Second ed. Van Nostrand Reinhold, Thorne, C.R., 1990. Effects of vegetation on riverbank erosion and stability.
New York, NY. In: Thornes, J.B. (Ed.), Vegetation and Erosion. Wiley, Chichester, pp. 125–144.
Mossa, J., 1996. Sediment dynamics in the lowermost Mississippi River. Viles, H., 1995. Ecological perspectives on rock surface weathering: towards a
Engineering Geology 45, 457–479. conceptual model. Geomorphology 13, 21–35.
Naiman, R.J., Décamps, H., Pollock, M., 1993. The role of riparian corridors in Viles, H.A., 1988. Cyanobacterial and other biological influences on terrestrial
maintaining regional diversity. Ecological Applications 3, 209–212. limestone weathering on Aldabra: implications for landform development.
Nakamura, F., Shin, N., 2001. The downstream effects of dams on the regeneration Bulletin of the Biological Society of Washington 8, 5–13.
of riparian tree species in northern Japan. In: Dorava, J.M., Montgomery, D.R., Wharton, C.H., Kitchens, W.M., Pendleton, E.C., Sipe, T.W., 1982. The ecology
Palcsak, B.B., Fitzpatrick, F.A. (Eds.), Geomorphic Processes and Rivervine of bottomland hardwood swamps of the southeast: a community profile.
Habitat. American Geophysical Union, Washington, DC, pp. 173–181. FWS/OBS-81/37. U.S. Fish and Wildlife Service, Biological Services Program,
Norman, S.A., Schaetzl, R.J., Small, T.W., 1995. Effects of slope angle on mass Washington, DC.
movement by tree uprooting. Geomorphology 14, 19–27. White, P.S., Pickett, S.T.A., 1985. Natural disturbance and patch dynamics: an
Osterkamp, W.R., Hupp, C.R., 1984. Geomorphic and vegetative characteristics introduction. In: Pickett, S.T.A., White, P.S. (Eds.), The Ecology of Natural
along three northern Virginia streams. Bulletin of the Geological Society of Disturbance and Patch Dynamics. Academic Press, Orlando, FL, pp. 3–13.
America 95, 501–513. Williams, G.P., Wolman, M.G., 1984. Downstream effects of dams on alluvial rivers.
Osterkamp, W.R., Hupp, C.R., 1996. The evolution of geomorphology, ecology, and U.S. Geological Survey Professional Paper 1286, 83 pp.
other composite Sciences. In: Thorn, C., Rhoads, B. (Eds.), The Scientific Nature Zimmermann, R.C., Thom, B.G., 1982. Physiographic plant geography. Progress in
of Geomorphology. Wiley, New York, Ch. 17, pp. 414–441. Physical Geography 6, 45–59.
106 Vegetation Ecogeomorphology, Dynamic Equilibrium, and Disturbance

Biographical Sketch

Cliff R. Hupp, PhD has investigated riparian vegetation ecology in relation to fluvial landforms and processes for
30 years. Additional research includes studies on channel evolution, floodplain processes and forms, sedimen-
tation dynamics, and carbon sequestration in riparian ecosystems in the US and Western Europe. He has been
employed by the US Geological Survey since 1978 where he is research project chief of the Vegetation and
Hydrogeomorphology Relations Project. He was a student of the late John T. Hack at the George Washington
University, where he received his doctorate in 1984. Dr. Hupp is the 1993 recipient of the Ecological Society of
America, W.C. Cooper Award for outstanding research in physiographic ecology. He received the US Department
of Interior Superior Service Award in 2006. He has served as Section Editor for the ESA journals Ecology and
Ecological Monographs since 1999.

W. R. Osterkamp, National Research Program, US Geological Survey (USGS), is a geomorphologist at the Desert
Laboratory, Tucson, Arizona. He is interested in the factors that govern landscape stability, from hillslopes to
bottomlands, and studies the nature and rate of recovery of these surfaces from natural and anthropogenic
disturbance, and the role that vegetation plays in this recovery. This information can be used to plan engineering
works and to manage riparian settings. Dr. Osterkamp also has investigated hydrologic processes, such as those
involved in the origin and development of playa-lake basins in the Southern High Plains of Texas and New
Mexico, the evolution of armored hillslopes, the geomorphic and hydrologic effects of transmission loss from
arid-zone streamflow, and the processes of sediment storage, alluvial-landform development, and revegetation
that follow catastrophic floods. He holds undergraduate degrees in geology and chemistry from the University of
Colorado and MS and PhD degrees in geology and hydrology from the University of Arizona. He has been Project
Chief, Sediment Impacts from Disturbed Lands, since 1980, is a former Research Advisor for the Geomorphology
and Sediment Transport Discipline of the National Research Program, and is Adjunct Professor, University of
Denver and University of Arizona.
 12.8 The Reinforcement of Soil by Roots: Recent Advances and Directions
for Future Research
N Pollen-Bankhead and A Simon, USDA-ARS National Sedimentation Laboratory, Oxford, MS, USA
RE Thomas, University of Tennessee, Knoxville, TN, USA
Published by Elsevier Inc.

12.8.1 Introduction 108

12.8.1.1 Root Tensile Strength 108
12.8.2 Calculating Root Reinforcement 111
12.8.2.1 The Use of Fiber-Bundle Models in Root-Reinforcement Modeling 113
12.8.2.1.1 Load apportionment alternatives in FBMs 114
12.8.2.1.2 Effect of changing root orientations in an FBM 115
12.8.2.1.3 Displacement over which root reinforcement is effective 115
12.8.2.1.4 Using a Monte Carlo approach in FBMs 116
12.8.2.2 Root Architecture Measurement and Modeling 117
12.8.2.3 Hydraulic and Hydrologic Effects of Vegetation 118
12.8.3 Root-Reinforcement and Geomorphologic Processes at Different Spatial Scales 119
12.8.4 Conclusions and Direction of Future Research 120
References 121

Glossary Root tensile strength A measure of the ability of plant

Fiber-bundle models Models developed by the materials roots to withstand a longitudinal stress, expressed as the
science industry to simulate the reinforcing effect of fibers greatest stress that the material can stand without
contained within a matrix (in this case roots within a soil breaking.
matrix). These models use simple rules to distribute stress Soil pore-water pressure The pressure exerted on its
between fibers according to their tensile strengths, allowing surroundings by water held in pore spaces in rock or soil.
for progressive breaking of the weakest fibers, and The pressure is positive when a soil is fully saturated,
redistribution of stresses to remaining intact fibers during and is then proportional to the height of the water
loading of a reinforced matrix. measured in an open tube (a piezometer) above the point
Riparian zone Riparian zones are vegetated areas along of interest. A buoyancy effect is achieved and the shear
both sides of water bodies that generally consist of trees, strength of the soil is reduced. The pressure is zero when the
shrubs, and grasses and are transitional boundaries between soil voids are filled with air, and is negative when the
land and water environments. voids are partly filled with water. When pore-water pressure
Root reinforcement The additional strength provided to a is negative, surface-tension forces operate to achieve a
soil matrix by the roots of vegetation growing within that suction effect and the shear strength of the soil is
soil. increased.

Abstract

Mechanical strengthening of a soil occurs as a result of reinforcement by the roots, due to their tensile strength, frictional,
and adhesional properties. Soil is strong in compression but weak in tension; conversely, plant roots are weak in com-
pression but strong in tension. When the two are combined they, therefore, produce a matrix of reinforced earth, which is
stronger than the soil or the roots separately. Root tensile strength tends to decrease nonlinearly with increasing root
diameter, with root strength having been found to be closely correlated to cellulose content and, therefore, the biotic and
abiotic factors that determine cellulose production in roots. Techniques for modeling root–soil interactions have improved
significantly over the last 5–10 years with the introduction of fiber-bundle models to the study of root reinforcement.
Although the mechanics of root breaking, stretching, compression, and pullout can be modeled successfully, several
obstacles to more accurate root-reinforcement predictions remain, including the collection of accurate field data to

Pollen-Bankhead, N., Simon, A., Thomas, R.E., 2013. The reinforcement of soil
by roots: recent advances and directions for future research. In: Shroder, J.
(Editor in Chief), Butler, D.R., Hupp, C.R. (Eds.), Treatise on Geomorphology.
Academic Press, San Diego, CA, vol. 12, Ecogeomorphology, pp. 107–124.

Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00325-0 107

108 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research

parameterize these models over different spatial and temporal scales. Root-reinforcement modeling has, however, shown
that roots can be of significant importance in the stabilization of slopes and streambanks affecting both the timing and
magnitude of mass failure events, and the geomorphic processes that cause them.

12.8.1 Introduction • root length/diameter ratio,

• soil–root bond strength,
Vegetation exerts a number of controls on the geomorphic • alignment – angularity/straightness of the roots, and
processes affecting slope and streambank stability. The mani- • orientation of the roots relative to the direction of principal
festation of these controls is a number of hydraulic, hydro- strains.
logic, and mechanical effects, some of which have positive and
some of which have negative impacts on soil stability. To Two methods have commonly been used to quantify the
evaluate whether a particular slope or streambank will be effects of root reinforcement on soil strength. First, values
strengthened or weakened by vegetation of a given species and collected from in situ shear tests of root-permeated soils have
age, the balance between these effects must be considered. been used to replace the value of the soil strength alone (e.g.,
Vegetation has been increasingly used as an effective, eco- Wu et al., 1988b). However, in situ shear tests present a
nomical, and environmentally friendly solution for slope and number of problems. Although they do give a direct assess-
streambank stabilization. Methods to measure and predict the ment of the amount of increased shear strength provided to
effects of vegetation on mass-wasting processes have become the soil by the roots, isolating a block of root-permeated soil
of particular interest to both the academic and engineering to shear is not an easy task and both the soil and the soil–root
communities. bond may be disturbed before shearing is undertaken. In
The roots of plants are anchored into the soil to support addition, Wu et al. (1988a) and Shewbridge and Sitar (1990)
the above-ground parts of vegetation, thereby creating a re- showed that when carrying out in situ tests of root-permeated
inforced soil matrix in which stress is transferred from the soil soil, the forces developed in the roots, and, therefore, their
to the roots, increasing the overall strength of the matrix contribution to soil shear strength, are dependent on the di-
(Greenway, 1987). The strength of rooted soil is, therefore, mensions of the shear box. Second, measurements of root
due to a combination of soil strength, root strength, and the strength have been carried out in field and laboratory experi-
strength of the bonds between the soil and roots (Waldron, ments, in the absence of the soil matrix, so that empirical and
1977; Waldron and Dakessian, 1981; Ennos, 1990). The the- physically based relationships can be established between the
ory for computing the magnitude of root reinforcement ori- root and soil properties that allow the roots to increase the
ginated in the materials science literature and the construction shear strength of the soil.
industry. Within the construction industry, although fiber- In this chapter, we review the literature pertaining to
reinforced concrete has been used since the beginning of the mechanical reinforcement of soils by roots, including meas-
twentieth century, it is only in the last 30 years or so that urement and modeling techniques, the influence of roots on
methods have been developed to quantify the extent of the slope and streambank geomorphic processes, and potential
reinforcement provided (Beaudoin, 1990). For a matrix of one directions for future research.
material to be strengthened by the fibers of another, the two
materials must have different tensile and compressive strength
properties (Beaudoin, 1990). For example, in the case of re- 12.8.1.1 Root Tensile Strength
inforced concrete, load transfer from the concrete matrix to
Root tensile strength can either be measured in situ at a field
the (usually steel) fibers occurs because the fibers have a
site (e.g., Wu et al., 1979; Abernethy and Rutherfurd, 2001;
higher tensile strength than the surrounding concrete matrix,
Simon and Collison, 2002; Pollen and Simon, 2005) or roots
allowing the concrete to resist the development of cracks. In a
can be removed and taken back to the laboratory for testing
similar way, the tensile strength of roots and the frictional and
(e.g., Genet et al., 2005; De Baets et al., 2008). There are ad-
adhesional bonds between the roots and soil lead to mech-
vantages and disadvantages of both methods. During in situ
anical strengthening of the soil. Soil is strong in compression
tests, one end of the root is attached to a load cell, whereas the
but weak in tension; conversely, plant roots are weak in
other end of the root remains attached to the tree or plant
compression but strong in tension. When the two are com-
being studied. Soil and plant moisture conditions are, there-
bined they, therefore, produce a matrix of reinforced earth,
fore, maintained (depending on the time-frame over which
which is stronger than the soil or the roots separately (Thorne,
the testing takes place), but root tortuosity (the twists and
1990). Hence, roots provide reinforcement by transferring
turns a root takes as it grows in the soil matrix), angle of
shear stress in the soil to tensile stress in the roots, thereby
orientation, and branching can interfere with accurate strength
increasing the confining stress of the soil (Gray and Sotir,
testing. Conversely, during laboratory tests, extruded roots are
1996).
clamped at both ends, eliminating the problems of root
Greenway (1987) noted that the magnitude of root re-
orientation and tortuosity but the absence of the soil and the
inforcement is a function of a number of following factors:
remainder of the plant make maintaining field moisture
• root density, conditions problematic. In addition, root tensile strengths
• root tensile strength, obtained from laboratory tests (Bischetti et al., 2005) are
• root tensile modulus, commonly much larger than those measured under field
 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research 109

conditions (Tosi, 2007; Docker and Hubble, 2008). This is combination of a dense rooting pattern of fine roots in the top
thought to be a result of root branches that influence the layer (where resistance in tension is most important) with
stress–strain distribution and ultimate resistance to failure coarse, deeply penetrating roots (crossing the potential shear
under field conditions. surface) is most beneficial to stability.
Experiments have shown that root tensile strengths de- The values shown in Table 1 and Figure 2 indicate that for
crease exponentially with increasing root diameter (examples tree species there is a wide range in the reported regression
include but are not limited to: Waldron and Dakessian, 1981; parameters a and b for root tensile strength. Values for the
Riestenberg and Sovonick-Dunford, 1983; Greenway, 1987; coefficient, a, range from 87.04 to 18.4, whereas those for the
Coppin and Richards, 1990; Gray and Sotir, 1996; Abernethy exponent, b, range from –1.11 to 0.13. The reported range of a
and Rutherfurd, 2001; Simon and Collison, 2002; Genet et al., for shrubs is very similar to that for grasses, but the reported
2005; Pollen and Simon, 2005; De Baets et al., 2008; Fan and range of b for shrubs is within the range of those for trees. The
Su, 2008; Hales et al., 2009). However, breaking stress (the b values for grass species were all more negative than those
maximum stress that can be applied to the root before it fails) reported for shrubs and trees. The exponent, b, has a larger
and the rate of decrease in tensile strength with increasing root impact on calculations of root tensile strength values than the
diameter vary between species. Typical tensile strength values coefficient, a. The strongly negative exponent for grass species
range from 4 to 20 MPa for grass roots and 5–70 MPa for tree results in grass roots less than 1 mm in diameter being cal-
roots (Coppin and Richards, 1990). The relationship between culated to be stronger than similar-sized shrub and tree roots.
root diameter and root tensile strength is usually best de- Conversely, strongly negative exponents result in flat Tr  d
scribed using a power law function of the form: curves for roots larger than 1 mm in diameter, with larger
diameter tree roots, therefore, tending to be stronger than
Tr ¼ adb ½1 similar-sized shrub and grass roots.
A study conducted by Genet et al. (2005) showed that the
where Tr is the ultimate tensile strength (MPa), d the root power relationship between tensile strength and root diameter
diameter (mm), and a and b the regression parameters. cannot simply be explained by the scaling effect typical of that
Results from a selection of root tensile strength studies found in fracture mechanics. Genet et al. (2005) suggested
carried out in the past 10 years are summarized in Table 1. that an additional explanation for the change in root strength
These studies cover a wide range of geographic locations and with root diameter was the potential change in cellulose
vegetation types. Mean values for the regression parameters a content in roots of varying diameters. In their study on sweet
and b were calculated for the broad vegetation types of trees, chestnut (Castanea sativa Mill.) and Maritime pine (Pinus
shrubs, and grasses (Figure 1). The regression parameters for pinaster Ait.), cellulose content was shown to increase with
the three broad vegetation types suggest that, at the smallest decreasing root diameter, as did root tensile strength. Cellu-
root diameters, grass roots have the highest tensile strength lose content also varied between the two species, which may
and shrubs the lowest. This finding is in agreement with a explain, in part, the differences in root tensile strength that can
similar analysis conducted by Mao et al. (2012). The tensile be seen between species of the same vegetation type.
strengths of the grass and tree species tend to converge at root Root architecture can also affect the distribution of tensile
diameters above 5 mm. For root diameters greater than 1 mm, strengths in a given root network (Stokes and Mattheck, 1996).
grass roots are, however, weaker per unit area than corres- For example, in addition to variations in root tensile strength
ponding tree and shrub species. As root reinforcement is a with root diameter, variations in root tissue strength have been
function of not only root strength but also the density of roots found to occur even along roots that have a fairly constant
in a soil, grasses may provide significant reinforcement to the diameter. The degree of variation in tensile, compressive, and
upper part of a soil profile where thousands of fine, fibrous bending strengths along structural roots of constant diameter
grass roots are concentrated. Conversely, the woody roots of may be related to the way in which cellulose is distributed in root
trees and shrubs will provide reinforcement over a greater systems with different architectures. In the case of plate root
depth of soil through a combination of both fine, fibrous systems having many shallow lateral roots, loading forces are
roots and coarser, woody roots. Grasses may, therefore, pro- transmitted into the soil further away from the stem, as they are
vide significant reinforcement when potential failure planes less branched than heart and tap root systems that are centered
are shallow, but provide less resistance when failure planes on a main vertical root. Root wood strength is, therefore, high
extend deeper into a slope or streambank as a smaller pro- along the entire length of the root in order to resist mechanical
portion of the failure plane length is affected by roots. As stress. Conversely, heart and tap root systems branch more near
Reubens et al. (2007) noted, different root diameters play the tree stem and the strength provided by the main root means
different roles in strengthening a soil matrix. During soil that a high investment in strength further along the roots in
shearing, fine roots tend to break but stay in position relative unnecessary (Stokes and Mattheck, 1996). Indeed, Genet et al.
to the adjacent soil particles, but coarse roots can be pulled (2005) suggested that a highly branched root system will most
out of the soil intact (Ennos, 1990), suggesting that fine roots likely have a different cellulose content to a root network con-
are of greatest importance for soil stabilization. However, taining fewer, thicker roots, but cellulose measurements obtained
deeper-penetrating coarse roots impact a greater proportion of from different root architectures are, to date, lacking from the
the length of potential failure planes. It should also be noted literature.
that fine roots have a low bending stiffness, whereas coarse Hales et al. (2009) investigated how cellulose content and
roots can resist both tension and bending (Bischetti et al., resultant root tensile strengths varied with hillslope topo-
2005). Reubens et al. (2007), therefore, concluded that a graphy. Their study found that tree roots located on the convex
110 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research

Table 1 Summary for tensile strength regression parameters a and b measured in a sample of such studies over the past 10 years

Species Latin name Vegetation Location of Tensile strength parameters Reference

type testing
a b

Hopea odorata Tree Thailand 87.04  0.69 Nilaweera and Nutalaya (1999)
Acacia floribunda Tree Australia 85.35  0.13 Docker (2003)
Dipterocarpus alatus Tree Thailand 68.95  0.45 Nilaweera and Nutalaya (1999)
Fagus sylvatica Tree SW France 63.51  0.61 Genet et al. (2005)
Corylus avellana Tree N. Italy 60.15  0.75 Bischetti et al. (2005)
Alangium kurzii Tree Thailand 52.95  0.53 Nilaweera and Nutalaya (1999)
Eucalyptus amplifolia Tree Australia 52.65  0.17 Docker (2003)
Liquidamber stryaciflua Tree MS, USA 52.10  1.04 Pollen and Simon (2005)
Hibiscus macrophyllus Tree Thailand 51.49  0.56 Nilaweera and Nutalaya (1999)
Eucalyptus elata Tree Australia 50.71  0.14 Docker (2003)
Plantus occidentalis Tree MS, USA 50.50  0.94 Pollen and Simon (2005)
Eucalyptus camaldulensis Tree Australia 49.39  0.77 Abernethy and Rutherfurd (2001)
Melaleuca ericifolia Tree Australia 49.39  0.77 Abernethy and Rutherfurd (2001)
Salix nigra Tree MS, USA 45.90  1.10 Pollen and Simon (2005)
Betula nigra Tree MS, USA 45.80  0.66 Pollen and Simon (2005)
Fagus sylvatica Tree N. Italy 41.65  0.97 Bischetti et al. (2005)
Ficus benjamina Tree Thailand 40.49  0.72 Nilaweera and Nutalaya (1999)
Eucalyptus elata Tree Australia 39.96  0.41 Docker and Hubble (2008)
Eucalyptus amplifolia Tree Australia 38.95  0.43 Docker and Hubble (2008)
Casuarina glauca Tree Australia 38.22 0.13 Docker (2003)
Picea abies Tree SW Trance 37.86  0.51 Genet et al. (2005)
Hevea brasiliensis (para rubber) Tree Thailand 36.66  0.77 Nilaweera and Nutalaya (1999)
Fraxinus excelsior Tree N. Italy 35.73  1.11 Bischetti et al. (2005)
Alnus viridis Tree N. Italy 34.70  0.69 Bischetti et al. (2005)
Salix caprea Tree N. Italy 34.50  1.02 Bischetti et al. (2005)
Larix decidua Tree N. Italy 33.45  0.75 Bischetti et al. (2005)
Castanea sativa Tree SW France 31.92  0.73 Genet et al. (2005)
Tamarix canariensis Tree SE Spain 31.74  0.89 De Baets et al. (2008)
Pinus palustrus Tree MS, USA 30.00  0.99 Pollen and Simon (2005)
Casuarina glauca Tree Australia 29.67  0.41 Docker and Hubble (2008)
Alstonia macrophulla Tree Thailand 28.27  0.46 Nilaweera and Nutalaya (1999)
Picea abies Tree N. Italy 28.10  0.72 Bischetti et al. (2005)
Salix purpurea Tree N. Italy 26.33  0.95 Bischetti et al. (2005)
Salix lemmonii Tree CA, USA 25.93  0.86 Simon, Pollen and Langendoen (2006)
Salix exigua Tree KS, USA 25.20  0.68 Pollen and Simon (2005)
Fraxinus latifolia Tree OR, USA 24.30  0.50 Pollen and Simon (2005)
Tamarisk ramosissima Tree AZ, USA 23.60  0.90 Pollen-Bankhead et al. (2009)
Pinus pinaster Tree SW France 23.40  0.87 Genet et al. (2005)
Elaeagnus angustifolia Tree AZ, USA 22.10  1.00 Pollen-Bankhead et al. (2009)
Pinus contorta Tree CA, USA 19.06  0.65 Simon, Pollen and Langendoen (2006)
Populus fremontii Tree OR, KS, USA 18.90  0.64 Pollen and Simon (2005)
Pinus nigra Tree SW France 18.40  0.52 Genet et al. (2005)
Atriplex halimus Shrub SE Spain 45.59  0.56 De Baets et al. (2008)
Salsola genistoides Shrub SE Spain 44.23  0.51 De Baets et al. (2008)
Thymelaea hirsuta Shrub SE Spain 33.31  0.64 De Baets et al. (2008)
Artemisia barrelieri Shrub SE Spain 30.12  0.51 De Baets et al. (2008)
Spirea douglasii Shrub OR, USA 22.90  0.54 Pollen and Simon (2005)
Rubus discolor Shrub OR, USA 19.50  0.69 Pollen and Simon (2005)
Thymus zygis Shrub SE Spain 19.31  0.73 De Baets et al. (2008)
Dittrichia viscosa Shrub SE Spain 18.94  0.45 De Baets et al. (2008)
Teucrium capitatum Shrub SE Spain 18.72  0.45 De Baets et al. (2008)
Dorycnium pentaphyllum Shrub SE Spain 16.32  0.62 De Baets et al. (2008)
Fumana thymifolia Shrub SE Spain 15.71  0.66 De Baets et al. (2008)
Rosmarinus officinalis Shrub SE Spain 12.89  0.77 De Baets et al. (2008)
Nerium oleander Shrub SE Spain 4.41  1.75 De Baets et al. (2008)
Brachypodium retusum Grass SE Spain 45.05  0.61 De Baets et al. (2008)
Tripsacum dactyloides Grass MS, USA 43.10  1.00 Pollen and Simon (2005)
Panicium virgatum Grass MS, USA 35.20  1.78 Pollen and Simon (2005)
(Continued )
 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research 111

Table 1 Continued

Species Latin name Vegetation Location of Tensile strength parameters Reference

type testing
a b

Stipa tenacissima Grass SE Spain 24.34  0.61 De Baets et al. (2008)

Lygeum spartum Grass SE Spain 19.28  0.68 De Baets et al. (2008)
Helictotrichon filifolium Grass SE Spain 14.51  1.08 De Baets et al. (2008)
Piptatherum miliaceum Grass SE Spain 11.49  1.77 De Baets et al. (2008)
Avenula bromoides Grass SE Spain 4.77  1.52 De Baets et al. (2008)
Juncus acutus Rush SE Spain 23.23  0.89 De Baets et al. (2008)
Phragmites australis Reed SE Spain 34.29  0.78 De Baets et al. (2008)
Limonitum supimum Herb SE Spain 33.82  0.85 De Baets et al. (2008)
Plantago albicans Herb SE Spain 16.75  0.52 De Baets et al. (2008)

Species are sorted by vegetation type, and then according to tensile strength regression parameter, a.

200 root tensile strength values within and between species

Mean tree tensile strength curve y = 40.12 × −0.67 (Schiechtl, 1980; Genet et al., 2005).
Mean shrub tensile strength curve y = 23.23 × −0.62
Root tensile strength (MPa)

150 Mean grass tensile strength curve y = 24.72 × −1.39

12.8.2 Calculating Root Reinforcement

100 The first attempts to quantify the reinforcement due to

roots involved the use of simple perpendicular root models
(Waldron, 1977; Wu et al., 1979) that included root re-
50 inforcement as an additional term in the Mohr–Coulomb shear
strength criterion for unsaturated soils (Fredlund et al., 1978):

0 S ¼ c0 þ ðma  mw Þtan fb þ ðs  ma Þtan f0 þ DS ½2

0.0 0.5 1.0 1.5 2.0 2.5 3.0
Root diameter (mm)
where S is the soil-shearing resistance (kPa), c’ the effective
cohesion (kPa), ma the pore-air pressure (kPa), mw the pore-
Figure 1 Mean tensile strength curves for tree, shrub, and grass water pressure (kPa), fb the angle describing the increase in
species collated from the studies of root strength listed in Table 1. shear strength due to an increase in matric suction (ma  mw)
(1), s the normal stress on the shear plane (kPa), f’ the effective
noses of a hillslope consistently had stronger tensile strengths soil friction angle (1), and DS the increase in shear strength due
than those found in concave hollows of the hillslope. Hales to roots (kPa). Gray (1974) reported that the angle of internal
et al. (2009) suggested that this finding may be linked to friction of the soil appeared to be affected little by the presence
differences in root structural chemistry, and, therefore, cellu- of roots. The first two terms in eqn [2] are the apparent co-
lose content, associated with soil water potential at the dif- hesion, whereas the third term is the frictional resistance.
ferent topographic locations, with higher cellulose contents Waldron (1977) assumed that all roots extend vertically
occurring in the stronger roots on hillslope noses. Variations across a horizontal shear zone, and that the roots act like lat-
in soil water potential between hollows and noses can also erally loaded piles, so tension is transferred to them as the soil is
result in distinctly different assemblages of vegetation being sheared. The tension developed in each root is resolved into a
supported (Stevenson and Mills, 1999); where combined with tangential component that augments the apparent cohesion and
variations in root tensile strength with topographic location, a normal component that augments the frictional resistance. In
these vegetation patterns may reinforce any positive feedback reality, the angle of each root in relation to the direction of the
loops in the landscape that create these variations in hillslope applied force is important, as this dictates the distribution of
topography. Thus, although a power law relation between root stresses within the root volume and hence the maximum tensile
diameter and tensile strength can commonly be seen for a strength reached before root failure (Niklas, 1992). Gray and
given species at a given site, cellulose content, and thus root Leiser (1982), therefore, generalized the Waldron (1977) model
tensile strength, may also vary with environmental factors. to the case where roots may be oriented at any angle relative to
These environmental factors include but are not limited to soil the failure plane. These authors expressed DS as
fertility, nutrient supply, soil moisture content, and soil
mechanical factors such as bulk density. Further studies that DS ¼ Tr ðAr =AÞ½sinð90  cÞ þ cosð90  cÞtan f0  ½3
compare the tensile strength of roots from trees with different
mechanical functions, and growing in different soil conditions where Tr is the root failure strength (tensile, frictional, or
are, therefore, needed to understand fully the possible range in compressive) of roots per unit area of soil (kPa), Ar/A the
112 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research

100 Vegetation type

Trees Shrubs Grasses
90
0.5
80
Range for multiplier, a (no units) 70
0.0

Range for exponent, b (no units)

60

50
−0.5
40

30
−1.0
20

10
−1.5
0
Trees Shrubs Grasses
Vegetation type −2.0
(a) (b)

Figure 2 Range of reported values for root tensile strength and diameter curve parameters a and b. The vertical lines represent the minimum
and maximum values reported and the dots represent mean values for each plant type.

root area ratio(dimensionless), and c the angle of the root (0  901), multiplier values selected by most authors have
at rupture relative to the failure plane (1), which can be tended to be within the range of 1.0–1.2 (Table 2). However,
expressed as Gray and Leiser (1982: 42–43), Riestenberg and Sovonick-
  Dunford (1983: 515), and Danjon et al. (2008: 1289) pro-
1 duced figures demonstrating much larger potential variability
c ¼ tan1 ½4
tan y þ 1=tan i in Rf. Danjon et al. (2008) obtained field measurements of
root angles crossing a failure plane and noted significant
where y is the angle of shear distortion (1) and i the initial root variations in the value of Rf. Greenway (1987) calculated Rf
orientation relative to the failure plane (1). Equation [3] is using several values of f’ and y and concluded that a value of
identical to the model of Waldron (1977) for a root orien- 1.2 was only attained when f’ was greater than 351 and y was
tation of 901 and Gray and Ohashi (1983) showed that per- between 501 and 601. Similarly, the analysis of Danjon et al.
pendicular orientations of reinforcing fibers provided (2008) showed that an Rf value of 1.2 was only realistic when
comparable reinforcement to randomly oriented fibers. This values of f’ neared 401 and root angles between 401 and 501
work lent support to the use of simple perpendicular root were used. Docker and Hubble (2008) used shear distortion
models in cases where it may be assumed that the roots are angles calculated using their own field data, which ranged
randomly oriented in the soil. from 11 to 251, with resulting Rf values of 0.62–0.98. As it is
Sensitivity analyses carried out by Wu (1976, cited by very difficult to measure field values for y, it should be rec-
Gray and Megahan, 1981), showed that the value of the ognized that there are few accurate estimations of this par-
term in square brackets in eqn [3] (hereafter referred to as the ameter (Thomas and Pollen-Bankhead, 2010). Similar to other
root orientation factor, Rf) was fairly insensitive to normal studies of Rf, Thomas and Pollen-Bankhead (2010) found that
variations in y and f’ (40–701 and 25–401, respectively if root orientation is allowed to vary between 01 and 1801 with
(Figure 3)), with values ranging from 0.92 to 1.31. Wu et al. f’ ranging from 101 to 441 and y ranging from 01 to 901, the
(1979) assumed that all roots were perpendicular to the fail- values for Rf vary from 0.1 to 1.4, presenting a far greater range
ure plane. A constant value of 1.2 was, therefore, selected in values and bringing into question the use of the 1.2 Rf value
by Wu et al. (1979) to replace the bracketed term and the under all conditions. Indeed, their sensitivity analysis showed
simplified equation became that for friction angles ranging from 51 to 451 and failure
surface angles ranging from 101 to 901 an Rf value close to 1.0
DS ¼ 1:2Tr ðAr =AÞ ½5 was found to be more appropriate than 1.2. As the dominance
of particular root angles varies for different root architectures,
As Thomas and Pollen-Bankhead (2010) pointed out, such as heart root, tap root, and plate root systems (Danjon
many researchers have explored the variability of Rf using et al., 2008), it is possible that different Rf values should
different assumptions for f’ and y (Table 2). Despite vari- be used in eqn [5], not only for different soil types with
ations in f’ for the soils used in studies of root tensile strength varying friction angles but also for plants with different root
and the potential range of values that can be used for y architectures.
 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research 113

(a) Stiff clay: friction angle = 10° (b) Silt: friction angle = 25° (c) Rounded sand: friction angle = 27°

1.0 1.0 1.0

Rf (dimensionless)

0.5 0.5 0.5

0.0 0.0 0.0

−0.5 −0.5 −0.5

−1.0 −1.0 −1.0

70 7060 7060
dis An 605040 160
50
40 30 160
50
40 30 140 160
tort gle of 30 20 10 60 80
100 120 140 20
10 60 80
100 120 140 20
10 60 80
100 120
ion of root
,  shear 20 40 of root 20 40 root 20 40
(de Initial a
ngle ngle of Initial a
ngle
plane, i
gre plane, i Initial a ilure plane, i
es) to fa ilu re fa to failure
relative egrees) to
relative egrees) relative s)
(d (d (degree

(d) Soft clay: friction angle = 30° (e) Angular sand: friction angle = 36° (f) Maximum Wu et al. (1979) friction angle = 44°

1.0 1.0 1.0

Rf (dimensionless)

0.5 0.5 0.5

0.0 0.0 0.0

−0.5 −0.5 −0.5

−1.0 −1.0 −1.0

70 7060 7060
6050
dis Angle
50 50
40 30 160 40 30 160 40 30 140 160
tort 120 140 120 140 100 120
ion of she 10
20 100 20 100 20
60 80 10 60 80 10 60 80
, 20 40 root 20 40 root 20 40 root
(de ar ngle of ne, i ngle of ngle of ne, i
gre Initial a la Initial a ne, i Initial a la
es) v e to failure p e to fa ilure pla e to failure p
rela ti s) relati v e s ) rela ti v s)
(degree (de g re (degree

Figure 3 Rf values ½sinð90  cÞ þ cosð90  cÞtanf0 varying with angle of shear distortion, initial root orientation relative to the failure plane,
and soil friction angles of (a) 101, (b) 251, (c) 271, (d) 301, (e) 361, and (f) 441. Reproduced from Thomas, R.E., Pollen-Bankhead, N., 2010.
Modeling root-reinforcement with a fiber-bundle model and Monte Carlo simulation. Ecological Engineering 36(1), 47–61.

Table 2 Range of soil-friction angles and shear-distortion angles used in the literature to calculate Rf values ½sinð90  cÞ þ cosð90  cÞtan f0 
in eqn [3]

Source Range of soil friction angles, f0 (1) Angle of shear distortion, f (1) Rf range
(dimensionless)

Wu (1976, cited by Gray and Megahan, 1981) 20.0–40.0 40.0–70.0 0.92–1.31

Waldron and Dakessian (1981) 30.0 0.0–90.0 0.58–1.16
Gray and Leiser (1982) 20.0–40.0 40.0–70.0 1.00–1.30
Riestenberg and Sovonick-Dunford (1983) 12.0 78.0–90.0 1.01–1.06
Greenway (1987) 435.0 50.0–60.0 1.20
Abernethy and Rutherfurd (2001) 16.0 43.0–66.0 1.00
Simon and Collison (2002) 20.0–40.0 40.0–70.0 1.20
Reubens et al. (2007) – – 1.15
Danjon et al. (2008) B40.0 40.0–50.0 1.20
Docker and Hubble (2008) 27.0–39.6 1.0–25.0 0.62–0.98

Source: From Thomas, R.E., and Pollen-Bankhead, N., 2010. Modeling root-reinforcement with a fiber-bundle model and Monte Carlo simulation. Ecological Engineering 36(1), 47–61.

12.8.2.1 The Use of Fiber-Bundle Models in Root- that the full tensile strength of each root is mobilized during
Reinforcement Modeling soil shearing, and that the roots all break simultaneously
(Waldron and Dakessian, 1981; Greenway, 1987; Pollen et al.,
In addition to the uncertainty discussed with the use of a 2004; De Baets et al., 2008). Waldron and Dakessian (1981)
constant Rf value, it has often been noted that the per- showed that the largest simulated values of DS were only 56%
pendicular root model of Wu et al. (1979) tends to over- of those calculated if the strength of all of the roots were
estimate root reinforcement due to the inherent assumption mobilized at once (i.e., overestimation of 79%). Similarly,
114 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research

Operstein and Frydman (2000) showed overestimation of additional comparisons of FBM root-reinforcement models
root-reinforcement values by up to 400%, and Docker and with the approach of Wu et al. (1979). Overall, the consensus
Hubble (2008) reported overestimates of shear resistance by of these papers has been that FBMs provide more accurate
50–215%. Where driving forces were great enough to break all estimates of root reinforcement than simple perpendicular
of the fibers contained in a soil matrix, Pollen and Simon root models, with most of the overestimation due to the as-
(2005) noted overestimation of DS of up to 100%, but found sumption of simultaneous breaking, being overcome. Pollen
that values of DS could be an order of magnitude too large and Simon (2005) and Docker and Hubble (2008) both ob-
in cases where driving forces did not exceed the strength of served that the magnitude of overestimation by eqn [5] was
the roots contained in the matrix. To correct for this over- species specific: the simplified model of Wu et al. (1979)
estimation, Pollen and Simon (2005) and Pollen (2007) de- tended to provide better predictions for species with narrower
veloped a fiber-bundle model (FBM) (RipRoot) to account for root–diameter distributions, such as grasses, because the ul-
progressive root breaking during shear failure. This model timate tensile strength of plant roots is diameter dependent
used the measured diameters and tensile strengths of roots (Coppin and Richards, 1990; Gray and Sotir, 1996). Docker
crossing the shear plane and the constant Rf of 1.2 used by Wu and Hubble (2008) also noted that the magnitude of over-
et al. (1979). The root reinforcement estimated using RipRoot estimation by eqn [5] depended not only on the species but
(DS) was then substituted into eqn [2]. also on the dominant root failure mode. The results of Loades
FBMs have been widely used in the materials industry to et al. (2010) showed good agreement between FBM predic-
aid in the understanding of composite materials (starting with tions of root reinforcement for their controlled greenhouse
the work of Daniels (1945)). They are easy to parametrize and experiments and for younger plants in the field, but predic-
incorporate the most important aspects of soil–root inter- tions were not as accurate for older plants. They concluded
actions, using a dynamic approach to remove the assumption that this might be corrected with the inclusion of root orien-
that all of the roots in the soil matrix break simultaneously. tations and root elasticity to their FBM. Schwarz et al. (2010b)
FBMs work by apportioning an applied load between N par- applied an FBM approach to modeling of root reinforcement
allel fibers comprising a bundle. Once the load has increased on slope stability in Tuscany, Italy. In this chapter, it was again
sufficiently for a fiber to break, the load that was carried by the confirmed that the Wu model overestimated root reinforce-
broken fiber is redistributed to the remaining (N  1) intact ment and, therefore, over-predicted slope factor of safety by up
roots, each of which then bears a larger load, and is hence to 10%, with this error increasing exponentially for smaller
more likely to break. If this redistribution causes further roots landslides (Schwarz et al., 2010b).
to break, additional redistribution of load occurs until no We envisage that soon the use of FBMs will be extended to
more breakages occur (in this type of model this is known as studies of windthrow in forests, and resistance of emergent
an avalanche effect). Another increment of load is then added plants to removal by fluid drag. In addition, with increased
to the system, and the process is repeated until either all of the data availability and computing power, root-reinforcement
fibers have broken or the maximum driving force acting on FBMs may be able to progress from GLS models to LLS models
the matrix is supported by the fibers contained within it. The that incorporate root geometry and topological information.
simplest FBMs assume that the load from the broken fibers is
redistributed between the remaining intact fibers regardless of 12.8.2.1.1 Load apportionment alternatives in FBMs
their proximity to the broken fiber; this type of redistribution Mickovski et al. (2009) raised an important issue regarding the
is known as global load sharing (GLS) (Hidalgo et al., 2001). assumptions made during the construction of an FBM for
The alternative to GLS is known as local load sharing (LLS) root-reinforcement modeling. Their work highlighted the
(Kun et al., 2007). In this type of FBM, the load is redistributed importance of the decision of how to apportion the load being
according to the proximity of intact fibers to the broken fibers applied to the soil to the roots contained within the soil
(Hidalgo et al., 2001). GLS FBMs assume that the interactions matrix. In the original model of Pollen and Simon (2005),
between the fibers in the bundle are long range, whereas load was added to the roots according to the ratio of root
LLS FBMs assume that short-range interactions between the diameter to the shear plane length. Other alternatives include
fibers are dominant. Hidalgo et al. (2001) stated that in equal load apportionment regardless of root diameter
homogeneous materials, the load sharing rules should fall (Daniels, 1945), or equal stress apportionment where load is
somewhere between GLS and LLS. This is because an im- distributed based on root cross-sectional area (e.g., Hidalgo
portant fraction of load is redistributed to the entire matrix, et al., 2001; Mickovski et al., 2009). An important result in the
even though the concentration of redistribution remains in paper of Mickovski et al. (2009) was that although their FBM
the locality of the breakage. However, fieldwork to examine results matched closely their experimental values for soil–root
root networks can be invasive and fraught with difficulties. shearing, their model always predicted the largest roots to
Therefore, the exact locations of individual roots, and their break first, contrary to observations of their experiments in
locations relative to one another, are rarely known. For this which the smallest roots broke first. This result was examined
reason, it seems more sensible to use a GLS model to simulate further by Thomas and Pollen-Bankhead (2010), who found
reinforcement of roots across a soil shear plane, as the amount that the order of root breaking using three apportionment
of data required to input to a LLS model would be difficult to methods was determined not only by the selected load ap-
obtain. portionment method but also by the tensile strength curve
Since the initial paper of Pollen and Simon (2005), several parameters. Their results showed that when the load applied
papers (e.g., Docker and Hubble, 2008; Mickovski et al., 2009; to the bundle of roots was equally apportioned regardless of
Schwarz et al., 2010b; Loades et al., 2010) have provided root diameter or area, the smallest roots broke first and the
 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research 115

largest roots broke last. Conversely, when the load was ap- streambank, or on top of a potential failure block. For ex-
portioned according to stress (i.e., load normalized by root ample, for 500 eastern sycamore roots median reinforcement
cross-sectional area), the biggest roots broke first and the (DS) varied from 4.86 to 15.08 kPa on a slope and from 9.49
smallest roots broke last. The case where load was appor- to 14.82 kPa on a horizontal surface (Thomas and Pollen-
tioned according to root diameter (as in Pollen and Simon Bankhead, 2010; Figure 4).
(2005)) presented interesting results with FBM output show-
ing that for some species the largest roots broke first and for
some species the smallest roots broke first; the result was 12.8.2.1.3 Displacement over which root reinforcement
dependent on the parameters a and b of the tensile strength is effective
curve regression. In addition, different peak root-reinforce- The peak shear strength of a soil is generally reached at a
ment values were obtained from the FBM for the three smaller shear displacement than that for roots (Pollen et al.,
apportionment functions. The results of Thomas and 2004; Mickovski et al., 2009). The presence of roots can,
Pollen-Bankhead (2010) suggest that only equal load appor- therefore, increase the displacement permitted by a soil matrix
tionment results in the correct system dynamics being before failure by several centimeters (Waldron, 1977; Waldron
modeled in all cases, where the smallest roots are predicted and Dakessian, 1981; Mickovski et al., 2009). This behavior of
to break first. An additional benefit of using equal load the soil–root system is particularly important in streambank
apportionment is the fact that estimated root-reinforcement and slope-stability analyses as the roots may provide resistance
values using this approach were also the smallest, leading to against failure even at large soil displacements (Chiatante
less potential for overestimation of the effects of vegetation on et al., 2002; Kun et al. 2000; Thomas and Pollen-Bankhead,
soil stability. 2010). To account for the displacement over which roots can
Perhaps one of the most interesting recent developments in act to reinforce a soil, it is necessary to know the Young’s
fiber-bundle modeling is the introduction of true stress–strain modulus of the roots being modeled, which can vary de-
behavior (e.g., Kun et al., 2000, 2007; Hidalgo et al., 2002; pending on species and environmental conditions. Including
Schwarz et al., 2010a, 2010b) and the incorporation of damage the ability of a root to stretch in FBMs is made even more
mechanics and strain hardening (e.g., Kun et al., 2000; complicated by the fact that roots do not tend to grow straight
Hidalgo et al., 2001, 2002; Raischel et al., 2006). This research in a soil, but instead have a degree of tortuosity, which, again
has led to the apportionment of load as a function of the varies according to species, soil type, and other environmental
Young’s modulus, so that stiffer fibers, therefore, receive more factors. Several FBM studies have, however, taken the effect of
load (Kun et al., 2007). In the first application of these con- root stretching into account (Kun et al., 2000, 2007; Raischel
cepts to the modeling of plant roots, Schwarz et al. (2010b) et al., 2006; Schwarz et al., 2010a, b), but opportunities for
showed that their FBM produced realistic and encouraging further research in this area of root-reinforcement modeling
results when applied to a landslide scarp in Tuscany, Italy. still exist.
Thomas and Pollen-Bankhead (2010) proposed that the
12.8.2.1.2 Effect of changing root orientations in an FBM output from the RipRoot FBM could be used to infer a time, or
An important improvement made to the original RipRoot shear displacement over which the roots act to reinforce the
model by Thomas and Pollen-Bankhead (2010) was explor- soil. It was recognized that roots that are oriented obtusely to
ation of the effects of root orientations on estimates of root the failure plane (i.e., 1801Zi4901) are initially compressed
reinforcement. Thomas and Pollen-Bankhead (2010) em- when shearing occurs but that after initially buckling, those
ployed statistical distributions to approximate root orien- roots are held in tension as failure progresses. The Monte
tations based on the typical root architecture for a plant Carlo simulations of Thomas and Pollen-Bankhead (2010),
species. Species forming taproot-dominated or herringbone therefore, identified the presence of two independent sub-
networks with multiple vertical sinker roots (e.g., Pinus sp., bundles of roots. The first subbundle contained both the roots
grasses, and some crops) were approximated using a Gaussian that were initially compressed and the roots that were im-
(normal) distribution with the mean perpendicular to the mediately tensioned at the onset of shear failure, and the
ground surface (mean ¼ 901, standard deviation ¼ 22.51), second subbundle contained the roots that buckled and then
species forming plateroot networks with many lateral roots went into tension as shear failure progressed (Figure 4).
(e.g., shallow rooting riparian Salix sp.) were approximated Thomas and Pollen-Bankhead (2010), therefore, hypothesized
using a Gaussian (normal) distribution with the mean that roots that were initially compressed at the onset of
parallel to the ground surface (mean ¼ 01 or 1801, standard shearing extended the time or displacement over which soil
deviation ¼ 22.51), and species forming heartroot networks was reinforced and could have an important influence on both
(e.g., Betula sp. and Plantanus sp.) were approximated using a the form of loading curves and the peak load attained. For
uniform distribution (equal probability of occurrence between plants growing on slopes, species with predominantly slope-
01 and 1801). Using these approximations within a Monte perpendicular roots provided the largest reinforcement, but
Carlo simulation approach, Thomas and Pollen-Bankhead over a short shear displacement or time. Conversely, species
(2010) showed that different distributions of root orientation with many slope-parallel roots provided reduced reinforce-
resulted in considerable differences in both the form of the ment but over a larger soil displacement or time because the
loading curves and the peak load attained for bundles of roots roots were more evenly distributed between those that were
of the same size. They also showed that the same plant would compressed and those that were tensioned at the onset of
provide a different amount of reinforcement when growing on shearing. For plants growing on horizontal surfaces (e.g.,
a slope or streambank face, on the floodplain abutting a floodplains abutting streambanks), the opposite was true.
116 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research

(1) Slopes (2) Horizontal surfaces

16

Load supported by root bundle (kN)

Eastern sycamore: soil friction angle = 30° Eastern sycamore: soil friction angle = 30°
14
12
10
8
6
4
2
0
(a)
16
Load supported by root bundle (kN)

Eastern sycamore: soil friction angle = 30° Eastern sycamore: soil friction angle = 30°
14
12
10
8
6
4
2
0
(b)
16
Load supported by root bundle (kN)

Eastern sycamore: soil friction angle = 30° Eastern sycamore: soil friction angle = 30°
14
12
10
8
6
4
2
0

0 100 200 300 400 500 600 700 800 0 200 400 600 800 1000
(c) Number of broken roots Number of broken roots
Figure 4 Results of Monte Carlo simulations of eastern sycamore trees growing on (1) sloping surfaces (e.g., slopes and streambank faces)
and (2) horizontal surfaces (e.g., streambank tops and floodplains) showing the load supported by root bundles during progressive failure, for
roots distributed (a) uniformly about the growing surface, (b) normally, with the mean parallel to the growing surface and (c) normally, with the mean
perpendicular to the growing surface. Dashed lines indicate mean values for each set of simulations. Reproduced from Thomas, R.E., Pollen-Bankhead, N.,
2010. Modeling root-reinforcement with a fiber-bundle model and Monte Carlo simulation. Ecological Engineering 36(1), 47–61.

12.8.2.1.4 Using a Monte Carlo approach in FBMs Monte Carlo simulations of different statistical distributions
A benefit of the Monte Carlo approach used by Thomas and of root orientations for 500 eastern sycamore roots were
Pollen-Bankhead (2010) is that the resulting loading curves applied to a streambank of Goodwin Creek, Mississippi
form an envelope around potential root-reinforcement values. (Figure 5) using the Bank Stability and Toe Erosion Model
The central tendency and variability of this envelope, in the (BSTEM) version 5.4 (Simon et al., 1999, 2000).
form of the mean and standard deviation in either normal or For each idealized root orientation distribution (approxi-
logarithmic space, can then be input into either deterministic mating those of heart root, plate root, and tap root networks),
models to allow estimation of the range of potential factor the minimum, maximum, and mean values (Table 3), ob-
of safety (Fs) values or probabilistic models to estimate the tained for root reinforcement in the Monte Carlo simulations
reliability index of a slope. These factors of safety or reliability of Thomas and Pollen-Bankhead (2010), were applied to
index values provide a probability of failure of a given BSTEM for a bank at the Goodwin Creek Bendway (Simon
streambank or slope. In the following example, the results of et al., 1999, 2000; Simon and Collison, 2002). The cohesions
 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research 117

90
Layer 1
c′ = 1.41 kPa
89  = 16.9 kN m−3
b = 17.0°
′ = 28.5°
88 Layer 2
c′ = 2.70 kPa
 = 19.3 kN m−3
b= 10.2°
87
′ = 28.1°
Layer 3
c ′ = 2.70 kPa
86  = 19.3 kN m−3
b = 10.2°
Elevation (m)

′ = 28.1°
85 Layer 4
c ′ = 6.30 kPa Right bank
 = 20.0 kN m−3 Failure plane
84 b = 17.0° Layer 1
′ = 27.0° Layer 2
Layer 5
83 c ′ = 6.30 kPa Layer 3
 = 19.7 kN m−3
b = 17.0° Layer 4
82 ′ = 27.0°

81 Layer 5

80

20 25 30 35 40
Station from left bank (m)
Figure 5 Bank geometry, layers, and geotechnical properties of the bank profile modeled at Goodwin Creek, Mississippi modified from a figure
published in Simon, A., Collison, A.J.C., 2002. Quantifying the mechanical and hydrologic effects of riparian vegetation on stream-bank stability.
Earth Surface Processes and Landforms 27(5), 527–546, with permission from Wiley.

ranged from 0.98 (unstable) to 1.08 (conditionally stable)

Table 3 Minimum, maximum, and mean values of root- with a mean of 1.03 (conditionally stable). For the plate root
reinforcement (DS) (kPa) obtained in the Monte Carlo simulations of distribution of roots, Fs ranged from 1.13 to 1.17, with a mean
Thomas and Pollen-Bankhead (2010) for an eastern sycamore tree of 1.15. Finally, with the tap root distribution Fs ranged from
(500 roots) growing on top of a streambank. (f0 ¼ 301) 0.98 to 1.13, with a mean value of 1.08. The results shown in
Root reinforcement (kPa) this procedure provide just one example of the potential use of
Monte Carlo simulations in modeling root-reinforcement and
Heart root Plate root Tap root streambank stability. Additional factors that could be ac-
counted for in a Monte Carlo framework for root-
Minimum 7.36 14.02 7.59 reinforcement modeling and/or streambank or slope-stability
Maximum 11.65 15.41 13.71
modeling include the number of roots, diameter distribution
Mean 9.53 14.79 11.63
of the roots, variability in tensile strength of roots, soil friction
For reference, typical values for effective cohesion of sandy, loamy and clayey soils are angle, and failure plane angle, to name but a few.
0.4, 4.3, and 8.2 kPa, respectively (WPPRU, 2010).

of the upper two bank layers (totaling 1.0-m depth) were

12.8.2.2 Root Architecture Measurement and Modeling
modified to include cohesion due to roots. An unstable bank
configuration was selected as a control case with no root re- As the previous sections have shown, in addition to under-
inforcement being present (2.7-m-flow depth, with water standing variations in root strength it is important to consider
table height set at the same height in the bank). Without ac- root architecture in calculations of root reinforcement.
counting for the presence of roots, the factor of safety (Fs) was Nielson et al. (1997) stated, however, that the root architecture
estimated to be equal to 0.80 (Figure 6). of plants and trees has intrigued scientists for at least 100 years.
BSTEM model results (Figure 6) showed that with the The organization of roots has long proved difficult to study,
addition of a heart-root distribution of roots, streambank Fs both due to their complexity and the practical problems
118 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research

1.20 change significantly over time, not just in its vertical and lat-
1.15 eral extent, but also in its topology. Reubens et al. (2007), for
Streambank Fs (no units) example, note that as seedlings, woody plants tend to form a
1.10
1.05 taproot with laterals, forming a herringbone root structure,
Fs = 1 but over time this pattern becomes more complex.
1.00
Although, historically, the collection of root data has been
0.95
time consuming and difficult (Nielson et al., 1997; Jansen
0.90 and Coelho Netto, 1999; Gyssels and Poesen, 2003), the
0.85 distribution of most root-architectural parameters has been
0.80 Fs with made easier with the use of three-dimensional (3D) digitiza-
no roots tion (Danjon et al. 2006, 2008) and 3D modeling (Dupuy
0.75
0.70 et al. 2005b, c; Pages et al., 2000). Only a few root-
Heartroot Plateroot Taproot reinforcement models have considered the spatial distribution
Root structure of root reinforcement with explicit consideration of root
Figure 6 Minimum, maximum (represented by the vertical lines), architecture (Sakals and Sidle, 2004; Kokutse et al., 2006;
and mean values (represented by the dots) of Fs at Goodwin Creek, Genet et al., 2008; Hubble et al., 2010). A great deal of lit-
Mississippi for an eastern sycamore tree growing on the bank top, erature relating to modeling of root architecture exists, and is
and for statistical distributions for root orientations approximating a not reviewed here (for a full review of the most commonly
heartroot, plateroot, or taproot type of root network. used methods to measure or analyze root architecture see
Reubens et al., 2007). Our ability to predict slope and
streambank stability would be greatly enhanced if root archi-
concerned with the fact that root systems are dynamic entities tecture models for different species were coupled with root-
that grow in a heterogeneous, opaque medium that is difficult reinforcement FBMs.
to noninvasively monitor or control (Nielson et al., 1997).
Root systems are dynamic and are subject to cyclical growth
and death at different stages of development (Fitter, 1996; 12.8.2.3 Hydraulic and Hydrologic Effects of Vegetation
Stokes et al., 2009). Root branching and architecture will,
Thus far, this chapter has focused upon the mechanical effects
therefore, change over the lifetime of a plant, and may even
of vegetation on soil stability. This section provides an over-
change seasonally as environmental conditions vary. The
view of studies that have explored the hydraulic and hydro-
physical and chemical properties of the soil are of major im-
logic effects of vegetation on soil stability. For the sake of
portance in determining the extent and nature of root growth.
brevity, this overview is necessarily short and our aim is not to
Important physical properties of the soil include bulk density,
provide the reader with an exhaustive review, but rather with
mechanical impedance, drainage, and aeration of the soil,
starter literature from which greater detail may be sought.
whereas chemical considerations include soil pH and the
The hydrologic effects of vegetation on slope and bank
concentration of minerals and toxins in the soil (Kozlowski
stability include (see also Gray and Leiser, 1982; Greenway,
et al., 1991). Where soils are compacted, the bulk density of
1987; Gray and Sotir, 1996):
the soil increases and the number of large pores present de-
creases, inhibiting propagation of roots into the soil. An in- • canopy interception, which reduces the amount of water
crease in soil strength, therefore, tends to lead to a reduction needed to be infiltrated by the soil (e.g., Simon and Col-
in root elongation rates and this is also correlated with an lison, 2002), the beneficial effect of which is tempered by
increase in root diameters due to greater radicle expansion the flow of water along the stems of plants, creating locally
under these conditions (Stokes et al., 2009). In addition to the elevated pore-water pressures at the base of stems
increased diameter of roots in stronger soils, the development (Durocher, 1990);
of more lateral roots initiating close to the root apex may • the creation of preferential flow paths (macropores) by
occur, producing a higher density of rooting per unit root developing and decaying roots, and associated biological
length (Marschner, 1991). activity, accelerating the delivery of water into the slope or
Root-architectural characteristics can have a significant in- bank (Simon and Collison, 2002). The extent to which
fluence on the anchorage of roots. The effect of some of these preferential flow affects slope and streambank stability
parameters, including the presence of lateral roots, and depends on the soil properties and their stratigraphy; and
branching angles, was investigated by Dupuy et al. (2005a) • the uptake of water by roots prior to evapotranspiration,
through numerical modeling. Results of this study showed during which water is extracted from the root zone, low-
that root failure forces were largely controlled by the root ering the local phreatic surface (Selby, 1982; Terwilliger,
diameter and the patterns of root branching, similar to other 1990).
studies by El-Khouly (1995), Wu et al. (1988a, b), and Wu
(1995). Root branching is, however, largely ignored in most The lowering of the phreatic surface, and the consequent
root-reinforcement models as this parameter can be highly reduction of pore-water pressures and increase in matric suc-
spatially variable. As previously discussed, root orientations tion within streambanks, has been found to provide signifi-
relative to failure planes are also of significant importance, but cant amounts of additional shear strength, in some cases
are rarely incorporated into models, largely because of a lack greater than the strength provided by the soil matrix alone
of data. Further, the root architecture of a given plant will (i.e., the effective cohesion; Simon et al., 1999; Shields et al.,
 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research 119

2009) and by the soil–root matrix (Simon and Collison, 2002; subsequent transpiration caused negative pore-water pressure
Pollen-Bankhead and Simon, 2010). The hydrologic dis- (matric suction) in the soil mass, significantly increasing soil
advantages of vegetation on soil stability are related to the shear strength. Pollen-Bankhead and Simon (2010) suggested
manner by which soil infiltration characteristics are altered, that this increase in strength was of potentially greater benefit
both at the soil surface and deeper within the soil profile. to slope and/or streambank stability than the other mechan-
Plant roots can also have a significant influence on the ical, hydrologic, and hydraulic effects, but stressed that this
hydraulic scour of slope and streambank material. In recent was limited only to the summer months. Therefore, in a given
years, a significant amount of work has investigated the climatic region, the hydrologic effects of vegetation may be
hydraulic effects of vegetation on geomorphic processes. Sev- particularly important if the most critical time of year for slope
eral laboratory flume and field studies have examined the ef- or streambank instability coincides with a period when plants
fects of plant roots on upland concentrated flows (Mamo and are transpiring. Conversely, the mechanical effect of roots will
Bubenzer, 2001a, b; Gyssels and Poesen, 2003; Gyssels et al., likely provide the greatest reinforcement where periods of
2005; Zhou and Shangguan, 2005; De Baets et al., 2006, slope and streambank instability coincide with dormancy of
2007), and have shown an exponential decline in rill erod- vegetation.
ibility and soil-detachment rates with increasing root-length
densities and root biomass. Flume studies have shown that
root architecture can play an important role in reducing soil 12.8.3 Root-Reinforcement and Geomorphologic
erosion, with fine-rooted grasses being particularly effective at Processes at Different Spatial Scales
preventing soil detachment (De Baets et al., 2006). The results
of these studies are also relevant for streambanks with exposed The scale at which root–soil interactions are modeled varies
root zones near to the water surface, because the root zone of greatly between studies, ranging from calculation of the
vegetation growing on the top of streambanks, rather than the strength of individual roots (Wu et al., 1988b) to entire
plant canopy, interacts with flowing water. Conversely, sub- vegetated slopes (Schwarz et al., 2010b). Additionally, the
merged roots react to the drag exerted by water by either re- spatial scales over which streambank failures and slope fail-
maining erect, oscillating in response to turbulent ures occur can vary dramatically, as can the geometries and
fluctuations, or bending. The magnitude of the drag force is a types of failures. As the factors affecting root growth and root
function of root flexibility, frontal projected area, relative reinforcement of a slope or streambank vary greatly both
depth of submergence, and density (Li and Shen, 1973; Petryk temporally and spatially, consideration needs to be given to
and Bosmajian, 1975; Pasche and Rouvé, 1985; Fathi- the scale at which variables affecting root growth and resulting
Moghadam and Kouwen, 1997; Nepf, 1999; Järvelä, 2002, root-reinforcement act, and whether these variables influence
2004; Wilson et al., 2003, 2006; McBride et al., 2007; White individual roots, or entire stands of vegetation.
and Nepf, 2008), which may all vary by plant type and age. A single root-reinforcement value is commonly applied to
According to Newton’s second law, vegetation must also affect an entire stand of vegetation (Schwarz et al., 2010b), with
the flow patterns. It does so by adding roughness and hence smaller-scale variations within a stand usually being ignored.
reducing the velocity in vegetated areas (Simon et al., 2004), Root-reinforcement estimates have commonly been applied to
introducing turbulence and inducing scour along the vegeta- 2D slope (e.g., Bathurst et al., 2007) and streambank models
tion–channel interface, and forcing flow back toward the open (e.g., Pollen and Simon, 2005; Pollen-Bankhead et al., 2009;
channel (McBride et al., 2007; White and Nepf, 2008). The van de Wiel and Darby, 2007), with only a small number
result of these interactions with transported sediment is the of papers applying root reinforcement in three dimensions
trapping and sorting of fine sediment (Lowrance et al., 1988; (e.g., Schmidt et al., 2001; Kokutse et al., 2006). The effects
Tsujimoto, 1999). of roots are also generally limited to the uppermost soil layers
Pollen-Bankhead and Simon (2010) designed a series of where roots tend to be concentrated (Bischetti et al., 2005;
experiments to compare the relative magnitude of the Pollen-Bankhead and Simon, 2009). However, it is commonly
hydraulic, hydrologic, and mechanical effects of the roots of accepted that root density declines nonlinearly with increasing
five tree species and one grass species in North Mississippi. depth in the soil profile (Pollen-Bankhead and Simon, 2009;
The hydrologic effects of riparian vegetation on soil shear Mao et al., 2012), a relationship that has been incorporated
strength were considerable but varied seasonally. When com- into some streambank stability models (van de Wiel and
pared to a control, it was shown that at a depth of 0.3 m, the Darby, 2007; Pollen-Bankhead and Simon, 2009). Ideally,
change in soil shear strength ranged from a decrease of 0.95 lateral and spatial variations in root reinforcement should be
kPa to an increase of 3.2 kPa. At a depth of 0.7 m, the change accounted for in slope and streambank stability models,
in soil shear strength ranged from a decrease of 0.54 kPa to an instead of applying a single root-reinforcement value for
increase of 5.0 kPa. During or directly following a storm event, an entire slope or streambank. Obtaining root data to
vegetation tended to increase pore-water pressure in the soil parametrize models at a finer resolution can, however, be
and hence decrease soil shear strength. This negative effect was difficult and time consuming. In addition, the inherent
rapidly overcome during times of the year when plants were variability in rooting characteristics associated with changing
actively transpiring, but was longer lasting during the winter environmental factors generally precludes the extrapolation
months when evapotranspiration was limited. Therefore, of data collected for a given species at one site to other
during winter and spring, mechanical root reinforcement had sites.
the largest beneficial effect on soil stability. Conversely, during Streambank vegetation commonly grows on both the
the hot summer months, the uptake of water by roots and its sloping face of the bank and on the bank top, or floodplain.
120 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research

The effect of root reinforcement on both sections of the bank, root reinforcement provided. As with streambanks, the pres-
therefore, needs to be considered. As Thomas and Pollen- ence of vertical sinker roots will act to reinforce the bottom of
Bankhead (2010) have shown, different types of root networks a given failure surface through a slope, whereas lateral roots
that can provide varying amounts of strength depending on will provide greater reinforcement to the back wall of a po-
their location relative to the failing block of streambank. Their tential landslide location (Reubens et al., 2007; Danjon et al.,
results suggested that species with a tap root-dominated root 2008). Root strength can also vary according to slope topo-
network will provide the greatest stability where growing on a graphy that varies more than along a streambank, setting up a
sloping bank face, as this tap root should grow deep enough to positive feedback mechanism in which concave hollows tend
cross potential failure planes within the bank. By contrast, the to be more unstable than convex noses (Hales et al., 2009),
results of Thomas and Pollen-Bankhead (2010) suggested that thereby causing the same topographic patterns to repeatedly
trees planted on the top of a bank will have the greatest re- develop on a given hillslope. As mass failures on slopes tend
inforcing effect if they have a plate root structure, with many to occur at larger scales than those on streambanks, failure
lateral roots growing across the upper part of potential failure plane lengths tend to be longer, and the mass of soil involved
surfaces through the bank. Studies by Zhou et al. (1998) and is generally larger. Therefore, at the scale of slopes the re-
Stokes (2002) have also shown that lateral roots play a par- inforcement provided by a single tree is likely to be fairly
ticularly important role where they cross the upper parts of a insignificant to overall slope stability, with root reinforcement
failure surface. having to be present over a larger area to provide a significant
The ratio of bank height to rooting depth is particularly increase in slope stability.
important in determining which types of vegetation will re-
inforce a given bank. For example, grasses may provide sig-
nificant reinforcement to lower bank heights through their 12.8.4 Conclusions and Direction of Future
dense mat of fine roots, but where growing on higher banks, Research
the rooting zone tends to become undercut, with cantilever
failures resulting. Grass roots also tend to grow vertically ra- A great deal is now known about the mechanics of root re-
ther than laterally, providing little reinforcing effect during a inforcement, and techniques for modeling root–soil inter-
cantilever failure, as the roots are growing parallel to the actions have improved significantly over the past 5–10 years
failure plane. For higher banks, the deeper-penetrating root with the introduction of FBMs to the field of root reinforce-
networks of trees, therefore, provide greater increases in sta- ment. A number of recent articles have confirmed that the use
bility. For higher streambanks, where bank-top vegetation of FBMs produces more accurate root-reinforcement values
protects only the upper bank, the dominant bank failure than the older perpendicular root models such as that of Wu
process may actually be changed by the vegetation. Where et al. (1979) (Pollen and Simon, 2005; Docker and Hubble,
roots reinforce only the upper part of the soil, they increase 2008; Mickovski et al., 2009; Thomas and Pollen-Bankhead,
resistance against planar and rotational failure but hydraulic 2010; Schwarz et al., 2010a, b; Loades et al., 2010). These
or seepage erosion of the bank toe may eventually lead to studies have also highlighted that it is not just the strength of
undercutting of the root zone, resulting in cantilever failure roots that affects total root reinforcement of a slope or
instead of a planar failure. At the scale of streambanks, streambank, but also the number, diameter, and orientation of
therefore, even individual trees can affect the location and these roots relative to the potential failure plane.
timing of mass failures. Indeed, an eroding bank with trees Other important findings from recent root-reinforcement
growing on the bank top commonly has a scalloped planform, studies include an explanation for changing root tensile
where mass wasting occurs preferentially between the trees to strength with root diameter due to changes in cellulose con-
set the bank angle back to the effective friction angle of the soil tent with increasing diameter (Genet et al., 2005), the pres-
types present. Although hydraulic scour and gravity may result ence of zones of different root strengths with varying slope
in soil removal and exposure of the root zone under the trees, topography (Hales et al., 2009), and the use of Monte Carlo
a steeper bank geometry tends to remain in the vicinity of an simulations to model the variability of root reinforcement for
individual tree until undercutting causes the driving forces different statistical distributions of root angles, relative to a
acting on the bank to overcome the resisting forces of the roots failure plane (Thomas and Pollen-Bankhead, 2010). However,
and soil. Because failure blocks along streambanks tend to be although the processes by which roots fail are well under-
limited to the scale of a few meters, it is the type and density of stood, and the mechanics of breaking, stretching, com-
the vegetation growing specifically on the bank edge and bank pression, and pullout can be modeled successfully, several
face that controls how much reinforcement is provided to a obstacles to more accurate root-reinforcement predictions re-
streambank by vegetation. main. First, the interactions between these mechanisms vary as
Mass failures of slopes tend to occur at a larger scale than a function of a number of factors, including species, root
streambanks. The presence of roots may force a deeper failure elasticity, soil type, biotic interaction, and moisture avail-
surface for a given slope, and similar to banks, lines of ability. Second, obtaining good-quality root data is difficult
weakness will be exploited through zones of soil weakness and due to the opaque medium they grow in, and variability in
where vegetation provides less reinforcement. Vegetation root parameters can be considerable not only between species
growing on slopes is, however, unlikely to change the mode of but also within the same species growing in different con-
mass failure, simply the geometry of the failure. Similar to ditions. It, therefore, appears that it is not our ability to
banks, the location of different types of root networks relative quantify and model the dominant processes that are the
to potential failure surfaces can greatly affect the magnitude of limiting factor in producing accurate root-reinforcement
 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research 121

predictions, but rather the availability of high-quality input Beaudoin, J.J., 1990. Handbook of Fiber-Reinforced Concrete. Noyes Publications,
data to parameterize the models over a diverse range of scales Park Ridge, NJ, 332 pp.
Bischetti, G.B., Chiaradia, E.A., Simonato, T., Speziali, B., Vitali, B., Vullo, P., Zocco,
and environments.
A., 2005. Root strength and root area ratio of forest species in Lombardy
Answers to several remaining research questions in the field (northern Italy). Plant and Soil 278, 11–22.
of root reinforcement could provide significant advances to Chiatante, D., Scippa, S.G., Di Iorio, A., Sarnataro, M., 2002. The influence of steep
future slope and streambank stability predictions. First, al- slopes on root system development. Journal of Plant Growth Regulation 21,
though the concentration of cellulose in a root of a given 247–260.
Coppin, N.J., Richards, I.G., 1990. Use of Vegetation in Civil Engineering.
diameter appears to be related to its tensile strength (Genet Butterworths, London, 238 pp.
et al., 2005), further investigation of the biotic and abiotic Daniels, H.E., 1945. The statistical theory of the strength of bundles of threads. I.
variables that affect cellulose content could yield additional Proceedings of the Royal Society of London, Series A 183(995), 405–435.
insight into variations in strength within and between species, Danjon, F., Barker, D.H., Drexhage, M., Stokes, A., 2008. Using three-dimensional
plant root architecture in models of shallow-slope stability. Annals of Botany
and at different topographic and geographic locations. Second,
101, 1281–1293.
root tensile strength measurements at different root diameters Danjon, F., Pagés, L., Descorps, M.C., 2006. Root diameter as predictor of borne
have rarely taken into account the age of the roots as well as root volume: estimating the missing root characteristics in Pinus pinaster (Ait)
their diameter. As the internal structure and texture of roots root systems. In: Ephrath, J., Godbold, D.L. (Eds.), Proceedings of the COST
changes with age, it may be that roots of the same species, with E38 Meeting on Woody Root Processes, Revealing the Hidden Half. Sede Boqer,
Israel, 4–8 February 2006.
the same diameter may have different tensile strengths at dif-
De Baets, S., Poesen, J., Galindo-Morales, P., Knapen, A., 2007. Impact of root
ferent ages of growth. No studies have tested how root strength architecture on the erosion-reducing potential of roots during concentrated flow.
changes with root age, although preliminary findings from a Earth Surface Processes and Landforms 23, 1323–1345.
study by M Dumlao et al. (W Silk, personal communication) De Baets, S., Poesen, J., Gyssels, G., Knapen, A., 2006. Effects of grass roots
show that in young oat seedlings, the more mature part of the on the erodibility of topsoils during concentrated flow. Geomorphology 76,
54–67.
main root nearer the stem of the seedling is stronger than the De Baets, S., Poesen, J., Reubens, B., Wemans, K., De Baerdemaeker, J., Muys, B.,
newer tissue in the elongation zone of the same root, even 2008. Root tensile strength and root distribution of typical Mediterranean plant
though the root diameter remains fairly constant. This finding species and their contribution to soil shear strength. Plant and Soil 305,
confirms that as the cellular structure and texture of a root 207–226.
Docker, B.B., 2003. Biotechnical engineering on alluvial riverbanks of south eastern
changes during elongation and growth, the tensile strength of a
Australia: a quantified model of the earth-reinforcing properties of some native
root can also be expected to change. Third, although the riparian trees. Ph.D. thesis, University of Sydney, Australia.
mechanisms of root failure can be accurately modeled, most Docker, B.B., Hubble, T.C.T., 2008. Quantifying root-reinforcement of river bank
models assume that the roots are not branched (although see soils by four Australian tree species. Geomorphology 100, 401–418.
Schwarz et al. (2010a) for a recent exception). During soil Dupuy, L., Fourcaud, T., Stokes, A., 2005a. A numerical investigation into factors
affecting the anchorage of roots in tension. European Journal of Soil Science
shearing, this assumption can lead to overestimation of the 56, 319–327.
number of roots that are pulled out of the soil intact relative to Dupuy, L., Fourcaud, T., Stokes, A., 2005b. A numerical investigation into the
the number of roots that break (Dupuy et al., 2005a; Mickovski influence of soil type and root architecture on tree anchorage. Plant and Soil
et al., 2007). Patterns of branching also vary considerably be- 278, 119–134.
Dupuy, L., Fourcaud, T., Stokes, A., Danjon, F., 2005c. A density-based
tween species, and with changing soil environments (Wang
approach for the modelling of root architecture: application to Maritime
et al., 2006). Branching of roots can, therefore, be highly spa- Pine (Pinus pinaster Ait.) root systems. Journal of Theoretical Biology 236,
tially variable, presenting difficulties for modelers. Investi- 323–334.
gations relating different branching patterns to different species Durocher, M.G., 1990. Monitoring spatial variability in forest interception.
and soil conditions would greatly aid the addition of this Hydrological Processes 4, 215–229.
El-Khouly, M.A., 1995. Analysis of soil-reinforcement interaction. Ph.D. dissertation.
variable to future FBMs. The development of characteristic Ohio State University, Columbus, OH.
fractal dimensions (a parameter that provides simple charac- Ennos, A.R., 1990. The anchorage of leek seedlings: the effect of root length and
terizations of how complex branching structures fill space) for soil strength. Annals of Botany 65, 409–416.
specific root topologies or soil conditions may be one way that Fan, C.C., Su, C.F., 2008. Role of roots in the shear strength of root-reinforced
soils with high moisture content. Ecological Engineering 33, 157–166.
root branching can be accounted for in FBMs (Reubens et al.,
Fathi-Moghadam, M., Kouwen, N., 1997. Nonrigid, nonsubmerged, vegetative
2007; Schwarz et al., 2010a). Finally, improving the linkages roughness on floodplains. Journal of Hydraulic Engineering, ASCE 123(1),
between root-reinforcement, root architecture, and root growth 51–57.
models would greatly enhance our ability to provide accurate Fitter, A.H., 1996. Characteristics and functions of root systems. In: Waisel, Y.,
streambank and slope-stability predictions with and without Eshel, A., Kafkafi, U. (Eds.), Plant Roots: The Hidden Half, 2nd edn. Dekker,
New York, NY, pp. 1–21.
vegetation. Fredlund, D.G., Morgenstern, N.R., Widger, R.A., 1978. The shear strength of
unsaturated soils. Canadian Geotechnical Journal 15, 313–321.
Genet, M., Kokutse, N., Stokes, A., Fourcaud, T., Cai, X., Ji, J., Mickovski, S.B.,
2008. Root reinforcement in plantation of Cryptomeria japonica D. Don: effect of
References tree age and stand structure on slope stability. Forest Ecology and Management
256, 1517–1526.
Genet, M., Stokes, A., Salin, F., Mickovski, S.B., Fourcaud, T., Dumail, J.-F., van
Abernethy, B., Rutherfurd, I.D., 2001. The distribution and strength of riparian Beek, R., 2005. The influence of cellulose content on tensile strength in tree
tree roots in relation to riverbank reinforcement. Hydrological Processes 15, roots. Plant and Soil 278, 1–9.
63–79. Gray, D.H., 1974. Reinforcement and stabilization of soil by vegetation.
Bathurst, J.C., Moretti, G., El-Hames, A., Begueira, S., Garzia-Ruiz, J.M., 2007. Proceedings of the ASCE. Journal of the Geotechnical Division 100(GT6),
Modelling the impact of forest loss on shallow landslide sediment yield, Ijuez 695–699.
River catchment, Spanish Pyrenees. Hydrologic Earth System Science 11(1), Gray, D.H., Leiser, A.T., 1982. Biotechnical Slope Protection and Erosion Control.
569–583. Van Nostrand Reinhold Company, New York, NY, 271 pp.
122 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research

Gray, D.H., Megahan, W.F., 1981. Forest vegetation removal and slope stability in potential affect the pullout resistance of model root systems. European Journal
the Idaho Batholith. Intermountain Forest and Range Experiment Station of Soil Science 58, 1471–1481.
Research Paper INT-271. USDA Forest Service, Ogden, UT. Mickovski, S.B., Hallett, P.D., Bransby, M.F., Davies, M.C.R., Sonnenberg, R.,
Gray, D.H., Ohashi, H., 1983. Mechanics of fiber reinforcement in sand. Journal of Bengough, A.G., 2009. Mechanical reinforcement of soil by willow roots:
Geotechnical Engineering 109, 335–353. impacts of root properties and root failure mechanism in controlled laboratory
Gray, D.H., Sotir, R.B., 1996. Biotechnical and Soil Bioengineering Slope tests. Soil Science Society of America Journal 73, 1276–1285.
Stabilization: A Practical Guide for Erosion Control. Wiley, New York, NY, 400 Nepf, H.M., 1999. Drag, turbulence, and diffusion in flow through emergent
pp. vegetation. Water Resources Research 35(2), 479–489.
Greenway, D.R., 1987. Vegetation and slope stability. In: Anderson, M.G., Richards, Nielson, K.L., Lynch, J.P., Weiss, H.N., 1997. Fractal geometry of bean root
K.S. (Eds.), Slope Stability. Wiley, Chichester, pp. 187–230. systems: correlations between spatial and fractal dimension. American Journal of
Gyssels, G., Poesen, J., 2003. The importance of plant root characteristics in Botany 84(1), 26–33.
controlling concentrated flow erosion rates. Earth Surface Processes and Niklas, K.J., 1992. Plant Biomechanics: An Engineering Approach to Plant Form
Landforms 28, 371–384. and Function. The University of Chicago Press, Chicago, IL, 622 pp.
Gyssels, G., Poesen, J., Bochet, E., Li, Y., 2005. Impact of plant roots on the Nilaweera, N.S., Nutalaya, P., 1999. Role of tree roots in slope stabilization. Bulletin
resistance of soils to erosion by water: a review. Progress in Physical of Engineering Geology and the Environment 57, 337–342.
Geography 29, 189–217. Operstein, V., Frydman, S., 2000. The influence of vegetation on soil strength.
Hales, T.C., Ford, C.R., Hwang, T., Vose, J.M., Band, L.E., 2009. Topographic and Ground Improvement 4, 81–89.
ecologic controls on root reinforcement. Journal of Geophysical Research, 114. Pages, L., Asseng, S., Pellerin, S., Diggle, A., 2000. Modelling root system growth
http://dx.doi.org/10.1029/2008JF001168. and architecture. In: Smit, A.L., Bengough, A.G., Engels, C., van Noordwijk, M.
Hidalgo, R.C., Kun, F., Herrmann, H.J., 2001. Bursts in a fiber bundle model with (Eds.), Root Methods: A Handbook. Springer, Berlin, pp. 113–146.
continuous damage. Physical Review E 64(6), 066122. http://dx.doi.org/ Pasche, E., Rouvé, G., 1985. Overbank flow with vegetatively roughened flood
10.1103/PhysRevE.64.066122. plains. Journal of Hydraulic Engineering, ASCE 111(9), 1262–1278.
Hidalgo, R.C., Kun, F., Herrmann, H.J., 2002. Creep rupture of viscoelastic fiber Petryk, S., Bosmajian, G., 1975. Analysis of flow through vegetation. Journal of the
bundles. Physical Review E 65(3), 032502. http://dx.doi.org/10.1103/ Hydraulics Division, ASCE 101(HY7), 871–884.
PhysRevE.65.032502. Pollen, N., 2007. Temporal and spatial variability in root reinforcement of
Hubble, T.C.T., Docker, B.B., Rutherfurd, I.D., 2010. The role of riparian trees in streambanks: accounting for soil shear strength and moisture. Catena 69,
maintaining riverbank stability: a review of Australian experience and practice. 197–205.
Ecological Engineering 36(3), 292–304. Pollen, N., Simon, A., 2005. Estimating the mechanical effects of riparian vegetation
Jansen, R.C., Coelho Netto, A.L., 1999. Root systems distribution and functions in on streambank stability using a fiber bundle model. Water Resources Research
a mountainous tropical rainforest environment. Geomorphic responses to 41, W07025. http://dx.doi.org/10.1029/2004WR003801.
vegetation changes: problems and remedial work. Proceedings of the Pollen, N., Simon, A., Collison, A.J.C., 2004. Advances in assessing the
International Geographical Union, Commission on Geomorphic Responses to mechanical and hydrologic effects of riparian vegetation on streambank stability.
Environmental Changes, GEOVEG99 Meeting, Rio de Janeiro. In: Bennett, S.J., Simon, A. (Eds.), Riparian Vegetation and Fluvial
Järvelä, J., 2002. Flow resistance of flexible and stiff vegetation: a flume study with Geomorphology. Water Science and Applications 8. AGU, Washington, DC, pp.
natural plants. Journal of Hydrology 269(1–2), 44–54. 125–139.
Järvelä, J., 2004. Determination of flow resistance caused by non-submerged Pollen-Bankhead, N., Simon, A., 2009. Enhanced application of root-reinforcement
woody vegetation. International Journal of River Basin Management 2(1), algorithms for bank-stability modeling. Earth Surface Processes and Landforms
61–70. 34(4), 471–480.
Kokutse, N., Fourcaud, T., Kokou, K., Neglo, K., Lac, P., 2006. 3D numerical Pollen-Bankhead, N., Simon, A., 2010. Hydrologic and hydraulic effects of riparian
modelling and analysis of the influence of forest structure on hillslopes stability. root networks on streambank stability: is mechanical root-reinforcement the
In: Marui, H., Marutani, T., Watanabe, N., et al. (Eds.), Interpraevent 2006: whole story? Geomorphology 116, 353–362.
Disaster Mitigation of Debris Flows, Slope Failures and Landslides. Niigata, Pollen-Bankhead, N., Simon, A., Jaeger, K., Wohl, E., 2009. Destabilization of
Japan, 25–27 September 2006. Universal Academy Press, Tokyo, pp. 561–567. streambanks by removal of invasive species in Canyon de Chelly National
Kozlowski, T.T., Kramer, P.J., Pallardy, G.S., 1991. The Physiological Ecology of Monument, Arizona. Geomorphology 103, 363–374.
Woody Plants. Academic Press, New York, NY, 464 pp. Raischel, F., Kun, F., Herrman, H.J., 2006. Failure process of a bundle of
Kun, F., Raischel, F., Hidalgo, R.C., Herrmann, H.J., 2007. Extensions of fiber plastic fibers. Physics Review E 73(6). http://dx.doi.org/10.1103/
bundle models. Lecture Notes in Physics 705, 57–92. PhysRevE.73.066101.
Kun, F., Zapperi, S., Herrmann, H.J., 2000. Damage in fiber bundle models. Reubens, B., Poesen, J., Danjon, F., Geudens, G., Muys, B., 2007. The role of fine
European Physical Journal B 17, 269–279. and coarse roots in shallow slope stability and soil erosion control with a focus
Li, R.-M., Shen, H.W., 1973. Effect of tall vegetations on flow and sediment. on root system architecture: a review. Trees 21, 385–402.
Journal of the Hydraulics Division, ASCE 99(HY5), 793–814. Riestenberg, M., Sovonick-Dunford, S.S., 1983. The role of woody vegetation in
Loades, K.W., Bengough, A.G., Bransby, M.F., Hallet, P.D., 2010. Planting density stabilizing slopes in the Cincinnati area, Ohio. Geological Society of America
influence on fibrous root reinforcement of soils. Ecological Engineering 36(3), Bulletin 94(4), 506–518.
276–284. Sakals, M.E., Sidle, R.C., 2004. A spatial and temporal model of root cohesion in
Lowrance, R.R., McIntyre, S., Lance, C., 1988. Erosion and deposition in a field/ forest soils. Canadian Journal of Forest Research 34, 950–958.
forest system estimated using cesium-137 activity. Journal of Soil and Water Schiechtl, H.M., 1980. Bioengineering for Land Reclamation and Conservation.
Conservation 43, 195–199. University of Alberta Press, Edmonton, Alberta, 404 pp.
Mamo, M., Bubenzer, G.D., 2001a. Detachment rate, soil erodibility and soil Schmidt, K.M., Roering, J., Stock, J., Dietrich, J.D., Montgomery, W.E., Schaub,
strength as influenced by living plant roots: part II. Field study. American D.R.T., 2001. The variability of root cohesion as an influence on shallow
Society of Agricultural Engineers 44, 1175–1181. landslide susceptibility in the Oregon Coast Range. Canadian Geotechnical
Mamo, M., Bubenzer, G.D., 2001b. Detachment rate, soil erodibility and soil Journal 38, 995–1024.
strength as influenced by living plant roots: part I. Laboratory study. American Schwarz, M., Lehmann, P., Or, D., 2010a. Quantifying lateral root reinforcement in
Society of Agricultural Engineers 44, 1167–1174. steep slopes – from a bundle of roots to tree stands. Earth Surface Processes
Mao, Z., Saint-Andre, L., Genet, M. et al. 2012. Engineering ecological protection and Landforms. doi:10.1002/esp.1927.
against landslides in mountain forests: choosing cohesion models. Ecololgical Schwarz, M., Preti, F., Giadrossich, F., Lehmann, P., Or, D., 2010b. Quantifying the
Engineering 45, 55–69. role of vegetation in slope stability: a case study in Tuscany (Italy). Ecological
Marschner, H., 1991. Mineral Nutrition of Higher Plants, 2nd edn. Academic Press, Engineering 36(3), 281–291.
New York, NY, 889 pp. Selby, M.J., 1982. Hillslope Materials and Processes. Oxford University Press,
McBride, M., Hession, W.C., Rizzo, D.M., Thompson, D.M., 2007. The influence of Oxford, 480 pp.
riparian vegetation on near-bank turbulence: a flume experiment. Earth Surface Shewbridge, S.E., Sitar, N., 1990. Deformation characteristics of reinforced sand in
Processes and Landforms 32(13), 2019–2037. direct shear. Journal of Geotechnical Engineering 116(7), 1153–1170.
Mickovski, S.B., Bengough, A.G., Bransby, M.F., Davies, M.C.R., Hallett, P.D., Shields, Jr. F.D., Simon, A., Dabney, S., 2009. Streambank dewatering for increased
Sonnenberg, R., 2007. Material stiffness, branching pattern and soil matric stability. Hydrological Processes 23, 1537–1547.
 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research 123

Simon, A., Bennett, S.J., Neary, V.S., 2004. Riparian vegetation and fluvial Waldron, L.J., 1977. The shear resistance of root-permeated homogeneous and
geomorphology: problems and opportunities. In: Bennett, S.J., Simon, A. (Eds.), stratified soil. Soil Science Society of America Journal 41, 843–849.
Riparian Vegetation and Fluvial Geomorphology. Water Science and Applications Waldron, L.J., Dakessian, S., 1981. Soil reinforcement by roots: calculation of
8. AGU, Washington, DC, pp. 1–10. increased soil shear resistance from root properties. Soil Science 132(6),
Simon, A., Collison, A.J.C., 2002. Quantifying the mechanical and hydrologic 427–435.
effects of riparian vegetation on stream-bank stability. Earth Surface Processes Wang, Z., Guo, D., Wang, X., Gu, J., Mei, L., 2006. Fine root architecture,
and Landforms 27(5), 527–546. morphology, and biomass of different branch orders of two Chinese temperate
Simon, A., Curini, A., Darby, S.E., Langendoen, E.J., 1999. Stream-bank mechanics tree species. Plant and Soil 288, 155–171.
and the role of bank and near-bank processes in incised channels. In: Darby, White, B.L., Nepf, H.M., 2008. A vortex-based model of velocity and shear stress in
S.E., Simon, A. (Eds.), Incised River Channels. Wiley, New York, NY, pp. a partially vegetated shallow channel. Water Resources Research 44(1), W01412.
123–152. http://dx.doi.org/10.1029/2006WR005651.
Simon, A., Curini, A., Darby, S.E., Langendoen, E.J., 2000. Bank and near-bank Wilson, C.A.M.E., Stoesser, T., Bates, P.D., Batemann Pinzen, A., 2003. Open
processes in an incised channel. Geomorphology 35, 193–217. channel flow through different forms of submerged flexible vegetation. Journal
Stevenson, S.L., Mills, H.H., 1999. Contrasting vegetation of noses and hollows in of Hydraulic Engineering, ASCE 29(11), 847–853.
the Valley and Ridge province, southwestern Virginia. Journal of the Torrey Wilson, C.A.M.E., Yagci, O., Rauch, H.-P., Stoesser, T., 2006. Application of the
Botanical Society 126, 197–212. drag force approach to model the flow-interaction of natural vegetation.
Stokes, A., 2002. Biomechanics of tree root anchorage. In: Waisel, Y., Eshel, A.,
International Journal of River Basin Management 4(2), 137–146.
Kafkafi, U. (Eds.), Plant Roots: The Hidden Half, 3rd edn. Dekker, New York, NY,
WPPRU (Watershed Physical Processes Research Unit), 2010. USDA-ARS-NSL
pp. 175–186.
Watershed Physical Processes Research Unit: Bank Stability and Toe Erosion
Stokes, A., Atger, C., Bengough, A.G., Fourcaud, T., Sidle, R.C., 2009. Desirable
Model. http://www.ars.usda.gov/Research/docs.htm?docid ¼ 5044 (accessed
plant root traits for protecting natural and engineered slopes against landslides.
December 2010).
Plant and Soil 324, 1–30. http://dx.doi.org/10.1007/s11104-009-0159-y.
Wu, T.H., 1976. Investigation of Landslides on Prince of Wales Island. Geotechnical
Stokes, A., Mattheck, C., 1996. Variation of wood strength in tree roots. Journal of
Experimental Botany 47(298), 693–699. Engineering Report 5. Civil Engineering Department, Ohio State University,
Terwilliger, V.J., 1990. Effects of vegetation on soil slippage by pore pressure Columbus, OH.
modification. Earth Surface Processes and Landforms 15(6), 553–570. Wu, T.H., 1995. Slope stabilization. In: Morgan, R.P.C., Rickson, R.J. (Eds.), Slope
Thomas, R.E., Pollen-Bankhead, N., 2010. Modeling root-reinforcement with a fiber- Stabilization and Erosion Control: A Bioengineering Approach. E. and F.N. Spon,
bundle model and Monte Carlo simulation. Ecological Engineering 36(1), pp. 221–264.
47–61. Wu, T.H., Beal, P.E., Lan, C., 1988a. In-situ shear test of soil–root systems. Journal
Thorne, C.R., 1990. Effects of vegetation on riverbank erosion and stability. of Geotechnical Engineering, ASCE 114(GT12), 1376–1394.
In: Thornes, J.B. (Ed.), Vegetation and Erosion. Wiley, Chichester, Wu, T.H., McKinnell, III W.P., Swanston, D.N., 1979. Strength of tree roots and
pp. 125–144. landslides on Prince of Wales Island, Alaska. Canadian Geotechnical Journal 16,
Tosi, M., 2007. Root tensile strength relationships and their slope stability 19–33.
implications of three shrubs species in the Northern Apennines (Italy). Wu, T.H., McOmber, R.M., Erb, R.T., Beal, P.E., 1988b. Study of soil–root
Geomorphology 87, 268–283. interaction. Journal of Geotechnical Engineering, ASCE 114(GT12), 1351–1375.
Tsujimoto, T., 1999. Fluvial processes in streams with vegetation. Journal of Zhou, Y., Watts, D., Li, Y., Cheng, X., 1998. A case study of effect of lateral roots
Hydraulic Research 37, 789–803. of Pinus yunnanensis on shallow soil reinforcement. Forest Ecology and
Van De Wiel, M.J., Darby, S.E., 2007. A new model to analyse the impact of woody Management 103, 107–120.
riparian vegetation on the geotechnical stability of river banks. Earth Surface Zhou, Z.C., Shangguan, Z.P., 2005. Soil anti-scouribility enhanced by plant roots.
Processes and Landforms 32, 2185–2198. Journal of Integrated Plant Biology 47, 676–682.

Biographical Sketch

Natasha Pollen-Bankhead is a quantitative geomorphologist, and is currently a postdoctoral research associate at

the USDA-ARS National Sedimentation Laboratory. Natasha received her BSc and PhD from the Department of
Geography, King’s College, University of London. Her PhD, entitled, ‘The effects of riparian vegetation on stream-
bank stability: mechanical and hydrological interactions’ led to advances in the way that the strength of root
networks is modeled, with the introduction of the use of fiber bundle models to this field, and the creation of the
root-reinforcement model, RipRoot. She has worked on projects involving the management of riparian species on
rivers across the USA, and has published in some of the top journals in her field. Her main areas of research are
the measurement and modeling of effects of riparian plants on streambank stability, and the resistance of root
networks of submerged vegetation to removal by flow.
124 The Reinforcement of Soil by Roots: Recent Advances and Directions for Future Research

Andrew Simon is a research geologist at the USDA-ARS, National Sedimentation Laboratory in Oxford Missis-
sippi. He received his PhD from Colorado State University under the direction of Professor Stanley Schumm. He
has 30 years of research experience (16 with the USGS) in channel response of unstable channels, streambank
processes and bank-stability modeling, and quantifying the role of vegetation on fluvial processes. He is the
author of more than 100 technical publications, has edited several books and journals and manages the devel-
opment of the Bank-Stability and Toe-Erosion Model (BSTEM). Dr. Simon is an adjunct professor at the Uni-
versity of Mississippi and Oregon State University.

Robert Thomas is a quantitative fluvial geomorphologist with 10 years’ experience of studying fluvial dynamics in the UK, US, and Canada. He has interests in
remote sensing, hydraulics, sediment transport, and slope stability with and without vegetation. During his PhD, obtained from the University of Leeds, UK
in 2006, Rob studied the behavior of the sandy, braided, South Saskatchewan River, Saskatchewan as it responded to impoundment by Gardiner Dam. He has
been the lead algorithm developer of the Bank-Stability and Toe-Erosion Model (BSTEM) since 2002.
 12.9 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings
M Stoffel, University of Berne, Berne, Switzerland, and University of Geneva, Carouge, Switzerland
BH Luckman, University of Western Ontario, London, ON, Canada
DR Butler, Texas State University-San Marcos, San Marcos, TX, USA
M Bollschweiler, University of Berne, Berne, Switzerland, and University of Geneva, Carouge, Switzerland
r 2013 Elsevier Inc. All rights reserved.

12.9.1 Introduction 126

12.9.2 Tree Rings and Earth-Surface Processes 126
12.9.2.1 Basic Patterns of Tree Growth 126
12.9.2.2 How Do Earth-Surface Processes Affect Tree Growth? 126
12.9.2.2.1 Wounding of trees (scars) and resin-duct formation 127
12.9.2.2.2 Tilting of stems 128
12.9.2.2.3 Stem burial 128
12.9.2.2.4 Decapitation of trees and elimination of branches 129
12.9.2.2.5 Root exposure and damage 129
12.9.2.2.6 Elimination of neighboring trees 130
12.9.2.2.7 Colonization of landforms after surface-clearing disturbances 131
12.9.2.3 Sampling Design and Laboratory Analyses 131
12.9.2.3.1 Field approach and sampling design 131
12.9.2.3.2 Laboratory approach: sample preparation and analysis 133
12.9.3 What Earth-Surface Processes Have Been Analyzed with Tree Rings? 134
12.9.3.1 Surface Erosion 134
12.9.3.2 Hydrological Processes 134
12.9.3.3 Landslides 135
12.9.3.4 Snow Avalanches 135
12.9.3.5 Rockfalls 135
12.9.3.6 Earthquakes and Volcanic Activity 135
12.9.3.7 Permafrost Processes 136
12.9.3.8 Dendroglaciology 136
12.9.4 Research Perspectives: Looking to Future Developments 137
References 139

Glossary Gymnosperm Gymnosperms are a group of seed-bearing

Angiosperms Angiosperms are seed-producing plants, plants that includes conifers, cycads, Ginkgo, and Gnetales.
also known as flowering plants. The flowers, which are the Conifers (pines, cypresses, and relatives) form by far the
reproductive organs of flowering plants, are the most largest group of living gymnosperms.
remarkable feature distinguishing them from other seed Resin Most trees from temperate coniferous forests
plants. Flowers aid angiosperms by enabling a wider range produce resin as protectants and wood preservatives
of adaptability and broadening the ecological niches open following mechanical or biological disturbance. The resin is
to them. This has allowed flowering plants to largely produced by parenchymatous epithelial cells that surround
dominate terrestrial ecosystems. the resin ducts. Following mechanical disturbance, resin
Dendrogeomorphology The science that uses ducts are formed in tangentially aligned rows allowing for
dendrochronology (tree-ring dating) to study changes to the an accurate dating of past geomorphic process activity.
Earth’s surface over time.

Abstract

The initial employment of tree rings in geomorphic studies was simply as a dating tool and rarely exploited other
environmental information and records of damage contained within the tree. However, these unique, annually resolved,

Stoffel, M., Luckman, B.H., Butler, D.R., Bollschweiler, M., 2013.

Dendrogeomorphology: dating earth-surface processes with tree rings. In:
Shroder, J. (Editor in Chief), Butler, D.R., Hupp, C.R. (Eds.), Treatise on
Geomorphology. Academic Press, San Diego, CA, vol. 12,
Ecogeomorphology, pp. 125–144.

Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00326-2 125

126 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings

tree-ring records preserve valuable archives of past Earth-surface processes on timescales of decades to centuries. As many of
these processes are significant natural hazards, understanding their distribution, timing, and controls provides valuable
information that can assist in the prediction, mitigation, and defense against these hazards and their effects on society.
This chapter provides many illustrations of these themes, demonstrating the application of tree rings to studies of
snow avalanches, rockfalls, landslides, floods, earthquakes, wildfires, and several other processes. It illustrates the breadth
and diverse applications of contemporary dendrogeomorphology and underlines the growing potential to expand
dendrogeomorphic research, possibly leading to the establishment of a range of techniques and approaches that
may become standard practice in the analysis and understanding of Earth-surface processes and related natural hazards in
the future.

12.9.1 Introduction growth rings. Information about these patterns is derived from
increment cores or by full or partial cross sections of the stem
A major key to the assessment of ongoing hazard is the docu- (see discussion of sampling, below). In conifers (gymno-
mentation of Earth-surface processes and past natural disasters. sperms), reproductive cambium cells form large, thin-walled
In many cases, because of the absence of documentary records, earlywood tracheids during the early stages of the growing
this information must be developed from natural archives or season, which primarily serve the transport of nutrients and
‘silent witnesses’ (Aulitzky, 1992) that remain visible in the water. Later in the season, smaller and denser latewood trac-
landscape after an incident. In addition to the geomorphic or heids are developed. These layers are darker in appearance due
sedimentological evidence, key information is required on the to thicker cell walls and serve to increase the stability of the
dating and history of past events. The significant contribution tree. Tissue formation in broadleaved trees (angiosperms or
of tree rings to these endeavors lies in their capacity to both flowering plants) is more complex and diverse than in gym-
preserve evidence of past geomorphic activity and provide nosperms. In addition to the tracheids found in gymno-
critical information on their dating with annual or subannual sperms, the dividing cambium of broadleaved trees also
resolution. Therefore, in many climates, the tree-ring record produces vessels. Figure 1 illustrates the appearance of tree
may represent the most valuable and precise natural archive for rings in conifers (Figures 1(a) and 1(b)) and broadleaved
the reconstruction and understanding of past and ongoing trees (Figures 1(c) and 1(d)).
Earth-surface processes during the last several hundred years. The width and character of each tree ring are influenced by
The initial employment of tree rings in geomorphic studies biotic and abiotic factors. Biotic factors include the genetic
was simply as a dating tool: it rarely exploited other en- makeup as well as the aging of trees and are individual for
vironmental information that could be derived from studies of each species and each tree. Abiotic factors include a wide range
ring-width variations and records of growth anomalies con- of environmental variables, for example, light, temperature,
tained within the tree. However, these unique, annually re- water, nutrient supply, or influence of strong wind, which are
solved, tree-ring records preserve potentially valuable archives more or less common for all trees growing at a specific site
of past Earth-surface processes on timescales of decades to (Fritts, 1976). Therefore, trees growing at the same site can
centuries. As many geomorphic processes are significant nat- record the same environmental impacts and fluctuations (e.g.,
ural hazards, understanding their distribution, timing, and temperature or precipitation) in their tree-ring series. More
controls provides valuable information that can assist in the details on tree growth can be found in Cook and Kairiukstis
development of mitigation and defense against these hazards (1990) or Schweingruber (1996).
and their effects on society.
This chapter provides an overview of tree-ring-based re-
constructions of Earth-surface processes resulting from different
12.9.2.2 How Do Earth-Surface Processes Affect Tree
geomorphic agents. It includes a brief background to tree
Growth?
growth and dendrochronology and outlines the impact of these
events on tree morphology and wood anatomy. In addition, it Apart from the site-specific information common to many
clarifies the approaches used for tree-ring reconstruction of past trees at any site, individual trees also record the effects of
activity of Earth-surface processes. The chapter also illustrates mechanical disturbance caused by external processes. Trees
the breadth and diverse applications of contemporary den- can be injured, suffer breakage of their crown or branches,
drogeomorphology and underlines the growing potential to have their stems tilted, partially buried, or their roots exposed.
expand these studies, possibly leading to the establishment of a Evidence of these events can be recorded in the tree-ring series
range of techniques and approaches that may become standard of the tree affected (Figure 2). The analysis of geomorphic
practice in the analysis of specific hazards. processes through the study of growth anomalies in tree rings
is called ‘dendrogeomorphology’ (Alestalo, 1971). Den-
drogeomorphic research is normally based on the ‘pro-
cess–event–response’ concept as defined by Shroder (1978)
12.9.2 Tree Rings and Earth-Surface Processes
where the ‘process’ is represented by any geomorphic agent,
for example, debris flows, snow avalanches, or fires. Individual
12.9.2.1 Basic Patterns of Tree Growth
geomorphic ‘events’ that affect the tree may result in a range of
Dendrochronology depends on the fact that trees growing in growth ‘responses’. These ‘events’ and associated ‘responses’
areas with strong seasonal climates can form distinct annual are illustrated in the following paragraphs.
 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings 127

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

Figure 1 Microsections of tree rings prepared from conifer and broadleaved trees: In (a) Picea abies and (b) Pinus cembra, bands of tracheids
form the individual increment rings. In broadleaved trees, tracheids and vessels are formed by the dividing cambium. Depending on the
distribution of vessels in the ring, we distinguish between (c) ring-porous (Fraxinus excelsior) and (d) diffuse-porous angiosperms (Acer
pseudoplatanus). Reproduced from Stefanini, M.C., 2004. Spatio-temporal analysis of a complex landslide in the Northern Apennines (Italy) by
means of dendrochronology. Geomorphology 63, 191–202.

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

Figure 2 Injuries in Larix decidua: (a) injured stem, (b) cross section with overgrowth starting from the lateral edges of the injury, (c) callus
tissue as observed in the overgrowing cell layers bordering the injury, and (d) tangential row of traumatic resin ducts migrating from earlywood
toward later portions of the tree ring with increasing distance from the wound. Reproduced from Bollschweiler, M., 2007. Spatial and temporal
occurrence of past debris flows in the Valais Alps – results from tree-ring analysis. Geofocus 20, 1–182. with permission from Geofocus.

12.9.2.2.1 Wounding of trees (scars) and resin-duct Following injury, tangential rows of traumatic resin ducts
formation (TRDs) are produced in the developing secondary xylem of
Scratches or partial removal of the outer bark is a very com- certain conifer species, for example, Larix, Picea, Pseudotsuga, or
mon feature in trees affected by geomorphic processes Abies (Bannan, 1936; Stoffel, 2008; Butler et al., 2010;
(Lundström et al., 2008, 2009). Wounds can be observed on Figure 2(d)). They extend both tangentially and axially from the
the tree’s stem (Figure 2(a)), its branches, or roots. When injury (Bollschweiler et al., 2008b; Schneuwly et al., 2009a,
impact locally destroys the cambium, increment cell for- 2009b). When wounding occurs during the vegetation period of
mation is disrupted and new cell formation ceases in the in- the tree, resin production will start a few days after the event and
jured segment of the tree. To minimize rot and insect attacks ducts emerge within 3 weeks after the disturbing event (Ruel
after damage, the injured tree will (1) compartmentalize the et al., 1998; Luchi et al., 2005; Kaczka et al., 2010). Therefore,
wound (Shigo, 1984) and (2) almost immediately start pro- when analyzing cross sections, the intraseasonal position of the
duction of chaotic callus tissue at the edges of the injury first series of TRDs can be used to reconstruct previous events
(Figure 2(c)). Through the production of callus tissue, cam- with monthly precision (Stoffel et al., 2005b, 2008b; Stoffel and
bium cells will continuously overgrow the injury from its Hitz, 2008), provided the incidence occurred during the vege-
edges (Figure 2(b); Sachs, 1991; Larson, 1994) and ideally can tation period. With increasing axial and tangential distance from
lead to the complete closure of the wound. The extent of the impact, however, TRDs tend to migrate to later portions of
wound healing greatly depends on the annual increment rate, the tree ring (Bollschweiler et al., 2008b; Schneuwly et al.,
the age of the tree, and on the size of the scar. 2009a). Therefore, intraseasonal dating with monthly precision
128 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings

has to be based on cross sections or a large number of increment dating of the disturbance. In coniferous trees, compression
cores at the same elevation on the stem. This technique cannot wood (also known as reaction wood) will be produced on the
be used in pines as TRDs are not produced in this genus, which underside of the trunk. Individual rings will be considerably
produces copious amounts of resin and resin ducts unrelated to larger and slightly darker in appearance as compared to the
mechanical wounding (Phillips and Croteau, 1999; Ballesteros upslope side (Figure 3(b)). The difference in color results
et al., 2010). from much thicker and rounded cell walls of early- and late-
An injured tree will concentrate the formation of tree rings wood tracheids (Timell, 1986; Du and Yamamoto, 2007).
to those parts essential for survival and limit growth in other Multiple tilting events in the same stem may be recognized by
segments in the years succeeding the impact (Bollschweiler, changes in the amount, color, or orientation of reaction wood
2007). This may result in changes in ring width in other parts series in the ring series. By contrast, stem tilting in broadleaved
of the tree away from the wound (e.g., a growth reduction on trees leads to the formation of tension wood (Westing, 1965;
the opposite radius for a period of years). Schweingruber, 1996) on the upper side facing the tilting
agent. Broadleaved trees also react upon tilting with ultra-
12.9.2.2.2 Tilting of stems structural modifications (e.g., a gelatinous layer oriented
Inclination of the stem may result from the sudden pressure nearly parallel to the fiber axis) that are only visible when
induced by geomorphic processes directly, by the associated studied on microsections (Pilate et al., 2004).
deposition of material (e.g., avalanche snow and debris-flow In addition to the formation of different types of reaction
material) or by the slow but ongoing destabilization of a tree wood, trees may also respond with reduced growth after tilting
through landslide activity or erosion (Lundström et al., 2007, (Mayer et al., 2010). It is believed that such reductions in
2008). Tilted trees are common in many areas affected by annual tree-ring size are related to the destruction of roots
geomorphic processes (Figure 3(a)) and have, therefore, been resulting from abrupt or severe tilting. It is worthwhile to note
used in many dendrogeomorphic studies to date previous that the growth decrease will be normally less visible on the
events (e.g., Clague and Souther, 1982; Braam et al., 1987a, side where the reaction wood (i.e., compression or tension
1987b; Fantucci and Sorriso-Valvo, 1999). wood) is being formed. Figure 3(c) provides an example of
Subsequent growth in the trunk of a tilted tree will attempt the appearance of differing yearly increments in a Picea abies
to restore its vertical position and the reaction will be most tree tilted by a debris flow in 1922.
clearly visible in that segment of the tree to which the center of
gravity has been moved to through the inclination of the stem 12.9.2.2.3 Stem burial
axis (Mattheck, 1993). In the tree-ring record, eccentric growth Debris flows, floods, landslides, or ‘dirty’ snow avalanches
will be visible after a tilting event and, thus, allow accurate may bury trees by depositing material around their stem base

(b)
300
Impact
250
Ring width (1/100 mm)
c
200

150
1922
d 100

50

0
1860

1880

1900

1920

1940

1960

1980

2000

(a) (c)

Figure 3 Evidence of tilting (a) tree morphology and (b) cross sections of a tilted Larix decidua Mill. (D. M. Schneuwly). (c) Increment curves
of the upslope (dashed line) and downslope (solid line) of two radii from a Picea abies tree tilted by a debris flow in 1922 (Stoffel et al., 2005b).
Reproduced from Stoffel, M., 2005. Spatio-temporal analysis of rockfall activity into forests – results from tree-ring and tree analyses. Ph.D.
thesis, University of Fribourg. GeoFocus 12, 1–188, with permission from Geofocus.
 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings 129

(a) (b) (c)

Figure 4 Burial by sedimentation (a) sedimentation and subsequent die-off of trees after sedimentation. (b) Microsection showing an abrupt
growth decrease in Castanea sativa following an event. (c) Several levels of adventitious roots in Populus deltoides exposed by subsequent
erosion. Photo: F. H. Schweingruber.

(Figure 4(a)). Growth in these trees will normally be reduced branches. The loss of the crown or branches is more common
as the supply of water and nutrients will be temporarily dis- in bigger trees, when stems have lost their flexibility. Apex loss
rupted or at least limited (Figure 4(b); LaMarche, 1966; Hupp has also been observed as a result of rockfall impacts close to
et al., 1987; Friedman et al., 2005). Exceptionally, the burial of the ground level. In such cases, the sinusoidal propagation of
a stem can cause a growth increase if the material deposited is shockwaves in the stem results in the break-off of the crown.
rich in nutrients, the water supply guaranteed, and the depth This phenomenon has been described as whiplash or a ‘hula-
of the deposited material is not too great (Strunk, 1995). hoop’ effect (Dorren and Berger, 2006; Lundström et al.,
If stem burial exceeds a certain threshold, trees will die 2009).
from a shortage in water and nutrient supply (Figure 4(a)). Trees react upon decapitation or branch (foliage) loss with
According to case study results from the Italian Dolomites distinct radial growth suppression in the years following the
(Strunk, 1991), P. abies may tolerate a maximum burial depth impact (see Figure 5(b)). One or several lateral branches will
of 1.6–1.9 m in environments dominated by fine-grained form a ‘leader’ that replaces the broken crown, resulting in the
debris flows composed of calcareous and dolomitic material tree morphology called ‘candelabra’ growth (Figure 5(b);
(Strunk, 1997). Although no data are available for other spe- Butler and Malanson, 1985; Stoffel et al., 2005c). In addition,
cies or lithologies, it is believed that survivable burial depths it is not unusual that the shock of the impacting material
will be much smaller in regions where debris flows are com- causes injuries and provokes the formation of TRDs. ‘Leaders’
posed of massive or larger materials. may also be formed from prostrated trunks knocked over by
Occasionally, buried trees produce adventitious roots close mass-wasting events (Figure 5(c)).
to the new ground surface (Figure 4(c); Bannan, 1941). As
adventitious roots are normally formed in the first years suc-
ceeding burial (Strunk, 1995), the moment of root sprouting 12.9.2.2.5 Root exposure and damage
can be used to derive an approximate date for the sedimen- Erosional processes and the (partial) denudation of roots may
tation processes, as shown by, for example, Strunk (1989, generate different growth reactions, both in the stem and in
1991) or Marin and Filion (1992). When a tree has been re- the exposed roots. The type and intensity of the reaction(s)
peatedly buried and formed several layers of adventitious will depend on the nature of the erosive event, which may be
roots, it is possible to estimate the thickness of sedimentation instantaneous or progressive and gradual. If several roots are
from individual events at the location of the tree (Strunk, completely denuded during a sudden erosive event (e.g.,
1997). debris flow, lahar, flood, or landslide), they will no longer able
to fulfill their primary functions and quickly die. The tree
subsequently suffers from a shortage of water and nutrient
12.9.2.2.4 Decapitation of trees and elimination of supply, resulting in suppressed tree growth and the formation
branches of narrow rings in the stem (see Figure 4(b); LaMarche, 1968;
Bouncing rocks and boulders, debris in flowing water, debris Carrara and Carroll, 1979; McAuliffe et al., 2006).
flows and lahars, or the windblast of snow avalanches may In cases where only part of a root is exposed (Figure 6(a))
cause decapitation of trees (Figure 5(a)) or the removal of and its outer end remains in the ground, the root will
130 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings

(a) (b) (c)

Figure 5 Breakage and stem removal (a) Picea abies topped by rockfall. (b) Candelabra growth in Larix decidua following apex loss. (c) Picea
engelmannii with ‘two leaders’ developed from a prostrated stem knocked to the ground by a snow avalanche.

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

Figure 6 Root damage: (a) exposed roots of Pinus sylvestris, (b) larger increment rings with distinct latewood are formed in Abies alba after
sudden exposure (dashed red line), (c) change in the arrangement of vessels in Fraxinus excelsior from diffuse-porous to ring-porous following
sudden exposure, (d) in addition to cell changes, tension wood formed in a root of Acer pseudoplatanus in addition to cell changes. Reproduced
from Hitz, O.M., Gärtner, H., Heinrich, I., Monbaron, M., 2008. First time application of ash (Fraxinus excelsior L.) roots to determine erosion
rates in mountain torrents. Catena 72, 248–258.

continue to grow and fulfill its functions. In the exposed root, 1999; Vittoz et al., 2001). As a result of root damage, tree-ring
however, anatomical changes will occur and individual growth growth will be suppressed or eventually cease (Figure 7).
rings similar to those in the stem or branches will be formed. Previous studies using ring-width series to determine landslide
The localization of this change in the tree-ring series may and earthquake activity include Meisling and Sieh (1980), Lin
allow determination of the moment of exposure (Figures and Lin (1998), Carrara and O’Neill (2003), or Papadopoulos
6(b)–6(d); Bodoque et al., 2005; Hitz et al., 2008). The et al. (2007). Rizzo and Harrington (1988) showed that per-
continuous slow exposure of roots is usually caused by gradual iods of decreased growth of P. rubens and Abies balsamea trees
processes and relatively low denudation rates, for example, by from the Appalachian Mountains were significantly correlated
overland flow, the slow opening of cracks in soils (e.g., soil with wind exposure and related root and crown damage
creep and landslides) and in disintegrated bedrock, along variables.
rivers, streams, lakes, and oceans (floods, shore erosion), as
well as with faulting activity and displacements in relation to 12.9.2.2.6 Elimination of neighboring trees
earthquake activity. Provided the roots are gradually exposed Geomorphic processes can eliminate trees along channels or
with time, it is also possible to determine the erosion rates in couloirs through uprooting and stem breakage and leave
such cases (Carrara and Carroll, 1979). neighboring trees intact (Figure 8(a)). This phenomenon can
Root shearing and root damage frequently occur in areas be observed with, for example, rockfalls, debris flows, lahars,
affected by landsliding or along earthquake faults (Allen et al., extreme floods, landslides, or snow avalanches (Butler, 1979,
 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings 131

1955−60 1979−84
1950 1970 1990

2.5

CSL01A
CSL01B B
2
CSL01C
A
C
Ring width (mm)

1.5

0.5

0
1820 1840 1860 1880 1900 1920 1940 1960 1980
Year
Figure 7 Radial growth suppression caused by landsliding. This tree was growing on a slope destabilized by downwasting of Columbia Glacier,
Alberta. The abrupt radial suppressions in 1887–94, 1955–60, and 1979–84 are thought to result from root damage as the block moved
downslope 3 times. The tree was cut in 1996. The enlargement shows part of radius A (Luckman, unpublished data).

1985; Stoffel et al., 2005c). The uninjured survivor trees have minimum time elapsed since the last devastating event in
less competition, more light, nutrients, and/or water. As a snow avalanche couloirs, debris-flow channels, or floodplains
consequence, they may start to produce larger increment rings. (Sigafoos and Hendricks, 1969; McCarthy and Luckman,
However, observations indicate that this growth release in 1993; Winter et al., 2002; Pierson, 2007; Bollschweiler et al.,
survivor trees can be delayed and, therefore, this reaction 2008a). It involves estimating the time between the exposure
cannot always be used to date past destructive events with of the surface and the germination of the first surviving
yearly precision (Figure 8(b)). Nevertheless, the growth re- seedling on that surface. This ‘ecesis’ estimate varies with the
lease in survivor trees can corroborate the dating of events that environment, substrate, available seed sources, and several
have been identified from evidence in other trees at the same other factors. Problems of ecesis determination have been
site, for example, from scars and tilted stems (Stoffel et al., extensively discussed in studies that attempt to date the for-
2005a). mation of glacier moraines (e.g., McCarthy and Luckman,
1993; Koch, 2009) where ecesis estimates may vary from a few
years to several decades.
12.9.2.2.7 Colonization of landforms after surface-
clearing disturbances
Many different Earth-surface processes can eliminate surface
12.9.2.3 Sampling Design and Laboratory Analyses
vegetation including entire forest stands, leaving no direct
dendrogeomorphic evidence. In such cases, germination ages 12.9.2.3.1 Field approach and sampling design
of trees growing on these bare surfaces can be used to estimate The types of damage to trees described in the preceding sec-
the time of creation of new landforms and the disturbance tion may result from a wide variety of Earth-surface processes.
event. This approach provides a minimum age for that surface Linkage between the damage and the causative process de-
and has frequently been used to date landforms or assess the pends on a critical evaluation of the sampling site. In many
132 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings

450
400

Ring width (1/100 mm)

c
350
300
250
d 200
1890
150
100
50
0

1800

1820

1840

1860

1880

1900

1920

1940

1960

1980

2000
(a) (b)

Figure 8 (a) Elimination of forest cover by a rockslide leaving uninjured survivor trees at both sides of the scar. (b) Increment curves of two radii
from an intact surviving, Larix decidua after a major debris flow in 1890 (Stoffel et al., 2005b). Reproduced from Stoffel, M., 2005. Spatio-temporal
analysis of rockfall activity into forests – results from tree-ring and tree analyses. Ph.D. thesis, University of Fribourg. GeoFocus 12, 1–188, with
permission from Geofocus.

cases, investigations are undertaken with a specific context The nature of any visible growth defect observed in the tree’s
where the process or processes involved are known and ap- morphology will strongly influence the sampling height, sam-
propriate sampling methods and sampling design may be pling directions, and the minimum number of samples to be
applied. These vary considerably with the nature of the process taken per tree. In trees with visible ‘scars’, previous geomorphic
or investigation and are best discussed in the specific context events are most easily dated through the destructive sampling of
of individual processes, for example, the sampling network trees and the preparation of cross sections taken at the location
needed to define past earthquake activity may be very different where the injury is largest. This approach will facilitate an ac-
from that required to reconstruct the history of snow ava- curate and intraseasonal identification of the onset of ‘callus
lanches and debris flows on a large debris cone. tissue’ production (and TRD formation in certain conifer spe-
At an individual site, the choice of sampling design and cies) and, therefore, allows a reconstruction of the impacting
selection of the trees to be sampled will depend on the pur- event with very high temporal resolution. Alternatively, wedges
pose of the investigation and the processes or hazards being can be sawn from the overgrowing callus and an increment core
sampled. The sampled area and number of trees sampled may extracted from the side opposite to the wound (Arbellay et al.,
be very small and specific (e.g., on a small landslide) or may 2010). In this case, the sampled tree will survive and a re-
involve systematic sampling of a larger area (e.g., within a construction of the wounding event will be possible. When
large run-out zone of a snow avalanche) or target individual cross sections and wedges cannot be taken from the injured
trees damaged at the margins of an event. The choice of the trees, at least two increment cores need to be extracted to
individual trees sampled is based (1) on an inspection of the sample both the overgrowing callus and the side opposite to the
stem (i.e., does the tree show obvious signs of past disturb- wound. Special attention needs to be given to sampling of cores
ance) and (2) on the location of the trees with relation to the from the overgrowing callus, and Figure 9 illustrates the rec-
processes/hazards studied (e.g., whether the tree located on or ommended position for the extraction of increment cores in
adjacent to the area is influenced by the process studied, such injured trees. In addition and in the case of certain conifer
as a debris-flow levee). A detailed description on the docu- species, TRD formation will be delayed with increasing distance
mentation and numbering of trees in the field is provided by from the wound and improper core location may influence the
Stoffel (2005). intraseasonal dating quality (Bollschweiler et al., 2008b;
The tree-ring record of growth disturbances created by past Schneuwly et al., 2009a, 2009b).
events is analyzed with cross sections, wedges, or increment ‘Tilted stems’ are usually analyzed with at least two incre-
cores. Cross sections or wedges are normally taken at the lo- ment cores extracted per tree, one in the direction of the tilting
cation of the growth disturbance and provide an excellent and and the other on the opposite side of the trunk. The reaction
very complete insight into the tree’s history. However, in many wood will be visible on the tilted side in conifer trees
locations (e.g., protection forests and National Parks) felling ( ¼ compression wood) and on the side opposite to the tilting
may be prohibited or ill-advised for esthetic or safety reasons. direction in broadleaved trees ( ¼ tension wood). Individual
Most tree-ring studies, therefore, use cores extracted with an cores are best extracted at the location where tilting is strongest
increment borer. Grissino-Mayer (2003) provided a technical based on an outer inspection of the tree morphology. How-
description of the correct use of increment borers. In some cases, ever, in situations where multiple tilting events may vary in
useful information may also be obtained from the analysis of direction, for example, in snow avalanche tracks, additional
cross sections sampled from tree stumps remaining on the cores, transverse to the downslope direction, may be needed to
slopes after logging (Swetnam, 1993; Stoffel and Perret, 2006). develop a more complete record.
 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings 133

a correction can be undertaken using a transparent sheet with

b
concentric rings (Bosch and Gutierrez, 1999; Gutsell and
Johnson, 2002). An age correction also needs to be applied for
the assessment of the colonization time or ecesis (see
McCarthy and Luckman, 1993; Pierson, 2007; Koch, 2009).
In some cases, a reference tree-ring chronology may need
c to be developed from a nearby forest stand to assist in cross-
dating of trees at the sampled site to verify the age assignment
of rings. The reference site should be undisturbed and the trees
sampled should not show obvious signs of external injury,
disease, or anomalous tree-ring growth to minimize noncli-
matic influences. Usually, the oldest trees are selected to
maximize the length of the tree-ring chronology. In contrast to
the disturbed trees, two increment cores are extracted per-
pendicular to the slope and at breast height (E130 cm). Ad-
equate sample size varies between species and also depends on
d the strength of the common signal. In general, 20–30 trees are
Figure 9 When coring injured trees, special attention needs to be a useful guide for most species to minimize the potential in-
given to the sampling position. Samples taken inside the wound ‘a’ or fluence of geomorphic processes and hidden growth
from the overgrowing callus tissue ‘b’ will provide an incomplete disturbances.
tree-ring record, as wounds are closed from their edges. Ideally,
increment cores are extracted just next to the injury ‘c’ where the 12.9.2.3.2 Laboratory approach: sample preparation and
presence of overgrowing callus tissue and TRD will allow accurate analysis
dating. Cores taken too far away from the wound ‘d’ will not
In the laboratory, samples are normally analyzed and data
necessarily show signs of the disturbing event and thus prevent
dating. A combination of ‘a’ and ‘d’ may allow approximate dating of
processed following the standard procedures described in
the scar if no rings have been removed from the impacted surface. Bräker (2002) or Stokes and Smiley (1968). Single steps of
sample analysis generally include surface preparation, count-
ing of tree rings, skeleton plots (Schweingruber et al., 1990),
In the case of ‘buried stem bases’, the sampling of two as well as ring-width measurements using a digital positioning
increment cores in opposite directions (ideally from the table connected to a stereo microscope and a time-series an-
upslope and the downslope side) will normally allow accurate alysis program. Cross-dating and, where required, the devel-
reconstruction of the event that has led to the sedimentation opment of a reference chronology follows standard
of material around the stem’s base. It is best to sample these procedures of measurement, cross-dating, and standardization
trees as close to the ground level as possible (c. 20–40 cm) to (e.g., Fritts, 1976; Cook, 1987; Cook and Kairiukstis, 1990;
extract a maximum number of tree rings and obtain maximum Grissino-Mayer, 2001; Vaganov et al., 2006).
information. However, the influence of roots should be ‘Ring widths of the disturbed samples’ may also be meas-
avoided. ured and the series compared graphically or statistically with
The analysis of ‘exposed roots’ is normally based on cross the reference chronology. These tools also allow for an as-
sections, as they regularly show partially or completely miss- sessment of cross-dating accuracy between ring-width series of
ing rings. The position of samples needs to be carefully noted individual disturbed trees and the reference chronology. Once
with respect to the ground surface or the noneroded parts of all tree-ring series are checked and missing rings identified,
the root. These samples are essential for the understanding individual growth curves may be analyzed visually to identify
and interpretation of continuous erosion processes and the the tree’s reactions to geomorphic processes, for example, the
determination of denudation rates (LaMarche, 1968; Bodoque initiation of abrupt growth reduction or releases (McAuliffe
et al., 2005). et al., 2006). In the case of tilted stems, the growth curve data
The ‘germination of trees’ on new landforms is best de- may be analyzed to define the dating of the tilt (Braam et al.,
termined by counting the annual growth rings in increment 1987a, 1987b; Fantucci and Sorriso-Valvo, 1999; Figure 3(c)).
cores taken from the root crown level. However, branches, In addition, the microscopic appearance of the cells (e.g.,
obstacles, and rot may sometimes require sampling positions structure of the reaction wood cells) may be investigated to
higher up on the stem. In these cases, an age-correction factor yield further evidence of disturbance (Wilson and Archer,
needs to be added to compensate for the time a tree takes to 1977). Other features, for example, callus tissue overgrowing
grow to sampling height (McCarthy et al., 1991). A simple scars or the presence of TRD formed following cambium
height–age correction can be achieved by dividing the tree damage, can only be identified through a visual inspection of
height by the number of tree rings to obtain an average rate for the cores and cross sections (e.g., Stoffel et al., 2005c; Perret
the yearly apical increment. Thereafter, the sampling height is et al., 2006; Bollschweiler et al., 2008b). Finally, the first
divided by this yearly increment so as to obtain an estimate of decade of juvenile growth should be excluded from analyses,
the number of rings between sampling height and the root as tree rings in seedlings tend to be more susceptible to any
crown (e.g., McCarthy et al., 1991; McCarthy and Luckman, kind of disturbance than those in larger trees.
1993; Bollschweiler et al., 2008a). The number of rings to pith All growth reactions identified in the samples are noted in
has to be estimated if the pith is not present on the core. This order to identify disturbance events. Except for processes with
134 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings

limited volumes (e.g., single rocks or boulders involved in Laurence Estuary in Québec (Bégin et al., 1991) and at Lago
rockfall events; Stoffel et al., 2005c, 2006b; Moya et al., 2010), Bolsena in Central Italy (Fantucci, 2007).
one growth disturbance identified in a single tree is not con-
sidered sufficient evidence to identify an event and, therefore,
the identification of events needs to be based on quantitative
12.9.3.2 Hydrological Processes
(i.e., indices; Butler et al., 1987) or semiquantitative (Stoffel
and Bollschweiler, 2008, 2009) thresholds. Some of the earliest studies of Earth-surface processes based on
growth rings extracted from stems focused on floods. The the-
oretical background for the study of flood and sedimentation
processes was provided by Sigafoos (1964). Helley and
12.9.3 What Earth-Surface Processes Have Been LaMarche (1968) applied Sigafoos’ techniques to determine the
Analyzed with Tree Rings? recurrence interval of flood events in Northern California. Later
studies in the United States focused on floods in the White
Dendrogeomorphology has been used in the analysis and River (Arkansas; Bedinger, 1971) or high summer flows and
reconstruction of a wide range of Earth-surface processes. This floods in the Potomac River (Yanosky, 1982, 1983, 1984). St.
chapter does not claim to be exhaustive, but references se- George and Nielson (2000, 2003) analyzed anatomical changes
lected ‘classic studies’ and significant follow-up papers in the (‘flood rings’) in Quercus macrocarpa to identify nineteenth-
fields of Earth-surface processes. century high-magnitude floods in the Red River (Canada). In an
early study, Parker and Josza (1973) used dendrogeomorphic
techniques to study fluvially triggered landslides and ice-jam
flooding along the Mackenzie and Liard Rivers in Canada. More
12.9.3.1 Surface Erosion
recently, Tardif and Bergeron (1997) also reconstructed ‘ice-jam
Some of the earliest applications of dendrochronological re- flood events’ based on a dendrochronological analysis of ice-
search to Earth-surface processes utilized roots rather than scarred Thuja occidentalis growing on the shore of Lake Dupar-
stem samples. In the 1960s, LaMarche (1961, 1968) used tree quet (Canada). In Europe, only a few studies have focused on
roots exposed by surface lowering as an indicator of ground floods so far, for example, Astrade and Bégin (1997) docu-
surface at the time of germination of 5000-year-old Pinus mented recent ‘spring floods’ in the Saône River (France) using
aristata in the White Mountains, California. More recently, rings of Populus tremula and Q. robur.
McAuliffe et al. (2006) and Scuderi et al. (2008) documented In contrast to erosion and floods, abundant studies of debris
four centuries of erosion and sedimentation history in a small flows have been carried out over the last 20 years. After the
drainage basin in the Colorado Plateau of northeastern Ari- pioneering studies of Hupp (1984; Hupp et al., 1987) con-
zona. Their results demonstrate that the most recent episode ducted on the slopes of Mount Shasta (California), tree-ring
of growth anomalies was associated with the largest precipi- records were used extensively by Strunk (1989, 1997) to re-
tation shift (drought to wet interval) in the last 400 years. construct debris-flow activity in the Italian Dolomites. Strunk
In Europe, the assessment of erosion rates based on in- also investigated the germination of adventitious roots in bur-
crement-ring records only started in the past few years. On the ied stems to reconstruct burial depths and the history of debris
Iberian Peninsula, rates of ‘sheet erosion’ were determined flows. More recently, the frequency and spatial patterns (i.e.,
with P. sylvestris (Bodoque et al., 2005; Rubiales et al., 2008) lateral spread and extent of flows) have been analyzed for more
along a popular walking trail. Corona et al. (in press, 2011) than 30 torrents in the Swiss Alps (e.g., Baumann and Kaiser,
focused on areal denudation in marly badlands of the Alpes 1999; Bollschweiler et al., 2007, 2008a; Bollschweiler and
de Haute-Provence and illustrated limitations and possible Stoffel, 2007, 2010; Stoffel et al., 2008a, 2008b). The largest
caveats of past approaches using growth rings of roots to re- tree-ring-based study was performed for the Ritigraben torrent
construct erosion rates. The authors calibrated reconstructed (Valais, Swiss Alps), where the analysis of 2450 increment cores
erosion rates with highly resolved, long-term measurements of of Larix decidua, P. abies, and P. cembra yielded spatiotemporal
denudation on slopes and sediment yield in traps at the data on 124 debris-flow events since AD 1570 (Stoffel et al.,
basins’ outlets. Stoffel et al. (2012) reconstructed sporadic but 2008b), dated with monthly precision (Stoffel and Beniston,
intense ‘gullying’ in the Patagonian Andes of Argentina using 2006). Based on the large amount of data, it was possible for
root-ring series of Southern Hemisphere species (Austrocedrus the first time to realize magnitude–frequency relationships for
chilensis, Nothofagus dombeyi, and Pseudotsuga menziesii) and debris flows (Stoffel, 2010) and quantify precipitation totals
applying new tree-ring parameters for the assessment of ero- and storm types involved in past events (Stoffel et al., 2011).
sive pulses and the localization of erosion ‘hot spots’ in In the mountains of central Spain (Sistema Central) Sys-
ephemeral gullies. All of the aforementioned studies used tem, Ballesteros et al. (2010, 2011) and Ruiz et al. (2010) used
coniferous species for the quantification of erosion rates. increment series and stem wedges of P. pinaster as well as Alnus
Roots from broadleaved species were only used exceptionally, glutinosa, F. angustifolia, and Q. pyrenaica to date ‘flash floods’
although Hitz et al. (2008) have recently demonstrated the and to distinguish higher magnitude–lower frequency from
potential of Fraxinus excelsior for the identification of sudden lower frequency–higher magnitude events for the last 50 years.
or continuous erosion processes. Finally, LaMarche (1966) The first dendrogeomorphic study of ‘debris floods’ was re-
was again the first to work on ‘shore erosion’ processes and cently completed in the Austrian Alps, allowing for the dating
reconstructed 800 years of geomorphic activity. More recent and analysis of the extent of 37 events based on growth
studies have been carried out on shores of the Upper St. anomalies in P. abies since AD 1800 (Mayer et al., 2010).
 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings 135

12.9.3.3 Landslides dating of large ‘rock avalanches’ (a type of landslide – e.g.,

Moore and Mathews, 1978; Butler et al., 1986) rather than on
Following early studies by Parker and Josza (1973), tree-ring-
the reconstruction of the much more frequent release of indi-
based analysis of landslides include case studies from Northern
vidual rocks and boulders during much smaller rockfalls.
Québec, Canada (Bégin and Filion, 1985, 1988; Filion et al.,
Lafortune et al. (1997) were probably the first to focus on
1991) that document nineteenth-century landslides along the
rockfalls, but with the aim of assessing sedimentation rates and
Great Whale River using buried trunks in flowing sediments or
forest-edge dynamics. The first studies analyzing the season-
tilted P. glauca trees. In the Barcelonnette region (French Alps),
ality, frequency, and spatial patterns of rockfalls using tree rings
Braam et al. (1987a, 1987b) assessed recent landslide activity
were only very recently realized by Stoffel et al. (2005b, 2005c),
from eccentricity variations in L. decidua trees. More recently,
yielding data on 400 years of rockfall activity on a forested
Corominas and Moya (1999) have analyzed landslides in
slope in the Swiss Alps. The methods have since been applied
the basin of the Llobregat River (Pyrenees) and identified dif-
to other sites in Switzerland by Perret et al. (2006) and
ferent rainfall patterns triggering landslides. In Italy, den-
Schneuwly and Stoffel (2008a, 2008b), providing a reliable
drogeomorphology has been used repeatedly for the
basis for the analysis of rockfall risks and the accuracy assess-
identification of landslide dynamics in Calabria or the Apen-
ment of rockfall models (Stoffel et al., 2006b). Based on Stoffel
nines based on growth series of Q. cerris and Q. pubescens
and Perret’s (2006) documentation of the healing of wounds,
(Fantucci and McCord, 1995; Fantucci and Sorriso-Valvo, 1999;
Moya et al. (2010) presented different strategies of tree sam-
Stefanini, 2004) and results have been compared with me-
pling for the reconstruction of complete rockfall chronologies
teorological data. In the first study in the subtropics, Paolini
from visible injuries in Q. robur from Solà d’Andorra (Eastern
et al. (2005) used suppression-release patterns and inception of
Pyrenees). They show that rockfall frequency cannot be as-
A. acuminate trees to date a number of landslides in Argentina.
sessed directly from a chronology of visible tree injuries, as it
also requires taking account of the resolution of the dating and
12.9.3.4 Snow Avalanches the magnitude of the rockfall events.

The occurrence of snow avalanches has quite frequently been

analyzed with tree rings over the last four decades, beginning
with Potter’s (1969) study and Schaerer’s (1972) publication
on vegetation in avalanche terrain at Rogers Pass in British 12.9.3.6 Earthquakes and Volcanic Activity
Columbia (Canada). For a long time, tree-ring-based analysis Tree rings have been extensively used to study century-old
of snow avalanche activity was almost exclusively used in earthquakes along the San Andreas Fault in California or in
North America, especially in the Rocky Mountains of Color- the Seattle region of the USA (Jacoby et al., 1988, 1992;
ado (Ives et al., 1976; Carrara, 1979) and in Glacier National Jacoby, 1997). More recently, Carrara and O’Neill (2003) and
Park, where Butler (1979, 1985) published a series of papers Bekker (2004) have analyzed spatial differences in the re-
on tree-ring-based records of snow avalanches and in the sponse of A. lasiocarpa, P. contorta, P. flexilis, and P. menziesii to
Canadian Rockies (Frazer, 1986; McCarthy, 1985; Luckman normal faulting caused by the Hebgen Lake earthquake.
and Frazer, 2001). More recently, and in the light of risk as- Hamilton (2010) sampled P. contorta, P. menziesii, and Tsuga
sessment, attention has turned toward the identification of heterophilla trees 6–24 m above ground level from Olympic
more ‘extreme events’ and the dating of high-magnitude snow and Yellowstone National Parks. Stem whiplash produced by
avalanches on the Gaspé Peninsula, Quebec (Boucher et al., ground motion left obvious macrofeatures in the form of
2003; Dubé et al., 2004; Germain et al., 2009) and in Glacier healed fracture and ‘chevron’ structures in these trees and
National Park (Butler and Sawyer, 2008). Only a few studies allowed for a dating of the AD 1700 Cascadia or the AD 1959
have been published on snow avalanche activity in the Euro- Hebgen lake earthquakes.
pean Alps. Stoffel et al. (2006a) studied a cone affected by Since the catastrophic explosion of Mount St. Helens in
snow avalanches descending from three different couloirs and 1980, tree rings have also been used for the analysis and cal-
were able to differentiate the damage induced by the wind- endar dating of volcanic eruptions. For example, Yamaguchi
blast from that induced by snow and transported material. At (1983, 1985) calendar-dated the explosive eruption of Mount
the same time, dendrogeomorphic analyses of snow ava- St. Helens in 1482. Biondi et al. (2003) identified clear signs
lanches have been initiated in the Spanish Pyrenees (Muntán of the AD 1913 Volcán de Fuego de Colima (Mexico) eruption
et al., 2004, 2009), the French Alps (Corona et al., 2010) and in tree-ring series of P. hartwegii, reacting to the deposition of
in Patagonia (Mundo et al., 2007; Casteller et al., 2008). tephra with a sudden growth decrease. Based on the analysis
Larocque et al. (2001) used impact scars, TRDs, and reaction of frost rings and other dendrochronological evidence,
wood to analyze frequency–magnitude relationships of D’Arrigo et al. (2001) dated millennia-old volcanic eruptions
‘slushflows’ (i.e., liquefied snow) on the Gaspé Peninsula. with Mongolian and Northern Siberian trees, whereas Salzer
and Hughes (2007) managed to identify 5000 years of vol-
canic eruptions through the study of P. aristata trees. Briffa
12.9.3.5 Rockfalls
et al. (1992, 2001) have used a tree-ring network of Northern
Surprisingly, and despite the potential of dendrogeomorphic Hemisphere treeline sites to identify major global eruptions
methods, rockfall activity has only rarely been studied through over the last millennium using maximum density data. Most
the analysis of tree-ring sequences (Stoffel, 2006). First studies recently, Sheppard et al. (2010) used dendrochemical evidence
of rock–tree interactions focused on the identification and in P. engelmannii, P. menziesii, and T. plicata trees to date the
136 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings

AD 1781 eruption of Mount Hood, a stratovolcano of the application of tree-ring data to study the past glacial history.
Cascade Volcanic Arc. However, minimum tree-ring counts from trees growing on
Solomina et al. (2008) have dated a seventeenth-century glacier moraines have been used intermittently throughout the
eruption of Sheveluch volcano (Kamchatka, Russia) through twentieth century, beginning with Tarr and Martin (1914), and
tree-ring dating of logs that were conserved in the deposits of a were probably the first dendrogeomorphic applications. Recent
‘pyroclastic flow’. The first dating of volcanic mass movements comprehensive discussions of dendroglaciology occur in
in survivor trees was performed in Mexico, where Bollsch- Luckman (1998), Smith and Lewis (2007), and Koch (2009).
weiler et al. (2010) used increment cores of A. religiosa, P. The first, simplest, applications involve determining the in-
ayacahuite, and P. hartwegii to date twentieth-century lahars at ception date of the oldest tree growing on a moraine or other
Volcán Popocatépetl. Similar techniques are used by Pringle surface to provide a minimum age for the feature or, more
et al. (2010) to date a major lahar from the AD 1781 Old Maid rarely, estimates of the rates of glacier recession or succession
eruption (Mount Hood, Oregon) and possibly in 1801 based (e.g., Cooper, 1916). Ideally, cores or, more rarely, the cross
on survivor P. menziesii and T. plicata trees. sections used should contain pith and be as close to the root
crown as possible. However, in many cases, a correction factor
needs to be added to the earliest ring sampled to account for
12.9.3.7 Permafrost Processes
pith correction and sampling height, plus an estimate of ecesis
Dendrogeomorphic research has only occasionally focused on at the site (see discussion above). Unless there is evidence of
‘periglacial processes’. Zoltai (1975) and Jakob (1995) ana- significant deadfall, trees must be assumed to be first growth
lyzed movement rates of gelifluction lobes, and there were but finding the oldest tree is not easy. Tree age is rarely strongly
early studies of the development of ‘thermokarst lakes’ in the related to size and the oldest trees may have suppressed early
Yukon, Canada (Burn and Smith, 1988; Burn and Friele, growth near the pith and be overtaken in growth by younger,
1989). More recently, Agafonov et al. (2004) documented the more virile, individuals. The oldest trees are more easily found
rise and evolution of thermokarst that was primarily inter- in sparsely vegetated sites, but these sites probably have longer
preted as resulting from increasing precipitation rather than ecesis intervals than more forested sites where, conversely,
increasing air temperature. The movement of ‘rock glaciers’ more trees must be sampled to find the oldest individual. In-
and related features was assessed initially by Shroder’s (1978) evitably, however, these dates have error estimates of one or
pioneer study of a glacier-like boulder deposit on the Table more decades because of ecesis variability and should
Cliffs Plateau (Utah). He documented movement over a 200- be treated as minimum dates. More precise, annually re-
year period and suggested that precipitation was possibly the solved, estimates can be derived from trees that
trigger for the main episodes of movements. Other studies on were damaged or killed at the glacier margin or during moraine
movements in permafrost complexes have subsequently been building (Figure 10(a)), but only usually on the outermost
carried out elsewhere in North America (Giardino et al., 1984; moraine (e.g., Luckman, 1988, 2006). When dead, such trees
Carter et al., 1999; Cannone and Gerdol, 2003; Bachrach must be cross-dated against a suitable reference chronology.
et al., 2004), although not in other alpine regions. In many areas around the world, the glacier advances of the
last few centuries were more extensive than earlier Holocene
events and significant dendroglaciological work has been done
12.9.3.8 Dendroglaciology
using recently exposed stumps or logs in glacier forefields to
The term ‘Dendroglaziologie’ was coined by Schweingruber estimate the dates of earlier events or ice-free periods, pro-
(1993) and has subsequently been used to describe the viding the tree-ring records can be cross-dated against

(a) (b)

Figure 10 Dendroglaciological examples (a) Nothofagus sp. tilted and killed during emplacement of the lateral moraine of Glaciar Seco, Santa
Cruz, Argentina and (b) standing snags in a former ice-marginal Lake, Brady Glacier, Glacier Bay, Alaska. The trees died between 1830 and 1839
as the lake filled. (a) Photo: Ricardo Villalba. (b) Denny McLane Capps, photo and personal communication.
 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings 137

reference chronologies. In such studies, an important dis- replication decreases in the earlier parts of the chronologies.
tinction must be made between in situ (i.e., rooted) stumps Techniques have been developed in dendroclimatology (e.g.,
and detrital wood that has been transported by the glacier. Fritts, 1976; Briffa and Jones, 1992; Vaganov et al., 2006) to
Calendar-cross-dated ring sequences from both types of sam- assess signal strength and the fidelity of the common signal as
ples can provide a minimum date for the glacier advance that sample replication decreases back in time. In the building of
killed the trees and the minimum duration of ice-free con- event chronologies of recurring Earth-surface processes, it is
ditions immediately prior to that glacier advance (e.g., Holz- important to develop similar techniques and criteria that can
hauser, 1984). However, only kill dates from in situ materials be applied to assess the quality of these event chronologies
provide the precise location of the glacier when the tree was and address four major problems.
killed, detrital material having been killed some unknown First, different processes can produce disturbance in tree-
distance up-glacier cannot yield such data. Although dating ring series, which is why there are so many potential appli-
the outermost rings provides calendar-dating precision, the cations of these data. However, in specific studies, the problem
dates remain minimum ages for the glacier event because, is to identify and isolate the signal related to the process under
except at the glacier margin, there is no indication of the study from other potential sources of disturbance. A detailed
position and timing of tree death with regard to the duration identification of processes and sources of disturbances,
or extent of the glacier advance. The presence of bark and good therefore, represents the initial and probably most crucial
radial preservation are good indicators that the outermost point in any dendrogeomorphic study.
rings are present: often sapwood weathers easily from these Second, the establishment of criteria to distinguish be-
trees when exposed. In some cases, multiple in situ dates can tween evidence of the events/process of interest (i.e., the ‘sig-
indicate the rate of glacier advance from a mapped pattern of nal’) from other causes (‘noise’). One significant approach in
kill dates (e.g., Luckman, 1995). Several studies (e.g., Luckman this respect is to develop measures such as the ‘tree-ring
and Villalba, 2000) have compared tree-ring-derived moraine response index’ suggested by Butler et al. (1987) or Ruiz
dates with dendroclimatic estimates of temperatures or pre- et al. (2010). Both approaches attempt to discern acceptable
cipitation to illustrate the linkage between climate and glacier minimum sample sizes and minimum numbers of trees
fluctuations. reacting simultaneously to a specific event. Such approaches
Glacier fluctuations may also be inferred from the dating of are applicable to processes with strong spatial footprints such
features such as the shorelines of former glacier lakes. The ages as floods, windstorms, snow avalanches, or wildfires. Accurate
of trees colonizing former shorelines provide minimum ages dating of past occurrences can be, therefore, done with rea-
for the lakes, whereas detrital logs stranded on former beaches sonable confidence based on the temporal synchronicity of
provide maximum ages (the logs died before they were signals between samples across the sampled area and defined
washed up on the beach, e.g., Clague et al., 2006). In some thresholds for acceptable quality of evidence. Nevertheless,
cases, trees killed by impoundment and flooding of lakes may more studies are still needed to determine threshold levels for
provide more precise information on the timing of lake in- the reconstruction of other processes and phenomena that are
ception and the related glacier advance (e.g. Masiokas et al., point-based and cover-limited areas where spatial replication
2010, Figure 10(b)). Finally, with the increased development of events is rarely possible. Rockfalls, for example, often occur
of tree-ring reconstructions of precipitation and temperature as single or small-scale events with local rather than areal
variables, recent studies have developed reconstructions of impacts (Moya et al., 2010).
proxy glacier mass balance series several centuries in length Third, at many sites, later events remove or truncate the
that can be compared against the dating of moraines based on evidence for earlier events resulting in a sample population
more traditional means (Nicolussi and Patzelt, 1996; Lewis that is increasingly censored as one goes back in time.
and Smith, 2004; Watson and Luckman, 2004). Finally, although tree-ring series are continuous, trees are
not equally sensitive responders to geomorphic disturbance
over their lifetimes, for example, the response of trees to im-
pact will vary with their age and/or size. Consequently, the
12.9.4 Research Perspectives: Looking to Future response and likelihood of recording an event in a given tree
Developments may vary over time – and may vary differently for different
locations within a sampled network.
As in most subfields of dendrochronology, the application of In many cases, with repeated hazard in a specific locality,
tree rings to geomorphology began with relatively simple the process under study damages or removes the record of past
chronological investigations that involved increment ring activities, thus truncating or censoring the population avail-
counting, dating of individual events, or minimum age dating able for sampling and ‘eliminating’ the record of moderate
of disturbed surfaces (Alestalo, 1971). Subsequently, the im- and smaller events back in time. In such cases, estimates of the
portance of cross-dating and verification has been recognized magnitude and frequency of moderate and smaller events may
and studies have expanded to utilize a wide array of data types be based on the more recent, better-replicated record and the
at a variety of scales. One of the novel contributions has been older records used to document the single largest events with
the development of event chronologies (originally event- less precision. However, these sampling problems make it
response chronologies; Shroder, 1978) that may be used to difficult to estimate changes in the magnitude–frequency dis-
document the magnitude–frequency characteristics of recur- tribution of events over time and underline the point that
rent processes. A central problem with the interpretation of they only provide conservative estimates that may under-
these chronologies is the limitations imposed as sample estimate the true magnitude–frequency spectrum at a site. The
138 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings

sampling problem may also be compounded by the differ- example, dendrochemical analyses have been shown to be a
ential response of trees to damage based on their size, age, and promising tool for the analysis of past volcanic activity
genetics. Experimental research is needed to determine the (Sheppard et al., 2010) and progress could be made in paleo-
optimal size and age (or range of ages) of trees of different hurricane dating through the study of oxygen isotope ratios of
species as recorders of, for example, reaction wood series or tree-ring a-cellulose (Gentry et al., 2010; Grissino-Mayer et al.,
TRDs that may further restrict the usable part of the tree-ring 2010). Further, future progress could be expected through the
records. use of densitometric, resistograph (Lopez Saez et al., 2010), or
Classic dendrogeomorphic work has usually utilized con- computer tomography analyses, mainly for the identification
iferous species, but there is excellent scope for the develop- of tension wood in broadleaved trees or the identification of
ment of similar studies using deciduous, and even tropical, completely overgrown injuries.
species. Research performed with these new species or parts of The vast majority of hazard-related, dendrogeomorphic
trees (e.g., roots and branches) needs to go beyond simple ring research has focused on snow avalanches, landslides, floods,
counting and should include more rigorous statistical com- debris flows, and, most recently, rockfalls. In conjunction with
parisons of ring chronologies or event–response replications dendroclimate studies, data on magnitude and frequency
to avoid spurious errors. The dendrogeomorphic research could develop a better understanding of climatic triggers and
community could profitably examine wood anatomy in more endogenic causes of hazardous processes and related disasters
detail, seeking to exploit new lines of physiological evidence (Solomina, 2002). Considerable potential also exists to ex-
from tree rings. Initial results from wood-anatomical changes pand studies to other catastrophic hazards such as glacier-lake
following flooding exist for several broadleaved species (e.g., outburst floods, ice avalanches, or flash floods. Regionally
Astrade and Bégin, 1997; St. George and Nielsen, 2003; Tardif these studies have generally focused on mountains and the
et al., 2010), and future research should continue to investi- North American cordillera and European Alps in particular.
gate these research lines, including more tree species as well as Ample opportunities arise for the application of these tech-
shrubs or perennial herbs. niques to sites in Latin America, the Indian subcontinent,
In conifers, the effects of wounding and the subsequent Africa, Northern and Eastern Europe, or Russia.
formation of tangential rows of TRDs have been analyzed re- To date, most tree-ring analyses of natural hazards have
cently for L. decidua and P. abies (Stoffel and Hitz, 2008; primarily focused on isolated case studies, for example, on
Bollschweiler et al., 2008b; Schneuwly et al., 2009a, 2009b) debris flows at specific sites, and/or the demonstration of the
following debris-flow, rockfall, or snow avalanche activity. utility and potential of the techniques. Single-site studies are
More fundamental studies are needed to understand whether important but regional studies may also contribute signifi-
other species produce TRDs as a result of geomorphic activity, cantly to hazard management. For example, although debris
and how TRDs intensity and distribution vary in species and flows may have recurrence intervals of 20–30 years at any site
in response to different Earth-surface processes. Many other within a specific valley, other potential sites in the same valley
unresolved questions remain relating to the response of trees may have different histories and patterns of flow activity (e.g.,
to disturbance by geomorphic processes. For example, does Bollschweiler and Stoffel, 2007, 2010). Combining studies
the intensity of an impact or the duration of an event influ- from a network of sites in the valley may indicate that there
ence the nature and intensity of the reaction? What is the lag was a reasonable chance of a debris flow somewhere in the
between an event and the evidence of reaction to that event valley once in every 3–5 years. Such information might be
that is visible in the tree ring? Does response time differ for particularly useful in deciding whether it is practical to put in
different types of responses (e.g., reaction wood and TRD) or place a regional emergency response system to deal with
between tree species? Collaborative and possibly experimental events likely to occur once in 3 years, rather than having to
research is needed with tree phytologists and physiologists to initiate a new response every time there is a debris flow.
address these questions and improve understanding of how Therefore, in the future , particularly in mountainous areas,
biochemical agents initiate tree response to disturbance and it may be important to establish regional or supraregional
how and when the actual formation of reaction wood, flood event chronologies and databases for natural hazards similar
rings, or TRD become visible in the tree ring. This knowledge to those existing in dendroclimatology. In addition, there is
will be particularly beneficial to studies of intraring dating and also a need and scope for our databases to be compared with
seasonal dating accuracy. New insights could also be generated dendroclimatological networks (Briffa et al., 1992, 2001) to
relatively easily through artificially inflicting wounds, burial of explore climate–natural hazard interactions. On an individual
stems, root exposure, or destabilization (tilting) of trees. Al- site-basis tree-ring data can be used for an accuracy assessment
ternatively, it would be feasible to analyze trees that have been or calibration of models of different geomorphic processes.
damaged during well-documented events in the past For example, dendrogeomorphic data have been used spor-
(Bollschweiler et al., 2008b; Ballesteros et al., 2010; Kaczka adically in the past for the testing of rockfall (Stoffel et al.,
et al., 2010). Developing a better understanding of tree re- 2006a), snow avalanche (Casteller et al., 2008) and debris-
actions following geomorphic activity will facilitate identifi- flow models (Graf et al., 2009). These results clearly pinpoint
cation of responses in tree-ring series and avoid abundant the high value added to model verification and improvement
destructive sampling in the future. with on-site field data.
As the general field of dendrochronology develops, it offers Examination of the past and contemporary nature, occur-
increasing opportunities to explore the application of newer rence, and distribution of Earth-surface processes using den-
techniques to dendrogeomorphic analyses and several prom- drogeomorphology provides extended records of the spatial
ising approaches are illustrated in Stoffel et al. (2010). For and temporal distribution of many significant natural hazards.
 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings 139

In many parts of the world, this unique information may be Bollschweiler, M., 2007. Spatial and temporal occurrence of past debris flows in the
the only available source of long-term data to document Valais Alps – results from tree-ring analysis. Geofocus 20, 1–182.
Bollschweiler, M., Stoffel, M., 2007. Debris flows on forested cones –
changing process dynamics over time and space, thereby
reconstruction and comparison of frequencies in two catchments in Val Ferret,
providing a vital tool to show the influence of past and current Switzerland. Natural Hazards and Earth System Sciences 7, 207–218.
conditions on a range of natural hazards. This brief chapter Bollschweiler, M., Stoffel, M., 2010. Variations in debris-flow occurrence in an
has outlined the contribution that dendrogeomorphology can Alpine catchment – Reconstruction and implications for the future. Global and
make to the study of many of these hazards. In conclusion, we Planetary Change 73, 186–192.
Bollschweiler, M., Stoffel, M., Ehmisch, M., Monbaron, M., 2007. Reconstructing
suggest that these techniques merit serious consideration as an spatio-temporal patterns of debris-flow activity with dendrogeomorphological
integral component of future studies of hazard assessment and methods. Geomorphology 87(4), 337–351.
management at suitable sites. Bollschweiler, M., Stoffel, M., Schneuwly, D.M., 2008a. Dynamics in debris-flow
activity on a forested cone – A case study using different dendroecological
approaches. Catena 72, 67–78.
Bollschweiler, M., Stoffel, M., Schneuwly, D.M., Bourqui, K., 2008b. Traumatic resin
References ducts in Larix decidua stems impacted by debris flows. Tree Physiology 28,
255–263.
Bollschweiler, M., Stoffel, M., Vazquez-Selem, L., Palacios, D., 2010. Spatio-
Agafonov, L., Strunk, H., Nuber, T., 2004. Thermokarst dynamics in Western Siberia:
temporal reconstruction of lahar activity in Barranca Huiloac (Volcán
insights from dendrochronological research. Palaeogeography, Palaeoclimatololy,
Popocatépetl, México). Holocene 20, 265–274.
Palaeoecology 209, 183–196.
Bosch, O., Gutierrez, E., 1999. La sucésion en los bosques de Pinus uncinata del
Alestalo, J., 1971. Dendrochronological interpretation of geomorphic processes.
Pirineo. De los anillos de crecimiento a la historia del bosque. Ecologia 13,
Fennia 105, 1–139.
133–171.
Allen, R., Bellingham, P., Wiser, S., 1999. Immediate damage by an earthquake to a
Boucher, D., Filion, L., Hétu, B., 2003. Reconstitution dendrochronologique et
temperate montane forest. Ecology 80, 708–714. fréquence des grosses avalanches de neige dans un couloir subalpin du mont
Arbellay, E., Stoffel, M., Bollschweiler, M., 2010. Dendrogeomorphic reconstruction Hog’s Back, en Gaspésie centrale (Québec). Géographie physique et Quaternaire
of past debris-flow activity using injured broad-leaved trees. Earth Surface 57, 159–168.
Processes and Landforms 35, 399–406. Braam, R.R., Weiss, E.E.J., Burrough, P.A., 1987a. Dendrogeomorphological
Astrade, L., Bégin, Y., 1997. Tree-ring response of Populus tremula L. and Quercus analysis of mass movement – a technical note on the research method. Catena
robur L. to recent spring floods of the Saône river, France. Ecoscience 4, 14, 585–589.
232–239. Braam, R.R., Weiss, E.E.J., Burrough, P.A., 1987b. Spatial and temporal analysis of
Aulitzky, H., 1992. Die Sprache der ‘‘Stummen Zeugen’’. Proceedings of the mass movement using dendrochronology. Catena 14, 573–584.
International Conference Interpraevent 1992, Bern, Switzerland, pp. 139–174. Bräker, O.U., 2002. Measuring and data processing in tree-ring research – a
Bachrach, T., Jakobsen, K., Kinney, J., Nishimura, P., Reyes, A., Laroque, C.P., methodological introduction. Dendrochronologia 20, 203–216.
Smith, D.J., 2004. Dendrogeomorphological assessment of movement at Hilda Briffa, K.R., Jones, P.D., 1992. Basic chronology statistics and assessment. In: Cook,
rock glacier, Banff National Park, Canadian Rocky Mountains. Geografiska E.R., Kairiukstis, L.A. (Eds.), Methods of Dendrochronology: Applications in the
Annaler 86A, 1–9. Environmental Sciences. Kluwer Academic New York, pp. 137–152.
Ballesteros, J.A., Eguibar, M., Bodoque, J.M., Dı́ez, A., Gutiérrez, I., Stoffel, M., Briffa, K.R., Jones, P.D., Schweingruber, F.H., 1992. Tree-ring density
2011. Estimating flash flood discharge in an ungauged mountain catchment with reconstructions of summer temperature patterns across western North America
2D hydraulic models and dendrogeomorphic palaeostage indicators. since 1600. Journal of Climate 5, 735–754.
Hydrological Processes 25, 970–979. Briffa, K.R., Osborn, T.J., Schweingruber, F.H., Harris, I.C., Jones, P.D., Shiyatov,
Ballesteros, J.A., Stoffel, M., Bodoque, J.M., Bollschweiler, M., Hitz, O.M., Diez, A., S.G., Vaganov, E.A., 2001. Low-frequency temperature variations from a
2010. Changes in wood anatomy in tree rings of Pinus pinaster Ait. following northern tree ring density network. Journal of Geophysical Research 106,
wounding by flash floods. Tree-Ring Research 66(2), 93–103. 2929–2942.
Bannan, M.W., 1936. Vertical resin ducts in the secondary wood of the Abietineae. Burn, C.R., Friele, P.A., 1989. Geomorphology, vegetational succession, soil
New Phytologist 35, 11–46. characteristics and permafrost in retrogressive thaw slumps near Mayo, Yukon
Bannan, M.W., 1941. Vascular rays and adventitious root formation in Thuja Territory. Arctic 33, 31–40.
occidentalis L. American Journal of Botany 28, 457–463. Burn, C.R., Smith, M.W., 1988. Thermokarst lakes at Mayo, Yukon Territory.
Baumann, F., Kaiser, K.F., 1999. The Multetta debris fan, eastern Swiss Alps: a 500- Proceedings of the Fifth International Permafrost Conference. Tapir , Trondheim,
year debris flow chronology. Arctic, Antarctic, and Alpine Research 31(2), Norway, pp. 700–705.
128–134. Butler, D.R., 1979. Snow avalanche path terrain and vegetation, Glacier National
Bedinger, M.S., 1971. Forest species as indicators of flooding in the lower White Park, Montana. Arctic and Alpine Research 11, 17–32.
River Valley, Arkansas. US Geological Survey Professional Paper 750-C. US Butler, D.R., 1985. Vegetational and geomorphic change on snow avalanche paths,
Glacier National Park. Great Basin Naturalist 45, 313–317.
Geological Survey, Alexandria, VA, pp. C248–C253.
Butler, D.R., Malanson, G.P., 1985. A history of high-magnitude snow avalanches,
Bégin, C., Filion, L., 1985. Analyse dendrochronologique d’un glissement de terrain
Southern-Glacier-National-Park, Montana, USA. Mountain Research and
dans la region du Lac de l’Eau Claire (Québec nordique). Canadian Journal of
Development 5, 175–182.
Earth Science 22, 175–182.
Butler, D.R., Malanson, G.M., Oelfke, J.G., 1987. Tree-ring analysis and natural
Bégin, C., Filion, L., 1988. Age of landslides along Grande Rivière de la
hazard chronologies: minimum sample sizes and index values. Professional
Baleine estuary, eastern coast of Hudson Bay, Quebec (Canada). Boreas 17, Geographer 39, 41–47.
289–299. Butler, D.R., Oelfke, J.G., Oelfke, L.A., 1986. Historic rockfall avalanches,
Bégin, Y., Langlais, D., Cournoyer, L., 1991. Tree-ring dating of shore erosion northeastern Glacier National Park, Montana, USA. Mountain Research and
events (Upper St. Lawrence estuary, eastern Canada). Geografiska Annaler 73A, Development 6, 261–271.
53–59. Butler, D.R., Sawyer, C.F., 2008. Dendrogeomorphology and high-magnitude snow
Bekker, M.F., 2004. Spatial variation in the response of tree rings to normal faulting avalanches: a review and case study. Natural Hazards and Earth System Science
during the Hebgen Lake Earthquake, Southwestern Montana, USA. 8, 303–309.
Dendrochronologia 22, 53–59. Butler, D.R., Sawyer, C.F., Maas, J.A., 2010. Tree-ring dating of snow avalanches in
Biondi, F., Estrada, I.G., Gavilanes Ruiz, J.C., Torres, A.E., 2003. Tree growth Glacier National Park, Montana, USA. In: Stoffel, M., Bollschweiler, M., Butler,
response to the 1913 eruption of Volcán de Fuego de Colima, Mexico. D.R., Luckman, B.H. (Eds.), Tree Rings and Natural Hazards: A State-of-the-Art.
Quaternary Research 59, 293–299. Springer, Dordrecht, Heidelberg, London, New York, NY, pp. 35–46.
Bodoque, J.M., Diez-Herrero, A., Martin-Duque, J.F., et al., 2005. Sheet erosion Cannone, N., Gerdol, R., 2003. Vegetation as an ecological indicator of
rates determined by using dendrogeomorphological analysis of exposed tree surface instability in rock glaciers. Arctic, Antarctic and Alpine Research 35,
roots: two examples from Central Spain. Catena 64(1), 81–102. 384–390.
140 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings

Carrara, P.E., 1979. Geological Society of America Bulletin. The determination of Hazards: A State-of-the-Art. Springer, Dordrecht, Heidelberg, London, New York,
snow avalanche frequency through tree-ring analysis and historical records at NY, pp. 309–319.
Ophir, Colorado 90, 773–778. Germain, D., Filion, L., Hétu, B., 2009. Snow avalanche regime and climatic
Carrara, P.E., Carroll, T.R., 1979. Determination of erosion rates from exposed tree conditions in the Chic-Chocs Range, eastern Canada. Climatic Change 92,
roots in the Piceance Basin, Colorado. Earth Surface Processes and Landforms 141–167.
4, 307–317. Giardino, J.R., Shroder, J.F., Lawson, M.P., 1984. Tree ring analysis of movement in
Carrara, P.E., O’Neill, J.M., 2003. Tree-ring dated landslide movements and their a rock-glacier complex on Mount Mestas, Colorado, USA. Arctic and Alpine
relationship to seismic events in southwestern Montana, USA. Quaternary Research 16, 299–309.
Research 59, 25–35. Graf, C., Stoffel, M., Grêt-Regamey, A., 2009. Enhancing debris flow modeling
Carter, R., LeRoy, S., Nelson, T., Laroque, C.P., Smith, D.J., 1999. parameters integrating Bayesian networks. Geophysical Research Abstracts 11,
Dendroglaciological investigations at Hilda Creek rock glacier, Banff National 10725.
Park, Canadian Rocky Mountains. Géographie physique et Quaternaire 53, Grissino-Mayer, H.D., 2001. Evaluating cross-dating accuracy: a manual and tutorial
365–371. for the computer program COFECHA. Tree-Ring Research 57, 205–221.
Casteller, A., Christen, M., Villalba, R., Martı́nez, H., Stöckli, V., Leiva, J.C., Bartelt, Grissino-Mayer, H.D., 2003. A manual and tutorial for the proper use of an
P., 2008. Validating numerical simulations of snow avalanches using increment borer. Tree-Ring Research 59, 63–79.
dendrochronology: The Cerro Ventana event in Northern Patagonia, Argentina. Grissino-Mayer, H.D., Miller, D.L., Mora, C.L., 2010. Dendrotempestology and the
Natural Hazards and Earth System Science 8, 433–443. isotopic record of tropical cyclones in tree rings of the southeastern United
Clague, J.J., Luckman, B.H., Van Dorp, R.D., Gilbert, R., Froese, D., Jensen, B.J.L., States. In: Stoffel, M., Bollschweiler, M., Butler, D.R., Luckman, B.H. (Eds.), Tree
Reyes, A.V., 2006. Rapid changes in the level of Kluane Lake, Yukon Territory, Rings and Natural Hazards: A State-of-the-Art. Springer, Dordrecht, Heidelberg,
over the last millennium. Quaternary Research 66, 342–355. London, New York, NY, pp. 291–303.
Clague, J.J., Souther, J.G., 1982. The Dusty Creek landslide on Mount Cayley, Gutsell, S.L., Johnson, E.A., 2002. Accurately ageing trees and examining their
British Columbia. Canadian Journal of Earth Sciences 19, 524–539. height-growth rates: implications for interpreting forest dynamics. Journal of
Cook, E.R., 1987. The decomposition of tree-ring series for environmental studies. Ecology 90, 153–166.
Tree-Ring Bulletin 47, 37–59. Hamilton, W.L., 2010. Seismic damage in conifers from Olympic and Yellowstone
Cook, E.R., Kairiukstis, L.A., 1990. Methods of Dendrochronology – Applications in National Parks, United States. In: Stoffel, M., Bollschweiler, M., Butler, D.R.,
the Environmental Sciences. Kluwer, London, 408 pp. Luckman, B.H. (Eds.), Tree Rings and Natural Hazards: A State-of-the-Art.
Cooper, W.S., 1916. Plant succession in the mount robson region, British Springer, Dordrecht, Heidelberg, London, New York, NY, pp. 437–440.
Columbia. Plant World 19, 211–238. Helley, E.J., LaMarche, V.C., 1968. December 1964, a 400-year flood in Northern
Corominas, J., Moya, J., 1999. Reconstructing recent landslide activity in relation to California. US Geological Survey Professional Paper 600-D. US Geological
rainfall in the Llobregat River basin, Eastern Pyrenees, Spain. Geomorphology Survey, Alexandria, VA, pp. D34–D37.
30, 79–93. Hitz, O.M., Gärtner, H., Heinrich, I., Monbaron, M., 2008. First time application of
Corona, C., Lopez, J., Rovéra, G., Astrade, L., Stoffel, M., Berger, F., in press. ash (Fraxinus excelsior L.) roots to determine erosion rates in mountain
Quantification of erosion rates from exposed roots: methodological validation in torrents. Catena 72, 248–258.
marly badlands in the experimental basins of Draix (Alpes de Haute-Provence). Holzhauser, H.P., 1984. Zur Geschichte der Aletschgletscher und des
Géomorphologie: Relief, Processus, Environnement. Fieschergletschers. Physische Geographie 13, 488 pp.
Corona, C., Lopez, J., Rovéra, G., Stoffel, M., Astrade, L., Berger, F., 2011. High- Hupp, C.R., 1984. Dendrogeomorphic evidence of debris flow frequency and
resolution, quantitative reconstruction of erosion rates based on anatomical magnitude at Mount Shasta, California. Environmental Geology and Water
changes in exposed roots (Draix, Alpes de Haute-Provence) – critical review of Science 6, 121–128.
existing approaches and independent quality control of results. Geomorphology Hupp, C.R., Osterkamp, W.R., Thornton, J.L., 1987. Dendrogeomorphic evidence
125, 433–444. and dating of recent debris flows on Mount Shasta, northern California. U.S.
Corona, C., Rovéra, G., Lopez, J., Stoffel, M., Perfettini, P., 2010. Spatio- Geological Survey Professional Paper 1396B. US Geological Survey, Alexandria,
temporal reconstruction of snow avalanche activity using tree rings: Jean VA, pp. 1–39.
Jeanne avalanche talus, Massif de l’Oisans, France. Catena 83, 107– Ives, J.D., Mears, A.I., Carrara, P.E., Bovis, M.J., 1976. Natural hazards in
118. mountain Colorado. Annals of the Association of American Geographers 66,
D’Arrigo, R., Frank, D., Jacoby, G., Pederson, N., 2001. Spatial response to major 129–144.
volcanic events in or about AD 536, 934 and 1258: frost rings and other Jacoby, G.C., 1997. Application of tree ring analysis to paleoseismology. Reviews of
dendrochronological evidence from Mongolia and Northern Siberia. Climatic Geophysics 35, 109–124.
Change 49, 239–246. Jacoby, G.C., Sheppard, P.R., Sieh, K.E., 1988. Irregular recurrence of large
Dorren, L.K.A., Berger, F., 2006. Stem breakage of trees and energy dissipation earthquakes along the San Andreas Fault. Evidence from trees. Science 241,
during rockfall impacts. Tree Physiology 26, 63–71. 196–199.
Du, S., Yamamoto, F., 2007. An overview of the biology of reaction wood Jacoby, G.C., Williams, P.L., Buckley, B.M., 1992. Tree ring correlation between
Formation. Journal of Integrative Plant Biology 49, 131–143. prehistoric landslides and abrupt tectonic events in Seattle, Washington. Science
Dubé, S., Filion, L., Hétu, B., 2004. Tree-ring reconstruction of high-magnitude 258, 1621–1623.
snow avalanches in the Northern Gaspé Peninsula, Québec, Canada. Arctic, Jakob, M., 1995. Dendrochronology to measure average movement rates of
Antarctic and Alpine Research 36, 555–564. gelifluction lobes. Dendrochronologia 13, 141–146.
Fantucci, R., 2007. Dendrogeomorphological analysis of shore erosion along Kaczka, R.J., Deslauriers, A., Morin, H., 2010. High-precision dating of debris-flow
Bolsena lake (Central Italy). Dendrochronologia 24, 69–78. events within the growing season. In: Stoffel, M., Bollschweiler, M., Butler, D.R.,
Fantucci, R., McCord, A., 1995. Reconstruction of landslide dynamic with Luckman, B.H. (Eds.), Tree Rings and Natural Hazards: A State-of-the-Art.
dendrochronological methods. Dendrochronologia 13, 43–58. Springer, Dordrecht, Heidelberg, London, New York, NY, pp. 227–229.
Fantucci, R., Sorriso-Valvo, M., 1999. Dendrogeomorpholigical analysis of a slope Koch, J., 2009. Improving age estimates for late Holocene glacial landforms using
near Lago, Calabria (Italy). Geomorphology 30, 165–174. dendrochronology – some examples from Garibaldi Provincial Park, British
Filion, L., Quinty, F., Bégin, C., 1991. A chronology of landslide activity in the Columbia. Quaternary Geochronology 4, 130–139.
valley of Rivière du Gouffre, Charlevoix, Quebec. Canadian Journal of Earth Lafortune, M., Filion, L., Hétu, B., 1997. Dynamique d’un front forestier sur un
Science 28, 250–256. talus d’éboulis actif en climat tempéré froid (Gaspésie, Québec). Géographie
Frazer, G.W., 1986. Dendrogeomorphic evaluation of snow avalanche history at two physique et Quaternaire 51, 1–15.
sites in Banff National Park. MSc thesis, University of Western Ontario. LaMarche, V.C., 1961. Rate of slope erosion in the White Mountains, California.
Friedman, J.M., Vincent, K.R., Shafroth, P.B., 2005. Dating floodplain sediments Geological Society of America Bulletin 72, 1579–1580.
using tree-ring response to burial. Earth Surface Processes and Landforms 30, LaMarche, V.C., 1966. An 800-year history of stream erosion as indicated by
1077–1091. botanical evidence. US Geological Survey Professional Paper 550D. US
Fritts, H.C., 1976. Tree Rings and Climate. Academic Press, London, 567 pp. Geological Survey, Alexandria, VA, pp. 83–86.
Gentry, C.M., Lewis, D., Speer, J.H., 2010. Dendroecology of hurricanes and the LaMarche, V.C., 1968. Geomorphic and dendroecological impacts of slushflows in
potential for isotopic reconstructions in Southeastern Texas. In: Stoffel, M., central Gaspé Peninsula (Québec, Canada). US Geological Survey Professional
Bollschweiler, M., Butler, D.R., Luckman, B.H. (Eds.), Tree Rings and Natural Paper 352-I. US Geological Survey, Alexandria, VA.
 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings 141

Larocque, S.J., Hétu, B., Filion, L., 2001. Geomorphic and dendroecological Bollschweiler, M., Butler, D.R., Luckman, B.H. (Eds.), Tree Rings and Natural
impacts of slushflows in central Gaspé Peninsula (Québec, Canada). Geogriska Hazards: A State-of-the-Art. Springer, Dordrecht, Heidelberg, London, New York,
Annaler 83, 191–201. NY, pp. 161–180.
Larson, P.R., 1994. The Vascular Cambium. Development and Structure. Springer, Mundo, I.A., Barrera, M.D., Roig, F.A., 2007. Testing the utility of Nothofagus
Berlin, 725 pp. pumilio for dating a snow avalanche in Tierra del Fuego, Argentina.
Lewis, D.H., Smith, D.J., 2004. Dendrochronological mass balance reconstruction, Dendrochronologia 25, 19–28.
Strathcona Provincial Park, Vancouver Island, British Columbia, Canada. Arctic, Muntán, E., Andreu, L., Oller, P., Gutiérrez, E., Martı́nez, P., 2004.
Antarctic and Alpine Research 36, 598–606. Dendrochronological study of the Canal del Roc Roig avalanche path: first results
Lin, A., Lin, S., 1998. Tree damage and surface displacement: the 1931 m 8.0 of the Aludex project in the Pyrenees. Annals of Glaciology 38, 173–179.
Fuyun earthquake. Journal of Geology 106, 751–757. Muntán, E., Garcı́a, C., Oller, P., Martı́, G., Garcı́a, A., Gutiérrez, E., 2009.
Lopez Saez, J., Corona, C., Berger, F., Stoffel, M., 2010. L’utilisation de la Reconstructing snow avalanches in the Southeastern Pyrenees. Natural Hazards
résistographie en dendrogéomorphologie: retour d’expériences. In: Astrade, L. and Earth System Science 9, 1599–1612.
(Ed.), Panorama de la dendrochronologie en France. Collection EDYTEM, Nicolussi, K., Patzelt, G., 1996. Reconstructing glacier history in Tyrol by means of
Cahiers de Géographie. tree-ring investigations Zeitschrift für Gletscherkunde und Glazialgeologie 32,
Luchi, N., Ma, R., Capretti, P., Bonello, P., 2005. Systemic induction of traumatic 207–215.
resin ducts and resin flow in Austrian pine by wounding and inoculation with Paolini, L., Villalba, R., Grau, H.R., 2005. Precipitation variability and landslide
Sphaeropsis sapinea and Diplodia scrobiculata. Planta 221, 75–84. occurrence in a subtropical mountain ecosystem of NW Argentina.
Luckman, B.H., 1988. Dating the moraines and recession of Athabasca and Dome Dendrochronologia 22, 175–180.
Glaciers, Alberta, Canada. Arctic and Alpine Research 20, 40–54. Papadopoulos, A.M., Mertzanis, A., Pantera, A., 2007. Dendrogeomorphological
Luckman, B.H., 1995. Calendar-dated, early Little Ice Age glacier advance at observations in a landslide on Tymfristos mountain in Central Greece. In:
Robson Glacier, British Columbia, Canada. Holocene 5, 149–159. Stokes, A., Spanos, I., Norris, J.E., Cammeraat, E. (Eds.), Eco- and Ground Bio-
Luckman, B.H., 1998. Dendroglaciologie dans les Rocheuses du Canada. Engineering: The Use of Vegetation to Improve Slope Stability. Springer, Berlin,
Géographie physique et Quaternaire 52, 137–149. Heidelberg, New York, NY, pp. 223–230.
Luckman, B.H., 2006. The Neoglacial history of Peyto Glacier. In: Demuth, M., Parker, M.l., Josza, L.J., 1973. Dendrochronological Investigations along the
Munro, S., Young, G. (Eds.), Peyto Glacier: One Century of Science. National Mackenzie, Liard and South Nahanni Rivers N.W.T. Hydrologic Affects of
Hydrological Research Institute Science Report Number 8. NHRI, Saskatoon, pp. Northern Pipeline Development, Report 73-3. Environmental-Social Committee
25–57. Northern Pipelines, Task Force on Northern Oil Development, Information
Luckman, B.H., Frazer, G.W., 2001. Dendrogeomorphic investigations of snow Canada Cat.No R27-172, pp. 313–464.
avalanche tracks in the Canadian Rockies. International Conference on Tree Perret, S., Stoffel, M., Kienholz, H., 2006. Spatial and temporal rockfall activity in a
Rings and People. Davos, Switzerland, p. 180. forest stand in the Swiss Prealps – a dendrogeomorphological case study.
Luckman, B.H., Villalba, R., 2000. Assessing synchroneity of glacier fluctuations in Geomorphology 74(1–4), 219–231.
the western cordillera of the Americas during the last millennium. In: Markgraf, Phillips, M.A., Croteau, R.B., 1999. Resin-based defenses in conifers. Trends in
V. (Ed.), Interhemispheric Climate Linkages. Academic Press, New York, NY, pp. Plant Science 4, 184–190.
119–140. Pierson, T.C., 2007. Dating young geomorphic surfaces using age of colonizing
Lundström, T., Heiz, U., Stoffel, M., Stöckli, V., 2007. Fresh-wood bending: linking Douglas fir in southwestern Washington and northwestern Oregon. USA. Earth
the mechanical and growth properties of a Norway spruce stem. Tree Physiology Surface Processes and Landforms 32, 811–831.
27, 1229–1241. Pilate, G., Chabbert, B., Cathala, B., et al., 2004. Lignification and tension wood.
Lundström, T., Jonsson, M.J., Volkwein, A., Stoffel, M., 2009. Reactions and energy Comptes Rendus Biologies 327, 889–901.
absorption of trees subject to rockfall: a detailed assessment using a new Potter, Jr. N., 1969. Tree-ring dating of snow avalanches, northern Absaroka
experimental method. Tree Physiology 29, 345–359. mountains, Wyoming. In: Schumm, S.A., Bradley, W.S. (Eds.), Contributions to
Lundström, T., Stoffel, M., Stöckli, V., 2008. Fresh-stem bending of silver fir and Quaternary Research INQUA VIII. Geological Society of America 123. Geological
Norway spruce. Tree Physiology 28, 355–366. Society of America, Boulder, CO, pp. 141–165.
Marin, P., Filion, L., 1992. Recent dynamics of Sub-Arctic dunes as determined by Pringle, P.T., Pierson, T.C., Cameron, K.A., Sheppard, P.A., 2010. Late eighteenth
Tree-Ring analysis of White Spruce, Hudson-Bay, Quebec. Quaternary Research century Old Maid eruption and lahars at Mount Hood, Oregon (USA) dated
38, 316–330. with tree rings and historical observations. In: Stoffel, M., Bollschweiler, M.,
Masiokas, M.H., Luckman, B.H., Villalba, R., Ripalta, A., Rabassa, J., 2010. Little Butler, D.R., Luckman, B.H. (Eds.), Tree Rings and Natural Hazards: A
Ice Age fluctuations of Glaciar Rı́o Manso in the north Patagonian Andes of State-of-the-Art. Springer, Dordrecht, Heidelberg, London, New York, NY,
Argentina. Quaternary Research 73, 96–106. pp. 487–491.
Mattheck, C., 1993. Design in der Natur. Der Baum als Lehrmeister. Rombach Rizzo, D.M., Harrington, T.C., 1988. Root movement and root damage of red spruce
Wissenschaft, 326 pp. and balsam fir on subalpine sites in the White Mountains, New Hampshire.
Mayer, B., Stoffel, M., Bollschweiler, M., Hübl, J., Rudolf-Miklau, F., 2010. Canadian Journal of Forest Research 18(8), 991–1001.
Frequency and spread of debris floods on fans: a dendrogeomorphic case-study Rubiales, J.M., Bodoque, J.M., Ballesteros, J.A., Diez-Herrero, A., 2008. Response
from a dolomite catchment in the Austrian Alps. Geomorphology 118, 199–206. of Pinus sylvestris roots to sheet-erosion exposure: an anatomical approach.
McAuliffe, J.R., Scuderi, L.A., McFadden, L.D., 2006. Tree-ring record of hillslope Natural Hazards and Earth System Science 8, 223–231.
erosion and valley floor dynamics: landscape responses to climate variation Ruel, J.J., Ayres, M.P., Lorio, P.L., 1998. Loblolly pine responds to mechanical
during the last 400yr in the Colorado Plateau, northeastern Arizona. Global and wounding with increased resin flow. Canadian Journal of Forest Research 28,
Planetary Change 50, 184–201. 596–602.
McCarthy, D.P., 1985. Dating holocene geomorphic activity of selected landforms in Ruiz, V., Dı́ez, A., Stoffel, M., Bollschweiler, M., Bodoque, J.M., Ballesteros, J.A.,
the Geikie Creek Valley, Mount Robson Provincial Park, British Columbia. MSc 2010. Dating flash flood events by means of dendrogeomorphic analysis in a
thesis, University of Western Ontario. small ungauged mountain catchment (Spanish Central System). Geomorphology
McCarthy, D.P., Luckman, B.H., 1993. Estimating ecesis for tree-ring dating of 118, 383–392.
moraines – a comparative study from the Canadian Cordillera. Arctic and Alpine Sachs, T., 1991. Pattern Formation in Plant Tissue. Cambridge University Press,
Research 25, 63–68. Cambridge.
McCarthy, D.P., Luckman, B.H., Kelly, P.E., 1991. Sampling height-age error Salzer, M.W., Hughes, M.K., 2007. Bristlecone pine tree rings and volcanic
correction for spruce seedlings in glacial forefields, Canadian Cordillera. Arctic eruptions over the last 5000 yr. Quaternary Research 67, 57–68.
and Alpine Research 23, 451–455. Schaerer, P.A., 1972. Terrain and vegetation of snow avalanche sites at Rogers
Meisling, K., Sieh, K., 1980. Disturbance of trees by the 1857 Fort Tejon Pass, British Columbia. B.C. Geographical Series 14, 215–222.
earthquake. Journal of Geophysical Research 85, 3225–3238. Schneuwly, D.M., Stoffel, M., 2008a. Spatial analysis of rockfall activity, bounce
Moore, D.P., Mathews, W.H., 1978. The Rubble Creek landslide, heights and geomorphic changes over the last 50 years – a case study using
southwestern British Columbia. Canadian Journal of Earth Science 15, dendrogeomorphology. Geomorphology 102, 522–531.
1039–1052. Schneuwly, D.M., Stoffel, M., 2008b. Tree-ring based reconstruction of the seasonal
Moya J., Corominas, J., Pérez Arcas, J., 2010. Assessment of the rockfall frequency timing, major events and origin of rockfall on a case-study slope in the Swiss
for hazard analysis at Solà d’Andorra (Eastern Pyrenees). In: Stoffel, M., Alps. Natural Hazards and Earth System Sciences 8, 203–211.
142 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings

Schneuwly, D.M., Stoffel, M., Bollschweiler, M., 2009a. Formation and spread of Stoffel, M., Bollschweiler, M., Hassler, G.R., 2006a. Differentiating events on a cone
callus tissue and tangential rows of resin ducts in Larix decidua and Picea abies influenced by debris-flow and snow avalanche activity – a
following rockfall impacts. Tree Physiology 29, 281–289. dendrogeomorphological approach. Earth Surface Processes and Landforms 31,
Schneuwly, D.M., Stoffel, M., Dorren, L.K.A., Berger, F., 2009b. Three-dimensional 1424–1437.
analysis of the anatomical growth response of European conifers to mechanical Stoffel, M., Bollschweiler, M., Leutwiler, A., Aeby, P., 2008a. Large debris-flow
disturbance. Tree Physiology 29, 1247–1257. events and overbank sedimentation in the Illgraben torrent (Valais Alps,
Schweingruber, F.H., 1993. Jahrringe und Umwelt. Dendroökologie. Eidgenössische Switzerland). Open Geology Journal 2, 18–29.
Forschungsanstalt für Wald. Schnee und Landschaft, Birmensdorf, 474 pp. Stoffel, M., Casteller, A., Luckman, B.H., Villalba, R., 2012. Spatio-temporal analysis
Schweingruber, F.H., 1996. Tree Rings and Environment. Dendroecology. Paul of channel wall erosion in ephemeral torrents – an example from the Patagonian
Haupt, Bern, Stuttgart, Wien, 609 pp. Andes. Geology 40(3), 247–250.
Schweingruber, F.H., Eckstein, D., Serre-Bachet, F., Bräker, O.U., 1990. Identification, Stoffel, M., Conus, D., Grichting, M.A., Lievre, I., Maitre, G., 2008b. Unraveling the
presentation and interpretation of event years and pointer years in patterns of late Holocene debris-flow activity on a cone in the Swiss Alps:
dendrochronology. Dendrochronologia 8, 9–39. chronology, environment and implications for the future. Global and Planetary
Scuderi, L.A., McFadden, L.D., McAuliffe, J.R., 2008. Dendrogeomorphically Change 60, 222–234.
derived slope response to decadal and centennial scale climate variability: Stoffel, M., Hitz, O.M., 2008. Rockfall and snow avalanche impacts leave different
Black Mesa, Arizona, USA. Natural Hazards and Earth System Sciences 8, anatomical signatures in tree rings of juvenile Larix decidua. Tree Physiology
869–880. 28, 1713–1720.
Sheppard, P.R., Weaver, R., Pringle, P.T., Kent, A.J., 2010. Dendrochemical evidence Stoffel, M., Lièvre, I., Conus, D., Grichting, M.A., Raetzo, H., Gärtner, H.W.,
of the 1781 eruption of Mount Hood, Oregon. In: Stoffel, M., Bollschweiler, M., Monbaron, M., 2005a. 400 years of debris-flow activity and triggering weather
Butler, D.R., Luckman, B.H. (Eds.), Tree Rings and Natural Hazards: A State-of- conditions: Ritigraben, Valais, Switzerland. Arctic, Antarctic and Alpine Research
the-Art. Springer, Dordrecht, Heidelberg, London, New York, NY, pp. 465–467. 37, 387–395.
Shigo, A.L., 1984. Compartmentalization – a conceptual framework for Stoffel, M., Lièvre, I., Monbaron, M., Perret, S., 2005b. Seasonal timing of rockfall
understanding how trees grow and defend themselves. Annual Review of activity on a forested slope at Taschgufer (Swiss Alps) – a dendrochronological
Phytopathology 22, 189–214. approach. Zeitschrift für Geomorphologie 49, 89–106.
Shroder, J.F., 1978. Dendrogeomorphological analysis of mass movement on Table Stoffel, M., Perret, S., 2006. Reconstructing past rockfall activity with tree rings:
Cliffs Plateau, Utah. Quaternary Research 9, 168–185. some methodological considerations. Dendrochronologia 24, 1–15.
Sigafoos, R.H., 1964. Botanical evidence of floods and floodplain deposition. US Stoffel, M., Schneuwly, D., Bollschweiler, M., Lièvre, I., Delaloye, R., Myint, M.,
Geological Survey Professional Paper 485-A. US Geological Survey, Alexandria, Monbaron, M., 2005c. Analyzing rockfall activity (1600–2002) in a protection
VA. forest – a case study using dendrogeornorphology. Geomorphology 68,
Sigafoos, R.H., Hendricks, E.L., 1969. The time interval between stabilization of 224–241.
alpine glacial deposits and establishment of tree seedlings. US Geological Stoffel, M., Wehrli, A., Kühne, R., Dorren, L.K.A., Perret, S., Kienholz, H., 2006b.
Survey Professional Paper 650B. US Geological Survey, Alexandria, VA, pp. Quantifying the protective effect of mountain forests against rockfall using a 3D
B89–B93. simulation model. Forest Ecology and Management 225, 113–122.
Smith, D.J., Lewis, D.H., 2007. Dendroglaciology. In: Scott, E. (Ed.), Encyclopedia Stokes, M.A., Smiley, T.L., 1968. An Introduction to Tree-Ring Dating. University of
of Quaternary Science. Elsevier, Oxford, pp. 988–994. Chicago Press, Chicago, 73 pp.
Solomina, O., 2002. Dendrogeomorphology: research requirements. Strunk, H., 1989. Dendrogeomorphology of debris flows. Dendrochronologia 7,
Dendrochronologia 20, 233–245. 15–25.
Solomina, O., Pavlova, I., Curtis, A., Jacoby, G., Ponomareva, V., Pevzner, M., Strunk, H., 1991. Frequency distribution of debris flow in the Alps since the ‘‘Little
2008. Constraining recent Shiveluch volcano eruptions (Kamchatka, Russia) by Ice Age’’. Zeitschrift für Geomorphologie Supplement 83, 71–81.
means of dendrochronology. Natural Hazards and Earth System Sciences 8, Strunk, H., 1995. Dendrogeomorphologische Methoden zur Ermittlung der
1083–1097. Murfrequenz und Beispiele ihrer Anwendung. Roderer, Regensburg, 196 pp.
St. George, S., Nielsen, E., 2000. Signatures of high-magnitude 19th-century floods Strunk, H., 1997. Dating of geomorphological processes using
in Quercus macrocarpa tree rings along the Red River, Manitoba, Canada. dendrogeomorphological methods. Catena 31, 137–151.
Geology 28, 899–902. Swetnam, T.R., 1993. Fire history and climate change in Giant Sequoia groves.
St. George, S., Nielsen, E., 2003. Palaeoflood records for the Red River, Manitoba, Science 262, 885–889.
Canada, derived from anatomical tree-ring signatures. Holocene 13, 547–555. Tardif, J., Bergeron, Y., 1997. Ice-flood history reconstructed with tree-rings
Stefanini, M.C., 2004. Spatio-temporal analysis of a complex landslide in the from the southern boreal forest limit, western Québec. Holocene 7,
Northern Apennines (Italy) by means of dendrochronology. Geomorphology 63, 291–300.
191–202. Tardif, J., Kames, S., Bergeron, Y., 2010. Spring water levels reconstructed
Stoffel, M., 2005. Spatio-temporal analysis of rockfall activity into forests – results from ice-scarred trees and cross-sectional area of the earlywood vessels in
from tree-ring and tree analyses. Ph.D. thesis, University of Fribourg. GeoFocus tree-rings from eastern boreal Canada. In: Stoffel, M., Bollschweiler, M.,
12, 1–188. Butler, D.R., Luckman, B.H. (Eds.), Tree Rings and Natural Hazards: A State-of-
Stoffel, M., 2006. A review of studies dealing with tree rings and rockfall activity: the-Art. Springer, Dordrecht, Heidelberg, London, New York, NY,
the role of dendrogeomorphology in natural hazard research. Natural Hazards pp. 257–261.
39, 51–70. Tarr, R.S., Martin, L., 1914. Alaskan studies of the National Geographic Society in
Stoffel, M., 2008. Dating past geomorphic processes with tangential rows of the Yakutat Bay, Prince William Sound and lower Copper River regions. National
traumatic resin ducts. Dendrochronologia 26, 53–60. Geographical Society, Washington, DC.
Stoffel, M., 2010. Magnitude-frequency relationships of debris flows – a case study Timell, T.E., 1986. Compression Wood in Gymnosperms. Springer, Berlin, 2150 pp.
based on field surveys and tree-ring records. Geomorphology 116, 67–76. Vaganov, E.A., Hughes, M.K., Shashkin, A.V., 2006. Growth Dynamics of Conifer
Stoffel, M., Beniston, M., 2006. On the incidence of debris flows from the early Tree Rings. Images of Past and Future Environments. Springer Berlin,
Little Ice Age to a future greenhouse climate: a case study from the Swiss Alps. Heidelberg, New York, NY, 354 pp.
Geophysical Research Letters 33, 16404. Vittoz, P., Stewart, G., Duncan, R., 2001. Earthquake impacts in old-growth
Stoffel, M., Bollschweiler, M., 2008. Tree-ring analysis in natural hazards research – Nothofagus forests in New Zealand. Journal of Vegetation Science 12, 417–426.
an overview. Natural Hazards and Earth System Sciences 8, 187–202. Watson, E., Luckman, B.H., 2004. Tree-ring estimates of Mass Balance at Peyto
Stoffel, M., Bollschweiler, M., 2009. What tree rings can tell about earth-surface Glacier for the last three centuries. Quaternary Research 62, 9–18.
processes. Teaching the principles of dendrogeomorphology. Geography Westing, A.H., 1965. Formation and function of compression wood in
Compass 3, 1013–1037. gymnosperms II. Botanical Reviews 34, 51–78.
Stoffel, M., Bollschweiler, M., Beniston, M., 2011. Rainfall characteristics for Wilson, B.F., Archer, R.R., 1977. Reaction wood: induction and mechanical action.
periglacial debris flows in the Swiss Alps: past incidences – potential future Annual Review of Plant Physiology 28, 23–43.
evolutions. Climatic Change 105, 263–280. Winter, L.E., Brubaker, L.B., Franklin, J.F., Miller, E.A., DeWitt, D.Q., 2002.
Stoffel, M., Bollschweiler, M., Butler, D.R., Luckman, B.H., 2010. Tree Rings and Initiation of an old-growth Douglas-fir stand in the Pacific Northwest: a
Natural Hazards: A State-of-the-Art. Springer, Dordrecht, Heidelberg, London, reconstruction from tree-ring records. Canadian Journal of Forest Research 32,
New York, NY, 505 pp. 1039–1056.
 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings 143

Yamaguchi, D.K., 1983. New tree ring dates for recent eruptions of Mount St. Yanosky, T.M., 1983. Evidence of floods on the Potomac River from anatomical
Helens. Quaternary Research 20, 246–250. abnormalities in the wood of flood-plain trees. US Geological Survey
Yamaguchi, D.K., 1985. Tree-ring evidence for a two-year interval between recent Professional Paper 1296. US Geological Survey, Alexandria, VA.
prehistoric explosive eruptions of Mount St. Helens. Geology 13, 554–557. Yanosky, T.M., 1984. Documentation of high summer flows on the Potomac River
Yanosky, T.M., 1982. Effects of flooding upon woody vegetation along parts of the from the wood anatomy of ash trees. Water Research Bulletin 20, 241–250.
Potomac River flood plain. US Geological Survey Professional Paper 1206. US Zoltai, S.C., 1975. Tree ring record of soil movements on permafrost. Arctic and
Geological Survey, Alexandria, VA. Alpine Research 7, 331–340.

Biographical Sketch

Dr. Markus Stoffel is a geomorphologist and currently works at the Institute for Environmental Sciences (ISE),
University of Geneva and at the Institute of Geological Sciences at the University of Berne (Switzerland), where he
directs the Laboratory of Dendrogeomorphology. His research interests are in hydrogeomorphic and Earth-surface
processes, climate change impacts, and dendrogeomorphology. He has authored more than 60 peer-reviewed
papers on hydrologic, geologic, and geomorphic processes in mountain and hillslope environments, re-
constructions of time series on frequency and magnitude of (hydro-)geomorphic processes, dendroecology and
wood anatomy of trees and shrubs, and integrated water resources management. He has (co-)edited books on Tree
Rings and Natural Hazards (2010) and Tracking torrential processes on fans and cones (2011) with Springer and the
Mountain and Hillslope Geomorphology volume of the Treatise on Geomorphology series with Elsevier.
Markus holds BSc and MSc degrees in physical geography, an MSc degree in media and communication
sciences, and a PhD in dendrogeomorphology from the University of Fribourg (Switzerland). Since 2010, he is a
Distinguished Professor in physical geography of the University Babes- -Bolyai (Romania).

Dr. Brian Luckman is Emeritus Professor of geography at the University of Western Ontario in London, Canada
where he has taught for almost 40 years. His major research contributions have been to the study of geomorphic
processes (snow avalanches, talus slopes, rockfall, etc.), environmental change, and Holocene glacier history in
the Canadian Rockies (and latterly in Patagonia). He has also carried out pioneering dendrochronological studies
throughout the Canadian Cordillera and was presented with the Harold C. Fritts Award for Lifetime Achievement
in Dendrochronology by the Tree-Ring Society in 2008. He has published over 150 papers in refereed journals
and edited volumes and co-edited two books, Darkening Peaks (University of California Press) in 2008 and Tree
Rings and Natural Hazards (Springer, 2010).
Brian holds BA and MA degrees in geography from the University of Manchester in England and a doctorate
from McMaster University in Hamilton, Ontario, Canada.

Dr. David R. Butler is the Texas State University System Regents’ Professor of geography, and a University Dis-
tinguished Professor at Texas State University-San Marcos where he has been on faculty since 1997. His research
interests are in the areas of mountain geomorphology, zoogeomorphology, biogeomorphology, and den-
drogeomorphology, focusing his work in the Glacier National Park region of Montana, USA. He has published
over 150 refereed papers in journals and conference volumes, and more than 35 book chapters. He is the author
of Zoogeomorphology – Animals as Geomorphic Agents (Cambridge University Press, 1995 and 2007), and co-editor
of Tree Rings and Natural Hazards (Springer, 2010), The Changing Alpine Treeline – The Example of Glacier National
Park, Montana (Elsevier, 2009), and Mountain Geomorphology – Integrating Earth Systems (Elsevier, 2003). He has
received the G.K. Gilbert Award for Excellence in Geomorphological Research from the Geomorphology Specialty
Group of the Association of American Geographers (AAG), and the Distinguished Career Award and the Out-
standing Recent Accomplishment Award from the AAG’s Mountain Geography Specialty Group.
David holds a BA in geography from the University of Nebraska-Omaha, the MSc in geography from the
University of Nebraska, and the PhD in geography from the University of Kansas.
144 Dendrogeomorphology: Dating Earth-Surface Processes with Tree Rings

Dr. Michelle Bollschweiler holds BSc and MSc degrees in physical geography and a PhD in dendrogeomorphology
from the University of Fribourg (Switzerland). She currently works as a senior scientist at the Institute for
Environmental Sciences (ISE), University of Geneva and at the Institute of Geological Sciences at the University of
Berne (Switzerland). Her research interests include various aspects of tree rings, mass-movement processes, and
climate change impacts. Her recent research focused mainly on the impact of mass-movement processes on tree
rings, the reconstruction of past mass-movement activity using tree rings and on the reconstruction of hydro-
climatic triggers of debris flows. Michelle is a (co-)author of more than 30 peer-reviewed papers and (co-)editor of
two books entitled Tree Rings and Natural Hazards (2010) and Tracking torrential processes on fans and cones (2011),
published with Springer.
 12.10 Tree-Ring Records of Variation in Flow and Channel Geometry
MF Merigliano, University of Montana, Missoula, MT, USA
JM Friedman and ML Scott, US Geological Survey, Fort Collins, CO, USA
r 2013 Elsevier Inc. All rights reserved.

12.10.1 Introduction 145

12.10.2 Tree-Ring Methods in the Riparian Setting 146
12.10.2.1 Dating of Tree Establishment, Injury, and Shifts in Growth Rate 146
12.10.2.2 Dating of Tree Burial 147
12.10.2.3 Tree-Ring Dating in Comparison to Other Methods Used in Riparian Zones 148
12.10.3 Using Establishment Dates of Riparian Pioneer Trees to Determine Flood History and Flood-Plain
Dynamics 148
12.10.3.1 Restrictive Establishment Requirements of Riparian Pioneer Trees in Arid and Semi-Arid Regions 148
12.10.3.2 Reproduction of Riparian Pioneer Trees Depends upon Channel Change Driven by Floods 149
12.10.3.3 The Spatial Pattern of Trees of Different Ages Provides a Record of Channel Change 150
12.10.3.4 Methodology for Determining Area-Age Distributions of Pioneer Trees 151
12.10.3.5 Steady-State Flood Plain with Exponential Area–Age Relation – A Null Model 153
12.10.3.6 Flood Hydrology, Channel Change, and Dominant Discharge 153
12.10.4 Forest Area–Age Distributions in Cottonwood-Dominated Systems: An Illustration of the Use of Tree Rings
to Investigate Fluvial Dynamics 154
12.10.4.1 Regional Climate and Riparian Vegetation 154
12.10.4.2 Small, Semi-Arid Low-Elevation Watersheds in Eastern Colorado 156
12.10.4.3 Large, Humid, High-Elevation Watersheds in the Northern Rocky Mountains 157
12.10.4.4 Intermediate Cases: Watersheds Dominated by Low-Elevation Snowmelt and Thunderstorms 159
12.10.4.5 Summary of Section 12.10.4 161
References 162

Glossary Establishment surface The surface on which a tree

Annual growth increment The width of a tree ring; in germinated. In a flood-plain environment, the
other words, the amount of wood produced by a tree in a year. establishment surface is often buried.
Area age distribution Areas of flood plain or forest Flood ring An abnormal tree ring produced in a flood
occupied by particular ages or age-classes of trees. year.
Area age relation A mathematical representation of the Pioneer tree A tree species that germinates and survives
area age distribution. on sites previously uninhabited by organisms such as plants
Corrasion Wearing away by abrasion. or lichen.
Cross-dating Matching years between cores or cross
sections from the same or different trees.

Abstract

We review the use of tree rings to date flood disturbance, channel change, and sediment deposition, with an emphasis on
rivers in semi-arid landscapes in the western United States. As watershed area decreases and aridity increases, large floods
have a more pronounced and sustained effect on channel width and location, resulting in forest area-age distributions that
are farther from a steady-state exponential relation. Furthermore, forests along three major snowmelt rivers in the northern
Rocky Mountains, USA, have smaller than expected areas of young trees, suggesting that high flows and channel migration
have decreased since the late 1800s.

12.10.1 Introduction

Establishment and growth of riparian trees are strongly in-

Merigliano, M.F., Friedman, J.M., Scott, M.L., 2013. Tree-ring records of
fluenced by stream flow and flood disturbance. As a result,
variation in flow and channel geometry. In: Shroder, J. (Editor in chief),
Butler, D.R., Hupp, C.R. (Eds.), Treatise on Geomorphology. Academic annual rings of riparian trees may record past flow variation
Press, San Diego, CA, vol. 12, Ecogeomorphology, pp. 145–164. and channel change. This chapter reviews the use of tree rings

Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00319-5 145

146 Tree-Ring Records of Variation in Flow and Channel Geometry

by geomorphologists in riparian systems, with an emphasis on

semi-arid landscapes in the western United States. To illustrate
how tree rings are used to approach a classic geomorphic
problem, we review in detail studies of pioneer tree-age dis-
tributions along rivers of the western United States, and test 3 years since
Wolman and Gerson’s (1978) prediction that geomorphic damage
effectiveness of large floods is greater in smaller, more arid
watersheds than in larger more humid watersheds. Combining
published data and new results, we demonstrate that as
watershed area decreases and aridity increases, large floods (a)
have a more pronounced and sustained effect on channel
width and location, resulting in forest area-age distributions
that are farther from a steady-state exponential relation. Fur- 10 years
thermore, we find that forests along three major snowmelt since
damage
rivers in the northern Rocky Mountains, USA, have smaller
than expected areas of young trees, suggesting that high flows
and channel migration have decreased since the late 1800s. 5 years since
(b)
tilting

5 years since
12.10.2 Tree-Ring Methods in the Riparian Setting (c)
tilting

12.10.2.1 Dating of Tree Establishment, Injury, and Shifts

in Growth Rate
Tree-ring dating involves the observation of the number or (d)
character of tree rings to date the occurrence of an event that
Figure 1 Dating flood damage to trees: (a) corrasion scar; (b) split
influenced the establishment or growth of the tree. This group base sprouts; (c) sprouts from tilted parent; and (d) eccentric
of methods takes advantage of the long life of many trees, growth. Reproduced with permission from Hupp, C.R., 1988. Plant
the annual nature of tree rings in many species in temperate ecological aspects of flood geomorphology and paleoflood history. In:
regions, and the fact that the macro- and micro-anatomy of Baker, V.R., Kochel, R.C., Patton, P.C. (Eds.), Flood Geomorphology.
wood are influenced in predictable ways by disturbances of Wiley, New York, NY, pp. 335–356.
interest, such as floods, droughts, and ice impact.
The number of annual rings in a tree core or cross section is may have removed scarred trees, or because old scars may have
an estimate of the number of years since the establishment become externally invisible by healing over the injury. The
of the tree (Stokes and Smiley, 1968). This fact can be used to elevation of the scar may underestimate the peak stage because
determine the time and location of occurrence of conditions the injury happened before or after the peak; conversely,
suitable for tree establishment, which is especially useful the elevation of the scar may overestimate the peak stage
information if such conditions are rare in the landscape. In because the debris was floating above the water surface or
addition, establishment dates can be used to determine a because of the local increase in water surface on the upstream
minimum age for the establishment surface or a maximum side of a tree. Finally, tilting of stems by floods induces a
age of an overlying sediment deposit (Hupp, 1988; Hupp and change in the direction of stem growth resulting in asym-
Simon, 1991; Friedman et al., 2005b). Difficulties encountered metrical ring growth along a stem cross section (Sigafoos,
using this method include lack of annual pattern of wood 1964; Hupp, 1988). The year of onset of this asymmetrical
deposition in some species, false and missing rings (Patterson, growth dates the tilting event.
1959), the difficulty of locating the wood produced in the first The anatomy of individual tree rings may record the occur-
year of life, which may be small, buried, rotten, or consumed rence of past events. Relatively large rings indicate occurrence of
by herbivores (Everitt, 1968, 1995), and the decreased dis- favorable growing conditions. Annual growth increment can be
tinctness of tree rings after burial (Friedman et al., 2005b). influenced by many factors including temperature; availability
When floating ice or flood debris collides with a tree, bark of water, light or nutrients; soil anoxia; death of competitors;
may be damaged or removed, and the cambium will produce and disease or injury (Sigafoos, 1964; Stokes and Smiley, 1968;
a characteristic scar to cover the injury (Shroder, 1980). Hupp, 1988). Because flooding may influence most of these
Counting the rings in the scar reveals the year and sometimes factors, linking annual growth to flooding can be complex.
season of damage (Gottesfeld and Gottesfeld, 1990), and the Flooding may increase (Reily and Johnson, 1982; Dudek et al.,
elevation of the top of the scar may reveal an approximate 1998) or suppress (Martens, 1993) annual ring width. Mech-
stage of the flow that carried the debris, allowing analysis anisms of growth suppression by floods include development of
of flood frequency where gaging station records are lacking soil anoxia from prolonged inundation and physical damage to
(Harrison and Reid, 1967; Hupp, 1988; Figure 1). This ap- the tree. Mechanisms of growth enhancement by floods include
proach may be complicated because of the limited locations reduction of drought, elimination of competitors, and fertiliza-
on the flood plain where moving debris might encounter a tion. Finally, the current year’s growth may be influenced by
tree (Gottesfeld and Gottesfeld, 1990), because large floods conditions in past years.
 Tree-Ring Records of Variation in Flow and Channel Geometry 147

Unusual events may affect the typical seasonal transition from ring abnormalities caused by other factors including
from large to small vessels. A slice through an annual ring insect outbreaks (Sutton and Tardif, 2005).
perpendicular to the direction of stem growth (Figure 2)
shows abundant circles of varying diameter. These circles are
cross sections of the xylem vessels, which transport water up
12.10.2.2 Dating of Tree Burial
from the roots to the leaves. The diameter of xylem vessels
is limited by the tendency of the water column in a vessel Tree rings can be used to relate sediment deposition to a
to break under high tension (Pockman and Sperry, 2000). datable event during the life of a tree, such as germination or
Narrower vessels can sustain higher tensions, while wider flood injury (Sigafoos, 1964; Hupp and Simon, 1991; Strunk,
vessels transport water more rapidly. Therefore, xylem vessels 1997). To determine the time and net amount of sediment
are larger in the spring when water is abundant (Panshin and deposited since establishment, a buried stem is excavated to
de Zeeuw, 1980). Flooding may alter this normal progression the establishment surface and the annual rings at that surface
in a characteristic fashion allowing identification of flood are counted (Alestalo, 1971; Hupp, 1988; Figure 3). Three
rings (Yanosky, 1983; Hupp, 1988; St. George and Nielsen, anatomical characteristics can be used to identify the estab-
2000), but it is important to be able to distinguish flood rings lishment level of a tree. First, the establishment level is com-
monly indicated in the field by the presence of large lateral
roots (Figure 3). Second, because stems have central pith
whereas roots have little or none, the establishment level can
generally be determined in the lab by locating the lowest cross
Bark section that includes pith (Gutsell and Johnson, 2002). Third,
the establishment level is always the oldest cross section of the
tree. If the tree is alive, the establishment year can be deter-
mined by counting back from the present (Hereford, 1984;
Scott et al., 1997). If the tree is dead, the establishment year
1973−2000
may be determined by cross-dating with living trees (Karl-
strom, 1988). The establishment date provides a maximum
age of the overlying sediment deposit and a minimum age of
the underlying deposit. Shortcomings of this method include
the difficulty of precisely locating the establishment level and
reduced precision if there was a long time between deposition
1972 of the overlying and underlying sediments. Finally, changes
1971 in ring anatomy following deep burial obscure the annual
transitions, making ring counts difficult. These burial-induced
1970
changes in anatomy, however, provide additional information

1969

1968

1967

Establishment
1966 level

Figure 2 Tamarisk stem cross section cut 2.8 m below the ground
surface. Following initial burial of 68 cm in 1969, ring-width
decreased, vessel size increased, and annual transitions became less
distinct. After a second burial by 41 cm in 1972, annual transitions
became indistinguishable. Reproduced from Friedman, J.M., Vincent, Figure 3 Excavated tamarisk tree along Chaco Wash, NM. Pink
K.R., Shafroth, P.B., 2005b. Dating floodplain sediments using tree- flags and aluminum tags tie stratigraphic contacts to the tree. Lower
ring response to burial. Earth Surface Processes and Landforms sections of the buried trunks are narrower because burial suppresses
30(9), 1077–1091, with permission from Wiley. stem diameter growth.
148 Tree-Ring Records of Variation in Flow and Channel Geometry

about the timing and depth of sediment deposition as de- have limited vertical precision, and historical measurements
scribed below. and photographs are available only where previous researchers
Sigafoos (1964) observed a burial-induced change in ring have had the foresight to collect them (Malde, 1973; Moody
anatomy of green ash (Fraxinus pennsylvanica) and demon- et al., 1999). The lead isotope 210Pb is delivered to the Earth’s
strated how this could be used to date sediment deposition surface through atmospheric deposition and has a half-life
events. Nanson and Beach (1977) observed a sudden increase of 22.3 years. Therefore, its concentration can be used to
in vessel size within annual rings of buried balsam poplar determine the age of sediment deposited in the last century
trees on point bars along a British Columbia river. They at- (Noller, 2000), but this interpretation can be complicated
tributed this change to burial and dated stratigraphic units in alluvial settings by differences in the concentration of lead
exposed in pits by dating the anatomical change within several in sediment from different sources (Popp et al., 1988). Tree
slabs from a single stem. Rubtsov and Salmina (1983) dated rings, because of their annual precision, present an oppor-
sediment deposits along a river in northern Russia on the basis tunity for improved dating of recent sediment deposition or
of a decrease in annual ring width in buried willow stems. other geomorphic events, such as landslides (Carrara, 2007),
Friedman et al. (2005b) dated complex stratigraphy along debris flows (Bollschweiler and Stoffel, 2010), glacial retreat
the Rio Puerco in New Mexico using observations of both an (Luckman, 2000), or changes in lake levels (Denneler et al.,
increase in vessel size and a decrease in annual ring width 2008).
(Figure 2) in buried stems of tamarisk (Tamarix sp.) and
sandbar willow (Salix exigua). Because the annual transitions
characteristically become unreadable in deeply buried stems,
12.10.3 Using Establishment Dates of Riparian
it may be necessary to cross-date buried sections of the stem
Pioneer Trees to Determine Flood History
with higher unburied sections (Friedman et al., 2005b).
and Flood-Plain Dynamics
Additional complications of this method include the difficulty
of excavating large trees, rot in buried stems, especially in
12.10.3.1 Restrictive Establishment Requirements of
S. exigua, the subtle effects of multiple shallow burials, and
Riparian Pioneer Trees in Arid and Semi-Arid
individualistic response of different stems. For example, the
Regions
effect of burial on ring anatomy may be more pronounced on
small than on large stems (Friedman et al., 2005b). Trees can only occur where there is sufficient water and
light. In humid regions adequate moisture may be generally
available across the landscape, but in dry regions trees may be
restricted to locations with a water subsidy, especially parts of
12.10.2.3 Tree-Ring Dating in Comparison to Other
the riparian zone (Friedman and Auble, 2000). Effects of water
Methods Used in Riparian Zones
scarcity are most extreme when trees are small because of
Tree-ring analysis, even in light of the caveats discussed above, the lack of stored water and deep roots capable of accessing
is one of the most precise techniques available to date events remote water. Riparian trees in arid and semi-arid regions
of the last few hundred years in a riparian setting, especially are typically pioneer species adapted for rapid and distant
where gage or historical records are lacking. Other tools in- dispersal to establishment sites that may be available for
clude 14C, 137Cs, 7Be, optically stimulated luminescence only a short time. This strategy involves production of many
(OSL), 210Pb, flow records, historical artifacts, and repeat cross small widely dispersed seeds and capability of rapid growth
sections and photographs. 14C in organic matter can be used and early maturity (Johnson, 1994; Merritt et al., 2010).
in some circumstances for precise dating of recent sediment, Capability of rapid growth entails a high requirement for
but this method is limited by the oscillations of the curve light, and small initial size limits the ability of seedlings to
relating 14C to calendar years (Trumbore, 2000). The presence reach light and moisture. As a result, seedlings of pioneer
of 137Cs indicates that sediments were deposited after the species compete poorly for light with pre-existing vegetation
onset of nuclear weapons testing in 1950 (Ely et al., 1992), (Johnson et al., 1976; Bazzaz, 1979). Finally, open, wet sites
and the presence of 7Be indicates that sediment was deposited will not produce trees if subsequent disturbances kill the
within the last year (Geyh and Schleicher, 1990), but neither seedlings, a serious limitation because most open wet sites are
of these isotopes date deposits, other than that of the current close to the channel and subject to frequent fluvial disturb-
year, with annual precision. OSL takes advantage of the fact ance. In summary, establishment of riparian pioneer trees in
that quartz and feldspar gradually accumulate charged par- dry regions is limited to unoccupied, moist sites that are
ticles in the absence of light. The quantity of these particles relatively safe from future disturbance (McBride and Strahan,
precisely indicates the elapsed time since the sediment was last 1984; Scott et al., 1996). Such locations are rare on the
exposed to light, presumably during sediment transport. In landscape and may only be produced by specific fluvial events.
the riparian situation, the precision of this method can be The spatial and temporal patterns of adult trees, therefore,
reduced by incomplete bleaching of sediments carried for a provide a record of past occurrence of such events. Dominant
short time at night or in turbid water (Lepper et al., 2002). riparian pioneer trees in arid and semi-arid areas include the
Analysis of flow records can identify specific events that family Salicaceae (willows and cottonwoods), especially in
could have deposited sediment on a particular alluvial surface, the Northern Hemisphere (Stettler et al., 1996; Argus, 1997);
but does not resolve which sediment was deposited by each the family Tamaricaceae (especially genus Tamarix) from Asia,
event. Historical artifacts are typically scarce, and can only Europe, and Africa, but now dominant along rivers in much of
give an approximate age. Stereo pairs of aerial photographs the western US and Australia (Baum, 1978); and river red gum
 Tree-Ring Records of Variation in Flow and Channel Geometry 149

(Eucalyptus camaldulensis) and paperbark (Melaleuca argentea) 2004). Scott et al. (1996) described three common processes
in Australia (House, 1997; Fielding et al., 1997). promoting establishment of riparian pioneer trees: channel
narrowing, meandering, and flood deposition (Figure 4).
Although a single location may show complicated combin-
12.10.3.2 Reproduction of Riparian Pioneer Trees ations of these and other processes (Cooper et al., 2003),
Depends upon Channel Change Driven by Floods their relative importance varies predictably among rivers
(Scott et al., 1996).
Establishment of riparian pioneer trees depends upon channel During channel narrowing, establishment occurs on por-
change, usually driven by floods (Bradley and Smith, 1986; tions of the channel bed safe from future disturbance because
Stromberg et al., 1991; Merritt et al., 2010). Along a static of the inability of a stream to rework its entire bed (Johnson,
channel, locations that are moist and open enough for ger- 1994; Figure 4). This process is especially important in
mination are unsuitable for establishment because of frequent streams with large fluctuations in width promoted by vari-
flood disturbance. Channel change provides sunny, moist sites ability in flow and banks dominated by noncohesive and
that ate relatively safe from future disturbance. At the easily eroded sand (Schumm and Lichty, 1963). For example,
decadal scale, this channel change is usually driven by flood- along streams in eastern Colorado, channel change is driven
ing, which transports sediment, removes vegetation, and by infrequent floods that greatly widen the sandbed channels,
moves or widens the channel. Flow conditions during the year and tree establishment occurs in association with post-flood
of establishment vary widely, however, depending on the channel narrowing lasting roughly two decades (Friedman
specific processes leading to establishment (Lytle and Merritt, and Lee, 2002; Katz et al., 2005). Damming of braided rivers

Germination
First year: Peak First year: Germination
normal peak low peak
Base Peak
Base

Flooding Drought Flooding Drought

mortality mortality mortality mortality

Future years: Future years:

Peak
no change continued
Base low peaks Peak
Base
No tree recruitment Tree recruitment on
(a) (b) former channel bed

Germination
Germination
First year: First year:
Peak Peak
moderate flood peak
peak
Base Base

Flooding Drought Flooding Drought

mortality mortality mortality mortality
Future years:
continued Future years:
channel normal peaks
Peak Peak
meandering
Base Base

Tree recruitment on Tree recruitment on flood

(c) accreting point bars (d) deposition surfaces
Figure 4 Hydrogeomorphic control of cottonwood recruitment: diagrammatic representations of cottonwood seed germination, early seedling
mortality, and tree recruitment in relation to annual high and low flow lines along a riparian cross section. Four idealized situations are depicted:
(a) little or no tree recruitment in the absence of inter-annual flow variability and channel movement, (b) channel narrowing with recruitment on
the former channel bed, (c) recruitment on point bars of a meandering river, and (d) tree recruitment at high elevations associated with
infrequent floods and no channel movement. Reproduced from Auble, G.T., Scott, M.L., 1998. Fluvial disturbance patches and cottonwood
recruitment along the upper Missouri River, MT. Wetlands 18(4), 546–556, with permission from SWS.
150 Tree-Ring Records of Variation in Flow and Channel Geometry

in the Great Plains of the USA decreased flood magnitude and tree establishment dates to map channel change along the
sand transport, resulting in extreme narrowing and establish- Snake (Figures 5 and 6) and Redwater rivers, and Merigliano
ment of extensive pioneer forests (Williams and Wolman, and Polzin (2003) used this method to map changes along the
1984; Johnson, 1994; Friedman et al., 1998; Graf, 2006). Yellowstone River, Montana.
Channel narrowing may also be driven by climate change It is important to be aware that the above-ground age of
(Schumm, 1969; Gottesfeld and Gottesfeld, 1990), changes a pioneer tree may underestimate the age of the underlying
in land management (Nadler and Schumm, 1981), or intro- flood plain. The most common problem is underestimating
duction of new plant species adept at trapping sediment or tree age by not finding wood produced when the tree was
retarding erosion (Nevins, 1969; Graf, 1978). young. Overbank deposition after tree establishment can bury
Meandering streams are characterized by low width-to- many years of growth. In addition, if a tree is cored above the
depth ratios, sediment load dominated by suspended sediment, ground, the number of years of growth required to reach the
and progressive channel migration through meander-loop coring height must be accounted for. This time cannot always
extension (Hickin and Nanson, 1975). Point bars migrate be precisely estimated because herbivory or flood damage
outward and downstream, and most sediment deposition can delay growth to coring height – sometimes by decades
on point bars occurs during relatively frequent high flows (Everitt, 1968). Therefore, determining the precise age requires
(Wolman and Miller, 1960; Bradley and Smith, 1986). During excavation to the original establishment surface (Scott et al.,
channel meandering, establishment occurs on point bars 1997). Because ring transitions may be more difficult to ob-
that are relatively safe from disturbance because of channel serve in buried stems (Friedman et al., 2005b), it may be ne-
migration away from the point bar (Everitt, 1968; Miller and cessary to cross-date buried cross sections of the stem with
Friedman, 2009; Figure 4). Therefore, forests produced as cross sections collected above the surface. Finally, the tree may
a result of meandering generally take the form of a series of have become established on an old surface that was surficially
parallel, arcuate even-aged bands of trees, increasing in age disturbed but not totally replaced by a flood. In this case, the
with distance from the channel and parallel to the direction of tree accurately dates the disturbance that allowed establish-
flow at the time of establishment (Nanson and Beach, 1977; ment, but not the formation of the fluvial surface. Because of
Noble, 1979). The establishment surface is a young point bar, these issues, the age of a tree growing on a flood plain is a
intermediate in elevation between the channel bed and the minimum age for the underlying surface or causative flood.
flood plain. Because young trees are frequently battered and Aging trees at the establishment surface improves pre-
occasionally knocked over by floods, it may be difficult to cision, but does not necessarily date formation of that surface
locate the original establishment surface of a tree along a with annual precision. Along a laterally constrained section of
meandering river (Everitt, 1968). A variant of the meandering the Missouri River, Montana, Scott et al. (1997) found a lag of
process occurs on anabranching channels, where channels 0–2 years between flood disturbance and establishment dates
migrate and create mid-channel bars as well as point bars indicated by ring counts of cottonwood trees at the excavated
that can develop into stable, vegetated islands (Nanson and establishment surface; aging trees at the ground surface de-
Knighton, 1996). creased age estimates by a mean of 5.1 years and a maximum
In the weeks following an extreme flood, pioneer tree es- of 34 years (Scott et al., 1997). On a young island along a
tablishment may occur on flood deposits that are high enough braided reach of the Salmon River in Idaho, ages for adjacent
to be relatively safe from future disturbance (Baker, 1990; pairs of excavated black cottonwood (Populus trichocarpa)
Stromberg et al., 1991; Auble and Scott, 1998; Figure 4). Be- seedlings or saplings varied up to 4 years (Merigliano,
cause the establishment elevation is at the level of the flood 1998). Along the meandering Little Missouri River, tree ex-
plain, trees may experience relatively little additional sediment cavations (Everitt, 1968) and comparisons with sequential
deposition (Scott et al., 1997). Flood deposition and erosion aerial photographs (Miller and Friedman, 2009) showed a
occur along most streams, but may be especially important for typical lag of a decade or two between surface formation and
pioneer tree establishment where lateral channel movement is tree establishment determined by ring counts made above the
constrained by a narrow valley (Scott et al., 1996, 1997). ground surface.
Tree age can be rapidly estimated using a local relation
between age and diameter (Howe and Knopf, 1991; Stromberg
and Patten, 1992). Because of the considerable scatter in
12.10.3.3 The Spatial Pattern of Trees of Different Ages
cottonwood diameter-age relations, however (Figure 7), this
Provides a Record of Channel Change
method is relatively imprecise (Mahoney et al., 1991). Stem
As sediment storage units, flood plains preserve an extended cross sections yield more accurate ages than increment cores
record of climatic and sedimentary conditions, and the ages because false or missing rings are more easily detected, but
and areas occupied by riparian pioneer trees provide an inte- collecting cross sections from large trees involves considerable
grated measure of that legacy. For example, along the me- time and disturbance. In a unique study, Mahoney et al.
andering Little Missouri River in North Dakota, USA, Everitt (1991) paired increment core and cross-section ages for cot-
(1968) reconstructed a 250-year history of channel migration tonwood trees older than 20 years, and found that core ages
by coring trees and mapping the locations of patches of dif- were typically about 2–7% less than cross-section ages.
ferent ages. Gottesfeld and Gottesfeld (1990) used patterns Study objectives and intended analysis needs should guide
of tree ages along the Morice River, British Columbia, to the desired level of accuracy, as effort increases substantially
document narrowing in the mid-1800s in association with the with aging accuracy, especially in settings with thick overbank
end of the neoglacial period. Merigliano (1996, 2007) used deposits and flood-trained or damaged stems. Stem sections
 Tree-Ring Records of Variation in Flow and Channel Geometry 151

(a)

(b)

Figure 5 Channel migration and riparian vegetation development along the South Fork Snake River near Heise, Idaho. View is upstream. The
upper scene (a) is c. 1908–1911 (W. B. Heroy #28, US Geological Survey); the lower scene (b) was photographed from Heroy’s standpoint on
18 June 1992 by M. Merigliano. The area bounded by canyon walls within (b) shows in the right third of the map in Figure 6. The large island
complex in (b) center, aged in Figure 6, originated in part during the 1918 flood (1472 m3 s–1) whereas the mid-channel island in (b) originated
in 1927 (1699 m3 s–1). The 2-year flood magnitude for the un-dammed annual flood series is 730 m3 s–1.

from mature trees are commonly bulky and can exceed 100 kg, can be weighted by the area they represent to produce an area-
and excavating, excising, and collecting stem sections from age distribution (Figure 6).
remote settings presents logistical challenges. In addition to Selecting the closest tree to a set of randomly selected
individual tree ages, the spatial pattern of tree ages can be points requires the least amount of field work and training.
complex and require ages from numerous trees, so one must The point selection can be done in advance of field work.
consider the tradeoff between individual tree accuracy and the Points can be selected by means of a dot grid (Friedman and
spatial resolution of those ages. Lee, 2002) or by an automated point selection routine in a
geographic information system. Transects may also be used,
but care must be taken to ensure that they produce a repre-
12.10.3.4 Methodology for Determining Area-Age
sentative sample (Kondolf and Micheli, 1995; Scott et al.,
Distributions of Pioneer Trees
1997). The flood plain can be divided into two or more classes
In order to determine the areas occupied by flood-plain pat- with different sampling intensities (Friedman and Lee, 2002).
ches of differing age (the area-age distribution), trees need to For example, if an area of a certain apparent age class is of
be selected to represent specific areas of flood plain. Two special interest, this area can be delineated, and sampling in-
general approaches have been used: (1) selection of the closest tensity within it can be increased. To correct for different
tree to a set of randomly selected points and (2) delineation of sampling intensities in different strata, area can be weighted
relatively even-aged patches, followed by selection of trees by sampling intensity in the final area-age distribution
within each patch. Once selected trees have been aged, they (Friedman and Lee, 2002).
 Cottonwood age classes

152
XI 201−250
X 151−200

Tree-Ring Records of Variation in Flow and Channel Geometry

IX 126−150
VIII 101−125
VII 81−100
VI 66−80
V 51−65
IV 36−50
III 21−35
II 11−20
I 1−10

0 2 4 6 8 10
Hectares

Water

Exposed channel bed at 6000 cfs

R Riparian grass and forbs

W Sandbar willow

MS Mixed riparian shrub, F = burned

S = sagebrush/grass, G = upland grass,
J = Juniper, H = goldenaster, E = silverberry

200 m

Figure 6 Flood-plain aging process for a sampled section of the South Fork Snake River. The progression begins with an aerial photo image which guides selection of tree aging within putative
even-aged patches. The spatial pattern of tree ages and geomorphic features define the mapped patch boundaries using a pre-determined set of age classes, and the age distribution is obtained from
mapped patch areas within each age class. The sampled unit depicted in the map is 102 ha, so the graphed areas approximate percentages of the flood plain-channel system. The aged-tree map
corresponds to the right-hand portion of the larger flood plain map. Downstream is to the left. Ages are as of 1993.
 Tree-Ring Records of Variation in Flow and Channel Geometry 153

300 age (Everitt, 1968):

250 Fðt Þ ¼ F0 ekt ½1

200 where F0 is the area of flood plain formed each year normal-
Age (years)

ized to the total area sampled, in m2 ha–1 yr–1; F(t) is the area
150 remaining at time t, also in m2 ha–1 yr–1; t is the time since
formation, or equivalently, the flood-plain age in years; and
100 k is the rate of flood-plain destruction, expressed as the pro-
portion of flood plain destroyed each year, in yr1. For con-
50 venience, flood-plain patches can be binned in age classes, and
the mid-points of these classes can be used in curve fitting.
0 Under stable conditions, Little’s law (Little, 1961) applies,
0 10 20 30 40 50 60 70 80 90 and the mean residence time (W) for flood-plain sediments is
Diameter at 1.37 m aboveground (cm)
W ¼ 1=k ½2
Figure 7 Relation between stem diameter and tree age for
narrowleaf cottonwood (Populus angustifolia) along the South Fork
Snake River in eastern Idaho. Data from Merigliano, M.F., 1996. while the turnover time T (the time to recycle a particular
Ecology and management of the South Fork Snake River cottonwood proportion of flood plain, P) is
Forest. Bureau of Land Management, Idaho State Office, Technical
T ¼ ðlnð1  P ÞÞ=k ½3
Bulletin 96-9, Boise, ID, 79 pp.

where ln is the natural logarithm.

Apparently even-aged patches of riparian trees can be
The parameters F0 and k indicate fluvial activity. Flood
delineated using aerial photo-interpretation and field obser-
plains with large F0 and k have high channel migration rates
vations of tree characteristics and geomorphic breaks. On
and rapid recycling of flood plain driven by large floods or
aerial photographs, vegetation patterns indicate geomorphic
erodible banks. For example, the high values of F0 and k for
surfaces including scroll bars and abandoned channels
the Little Missouri River (Figure 8) result from the erodible
(Figure 6). Pioneer species such as cottonwood and willow
sand banks and a history of large floods in the last 150 years.
often establish at distinct elevation zones, and their spatial
Along the Yellowstone River, abundant cobbles and boulders in
patterns indicate micro-topography and the originating flow
the channel boundary result in relatively low F0 and k, and
direction. Tree characteristics such as stem diameter, bark
confined or entrenched reaches have even lower values.
thickness, tree height, and branching pattern indicate relative
So far, forest area has been assumed equivalent to flood-
ages, and with calibrating age data, actual ages can be esti-
plain area, but typically flood plains are not completely
mated. For example, as cottonwood ages, bark generally be-
covered in forest. This problem can be addressed either by
comes thicker and rougher, this characteristic extends higher
ignoring the nonforested flood plain or by expanding forest-
up the tree, branches become more horizontal, and the crown
cover data systematically to represent the nonforested areas.
top flattens (Hickin and Nanson, 1975). Once patches have
The steady-state exponential decay relation [1] is a null
been delineated, trees can be selected within each patch, either
model representing the expected area–age distribution for a
at random or systematically, and the ages of these trees can be
flood plain in steady state. Deviations from this steady-state
used to refine the original delineation. Tree selection should
exponential distribution provide information about past flow
avoid root or stem sprouts, which underestimate the age of the
variations. For example, if a large flood in a meandering
underlying surface. Ideally, all aged trees within delineated
channel temporarily increases the rates of channel migration
patches would be exactly the same age, but this is unusual in
and tree establishment, the area–age distribution will include
our experience for the reasons stated earlier. However, if the
more area for the flood year than predicted by the exponential
local forest became established after a flood-plain forming
distribution. Conversely, if a period of low peak flows slows
event, the patch age distribution will approach a maximum
channel migration and forest establishment along a me-
representing the surface age, and similar patches elsewhere on
andering channel for a decade, the area–age distribution will
the flood plain will approach the same age.
include less area for the low-flow decade than predicted by the
exponential distribution.
12.10.3.5 Steady-State Flood Plain with Exponential
Area–Age Relation – A Null Model 12.10.3.6 Flood Hydrology, Channel Change, and
Dominant Discharge
The area–age distribution provides information about flood-
plain dynamics and flow variation. The area of flood plain The relation between flow and riparian pioneer tree establish-
of a given age is the difference between the initial area formed ment is conditioned by watershed characteristics. The area–age
and the areas destroyed by subsequent floods and other distribution of a riparian forest will be closest to the steady-
disturbances. If the annual rates of flood-plain formation and state exponential distribution where peak flows vary little from
destruction are constant and equal, and if destruction is uni- year to year. The area–age distribution is likely to be far from
formly or randomly distributed across the flood plain, then the steady state in systems dominated by the long-term effects
flood-plain (or forest) area will decrease exponentially with of rare floods. Year-to-year variation in peak flows is strongly
154 Tree-Ring Records of Variation in Flow and Channel Geometry

100

Normalized flood-plain area (m2 ha−1 yr−1)

75

Lit
tle
Mi

Bi
50 ss

gh
ou

or
SF ri k

n
Sn =0

k
ake

=
k= .01

0.
0.0 00

01
074

75
25 Yellowstone W
Yellowston GB k = 0.003
e CWGB k 2
= 0.0026

Yellowstone CCT k =
0.0074
0 Yellowstone E k = 0.0023

0 50 100 150 200 250 300

Age (years)
Figure 8 Exponential area–age relations (F(t) from eqn [1]) for seven river reaches in the western USA. Normalized flood-plain area is the area
of flood plain per year divided by the total flood-plain area sampled at the site. F0 for each curve is the y intercept. Data sources are: Little
Missouri River (Everitt, 1968), South Fork Snake River (Merigliano, 1996), Bighorn River (Akashi, 1988), and Yellowstone River (Merigliano and
Polzin, 2003). The Yellowstone River reach strata are: WGB ¼ wandering gravel bed, CWGB ¼ confined wandering gravel bed, CCT ¼ confined
course texture, and E ¼ Entrenched (Merigliano and Polzin, 2003; Nanson and Croke, 1992). The curves for the Yellowstone River are for reaches
with similar water and sediment flux. We include the channel in calculating flood-plain area; therefore, the F0 value reported here for the Little
Missouri River is slightly lower than that in Everitt (1968). For regulated rivers post-dam data are excluded. For the Bighorn River, we used data
from 1944 (Akashi, 1988), but we excluded a stable low terrace from calculations of the total flood-plain area.

related to watershed characteristics, especially watershed size lowlands with low precipitation and high potential evapor-
and aridity (Baker, 1977). In large, humid watersheds, peak ation. Large watersheds are commonly dominated by mon-
flows integrate precipitation and runoff over large areas and tane snowmelt, which integrates months of winter weather
long times, and flows between the peaks are relatively high; as a across the entire montane portion of the watershed, resulting
result, year-to-year variation in peak flows is small and flood- in broad sustained annual peak flows lasting weeks or more
plain geometry is dominated by frequent events of modest with modest interannual variation (Table 1). By contrast, in
magnitude (Wolman and Miller, 1960). In small, dry water- small low-elevation watersheds, annual precipitation is rela-
sheds, peak flows result from local short-term extreme pre- tively low, snowfall is relatively slight, snow does not accu-
cipitation events; as a result, year-to-year variation in peak mulate throughout the winter, and peak flows are dominated
flows is large, and because flows between peaks in an arid or by the effects of local intense monsoonal thunderstorms
semi-arid watershed are relatively low, the effects of extreme lasting for minutes to hours. Peak flows resulting from these
events are persistent (Wolman and Gerson, 1978; Baker, thunderstorms are high in magnitude and short in duration
1977; Graf, 1988). Therefore, geomorphic theory predicts that with large interannual variation (Table 1). Watersheds in this
area–age distributions of riparian pioneer forests should be region, therefore, provide an opportunity to test predictions of
closer to a steady-state exponential distribution in large humid the effects of watershed size and aridity on patterns of channel
watersheds than in small dry watersheds (Friedman and Lee, change and forest development.
2002). This prediction can be tested by analysis of area–age Cottonwood (Populus spp.) is the most abundant riparian
distributions of cottonwood (Populus spp.) in watersheds of tree in the interior western United States (Friedman et al.,
contrasting size and aridity in the western United States; the 2005a). Pioneer attributes of this genus include production of
following section presents such a test. abundant, small, short-lived, and widely dispersed seeds
(Fenner et al., 1984; Mahoney and Rood, 1998; Auble and
Scott, 1998), high requirements for light and moisture
12.10.4 Forest Area–Age Distributions in
(Johnson et al., 1976; Johnson, 1994), and formation of even-
Cottonwood-Dominated Systems: An
aged stands (Scott et al., Johnson, 1997; Bradley and Smith,
Illustration of the Use of Tree Rings to
1986). North American Populus is capable of reproducing
Investigate Fluvial Dynamics
vegetatively by root sprouts in the absence of disturbance, but
this ability varies by species. Populus deltoides (including eastern
12.10.4.1 Regional Climate and Riparian Vegetation
cottonwood, plains cottonwood, and Fremont cottonwood)
In the interior western United States, mountains with high rarely forms root sprouts, but Populus balsamifera (balsam
precipitation and low potential evaporation are surrounded by poplar), Populus angustifolia (narrowleaf cottonwood), and
Table 1 Flow characteristics at gaging stations near location of tree aging

Gage name Period of record Gage Drainage Watershed Gage Mean daily Q Percent Q100 (m3 s1) Q2 (m3 s1) Q100/Q2
number area (km2) maximum elevation (m) m3 s1 snowmelt
elevation (m) peaks

Tree-Ring Records of Variation in Flow and Channel Geometry

Yellowstone River near Livingston, MT 1897–2009 06192500 9 194 43500 1385 106.5 100 1017 578 1.8
Green River near LaBarge, WY 1947–2009 09209400 10 123 43500 1987 45.3 100 562 244 2.3
Snake River near Heise, ID 1911–2009 13037500 14 892 43500 1528 197.6 97 1478 623 2.4
Little Missouri R. nr Watford City, ND 1935–2009 06337000 21 515 1250 588 15.3 69 2093 428 4.9
Redwater River at Circle, MT 1929–2004 06177500 1 416 1105 730 0.3 62 306 24 12.7
S. Fork Republican River near Idalia, CO 1935, 1951–1979 06825000 3 366 1676 1122 0.9 14 2611 92 28.4
Arikaree River at Haigler NE 1932–2009 06821500 4 246 1707 991 0.5 18 1317 58 22.6
Kiowa Creek at Kiowa, CO 1956–1965 06758200 287 2286 1935 0.1 30 2628 22 119.3

Q is discharge, and Q100 and Q2 are floods with recurrence intervals of 100 and 2 years. Percent snowmelt peaks is the percentage of peak instantaneous annual discharges resulting from snowmelt. Gaging stations are arranged by increasing
Q100/Q2.
Source: Data are from the National Water Information System (2010) and StreamStats (2010).

155
156 Tree-Ring Records of Variation in Flow and Channel Geometry

Populus tremuloides (aspen) all form abundant root sprouts, floods exceeding 1500 m3 s–1 in the last 100 years (Friedman
degrading the even-aged character of stands (Gom and Rood, et al., 1996), and these floods have caused major changes
1999). Cottonwoods are commonly the most abundant ripar- in channel width and alignment (Figures 9–11); (Osterkamp
ian tree on the flood plain, and the trees may live for hundreds and Costa, 1987; Friedman and Lee, 2002). In the decades
of years, preserving information on flood history from long between large floods, peak instantaneous annual discharge is
before the advent of stream gages in the region around 1900. much lower, and therefore, the ratio of the 100-year to the
Old stands are eventually replaced by shade-tolerant trees, 2-year peak magnitudes is large (Table 1; Lewin, 1989). As a
shrubs, or grasses (Johnson et al., 1976; Boggs and Weaver, result, geomorphic work between extreme floods is limited,
1994; Friedman and Lee, 2002). and the effects of major floods are long lived. Response of
these channels to flooding appears to be typical of sandbed
channels in semi-arid to arid regions. Channel widening and
12.10.4.2 Small, Semi-Arid Low-Elevation Watersheds in
migration during floods are typically followed by decades of
Eastern Colorado
narrowing (Figures 9–11). The sandy flood-plain sediments
The Plains of northeastern Colorado are semi-arid with a form noncohesive, easily eroded banks that allow large fluc-
continental climate characterized by low precipitation, high tuations in channel width (Schumm and Lichty, 1963).
daily and annual temperature ranges, and high potential Sandbed streams in northeastern Colorado illustrate the
evapotranspiration rates. In this region snow generally does effects of aridity on forest age structure. Friedman and Lee
not accumulate throughout the winter, and floods occur (2002) and Katz et al. (2005) examined accessible forests in
between mid-May and late September resulting from local, six locations along West Bijou, Kiowa, and Coal creeks in the
intense thunderstorms that can produce as much as the mean Colorado Piedmont, and nine locations along the South Fork
annual precipitation in a few hours (Hansen et al., 1978; Republican and Arikaree rivers in the High Plains, all sandbed
Osterkamp and Costa, 1987). Many streams have experienced streams in northeastern Colorado (Figures 9–11). Elevation of

Establishment year

1980 1960 1940 1920 1900 1880 1860

600
300 Upper West Bijou Creek

0
0 20 40 60 80 100 120 140
900
600 Lower West Bijou Creek
300
0
Normalized forest area (m2 ha−1 yr−1)

0 20 40 60 80 100 120 140

600
Upper Kiowa Creek
300
0
0 20 40 60 80 100 120 140
600
Lower Kiowa Creek
300
0
0 20 40 60 80 100 120 140

400 Upper Coal Creek

200
0
0 20 40 60 80 100 120 140

400
Lower Coal Creek
200
0
0 20 40 60 80 100 120 140
Age (years)
Figure 9 Area–age distributions of plains cottonwood (Populus deltoides subsp. monilifera) along six streams in eastern Colorado. Normalized
forest area is the area of forest per year divided by the total forest area sampled at the site. Arrows indicate years of known major floods.
 Tree-Ring Records of Variation in Flow and Channel Geometry 157

by infrequent local events, channel width and tree age distri-

1937 butions vary greatly over time and among sites (Figure 9).
A steady-state forest area–age relation is rarely approached on a
given stream.
Water management, tree planting, and climate change have
all influenced forest patterns in eastern Colorado, but these
human-induced factors are not necessarily responsible for the
observed deviations from steady state. For example, to exam-
ine the effects of flow regulation on sand-bed streams in
eastern Colorado, Katz et al. (2005) studied the riparian forest
on three river segments, the dam-regulated South Fork Re-
publican River downstream of Bonny Dam, the unregulated
1993 South Fork Republican River upstream of Bonny Dam, and the
nearby unregulated Arikaree River. Although Bonny Dam
significantly reduced peak and mean discharge downstream
since 1951, there was little difference in forest structure be-
tween the regulated and unregulated segments. On all river
segments, the riparian forest was dominated by plains cot-
tonwoods that became established during a period of channel
narrowing beginning after the 1935 flood of record and
ending by 1965 (Figures 10 and 11). The lack of contrast in
forest structure between regulated and unregulated reaches
resulted primarily from the fact that no large floods occurred
Figure 10 Channel narrowing on the South Fork Republican River,
on any of the study segments since dam construction.
Colorado, upstream of Bonny Reservoir. A flood in 1935, 2 years
before the upper photograph, greatly widened the channel. In the Although Bonny Dam, whose storage capacity is 95% of mean
following three decades, the channel narrowed and plains cottonwood annual inflow, had the potential to significantly influence
(Populus deltoides subsp. monilifera) became established on the downstream riparian ecosystems, this influence had not been
former channel bed. Modified from Katz, G.L., Friedman, J.M., Beatty, expressed, and may never be if a large flood does not occur
S.W., 2005. Delayed effects of flood control on a flood-dependent within the lifetime of the dam.
riparian forest. Ecological Applications 15(3), 1019–1035, with
permission from GSA.

12.10.4.3 Large, Humid, High-Elevation Watersheds in the

these sites ranges from 1010 to 1783 m, drainage area ranges
Northern Rocky Mountains
from 110 to 4727 km2 , and annual precipitation ranges from
about 36 to 47 cm (Colorado Climate Center, unpublished The northern Rocky Mountains in the coterminous United
data). All of these watersheds are entirely within the Plains States form a humid core within a broader semi-arid region.
with no montane contributing area. Lithology is dominated Snow is the dominant precipitation, and snowmelt drives
by sandstone (Dawson Arkose) on the Colorado Piedmont streamflow. The Yellowstone, Green, and Snake rivers, three
and loosely to tightly cemented gravel (Ogallala Formation) large rivers draining the Rocky Mountains, all originate in an
on the High Plains. The study reaches occupy valleys more area of the Rocky Mountains in southern Montana, western
than 0.5 km wide consisting of aeolian deposits, rarely inun- Wyoming, and eastern Idaho. Merigliano and Polzin (2003)
dated Quaternary alluvial terraces, and frequently inundated examined the Yellowstone River from Gardiner to Springdale,
alluvial deposits. Bedrock is exposed locally in banks but Montana. Glass (2002) studied the Green River, Wyoming,
rarely in stream beds. The streams in this set range from upstream and downstream of Fontenelle Reservoir. Merigliano
ephemeral to perennial, but even the perennial sites have (1996) mapped forest ages along the South Fork Snake River
mean annual discharges below 1 m3 s–1. The primary land use near Heise, Idaho. The study reaches traverse the semi-arid
is livestock grazing. terrain downstream of the montane zone, and have wandering
The riparian forests along these streams are dominated by gravel-bed channels (Desloges and Church, 1989) bounded
plains cottonwood (P. deltoides subsp. monilifera). Reproduction by glacio-fluvial sediments. Flood plains along all three rivers
of cottonwood trees along these eastern Colorado streams has are dominated by narrowleaf cottonwood. Because mean an-
occurred mostly in the former channel bed during periods of nual precipitation is high in the mountains, the mean daily
channel narrowing beginning after floods in 1935 or 1965 and discharge of these rivers is large (Table 1). Peak annual dis-
continuing for as long as two decades (Figure 9). Thus, cot- charge always results from snowmelt, which integrates months
tonwood establishment is related to low flows at the timescale of winter weather across the montane portion of the water-
of a year, but to extreme high flows at the timescale of decades. shed, resulting in broad sustained annual peak flows lasting
At the two Coal Creek sites, which have not experienced major for weeks with small interannual variation. The ratio of the
floods in the last 80 years, little channel change has occurred, 100-year and 2-year flood magnitudes in these systems is only
cottonwood reproduction has been limited, tree density has about 2 (Table 1), indicating that extreme floods are not im-
declined, and succession to grassland is occurring. Because portant in controlling channel change. Because extreme flows
channel change and tree reproduction in this region are driven are not important in these systems, the forest area–age
158 Tree-Ring Records of Variation in Flow and Channel Geometry

Establishment year
1980 1960 1940 1920 1900
500
400 Arikaree River
300 Flood
200
100

Normalized forest area (m2 ha−1 yr−1)

0
0 20 40 60 80 100
800
Flood
600
S. Fk. Republican River,
400
unregulated
200

0
0 20 40 60 80 100
1000 Dam construction
Flood
800
600 S. Fk. Republican
400 River, regulated
200
0
0 20 40 60 80 100
Age (years)
Figure 11 Area–age distributions of plains cottonwood (Populus deltoides subsp. monilifera) in cross sections along the unregulated Arikaree
River and regulated and unregulated sections of the South Fork Republican River, Colorado (three cross sections per site). Normalized forest
area is the area of forest per year divided by the total forest area sampled at the site.

distribution should be close to a steady-state exponential regulated by Jackson Lake and Palisades Reservoirs, completed
pattern if the rate of flood-plain formation has not changed in 1917 and 1956. Jackson Lake dam enlarged a piedmont
over time. glacial lake, so sediment supply to downstream reaches was
The area–age distributions at all three sites are roughly bell- not changed. Both reservoirs are primarily for irrigation
shaped, peaking in the late 1800s on the Yellowstone and storage, but flood-control objectives limit peak flows to about
Green rivers and in the early 1900s on the South Fork Snake the pre-dam, mean-annual flood magnitude at Heise. The
River (Figures 12–14). The decrease in forest area with flood-control effects from these reservoirs are proportional
increasing age to the right of the peak at all three rivers ap- to their active storage capacities, which are 17% and 24%
proximates the exponentially decreasing area–age distribution of the average inflow at Heise; however, the decrease in
expected in a system close to steady state. The decrease in channel activity suggested by the forest area–age distribution
forest area with decreasing age to the left of the peak at all began 30–45 years prior to completion of the reservoir. In
three sites indicates a strong decrease in the rate of flood-plain summary, the decrease in river activity in the last century
formation in the last century, a conclusion that has been indicated by the forest area–age relations of these three rivers
verified by the small amount of channel change since the first is unlikely due to flow regulation alone. Analyses of the oldest
detailed map of the Upper Green River was prepared by the stream gage records along the Green River have demonstrated
General Land Office in 1890 (G.T. Auble and M.L. Scott, un- a decrease in peak flows beginning in the 1920s prior to
published result). The dramatic decline in flood-plain activity construction of major dams (Allred and Schmidt, 1999; Glass,
suggests a decrease in peak flows since the late 1800s or early 2002).
1900s, possibly enhanced by flow regulation on the Snake Long-term reconstructions of annual flow volume based on
River, channelization, and woody debris removal. correlations with ring widths of upland conifer trees show
Such a decrease in flows could be explained by flow periods of rapid growth (and inferred high flow volume)
regulation or climatic fluctuation. The Yellowstone River around 1850, 1870, and 1920 in the Upper Colorado River
above Springdale has negligible flow regulation. The upper (Woodhouse et al., 2006) and around 1880 and 1915 in the
Green River is regulated by Fontenelle Reservoir, but this Yellowstone River (Graumlich et al., 2003), but these records
reservoir was not completed until 1964, and forest area–age do not show slower growth (inferred lower flow volumes)
distributions were similar upstream and downstream of this in the 1900s compared to the previous several centuries.
reservoir (Glass, 2002). The South Fork Snake River is These results suggest that our inferred decrease in peak flows
 Tree-Ring Records of Variation in Flow and Channel Geometry 159

Establishment year
2000 1975 1950 1925 1900 1875 1850 1825 1800 1775 1750 1725 1700 1675 1650

Normalized flood-plain area (m2 ha−1 yr−1)

40
Upper Yellowstone River
Confined wandering gravel bed
Wandering gravel bed
30

20

10

0 25 50 75 100 125 150 175 200 225 250 275 300 325 350
Age (years)
Figure 12 Area–age distributions of narrowleaf cottonwood (Populus angustifolia) along the upper Yellowstone River between Gardner and
Springdale, Montana. Normalized flood-plain area is the area of flood plain per year divided by the total flood-plain area sampled at the site.
Steady-state models are Ft ¼ 25.8e(–0.0026t) for sections categorized as confined wandering gravel bed river, and Ft ¼ 32.7e(–0.0032t) for sections
categorized as wandering gravel bed river.

Establishment year
2000 1975 1950 1925 1900 1875 1850 1825 1800 1775 1750 1725 1700
Normalized flood-plain area (m2 ha−1 yr−1)

80
Green River
Above Fontenelle Dam
60 Below Fontenelle Dam

40

20

0
0 25 50 75 100 125 150 175 200 225 250 275 300
Age (years)
Figure 13 Area-age distribution of narrowleaf cottonwood (Populus angustifolia) along the upper Green River, Wyoming. Normalized flood-plain
area is the area of flood plain per year divided by the total flood-plain area sampled at the site. Data are from 79 patches both upstream
and downstream of Fontenelle Reservoir (Glass, 2002 and unpublished results of Michael Scott and Lamont Glass). The steady-state model is
Ft ¼ 43.7e(–0.0052t). The curve is fit through the mean of the points representing unregulated conditions before 1964.

in the 1900s resulted from some factor that did not decrease sustained increase in temperature in the region may be linked
the growth of upland trees. One such factor could be an to the sustained decrease in peak flows since the late 1800s
increase in temperature that decreased snowmelt peaks by suggested by our results.
decreasing the percentage of precipitation falling as snow.
Examination of stable isotopes in an ice core from the
12.10.4.4 Intermediate Cases: Watersheds Dominated by
Fremont Glacier near the divide between the Green and
Low-Elevation Snowmelt and Thunderstorms
Yellowstone basins, showed a strong increase in temperature
marking the end of the Little Ice Age beginning around 1845 The Little Missouri River in western North Dakota and the
and continuing to the present (Schuster et al., 2000). This Redwater River in eastern Montana have watersheds of
160 Tree-Ring Records of Variation in Flow and Channel Geometry

Establishment year
1975 1950 1925 1900 1875 1850 1825 1800 1775 1750 1725 1700

Normalized flood-plain area (m2 ha−1 yr−1)

40
South Fork Snake River

30

20

10

0
0 25 50 75 100 125 150 175 200 225 250 275 300
Age (years)
Figure 14 Area–age distribution of narrowleaf cottonwood (Populus angustifolia) along the South Fork Snake River, Idaho. Normalized
flood-plain area is the area of flood plain per year divided by the total flood-plain area sampled at the site. The steady-state model is
Ft ¼ 44.0e(–0.0074t). The curve is fit through the points representing unregulated conditions before 1956.

intermediate flow variability that are largely or wholly within watersheds of the Snake, Yellowstone, and Green rivers, but
the northern Great Plains. The Little Missouri River flows from less than that of the small Plains watersheds of northeastern
northeastern Wyoming through Montana and South Dakota, Colorado (Table 1).
and joins the Missouri River in North Dakota. The watershed, Channel change along the Little Missouri River is driven by
located within the Great Plains Physiographic Province, in- high flows that occur every 5–10 years, but the flood of record
cludes plains, badlands, and foothills, but no mountains and in 1947 caused an unusual amount of channel migration
no major dams. The study area is within the North Unit of and cottonwood establishment (Miller and Friedman, 2009).
Theodore Roosevelt National Park, North Dakota. Uplands in Cottonwood seedling establishment occurs on point bars and
the study area are sparsely vegetated, highly erosive badlands the forest consists of even-aged bands of trees increasing in
in the Sentinel Butte and Bullion Creek Formations of the age with distance from the channel, similar to the pattern
Paleocene Fort Union Group. The predominant land use in occurring along the Yellowstone, Snake and Green rivers. An
the watershed is grazing. The climate of the Little Missouri area–age distribution developed in 1965 (Everitt, 1968) was
River watershed is semi-arid with a mean annual precipitation close to the steady-state exponential relation (Figure 15).
of 35 cm (Western Regional Climate Center Station 391294, A more recent analysis based on data up to 2003 (Miller
period of record 1 January 1896–30 June 2007). Because and Friedman, 2009) showed that forest establishment has
the watershed is relatively low in elevation, mean precipitation decreased in recent years associated with lower peak flows and
in the headwaters is low and as a result mean daily discharge that there was a moderate deviation from steady state resulting
is lower than in the South Fork Snake, upper Yellowstone, from increased tree establishment in the decades following the
or upper Green rivers even though the watershed is larger flood of record in 1947. Because interannual variation in peak
(Table 1). On the flood plain, the dominant trees are plains flows at this site is large and strongly influenced by the local
cottonwood. Upland vegetation is mostly shrub steppe and factors controlling ice jams, evidence of long-term climate
grassland. change would be more difficult to detect here than at the high-
Peak flow along the Little Missouri River typically occurs in elevation sites.
late March or early April as a result of snowmelt and spring The Redwater River is a relatively small tributary to the
rain. Because the watershed is relatively low in elevation, Missouri River. At Vida, Montana, the 5480 km2 watershed
and because the river flows to the north, snowmelt runoff is prairie and dryland agriculture on gentle rolling terrain.
coincides with ice breakup. As a result, ice jams often Streamflow is moderately regulated by widely scattered stock
increase the peak flow (Harrison and Reid, 1967). By contrast, ponds and small reservoirs with a ratio of storage volume
the montane watersheds described above produce snowmelt to mean annual flow volume of about 0.1 (Merigliano, 2007).
peaks in June, long after ice breakup at the gages examined. As along the Little Missouri River, peak flows are a mix
Along the Little Missouri River, summer floods resulting from of snowmelt flows in March and April and thunderstorm
local thunderstorms are usually, but not always, smaller in peaks, usually in June (Merigliano, 2007) with moderate
magnitude and duration than snow-melt peaks (Table 1). For interannual variation (Table 1). The flood plain is dominated
these reasons, interannual variation in the magnitude of by plains cottonwood and the area–age distribution of
spring peaks is more than that of the large high-elevation the forest is far from steady state (Figure 16). As at the
 Tree-Ring Records of Variation in Flow and Channel Geometry 161

Establishment year
1965 1940 1915 1890 1865 1840 1815 1790 1765 1740 1715 1690 1665

Normalized flood-plain area (104 ft2 mi−1 yr−1)

Normalized flood-plain area (m2 ha−1 yr−1)

14
90
Little Missouri River
12 80

10 70
60
8
50
6 40
30
4
20
2
10
0 0
0 25 50 75 100 125 150 175 200 225 250 275 300
Age (years)
Figure 15 Area–age distribution of plains cottonwood (Populus deltoides subsp. monilifera) along the Little Missouri River, North Unit of
Theodore Roosevelt National Park, North Dakota. Normalized flood-plain area is the area of flood plain per year divided by the total flood-plain
area sampled at the site. The steady-state model is Ft ¼ 94.8e(–0.010t).

Establishment year
2000 1975 1950 1925 1900 1875 1850 1825 1800 1775 1750 1725 1700
50
Normalized flood-plain area (m2 ha−1 yr−1)

Redwater River
40

30

20

10

0
0 25 50 75 100 125 150 175 200 225 250 275 300
Age (years)
Figure 16 Area–age distribution for plains cottonwood (Populus deltoides subsp. monilifera) along the Redwater River in eastern Montana.
Normalized flood-plain area is the area of flood plain per year divided by the total flood-plain area sampled at the site. The steady-state model is
Ft ¼ 135.7e(–0.0143t). The curve is fit through the points representing conditions before 1925. Although there is very little flow regulation, the
channel became stable after the mid-1920s.

Yellowstone, Green, and Snake rivers, there was a scarcity 12.10.4.5 Summary of Section 12.10.4
of trees less than about 100 years old. Sequential aerial
photographs indicate that the channel has moved little since As predicted by Wolman and Gerson (1978), interannual
1938, suggesting that the decrease in cottonwood establish- variation in peak discharge is much larger in small, semi-arid
ment is related to a decrease in peak flows. Because flow watersheds than in large humid watersheds. The ratio of the
regulation is limited in this watershed, it is unlikely to have 100-year flood to the 2-year flood varied from about 2 in large
caused this change in flood peaks. It is unknown whether watersheds with peak flows dominated by high-elevation
the hydrologic shift reflects a change in snow melt or the snowmelt to more than 100 in a small ephemeral watershed
occurrence of a few intense summer thunderstorms, as nearly dominated by runoff from local summer thunderstorms
all channel change occurred before the flow record began (Table 1). In small semi-arid watersheds in the Great Plains,
in 1929. the area–age distribution of flood-plain cottonwood forests
162 Tree-Ring Records of Variation in Flow and Channel Geometry

varied widely over time and space even between sites along the Dudek, D.M., McClenahen, J.R., Mitsch, W.J., 1998. Tree growth responses of
same creek because of the persistent effects of extreme, local Populus deltoides and Juglans nigra to streamflow and climate in a bottomland
hardwood forest in central Ohio. American Midland Naturalist 140, 233–244.
thunderstorms on long-term channel change and tree estab-
Ely, L.L., Webb, R.H., Enzel, Y., 1992. Accuracy of post-bomb 137Cs and 14C in
lishment. By contrast, cottonwood forests along the upper dating fluvial deposits. Quaternary Research 38, 196–204.
Yellowstone, upper Green, and South Fork Snake rivers, where Everitt, B.L., 1968. Use of the cottonwood in an investigation of the recent history
floods are dominated by high-elevation snowmelt with little of a flood plain. American Journal of Science 266, 417–439.
interannual variation in peak discharge, had consistent Everitt, B.L., 1995. Hydrologic factors in regeneration of Fremont cottonwood along
the Fremont River, Utah. In: Costa, J.E., Miller, A.J., Potter, K.W., Wilcock, P.R.
area–age distributions closer to the pattern of exponential (Eds.), Natural and Anthropogenic Influences in Fluvial Geomorphology.
decay predicted by a steady-state model of channel change. Geophysical Monograph 89. American Geophysical Union, Washington DC,
Watersheds with peak instantaneous annual discharges con- 197–208.
sisting of a mix of snowmelt and thunderstorm events had Fenner, P., Brady, W.W., Patton, D.R., 1984. Observations on seeds and seedlings of
Fremont cottonwood. Desert Plants 6, 55–58.
forest patterns intermediate between those dominated by
Fielding, C.R., Alexander, J., Newman-Sutherland, E., 1997. Preservation of in situ,
snowmelt and those dominated by thunderstorms. arborescent vegetation and fluvial bar construction in the Burdekin River of
Because channel dynamics on high-elevation snowmelt north Queensland, Australia. Palaeogeography, Palaeoclimatology, Palaeoecology
rivers are dominated by frequent high flows, their forest 135, 123–144.
area–age distributions are relatively smooth and, therefore, Friedman, J.M., Auble, G.T., 2000. Floods, flood control, and bottomland
vegetation. In: Wohl, E. (Ed.), Inland Flood Hazards: Human, Riparian and
well suited for detecting effects of climate change. Flood plains
Aquatic Communities. Cambridge University Press, Cambridge, pp. 219–237.
along all three of these rivers differed from the steady-state Friedman, J.M., Auble, G.T., Shafroth, P.B., Scott, M.L., Merigliano, M.F., Freehling,
model in having small areas of forest established in the last M.D., Griffin, E.R., 2005a. Dominance of non-native riparian trees in western
75–100 years, suggesting a decrease in river activity caused, at USA. Biological Invasions 7, 747–751.
least in part, by a climate-induced decrease in peak flows. Friedman, J.M., Lee, V.J., 2002. Extreme floods, channel change and riparian
forests along ephemeral streams. Ecological Monographs 72, 409–425.
Friedman, J.M., Osterkamp, W.R., Lewis, W.M., Jr., 1996. Channel narrowing and
vegetation development following a Great Plains flood. Ecology 77, 2167–2181.
Friedman, J.M., Osterkamp, W.R., Scott, M.L., Auble, G.T., 1998. Downstream
References effects of dams: regional patterns in the Great Plains. Wetlands 18, 619–633.
Friedman, J.M., Vincent, K.R., Shafroth, P.B., 2005b. Dating floodplain sediments
Akashi, Y., 1988. Riparian vegetation dynamics along the Bighorn River, Wyoming. using tree-ring response to burial. Earth Surface Processes and Landforms 30,
M.S. thesis, University of Wyoming, Laramie, WY, 245 pp. 1077–1091.
Alestalo, J., 1971. Dendrochronological interpretation of geomorphic processes. Geyh, M.A., Schleicher, H., 1990. Absolute age determination: physical and
Fennia 105, 1–140. chemical dating methods and their application. Springer, Berlin, 503 pp.
Allred, T.M., Schmidt, J.C., 1999. Channel narrowing by vertical accretion along the Glass, L., 2002. Comparison of narrowleaf cottonwood forest condition and
Green River near Green River, Utah. Geological Society of America Bulletin 111, reproduction above and below a reservoir. Masters thesis, University of
1757–1772. Washington, Seattle, WA, 86 pp.
Argus, G.W., 1997. Infrageneric classification of Salix (Salicaceae) in the New Gom, L.A., Rood, S.B., 1999. Patterns of clonal occurrence in a mature cottonwood
World. Systematic Botany Monographs 52, 1–121. grove along the Oldman River, Alberta. Canadian Journal of Botany 77, 1095–1105.
Auble, G.T., Scott, M.L., 1998. Fluvial disturbance patches and cotton- Gottesfeld, A.S., Gottesfeld, L.M.J., 1990. Floodplain dynamics of a wandering river,
wood recruitment along the upper Missouri River, MT. Wetlands 18(4), dendrochronology of the Morice River, British Columbia, Canada.
546–556. Geomorphology 3, 159–179.
Baker, V.R., 1977. Stream-channel response to floods with examples from central Graf, W.L., 1978. Fluvial adjustments to the spread of tamarisk in the Colorado
Texas. Geological Society of America Bulletin 88, 1057–1071. Plateau region. Geological Society of America Bulletin 89, 1491–1501.
Baker, W.L., 1990. Climatic and hydrologic effects on the regeneration of Populus Graf, W.L., 1988. Fluvial processes in dryland rivers. Springer, Berlin, 346 pp.
angustifolia James along the Animas River, Colorado. Journal of Biogeography Graf, W.L., 2006. Downstream hydrologic and geomorphic effects of large dams on
17, 59–73. American rivers. Geomorphology 79, 336–360.
Baum, B.R., 1978. The Genus Tamarix. The Israel Academy of Sciences and Graumlich, L.J., Pisaric, M.F.J., Waggoner, L.A., Littell, J.S., King, J.C., 2003.
Humanities, Jerusalem, 209 pp. Upper Yellowstone River flow and teleconnections with Pacific Basin climate
Bazzaz, F.A., 1979. The physiological ecology of plant succession. Annual Review of variability during the past three centuries. Climatic Change 59, 245–262.
Ecology and Systematics 10, 351–372. Gutsell, S.L., Johnson, E.A., 2002. Accurately ageing trees and examining their
Boggs, K., Weaver, T., 1994. Changes in vegetation and nutrient pools during height-growth rates: implications for interpreting forest dynamics. Journal of
riparian succession. Wetlands 14, 98–109. Ecology 90, 153–166.
Bollschweiler, M., Stoffel, M., 2010. Tree rings and debris flows: recent Hansen, W.R., Chronic, J., Matelock, J., 1978. Climatography of the Front Range
developments, future directions. Progress in Physical Geography 34, 625–645. Urban Corridor and Vicinity, Colorado. US Geological Survey Professional Paper
Bradley, C.E., Smith, D.G., 1986. Plains cottonwood recruitment and survival on a 1019.
prairie meandering river floodplain, Milk River, southern Alberta and northern Harrison, S.S., Reid, J.R., 1967. A flood-frequency graph based on tree-scar data.
Montana. Canadian Journal of Botany 64, 1433–1442. Proceedings of the North Dakota Academy of Science 21, 23–33.
Carrara, P.E., 2007. Movement of a large landslide block B dated by tree-ring Hereford, R., 1984. Climate and ephemeral-stream processes: twentieth-century
analysis, tower falls area, Yellowstone National Park, Wyoming, Chapter B. geomorphology and alluvial stratigraphy of the Little Colorado River, Arizona.
In: Morgan, L.A. (Ed.), Integrated Geoscience Studies in the Greater Yellowstone Geological Society of America Bulletin 95, 654–668.
Area–Volcanic, Tectonic, and Hydrothermal Processes in the Yellowstone Hickin, E.J., Nanson, G.C., 1975. The character of channel migration on the Beatton
Geoecosystem. US Geological Survey Professional Paper 1717. US Geological River, northeast British Columbia, Canada. Geological Society of America
Survey, Alexandria, VA, pp. 43–52. Bulletin 86, 487–494.
Cooper, D.J., Andersen, D.C., Chimner, R.A., 2003. Multiple pathways for woody House, S.M., 1997. Reproductive biology of eucalypts. In: Williams, J.E., Woinarski,
plant establishment on floodplains at local to regional scales. Journal of J.C.Z. (Eds.), Eucalypt Ecology: Individuals to Ecosystems. Cambridge University
Ecology 91, 182–196. Press, Cambridge, pp. 30–55.
Denneler, B., Bergeron, Y., Bégin, Y., Asselin, H., 2008. Growth responses of Howe, W.H., Knopf, F.L., 1991. On the imminent decline of Rio Grande cottonwoods
riparian Thuja occidentalis to damming of a large boreal lake. Botany 86, in central New Mexico. Southwestern Naturalist 36, 218–224.
53–62. Hupp, C.R., 1988. Plant ecological aspects of flood geomorphology and paleoflood
Desloges, J.R., Church, M.A., 1989. Wandering gravel-bed rivers. Canadian history. In: Baker, V.R., Kochel, R.C., Patton, P.C. (Eds.), Flood Geomorphology.
landform examples – 13. Canadian Geographer 33, 360–364. Wiley, New York, NY, pp. 335–356.
 Tree-Ring Records of Variation in Flow and Channel Geometry 163

Hupp, C.R., Simon, A., 1991. Bank accretion and the development of vegetated Noller, J.S., 2000. Lead-210 geochronology. In: Noller, J.S., Sowers, J.M., Lettis,
depositional surfaces along modified alluvial channels. Geomorphology 4, 111–124. W.R. (Eds.), Quaternary Geochronology: Methods and Applications. American
Johnson, W.C., 1994. Woodland expansion in the Platte River, Nebraska: patterns Geophysical Union Reference, shelf 4, pp. 115–120.
and causes. Ecological Monographs 64, 45–84. Osterkamp, W.R., Costa, J.E., 1987. Changes accompanying an extraordinary flood
Johnson, W.C., Burgess, R.L., Keammerer, W.R., 1976. Forest overstory vegetation on a sandbed stream. In: Mayer, L., Nash, D. (Eds.), Catastrophic Flooding.
and environment on the Missouri River Floodplain in North Dakota. Ecological Allen and Unwin, Boston, MA, pp. 201–224.
Monographs 46, 69–84. Panshin, A.J., de Zeeuw, C., 1980. Textbook of wood technology. Structure,
Karlstrom, T.N.V., 1988. Alluvial chronology and hydrologic change of Black Mesa Identification, Properties, and Uses of the Commercial Woods of the United
and nearby regions. In: Gumerman, G.J. (Ed.), The Anasazi in a Changing States. McGraw Hill, New York, NY, 722 pp.
Environment. Cambridge University Press, Cambridge, pp. 45–91. Patterson, A.E., 1959. Distinguishing annual rings in diffuse porous tree species.
Katz, G.L., Friedman, J.M., Beatty, S.W., 2005. Delayed effects of flood control on a Journal of Forestry 57, 126.
flood-dependent riparian forest. Ecological Applications 15(3), 1019–1035. Pockman, W.T., Sperry, J.S., 2000. Vulnerability to xylem cavitation and the distribution
Kondolf, G.M., Micheli, E.R., 1995. Evaluating stream restoration projects. of Sonoran Desert vegetation. American Journal of Botany 87, 1287–1299.
Environmental Management 19, 1–15. Popp, C.J., Hawley, J.W., Love, D.W., Dehn, M., 1988. Use of radiometric (Cs-137,
Lepper, K., Mahan, S., Vincent, K., 2002. Luminescence dating of sediments from Pb-210), geomorphic, and stratigraphic techniques to date recent oxbow
Chaco Wash: Chaco Culture National Historical Park, New Mexico, USA. sediments in the Rio Puerco drainage, Grants Uranium Region, New Mexico.
Geological Society of America Abstracts with Programs 34, 110. Environmental Geology and Water Sciences 11, 253–269.
Lewin, J., 1989. Floods in fluvial geomorphology. In: Beve, K.J., Carling, P.A. Reily, P.W., Johnson, W.C., 1982. The effects of altered hydrologic regime on tree
(Eds.), Floods: Hydrological, Sedimentological, and Geomorphological growth along the Missouri River in North Dakota. Canadian Journal of Botany
Implications. Wiley, Chichester, pp. 264–284. 60, 2410–2423.
Little, J.D.C., 1961. A Proof of the Queueing Formula L ¼ l W. Operations Rubtsov, M.V., Salmina, I.U.N., 1983. Use of buried Salix thickets for determination
Research 9, 383–387. of the chronology of floodplain deposits. Lesovedenie, 69–70.
Luckman, B.H., 2000. The Little Ice Age in the Canadian Rockies. Geomorphology Schumm, S.A., 1969. River metamorphosis. Journal of Hydraulics Division.
32, 357–384. American Society of Civil Engineers 95, 255–273.
Lytle, D.A., Merritt, D.M., 2004. Hydrologic regimes and riparian forests: a Schumm, S.A., Lichty, R.W., 1963. Channel widening and floodplain construction
structured population model for cottonwood. Ecology 85, 2493–2503. along Cimarron River in southwestern Kansas. US Geological Survey
Mahoney, J.M., Koegler, P., Rood, S.B., 1991. The accuracy of tree ring analysis for Professional Paper 352 D, 88 pp.
estimating the age of riparian poplars. In: Rood, S.B., Mahoney, J.M. (Eds.), The Schuster, P.F., White, D.E., Naftz, D.L., Cecil, L.D., 2000. Chronological refinement
Biology and Management of Southern Alberta’s Cottonwoods. Proceedings of the of an ice core record at Upper Fremont Glacier in south central North America.
University of Lethbridge Conference, 4–6 May, 1990. University of Lethbridge, Journal of Geophysical Research 105(D4), 4657–4666. http://dx.doi.org/
Lethbridge, AB, pp. 28–30. 10.1029/1999JD901095.
Mahoney, J.M., Rood, S.B., 1998. Streamflow requirements for cottonwood seedling Scott, M.L., Auble, G.T., Friedman, J.M., 1997. Flood dependency of cottonwood
recruitment – an integrative model. Wetlands 18, 634–645. establishment along the Missouri River, Montana, USA. Ecological Applications
Malde, H.E., 1973. Geologic bench marks by terrestrial photography. US Geological 7, 677–690.
Survey Journal of Research 1, 193–206. Scott, M.L., Friedman, J.M., Auble, G.T., 1996. Fluvial process and the
Martens, D.M., 1993. Hydrologic inferences from tree-ring studies on the establishment of bottomland trees. Geomorphology 14, 327–339.
Hawkesbury River, Sydney, Australia. Geomorphology 8, 147–164. Shroder, J.F., Jr., 1980. Dendrogeomorphology: review and new techniques of
McBride, J.R., Strahan, J., 1984. Establishment and survival of woody riparian tree-ring dating. Progress in Physical Geography 4, 161–188.
species on gravel bars of an intermittent stream. American Midland Naturalist Sigafoos, R.S., 1964. Botanical evidence of floods and flood-plain deposition. US
112, 235–245. Geological Survey Professional Paper 485-A, 35 pp.
Merigliano, M.F., 1996. Ecology and management of the South Fork Snake River St. George, S., Nielsen, E., 2000. Signatures of high-magnitude 19th-century floods
cottonwood Forest. Bureau of Land Management, Idaho State Office. Technical in Quercus macrocarpa tree rings along the Red River, Manitoba, Canada.
Bulletin 96-9, Boise, ID, 79 pp. Geology 28, 899–902.
Merigliano, M.F., 1998. Cottonwood and willow demography on a young island, Stettler, R.F., Bradshaw, H.D., Jr., Heilman, P.E., Hinkley, T.M., 1996. Biology of
Salmon River, Idaho. Wetlands 18, 571–576. Populus and its implications for management and conservation. NRC Research
Merigliano, M.F., 2007. Senescent cottonwoods along the Redwater River: possible Press. National Research Council of Canada, Ottawa, ON, 539 pp.
causes that could guide remedies. Final Report to the McCone Conservation Stokes, M.A., Smiley, T.L., 1968. An introduction to tree-ring dating. University of
District, Circle, MT, 20 pp. Chicago Press, Chicago, IL, 73 pp.
Merigliano, M.F., Polzin, M.L., 2003. Temporal Patterns of Channel Migration, Stromberg, J.C., Patten, D.T., 1992. Mortality and age of black cottonwood stands
Fluvial Events, and Associated Vegetation Along the Upper Yellowstone River, along diverted and undiverted streams in the eastern Sierra Nevada, California.
Montana. College of Forestry and Conservation, University of Montana, Madrono 39, 205–223.
Missoula, MT, 41 pp. Stromberg, J.C., Patten, D.T., Richter, B.D., 1991. Flood flows and dynamics of
Merritt, D.M., Scott, M.L., Poff, N.L., Auble, G.T., Lytle, D.A., 2010. Theory, Sonoran riparian forests. Rivers 2, 221–235.
methods and tools for determining environmental flows for riparian vegetation: Strunk, H., 1997. Dating of geomorphological processes using
riparian vegetation-flow response guilds. Freshwater Biology 55, 206–225. dendrogeomorphological methods. Catena 31, 137–151.
Miller, J.R., Friedman, J.M., 2009. Influence of flow variability on floodplain Sutton, A., Tardif, J., 2005. Distribution and anatomical characteristics of white
formation and destruction, Little Missouri River, North Dakota. Geological rings in Populus tremuloides. IAWA Journal 26, 221–238.
Society of America Bulletin 121, 752–759. Trumbore, S.E., 2000. Radiocarbon geochronology. In: Noller, J.S., Sowers, J.M.,
Moody, J.A., Pizzuto, J.E., Meade, R.H., 1999. Ontogeny of a flood plain. Lettis, W.R. (Eds.), Quaternary Geochronology: Methods and Applications.
Geological Society of America Bulletin 111, 291–303. American Geophysical Union Reference, shelf 4, pp. 41–60.
Nadler, C.T., Schumm, S.A., 1981. Metamorphosis of South Platte and Arkansas Williams, G.P., Wolman, M.G., 1984. Downstream Effects of Dams on Alluvial
Rivers, eastern Colorado. Physical Geography 2, 95–115. Rivers. US Geological Survey Professional Paper 1286, 84 pp.
Nanson, G.C., Beach, H.F., 1977. Forest succession and sedimentation on a Wolman, M.G., Gerson, R., 1978. Relative scales of time and effectiveness
meandering-river floodplain, northeast British Columbia, Canada. Journal of of climate in watershed geomorphology. Earth Surface Processes 3,
Biogeography 4, 229–251. 189–208.
Nanson, G.C., Croke, J.C., 1992. A genetic classification of floodplains. Wolman, M.G., Miller, J.P., 1960. Magnitude and frequency of forces in geomorphic
Geomorphology 4, 459–486. processes. Journal of Geology 68, 54–74.
Nanson, G.C., Knighton, A.D., 1996. Anabranching rivers: their cause, character and Woodhouse, C.A., Gray, S.T., Meko, D.M., 2006. Updated streamflow
classification. Earth Surface Processes and Landforms 21, 217–239. reconstructions for the Upper Colorado River Basin. Water Resources Research
Nevins, T.H.F., 1969. River training–the single thread channel. New Zealand 42, W05415. http://dx.doi.org/10.1029/2005WR004455.
Engineering 1969, 367–373. Yanosky, T.M., 1983. Evidence of floods on the Potomac River from anatomical
Noble, M.G., 1979. The origin of Populus deltoides and Salix interior zones on abnormalities in the wood of flood-plain trees. US Geological Survey
point bars along the Minnesota River. American Midland Naturalist 102, 59–67. Professional Paper 1296, 42 pp.
164 Tree-Ring Records of Variation in Flow and Channel Geometry

Relevant Websites http://water.usgs.gov/osw/streamstats/

U.S. Geological Survey, StreamStats, 2010.
http://waterdata.usgs.gov/nwis/
U.S. Geological Survey, National Water Information System, 2010.

Biographical Sketch
Mike Merigliano is a plant ecologist who is especially interested in how plants respond to disturbance and their
environment. He is with the University of Montana College of Forestry and Conservation, and has conducted
research throughout the West since 1996. His education includes a bachelors of science degree in forest resources
from the University of Idaho and an MS and PhD from the University of Montana. His work experience also
includes forestry with the Idaho Department of Lands and the US Forest Service.

Dr. Jonathan M. Friedman is a plant ecologist and fluvial geomorphologist who has worked for the US Geological
Survey, Fort Collins Science Center in Colorado since 1990. He is also an affiliate faculty member at Colorado
State University and the University of Colorado. Dr. Friedman studies interactions among riparian vegetation,
streamflow, and channel change throughout the western United States with emphases on the effects of global
change and invasive species.

Dr. Michael L. Scott has been a research ecologist with the US Geological Survey, Fort Collins Science Center,
Colorado since 1987. He also is an affiliate faculty member at Colorado State University in the Graduate Degree
Program in Ecology and adjunct assistant professor in the Watershed Sciences Department, Utah State University,
Logan, Utah, where he is currently teaching a graduate seminar course on riparian ecosystems of the western US.
He has over 20 years of research experience with riparian ecosystems throughout Western North America, in-
cluding Mexico, and has published more than 60 peer-reviewed articles in scientific journals. From 2003 to 2006,
he served as an associate editor of the Journal Wetlands. Dr. Scott’s research emphasizes a broad interdisciplinary
and multi-process understanding of riparian ecosystems, with direct application to management. Focal research
activities are primarily centered in the upper Missouri River basin and Colorado Plateau and include: riparian
ecology, riparian restoration and monitoring, invasive species, fluvial geomorphic processes relative to the
establishment and persistence of riparian vegetation, and modeling of riparian vegetation response to flow
alteration.
 12.11 Peatland Geomorphology
M Evans, University of Manchester, Manchester, UK
r 2013 Elsevier Inc. All rights reserved.

12.11.1 Introduction 166

12.11.2 Definition of Peatlands 166
12.11.3 Geomorphology of Intact Peatlands 166
12.11.3.1 Hydrological Control of Raised-Mire Topography: The Groundwater-Mound Hypothesis 167
12.11.3.2 Streamlined Peatland Forms 168
12.11.3.3 Patterned Peatlands 168
12.11.3.4 Peatland Fluvial Systems 169
12.11.3.5 Palsa Mires and Polygonal Mires 170
12.11.4 Geomorphology of Eroding Peatlands 171
12.11.4.1 Causes and Occurrence of Peat Erosion 171
12.11.4.2 Patterns of Peat Erosion 172
12.11.4.3 Peat Erosion Processes 172
12.11.4.3.1 Gully erosion 172
12.11.4.3.2 Wind erosion 173
12.11.4.3.3 Peat mass movements 173
12.11.4.4 Magnitude of Peat Erosion 174
12.11.4.5 Erosion, Revegetation, and Restoration 174
12.11.5 Techniques in Peatland Geomorphology 175
12.11.5.1 Remote Sensing and Geospatial Analysis 175
12.11.5.2 Measuring Physical Properties of Peat 175
12.11.5.3 Measuring Peat Erosion 176
12.11.6 Putting It All Together: Peatland Function and Ecosystem Services 176
References 178

Glossary Minerotrophic Relating to peatlands. Minerotrophic

Carbon sequestration The removal of carbon from the peatlands receive runoff water from nonpeatland upslope
atmosphere (e.g., by photosynthesis) to relatively sources or springwater in addition to rainfall.
stable storage in the terrestrial system (e.g., peatland soils). Ombrotrophic Relating to peatlands. Ombrotrophic
DOC Dissolved organic carbon as for particulate organic peatlands are rain fed, i.e., their sole source of water input is
carbon (POC) but describes the portion of total organic from the atmosphere.
carbon which is in dissolved form. Particulate organic carbon (POC) Used in relation
Geomorphology The study of landforms and land- to aquatic systems, this relates to the portion of the
forming processes. total organic carbon in the water which is in
Hydraulic conductivity The constant which describes the particulate form.
ability of a substance to transmit water (the rate constant in Peatlands A generic term relating to landscapes where the
Darcy’s law). dominant soil cover is peat; may incorporate a number of
LiDAR Light detection and ranging technology for different types of peat landform.
determining distance by the measurement of backscatter of Sediment yield The flux of sediment from a
laser light. Used from airborne platforms, or increasingly catchment system over a given time. Typically has
from ground scanners to create digital elevation models. units of t km 2 a 1.

Abstract

Peatlands cover 3% of the Earth’s surface and play an important role in carbon and water cycling. The geomorphology of
these systems, however, has been relatively understudied. This chapter reviews work on the geomorphology of intact
peatlands and argues that although at the landscape scale the form of the topography underlying the peat has an important

Evans, M., 2013. Peatland geomorphology. In: Shroder, J. (Editor in Chief),

Butler, D.R., Hupp, C.R. (Eds.), Treatise on Geomorphology. Academic
Press, San Diego, CA, vol. 12, Ecogeomorphology, pp. 165–181.

Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00327-4 165

166 Peatland Geomorphology

influence, at smaller scales the geomorphology of intact peatlands is controlled by hydrological and ecological processes.
One exception to this generalization is the case of peatlands in the permafrost zone where the formation of segregation
ice has important impact on the nature of peat landforms. The potential for degradation of these permafrost-influenced
peatlands due to climate warming may lead to an increase in peat erosion. Physical degradation of peatlands is widespread
in the UK and Ireland, and in these conditions normal geomorphological processes such as gully erosion, wind
erosion, and mass movements become significant controls on peatland geomorphology. Erosion of peatlands significantly
impacts on their function and the ecosystem services that they provide. Management of peatlands under a changing
climate will require geomorphological understanding as well as knowledge of the ecology and hydrology of peatland
systems.

12.11.1 Introduction accumulations are considerable with commonly over 2 m of

peat depth and in deep peat basins up to 10 m. Therefore,
Globally, peatlands cover B4 million km2 (Lappalainen, peatland landscapes are those landscapes where the domin-
1996), which is approximately 3% of the Earth’s land area. ant soil type is peat. This is a more restrictive class than
Despite the areal extent of peatlands, the amount of geo- wetland which incorporates flooded mineral soils. Peatland
morphological work on these systems has been rather slight. is generally used interchangeably with mire. Peatlands are
Peatland science is a large and dynamic interdisciplinary field commonly subdivided into fens (areas that are groundwater
with much of the work carried out at the borders of hydrology, fed or fed by runoff from marginal sites) and bogs (areas fed
ecology, and biochemistry. Excellent reviews of this work are only by rainwater). The literature that this chapter draws on
available (e.g., Ingram, 1983; Charman, 2002; Blodau, 2002; relates largely to the peatlands of Europe/Russia and North
Rydin and Jeglum, 2006; Limpens et al., 2008). Analyses of America. These northern peatlands constitute over 80% of
landforms and the processes that drive topographic change are global peatlands (Immirzi et al., 1992). Tropical peatlands
not absent from this literature, but relatively little interaction constitute a different ecosystem that has been less well
has occurred with mainstream geomorphology. Peatland studied and are not considered in this work.
geomorphology differs from that of mineral systems in that
the sediment which constructs the landform is organic so that
its production and degradation (weathering) are strongly in- 12.11.3 Geomorphology of Intact Peatlands
fluenced by biochemical processes. However, particularly in
peatlands susceptible to erosion, conventional understanding A widely used approach to describe peatland landscapes
of the processes of erosion and sediment transport can be identifies three major scales of interest – the microtope, the
applied to understand both landforms and the impacts of mesotope, and the macrotope (Ivanov, 1981; see Table 1).
geomorphological change on peatland function. Although the From the viewpoint of peatlands as landforms, the
application of geomorphological techniques to peatland sys- mesotope is the scale of the peatland landform type, so a
tems has the potential to enhance understanding of their raised bog or a patterned fen is a mesotope. Macrotopes are
function, uncritical application of methods derived for min- amalgamations of peatland landforms into peatland land-
eral sediment systems is generally problematic in organic scapes, and the microtope represents the fine-scale geo-
systems. The aim of this chapter is to review the state of the morphological detail.
science of peatland geomorphology. This includes a review, Peatlands are akin to depositional landforms in that they
from a geomorphological perspective, of current understand- are accreted above the preexisting land surface. In the case of
ing of the controls on peatland landforms and landforming an active peatland, the accretion is typically a function of in
processes. In addition, the review will assess the impacts of situ accumulation of organic litter, rather than deposition of
geomorphological change on peatland function and sum- material eroded from elsewhere. The controls on peatland
marize the most significant technical developments that have growth are therefore a function of site conditions rather than
contributed to this understanding. off-site erosion, but in common with depositional landforms,
peatland form is to some degree derived from, and con-
strained by, the antecedent landscape. This association has
12.11.2 Definition of Peatlands
Table 1 Classification of scales of mire landforms
The scientific literature on peatlands is confused by a morass
of overlapping classification systems. Therefore, it is im- Microtope ‘‘A part of the mire where plant cover and all other
portant to establish what exactly this chapter means by physical components of the environment connected
‘peatland geomorphology’. Peat is organic soil with a thick A with it are uniform.’’ (Ivanov, 1981: 6)
horizon (commonly 440 cm), referred to as a Histosol in Mesotope Isolated mire massifs with distinct patterns of
both the Food and Agriculture Organisation of the United microtopes and a single center of peat formation.
Nations (FAO) and US soil classifications. Because of per-
Macrotope Complex mire massif formed from the fusion of isolated
sistently high water tables, peat forms create anaerobic con-
mesotopes through peat growth.
ditions within the soil and limit decomposition of fresh
plant litter. Peat has been accumulating across most northern Source: Modified from Ivanov, K., 1981. Water Movement in Mirelands (Translated
peatlands for at least 5000 years (Ovenden, 1990), so peat from the Russian by Thompson, A. and Ingram, H.A.P.). Academic Press, London.
 Peatland Geomorphology 167

been demonstrated empirically by Graniero and Price (1999). The one partial exception to this is the case of raised bogs.
In a study of bog patterning in the blanket bogs of New- Unlike other mire types, the large-scale form of a raised bog
foundland, they demonstrated that the strongest predictor of does not relate directly to the underlying mineral surface.
pattern was the pre-peat topography. Similarly, Almquist- Commonly raised bogs develop from lakes or wet depressions
Jacobson and Foster (1995) reported that both the initiation through hydroseral succession. Lake infilling and vegetation
and spread of bogs in southern Sweden are strongly controlled growth first produce emergent vegetation, then fen peat, and
by topographic constraints. finally raised bog peat, commonly with forest cover (Walker,
The importance of landscape position is recognized in a 1970). The precise nature of this succession is variable in space
number of approaches to the classification of mire types and time and controlled by both climate and local vegetation
(Hughes and Heathwaite, 1995; Lindsay, 1995; Charman, (Klinger et al., 1996; Hughes and Barber, 2004: for a useful
2002). Charman (2002) built on the approach of Lindsay review, see Charman, 2002; pp. 145–149). In their climax
(1995) and identified six main mire types based on landscape form, raised bogs are elevated above the surrounding land-
position, namely raised mire, blanket mire, basin fen, valley scape so that the topography of the preexisting mineral surface
mire, floodplain mire, and sloping mire (Figure 1). This is inverted. In contrast to the other hydromorphological mire
hydro-morphological classification can be categorized into the types in Figure 1, therefore, raised-mire morphology at the
water-shedding sites (raised and blanket mire) where peatland mesoscale is controlled by hydrological and ecological pro-
character is primarily ombrotrophic (receiving all its water and cesses, which, as will be shown, is more typical of smaller-
nutrients from rainfall) bog and water-gathering sites, which scale morphology in other mire types.
support a more minerotrophic (receiving some water and
nutrients from groundwater sources) fen peatland. The pri-
12.11.3.1 Hydrological Control of Raised-Mire
mary control on the mesoscale hydrological function of these
Topography: The Groundwater-Mound
sites is the preexisting topography so that the controls on
Hypothesis
peatland form are the long-term geological and geomorpho-
logical controls on topography. Similarly at the scale of the Although peatlands are in many ways analogous to de-
macrotope, given hydroclimatic conditions conducive to peat positional landforms, the accretion of peat is controlled not by
formation, the assemblage of peat landforms that make up the sedimentary processes but by biochemical processes, specific-
macrotope is strongly conditioned by the subpeat topography. ally the balance between litter production and decomposition.

Raised mire (‘bog’) 3

2
Valley mire (‘fen’)
Surface ‘raised’ in center. 1 Peat restricted to valley bottom
receiving water from surface and

2
0 runoff, groundwater, and stream flow.

1
0
2 3
1

0
Streams
0
0

sometimes
0 present 2

1
1

0
2 0
(a) 3 1 (d)

Blanket mire (‘bog’) Floodplain mire (‘fen’)

Peat covers most of landscape 5 Water supply from river floods,
excluding steepest ground. 4 surface runoff, and/or groundwater. Peat cover
3 Water table often fluctuating seasonally. over flood plain.
2 Deposits
Some areas receive 1 mixed with
surface runoff minerogenic
River sediment.
3
2
1

0
2 3

Where peat is
1

0
0

thin groundwater
1

may affect surface Fluctuating water table

(b) (e)
4
Basin mire (‘fen’) 3 Sloping mire(‘fen’)
2
Peat restricted to topographic low. Peat on sloping terrain.
Water table maintained by surface 1 Water from runoff and groundwater.
runoff and groundwater. 0 May be concentrated as spring.
Highly variable setting and morphology.
4

3
2

(c) (f)

Figure 1 Hydro-topographic classification of peatlands: (a) raised mire (bog); (b) blanket mire (bog); (c) basin mire (fen); (d) valley mire (fen);
(e) floodplain mire (fen); and (f) sloping mire (fen). Modified with permission from Charman, D., 2002. Peatlands and Environmental Change.
Wiley, Chichester, 301 pp.
168 Peatland Geomorphology

Because microbial decomposition of litter is strongly con- particular, they highlighted inability to deal with spatial vari-
trolled by the aerobic status of the upper peat layers (Clymo, ability, failure to adequately model interactions of hydrology
1983; Freeman et al., 2001, 2004), rates of decomposition are and peat growth, and failure to model rapid near-surface flows
strongly linked to the water table with suppression of de- as particularly problematic. They also highlighted the im-
composition in anaerobic conditions below the water table. portance of topographic constraints in controlling the vertical
Higher rates of decomposition occur above the water table, and lateral growth of raised bog systems. Belyea and Baird
with the highest rates occurring in the zone of water- (2006) argued that such is the degree of complexity and
table fluctuation (Belyea, 1996). feedback between hydrological processes controlled at large
Ingram (1982) argued that as the water table was the pri- scales and small-scale peat growth that the classic simple
mary control on peat accumulation, the form of raised bogs models of peatland development are inadequate. They advo-
could be modeled using purely hydrological parameters. The cated modeling peatlands as complex adaptive systems
widely cited ‘groundwater-mound hypothesis’ suggests that (Figure 2), integrated in a numerical modeling structure.
the low hydraulic conductivity typical of all but the surface
layers of the peat allows maintenance of higher water tables,
and hence more rapid peat accumulation in the center of the 12.11.3.2 Streamlined Peatland Forms
mire. Ingram demonstrated that the volume of water
A second form of hydrological control on the mesoscale form
throughput (lateral discharge) and the hydraulic conductivity
of peatland landscapes has been identified in detailed work on
can be solved analytically to predict, for typical peatland sys-
the continental peatlands of North America (Glaser, 1987,
tems, a hemi-elliptical groundwater mound. If, as the model
1992). The streamlined bog forms of the zone south of the
assumes, the water table is in fact the dominant control on
permafrost limit (Figure 3) depart from the simple forms
peat accumulation, then this groundwater form should be
predicted by autigenic bog-forming processes. The streamlined
mirrored in the gross topography of the peat surface. Empirical
bogs are separated by minerotrophic water tracks. Glaser
testing of the model against simple raised bog systems
(1987) proposed two mechanisms for the formation of
(Ingram, 1982, 1987) appears to support the assumption. The
streamlined bogs. The first, depositional, mechanism is the
groundwater-mound hypothesis in its original form assumes a
formation of bog islands downstream of bedrock obstructions
constant hydraulic conductivity (k) for the mire. In fact, bog
which shield the bog surface from flows of alkaline ground-
peats have been demonstrated to have significant vertical and
water. The second, erosional, form occurs where a large bog
lateral variability (Kneale, 1987; Baird et al., 2008; Quinton
develops internal flow lines which channel alkaline ground-
et al., 2008), and various attempts have been made to derive
waters, so that as a more minerotrophic vegetation develops,
more sophisticated numerical models of peatland form, which
the bog is separated into several streamlined bog fragments. It
account for variable k and can simulate more complex raised
is important to emphasize that in these intact bogs, neither
bog form (Armstrong, 1995; Bromley and Robinson, 1995).
erosion nor deposition is occurring, but the flow paths of
Similarly, there are alternative numerical approaches to
surface water control the nature of peatland vegetation and
modeling the groundwater mound that allow more com-
consequently peat accumulation rates producing characteristic
plexity but are numerically intensive (Waddington and Roulet,
streamlined forms. Streamlined bog forms are controlled both
1997; Reeve et al., 2001). Recent work has highlighted that
by subpeat sediment (controlling groundwater flow) and
hydraulic conductivity is anisotropic (k values in the vertical
topography (controlling initial bog growth locations) (Glaser,
plane differ from those in the horizontal plane) (Beckwith
1992), but as with the groundwater-mound model the
et al., 2003) and also that k can vary in time due to variable
hydrological control on peatland form is mediated by the
compression (Whittington et al., 2007) or the effects of bio-
interaction of hydrological and ecological processes.
genic gas bubbles in the peat (Waddington et al., 2009). These
observations add further complexity to spatial modeling ap-
proaches driven by hydraulic conductivity. Nevertheless,
12.11.3.3 Patterned Peatlands
although these complexities are real, the basic model of water-
table control on peatland form has been widely accepted and Patterning of peatlands occurs at a range of scales from the
has provided useful explanation of mesoscale peatland geo- mesoscale patterning described by Glaser (1987) through to
morphology (e.g., Glaser et al., 2004). smaller-scale pattern of pools and ridges or hummocks (flarks
The groundwater-mound approach predicts equilibrium and strings) which are common features on many fen peat-
forms for raised bogs rather than predicting rates of peat lands. Such surface patterning is a striking landscape feature,
growth. Mass-balance models of peat accumulation have been which, unsurprisingly, has been a significant focus of peatland
used to predict growth rates and ultimate bog elevation. research. Detailed reviews were provided by Glaser (1998).
Clymo (1984) demonstrated that slow decomposition of deep The generation of pattern is commonly thought to be
peat means that there is an ultimate limit to peat growth, autigenic. Belyea and Lancaster (2002) described how pat-
because overall decomposition increases as bog depth in- terning in oceanic bog environments results from initial for-
creases, whereas recruitment of fresh litter remains constant. mation of pools at low gradient sites in the wider peatland
Efforts have been made to integrate these two approaches landscape such as saddles. Various models have been sug-
(Almquist-Jacobson and Foster, 1995), but recent work by gested where the pattern derives from system self-organization
Belyea and Baird (2006) has argued that a range of approaches following relatively simple rule sets. Key controls proposed
to combining these two models all fail to adequately describe include the relative ability of hummocks and pools to transmit
the detail of peatland development in varying contexts. In water (Couwenberg and Joosten, 2005), lower peat growth
 Peatland Geomorphology 169

Timescale (years) Space scale (m) Internal processes External forcing

Macrotope
104−105 104−105 (e.g., bog−fen complex) Multimillennial climate
(e.g., glacial−interglacial
Energy balance Paludification and
cycle)
drainage networks terrestrialization

Mesotope
102−103 102−103 (e.g., raised bog)
Millennial climate
Flow network
Lateral expansion (e.g., Little ice age)
macrosuccession
landscape topography
storage change

Mircotope
101−102 101−102 (e.g., hummock−hallow
complex) Decadal climate
(e.g., North Atlantic
Peat formation Oscillation)
Water balance microsuccession
ecological neighborhood flow routing

Microform
1−101 1−101 (e.g., hummock
Annual climate
Litter production and decay
Water table depth competition (e.g., Seasonal variation)
species composition evaportranspiration
interception
Plant assemblage
10−1−1 10−2−1 (e.g., S. capillifolium +
C. vulgaris)

Figure 2 Conceptual model of multiscalar feedbacks in peatland systems. Modified from Belyea, L.R., Baird, A.J., 2006. Beyond ‘the limits to
peat bog growth’: cross scale feedback in peatland development. Ecological Monographs 76(3), 299–322, with permission from Sage.

rates in pools (Belyea and Clymo, 2001), and convective within peatland complexes. Generally in extensive peatlands,
transfer of nutrients (Rietkerk et al., 2004). Foster and Wright the major drainage channels pre-date the formation of the
(1990) argued that pool formation is an autigenic process and peat – they may flow on a mineral bed in a preexisting valley,
does not require external climate change; however, Charman or the river system may flow across an extensive fen wetland.
(1994) demonstrated the development of patterned fen In either case, the flashy hydrological regime and the impact of
overlying bog peats which he argued is due to hydrological peat growth on floodplains should lead us to expect that these
changes resulting from pipe development within the peat. peatland rivers have distinctive features.
Spectral analysis of peatland hummock structure by Kenkel Work on a fen system in Wisconsin has described a river
(1988) demonstrated sphagnum hummock frequencies re- that has a very low energy system with unusual features, in-
lated to spacing of vascular plants that provide internal scaf- cluding low width:depth ratios, long straight reaches, and
folding. The spacings are therefore controlled by intra-specific acute angled bends (Watters and Stanley, 2007). These features
competition between scaffold species. Patterned fens have also are consistent with previous findings by Jurmu (2002). Watters
been reported from the alpine zone, where it has been argued and Stanley concluded that in these systems, the influences of
that the presence of alpine permafrost, frost action, and spring groundwater, and, in particular, biological control in the form
melt are key controls on mire form (Vitek and Rose, 1980). of peaty banks and dense bank vegetation, exert a much
A complete explanation of the wide range of peatland greater control on fluvial form than in mineral systems. One
patterning observed across the northern peatlands has yet to characteristic associated with the well-vegetated stream banks
be proposed. It is likely that self-organization is a property of of peatland systems is cantilever bank failure (Figure 4), where
the complex feedbacks between hydrological change and typically the well-vegetated failure block remains against the
ecological change that characterize mire surfaces (Belyea and bank and provides some measure of protection from further
Baird, 2006), so that although the local cause of patterning bank erosion. This pattern has also been observed in upland
may vary in space and time, the development of pattern at the peatlands. Evans and Warburton (2001) identified the role of
level of the microtope is a property of most mire systems. cantilever bank failure in the supply of large blocks of peat to
upland river systems. Cantilever failure of peaty banks pro-
vides a source of coarse organic debris to the river system.
12.11.3.4 Peatland Fluvial Systems
These peat blocks can be a significant control on fluvial form
One aspect of peatland geomorphology that has been under- in upland rivers. Meter-scale blocks of peat form significant in-
studied is the function and activities of the fluvial network channel obstructions that persist over timescales of up to 10
170 Peatland Geomorphology

1 Streamlined bog
2 Water track

2 1

Figure 3 Patterns of streamlined bogs. Original copyrighted by Regents of the University of Colorado.

control on the stream long profile (Evans and Warburton,

2007). The role of peat blocks in peatland river systems is, in
many ways, analogous to the extensive work on large organic
debris in forested stream systems (e.g., Gurnell et al., 2002).
Where peatland river systems are laterally unstable, so that
peat blocks are produced by bank failure, these blocks con-
stitute a distinctive characteristic of the fluvial geomorphology
of peatland systems.

12.11.3.5 Palsa Mires and Polygonal Mires

A well-studied area of the geomorphology of intact mire sys-
tems are the systems affected by periglacial processes. In
temperate mire systems, the dominant controls on mire form
derive from interactions between mire hydrology and the mire
vegetation. However, in extensive areas of the northern peat-
Figure 4 Cantilever bank failure on a peatland stream after a large lands that lie at higher latitudes or in strongly continental
flood event. The trash line of a recent large flood is apparent on the conditions, mire geomorphology at the scale of the microtope
left-hand side of the image, and in the center of the image just above is strongly influenced by the role of water in the frozen form.
the rock step in the stream there is a large cantilever bank failure Palsa mires are relatively dry ombrotrophic systems typical
typical of this stream type. of marginal permafrost areas. Palsa mires are characterized by
the formation of raised palsa mounds with relief of up to 7–8
years and produce characteristic-associated sedimentary forms m. The mounds are ice cored and are raised by the formation
in the gravel bed, including scour holes and tails (Warburton of segregation ice in the core of the palsa mound. The mounds
and Evans, 2011). In more headwater systems, peat blocks can form as a response to changes in the thermal regime of the
form jams across the stream cross section which are a major upper peat due to the effects of wind blow of snow or
 Peatland Geomorphology 171

vegetation succession (Seppala, 1982; Gurney, 2001). Due to unsurprising that the majority of academic work on peat ero-
the elevated surface of the dry mound, palsas commonly ex- sion has been completed in the UK and Ireland. The geo-
hibit surface erosion of the peat caused by wind action and graphical focus of the second part of this chapter therefore will
abrasion by wind-blown ice crystals (Seppala, 2003). be relatively narrow. However, it is argued that the under-
Palsa mires occur in cold and dry conditions with mean standing of peatland erosion processes which is derived from
temperatures in the range –3 to –5 1C (Luoto et al., 2004). the areas of severe erosion has, in addition to its intrinsic
Parviainen and Luolo (2007) demonstrated that the tem- geomorphological interest, a wider applicability. As global
perature conditions are more important than rainfall in de- temperatures rise and increasing evidence of permafrost thaw
termining the distribution of palsa mires. Envelope modeling emerges, potential exists for significant physical instability in
of climate has suggested that the distribution of palsa mires is the extensive northern peatlands of the permafrost zone. Evi-
strongly sensitive to projected climate warming. Fronzek et al. dence of erosion associated with thaw exists already and is
(2006) modeled almost complete loss of palsa mires from the likely to become more widespread (Klaminder et al., 2010). The
Fennoscandia region by 2080. understanding of physical erosion processes derived from the
A second mire form strongly influenced by the formation UK and Ireland experience may become a more important part
of segregation ice within the permafrost zone is the polygonal of the management of extensive northern peatlands, particu-
mire. The polygonal form of these mires is controlled by the larly in areas of sloping bogs, if predicted climate warming
formation of ice-wedge polygons in the zone of continuous occurs.
permafrost. Polygonal ice-wedge formation creates raised rid- The extraordinary concentration of peatland erosion in the
ges around the edge of a central wet mire (diameter 7–30 m) UK and Ireland raises the obvious question as to what caused
(Minke et al., 2009). Over time, peat growth in the center of peat erosion in these areas. This has been a major area of re-
the polygons can create high-centered polygonal mires. These search over the last 40 years. The extensive peat erosion oc-
high-centered mires appear to be self-limiting in that the high curring across the UK and Ireland appears to be largely a
center dries out and is subject to aeolian erosion (Minke et al., product of the last millennium (Labadz et al., 1991; Rhodes
2007). Polygonal mire systems are widespread across the arctic and Stevenson, 1997; Tallis, 1997a), although dates from Ire-
region (Minke et al., 2007) and, in common with palsa mires, land (Bradshaw and McGee, 1988; Leira et al., 2007) and North
are likely to be strongly climate sensitive. Wales (Ellis and Tallis, 2001) show initiation in the range
Because of the key role of segregation ice in the formation 1500–3000 BP. A range of possible causes for erosion has been
of mire microtopography in the permafrost zone, the potential identified, including climate (both warmer, causing desiccation,
exists that climate warming will lead to significant physical and colder/stormier, causing greater erosion by runoff; Tallis,
instability in these peatlands. On low-relief sites, this in- 1997a, 1997b; Rhodes and Stevenson, 1997), overgrazing
stability is likely to lead to vegetation change and change in (Shimwell, 1974; Huang, 2002), atmospheric pollution (Fer-
mire microtopography. However, on sloping sites the potential guson et al., 1978), natural peatland instability (Tallis, 1985),
exists that such instability will trigger the onset of peat and moorland fire (Radley, 1960; Anderson et al., 1997).
erosion. The geomorphology of eroded mire systems is Although in some locations the specific cause of erosion
considered in more detail in the next section. can be identified, particularly for example where wildfire leads
to catastrophic removal of surface vegetation (e.g., Maltby
et al., 1990; Anderson et al., 1997), the search for a single
cause of widespread erosion has not convincingly identified
12.11.4 Geomorphology of Eroding Peatlands
one. Evans and Warburton (2007) argued that peatland sur-
face stability can be characterized as a balance of erosional
12.11.4.1 Causes and Occurrence of Peat Erosion
forces and resisting forces. The resisting forces are associated
Physical erosion of peatlands has been recorded around the with the binding of surfaces through a healthy moorland plant
world in locations as diverse as China (Joosten and Schumann, assemblage, whereas the erosional forces relate to climatic
2007; Joosten et al., 2008), Newfoundland (Glaser and Jans- conditions and runoff. As the balance between these forces
sens, 1986), Lapland (Luoto and Seppälä, 2000), Lesotho shifts with climate and land-use change, peatland systems can
(Grab and Deschamps, 2004), and sub-Antarctic Macquarie cross a threshold into an erosional state. Under this con-
Island (Selkirk and Saffigna, 1999). However, in most parts of ceptualization, anthropogenic impacts on the vegetation (in-
the world peat erosion is local and associated with specific cluding air pollution, fire, overgrazing, and footpath erosion)
anthropogenic impacts on the peatland surface, including reduce the resistance to erosion, whereas climatic and hydro-
overgrazing (Grab and Deschamps, 2004; VNPA, 2005) and logical changes such as increased storminess, greater frost ac-
peat mining (Lavoie et al., 2003), or it is associated with areas tion, or increased runoff increase the erosive force. When
of permafrost degradation (Gurney, 2001; Zuidhoff, 2002; viewed in this way, it can be seen that during the last mil-
Klaminder et al., 2010). The major exception to this statement lennium a combination of processes has acted to stress peat-
are the peatlands of the UK and Ireland, where approximately land vegetation surfaces. These processes include agricultural
30% of peatlands are, to some degree, eroded (Evans and intensification, industrial air pollution, and desiccation in the
Warburton, 2007), and peat erosion is severe and widespread. Medieval Warm Period. Furthermore, during the Little Ice Age,
In the worst-affected areas of the south Pennines in England, increased storminess and frost action have increased the ero-
erosion has reached such a degree that erosional gullies com- sional stress on the peatland surface. Local spatial variation in
prise 34% of the peatland area (Evans and Lindsay, 2010). As a the intensity of these impacts has created a mosaic of upland
consequence of this global concentration of erosion, it is peat erosion so that the particular causes of peat erosion can
172 Peatland Geomorphology

be identified at a site, but the regional impact is best understood 12.11.4.3 Peat Erosion Processes
in terms of changing thresholds of stability for the peat surface.
Three major mechanisms of peat erosion can be identified –
The multiple impacts on that peat surface are well exemplified
erosion by running water (predominantly gully erosion)
by Tallis (1997a), who described the cumulative impact of
(Labadz et al., 1991), aeolian erosion (Warburton, 2003), and
desiccation, fire, peat sliding, sphagnum loss due to pollution,
mass movement of peat (Dykes and Warburton, 2007).
overgrazing, and eventually the construction of a television
mast on a site in the southern Pennines, UK. These impacts,
spanning a period of 500 years, combined to cause extensive 12.11.4.3.1 Gully erosion
and severe erosion at the site. Similarly, Yeloff et al. (2005, The most areally extensive form of peatland erosion is gully
2006) have demonstrated increased peat erosion at another erosion. Gully erosion is widespread across the upland peat-
South Pennine site post-1959. The initial bare peat surface was lands of the UK and Ireland (Bower, 1961; Coupar et al., 1997;
exposed by wildfire, but they argued that the severe erosion was Cooper and Loftus, 1998; McHugh et al., 2002; Evans and
caused by high grazing numbers that suppressed natural Warburton, 2007). Once bare peat is exposed, vertical incision
revegetation. Suppression of revegetation allows accumulation of gullies occurs by normal fluvial erosion processes. Cross-
of erosion features in the landscape, so that anthropogenic sectional form of gullies is typically V-shaped in the initial
impacts on revegetation may be as important as the initial stages of incision. Once the base of the gully is eroded to the
triggers of erosion in the generation of the widespread peat mineral substrate at the base of the peat, the rate of vertical
erosion that occur in the UK (Evans and Warburton, 2007). incision is reduced and the gully cross sections become char-
acteristically rectangular in more mature gullies. Lateral ero-
sion of the bare peat walls exposed by gully erosion is rapid.
12.11.4.2 Patterns of Peat Erosion
Freshly exposed fibrous peat is relatively resistant to erosion by
When viewed from the air, the patterning of areas of peat surface flow (Carling et al., 1997), but frost action and des-
erosion is very distinctive. These patterns were described and iccation of the bare peat surface produce a friable weathered
classified from aerial photography by Bower (1960, 1961). He layer which is easily mobilized by surface flow across gully
identified two characteristic patterns of gully erosion, termed ‘type walls and by raindrop impacts (Francis, 1990; Labadz et al.,
1’ – densely wandering dendritic gullies on slopes of less than 1991; Klove, 1998; Evans and Warburton, 2007). Rates of
51 – and ‘type 2’ – weakly branching linear gullies running lateral gully wall recession have been widely measured using
downslope on steeper terrain (Figure 5). Bower also recog- erosion pins. Table 2 reports the range of recession rates re-
nized the potential importance of aeolian processes in strip- corded across the UK and more widely. Bare peat typically
ping peat from summit areas, a mechanism also emphasized shows recession rates of up to 40 mm yr–1. Consequently,
by Radley (1962) and Tallis (1964), who termed this ‘summit- where gully erosion progresses unchecked, it can lead to rapid
type’ erosion. The association of slope and pattern and process degradation. In the southern Pennines of the UK, which are
can be used to characterize a typical process regime across a probably the most eroded blanket peatlands in a global con-
slope profile (Evans and Warburton, 2007; Figure 6). The text, Evans and Lindsay (2010) mapped 34% of the land sur-
basic characteristics of this model are also discernible in the face as gullied, effectively a loss of 34% of peatland volume
mapping presented in Figure 5. over approximately the last millennium (Tallis, 1985).

632 m

516 m

400 m
2
1

200 m

(a) (b)

Figure 5 Patterns of peat erosion from the Bleaklow Plateau in the English southern Pennines: (a) digital elevation data of the Bleaklow summit
area and (b) patterns of gully erosion mapped using the approach of Evans and Lindsay (2010). On the steeper western flanks of the summit
linear gullies characteristic of Bowers type 2 can be seen (box 1). Nearer the summit on areas of lower slope, the high-density wandering gullies
with the dendritic form characteristic of Bowers type 1 can be seen. The area to the east of box 1 where there are no gullies is an area of
summit-type erosion where the peat has been entirely eroded away.
 Peatland Geomorphology 173

Summit/interfluve

Hillslope
Slope series

Valley bog /lowland

Drainage Anastomosing Dendritic Linear Dendritic Pool

Peat flux Erosion Deposition

Process Aeolian Fluvial Sedimentary

Figure 6 Hillslope model of peat erosion. Modified from Evans, M., Warburton, J., 2007. The Geomorphology of Upland Peat: Pattern, Process,
Form. Blackwell, Oxford, 262 pp, with permission from Wiley.

Table 2 Reported rates of erosion of bare peat surfaces measured by erosion pins

Location Context Surface retreat rate References

(mm a 1)

South Pennines, England Gully walls and bare peat 5–33 Philips et al. (1981), Tallis and Yalden (1983),
Anderson et al. (1997), and Authors (unpublished
data)
North Pennines, England Gully walls 10.4–19.3 Philips et al. (1981) and Evans and Warburton
(2005)
Wales Gully walls 16–23.4 Robinson and Newson (1986), Francis and Taylor
(1989), and Francis (1990)
Tasmania Bare peat 43 Selkirk and Saffigna (1999)
Shetland Islands, Scotland Bare peat 10–40 Birnie (1993)

12.11.4.3.2 Wind erosion slope such as summit peats. Small-scale ripples and
Wind erosion has long been identified as an important terracing are commonly observed, but the most significant
mechanism of erosion of upland peat (Radley, 1962) and is aeolian landforms are peat robofard forms (Evans and War-
one of the more commonly reported features of peat erosion burton, 2007), which are asymmetric tilted peat blocks pro-
in areas of peatland beyond the severely impacted areas of the duced by preferential windward erosion of remnant peat
UK (e.g., Selkirk and Saffigna, 1999; Zuidhoff, 2002). Until blocks.
recently, however, there has been very little research into the
processes of wind erosion. Warburton (2003) demonstrated 12.11.4.3.3 Peat mass movements
that, in the wet conditions typical of peatland environments, Where peat has formed on slopes, one of the most dramatic
application of standard dryland models of wind erosion is forms of peat erosion is mass failure of peat as peat slides or
problematic. He proposed that the dominant mechanism of bog bursts. There is a long history of describing these events in
wind erosion was wind-driven rainsplash. Measurements by the literature (Honohane, 1697; Griffiths, 1821; Mulvaney,
Foulds and Warburton (2006, 2007) confirmed the import- 1879; Bowes, 1960), but systematic analysis of mass failure
ance of rainsplash in mobilizing peat from the surface by mechanisms and of peat failures as a class of mass movements
demonstrating that lateral sediment fluxes recorded in aeolian is a more recent development (Carling, 1986; Mills, 2002).
sediment traps were strongly controlled by prevailing wind Dykes and Warburton (2007) presented a classification of
and that fluxes during wet periods exceed than those during peat mass movements based on the mode of failure (Table 3).
drought conditions by a factor of 10. The key distinction is between flow mechanisms where there is
Wind erosion has a demonstrable impact on the deflation liquefaction of a significant depth of well-humified peat and
of bare peat surfaces; however, because of the more significant sliding mechanisms where intact peat is transported above a
role of water erosion on sloping surfaces, the landforms as- shear zone that may be a weaker layer either within the peat or
sociated with wind erosion are most significant in areas of low within the substrate. Peat mass movements are most commonly
174 Peatland Geomorphology

Table 3 Classification of peat mass movements

Term Definition Morphology of failure scars

Bog bursts Flow failure of raised bog, failure in deep peat resulting in Shallow depressions with marginal concentric ridges
slurry-like flow
Bog flows Flow failure of blanket bog, failure in deep peat resulting in Pear-shaped failures with marginal peat blocks or concentric
slurry-like flow ridges
Bog slides Shearing failure of deep peat in blanket bog. Sliding of Failure shape varies but marginal blocks do not show
intact peat over shear surface significant volume loss (subsidence)
Peat slides Shear failure at the peat-mineral interface in blanket bogs Variable shape, margins defined by undisturbed peat
Peaty debris slides Sliding failure initiated by shear within the mineral Variable shape, margins defined by undisturbed peat
substrate below the peat
Peat flows Flow failure of fen peats and other peat types caused by Long, thin, or sometimes variable shape
head loading

After Dykes and Warburton (2007).

less than 50 000 m3 in magnitude, although flow-type failures budget. Evans et al. (2006) reported data from two contrasting
can be larger and may have magnitudes of 5–10 million m3 for catchments draining blanket peat in the southern Pennines,
the very largest events (Dykes and Warburton, 2007). UK. There, a near-complete sediment budget suggested that
Peat mass movements are distinct from extensive gully total aeolian losses are only B1% of total sediment yield from
erosion and, to some extent, from wind erosion, in that the a partially revegetated catchment. Assessing the relative im-
distribution of reported peat failures is global (Dykes and portance of peat mass movements is more difficult. In a small
Warburton, 2007; Evans and Warburton, 2007). Early workers (1 km2) catchment, a peat slide that evacuates 10 000 m3
argued that peat on slopes reached a limit of stability as peat (1000 t) of sediment is equivalent to perhaps 10 years of
depth increased (Pearsall, 1950), although later work has normal erosion. However, at the landscape scale peat slides are
demonstrated that at least for some peat types this is not the relatively rare in both time and space. Mills (2002) estimated
case (Carling, 1986). However, the wide distribution of peat the landscape-scale impact of peat sliding in the blanket
mass movements suggests that they are a natural component peatlands of the North Pennines as equivalent to 3% of total
of peatland landscapes. Failure events are commonly associ- sediment yield. Therefore, the limited available data support
ated with high-magnitude convective rainfall events or rain- the idea that gully erosion is the dominant mechanism of
on-snow events (Carling, 1986). Warburton et al. (2004) sediment export from eroding peatlands. However, in par-
identified an association between peat failures and preexisting ticular locations where low-slope angles minimize runoff
drainage lines, whether natural or artificial. Evidence exists erosion, or where mass movements are active, wind erosion
that the system can be pre-conditioned for peat failure by and landsliding can become locally important.
external changes that modify hydrological pathways. These
include not only artificial drainage but also summer desic-
12.11.4.5 Erosion, Revegetation, and Restoration
cation that may produce cracks in the peat, which can channel
runoff into deeper peat or to the peat mineral interface Comparative sediment yield data presented by Evans et al.
(Warburton et al., 2004). Peat mass movements are high- (2006) for two Pennine catchments, one actively eroding and
magnitude events that can cause significant local disruption to the other with significant gully revegetation, demonstrate the
transport and impact local stream hydrology. The relative importance of revegetation in controlling sediment yield from
importance of mass failure and other erosion forms in the eroding sites. In these catchments, natural revegetation of gully
wider sediment budget are considered in the next section. floors provides an effective sediment trap so that sediment yield
is reduced by an order of magnitude, despite the fact that the
steep gully walls remain bare. Natural revegetation of eroding
12.11.4.4 Magnitude of Peat Erosion
sites is common in the absence of continued pressures such as
Total sediment yields from eroding peatland systems are high. pollution and overgrazing, which prevent the establishment of
A review of published data on sediment yield from eroding vegetation cover. Crowe et al. (2008) demonstrated that com-
blanket peatlands in UK catchments shows losses in the range plete revegetation can occur at timescales on the order of 25
of 12–265 t km 2 a 1 (Evans and Warburton, 2007). To put years. Revegetation starts preferentially in locations where
these figures in context, Walling and Webb (1981a) reported stream power is reduced, either by partial blockage of gullies
typical sediment yield for upland catchments in the UK of up due to bank collapse or by gully widening leading to wandering
to 100 t km 2 a 1. The landscape effect of these losses from channel form on the gully floor (Wishart and Warburton, 2001;
peatland areas is further magnified by the relatively low dry Evans and Warburton, 2007; Crowe, 2007). Therefore, peat
density of peat (at least an order of magnitude less than erosion is not an irreversible process; natural recovery of vege-
mineral soils), which means that the volume of material being tation occurs at relatively short timescales. However, the mor-
removed from peatland catchments is large. Assessment of the phological impacts of erosion are likely to be irreversible at all,
relative proportional contributions of the peat erosion but the longest, timescales, with significant implications for
mechanisms reviewed above is hampered by the limited peatland function (see the next section). In locations where
studies that have measured the whole catchment sediment peat erosion is widespread, restoration of eroding systems,
 Peatland Geomorphology 175

remote sensing for the purposes of estimating peatland carbon

cycling has been reviewed recently by Harris and Bryant (2009).
Perhaps the most important development of remote sens-
ing from a geomorphological point of view has been the
availability of high-resolution digital elevation data derived
from airborne light detection and ranging technology
(LiDAR). The ability to resolve vertical changes on the order of
100 mm at spatial resolutions as low as 0.5 m has a wide range
of potential applications in the analysis of peat landforms. To
date, applications include assessing peat (carbon) loss due to
wildfire (Poulter et al., 2006; Ballhorn et al., 2009) and
mapping of peatland erosion (Evans and Lindsay, 2010).
LiDAR has also been used in combination with multispectral
data for peatland classification (Maxa and Bolstad, 2009),
mapping peatland condition (Milton et al., 2004), and for
mapping vegetation for ecohydrological classification
Figure 7 Early stages of peatland revegetation on the Bleaklow (Anderson et al., 2010). Despite the increasing application of
Plateau in the southern Pennines, UK. A nurse crop of green lawn LiDAR in the analysis and classification of peatland systems,
grass has been established over extensive bare peat through aerial arguably the potential of the available data for geomorpho-
seeding with the addition of lime and fertilizer. Over time a complete logical investigations has not yet been fully realized. The
vegetation cover is established and native sedge and heather begin to available high-resolution topographic data have potential
replace the grasses. The transformation from an unvegetated to a
both in the upscaling of small-scale process investigation and
fully vegetated surface dramatically reduces sediment yield but the
in the landscape-scale analysis of landforms.
eroded topography of the peatland is still apparent.
The main focus of the remote sensing of peatlands has
been the exploitation of spatially extensive remotely sensed
either by artificial revegetation (reseeding) or through artificial data sets, to study peatlands either at the landscape scale or as
gully blocking, has been attempted with some success (Ander- a route to upscaling smaller-scale measurements, for example,
son et al., 2009). Preliminary evidence suggests that reseeding of carbon flux (Harris and Bryant, 2009). However, there is
of bare peat (Figure 7) has the potential to reduce sediment flux also significant potential in the remote sensing of the interiors
from eroded landscapes to levels similar to pristine peatlands of peat landforms. For example, Holden et al. (2002) applied
(author’s unpublished data), because it reestablishes a complete ground penetrating radar (GPR) to map the connectivity of
vegetation cover and shuts down sediment production. As with pipe networks in peatland systems, whereas Holden (2004)
natural revegetation, however, restoration of peatland function demonstrated the use of GPR in active tracing of pipe flow.
is only partial because the gullied morphology significantly GPR has also been applied to map peatland stratigraphy and
modifies the hydrological functioning of the peatland. to determine depths of peat accumulation (Slater and Reeve,
2002). Similarly, Comas et al. (2004) used resistivity meas-
urements to demonstrate a correlation between subpeat stra-
tigraphy and surface vegetation. The remote measurement of
peat depth is an area that is increasingly of interest since
12.11.5 Techniques in Peatland Geomorphology
high-density measurements of peat depth are important for
accurate assessment of carbon storage in peatland landscapes.
12.11.5.1 Remote Sensing and Geospatial Analysis
Resistivity measurements have also been exploited by Slater
The use of techniques of remote sensing to assess peatland et al. (2007) to record the evolution of biogenic gases from
landscapes has a long history. The seminal work on patterns of peat systems, which play an important role in vertical changes
peat erosion by Bower (1960, 1961) built on painstaking to the mire surface as well as being important greenhouse
manual mapping and classification of pattern from aerial gasses. Therefore, the application of geophysical techniques of
photography. Aerial photography has continued to be import- remote sensing to peatlands has potential in mapping peat-
ant in identifying large-scale pattern in peatland (Glaser, 1987; land form and also in monitoring peatland processes.
Vitt et al., 1994). Increasingly however, high-resolution satellite
imagery has increased the available spatial coverage and the
ease of data acquisition. In addition, greater use of multispectral
12.11.5.2 Measuring Physical Properties of Peat
and hyper-spectral data has been made to map vegetation and
peat composition remotely (McMorrow et al., 2003). One of the challenges of geomorphological analysis of peat-
More recently, rapid advances in capability to acquire and lands is that the particular properties of peat as a landforming
process data have led to the development of a range of ap- material mean that standard methods of measurement gen-
proaches to remote sensing of peatland form and condition. erally require modification. Approaches to modeling peat
Remote-sensing applications to wetlands have been re- failure require an understanding of geotechnical properties of
viewed by Ozesmi and Bauer (2002). An area of considerable peat. Although the classic text on the geotechnical aspects of
focus has been distinguishing peatland types on ecological peat (Hobbs, 1986) provides an excellent starting point, the
grounds. The classification of northern peatland vegetation by practicalities of measuring peat properties have led to the
176 Peatland Geomorphology

development of a range of modified procedures (e.g., Yang and density materials (Francis, 1990), and the large range of peat
Dykes, 2006; Dykes, 2008). Similarly, the compressibility of particle sizes from silt fraction (Pawson et al., 2008) through
peat soils meant that, in order to calculate hydraulic con- to large peat blocks (Evans and Warburton, 2001).
ductivity from piezometer head recovery tests, a modified Sediment yield data provide a useful basis for comparison
approach for compressible soils is required (Baird and Gaff- between sites in a range of states of erosion (Evans and War-
ney, 1994; Holden and Burt, 2003). Even the most basic de- burton, 2007) but are limited in their ability to elucidate the
termination of physical parameters such as soil moisture and controls on sediment flux. Despite the relatively low density of
particle size requires careful consideration in peats because organic sediments in some catchments, erosion and de-
oven drying of peat can lead to loss of mass by oxidation, so position have been demonstrated to be an important control
air drying is preferred. Peat particles are organic so that de- on sediment flux even at scales of less than 1 km2 (Evans and
terminations of particle size using X-rays are inappropriate, Warburton, 2005). Therefore, a more complete picture of
and dried peat becomes impossible to disaggregate so that peatland geomorphology is derived from the application of
approaches that deal with wet samples such as particle-size sediment budgeting approaches (Swanson et al., 1982) to
determination by image analysis are most appropriate. eroding systems (Evans and Warburton, 2005; Evans et al.,
2006). In addition to elucidating material transfers at sub-
catchment scale, sediment budget approaches have two key
benefits to the study of eroding peatland systems. The first is
12.11.5.3 Measuring Peat Erosion
the potential for direct comparison with mineral and mixed-
A range of direct and indirect methods has been applied to es- sediment systems that help to establish the wider context of
timate rates of peatland erosion. Direct measurement of the re- peatlands within the landscape sediment system (e.g., War-
cession of bare peat surfaces using erosion pins is probably the burton et al., 2003; Warburton, 2009). The second is that a
most widely applied technique (e.g., Philips et al., 1981; proper understanding of spatial and temporal flows of organic
Anderson et al., 1997; Selkirk and Saffigna, 1999; Evans and sediment within the landscape is a prerequisite to landscape-
Warburton, 2005). A range of pin material has been used, in- scale carbon budgeting. Methodologically, it is but a short step
cluding wooden dowel and 3-mm stainless steel or brass rods. from the production of a sediment budget to a description of
Pins need to be of significant length (in excess of 600 mm) in particulate organic carbon (POC) flux in the landscape. All
order that they are anchored in deep peat to minimize frost that is required is an estimate of the carbon content of the
heave. Pins are typically deployed in groups of 10 or more to POC. The simplest approach is just to assume that carbon
provide both a spatial average and some averaging of potential content is 50% of the organic proportion of the suspended
error. Given the degree of noise induced by disturbance, meas- sediment load. The latter is typically estimated from loss on
urement error (in particular, the difficulty of measuring on fluffy ignition (LOI) (Belcher and Ingram, 1950) measurements on
frost-disturbed surfaces), and frost heave, measurements at much the filter papers used to derive suspended sediment concen-
less than annual resolution are liable to have high signal-to-noise tration (SSC) estimates. This approach has been widely em-
ratios. Annual measurement also provides some control of po- ployed. However, there are methodological difficulties. On
tential seasonal expansion and contraction of surface peats filter papers with low sediment concentrations, the signal-to-
caused by desiccation. Taking these precautions, it is possible to noise ratio of the LOI method becomes problematic, and at
produce replicable results from pin arrays, and the rates of sur- SSCs of less than 10 mg l–1 the noise is considerable.
face erosion derived from the technique are surprisingly con- An alternative approach is to analyze carbon content of
sistent (Evans and Warburton, 2007). filter papers directly in a standard carbon analyzer using
An alternative approach to measuring surface erosion of combustion, catalytic oxidation, and detection by infrared
peat is to derive catchment average erosion from estimates of absorption. This provides direct carbon measurement, but for
sediment yield. Annual sediment yields are derived by ex- filter papers with high sediment loads very careful homogen-
trapolation of point samples using a rating curve approach ization and subsampling of the filter paper is required to
(relating suspended sediment concentration to discharge; produce small samples for analysis.
Walling and Webb, 1981b). Extrapolation is preferred to in- The methodological potential for error in estimating POC
terpolation because of the extremely flashy nature of peatland flux from peatlands is, however, small in comparison to the error
stream systems (e.g., Labadz et al., 1991; Evans and Warbur- in the total carbon budget that arises from ignoring the POC flux.
ton, 2007). An alternative to point sampling is the use of In eroding peatlands, this can be the largest single component of
continuous turbidity probes as a proxy for particulate sedi- the carbon budget and has the potential to shift peatland systems
ment yield. Deriving good calibrations for turbidity probes in from being carbon sinks to carbon sources. Geomorphological
peatland streams is problematic because of the mixed organic engagement with the description and analysis of biogeochemical
and mineral nature of the sediments, which can create vari- cycling in these systems is, therefore, essential.
ability in particle size, shape, and color over time. Neverthe-
less, with careful ongoing field calibration, good results have
been achieved in some studies (Pavey et al., 2007; Grieve and 12.11.6 Putting It All Together: Peatland Function
Gilvear, 2008; Holden et al., 2007). and Ecosystem Services
Several generic problems add to the logistical difficulty of
estimating sediment loads from peatland streams. These in- Peatlands provide a range of ecosystem services (MA, 2005),
clude siltation problems due to very high sediment loads in including carbon storage, water supply, pollution regulation,
eroding systems, the difficulty of using sediment traps in low- biodiversity, and recreation. Provision of these services is
 Peatland Geomorphology 177

intimately linked to the maintenance of high water tables of peatlands. Erosion can impact water delivery, for example,
within peatland systems (Charman, 2002). A major focus of Holden et al. (2008) demonstrated higher overland flow vel-
peatland research in the last decade has addressed the role that ocities on bare peat surfaces as opposed to vegetated surfaces.
peatlands play in the global carbon cycle as important loci of Erosion also has demonstrable impacts on water quality. In-
carbon sequestration and storage (Limpens et al., 2008). creases in dissolved organic carbon loss caused by water-
Northern peatlands are estimated to store 455 PgC and have table reductions are observed as water color changes. Treatment
accumulated that carbon at an average rate of 0.096 PgC yr–1 of water color is a significant concern to the water industry
(Gorham, 1991). Maintenance of high water tables inhibits peat (Worrall et al., 2004).
decomposition through maintenance of anaerobic conditions A further important regulation function associated with
(Freeman et al., 2001). Rates of long-term peat accumulation peatlands is the effective storage of atmospheric pollution.
are strongly controlled by the water table, to the extent that Metal pollutants associated with industrial air pollution are
measures of peat decomposition are adopted as proxy for past effectively bound to organic matter in peat, so that metal
water-table conditions (e.g., Chiverrell, 2001). In an actively concentrations in the peat may be very high (Rothwell et al.,
accumulating peatland, therefore, it is the interactions between 2007). Physical erosion releases these sediment-bound metals
plant primary production and decomposition controlled by into watercourses (Rothwell et al., 2008). In addition, re-
local hydrology that control rates of carbon accumulation. In ductions in water table due to erosional drainage have the
eroding peatlands, geomorphological controls on the style and potential to enhance the mobility of stored pollutants, leading
rates of peat erosion have the potential to impact significantly to the release of metals in the dissolved form into the stream
on carbon accumulation rates. Direct losses of carbon from the system. For example, Rothwell et al. (2009) demonstrated
peatland system through erosion (POC) can be sufficient in the mobility of arsenic in periods of water-table variability in
most extreme cases to shift peatlands from carbon sinks to polluted peatlands in the south Pennines, UK.
carbon sources (Evans et al., 2006). Furthermore, the impact of It is clear that extensive erosion of peatland system disrupts
gully erosion on peatland drainage can additionally reduce ecosystem function. Reductions in water table and direct re-
carbon accumulation. The direct impacts of drainage in many moval of peat by erosion have dramatic impacts on the ability
peatlands are rather small because of the very low hydraulic of the peatlands to provide regulating ecosystem services.
conductivity of peat soils (e.g., Stewart and Lance, 1991), but a Whereas in intact peatlands their form and function are con-
significant reduction in water table over wider areas can be trolled primarily by the interaction of ecological and hydro-
caused by the way in which dense gully networks modify logical processes, in eroded systems geomorphological controls
drainage networks and reduce the area contributing runoff to dominate. Figure 8 presents a conceptual classification of the
downslope areas of the hillslope. This effect reduces average controls on peatland form and function. It builds on the dis-
water tables in the eroded peatland (Allott et al., 2009) and has cussion of both intact and eroded systems presented above and
the potential to increase carbon losses in both dissolved and plots the morphological features of peatland landscapes in
gaseous form due to increased rates of decomposition in the space defined by the scale of the features and the degree of
upper layers of the peatland (Evans and Warburton, 2010). erosion present in the peatland complex. The classification
Erosion also has potential to impact the water-supply function suggests that as scale increases, there is a shift from autigenic

Hydrology/ecology Geomorphology
Intact Eroding

Microtope Autigenic
Low ridge
High ridge Erosional microforms
Hummock
Peat mound Peat hagg
Erosional gully
Sphagnum pool Mud bottomed pool
Permanent pool
Drought sensitive pool

Mesotope Watershed
Spur Erosion pattern
Valley side
Saddle
Eccentric

Gully networks

Macrotope Mire complex

Allogenic

Figure 8 A conceptual model of the controls on peatland geomorphology.

178 Peatland Geomorphology

processes such as the formation of hollows and hummocks on Bower, M.M., 1961. The distribution of erosion in blanket peat bogs in the
peatland surfaces to an increasing influence of extrinsic factors, Pennines. Transactions of the Institute of British Geographers 29, 17–30.
Bowes, D.R., 1960. A bog-burst in the Isle of Lewis. Scottish Geographical Journal
such as preexisting topography and human impacts, on peat-
76, 21–23.
land vegetation. It also postulates that along the spectrum of Bradshaw, R., McGee, E., 1988. The extent and time-course of mountain blanket
peatlands from intact to eroded, there is a shift from the peat erosion in Ireland. New Phytologist 108(2), 219–224.
dominance of ecological and hydrological controls on peatland Bromley, J., Robinson, M., 1995. Groundwater. In: Hughes, J., Heathwaite, A.L.
function to an increasing importance of geomorphological (Eds.), Hydrology and Hydrochemistry of British Wetlands. Wiley, Chichester, pp.
45–409.
processes. The range of important ecosystem services that Carling, P.A., 1986. Peat slides in Teesdale and Weardale, Northern Pennines, July
peatland systems provide emphasizes the importance of 1983 – description and failure mechanisms. Earth Surface Processes and
understanding controls on peatland systems in order to be able Landforms 11(2), 193–206.
to manage negative external impacts on the ecosystem. This Carling, P.A., Glaister, M.S., Flintham, T.P., 1997. The erodibility of upland soils
and the design of pre-afforestation drainage networks in the United Kingdom.
chapter has demonstrated that peatland form is not simply a
Hydrological Processes 11(15), 1963–1980.
product of ecological processes; rather, it is a dynamic product Charman, D., 2002. Peatlands and Environmental Change. Wiley, Chichester, 301 pp.
of the interaction of ecological, hydrological, and geo- Charman, D.J., 1994. Patterned fen development in northern Scotland – developing
morphological processes. Management and restoration of the a hypothesis from palaeoecological data. Journal of Quaternary Science 9,
important global peatland resource will, therefore, require in- 285–297.
Chiverrell, R.C., 2001. A proxy record of late Holocene climate change from May
creasing engagement of geomorphologists, with the unique
Moss, northeast England. Journal of Quaternary Science 16(1), 9–29.
challenges of working in peatland systems and with the inter- Clymo, R.S., 1983. Peat. In: Gore, A.J.P. (Ed.), Mires: Swamp, Bog, Fen and Moor.
disciplinary work required for a complete understanding of the Ecosystems of the World 4A. Elsevier, Amsterdam.
system. Clymo, R.S., 1984. The limits to peat bog growth. Royal Society Philosophical
Transactions B 303, 605–654.
Comas, X., Slater, L., Reeve, A., 2004. Geophysical evidence for peat basin
morphology and stratigraphic controls on vegetation observed in a northern
References peatland. Journal of Hydrology 295(1–4), 173–184.
Cooper, A., Loftus, M., 1998. The application of multivariate land classification to
vegetation survey in the Wicklow Mountains, Ireland. Plant Ecology 135,
Allott, T.E.H., Evans, M.G., Lindsay, J.L., Agnew, C.T., Freer, J.E., Jones, A., Parnell, 229–241.
M., 2009. Water tables in peak district blanket peatlands. Report to Moors for Coupar, A., Immirzi, P., Reid, E., 1997. The nature and extent of degradation in
the Future. Scottish blanket mires. In: Tallis, J., Meade, R., Hulme, P. (Eds.), Blanket Peat
Almquist-Jacobson, H.A., Foster, D.R., 1995. Toward an integrated model for raised- Degradation: Causes, Consequences, Challenges. British Ecological Society,
bog development: theory and field evidence. Ecology 76(8), 2503–2516. Aberdeen.
Anderson, K., Bennie, J.J., Milton, E.J., Hughes, P.D., Lindsay, R., Meade, R., 2010. Couwenberg, J., Joosten, H., 2005. Self-organization in raised bog patterning: the
Combining LiDAR and IKONOS data for eco-hydrological classification of an origin of microtope zonation and mesotope diversity. Journal of Ecology 93(6),
ombrotrophic peatland. Journal of Environmental Quality 39, 260–273. 1238–1248.
Anderson, P., Tallis, J., Yalden, D. (Eds.), 1997. Restoring Moorland: Peak District Crowe, S.K., 2007. The Natural Re-Vegetation of Eroded Blanket Peat: Implications
Moorland Management Project Phase III Report. Bakewell, Peak District for Blanket Bog Restoration. Unpublished PhD Thesis, University of
Moorland Management Project. Manchester.
Armstrong, A.C., 1995. Hydrological model of peat-mound form with vertically Crowe, S.K., Evans, M.G., Allott, T.E.H., 2008. Geomorphological controls on the
varying hydraulic conductivity. Earth Surface Processes and Landforms 20(5), re-vegetation of erosion gullies in blanket peat: implications for bog restoration.
473–477. Mires and Peat 3, 1–14.
Baird, A.J., Eades, P.A., Surridge, B.W.J., 2008. The hydraulic structure of a raised Dykes, A., 2008. Tensile strength of peat: laboratory measurement and role in Irish
bog and its implications for ecohydrological modelling of bog development. blanket bog failures. Landslides 5(4), 417–429.
Ecohydrology 1(4), 289–298. Dykes, A., Warburton, J., 2007. Mass movements in peat: a formal classification
Baird, A.J., Gaffney, S.W., 1994. Cylindrical piezometer responses in a humified fen scheme. Geomorphology 86(1–2), 73–93.
peat. Nordic Hydrology 25, 167–182. Ellis, C.J., Tallis, J.H., 2001. Climatic control of peat erosion in a North Wales
Ballhorn, U., Siegert, F., Mason, M., Limin, S., 2009. Derivation of burn scar blanket mire. New Phytologist 152, 313–324.
Evans, M., Lindsay, J., 2010. High resolution quantification of gully erosion in
depths and estimation of carbon emissions with Lidar in Indonesian peatlands.
upland peatlands at the landscape scale. Earth Surface Processes and Landforms
Proceedings of the National Academy of Sciences 106(50), 21213–21218.
35(8), 876–886.
Beckwith, C.W., Baird, A.J., Heathwaite, A.L., 2003. Anisotropy and depth-related
Evans, M., Warburton, J., 2007. The Geomorphology of Upland Peat: Pattern,
heterogeneity of hydraulic conductivity in a bog peat. I: Laboratory
Process. Form. Blackwell, Oxford, 262 pp.
measurements. Hydrological Processes 17(1), 89–101. Evans, M., Warburton, J., 2010. Peatland geomorphology and carbon cycling.
Belcher, R., Ingram, G., 1950. A rapid micro-combustion method for the Geography Compass 4(10), 1513–1531.
determination of carbon and oxygen. Analytica Chimica Acta 4, 118–129. Evans, M., Warburton, J., Yang, J., 2006. Sediment budgets for eroding blanket
Belyea, L.R., 1996. Separating the effects of litter quality and microenvironment on peat catchments: global and local implications of upland organic sediment
decomposition rates in a patterned peatland. Oikos 77(3), 529–539. budgets. Geomorphology 79(1–2), 45–57.
Belyea, L.R., Baird, A.J., 2006. Beyond ‘the limits to peat bog growth’: cross scale Evans, M.G., Warburton, J., 2001. Transport and dispersal of organic debris (peat
feedback in peatland development. Ecological Monographs 76(3), 299–322. blocks) in upland fluvial systems. Earth Surface Processes and Landform 26,
Belyea, L.R., Clymo, R.S., 2001. Feedback control of the rate of peat formation. 1087–1102.
Proceedings of the Royal Society of London Series B – Biological Sciences Evans, M.G., Warburton, J., 2005. Sediment budget for an eroding peat-moorland
268(1473), 1315–1321. catchment in Northern England. Earth Surface Processes and Landforms 30(5),
Belyea, L.R., Lancaster, J., 2002. Inferring landscape dynamics of bog pools from 557–577.
scaling relationships and spatial patterns. Journal of Ecology 90(2), 223–234. Ferguson, P., Lee, J.A., Bell, J.N.B., 1978. Effects of sulfur pollutants on growth of
Birnie, R.V., 1993. Erosion rates on bare peat surfaces in Shetland. Scottish sphagnum species. Environmental Pollution 16(2), 151–162.
Geographical Magazine 109(1), 12–17. Foster, D.R., Wright, H.E., 1990. Role of ecosystem development and climate
Blodau, C., 2002. Carbon cycling in peatlands – a review of processes and change in bog formation in central Sweden. Ecology 71(2), 450–463.
controls. Environmental Reviews 10(2), 111–134. Foulds, S.A., Warburton, J., 2006. Wind erosion of blanket peat during a short
Bower, M.M., 1960. Peat erosion in the Pennines. Advancement of Science 64, period of surface desiccation, North Pennines, Northern England. Earth Surface
323–331. Processes and Landforms 32, 481–488.
 Peatland Geomorphology 179

Foulds, S.A., Warburton, J., 2007. Significance of wind-driven rain (wind-splash) in Huang, C.C., 2002. Holocene landscape development and human impact in the
the erosion of blanket peat. Geomorphology 83, 183–192. Connemara Uplands, Western Ireland. Journal of Biogeography 29(2), 153–165.
Francis, I.S., 1990. Blanket peat erosion in a mid-Wales catchment during two Hughes, J.M.R., Heathwaite, A., 1995. Hydrology and hydrochemistry of British
drought years. Earth Surface Processes and Landforms 15(5), 445–456. wetlands. In: Hughes, J., Heathwaite, A. (Eds.), Hydrology and Hydrochemistry
Francis, I.S., Taylor, J.A., 1989. The effect of forestry drainage operations on upland of British Wetlands. Wiley, Chichester, pp. 1–8.
sediment yields: a study of two peat-covered catchments. Earth Surface Hughes, P.D.M., Barber, K.E., 2004. Contrasting pathways to ombrotrophy in three
Processes and Landforms 14(1), 73–83. raised bogs from Ireland and Cumbria, England. Holocene 14(1), 65–77.
Freeman, C., Fenner, N., Ostle, N.J., et al., 2004. Export of dissolved organic Immirzi, C.P., Maltby, E., Clymo, R.S., 1992. The Global Status of Peatlands and
carbon from peatlands under elevated carbon dioxide levels. Nature 430(6996), Their Role in Carbon Cycling. Friends of the Earth, London.
195–198. Ingram, H.A.P., 1982. Size and shape in raised mire ecosystems – a geophysical
Freeman, C., Ostle, N., Kang, H., 2001. An enzymic ’latch’ on a global carbon store model. Nature 297(5864), 300–303.
– a shortage of oxygen locks up carbon in peatlands by restraining a single Ingram, H.A.P., 1983. Hydrology. In: Gore, A.J.P. (Ed.), Ecosystems of the World
enzyme. Nature 409(6817), 149. 4A. Mires: Swamp, Bog, Fen and Moor. Elsevier, Amsterdam, pp. 67–158.
Fronzek, S., Luoto, M., Carter, T.R., 2006. Potential effect of climate change on the Ingram, H.A.P., 1987. Ecohydrology of Scottish Peatlands. Transactions of the Royal
distribution of palsa mires in subarctic Fennoscandia. Climate Research 32, Society of Edinburgh Earth Sciences 78, 287–296.
1–12. Ivanov, K., 1981. Water Movement in Mirelands. Academic Press, London,
Glaser, P.H., 1987. The development of streamlined bog islands in the continental (Translated from the Russian by Thompson, A. and Ingram, H.A.P.). Academic
interior of North America. Arctic and Alpine Research 19(4), 402–413. Press, London.
Glaser, P.H., 1992. Peat landforms. In: Wright, D.B., Coffin, B.A., Aaseng, N. (Eds.), Joosten, J., Haberl, A., Schumann, M., 2008. Degradation and restoration of
The Patterned Peatlands of Minnesota. University of Minnesota Press, St. Paul, peatlands on the Tibetan Plateau. Peatlands International 1, 32–35.
MN, pp. 27–42. Joosten, H., Schumann, M., 2007. Hydrogenetic aspects of peatland restoration in
Glaser, P.H., 1998. The distribution and origin of mire pools. In: Standen, V., Tallis, Tibet and Kalimantan. Global Environmental Research 11, 195–204.
J.H., Meade, R. (Eds.), Patterned Mires and Mire Pools. British Ecological Jurmu, M.C., 2002. A morphological comparison of narrow, low gradient streams
Society, London, pp. 4–25. traversing wetland environments to alluvial streams. Environmental Management
Glaser, P.H., Hansen, B.C.S., Siegel, D.I., Reeve, A.S., Morin, P.J., 2004. Rates, 30, 831–856.
pathways and drivers for peatland development in the Hudson Bay Lowlands, Kenkel, N.C., 1988. Spectral analysis of hummock-hollow pattern in a weakly
northern Ontario, Canada. Journal of Ecology 92(6), 1036–1053. minerotrophic mire. Plant Ecology 78(1), 45–52.
Glaser, P.H., Janssens, J.A., 1986. Raised bogs in eastern North-America – Klaminder, J., Hammarlund, D., Kokfelt, U., Vonk, J.E., Bigler, C., 2010. Lead
transitions in landforms and gross stratigraphy. Canadian Journal of Botany – contamination of subarctic lakes and its response to reduced atmospheric
Revue Canadienne De Botanique 64(2), 395–415. fallout: can the recovery process be counteracted by the ongoing climate
Gorham, E., 1991. Northern Peatlands: role in the carbon cycle and probable change? Environmental Science and Technology 44(7), 2335–2340.
responses to climatic warming. Ecological Applications 1, 182–195. Klinger, L., Taylor, J.A., Franzen, L.G., 1996. The potential role of peatland
Grab, S.W., Deschamps, C.L., 2004. Geomorphological and geoecological controls dynamics in ice-age initiation. Quaternary Research 45(1), 89–92.
and processes following gully development in alpine mires, Lesotho. Arctic, Klove, B., 1998. Erosion and sediment delivery from peat mines. Soil and Tillage
Antarctic and Alpine Research 36(1), 49–58. Research 45(1–2), 199–216.
Graniero, P.A., Price, J.S., 1999. The importance of topographic factors on the Kneale, P.E., 1987. Sensitivity of the groundwater mound model for predicting mire
distribution of bog and heath in a Newfoundland blanket bog complex. Catena topography. Nordic Hydrology 18(4/5), 193–202.
36(3), 233–254. Labadz, J.C., Burt, T.P., Potter, A.W.R., 1991. Sediment yield and delivery in the
Grieve, I., Gilvear, D., 2008. Effects of wind farm construction on concentrations blanket peat moorlands of the Southern Pennines. Earth Surface Processes and
and fluxes of dissolved organic carbon and suspended sediment from peat Landforms 16(3), 255–271.
catchments at Braes of Doune, central Scotland. Mires and Peat 4(9), 1–11. Lappalainen, E., 1996. General review on world peatalnds and peatland resources.
Griffiths, R., 1821. Report relative to the moving bog of Kilmaleady, in the King’s In: Lappaleinen, E. (Ed.), Global Peat Resources. International Peat Society,
County, made by the Order of the Royal Dublin Society. Journal of the Royal Jyväskylä, pp. 53–56.
Dublin Society, 141–144. Lavoie, C., Grosvernier, P., Girard, M., Marcoux, K., 2003. Spontaneous revegetation
Gurnell, A.M., Piegay, H., Swanson, F.J., Gregory, S.V., 2002. Large wood and of mined peatlands: a useful restoration tool? Wetlands Ecology and
fluvial processes. Freshwater Biology 47(4), 601–619. Management 11(1–2), 97.
Gurney, S.D., 2001. Aspects of the genesis, geomorphology and terminology of Leira, M., Cole, E., Mitchell, F., 2007. Peat erosion and atmospheric deposition
palsas: perennial cryogenic mounds. Progress in Physical Geography 25(2), impacts on an oligotrophic lake in Eastern Ireland. Journal of Paleolimnology
249–260. 38(1), 49–71.
Harris, A., Bryant, R.G., 2009. Northern peatland vegetation and the carbon cycle: a Limpens, J., Berendse, F., Blodau, C., et al., 2008. Peatlands and the carbon cycle:
remote sensing approach. In: Baird, A.J., Belyea, L.R., Comas, X., Reeve, A.S., from local processes to global implications – a synthesis. Biogeosciences 5(5),
Slater, L.D. (Eds.), Carbon Cycling in Northern Peatlands. Geophysical 1475–1491.
Monograph Series 184. American Geophysical Union, Washington, DC, pp. Lindsay, R., 1995. Bogs: The Ecology, Classification and Conservation of
79–98. Obrotrophic Mires. Scottish Natural Heritage, Edinburgh, 119 pp.
Hobbs, N.B., 1986. Mire morphology and the properties and behaviour of some Luoto, M., Fronzek, S., Zuidhoff, F.S., 2004. Spatial modelling of palsa mires in
British and foreign peats. Quarterly Journal of Engineering Geology 19, 7–80. relation to climate in northern Europe. Earth Surface Processes and Landforms
Holden, J., 2004. Hydrological connectivity of soil pipes determined by ground- 29(11), 1373–1387.
penetrating radar tracer detection. Earth Surface Processes and Landforms 29(4), Luoto, M., Sepälä, M., 2000. Summit peats (‘peat cakes’) on the fells of Finnish
437–442. Lapland: continental fragments of blanket mires? Holocene 10, 229–241.
Holden, J., Burt, T.P., 2003. Hydraulic conductivity in upland blanket peat: MA, Millennium Ecosystem Assessment, 2005. Ecosystems and Human Wellbeing;
measurement and variability. Hydrological Processes 17(6), 1227–1237. Synthesis. Island Press, Washington, DC.
Holden, J., Burt, T.P., Vilas, M., 2002. Application of ground-penetrating radar to Maltby, E., Legg, C.J., Proctor, M.C.F., 1990. The ecology of severe moorland fire
the identification of subsurface piping in blanket peat. Earth Surface Processes on the North York Moors – effects of the 1976 fires, and subsequent surface
and Landforms 27(3), 235–249. and vegetation development. Journal of Ecology 78(2), 490–518.
Holden, J., Gascoign, M., Bosanko, N.R., 2007. Erosion and natural revegetation Maxa, M., Bolstad, P., 2009. Mapping northern wetlands with high resolution
associated with surface land drains in upland peatlands. Earth Surface Processes satellite images and Lidar. Wetlands 29(1), 248–260.
and Landforms 32(10), 1547–1557. McHugh, M., Harrod, T., Morgan, R., 2002. The extent of soil erosion in Upland
Holden, J., Kirkby, M.J., Lane, S.N., Milledge, D.G., Brookes, C.J., Holden, V., England and Wales. Earth Surface Processes and Landforms 27(1), 99–107.
McDonald, A.T., 2008. Overland flow velocity and roughness properties in McMorrow, J.M., Cutler, M.E.J., Evans, M.G., Al-Roichdi, A., 2003. Characterising
peatlands. Water Resources Research 44(6), W06415. http://dx.doi.org/10.1029/ upland peat composition with hyperspectral data. International Journal of Remote
2007WR006052. Sensing 24, 1–13.
Honohane, J., 1697. Kapanhane bog flow. Philosophical Transactions of the Royal Mills, A.J., 2002. Peat slides: Morphology, Mechanisms and Recovery. PhD Thesis,
Dublin Society Part XIX (Oct), 714–716. University of Durham.
180 Peatland Geomorphology

Milton, E.J., Hughes, P.D., Anderson, K., Schultz, J., Hill, C.T., Lindsay, R., 2004. Slater, L., Comas, X., Ntarlagiannis, D., Moulik, M.R., 2007. Resistivity-based monitoring
Development and testing methods. In: Proceedings of the Peterborough Remote of biogenic gases in peat soils. Water Resources Research 43(10), W10430.
Sensing Workshop 30 September 2004. English Nature, Peterborough, pp. Slater, L.D., Reeve, A., 2002. Investigating peatland stratigraphy and hydrogeology
26–36. using integrated electrical geophysics. Geophysics 67(2), 365–378.
Minke, M., Donner, N., Karpov, N., de Klerk, P., Joosten, H., 2009. Patterns in Stewart, A.J., Lance, A.N., 1991. Effects of moor drains on the hydrology and
vegetation composition, surface height and thaw depth in polygon mires in the vegetation of North Pennines blanket bog. Journal of Applied Ecology 28,
Yakutian Arctic (NE Siberia): a microtopographical characterisation of the active 1105–1117.
layer. Permafrost and Periglacial Processes 20(4), 357–368. Swanson, F.J., Janda, R.J., Dunne, T., Swanston, D.N. (Eds.), 1982. USDA General
Minke, M., Donner, N., Karpov, N.S., de Klerk, P., Joosten, H., 2007. Distribution, Technical Report PNW – 141.
diversity, development and dynamics of polygon mires: examples from NE Tallis, J.H., 1964. Studies on Southern Pennine peats. II: The pattern of erosion.
Yakutia (NE Siberia). Peatlands International 1, 36–40. Journal of Ecology 52, 333–344.
Mulvaney, J., 1879. Avalanche of peat in the Falkland Islands 1(12), 803–805,New Tallis, J.H., 1985. Mass movement and erosion of a southern Pennine blanket peat.
Monthly Series. Proceedings of the Royal Geographical Society and Monthly Journal of Ecology 73, 283–315.
Record of Geography. New Monthly Series 1(12), 803–805. Tallis, J.H., 1997a. The southern Pennine experience: an overview of blanket mire
Ovenden, L., 1990. Peat accumulation in northern wetlands. Quaternary Research degradation. In: Tallis, J.H., Meade, R., Hulme, P.D. (Eds.), Blanket Mire
33(3), 377–386. Degradation: Causes, Consequences and Challenges. Macaulay Land Use
Ozesmi, S.L., Bauer, M.E., 2002. Satellite remote sensing of wetlands. Wetlands Research Institute, Aberdeen, pp. 7–16.
Ecology and Management 10(5), 381–402. Tallis, J.H., 1997b. The pollen record of Empetrum nigrum in southern Pennine
Parviainen, M., Luoto, M., 2007. Climate envelopes of mire complex types in peats: implications for erosion and climate change. Journal of Ecology 85(4),
Fennoscandia. Geografiska Annaler: Series A, Physical Geography 89(2), 455–465.
137–151. Tallis, J.H., Yalden, D., 1983. Peak District Moorland Restoration Project Phase II
Pavey, B., Saint-Hilaire, A., Courtenay, S., Ouarda, T., Bobée, B., 2007. Exploratory Report: Re-Vegetation Trials. Peak Park Joint Planning Board, Bakewell.
study of suspended sediment concentrations downstream of harvested peat Vitek, J.D., Rose, E.M., 1980. Preliminary observations on a patterned fen in the
bogs. Environmental Monitoring and Assessment 135(1), 369–382. Sangre de Cristo Mountains, Colorado, U.S.A. Zeitschrift für Geomorphogie, N.F
Pawson, R.R., Lord, D.R., Evans, M.G., Allott, T.E.H., 2008. Fluvial organic carbon 24, 393–404.
flux from an eroding peatland catchment, southern Pennines, UK. Hydrology and Vitt, D.H., Halsey, L.A., Zoltai, S.C., 1994. The bog landforms of continental western
Earth System Science 12, 625–634. Canada in relation to climate and permafrost patterns. Arctic and Alpine
Pearsall, W., 1950. Mountains and Moorlands. Fontana, London. Research 26(1), 1–13.
Philips, J., Tallis, J., Yalden, D. (Eds.), 1981. Peak Park Joint Planning Board, Bakewell. Waddington, J.M., Harrison, K., Kellner, E., Baird, A.J., 2009. Effect of atmospheric
Poulter, B., Christensen, N.L., Halpin, P.N., 2006. Carbon emissions from a pressure and temperature on entrapped gas content in peat. Hydrological
temperate peat fire and its relevance to interannual variability of trace Processes 23(20), 2970–2980.
atmospheric greenhouse gases. Journal of Geophysical Research 111, D06301. Waddington, J.M., Roulet, N.T., 1997. Groundwater flow and dissolved carbon
http://dx.doi.org/10.1029/2005JD006455. movement in a boreal peatland. Journal of Hydrology 191(1–4), 122–138.
Quinton, W.L., Hayashi, M., Carey, S.K., 2008. Peat hydraulic conductivity in cold Walker, D., 1970. Direction and rate in some British post-gloacial hydroseres.
regions and its relation to pore size and geometry. Hydrological Processes In: Walker, D., West, R.G. (Eds.), Studies in the Vegetational History of the
22(15), 2829–2837. British Isles. Cambridge University Press, Cambridge, pp. 117–139.
Radley, J., 1962. Peat erosion on the high moors of Derbyshire and West Walling, D.E., Webb, B.W., 1981a. Water quality. In: Lewin, J. (Ed.), British Rivers.
Yorkshire. East Midlands Geographer 3(17), 40–50. Allen and Unwin, London, pp. 126–169.
Radley, J., 1965. Significance of major moorland fires. Nature 205, 1254–1259. Walling, D.E., Webb, B.W., 1981b. Reliability of Suspended Sediment Load Data. In:
Reeve, A.S., Warzocha, J., Glaser, P.H., Siegel, D.I., 2001. Regional ground-water Erosion and Sediment Transport Measurement: Symposium. IAHS Publication
flow modeling of the glacial lake Agassiz Peatlands, Minnesota. Journal of No. 133. IAHS Press, Wallingford, pp. 177–194.
Hydrology 243(1–2), 91–100. Warburton, J., 2003. Wind splash erosion of bare peat on UK upland moorlands.
Rhodes, N., Stevenson, T., 1997. Palaeoenvironmental evidence for the importance Catena 52, 191–207.
of fire as a cause of erosion of British and Irish blanket peats. In: Tallis, J., Warburton, J., Evans, M.G., 2011. Geomorphic, sedimentary and
Meade, R., Hulme, P. (Eds.), Blanet Mire Degradation: Causes, Consequences palaeoenvironmental significance of peat blocks in alluvial river systems.
and Challenges. British Ecological Society, Aberdeen, pp. 64–78. Geomorphology 130(3–4), 101–114.
Rietkerk, M., Dekker, S.C., Wassen, M.J., Verkroost, A.W.M., Bierkens, M.F.P., 2004. Warburton, J., Holden, J., Mills, A.J., 2004. Hydrological controls of surficial mass
A putative mechanism for bog patterning. American Naturalist 163(5), 699–708. movements in peat. Earth-Science Reviews 67(1–2), 139–156.
Rothwell, J., Taylor, K., Ander, E., Evans, M., Daniels, S., Allott, T., 2009. Arsenic Watters, J., Stanley, E., 2007. Stream channels in peatlands: the role of biological
retention and release in ombrotrophic peatlands. Science of the Total processes in controlling channel form. Geomorphology 89(1–2), 97–110.
Environment 407(4), 1405–1417. Whittington, P., Strack, M., van Haarlem, R., et al., 2007. The influence of peat
Rothwell, J.J., Evans, M.G., Daniels, S.M., Allott, T.E.H., 2008. Peat soils as a volume change and vegetation on the hydrology of a kettle-hole wetland in
source of lead contamination to upland fluvial systems. Environmental Pollution southern Ontario, Canada. Mires and Peat 2, 1–14.
153(3), 582–589. Wishart, D., Warburton, J., 2001. An assessment of blanket mire degradation and
Rothwell, J.J., Evans, M.G., Lindsay, J.B., Allott, T.E.H., 2007. Scale-dependent peatland gully development in the Cheviot Hills, Northumberland. Scottish
spatial variability in peatland lead pollution in the southern Pennines, UK. Geographical Journal 117(3), 185–206.
Environmental Pollution 145(1), 111–120. Worrall, F., Harriman, R., Evans, C.D., et al., 2004. Trends in dissolved organic
Rydin, H., Jeglum, J., 2006. The Biology of Peatlands. Oxford University Press, carbon in U.K. rivers and lakes. Biogeochemistry 70(3), 369–402.
Oxford, 343 pp. Yang, J., Dykes, A., 2006. The liquid limit of peat and its application to the
Selkirk, J.M., Saffigna, L.J., 1999. Wind and water erosion of a peat and sand area understanding of Irish blanket bog failures. Landslides 3(3), 205–216.
on subantarctic Macquarie Island. Arctic, Antarctic and Alpine Research 31(4), Yeloff, D.E., Labadz, J.C., Hunt, C.O., 2006. Causes of degradation and erosion of a
412–420. blanket mire in the southern Pennines, UK. Mires and Peat 1, 1–18.
Seppala, M., 1982. An experimental study of the formation of palsas. In: Alberta Yeloff, D.E., Labadz, J.C., Hunt, C.O., Higgitt, D.L., Foster, I.D.L., 2005. Blanket
French, H. (Ed.), Proceedings of the Fourth Canadian Permafrost Conference. peat erosion and sediment yield in an upland reservoir catchment in the
NRC, Ottawa, pp. 32–42. southern Pennines, U.K. Earth Surface Processes and Landforms 30(6),
Seppala, M., 2003. Surface abrasion of palsas by wind action in Finnish Lapland. 717–733.
Geomorphology 52(1–2), 141–148. Zuidhoff, F.S., 2002. Recent decay of a single palsa in relation to weather
Shimwell, D.W., 1974. Sheep grazing intensity in Edale, Derbyshire, 1692–1747, and conditions between 1996 and 2000 in Laivadalen, northern Sweden. Geografiska
its effect on blanket peat erosion. Derbyshire Archaeological Journal 94, 35–40. Annaler Series A – Physical Geography 84A(2), 103–111.
 Peatland Geomorphology 181

Biographical Sketch

Martin Evans is a reader in geomorphology at the University of Manchester. He studied for his first degree at the
University of Oxford before moving to the University of British Columbia where he completed an MSc and PhD
in physical geography. His work in British Columbia examined the impacts of climate change on alpine sediment
budgets; after completing his graduate work, he moved to the University of Durham where he began postdoctoral
work applying sediment budgeting techniques to peatland systems.
Since moving to the University of Manchester, the scope of this work has expanded to consider both the
geomorphology of peatland systems and the impacts of geomorphological change on peatland function. Major
foci have been the geomorphological impacts of peatland restoration, the impact of peat erosion on peatland
hydrology, and the impact of peat erosion and restoration on peatland carbon cycling.
Martin currently convenes the BSG working group on carbon and geomorphology. He is an associate editor of
Transactions of the Institute of British Geographers, a member of the UK NERC peer review college, a member of
the executive committee of the British Society for Geomorphology, and is currently a panel member for the IUCN
Committee of Enquiry on UK peatlands.
 12.12 Ecogeomorphology of Salt Marshes
S Fagherazzi, DM FitzGerald, RW Fulweiler, and Z Hughes, Boston University, Boston, MA, USA
PL Wiberg and KJ McGlathery, University of Virginia, Charlottesville, VA, USA
JT Morris, Belle Baruch Institute for Marine and Coastal Sciences, University of South Carolina, Columbia, SC, USA
TJ Tolhurst, School of Environmental Sciences, University of East Anglia, Norwich, UK
LA Deegan and DS Johnson, Marine Biological Laboratory, Woods Hole, MA, USA
r 2013 Elsevier Inc. All rights reserved.

12.12.1 Effects of Invertebrates and Vegetation on Marsh-Sediment Transport 183

12.12.1.1 Effects of Vegetation–Sediment Transport Interactions on Marsh Elevation 183
12.12.1.2 Effects of Vegetation and Invertebrates on the Erosion of the Marsh Edge 184
12.12.1.3 Effects of Vegetation–Sediment Transport Interaction on Marsh–Channel Networks 186
12.12.2 Feedbacks between Salt-Marsh Vegetation and Platform Elevation 187
12.12.3 Long-Term Marsh Stability and Biogeochemical Cycling 189
12.12.4 Modeling Intertidal Ecogeomorphology 192
Acknowledgments 195
References 195

Glossary muddy sediments enclose isolated sandy layers (sand

Belowground production The belowground production lenses).
of organic material from roots and rhizomes. Macrophytes Aquatic plants that are large enough to be
Benthic macroalgae Macroscopic algae that cover the apparent to the naked eye. Macrophytes may be emergent,
ocean bottom. submergent, or floating.
Biofilm Microorganisms (diatoms, cyanobacteria, and Microphytobenthos The assemblage of microalgae
green algae) and their extracellular products bound to a (mostly diatoms and cyanobacteria) living at the bottom of
surface. the ocean.
Bioturbation Mixing and reworking of sediment particles Nutrient enrichment Increased supply of nutrients, such
by animals and plants. as nitrogen and phosphorus, to an ecosystem, often caused
Cyanobacteria Bacteria that obtain energy through by human activities.
photosynthesis, also called blue-green algae. Rhizome An underground stem, usually horizontal, from
Detritus Dead particulate organic material, either which roots and shoots can grow.
accumulating in soil or carried by water in suspension. Ridge and runnel Mudflat bedforms characterized by
Diatoms Unicellular algae with a silica cell wall, which parallel ridges separated by shallow troughs (runnels).
form extensive biofilms. These bedforms are usually equispaced with a depth of tens
Extracellular polymeric substances (EPSs) A of centimeters.
mucilaginous extracellular carbohydrate matrix secreted by Root scalping The mechanical removal of the canopy
organisms (diatoms, cyanobacteria, and green algae) in mat, including roots, from a surface (for example, from a
biofilms. The EPS forms a cohesive matrix that helps to salt-marsh platform).
structure the sediment microbial environment. Seagrass meadows Marine flowering plants, with long
Flasers Sedimentary structures characterized by sand and narrow leaves, that grow in large patches (meadows).
ripples partly infilled with mud, usually created by Skimming flow The flow above a vegetation canopy. In a
intermittent flows in tidal environments. In flasers, the sand skimming flow, the entire velocity profile, and related
deposits are larger than the mud deposits. boundary layer, is displaced upward.
Lenticular bed A sedimentary layer that is thickest at one Wavy bedding Sedimentary structures characterized by
point and thins away in every direction, forming a lens. alternating rippled sand and mud layers. In wavy bedding,
Lenticular beds are typical of tidal environments where mud and sand deposits are equal.

Fagherazzi, S., FitzGerald, D.M.,. Fulweiler, R.W., et al., 2013.

Ecogeomorphology of salt marshes. In: Shroder, J. (Editor in Chief), Butler,
D.R., Hupp, C.R. (Eds.), Treatise on Geomorphology. Academic Press, San
Diego, CA, vol. 12, Ecogeomorphology, pp. 182–200.

182 Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00329-8

 Ecogeomorphology of Salt Marshes 183

Abstract

The coupling of geomorphological and ecological processes is critical for the maintenance or disappearance of salt marshes.
Emergent macrophytes dampen wave- and tide-generated shear stresses, promoting sediment deposition and marsh for-
mation. The complex interactions among marsh primary productivity, sea level, and sedimentation determine the equi-
librium elevation of the marsh platform and its resilience against external drivers. The strength of the root system can affect
the ability of marsh edge scarps to resist erosion. Some marshes are commonly inhabited by a number of species of crab,
whose burrows have been implicated in marsh-edge erosion, whereas mussels and other bivalves can be very important in
stabilizing marsh sediment by both slowing wave and current velocities and binding sediment to the root mat. Nutrient
enrichment has the potential to invoke negative feedbacks that will ultimately affect marsh geomorphic configuration and
biogeochemical cycling, by altering plant production and community characteristics and increasing microbial de-
composition of organic matter. In this chapter, we present a broad overview of the feedbacks between biota and sedi-
mentological processes in salt marshes, including recent numerical models that have been utilized to study the
ecogeomorphic evolution of intertidal areas.

12.12.1 Effects of Invertebrates and Vegetation on 12.12.1.1 Effects of Vegetation–Sediment Transport

Marsh-Sediment Transport Interactions on Marsh Elevation

Whereas tidal flats dominate in areas of shallow coastal bays Sediment deposition on tidal salt marshes occurs when the
characterized by persistent sediment resuspension due to tides vegetated marsh platform is flooded with water carrying a load
and waves, the presence of Spartina alterniflora (or other of suspended sediment. Deposition is greatest near marsh
emergent macrophytes) dampens wave- and tide-generated edges bordering tidal creeks or open water because: (1) creeks
shear stresses and promotes sediment deposition, thereby and bays are primary sediment sources; (2) the adjacent marsh
allowing salt marshes to form and persist in regions with edges are flooded more frequently than higher-elevation, in-
adequate sediment supply. land portions of the marsh; and (3) turbulent mixing, which
Salt marshes occupy a narrow range of elevations, generally maintains particles in suspension, is dampened as flow enters
between mean sea level (MSL) and mean higher high water. the marsh canopy. Whereas observations for marshes adjacent
Marshes are subdivided into low, mid and high marsh, de- to tidal creeks indicate maximum deposition on the creek
pending on vegetation type and tidal flooding frequency. Along bank (e.g., Leonard and Luther, 1995; Christiansen et al.,
the Mid-Atlantic and Northeast coastline of the United States, 2000), wave energy near marsh edges fronting open water may
S. alterniflora dominates in the low marsh where tidal flooding shift the depositional maximum inland several meters or
and inorganic sediment deposition are the primary controls more, depending on the density of marsh-edge vegetation,
on surface accretion. In the mid- and high marsh, other marsh water depths, and wave energy.
grasses, such as S. patens, Juncus roemerianus, and Distichlis The effect of marsh vegetation on flow has been examined
spicata typically replace S. alterniflora, and changes in marsh- in a number of field and laboratory studies. Leonard
surface elevation depend on a combination of inorganic sedi- and Luther (1995) documented decreases in flow velocity and
ment deposition, surface and subsurface biomass accumulation turbulence intensity with distance from the creek edge and
and decomposition, and shallow and deep subsidence. vegetation density for marshes in Louisiana and Florida,
Low-marsh elevation is primarily controlled by feedbacks which were dominated by S. alterniflora and J. roemerianus,
between flooding frequency and vegetation-enhanced sedi- respectively. Flow speeds within the dense vegetation canopy
ment deposition. If marsh elevation increases faster than sea were of the order of 2–3 cm s–1. Turbulence intensities drop-
level, flooding frequency decreases and deposition rates drop. ped by an order of magnitude within 10 m of the tidal
If marsh elevation increases more slowly than sea-level rise, creek–marsh edge. Wave energy has the potential to be re-
flooding frequency increases, thereby increasing the potential duced by over 80% when traveling over salt marshes (Möller
for sediment deposition and also the potential for water- et al., 1999; Knuston et al., 1982). Both canopy height and
logging marsh vegetation (Morris et al., 2002). In the latter stem density are important in the dissipation or attenuation of
case, the marsh can evolve into a tidal flat. The marsh–tide flat wave energies, as density affects horizontal friction and vege-
boundary can take several forms, depending on tidal range, tation height influences boundary-layer depth (Möller, 2006).
wave energy, marsh vegetation, and sediment supply (Mariotti The structure of the marsh vegetation is particularly important
and Fagherazzi, 2010). The marsh boundary along eroding near the marsh edge, as most wave forces are dissipated within
edges is characterized by scarps and/or slumps, whereas ac- the first 50 m of the marsh (Möller et al., 1999).
creting marsh boundaries can be more gently sloped. Marsh Christiansen et al. (2000) measured flow, turbulence, sus-
platforms are typically dissected by a dendritic network of tidal pended sediment, and sediment deposition on an S. alterni-
channels, the morphology and stability of which are strongly flora marsh on the Eastern Shore of Virginia over a range of
linked to the presence of marsh vegetation (Kirwan and tidal cycles along a transect extending B50 m from a tidal
Murray, 2007). The sediment–vegetation interactions that creek–marsh edge into the marsh interior. Flow velocities
control marsh elevation, marsh-boundary erosion and marsh- within the marsh canopy never exceeded 1 cm s–1 and shear
channel networks, as well as effects of invertebrates, are con- velocities never exceeded 0.05 cm s–1, irrespective of tidal
sidered in this section. amplitude. In addition to a clear pattern of reduction in
184 Ecogeomorphology of Salt Marshes

turbulent energy (TKE) with distance into the marsh, Chris- environments) or sediment removal. Surface elevation table
tiansen et al. (2000) observed a flood–ebb asymmetry in TKE, (SET; Cahoon et al., 2002) and marker-horizon (Cahoon and
with higher TKE near the marsh edge during the flooding Turner, 1989) measurements of marsh-surface accumulation
tides, although flow speeds were higher during ebbing tides. and subsidence within the root zone and below indicate that
The effect of the drop in TKE and low values of shear vel- mid-high mainland marshes in the VCR are increasing in
ocity on sediment transport is evident in the time series of surface elevation at a rate at least comparable to sea-level rise,
suspended-sediment concentration (Figure 1) measured by with a greater percentage of elevation change due to surface
Christiansen et al. (2000). The decrease in suspended- accretion (deposition) in more frequently flooded portions of
sediment concentration with distance from the tidal channel the marsh. Although tidal shear stresses on a vegetated marsh
is evidence of sediment deposition during flood tides. Low surface are too low to resuspend sediment, rainfall runoff
values of suspended-sediment concentration during ebb tides from an exposed but saturated marsh surface is a potentially
show that sediment is not resuspended from the vegetated important mechanism for sediment export from salt marshes
marsh platform as tidal water drains from the marsh, con- (Mwamba and Torres, 2002; Turaski, 2002), particularly where
sistent with the low measured values of shear velocity. In- raindrop detachment of sediment is able to create high
creases in tidal amplitude were observed to correlate with sediment concentrations in the runoff flow.
increases in suspended-sediment concentration in the tidal
creek, but not in the marsh interior (Christiansen et al., 2000),
suggesting deposition rates vary with tidal amplitude and 12.12.1.2 Effects of Vegetation and Invertebrates on the
flood duration. Erosion of the Marsh Edge
Estimated annual accumulation rates on the basis on Whereas aboveground vegetation acts to slow flow velocities,
sediment-trap measurements and sediment-flux gradients trap sediment, and attenuate waves and turbulence, below-
(1.0–1.2 g cm–2 yr–1 on the creek bank and 0.6–0.7 g cm–2 ground roots and rhizomes help to stabilize marsh sediment
yr–1 in the interior of the low marsh) are higher than 210-Pb- (Coops et al., 1996) and play an important role in limiting
based estimates of longer-term deposition rates on the mar- erosion along the marsh edge. Loss of marsh area through
shes of the Virginia Coast Reserve, VCR (B2 mm yr–1 or marsh-edge erosion occurs in many coastal environments
0.2 g cm–2 yr–1, Kastler and Wiberg, 1996) and rates of (e.g., Day et al., 1998; Wray et al., 1995; Schwimmer, 2001;
sea-level rise (B4 mm yr–1) (see Figure 2). Higher rates of Hughes and Paramor, 2004; Wilson and Allison, 2008), with
short-term than long-term deposition, commonly occurring rates ranging from B0.1 to 43 m yr–1. Analysis of historical
in marine environments (e.g., Sommerfield, 2006), suggest aerial photos in the VCR indicates erosion rates of 50–100 m
the presence of compaction/subsidence (including de- in 40 years on marshes located on a back barrier, marsh
composition), and periods of nondeposition (unlikely in tidal island, and mainland marsh peninsula, with various
Tidal elevation (m)

1.0

0.8

0.6

0.4
−2.5 −2.0 −1.5 −1.0 −0.5 0 0.5 1.0 1.5 2.0 2.5

100
Concentration (mg l−1)

Station 1 (creek bank)

80
Station 3 (transition)
60 Station 5 (interior)

40

20

0
−2.5 −2.0 −1.5 −1.0 −0.5 0 0.5 1.0 1.5 2.0 2.5
Time relative to high tide (h)
Figure 1 Suspended sediment concentration (bottom panel) above a salt marsh as a function of time through a tidal cycle and distance from
the tidal creek. Tidal elevation is shown in the top panel. Stations 1, 3, and 5 were 0, 7, and 46 m from the edge of the tidal creek. The study
site was located on a mainland marsh in the Virginia Coast Reserve. Reproduced from color version of Figure 8 in Christiansen, T., Wiberg, P.L.,
Milligan, T.G., 2000. Flow and sediment transport on a tidal salt marsh surface. Estuarine, Coastal and Shelf Science 50(3), 315–331.
 Ecogeomorphology of Salt Marshes 185

Deposition on Phillips creek marsh

0.1

0.08 Individual tidal cycles

kg m−2 tide−1
0.06

0.04

0.02

0
0 5 10 15 20 25

0.4
Included very large storm 21 May−3 Jun, 1997
3 Jun−17 Jun, 1997
0.3 17 Jun−2 Jul, 1997
2 Feb−14 Feb, 1998
kg m−2 tide−1

0.2
Included large storm

Included small storm

0.1

0
0 5 10 15 20 25
Distance from creek (m)
Figure 2 Sediment deposition on a salt marsh as a function of distance from the tidal creek. Deposition was measured on netted plates that
were flush with the marsh surface. Top panel: deposition measured over individual tidal cycles during fair-weather conditions. Bottom panel:
deposition measured during four 2-week periods, divided by the number of tidal cycles. The 17 June–2 July period was fair weather. The periods
21 May–3 June and 3 June–17 June included a small and large storm, respectively. The 2 February–14 February period included a very large
storm. Modified from Christiansen, T., 1998. Sediment deposition on a tidal salt marsh. Dissertation, University of Virginia, 134 pp.

orientations relative to the prevailing winds. Eroding marshes affect rates of marsh-edge erosion, except inasmuch as it
are typically characterized by a relatively steep scarp, com- relates to marsh sediment characteristics.
monly accompanied by slumping, undercutting, and/or root Salt marshes of the eastern USA are inhabited by a number
scalping (removal of the active root layer forming a denuded of species of crabs. Sesarma reticulatum, nocturnal crabs that
terrace) (Figure 3). can create interconnected and expansive burrows inhabited by
The strength of the marsh root system, determined through multiple individuals, have been implicated in marsh-edge
measures of biomass, root diameter, root length, and tensile erosion in North Carolina (Allen and Curran, 1974) and die-
strength, can affect the ability of marsh-edge scarps to resist offs of cordgrass in New England salt marshes (Holdredge
erosion (van Eerdt, 1985). Root strength also decreases with et al., 2009). Dense, interconnected burrows of Chasmagnathus
depth, thus making marsh edges susceptible to undercutting. granulatus near marsh-edge sites in salt marshes in the
A study of root effects on scarp erosion in the Oosterschelde southwest Atlantic (e.g., Argentina) decrease sediment shear
(the Netherlands) found that scarps dominated by a species strength and increase permeability and water content, thus
with weaker root strength (Limonium vulgare) experienced reducing soil strength and increasing erosion potential
greater erosion than those dominated by vegetation with (Escapa et al., 2008).
stronger root systems (S. anglica) (van Eerdt, 1985). By con- The fiddler crab Uca pugnax (Atlantic marsh or mud fid-
trast, recent laboratory and field measurements by Feagin et al. dler), abundant throughout marshes along the mid-Atlantic
(2009) found that marsh vegetation does not significantly and northeast coast of USA, particularly dominates the muddy
186 Ecogeomorphology of Salt Marshes

(b)

(a) (c)

Figure 3 Examples of different mechanisms of marsh boundary degradation by wave erosion: (a) slumping; (b) undercutting; and (c) root
scalping (removal of the active root layer forming a denudated terrace). These marsh sites are located in Hog Island Bay in the Virginia Coast
Reserve.

sediments near the creek edges and transitional zones (Mon- in the VCR is r0.5 m in 40 years, compared to rates of
tague, 1980; Allen and Curran, 1974). Burrow densities 1.0–1.5 m in 40 years at sites lacking protective oyster rock.
near creek banks are estimated to have an average range
of 40–120 per m2, with some reported estimates as large
12.12.1.3 Effects of Vegetation–Sediment Transport
as 700 per m2 (Montague, 1980). Mud-fiddler burrows
Interaction on Marsh–Channel Networks
are typically 1–2 cm in diameter, 15–25 cm deep, and
have twisting, U-shaped morphologies that are commonly Marshes are typically dissected by a network of sinuous tidal
interconnected with the burrows of other crab individuals channels of varying scale. These networks help to convey water
(Montague, 1980; Allen and Curran, 1974). Although rela- and sediment onto and off of the marsh platforms during
tively little work has been performed on the role of fiddler flood and ebb tides, respectively. Unlike fluvial meandering
burrows in marsh erosion, U. pugnax burrows may have an channels, tidal channels on vegetated salt marshes are rela-
effect on erosion similar to those of Sesarma or Chasmagnathus, tively fixed in position, at least on timescales of decades
due to their interconnectedness and proximity to the (e.g., Friedrichs and Perry, 2001; Kastler and Wiberg, 1996;
marsh edge. Fagherazzi et al., 1999). Their equilibrium cross section
Ribbed mussels commonly occur at very high densities generally varies with the size of the tidal prism (D’Alpaos
near the marsh edge and their numbers tend to decrease sig- et al., 2010), suggesting that channel widening will accompany
nificantly with distance from the edge (Bertness, 1984; Franz, increases in tidal prisms associated with sea-level rise.
1993, 2001). Bertness (1984) has argued that the presence of Model simulations of marsh–channel-network evolution
Geukensia demissa, a ribbed mussel that commonly attaches to by Kirwan and Murray (2007) indicate that, in the absence of
cordgrass stems, can increase the productivity of S. alterniflora. marsh vegetation, increasing rates of sea-level rise (or de-
However, ribbed mussels can also be very important in sta- creasing sediment supply) can lead to water depths on mar-
bilizing marsh sediment and reducing erosion. This occurs by shes that exceed the limit for plant survival, thereby causing
both slowing wave and current velocities and binding sedi- the marsh to evolve into a tidal flat. By contrast, the presence
ment to the root mat (Bertness, 1984). Oysters lining the edge of marsh vegetation allows marsh-surface accretion to gener-
and surface of marshes may serve a similar protective role to ally keep pace with sea-level rise, thereby preserving the marsh
ribbed mussels in that they have the potential to reduce wave surface and its channel network (Kirwan and Murray, 2007).
energies and stabilize the marsh sediment against erosion Temporary, localized disturbance to marsh vegetation, on
(Meyer et al., 1997). Piazza et al. (2005) have shown that the timescales of 5–10 years, can lead to channel expansion and,
creation of cultched oyster reefs near the edges of marshes can ultimately, to wetland loss, particularly under conditions of
reduce shoreline erosion in areas with low-energy wave con- rapid sea-level rise and/or reduced sediment supply (Kirwan
ditions. Natural patches of oyster shells along the marsh edge et al., 2008). Dynamic patchiness in marsh vegetation has also
may provide similar protection. For example, the erosion rate been suggested to contribute to higher channel densities
of a mainland marsh edge with relict oyster rock just offshore (Temmerman et al., 2007).
 Ecogeomorphology of Salt Marshes 187

12.12.2 Feedbacks between Salt-Marsh Vegetation accretion increased significantly above that of the controls
and Platform Elevation (Figure 5). Plant biomass increased dramatically in fertilized
plots beginning in 1997, the second year of the experiment,
Long-term measurements in salt marshes at North Inlet, SC and the elevations of these sites increased faster than controls
and Plum Island, MA, USA, of primary production of the salt- due to trapping of sediment by a dense canopy of S. alterni-
marsh cordgrass, S. alterniflora, and marsh elevation are illus- flora. Since about 2001, coincident with the declining phase of
trative of the complex interactions that exist among marsh the lunar nodal cycle, changes in elevation of fertilized and
primary productivity, sea level, and sedimentation. These control sites have been less dramatic.
interactions have been studied intensively at North Inlet for a The long-term stability of coastal wetlands is explained by
sufficiently long time to have paid off in terms of our under- interactions among sea level, primary production, and sedi-
standing of mechanisms that promote equilibrium between ment accretion that regulate the elevation of the sediment
the marsh and mean sea level (MSL) or, alternatively, that lead surface toward equilibrium with MSL (Morris et al., 2002).
to instability when the rate of sea-level rise exceeds a tipping We have shown that this equilibrium is adjusted upward
point. The model that has emerged from this research (Morris by increased production of S. alterniflora and downward by
et al., 2002) has been instrumental in changing our views an acceleration in the rate of sea-level rise. However, adjust-
about salt-marsh biogeomorphology. ments in marsh-surface elevation are slow in comparison to
Feedbacks between salt-marsh vegetation and sediments interannual anomalies and long-period cycles of sea level, and
actually serve to stabilize the marsh against changes in sea this lag in the response of marsh elevation explains the
level, at least to a degree. The feedback works because the variability in annual primary productivity that was described
growth of the vegetation and its ability to trap sediment are above (Figure 4).
affected by the relative elevation of the marsh within the tidal We developed a theoretical model that predicts how marsh
frame. The growth of marsh vegetation has been shown to elevation and productivity adjust to changes in the rate of sea-
increase with positive anomalies in MSL (Figure 4). However, level rise (Morris et al., 2002). The logic of the model derives
marsh vegetation responds positively to rising relative sea level in part from the observation that the sedimentation of mineral
only when marsh elevation is super-optimal for growth. matter onto the marsh surface is a positive function of the
At both North Inlet and Plum Island, the marsh platform is standing density of aboveground biomass Bs (g cm–2), dis-
situated at an elevation that is high in the tidal frame. cussed earlier, and is also proportional to the time that a
This constraint on productivity is an important factor in marsh remains flooded. Inundation time is roughly pro-
maintaining elevation because a rise in relative sea level at portional to the depth D (cm) of the marsh surface below
these high elevations brings about an increase in production mean high water (MHW). In addition to mineral inputs, or-
and biomass density (Morris et al., 2002). This enhances ganic matter contributes to the total accumulation rate. This is
sediment deposition by increasing the efficiency of sediment largely from the production of belowground organic matter,
trapping (Gleason et al., 1979; Leonard and Luther, 1995; which is given by the product of kr Br, where Br represents
Yang, 1998). belowground production and kr is the fraction of production
This was demonstrated experimentally at North Inlet that is refractory to decay. The model was simplified by
(Morris et al., 2002). Where primary production and biomass making Br equal to a proportion of aboveground biomass
density were increased by fertilization, the rate of sediment density Bs and redefining kr as the product of that

60 Last
Net primary production (mol C m−2 yr−1)

North Inlet Model fit

r 2 = 0.25
Plum Island
Observed controls
treatment 3.6 mm yr−1
50 r 2 = 0.85
p < 0.001 p < 0.0001 48 Observed fertilized
NAVD 88 relative elevation (cm)

40 46

30 44
1.2 mm yr−1
42
20 20 ks = 0.1029 cm2 g−1 yr−1
19 q = 0.0017 cm−2 yr−1
10 Slope = Slope = 2.9 mm yr−1
1 mol C m−2 yr−1 cm−1 4.3 mol C m−2 yr−1 cm−1 18 kr = 0.6 cm3 g−1 yr−1
0 17
65 70 75 80 140 145 150
Mean high water level (cm) 16
1996 1998 2000 2002 2004 2006 2008
Figure 4 Net above-ground primary production of Spartina
Year
alterniflora from fixed sites at North Inlet, SC, USA (tidal amplitude
E70 cm) and Plum Island, MA, USA (tidal amplitude E145 cm) vs. Figure 5 Observed elevations of North Inlet marshes and the results
the local mean high-water level. Mean high-water level was averaged of a model fit. Parameter kr was an input, whereas parameters ks and
from nearby gages for each year that productivity was measured. q were obtained by least-squares nonlinear regression.
188 Ecogeomorphology of Salt Marshes

proportionality constant and the refractory fraction of organic 1200

Bioassay data
production. Hence, Polynomial fit

Sranding biomass (g m−2)

1000 GPS survey
dZ=dt ¼ ðq þ ks Bs ÞD þ kr Bs ½1
800
Parameter q is the settling velocity of suspended particles in
600
the absence of vegetation, and ks is functionally equivalent to a
trapping coefficient. Values of q and ks will vary locally and
400
regionally as a function of sediment availability, tidal range
(e.g., Stevenson et al., 1986), and the drag exerted by different
200 B = 16.7D − 0.21D2 + 521
forms of vegetation (Bouma et al., 2005). Implicit in the P < 0.0001
parameters q (yr–1), ks (cm2 g–1 yr–1), and kr (cm3 g–1 yr–1) are 0
mass-to-volume conversions such that the products of the
multiples give the units of dZ/dt (cm yr–1). −40 −20 0 20 40 60 80 100
The model fit to elevation data from North Inlet is quite
Depth below mean high water (cm)
good for sites ranging from a low marsh site that was initiated
in 2001 to high marsh control and fertilized sites that were Figure 6 Results of growing S. alterniflora at different elevations
initiated in 1996 (Figure 5). The model was fitted using an (marsh organ data) within the intertidal zone and a global positioning
SAS nonlinear parameter estimation procedure. Parameters q system (GPS) survey in North Inlet marshes of the upper range of S.
and ks were unknown, and parameter kr, the soil organic alterniflora.
matter equivalent of aboveground biomass, was input as a
known 0.6 cm3 g–1 yr–1. The model fits the data equally well
any given marsh species, sensu Shelford (1931). As shown here
using lower values of kr, but as kr is decreased, the value ks
for S. alterniflora at North Inlet, the optimum lies at roughly
increases. For example, for kr ¼ 0, ks increased to 0.1412 from
40 cm below the MHW level (Figure 6). The upper and lower
0.1029 cm3 g–1 yr–1 while q declined to 0.0011 from 0.0017 yr1.
limits of relative elevation are likely determined by hypoxia at
In other words, kr and ks are compensatory to a large degree,
one extreme and osmotic stress at the other. The vertical dis-
meaning that biomass is equally effective in a mathematical
tribution of its biomass density can be approximated by a
sense in generating sediment volume by accreting mineral
parabola with an optimum depth (depth below MHW) that is
sediment (ks) or organic matter (kr). In practice, kr can be esti-
bounded by upper and lower limits:
mated from sediment cores. It should be the equivalent to the
quotient of a year’s worth of sediment organic matter from a Bs ¼ aD þ bD2 þ c ½2
stable soil horizon below the root zone divided by the standing
biomass. This curve can be viewed as dimensions of a species’ funda-
The results of the model fit (Figure 5) demonstrate the mental (in the absence of competitors) or realized (in the
importance of standing biomass and relative elevation to the presence of competitors) niche, sensu Hutchinson (1957).
process of sediment accretion. Accretion rates in the high Finally, making the substitution for Bs in eqn [1], the rate
marsh, as measured by the model fit, have been exactly 3 times of change of elevation of the marsh surface, dZ/dt, can then be
greater in the fertilized plots as in the controls due to en- expressed as a function of a single variable, depth (D):
hanced growth in fertilized plots. Sediment accretion rate in a
low marsh site, as measured by the model fit, was more than 2 dZ=dt ¼ Dðakr þ cks þ qÞ þ D2 ðbkr þ aks Þ þ D3 bks þ ckr ½3
times greater than in the high marsh control site (Figure 5),
and this is largely due to the effect of depth (D) or inundation This equation can be solved for the equilibrium condition
time. dZ/dt ¼ r, where r is the rate of sea-level rise, to give in-
Although this model fits data from marshes at North Inlet formative and testable predictions. For instance, the equi-
spanning a range of elevations and biomass values, extrapo- librium elevation of the marsh is a function of the rate of
lating to other estuaries with these particular parameter values sea-level rise (Figure 7). The faster the rate of rise, the greater is
is problematic. Abstract models such as this are better at the equilibrium depth. Biomass and accretion rates are func-
making generalities and insights than they are at generating tions of depth and, hence, are also functions of the rate of
quantitatively accurate forecasts. Estuaries differ in tidal sea-level rise. Furthermore, there is a tipping point. When the
amplitude and suspended sediment concentration. These are rate of sea-level rise exceeds about 0.5 cm yr–1 in this in silico
merely two important variables that are implicitly embedded marsh, the equilibrium depth of the marsh surface will drop
in the model’s ‘constants’. Nevertheless, there are a number of below the range of the vegetation. In other words, the vege-
powerful generalities that fall out of a careful analysis, as tated marsh is stable only when the rate of sea-level rise is less
demonstrated below. than about 0.5 cm yr–1 (at North Inlet). Once the depth of the
We can further simplify the model by substituting depth marsh has increased to a level where a further increase in sea
(D) for standing biomass. The standing biomass density Bs level will depress primary production, the vegetated marsh
changes with a number of environmental conditions, in- will take on a trajectory that leads to irreversible collapse.
cluding inundation time, which is related to the elevation of a Recovery is not possible, if the rate of sea-level rise remains
site relative to MSL, or depth of the marsh surface below some above the tipping point. The model does not, however, predict
tidal datum, such as MHW level. There is an optimal depth for that the marsh will collapse episodically. The loss of a marsh
 Ecogeomorphology of Salt Marshes 189

60 1000 0.6
Physical
Biological

Accretion rate (cm yr−1)

Equilibrium depth (cm)
50 control 800 0.5
control

Biomass (g m−2)

“Optimum depth”
40 0.4
600
30 0.3

Tipping
400
20 0.2

point
10 200 0.1
Kr = 0.2
0 0 0.0

0.0 0.2 0.4 0.6 0.8 0 10 20 30 40 50 60 0 20 40 60 80

Rate of sea-level rise (cm yr−1) Equilibrium depth (cm) Deapth (cm)

Figure 7 Model predictions showing (left) the equilibrium depth (ED) of the marsh surface at North Inlet, SC, as a function of the rate of sea-
level rise; (middle) the biomass of S. alterniflora as a function of ED; and (right) and the rate of sediment accretion as a function of depth. ED is
measured as depth to the marsh surface from mean high water.

should take decades after sea-level rise exceeds the tipping decomposition of organic matter, an electron donor, during
point, and the current long-term rate is just 0.3 cm yr–1, but is nitrate reduction to N2 gas (denitrification). Increased deg-
accelerating (Milne et al., 2009). radation of the organic matter can lead to less accumulation of
To summarize, this model and supporting data suggest that organic material belowground and smaller particle sizes, and
homeostatic interactions among plants, sea level, and sedi- thus, changes in porosity and bulk density. These factors
ments maintain the marsh at elevations within the intertidal combine to alter the geomorphology and biogeochemistry of
zone that are super-optimal with respect to productivity. The marshes, potentially leading to increased bank instability and
marsh can maintain its elevation within a stable range, pro- erosion. In this section, we review the evidence for these
vided the rate of sea-level rise does not exceed a tipping point. interactions.
This represents a significant advance in our understanding of In plot-level experiments, Spartina responds rapidly to
the ecology and geomorphology of coastal landscapes. N enrichment with increased above-ground production and
relatively less production allocated to below-ground growth, a
more shallow rooting system, and taller above-ground stems
12.12.3 Long-Term Marsh Stability and and leaves (Figure 9) (Mendelssohn and Morris, 2000; Tyler
Biogeochemical Cycling et al., 2003; Wigand et al., 2003; Deegan et al., 2007; Darby
and Turner, 2008). Nitrogen-driven shifts in above-ground/
Plants are characteristically the foundation species in an eco- below-ground production and plant-tissue characteristics, in
system and are considered to have primacy in controlling turn, can affect larger-scale ecosystem function, such as creek-
ecosystem geomorphology and biogeochemical function bank stability, accumulation, decomposition, and integrity of
(Hooper et al., 2005; Caraco et al., 2006; McGlathery et al., the organic matter (peat) and sediment characteristics. A sig-
2007). The long-term sustainability of salt marshes depends nificant contribution from vegetation toward bank stability
upon plants, predominately Spartina sp., that maintain the can be attributed to soil reinforcement, including increasing
morphology and elevation by a combination of below-ground soil shear strength and cohesion by the root system (Gray,
production, which serves to bind sediment and accumulates as 1995; Osman and Barakbah, 2006). Belowground biomass is
organic peat (Turner et al., 2009), and the trapping of in- an important component for sediment stability and marsh
organic sediment by aboveground plant stems and leaves accretion (Turner et al., 2009), where roots and rhizomes
(Morris et al., 2002). Spartina production characteristics are typically bind sediments and anchor the creek bank; thus, with
affected by nutrient availability, particularly N (Mendelssohn fewer binding agents (i.e., roots), banks destabilize and fail.
and Morris, 2000). Most Atlantic coast salt marshes are ex- A longer-term feedback of plants on marsh geomorph-
periencing rates of N loading at or above the level (B25 g ology involves the spatial rearrangement of species caused by
m–2 yr–1) that has induced plant species production and chronic N enrichment, with S. alterniflora outcompeting S.
composition changes for European wetlands (Bricker et al., patens under nutrient-enriched conditions (Levine et al., 1998;
1999; USEPA, 2005; Verhoeven et al., 2006), suggesting that Emery et al., 2001; Wigand et al., 2003; Crain, 2007). Because
coastal eutrophication may be affecting salt-marsh structure S. alterniflora has intrinsically lower below-ground production
and function. compared to S. patens, N enrichment could start a positive
Recent research suggests that chronic nutrient enrichment feedback loop with S. alterniflora expansion decreasing below-
has the potential to invoke negative feedbacks that will ground production, thus, lowering marsh-surface elevations,
ultimately affect marsh geomorphic configuration and bio- further favoring S. alterniflora. The combination of variations
geochemical cycling (Figure 8) by altering plant production in plant allocation and species composition may ultimately
and community characteristics and increasing microbial de- result in more surface detritus and less peat formation.
composition of organic matter by supplying nitrate, a terminal Nutrient enrichment can also decrease the amount of
electron acceptor. Nitrate availability can result in the structural compounds in aboveground biomass, which can
190 Ecogeomorphology of Salt Marshes

Geomorphic change Biogeochemical change

Creek bank collapse More direct denitrification
increased erosion More export to tidal flats

More tidal drag on Less soil reinforcement by Altered geotechnical Higher sediment flux to N2
above-ground biomass roots, lowered soil shear properties (organic tidal flats
strength and cohesion. matter size, bulk
density, loss on
ignition) Denitrification
Faster decomposition,
smaller particle size

Plant overgrowth Shallower rooting depth,

less structural lower root density, Organic
compounds fewer rhizomes Less
matter

OM
Less
More above-ground Less below-ground NO3−

Less coupled
Live
Spartina O2
NH4+
More direct
NO3−

Figure 8 Negative feedbacks induced by nutrient loading alter plant allocation to above- and below-ground production, and biogeochemical
cycling and sediment characteristics, leading to long-term marsh instability.

The combination of taller plants with less structural material

can result in falling over into the creek where accelerating and
10 decelerating tidal currents ‘tug’ on the plants, contributing to
R+R:shoot biomass

bank instability. Additionally, decreased lignin increases the

rate of decomposition of organic matter particularly under
nutrient-enriched conditions (Gill and Jackson, 2000; Ogram
1
et al., 2006). Fewer and less dense roots, combined with in-
creased drag on lodged aboveground biomass by tidal currents
and a more decomposed root mat, increases marsh suscepti-
Control
0.1 bility to creek-bank failure and erosion during regular low-
N energy events, such as tides and the passage of weather fronts.
N+P Marsh creek banks are hotspots of biogeochemical cycling
P in marshes with ecosystem-level impacts that are dis-
0.01 proportionate to their areal extent (Drake et al., 2008). De-
0 800 1600
composition of organic matter involving ammonia formation,
Above-ground biomass (g m−2) nitrification, and denitrification are central processes in the
Figure 9 Decrease in belowground biomass relative to aboveground salt-marsh nitrogen (N) cycle. Microbial decomposition is
biomass in fertilized salt-marsh plots. Adapted from Darby, F.A., generally expected to increase in response to nutrient enrich-
Turner, R.E., 2008. Effects of eutrophication on salt marsh root and ment with the greatest increase in lignin-poor organic matter
rhizome biomass accumulation. Marine Ecology Progress Series 363, (Ogram et al., 2006). Both nitrification and denitrification are
63–70. influenced by a variety of environmental parameters, including
the availability of oxygen and the residence time of nutrient-
affect both Spartina production characteristics and biogeo- enriched water. Therefore, processes that alter sediment geo-
chemical cycling. In an ecosystem experiment on level- technical characteristics, such as root structure and density,
salt-marsh fertilization (Deegan et al., 2007), plants grew taller porosity, and bulk density influence rates of these processes.
and lignin content decreased 50% in S. alterniflora leaves with An additional effect of coastal N loading, which is delivered
nutrient enrichment (Figure 10). Compared to the reference primarily as NO3, is a stimulation of the decomposition of
system, plants in the fertilized system more commonly ex- organic matter, an electron donor, during nitrate reduction to
ceeded the ‘critical height’ (the height at which plant stem N2 gas (denitrification). Increased rates of microbial de-
height increases beyond the biomechanical ability of the stem composition and potential denitrification have been reported
diameter to hold them upright; Harley and Bertness, 1996), for experimentally and culturally NO3-enriched Spartina mar-
resulting in the plants lodging (falling over). Lodging is a shes and mudflats (Wigand et al., 2004; Ma and Aelion, 2004;
common physiological response of graminoids to excess N. Bowen et al., 2009a, 2009b; Drake et al., 2009; Koop-Jakobsen
 Ecogeomorphology of Salt Marshes 191

Plant height

80
Measured

60 Critical

40 Fertilized Plant % Lignin

40

Frequency (%) 20
30

% Lignin
0
0.3 1.2 2.2 3.2 4.2 20
80
10
60

Reference 0
40
(b) Reference Fertilized
20

0
0.3 1.2 2.2 3.2 4.2
(a) Plant height (m)
Figure 10 (a) Plants in fertilized marshes grow taller (solid lines) with a higher fraction of the population exceeding the critical height (dashed
lines) at which they buckle and fall over (lodge) compared to reference plots; (b) plants also show a concomitant decrease in structural
components (lignin content) of Spartina leaves with fertilization, which also contributes to lodging and increased decomposition (Deegan,
unpublished).

and Giblin, 2010). The degradation of the organic matter leads The effects of nutrient enrichment on salt-marsh geo-
to less accumulation of organic matter and smaller particle morphology and biogeochemistry have broader implications
sizes, and, thus, changes in porosity and bulk density. These than just the persistence of salt marshes. The latter are widely
factors combine to alter the geomorphology, geotechnical recognized as some of the most productive ecosystems on
characteristics, and biogeochemistry of marshes, potentially Earth with primary productivity that rivals industrialized
leading to increased bank instability and erosion. Obviously, agriculture (Mitch and Gosselink, 2000). Tidal marshes can
geomorphic change, such as creek-bank failure, which cata- produce up to 8000 metric tons of plant material per year,
strophically lowers a wedge of high marsh dominated by continually removing CO2 from the atmosphere. Under an-
S. patens into the creek, altering the tidal inundation frequency aerobic conditions typical of salt marshes, decomposition
and duration, and converting the vegetation to S. alterniflora, rates are low, thus, most of this C is stored in the soil. Salt
will also have a large effect on biogeochemical cycling at the marshes are well known as major carbon-storing ecosystems
marsh edge. sequestering carbon at a rate about 10-fold higher on an area
The interactive and negative feedbacks among nutrient basis than any other wetland ecosystem (Brigham et al., 2006;
loading, changes to characteristics of salt-marsh plant pro- Chmura et al., 2003). Eutrophication may disproportionably
duction and biogeochemical cycling that constitute a ‘marsh alter the ability of marshes to sequester C through indirect
eutrophication syndrome’ may provide insight into the pat- effects of decomposition and altered plant allocations on
tern and rates of loss of salt marshes in nutrient-enriched areas erosion and creek-bank failure. By disrupting organic-matter-
along the northeastern Atlantic seaboard of USA. In the highly storage processes, altering the geomorphology, and exposing
eutrophic western Long Island Sound (LIS), six coastal em- stored organic C to aerobic conditions, salt marshes may
bayments lost from 27% to 54% of their low marsh (i.e., creek become a source of C to the atmosphere.
edge) habitat over 30 years (Tiner et al., 2006). Jamaica Bay, Additionally, decreased peat formation and creek-bank sta-
NY, has experienced high nutrient loads for decades (O’Shea bility reduces the ability of marshes to keep pace with sea-level
and Brosnan, 2000) and despite efforts to curtail salt-marsh rise, leading to a reduction in the buffering of coastlines from
loss in this area by restricting dredging and filling activities, storms (Morris, 2005). On 29 August 2005, the world became
marsh loss in this area has occurred at a rate of 1.4% yr–1 aware of the importance of coastal wetlands in protecting
(B17 ha yr–1) from 1974 to 1999 (Hartig et al., 2002). Bank inland populations as Hurricane Katrina slammed into the
collapse and creek widening has been reported for highly Louisiana and Mississippi Gulf Coasts. In Hancock County,
enriched marshes in Jamaica Bay, NY, although the underlying Mississippi, where there are little-to-no coastal wetlands,
mechanisms were not determined (Hartig et al., 2002). These a high-energy, 10-m storm surge powered B1 km inland un-
continued marsh losses, and the pattern of creek bank collapse impeded. The diminished extent of wetlands southeast of
and widening are consistent with a marsh-loss syndrome New Orleans contributed to the flooding of that city by a
resulting from nutrient loading. 5–6 m storm surge (Day et al., 2007). Hurricane Katrina
192 Ecogeomorphology of Salt Marshes

resulted in approximately 400 deaths in Mississippi and 1100 in the marsh model of the vegetation effects on sediment dy-
Louisiana. Conversely, when Hurricane Rita struck the Louisi- namics, accumulation rates, and organogenic production by
ana/Texas border less than a month later, with a tidal surge linking all these processes to the biomass of the halophyte
similar to that generated on the eastern side of New Orleans by vegetation that colonizes the marsh surface (Morris et al.,
Katrina (5–6 m), only 10 people died. This is due in part to 2002; Mudd et al., 2004; D’Alpaos et al., 2006).
attenuation as the surge crossed nearly 50 km of the coastal In the seminal paper of Morris et al. (2002), the above-
wetlands before reaching populated areas (Day et al., 2007). ground vegetation biomass of S. alterniflora is linked to the
Because of a renewed understanding of the importance of marsh elevation and hydroperiod, based on field data collected
coastal ecosystems for human and marine life, estuaries have in North Inlet, South Carolina. The availability of quantitative
become the focus of large-scale restoration strategies (Perry relationships between biological parameters (i.e., vegetation
et al., 2001; Warren et al., 2002). To be successful, these ap- biomass) and morphological parameters (i.e., marsh elevation)
proaches require an understanding of the degree and synergies allow the development of numerical models for the coupled
of degradation in an ecosystem context (Deegan, 2002; Adger evolution of biology and landforms in coastal systems. The first
et al., 2005; Callaway, 2005). model for salt-marsh ecogeomorphology was developed by
Mudd et al. (2004). In this model, a series of feedbacks be-
tween vegetation and sediment fluxes were explicitly treated
with parametric equations based on field data and laboratory
12.12.4 Modeling Intertidal Ecogeomorphology experiments. For instance, Mudd et al. (2004) included a drag-
coefficient term function of vegetation biomass (Nepf, 1999),
Several numerical models have been utilized in recent years to
marsh accretion due to belowground organic production
study the evolution of salt marshes and tidal flats (see
(Randerson, 1979), and sediment trapping due to deposition
Fagherazzi and Overeem, 2007 for a review). In the pioneering
on vegetation stems (Palmer et al., 2004). All these expressions
work of Krone (1987), changes in marsh elevation are calcu-
are linked to vegetation biomass either directly or indirectly
lated as a function of sediment concentration, settling velocity
through stem density and diameter, and therefore fully capture
of the suspended sediment flocs, and hydroperiod. In Krone’s
the feedbacks between biology and geomorphology.
model, as the marsh platform becomes emergent, the inun-
A similar framework has been applied to the role of
dation period decreases, so that fewer aggregates have time to
vegetation in the formation and evolution of tidal channels
deposit. This mechanism leads to a reduction in marsh ac-
(D’Alpaos et al., 2005; Temmerman et al., 2007) and the
cretion. The model has been improved by considering sedi-
evolution of the entire marsh surface (Kirwan and Murray,
ment supply and sea-level rise (Allen, 1990; French, 1993),
2007; D’Alpaos et al., 2007; Temmerman et al., 2005, see
sediment composition (Allen, 1995), differences in sedimen-
Figure 11). All these models point to the following effects of
tation rates between creek levees and marsh platform
vegetation on marsh landforms:
(Temmerman et al., 2004a), the effect of tidal channels
(Fagherazzi and Sun, 2004), and variations in sediment con- 1. vegetated surfaces increase friction thus reducing flow
centration as a function of tidal inundation and waves on the marsh platform; water fluxes are then concentrated
(Temmerman et al., 2004b; Fagherazzi and Priestas, 2010). in unvegetated areas leading to channel formation and
In recent years, a major development has been the inclusion in channel deepening;

Interaction vegetation−sediment transport

Temmerman et al., 2004
Mudd et al., 2004
Tidal channels and vegetation Interaction vegetation−marsh elevation
D’Alpaos et al., 2006 Morris et al., 2002
Kirwan and Murray 2007 Morris 2006
Temmerman et al., 2007
Seagrass and sediment transport
Carr et al., 2010
Halophyte vegetation
Mean high tide

Salt marsh Mean sea-level

Tidal channel
Scarp Seagrass
Microphytobentos
Below-ground organic production
Mudd et al., 2009
Tidal flats

Marsh−tidal flat boundary Tidal flats evolution

van Koppel et al., 2005 Cappucci et al., 2004
Mariotti and Fagherazzi 2010 Fagherazzi et al., 2007
Figure 11 Numerical models of intertidal ecogeomorphology.
 Ecogeomorphology of Salt Marshes 193

2. vegetation favors marsh accretion by trapping sediments

Marsh elevation (m)

1.0
and storing organic matter below ground; and 0.0
3. enhanced accretion leads to higher marsh elevations within −1.0 MHWL
the tidal range and, therefore, a smaller tidal prism,
−2.0 MSL
reducing discharges and channel cross sections. MLWL
−3.0 Vegetated scenario t = 55 years
−4.0 Unvegetated scenario
The result is therefore twofold: with vegetation, channels are
deeper for a given marsh elevation, but they get infilled faster −20 −10 0 10 20
(a) Channel width (m)
due to higher accretion (Figure 12).

Marsh elevation (m)

More realistic ecogeomorphic models need to quantify the 1.0
feedbacks between different plant species that colonize the 0.0
marsh surface (Silvestri et al., 2005; Morris, 2006). An im- −1.0
portant component of salt-marsh models is the production of
−2.0
belowground organic material, composed of roots, rhizomes,
−3.0
and detritus, which can either increase salt-marsh accretion or t = 70 years
−4.0
enhance subsidence, if the organic material decays in time. In
the model of Mudd et al. (2009), both refractory and labile −20 −10 0 10 20
(b) Channel width (m)
carbon pools are considered, and the balance between organic
matter production and decay is tracked in time along a vertical

Marsh elevation (m)

1.0
column of marsh sediments.
0.0
In recent years, a series of models focused on the mor-
−1.0
phological evolution of tidal flats and their interconnection
−2.0
to benthic ecosystems colonizing them (Fagherazzi et al.,
2007; Cappucci et al., 2004; Carr et al., 2010). Even in tidal −3.0
t = 90 years
flats, sediment-transport processes are mediated by biota. −4.0
Seagrass strengthens the bottom substrate, reduces tidal cur- −20 −10 0 10 20
(c) Channel width (m)
rents, thus favoring sediment deposition, and actively traps
sediment particles on leaves and stems (de Boer, 2007).
Marsh elevation (m)

1.0
Microalgal biofilms modify cohesive sediment properties, in- 0.0
creasing the critical shear stress for erosion, thus reducing the −1.0
potential erosion of tidal flats (Tolhurst et al., 2008). Models Channel comparison
−2.0
that aim to capture the feedbacks between biology and sedi- at same salt marsh
−3.0 elevation
ment transport in tidal flats need to include two processes that
−4.0
are negligible in salt marshes: (1) erosion, and, in particular,
−20 −10 0 10 20
the effect of wind waves, which resuspend sediments main-
(d) Channel width (m)
taining the tidal flat below MSL, and (2) water turbidity, which
depends on sediment resuspension triggered by waves and Figure 12 Channel cross sections on a salt marsh at different
currents because seagrasses and biofilms are often light lim- stages of channel development starting from a tidal flat below mean
sea level in the absence (solid brown lines) or in the presence of
ited. All these processes are correctly captured in the seagrass
vegetation (dashed green lines): (a) after 55 years; (b) after 70 years;
model of Carr et al. (2010), which shows that tidal flat systems
(c) after 90 years. Mean high-water level (MHWL), mean sea level
are bistable, evolving toward a configuration with seagrasses (MSL), and mean low-water level (MLWL) are also indicated. In this
and clear water or a configuration with bare bottom and tur- scenario, enhanced deposition favored by vegetation increases the
bid water. marsh elevation reducing tidal prism and therefore channel cross
Ecogeomorphic models of tidal flats are inherently more section; and (d) a comparison between channels developed with and
complex than salt-marsh models, and need to account for the without vegetation for identical marsh elevations (i.e., identical tidal
variability of wave climate as a function of metereological prism) shows that, with vegetation, the channel is deeper due to the
forcing as well as for the spatial variation in sediment con- concentration of flow from the vegetated platform to the unvegetated
centration in the water column, which again depends on wind channel.
waves and tidal circulation.
Of fundamental importance for the coupling between salt
marshes, tidal flats, and related ecosystems are the dynamics of The model of van de Koppel et al. (2005) shows that the
the boundary between the two landforms. Biological and edge between salt marshes and tidal flats approaches a critical
morphological gradients commonly produce a distinct scarp state becoming increasingly steep and vulnerable to wave
between these two environments (van de Koppel et al., 2005). attack. A storm can then cause a vegetation collapse and
Marsh deterioration is generally linked to erosion and collapse extensive erosion of the marsh edge, reducing marsh area.
of the marsh edges under the influence of wind waves (Möller, The model of Mariotti and Fagherazzi (2010) focuses on
et al., 1999; Möller, 2006; van de Koppel et al., 2005). The a typical intertidal transect composed of a salt marsh, a tidal
elevation of tidal flats bordering the marsh platform strongly flat, and the marsh edge separating the two environments.
influences the height of the incoming waves and the energy The model computes wind waves, tidal currents, sediment
dissipated at the marsh edge, thus determining erosion rates. erosion, and deposition, as well as the effect of vegetation on
194 Ecogeomorphology of Salt Marshes

sediment dynamics. Numerical simulations show that the and morphological parameters in a more general way. Exten-
boundary between salt marshes and tidal flats is seldom at sive fieldwork and laboratory experiments are deemed neces-
equilibrium, either prograding or retrograding as a function of sary to develop such relationships. The following criteria are
sediment supply and sea-level oscillations. Vegetation plays an important for the correct development of expressions to be
important role in the dynamics of the system, regulating the used in numerical models of intertidal ecogeomorphology:
rate of marsh progradation and regression (Figure 13). Finally,
Marani et al. (2007) explore with a point model the equi- 1. The feedback between biology and geomorphology must
librium states of the coupled tidal flat–salt-marsh system, be quantitative, that is, described by suitable equations
clearly showing the importance of vegetation in stabilizing the with process-based parameters that are directly measured
marsh platform. in the field or in the laboratory.
Most of the quantitative expressions at the base of the 2. Parameters derived from direct correlation between bio-
numerical models described herein have been derived from logical and morphological quantities should be avoided
limited field data, for specific vegetation species, or in selected because they are often valid only for the studied system and
locations with unique climatological and sedimentological not applicable elsewhere; if possible, the quantitative ex-
characteristics. An important research need for future eco- pressions should be made nondimensional and process
geomorphological models of intertidal landscapes is the based, in order to increase portability among different
availability of quantitative expressions that couple biological systems.

250

1.5 200

Marsh progrdation rate (cm yr−1)

Salt marsh Vegetated
1.0 Vegetated
Initial configuration 150
0.5 50 years
100 years
0 200 years
Elevation (m)

100
300 years
−0.5 500 years
50
−1.0 C=1 g l−1
C=0.9 g l−1
−1.5 0 C=0.8 g l−1
Tidal flat C=0.7 g l−1
−2.0 C=0.6 g l−1
−50 C=0.5 g l−1
C=0.4 g l−1
−2.5 C=0.3 g l−1
−1
C=0.2 g l−1
C=0.1 g l
−3.0 −100
0 50 100 150 200 250 300 350 400 450 500 0 5 10 15 20
(a) Transect distance (m) (c) RSLR (mm yr−1)

150
Marsh progradation rate (cm yr−1)

1 Salt marsh 100

Unvegetated Unvegetated
0.5 Initial configuration 50
10 years
0 25 years C=2.0 g l−1
Elevation (m)

50 years 0
−0.5 100 years C=1.5 g l−1
150 years
−1.0 200 years −50
C=1 g l−1
300 years C=0.9 g l−1
−1.5 C=0.8 g l−1
−100 C=0.7 g l−1
Tidal flat C=0.6 g l−1
−2.0
C=0.5 g l−1
−150 C=0.4 g l−1
−2.5 C=0.3 g l−1
C=0.2 g l−1
C=0.1 g l−1
−3.0 −200
0 50 100 150 200 250 300 350 400 450 500 0 5 10 15 20
(b) Transect distance (m) (d) RSLR (mm yr−1)

Figure 13 Simulation of coupled salt-marsh- and tidal-flat evolution along a transect perpendicular to the marsh boundary. Erosion of the
marsh scarp by wave attack imposing a sediment concentration equal to 0.1 g l–1 at the seaward boundary: (a) with vegetation; (b) without
vegetation. With vegetation, the marsh boundary retreats, conserving the scarp shape. Progradation and erosion rates of the marsh boundary as
a function of relative sea level rise (RSLR) and sediment concentration. Positive values indicate progradation, negative values indicate erosion;
(c) with vegetation; and (d) without vegetation. Adapted from Mariotti, G., Fagherazzi, S., 2010. A numerical model for the coupled
long-term evolution of salt marshes and tidal flats, Journal of Geophysical Research-Earth Surface, F01004, doi:10.1029/2009JF001326, with
permission from AGU.
 Ecogeomorphology of Salt Marshes 195

3. Biological processes must be linked to morphodynamic Allen, J.R.L., 1990. Salt-marsh growth and stratification – a numerical model with
processes; two different connections are possible: special reference to the Severn Estuary – Southwest Britain. Marine Geology
95(2), 77–96.
(a) Landscape Feedback: In this connection, the distri-
Allen, J.R.L., 1995. Salt-marsh growth and fluctuating sea level: implications of a
bution of biota is related to the characteristics of the simulation model for Flandrian coastal stratigraphy and peat-based sea-level
landscape and its variability in space. For example, curves. Sedimentary Geology 100(1–4), 21–45.
vegetation biomass varies as a function of salt-marsh Bertness, M.D., 1984. Ribbed mussels and Spartina alterniflora production in a
elevation and distance from the creeks. New England salt marsh. Ecology 65, 1794–1807.
de Boer, W.F., 2007. Seagrass–sediment interactions, positive feedbacks and critical
(b) Sediment Transport Feedback: This connection de- thresholds for occurrence: a review. Hydrobiologia 591, 5–24.
scribes how biological processes alter sediment trans- Bouma, T.J., DeVries, M.B., Low, E., et al., 2005. Flow hydrodynamics on a mudflat
port and, vice versa, how sediment fluxes affect and in salt marsh vegetation: identifying general relationships for habitat
vegetation and animals. For example, vegetation can characterisations. Hydrobiologia 540, 259–274.
Bowen, J.L., Crump, B.C., Deegan, L.A., Hobbie, J.E., 2009a. Salt marsh sediment
increase sediment trapping and sediment settling.
bacteria: their distribution and response to external nutrient inputs. ISME
These feedbacks are important because, by mediating Journal 3, 924–934.
sediment transport, biota directly influence erosion Bowen, J.L., Crump, B.C., Deegan, L.A., Hobbie, J.E., 2009b. Increased supply of
and deposition patterns, and therefore the evolution of ambient nitrogen has minimal effect on salt marsh bacterial production.
the intertidal landscape. Limnology and Oceanography 54(3), 713–722.
Bricker, S., Clements, C., Pirhalla, D., Orlando, S., Farrow, D., 1999. National
4. The feedbacks between biology and geomorphology
Estuarine Eutrophication Assessment. NOAA, Silver Springs, MD.
should be expressed in terms of only one of two physical Brigham, S.D., Megonigal, J.P., Keller, J.K., Bliss, N.P., Trettin, C., 2006. The
variables. For example, most of the biological processes carbon balance of North American wetlands. Wetlands 26, 889–916.
can be related to biomass, whereas all geomorphic pro- Cahoon, D.R., Lynch, J.C., Hensel, P., Boumans, R., Perez, B.C., Segura, B.,
cesses depend on critical shear stress for erosion and Day, J.W., 2002. High-precision measurements of wetland sediment elevation: I.
Journal of Sedimentary Research 72, 730–733.
sediment concentration in the water column. Fewer vari- Cahoon, D.R., Turner, R.E., 1989. Accretion and canal impacts in a rapidly subsiding
ables increase the significance of the results and allow wetland 2. Feldspar marker horizon technique. Estuaries 12(4), 260–268.
direct testing with field data. Moreover, the utilization of a Callaway, J.C., 2005. The challenge of restoring functioning salt marsh ecosystems.
prescribed set of variables favors a comparison among Journal of Coastal Research 40, 24–36.
Cappucci, S., Amos, C.L., Hosoe, T., Umgiesser, G., 2004. SLIM: a numerical
different models.
model to evaluate the factors controlling the evolution of intertidal mudflats in
5. A critical component of ecogeomorphic modeling is the Venice Lagoon, Italy. Journal of Marine Systems 51(1–4), 257–280.
dynamics of organic matter, both in terms of fluxes and Caraco, N., Cole, J., Findlay, S., Wigand, C., 2006. Vascular plants as engineers of
spatial distribution within the landscape. In fact, organic oxygen in aquatic systems. BioScience 56, 219–225.
matter in the intertidal area has a twofold importance: Carr, J., D’Odorico, P., McGlathery, K., Wiberg, P.L., 2010. Stability and bistability
of seagrass ecosystems in shallow coastal lagoons: role of feedbacks with
(a) it is the central component of any biological process sediment resuspension and light attenuation. Journal of Geophysical Research-
and ecological system; Biogeosciences 115, G03011.
(b) it is an important geomorphic material, which can be Chmura, G.L., Anisfeld, S.C., Cahoon, D.R., Lynch, J.C., 2003. Global carbon
stored and transported within the landscape, contrib- sequestration in tidal, saline wetland soils. Global Biogeochemical Cycles 17(4),
1111.
uting to the formation of landforms. Other important
Christiansen, T., Wiberg, P.L., Milligan, T.G., 2000. Flow and sediment transport on
constituents do not have this double function. Nutrients a tidal salt marsh surface. Estuarine, Coastal and Shelf Science 50(3), 315–331.
influence biogeochemical cycles but are not building Coops, H., Geilen, N., Verheij, H.J., Boeters, R., van der Velde, G., 1996.
blocks of the landscape, whereas mineral sediments Interactions between waves, bank erosion and emergent vegetation: an
only indirectly affect biology, being commonly origin- experimental study in a wave tank. Aquatic Botany 53, 187–198.
Crain, C., 2007. Shifting nutrient limitation and eutrophication effects in marsh
ated by physical processes occurring elsewhere. vegetation across estuarine salinity gradients. Estuaries 30, 26–34.
D’Alpaos, A., Lanzoni, S., Marani, M., Fagherazzi, S., Rinaldo, A., 2005. Tidal
network ontogeny: channel initiation and early development. Journal of
Geophysical Research-Earth Surface 110(F2), 1–18.
Acknowledgments D’Alpaos, A., Lanzoni, S., Mudd, S.M., Fagherazzi, S., 2006. Modeling the influence
of hydroperiod and vegetation on the cross-sectional formation of tidal channels.
This research was supported by the Department of Energy Estuarine Coastal and Shelf Science 69(3–4), 311–324.
D’Alpaos, A., Lanzoni, S., Marani, M., Rinaldo, A., 2007. Landscape evolution in
NICCR program award # TUL-538-06/07, by the National tidal embayments: modeling the interplay of erosion, sedimentation, and
Science Foundation, LENS project award # OCE 0924287 and vegetation dynamics. Journal of Geophysical Research-Earth Surface 112(F1),
OCE 0923689, TIDE project award # DEB 0816963, the VCR- F01008.
LTER program award # DEB 0621014, the PIE-LTER program D’Alpaos, A., Lanzoni, S., Marani, M., Rinaldo, A., 2010. On the tidal
prism–channel area relations. Journal of Geophysical Research 115, F01003.
award# OCE 9726921, and by the Office of Naval Research
http://dx.doi.org/10.1029/2008JF001243.
award # N00014-10-1-0269. Darby, F.A., Turner, R.E., 2008. Effects of eutrophication on salt marsh root and
rhizome biomass accumulation. Marine Ecology Progress Series 363, 63–70.
Day, Jr. J.W., Boesch, D., Clairain, E., et al., 2007. Restoration of the Mississippi
References Delta: lessons from hurricanes Katrina and Rita. Science 315, 1679–1684.
Day, Jr. J.W., Scarton, F., Rismondo, A., Are, D., 1998. Rapid deterioration of a salt
marsh in Venice Lagoon, Italy. Journal of Coastal Research 14, 583–590.
Adger, W.N., Hughes, T.P., Folke, C., Carpenter, S.R., Rockstrom, J., 2005. Deegan, L.A., 2002. Lessons learned: the effects of nutrient enrichment on the
Social–ecological resilience to coastal disasters. Science 309, 1036. support of nekton by seagrass and salt marsh ecosystems. Special SCOR
Allen, E.A., Curran, H.A., 1974. Biogenic sedimentary structures produced by crabs Volume. Estuaries 25(4b), 585–600.
in lagoon margin and salt marsh environments near Beaufort, North Carolina. Deegan, L.A., Bowen, J.L., Drake, D., et al., 2007. Susceptibility of salt marshes to
Journal of Sedimentary Petrology 44, 538–548. nutrient enrichment and predator removal. Ecological Applications 17(5), S42–S63.
196 Ecogeomorphology of Salt Marshes

Drake, D.C., Deegan, L.A., Harris, L.A., Miller, E.E., Peterson, B.J., Warren, R.S., Kirwan, M.L., Murray, A.B., Boyd, W.S., 2008. Temporary vegetation disturbance as
2008. Plant N dynamics in fertilized and natural New England salt marshes: a an explanation for permanent loss of tidal wetlands. Geophysical Research
paired 15N tracer study. Marine Ecology Progress Series 354, 35–46. Letters 35(5), L05403.
Drake, D.C., Peterson, B.J., Galván, K.A., et al., 2009. Salt marsh ecosystem Knuston, P.L., Brochu, R.A., Seelig, W.N., Inskeep, M., 1982. Wave damping in
biogeochemical responses to nutrient enrichment: a paired 15N tracer study. Spartina alterniflora marshes. Wetlands 2, 87–104.
Ecology 90(9), 2535–2546. Koop-Jakobsen, K., Giblin, A.E., 2010. The effect of increased nitrate loading on
van Eerdt, M.M., 1985. The influence of vegetation on erosion and accretion nitrate reduction via denitrification and DNRA in salt marsh sediments.
in salt marshes of the Oosterschelde, The Netherlands. Vegetatio 62, Limnology and Oceanography 55(2), 789–802.
367–373. van de Koppel, J., van der Wal, D., Bakker, J.P., Herman, P.M.J., 2005. Self-
Emery, N.C., Ewanchuk, P.J., Bertness, M., 2001. Competition and salt marsh plant organization and vegetation collapse in salt marsh ecosystems. American
zonation: stress tolerators may be dominant competitors. Ecology 82, Naturalist 165(1), E1–E12.
2471–2485. Krone, R.B., 1987. A method for simulating historic marsh elevations. In: Krause,
Escapa, M., Perillo, G.M.E., Iribarne, O., 2008. Sediment dynamics modulated by N.C. (Ed.), Coastal Sediments ’87. American Society of Civil Engineers, New
burrowing crab activities in contrasting SW Atlantic intertidal habitats. Estuarine, York, NY, pp. 316–323.
Coastal and Shelf Science 80, 365–373. Leonard, L.A., Luther, M.E., 1995. Flow hydrodynamics in tidal marsh canopies.
Fagherazzi, S., Bortoluzzi, A., Dietrich, W.E., Adami, A., Lanzoni, S., Marani, M., Limnology and Oceanography 40, 1474–1484.
Rinaldo, A., 1999. Tidal networks 1. Automatic network extraction and Levine, J., Brewer, S., Bertness, M., 1998. Nutrients, competition and plant zonation
preliminary scaling features from Digital Elevation Maps. Water Resources in a New England salt marsh. Journal of Ecology 86, 285–292.
Research 35(12), 3891–3904. Ma, H., Aelion, C.M., 2004. Ammonium production during microbial nitrate removal
Fagherazzi, S., Overeem, I., 2007. Models of deltaic and inner continental shelf in soil microcosms from a developing marsh estuary. Soil Biology and
landform evolution. Annual Review of Earth and Planetary Sciences 35, 685–715. Biogeochemistry 37, 1869–1878.
Fagherazzi, S., Palermo, C., Rulli, M.C., Carniello, L., Defina, A., 2007. Wind waves Marani, M., D’Alpaos, A., Lanzoni, S., Carniello, L., Rinaldo, A., 2007. Biologically-
in shallow microtidal basins and the dynamic equilibrium of tidal flats. Journal controlled multiple equilibria of tidal landforms and the fate of the Venice
of Geophysical Research-Earth Surface 112(F2), 12. lagoon. Geophysical Research Letters 34, L11402.
Fagherazzi, S., Priestas, A.M., 2010. Sediments and water fluxes in a muddy Mariotti, G., Fagherazzi, S., 2010. A numerical model for the coupled long-term
coastline: interplay between waves and tidal channel hydrodynamics. Earth evolution of salt marshes and tidal flats. Journal of Geophysical Research-Earth
Surface Processes and Landforms 35(3), 284–293. Surface F01004, http://dx.doi.org/10.1029/2009JF001326.
Fagherazzi, S., Sun, T., 2004. A stochastic model for the formation of channel McGlathery, K.J., Sundbeck, K., Anderson, I.C., 2007. Eutrophication in shallow
networks in tidal marshes. Geophysical Research Letters 31(21), L21503. http:// coastal bays and lagoons: the role of the coastal plant filter. Marine Ecology
dx.doi.org/10.1029/2004GL020965. Progress Series 348, 1–18.
Feagin, R.A., Lozada-Bernard, S.M., Ravens, T.M., Möller, I., Yeager, K.M., Mendelssohn, I.A., Morris, J.T., 2000. Ecophysiological controls on the growth of
Baird, A.H., 2009. Does vegetation prevent wave erosion of salt marsh edges? Spartina alterniflora. In: Weinstein, N.P., Kreeger, D.A. (Eds.), Concepts and
Proceedings of the National Academy of Sciences 106, 10109–10133. Controversies in Tidal Marsh Ecology. Kluwer, Dordrecht, pp. 59–80.
Franz, D.R., 1993. Allometry of shell and body weight in relation to shore level in Meyer, D.L., Townsend, E.C., Thayer, G.W., 1997. Stabilization and erosion control
the intertidal bivalve Geukensia demissa (Bivalvia: Mytilidae). Journal of value of oyster cultch for intertidal marsh. Restoration Ecology 5, 93–99.
Experimental Marine Biology and Ecology 174, 193–207. Milne, G.A., Gehrels, W.R., Hughes, C.W., Tamisiea, M.E., 2009. Identifying the
Franz, D.R., 2001. Recruitment, survivorship, and age structure of a New York causes of sea-level change. Nature Geoscience 2, 471–478.
ribbed mussel population (Geukensia demissa) in relation to shore level – a Mitch, W.J., Gosselink, J.G., 2000. Wetlands, Third ed. Wiley, New York, NY.
nine year study. Estuaries 24, 319–327. Möller, I., 2006. Quantifying saltmarsh vegetation and its effect on wave height
French, J.R., 1993. Numerical simulation of vertical marsh growth and adjustment dissipation: results from a UK East coast saltmarsh. Estuarine Coastal and Shelf
to accelerated sea-level rise, North Norfolk. UK. Earth Surface Processes and Science 69(3–4), 337–351.
Landforms 18(1), 63–81. Möller, I., Spencer, T., French, J.R., Leggett, D.J., Dixon, M., 1999. Wave
Friedrichs, C.T., Perry, J.E., 2001. Tidal salt marsh morphodynamics: a synthesis. transformation over salt marshes: a field and numerical modelling study from
Journal of Coastal Research, Special Issue 27, 3–37. north Norfolk, England. Estuarine Coastal and Shelf Science 49(3), 411–426.
Gill, R.A., Jackson, R.B., 2000. Global patterns of root turnover for terrestrial Montague, C.L., 1980. A natural history of temperate western Atlantic fiddler crabs
ecosystems. New Phytololgist 147, 13–31. (Genus Uca) with reference to their impact on the salt marsh. Contributions in
Gleason, M.L., Elmer, D.A., Pien, N.C., Fisher, J.S., 1979. Effects of stem density Marine Science 23, 25–55.
upon sediment retention by salt marsh cord grass, Spartina alterniflora Loisel. Morris, J.T., 2005. Effects of changes in sea level and productivity on the stability
Estuaries 2, 271–273. of intertidal marshes. UNESCO Proceeding Series on Lagoons and Coastal
Gray, D.H., 1995. Influence of vegetation on the stability of slopes. In: Barker, D.H. Wetlands in the Global Change Context: Impact and Management Issues. United
(Ed.), Vegetation and Slopes Stabilisation, Protection and Ecology. Thomas Nations, NY.
Telford House, London, pp. 2–23. Morris, J.T., 2006. Competition among marsh macrophytes by means of
Harley, C.D.G., Bertness, M.D., 1996. Structural interdependence: an ecological geomorphological displacement in the intertidal zone. Estuarine Coastal and
consequence of morphological responses to crowding in marsh plants. Shelf Science 69(3–4), 395–402.
Functional Ecology 10, 654–661. Morris, J.T., Sundareshwar, P.V., Nietch, C.T., Kjerfve, B., Cahoon, D.R., 2002.
Hartig, E.K., Gornitz, V., Kolker, A., Muskacke, F., Fallon, D., 2002. Anthropogenic Responses of coastal wetlands to rising sea level. Ecology 83(10), 2869–2877.
and climate-change impacts on salt marshes of Jamaica Bay, New York City. Mudd, S.M., Fagherazzi, S., Morris, J.T., Furbish, D.J., 2004. Flow, sedimentation,
Wetlands 22, 71–89. and biomass production on a vegetated salt marsh in South Carolina: toward a
Holdredge, C., Bertness, M.D., Altieri, A.H., 2009. Role of crab herbivory in die-off predictive model of marsh morphologic and ecologic evolution. In: Fagherazzi,
of New England salt marshes. Conservation Biology 23(3), 672–679. S., Marani, M., Blum, L.K. (Eds.), The Ecogeomorphology of Salt Marshes.
Hooper, D.U., Chapin, III F.S., Ewel, J.J., et al., 2005. Effects of biodiversity on Estuarine and Coastal Studies Series. American Geophysical Union, Washington,
ecosystem functioning: a consensus of current knowledge. Ecological DC, pp. 165–188.
Monographs 75, 3–35. Mudd, S.M., Howell, S.M., Morris, J.T., 2009. Impact of dynamic feedbacks
Hughes, R.G., Paramor, O.A.L., 2004. On the loss of saltmarshes in south-east between sedimentation, sea-level rise, and biomass production on near-surface
England and methods for their restoration. Journal of Applied Ecology 41, marsh stratigraphy and carbon accumulation. Estuarine Coastal and Shelf
440–448. Science 82(3), 377–389.
Hutchinson, G.E., 1957. Concluding remarks. Cold Spring Harbor Symposia on Mwamba, M.J., Torres, R., 2002. Rainfall effects on marsh sediment redistribution,
Quantitative Biology 22, 415–442. North Inlet, South Carolina, USA. Marine Geology 189, 267–287.
Kastler, J.A., Wiberg, P.L., 1996. Sedimentation and boundary changes of Virginia Nepf, H.M., 1999. Drag, turbulence, and diffusion in flow through emergent
salt marshes. Estuarine, Coastal and Shelf Science 42, 683–700. vegetation. Water Resources Research 35(2), 479–489.
Kirwan, M.L., Murray, A.B., 2007. A coupled geomorphic and ecological model of Ogram, A., Bridgham, S., Corstanje, R., et al., 2006. Linkages between microbial
tidal marsh evolution. Proceedings of the National Academy of Sciences of the community composition and biogeochemical processes across scales.
United States of America 104(15), 6118–6122. In: Bobbink, R., Beltman, B., Verhoeven, J.T.A., Whigham, D.F. (Eds.), Wetlands
 Ecogeomorphology of Salt Marshes 197

as a Natural Resource, Vol. 2, Ecological Studies. Springer, New York, NY, Vol. Tiner, R.W., Huber, I.J., Nuerminger, T., Marshall, E., 2006. Salt Marsh Trends in
190, pp. 239–268. Selected Estuaries of Southwestern Connecticut. U.S. Fish and Wildlife Service
O’Shea, M.L., Brosnan, T.M., 2000. Trends in indicators of eutrophication in Western National Wetlands Inventory Program, Northeast Region, Hadley, MA. Prepared
Long Island Sound and the Hudson-Raritan Estuary. Estuaries 23(6), 877–901. for the Long Island Studies Program, Connecticut Department of Environmental
Osman, N., Barakbah, S.S., 2006. Parameters to predict slope stability – soil water Protection, Hartford, CT. NWI Cooperative Report, 20 pp.
and root profiles. Ecological Engineering 28(1), 90–95. Tolhurst, T.J., Consalvey, M., Paterson, D.M., 2008. Changes in cohesive sediment
Palmer, M.R., Nepf, H.M., Pettersson, T.J.R., 2004. Observations of particle capture properties associated with the growth of a diatom biofilm. Hydrobiologia 596,
on a cylindrical collector: implications for particle accumulation and removal in 225–239.
aquatic systems. Limnology and Oceanography 49(1), 76–85. Turaski, S.J., 2002. Spatial and temporal controls on saturated overland flow in a
Perry, J.E., Barnard, T.A., Bradshaw, J.G., et al., 2001. Creating tidal salt marshes in regularly flooded salt marsh. M.S. Thesis. University of Virginia, 189 pp.
the Chesapeake Bay. In: Goodwin, P., Mehta, A.J. (Eds.), Tidal Wetlands: Turner, R.G., Howes, B., Teal, J., Milan, C., Swenson, E., Goehringer-Toner, D.D.,
Physical and Ecological Processes. Journal of Coastal Research, Special Issue, 2009. Salt marshes and eutrophication: an unsustainable outcome. Limnology
vol. 27, pp. 170–191. and Oceanography 54(5), 1634–1642.
Piazza, B.P., Banks, P.D., La Peyre, M.K., 2005. The potential for created oyster Tyler, A.C., Mastronicola, T.A., McGlathery, K.J., 2003. Nitrogen fixation and
shell reefs as a sustainable shoreline protection strategy in Louisiana. nitrogen limitation of primary production along a natural marsh chronosequence.
Restoration Ecology 13, 499–506. Oecologia 136, 431–438.
Randerson, P.F., 1979. A simulation model of salt-marsh development and plant USEPA (US Environmental Protection Agency), 2005. National Coastal Condition
ecology. In: Knights, B., Phillips, A.J. (Eds.), Estuarine and Coastal Land
Report. US Environmental Protection Agency, Washington, DC. http://
Reclamation and Water Storage. Saxon House, Farnborough, pp. 48–67.
www.epa.gov/owow/oceans/nccr/2005/ (accessed April 2011).
Schwimmer, R.A., 2001. Rates and processes of marsh shoreline erosion in
Verhoeven, J.T., Arheimer, A.B., Yin, C., Hefting, M., 2006. Regional and global
Rehoboth Bay, Delaware, U.S.A. Journal of Coastal Research 17, 672–683.
concerns over wetlands and water quality. Trends in Ecology and Evolution 21,
Shelford, V.E., 1931. Some concepts of bioecology. Ecology 12, 455–467.
96–103.
Silvestri, S., Defina, A., Marani, M., 2005. Tidal regime, salinity and salt marsh
Warren, R.S., Fell, P.E., Rozsa, R., et al., 2002. Salt marsh restoration in Connecticut:
plant zonation. Estuarine Coastal and Shelf Science 62(1–2), 119–130.
Sommerfield, C.K., 2006. On sediment accumulation rates and stratigraphic 20 years of science and management. Restoration Ecology 10, 497–513.
completeness: lessons from Holocene ocean margins. Continental Shelf Wigand, C., McKinney, R., Charpentier, M., Chintala, M., Thursby, G., 2003.
Research 26, 2225–2240. Relationship with nitrogen loadings, residential development, and physical
Stevenson, J.C., Ward, L.G., Kearney, M.S., 1986. Vertical accretion in marshes characteristics with plant structure in New England saltmarshes. Estuaries 26,
with varying rates of sea-level rise. In: Wolfe, E. (Ed.), Estuarine Variability. 1494–1504.
Academic Press, New York, NY, pp. 241–259. Wigand, C., McKinney, R., Chintala, M., Charpentier, M., Groffman, P., 2004.
Temmerman, S., et al., 2005. Impact of vegetation on flow routing and Denitrification enzyme activity of fringe saltmarshes in New England (USA).
sedimentation patterns: three-dimensional modeling for a tidal marsh. Journal of Journal of Environmental Quality 33, 1144–1151.
Geophysical Research-Earth Surface 110(F4), 18. Wilson, C.A., Allison, M.A., 2008. An equilibrium profile model for retreating marsh
Temmerman, S., et al., 2007. Vegetation causes channel erosion in a tidal shorelines in southeast Louisiana. Estuarine, Coastal and Shelf Science 80,
landscape. Geology 35(7), 631–634. 483–494.
Temmerman, S., Govers, G., Meire, P., Wartel, S., 2004a. Simulating the long-term Wray, R.D., Leatherman, S.P., Nicholls, R.J., 1995. Historic and future land loss for
development of levee-basin topography on tidal marshes. Geomorphology upland and marsh islands in the Chesapeake Bay, Maryland, U.S.A. Journal of
63(1–2), 39–55. Coastal Research 11, 1195–1203.
Temmerman, S., Govers, G., Wartel, S., Meire, P., 2004b. Modelling estuarine Yang, S.L., 1998. The role of Scirpus marsh in attenuation of hydrodynamics and
variations in tidal marsh sedimentation: response to changing sea level and retention of fine sediment in the Yangtze estuary. Estuarine, Coastal and Shelf
suspended sediment concentrations. Marine Geology 212(1–4), 1–19. Science 47, 227–233.

Biographical Sketch

Sergio Fagherazzi is an associate professor of surface processes and marine sciences at the Department of Earth
Sciences, Boston University. He obtained a Doctoral Degree in Hydrodynamics in 1999 at the Department of
Environmental, Maritime, Geotechnical, and Hydraulic Engineering, University of Padua, Italy. His primary study
interest concerns the evolution of coastal environments. His research activities cover all aspects of coastal mor-
phodynamics, including wetlands erosion, hurricanes impact on sandy beaches and mitigation, tsunami effects
on shorelines, and long-term evolution caused by sea level rise and climate change. He is a member of the
editorial board for Earth Surface Processes and Landforms.
198 Ecogeomorphology of Salt Marshes

Duncan FitzGerald is a Professor in the Department of Earth Sciences with an expertise in coastal processes,
stratigraphy, and morphodynamics. He received his Masters in Geological Oceanography at Texas A&M and his
PhD in Geology at the University of South Carolina. He is a marine geologist who studies sediment exchange,
hydrodynamics, and coastal evolution of marshes, estuaries, river deltas, barrier islands, and tidal inlets. His three
major research themes are presently focused on coastal response to accelerating sea-level rise, impact of major
storms along the Louisiana coast, and climatic and oceanographic controls on strand plain development in Brazil.
His textbook on coastal geology is used both nationally and internationally. He is a Fellow of the Geological
Society of America and has received numerous teaching awards at Boston University.

Robinson (Wally) Fulweiler is assistant professor of Earth Sciences at Boston University. She is a biogeochemist
and coastal ecosystem ecologist. Her research has included coastal watershed mass balances of major biogenic
elements in New England (C, N, P, Si), the biogeochemistry of nitrogen in coastal marine ecosystems, especially
sediments, and wetland ecology in coastal Louisiana. Her recent focus has been on how climate change may
influence nitrogen fixation and denitrification in estuarine and shelf systems and anthropogenic impacts on the
coastal silica cycle.

Zoe Hughes is a senior postdoctoral research associate in the Department of Earth Sciences, Boston University. She
received her PhD from the University of Southampton working at the National Oceanography Centre, South-
ampton. Hughes uses a combination of modeling and field investigations to examine the impacts of waves, tidal
currents, and sea level rise on coastal sediment transport and geomorphology. Her recent work involves col-
laborations with ecologists, undertaking interdisciplinary studies of feedbacks between hydrodynamics, sediment
movement, and both flora and fauna. She is involved in a range of studies within coastal barrier and saltmarsh
systems along the East and Gulf coasts of the US.

Patricia L. Wiberg is Professor and Chair of Environmental Sciences at the University of Virginia, where she has
taught since 1990. Wiberg received her PhD in oceanography from the University of Washington in 1987. Her
research focuses on the mechanics of sediment erosion, transport, and deposition as well as associated evolution
of sediment bed properties and morphology. Her research topics include mechanisms and rates of saltmarsh
erosion and accretion, sediment dynamics on tidal flats, effects of seagrass on turbidity in coastal lagoons, and the
impacts of sea-level rise and climate change on the evolution of coastal landscapes. She currently serves as the
chair of the Marine Working Group of the Community Surface Dynamics Modeling System (CSDMS).
 Ecogeomorphology of Salt Marshes 199

Karen McGlathery is a Professor of Environmental Sciences at the University of Virginia. Her research focuses on
the effects of long-term change, including climate, sea-level rise, land-use, and species invasions in coastal marine
ecosystems. She received her BS from Connecticut College, her PhD from Cornell University, and she did her post-
doctoral work at the University of Copenhagen and the National Environmental Research Institute in Denmark.
She is the Program Director of the Virginia Coast Reserve Long Term Ecological Research site, which is one of 25
sites in the nation funded by the National Science Foundation to study long-term environmental changes in
marine and terrestrial ecosystems. Karen serves on the editorial board of the journal Ecosystems.

James Morris is the Director of the Belle Baruch Institute for Marine and Coastal Sciences, Professor of Biological
Sciences, Distinguished Professor of Marine Studies at the University of South Carolina, and is a AAAS Fellow. He
served as a Program Officer at the National Science Foundation in the Division of Environmental Biology from
2003–2005 and was a visiting professor at Aarhus University, Denmark in 1990. His academic background
includes degrees in environmental sciences, (BA, Univ. Virginia), biology (MA, Yale), and forestry and environ-
mental studies (PhD, Yale). He held a postdoctoral fellowship at the Marine Biological Laboratory, Woods Hole
before taking a faculty position at the University of South Carolina in 1981. He currently serves on the Wetland
Carbon Modeling working group at the National Center for Ecological Synthesis, the Special Science, Engineering,
and Technology panel that is making recommendations on the restoration of the Mississippi River Delta, and an
NRC Gulf Oil Spill committee.

Trevor Tolhurst is a lecturer in coastal processes at the School of Environmental Sciences, University of East Anglia,
UK. He obtained a Doctoral Degree in Marine Biology in 2000 at the Gatty Marine Laboratory, University of St.
Andrews, Scotland. His primary interests concern the sedimentology and ecology of intertidal sedimentary
habitats. His research is multidisciplinary and field based, investigating the nature of direct and indirect inter-
actions between sediments and biota, particularly how biota affects the erosion of sediments. He has helped
develop various devices for measuring erosion, as well as remote sensing techniques for investigating microbial
biofilms.

Linda Deegan is a Senior Scientist at The Ecosystems Center, Marine Biological Laboratory at Woods Hole. She
obtained a doctorate in Marine Science in 1985 from Louisiana State University. Her primary interest is in how
animals through their behavior and interactions shape ecosystem processes, including geomorphic dynamics and
nutrient cycling. She works in many ecosystems, from temperate coastal wetlands to arctic and tropical streams.
She has been a member of the editorial boards of Ecological Applications and Estuaries and Coasts. She currently
serves on the Board of The Nature Conservancy and is a member of the Science and Engineering Special Team
(SEST) advising non-governmental organizations, especially Environmental Defense Fund, National Audubon
Society, and National Wildlife Federation, on issues underlying the restoration of the Mississippi River delta in the
Gulf of Mexico.
200 Ecogeomorphology of Salt Marshes

David Samuel Johnson is a research associate in the Ecosystems Center at the Marine Biological Laboratory in
Woods Hole, MA. He is a marine ecologist who is interested in species interactions and ecosystem processes.
David is particularly fond of invertebrates. He received his PhD from Louisiana State University, Baton Rouge,
Louisiana, in 2008.
 12.13 Ecogeomorphology of Tidal Flats
S Fagherazzi, DM FitzGerald, RW Fulweiler, and Z Hughes, Boston University, Boston, MA, USA
PL Wiberg and KJ McGlathery, University of Virginia, Charlottesville, VA, USA
JT Morris, Belle Baruch Institute for Marine & Coastal Sciences, University of South Carolina, Columbia, SC, USA
TJ Tolhurst, School of Environmental Sciences, University of East Anglia, Norwich, UK
LA Deegan and DS Johnson, Marine Biological Laboratory, Woods Hole, MA, USA
r 2013 Elsevier Inc. All rights reserved.

12.13.1 Physiography, Sedimentology, and Stratigraphy of Tidal Flats 202

12.13.1.1 Tidal Flats Deposits 203
12.13.2 Biofilms in Tidal Flat Sediments 208
12.13.2.1 What are Biofilms? 208
12.13.2.2 Diatom Biofilms 209
12.13.2.3 Cyanobacterial Biofilms 209
12.13.2.4 Green Algal Biofilms 209
12.13.2.5 Sediment Stabilization by Biofilms 209
12.13.2.6 Extracellular Polymeric Substances 209
12.13.2.7 Effects of Biofilms on Physical Properties and Processes 211
12.13.2.8 Biotic Mediation of Bedforms 212
12.13.2.9 Destabilization – Buoyant Biofilms 212
12.13.2.10 Biofilms and Rainfall 212
12.13.2.11 Biofilms as Geomorphological Agents 212
12.13.2.12 Biofilms Biogeochemistry 212
12.13.3 Tidal Flats Vegetation and Sediment Transport Interactions 213
12.13.3.1 Modification of Near-Bed Hydrodynamics 213
12.13.3.2 Vegetation Density Effects 214
12.13.3.3 Feedbacks and Bistability 215
Acknowledgments 216
References 216

Glossary helps to structure the sediment microbial

Belowground production The belowground production environment.
of organic material from roots and rhizomes. Flasers Sedimentary structures characterized by sand
Benthic macroalgae Macroscopic algae that cover the ripples partly infilled with mud, usually created by
ocean bottom. intermittent flows in tidal environments. In flasers, the sand
Biofilm Microorganisms (diatoms, cyanobacteria, and deposits are larger than the mud deposits.
green algae) and their extracellular products bound to a Lenticular bed A sedimentary layer that is thickest at one
surface. point and thins away in every direction, forming a lens.
Bioturbation Mixing and reworking of sediment particles Lenticular beds are typical of tidal environments where
by animals and plants. muddy sediments enclose isolated sandy layers (sand
Cyanobacteria Bacteria that obtain energy through lenses).
photosynthesis, also called blue-green algae. Macrophyte Aquatic plants that are large enough to be
Detritus Dead particulate organic material, accumulating apparent to the naked eye. Macrophytes may be emergent,
in soil or carried by water in suspension. submergent, or floating.
Diatoms Unicellular algae with a silica cell wall, which Microphytobenthos The assemblage of microalgae
forms extensive biofilms. (mostly diatoms and cyanobacteria) living at the bottom of
Extracellular polymeric substances (EPS) A the ocean.
mucilaginous extracellular carbohydrate matrix secreted Nutrient enrichment Increased supply of nutrients, such
by organisms in biofilms (diatoms, cyanobacteria, and as nitrogen and phosphorus, to an ecosystem, often caused
green algae). The EPS forms a cohesive matrix that by human activities.

Fagherazzi, S., FitzGerald, D.M., Fulweiler, R.W., et al., 2013.

Ecogeomorphology of tidal flats. In: Shroder, J. (Editor In Chief), Butler,
D.R., Hupp, C.R. (Eds.), Treatise on Geomorphology. Academic Press, San
Diego, CA, vol. 12, Ecogeomorphology, pp. 201–220.

Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00403-6 201

202 Ecogeomorphology of Tidal Flats

Rhizomes An underground stem, usually horizontal, Seagrass meadows Marine flowering plants, with long
from which roots and shoots can grow. and narrow leaves, that grow in large patches (meadows).
Ridge and runnel Mudflat bedforms characterized by Skimming flow The flow above a vegetation canopy. In a
parallel ridges separated by shallow troughs (runnels). skimming flow the entire velocity profile, and related
These bedforms are usually equispaced with a depth of tens boundary layer, is displaced upward.
of centimeters. Wavy bedding Sedimentary structures characterized by
Root scalping The mechanical removal of the canopy alternating rippled sand and mud layers. In wavy bedding,
mat, including roots, from a surface (for example, from a mud and sand deposits are equal.
salt marsh platform).

Abstract

Coastal landscapes are often dominated by extensive tidal flats. Tidal flats are characterized by near-horizontal topography and
are typically depositional environments that store sediments transported by rivers and nearshore currents. These environments
support a diverse biota that modifies the erosive characteristics of the substrate and mediates sediment transport processes.
Biofilms are a ubiquitous feature of intertidal mudflats and stabilize the bottom by secretion of extracellular polymeric
substances and formation of a tough layer protecting the underlying sediments. Seagrasses on subtidal flats also play an
important role in regulating near-bed flow and particle dynamics. A positive feedback between seagrasses, substrate sta-
bilization, and turbidity of the water column ultimately controls the morphological stability of these landforms. In this
chapter, an overview of the sedimentological and physical processes acting on tidal flat sediments have been presented. The
effects of biofilms and seagrasses on tidal flat substrates and sediment transport processes are then introduced.

12.13.1 Physiography, Sedimentology, and low because waves are attenuated as they propagate across
Stratigraphy of Tidal Flats wide low-gradient inner shelves. High wave-energy events can
occur due to the passage of infrequent, large magnitude
Tidal flats are low-relief sedimentary environments that are storms, such as in Shark Bay when hurricanes rework tidal flat
flooded and drained during the rise and fall of the tides. They sediments to form chenier ridges (M. O’Leary, Personal
are composed of a variety of sediment types and are found Communication). Tidal flats are locally affected by small tidal
throughout the world in a wide range of coastal settings creeks (Figure 2) or can be dominated by tidal processes
having low wave energy, such as behind barrier islands where the tidal flats abut or are surrounded by major tidal
(i.e., Friesian Islands, North Sea; Outer Banks, North Car- channels, such as those that occur within macrotidal embay-
olina), inside deep embayments (i.e., Jade, Germany; Skagit ments (i.e., Bay of Fundy; Dalrymple and Zaitlin, 1989).
Bay, Washington; Shark Bay, Australia; and Missionary Bay, In temperate zones and high latitudes, intertidal flats
Hinchinbrook Island, Australia), at the mouth of estuaries consist primarily of siliciclastics, contrasting to the tropics
(i.e., Chesapeake Bay, East Coast of USA and The Wash, UK), where they are dominated by carbonates, except in regions of
in deltaic regions (i.e., Irrawaddy Delta, Myanmar and Niger high terrigenous input, such as the Niger Delta and the mouth
Delta, Nigeria), landward of wide shallow shelves (i.e., west- of the Amazon. Typically, tidal flats exhibit a zonation in the
ern Andros Island, Bahamas), and along open coasts having grain size that reflects a gradual decrease in wave energy
macrotidal ranges (i.e., Bay of Fundy, Nova Scotia; Mont Saint whereby the lower, more exposed parts of the flat are com-
Michel, France; and German Bight, North Sea) (Figure 1). posed of sand and the increasingly landward, more elevated,
Tidal flats are primarily depositional environments where an and lower wave-energy portions of the flats transition to mud.
abundant supply of sediment produces vertical and lateral This trend is apparent in small bays (Manitounuk Sound,
accretion (Reineck, 1972). An exception occurs in Hudson Bay Hudson Bay, 0.7-km wide; Ruz et al., 1998), moderately sized
where the coast is undergoing isostatic rebound and tidal flats (18-km long; Lee et al., 2004; Jade, Germany, 15-km long;
have been created through the erosion of uplifted deepwater Reineck and Singh, 1980), and in deep estuaries where tidal
sediment (Ruz et al., 1998). The extent of tidal flats is gov- flats are highly separated from one another (Paranagua,
erned by: (1) antecedent topography, which dictates accom- Parana, Brazil, 40-km long, Netto and Lana, 1997) (Figure 3).
modation space and (2) volume of sediment delivered to the As seen in the Paranagua Estuary, the mean grain size of the
coast from onshore, longshore, and offshore sources, or tidal flats gradually coarsens toward the mouth of the estuary,
formed in place biogenically or chemically. Sedimentation fills reflecting an increase in wave and tidal energy (Figure 4, Netto
accommodation space creating the broad tidal flat platform and Lana, 1997). Likewise, sorting increases, whereas organic
(Figures 1 and 2); flats also form as a thin veneer overlying the content decreases.
bedrock or other planar surfaces (Semeniuk, 1981). Early studies of tidal flat sedimentation used the concepts
Although tidal flats exist in low wave-energy settings, of settling and scouring lags to explain the landward transport
sedimentation processes are primarily a product of wave- of particles on tidal flats (Postma, 1961, 1967; van Straaten
induced erosion and subsequent transport by tidal currents. In and Kuenen, 1957). Postma (1967) showed that the
protected environments, waves are small and locally wind time–velocity asymmetry of the currents could also help in the
generated. Wave energy at tidal flats along open coasts is also landward transport of sediment. Since that time, numerous
 Ecogeomorphology of Tidal Flats 203

Shark Bay

Pleasant Bay

(a) (b)

Niger Delta Coos Bay

(c) (d)

Figure 1 Aerial photographs of tidal flats around the world: (a) Shark Bay, western Australia; (b) Pleasant Bay, Cape Cod, MA, USA; (c) Niger
Delta, Nigeria; (d) Coos Bay, OR, USA.

studies have focused on the process–response of tidal flats and 12.13.1.1 Tidal Flats Deposits
the resulting sedimentary structures and facies relationships,
including work along the German coast (Reineck and Singh, The inundation and draining of tidal flats dictates that tidal
1980; Reineck and Wunderlich, 1968), at the Wash in currents in channels and creeks are constantly accelerating and
England (Evans, 1965), and in the Bay of Fundy (Klein, 1967; decelerating and wave-induced shear stresses on flats are
Knight and Dalrymple, 1975). In addition, there have likewise increasing and decreasing as a function of water
been several reference compilations dealing with tidal flats depth and wave energy. This cyclic regime in energy results in
deposits (Reineck and Singh, 1980; Klein, 1977; Ginsburg, alternating transport conditions between the bedload and
1975). suspended load. Sand is deposited during the periods of
204 Ecogeomorphology of Tidal Flats

(a) (b)

(c) (d)

(e) (f)

Figure 2 Tidal flat ground photographs: (a) Brewster tidal flats; (b) Stewart Island flats New Zealand; (c) Rills on mud; (d) Mudflat; (e) Fiddler
crabs on flat; (f) Ghost shrimp holes on tidal flat.
 Ecogeomorphology of Tidal Flats 205

37°00′N
N N

53°31′50″N

36°55′N

Sand
Muddy sand
36°50′N
Sandy mud
Fine sand
Mud 53°23′51″N
Muddy sand

0 1 2 3 4 5 Mud 0 1 2 3 4
(a) Km (b) Km
126°20′E 126°25′E 8°03′05″E 8°18′50″E

55°32′N N

Low tide zone: sand

Mid tidal flat zone: silty sand
Channel zone: sand
Upper tidal flat zone: silt clay

Kuugaapik
River
Manitounuk

55°33′N
0 500
Meters
(c)
77°16′W 77°17′W
Figure 3 Sediment zonation in tidal flats: (a) Garolim Bay. Reproduced from Lee, H.J., Jo, H.J., Chu, Y.S., Bahk, K.S., 2004. Sediment transport on
macrotidal flats in Garolim Bay, west Korea: significance of wind waves and asymmetry of tidal currents. Continental Shelf Research 24, 821–832;
(b) Jade, North Sea. Reproduced from Reineck, H.E., Singh, I.B., 1980. Depositional Sedimentary Environments. Springer-Verlag, Berlin, 439 pp,
and Gadow, S., 1970. Sedimente und Chemismus. In: Reineck, H.-E. (Ed.), Das Watt, Ablagerungs- und Lebenstraum. W. Kramer, Frankfurta. M.,
pp. 23–35, with permission from Springer; (c) Manitounuk Bay. Reproduced from Ruz, M., Allard, M., Michaud, Y., Héquette, A., 1998.
Sedimentology and evolution of subarctic tidal flats along a rapidly emerging coast, Eastern Hudson Bay, Canada. Journal of Coastal Research 14,
1242–1254.

current or wave-induced bedload transport and mud is members: (1) high-energy ripple cross-bedded sand with in-
deposited during low energy and slack water conditions. frequent mud flasers and (2) low-energy muddy layers having
Interbedded and interlaminated sand and mud are one of occasional sandy lenticular beds. Transitioning between flasers
the diagnostic characteristics of tidal flat deposits. Reineck and lenticular beds is wavy bedding in which sand and mud
and Wunderlich (1968) have classified the primary sedi- layers alternate to form continuous thin beds or laminae
mentary structures of mixed-sediment flats as having two end (Figure 5).
206 Ecogeomorphology of Tidal Flats

48° 40′W 30′ 20′

Brazil
1 N

25° 25′S 2 3
Antonina

4 9
5 7
Paranagua Bay Cotinga
10 Island
30′ 6 11 13
8 Mel Island

Pedras Paranagua 15 14 17 18
12
Island 16
20
5 Km 19
35′

35 30

Organic content (%)

30 25
20
Salinity

25
15
20
10
15 5
10 0
1 5 10 15 20 1 5 10 15 20

2.5 100
2 80
Sand (%)

1.5 60
Sorting

1 40
0.5 20
0 0
1 5 10 15 20 1 5 10 15 20

6 100
5.5
80
Grain size (φ)

5
4.5
Silt (%)

60
4
3.5 40
3
2.5 20
2 0
1 5 10 15 20 1 5 10 15 20
2 5
1.6
Carbonate (%)

4
Clay (%)

1.2 3
0.8 2
0.4 1
0
0
1 5 10 15 20 1 5 10 15 20
Inner Outer
Station number
Figure 4 Sediment trends in the Paranagua Estuary in Parana, Brazil. Reproduced from Netto S.A, Lana, P.C., 1997. Influence of Spartina
alterniflora on Superficial sediment characteristics of tidal flats in Paranagua Bay in Southeastern Brazil. Estuarine Coastal and Shelf Science 44,
641–648.
 Ecogeomorphology of Tidal Flats 207

(a) (b)

(c) (d)

Figure 5 Examples of primary sedimentary structures of tidal flats: (a) Flaser bedding; (b) Flasers at the surface and wavy bedding at the
bottom; (c) Wavy bedding; (d) Lenticular bedding. Reproduced with permission from Reineck, H.E., Singh, I.B., 1980. Depositional Sedimentary
Environments. Springer-Verlag, Berlin, 439 pp.

In creeks and tidal channels within tidal flats, migrating Tidal flats are reworked by the migration and meandering
bedforms produce cross-bedding indicative of the dominant of small creeks (depth o2 m) as well as major tidal channels
current direction. Bedforms range in size from ripples (lo0.6 m) (depth ¼ 4 to 410 m). Reineck (1972) reports that com-
to megaripples (0.6olo6.0 m). In channels where ebb and parison of historical maps behind Wangerooge Island
flood current velocities are nearly equal and migrating, bedforms along the German North Sea indicates that 58% of the tidal
are partly preserved, herring-bone stratification is produced, and flats in this region were reworked during a 68-year period
is a diagnostic characteristic of a tidal environment. Klein (1970) (1879–1947) to depths of 0.5 to 46 m. Commonly, the
showed that in sandy tidal flat systems, bedforms migrate in the surface of tidal flats is also reworked by organisms, which can
direction of the stronger current velocity, but when the current completely or partially destroy the primary sedimentary
reverses, the bedform crest is partially eroded producing a trun- structures. Typically, bioturbation structures are most abun-
cated surface. This becomes a reactivation surface when the dant in muddier sediments rather than the sandier portion
bedform resumes its dominant migration trend (Figure 6). of the flats due to the greater energy, which inhibits
208 Ecogeomorphology of Tidal Flats

Constructional event
Dominant current direction

Ripple migration
(a)

Destructional event
Subordinant current direction

Erosion
R
R

(b)

Constructional event
Dominant current direction

(c) R Reactivation surface

Figure 6 Formation of reactivation surface by bedform migration, erosion, and subsequent bedform migration. Reproduced from Klein, G. de V.,
1970. Depositional and dispersal dynamics of intertidal sand bars. Journal of Sedimentary Petrology 40, 1095–1127, with permission from Springer.

organisms; however, mixed sediments can be highly biotur- services. Each day, tidal flat sediments experience strong
bated as well. variations in light intensity as well as alternating periods of
The upper portion of tidal flats transitions to supratidal tidal flooding and ebbing, and aerial exposure. How these
regions have a range of environments from salt marshes factors interact determines the geomorphology, ecology, and
(temperate regions) and mangroves and sabkhas (tropics) to biogeochemistry of tidal flats. Biofilms are a ubiquitous feature
carbonate platforms and beaches and barriers. In a regressive of intertidal sand and mudflats, which are particularly im-
system, tidal flats exhibit a fining upward sequence containing portant because of their role as primary producers and in me-
a basal shelly lag produced by reworking of tidal channel diating numerous processes. The microorganisms in biofilms
sediments. This layer is overlain by cross-bedded sand transi- mediate both small scale (i.e., chemical zonations) and large
tioning to a mixed sediment containing flasers, wavy bedding, scale (i.e., sediment stability) properties and processes of tidal
and lenticular beds. Channel cut and fills can also be present. flats. Although the presence of biofilms on tidal flats is com-
The sequence is capped by mud and finally supratidal sedi- mon, still much is left to learn about how these organisms
ment. Carbonate settings produce tidal flat facies dominated interact with each other and the surrounding environment. For
by fine-grained sediment, shells, cemented layers, algal example, how biofilms and tidal flats will respond to anthro-
laminations, intraclasts, bioturbation structures, channel cut, pogenic impacts such as nutrient loading and climate change
and fills (Shinn et al., 1969). (i.e., increased water temperature, rising sea level, etc.) is cur-
rently of particular relevance. This section will primarily con-
sider the effects of photosynthetic biofilms on intertidal flats.

12.13.2 Biofilms in Tidal Flat Sediments

12.13.2.1 What are Biofilms?
Estuaries and other low-energy coastal environments are often
characterized by intertidal mudflats that were once considered Biofilms are a prominent (although not always visible to the
wastelands (Stal and De Brouwer, 2003). It is now known that naked eye) feature of tidal flats. They are composed of both
these are dynamic, highly productive environments, capable of autotrophic and heterotrophic microorganisms. The photo-
supporting a diverse biota, and provide numerous ecosystem synthetic eukaryotic and cyanobacteria group of organisms in
 Ecogeomorphology of Tidal Flats 209

biofilms are known as benthic algae or microphytobenthos many ways resemble true biofilms (Figures 7(e) and 7(f)).
(MPB). There are a variety of definitions of the term ‘biofilm’ Filamentous green algae are not motile, but grow across the
(or microbial mat), but they are basically microorganisms and surface of the sediment. They cannot, therefore, migrate back
their extracellular products bound to a surface (Neu, 1994). to the sediment surface if they are buried by deposited sedi-
A major component of biofilms is water, anywhere up to 99%. ment. Filaments ramify below the sediment surface, pre-
On intertidal flats, biofilms are a generally thin (mm to mm sumably representing older cells that have been gradually
thick) surface layer, consisting of the microbes, their exudates, buried by deposition of sediment as the algae grow. These
and sediment particles. Photosynthetic intertidal biofilms must algae do not secrete copious amounts of EPS, so they do not
be exposed to sufficient sunlight for the cells to be able to stabilize the substrate as effectively as diatoms or cyano-
photosynthesize, so they form at or near the sediment surface. bacteria, but their filaments are still able to trap and stabilize
Here they are alternately exposed to overlying water and air over sediment. This results in raised areas of sediment very similar
a tidal cycle (although some organisms are motile and can to those formed by diatoms or cyanobacteria (Figure 7(e)).
descend below the sediment surface if required). Biofilms on Intertidal green algal biofilms are found in both muddy and
intertidal sediments act as a barrier between the sediment and sandy sediments.
the overlying atmosphere/water and are often described as a
‘skin.’ They mediate biological, physical, and chemical processes
across the sediment–water/atmosphere interface making them 12.13.2.5 Sediment Stabilization by Biofilms
vital components in the structure, functioning, and dynamics of
Various processes and properties of intertidal flats are mediated
intertidal flats. There are a variety of different types of biofilm
by biofilms, but probably one of the most important influences
organisms, some of which are described below.
of biofilms on the geomorphology of intertidal flats is to sta-
bilize intertidal sediments, making them harder to erode
12.13.2.2 Diatom Biofilms (Paterson, 1989; Black et al., 2002; Larson et al., 2009). This is
achieved by a number of mechanisms and processes, among
Diatoms are unicellular algae with a silica cell wall, which them: (1) secretion of EPS, (2) smoothing of the sediment
form extensive, generally brown or golden brown coloured, surface roughness, (3) network effects, and (4) formation of a
biofilms (Figures 7(a) and 7(b)). They are motile and can tough layer or ‘skin’ protecting underlying sediments (see
move from the sediment surface deeper into the sediment. Paterson, 1997 and Black et al., 2002 for more details).
They often migrate below the sediment surface just before the These often occur in concert making the relative contri-
tide comes in to avoid being eroded and washed away, or to bution of each difficult to elucidate. The erodability and sta-
obtain essential nutrients (Tolhurst et al., 2003). When ex- bility of sediments also depends on the sedimentological
posed to the air during low tide, they can migrate into the properties of the substrate, including particle size, water con-
sediment to avoid high light intensities, which could damage tent, bulk density, organic content, depositional history, and
the cell (Mouget et al., 2008). Although the exact mechanism air exposure (Willows et al., 1998; Tolhurst et al., 2006a).
remains poorly understood, their movement seems to rely on
the secretion of extracellular polymeric substances (EPSs)
(Edgar and Pickett-Heaps, 1984). They also secrete EPS as a 12.13.2.6 Extracellular Polymeric Substances
part of unbalanced growth and as a food source for later use
(de Brouwer and Stal, 2002a). These EPS secretions play a vital Biofilm organisms, which include heterotrophic bacteria, have
role in stabilizing sediments (see later sections). This stabil- been shown to increase sediment stability (Grant and Gust,
izing effect often results in diatom biofilms being raised up 1987; Neumeier and Amos, 2006; Lanuru et al., 2007; Lundk-
compared with surrounding areas without a visible biofilm vist et al., 2007; Friend et al., 2003). Many of these organisms
(Figure 7(a)). Biofilm-forming diatoms tend to be found in secrete a mucilaginous extracellular carbohydrate matrix col-
muddier sediments, but can occur in sandy sediments. lectively called EPS (Underwood et al., 1995). Many studies
have shown a positive correlation between sediment stability
and extracellular carbohydrates (Dade et al., 1992; Madsen
12.13.2.3 Cyanobacterial Biofilms et al., 1993; Tolhurst et al., 2002; Friend et al., 2003; de
Brouwer et al., 2002b, 2005). The EPS forms a cohesive matrix
Cyanobacteria are bacteria that can photosynthesize, getting that helps to structure the sediment microbial environment,
their name from the bluish pigment phycocyanin (Figure 7(c)). which controls the physical properties of the biofilm that can
Colonial species form filaments that ramify through the sedi- range from a loose slime to a compact, cohesive gel (Decho,
ment surface forming thick leathery biofilms (Figure 7(d)). The 2000). EPS acts to both enhance the physicochemical cohesive
filaments are able to move, migrating toward light (Stal, 2010). forces between cohesive sediments and, at a larger scale, coats
Cyanobacteria also secrete EPS and are particularly effective at sediment particles, binding them together like a glue. This has
stabilizing sediment. Intertidal cyanobacteria tend to be found two main effects (1) it acts to encourage flocculation of fine
in sandier sediments, but can occur in muddy sediments. sediment, enhancing its deposition and (2) it increases the
strength of the sediment bed (Decho, 1990; Sutherland et al.,
1998a, b; Underwood and Paterson, 1993).
12.13.2.4 Green Algal Biofilms
The effects of EPS on sediment strength are obvious when
Although arguably not biofilms in the strictest sense, fila- natural sediments are cleaned to remove the EPS. Natural
mentous green algae can form mats at the surface, which in muddy sediments are viscoplastic, but when cleaned and
210 Ecogeomorphology of Tidal Flats

(a) (b)

(c) (d)

(e) (f)

Figure 7 (a) Diatom biofilm from Biezelingsche-Ham mudflat, the Netherlands. Note the brown colour and how the biofilm is raised up above the
surrounding areas without a visible biofilm. Scale bar is 4 cm. Image courtesy of Stal, L. (b) Low-temperature scanning electron microscope image
of the surface of a diatom biofilm from the Biezelingsche-Ham mudflat, the Netherlands. It consists of a number of different diatom species with
strands of EPS and a few small flocs of sediment (middle right hand side). Scale bar¼10 mm. (c) Cyanobacterial biofilm from Cairns, Australia.
Note the blue–green colour and oxygen bubbles forming on the surface of the biofilm under a thin film of water on the far left hand side. (d) Light
microscope image of Lyngbya spp. cyanobacteria filaments. Scale bar 40 mm. (e) Green algal mat from Brays Bay, Australia. Note the bright green
colour and how the mat is raised up compared with areas without a mat, where water pools. Small holes are crab burrows. (f) Close up of the
green algal filaments covering the sediment surface. Scale bar is 5 cm.

reconstituted to the same water content they become fluid, sediment, even at large water contents. This is probably due, in
with a consistency similar to that of milk or thin cream. EPS part, to the action of these molecules in stabilizing water (the
appears to be vital in maintaining the viscoplastic structure of pore water) as well as their ability to increase the physico-
natural cohesive sediments, imparting plasticity to the chemical cohesion between sediment particles.
 Ecogeomorphology of Tidal Flats 211

Diatom secretion of EPS is strongly correlated to the motility 12.13.2.7 Effects of Biofilms on Physical Properties and
of the diatom (Underwood and Smith, 1998). EPS secretion Processes
rates vary with species, growth phase, and physiological state
(Decho, 1990; Smith and Underwood, 1998). In addition, the One of the most notable effects of biofilms is the way they alter
chemical composition of the secreted exopolymer can be dif- physical properties and processes in sediments. In abiotic
ferent between different species or change during growth and muddy sediments, as water content decreases the strength (or
this has been connected to changes in sediment stability stability) of the sediment increases. This is because as water is
(Decho, 1990; de Brouwer et al., 2005). Finally, EPS production removed from the sediment, the fine-grained clay particles are
is intimately linked to the vertical migration patterns of epipelic brought closer together, increasing the strength of the inter-
diatoms (Smith and Underwood, 1998), which has been linked particle forces and leading to an increase in cohesion. Thus, as a
to temporal changes in sediment stability over minutes and muddy sediment becomes drier, it becomes stronger and more
hours, as the diatoms migrate to the surface during the day, or resistant to erosion. The presence of a biofilm can reverse this
stay below the sediment surface at night (Tolhurst et al., 2003, relationship. As a biofilm grows and becomes thicker, it in-
2006a; Friend et al., 2005). The production and extrusion of creases the water content of the surface sediment (Tolhurst
exopolymers is metabolically costly to organisms but it is also et al., 2008). It can do this in three different ways; first, the
important in protecting the organisms against a variety of water contained within the cells themselves will increase
stresses (i.e. desiccation, Daniel et al., 1987; Savage and the water content of any sample of sediment that includes the
Fletcher, 1985 and toxin resistance, Decho, 1990). EPS pro- biofilm; second, by enhancing the deposition of loose fine-
duction rates have been found to increase under environmental grained sediment flocs that have a large amount of pore space
stresses (Yallop et al., 2000). For example, as nutrients become filled with water; and finally by ‘stabilizing’ water within the
limiting in a mature biofilm, exopolymer production rates in- biofilm with EPS. Thus, biofilms tend to have large water
crease (Decho, 1990; Mylkestad, 1974). contents. However, because this water is ‘contained’ (either
EPS can also bind and therefore concentrate a variety of within the cells or stabilized by EPS), the sediment remains
metal ions, including Cd2 þ , Cu2 þ , Cr3 þ , Pb2 þ (Decho, plastic at much larger water contents than it would if there was
2000). The binding affinity strength varies and depends on the no biofilm. This results in natural sediments often having large
composition of the EPS, the pH, and salinity. In estuarine water contents and being very strong (Tolhurst et al., 2008), the
environments, pH varies greatly because of freshwater inputs opposite of what would be expected in abiotic sediments.
as well as photosynthesis and respiration (Decho, 2000). Diatoms tend to be found in muddy sediments, although
Typically, in acidic conditions (low pH) ions are released and there is something of a chicken and the egg situation; are
conversely during basic conditions (high pH) ions are bound diatoms preferentially living in muddy sediments, or are they
or chelation is favored. In the surface of intertidal sediments, merely enhancing the deposition and retention of fine grains?
chelation processes have distinct diel patterns that match the Experiments have shown that diatoms are good at stabilizing
cycle of photosynthesis and respiration. Metal binding effi- fine-grained sediments, but less effective in stabilizing coarse-
ciency has also been related to the physical characteristics of grained sediments (Figure 8). It is currently unclear why this
the secretion. EPS in a gel state have been observed to more is, but it has probably something to do with the surface area to
strongly bind metals than a looser, slimy EPS state. Thus, volume (and hence mass) relationship; as particles become
biofilms are vital in mediating heavy metal dynamics in larger, the effect of the EPS binding becomes proportionally
intertidal flats. less. Measurements in the Ems Dollard show that patches of

0.9 Empirical curve

Diatom biofilm
0.8
No biofilm
Erosion threshold (N m−2)

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0 200 400 600 800 1000

Grain size (μm)
Figure 8 Erosion thresholds for natural beach sands with and without a diatom biofilm. As grain size increases, the ability of the biofilm to
stabilize the sediment decreases. The empirical curve is the Shields erosion threshold for sand.
212 Ecogeomorphology of Tidal Flats

biofilm contain considerably more fine-grained sediment than biofilm with attached sediments under flow conditions much
patches a few centimeters away with no visible biofilm. The lower than that would otherwise be expected (Sutherland et al.,
biofilm areas are also raised a few centimeters above the ad- 1998a; Tolhurst et al., 2008). We have collected chunks of
jacent sediments. Measurements of suspended sediment con- diatom biofilm floating downstream after being disturbed by
centrations (SSCs) in the estuarine water column have shown walking over the substrate, it is easy to see a similar process
that SSC can decrease as a diatom biofilm forms, suggesting occurring due to disturbance from fish and/or wading birds.
that the diatom biofilms are acting to strip sediment from the
water column and retain it on the bed, creating the raised
biofilms (Kornman and de Deckere, 1998). This is possible 12.13.2.10 Biofilms and Rainfall
because the diatoms are capable of migrating up through The interaction of biofilms with physical processes can be
sediment deposited on top of them. Scaled up to a whole complex. Rain falling on intertidal flats erodes and mobilizes
estuary, this represents tons of sediment that would otherwise sediments, particularly fine-grained sediment (Mwamba and
be in the water column. All this would suggest that diatoms Torres, 2002; Pilditch et al., 2008; Tolhurst et al., 2006c, 2008).
are acting to deposit, stabilizes, and retain fine sediments, with It might be expected that biofilms would act to prevent this,
processes occurring at a scale of centimeters on the tidal flat stabilizing the sediment against rainfall in a similar way to that
being able to affect SSC estuary wide. It seems likely, therefore, by which they stabilize sediment against tidal flows. However,
that although diatoms may recruit to finer grained sediments, measurements show that rainfall is very effective at negating the
the development of a biofilm also acts to enhance fine-grained stabilizing effect of diatoms (Paterson, 1997; Tolhurst et al.,
deposition and retention, reducing the grain size of the sedi- 2003). It is unclear exactly why this is, but it seems to have
ments where diatoms grow. something to do with the combined effects of the physical
impact of the rain drops and the introduction of freshwater
reducing physicochemical bonding between particles, which is
12.13.2.8 Biotic Mediation of Bedforms particularly dispersive to intertidal muddy sediments. Reduction
One thing that should always be borne in mind when in- in biofilm biomass and increase in rainfall in winter have been
vestigating intertidal flats is that the various biological, phys- proposed as significant contributors to seasonal changes in
ical, and chemical processes are not acting in isolation, they sediment budgets on tidal flats (Pilditch et al., 2008).
may be synergistic or antagonistic. An example of this seems to
be the formation of ridge–runnel systems, shore normal
channels, and humps found on some intertidal mudflats. 12.13.2.11 Biofilms as Geomorphological Agents
Diatoms occur in greater amounts on the ridges, which drain Given the various ways in which biofilms act to stabilize
quickly and are therefore drier than the runnels, which have intertidal sediments, and given that all natural sediments
less diatoms and are wetter (often containing a surface film of contain these organisms (even if visible biofilms are not pre-
water). It has been shown that sediment submerged under- sent), it would seem that more finer grained sediment is de-
water is less stable than the same sediment exposed to air and posited on intertidal flats than would be the case if the
able to drain (Tolhurst et al., 2006b). Thus, the physical and biofilms were not present. The implication is that intertidal
biological features of ridges are acting in concert to increase flats would be less extensive and sandier were not for biofilms.
their stability, whereas in runnels, they are acting to decrease A study in Sydney, Australia, where algal biofilms were re-
stability; there is a synergy between the physical and biological moved with an algicide, showed that erosion of fine-grained
processes acting to maintain the ridge–runnel feature. sediment was enhanced, leaving bare sandier sediment (un-
In sandier sediments, biofilms can act to prevent the published data). This is how intertidal sediments may have
sediment bed from responding as expected to changes in been before life and biofilms evolved.
hydrodynamic conditions (Grant et al., 1986; Friend et al., The effect of biofilms on intertidal sediments has con-
2008). Ripple marks can become colonized by diatoms, which siderable implications for the interpretation of sedimentary
stabilize the bedform so that it no longer moves under features in the geological record. Ripples may have been re-
hydrodynamic forcing; the sediment bedform is no longer tained under hydrodynamic conditions very different to those
determined by the existing hydrodynamics, but is a relic of they were formed in. Extensive muddy sediments may have
previous conditions. Alternatively, biofilms can stabilize flat been deposited under much more energetic conditions than
sediments; changes in hydrodynamic conditions result in those expected if purely physical processes were in action.
ripples forming in adjacent areas, which then migrate over the Thus, biofilms are an integral component of the complex
top of the stabilized biofilm sediments. intertidal flat system, mediating numerous properties and
processes (Murphy and Tolhurst, 2009).

12.13.2.9 Destabilization – Buoyant Biofilms

12.13.2.12 Biofilms Biogeochemistry
Diatom and cyanobacterial biofilms often have a line of weak-
ness below the main biofilm along which erosion can occur. The importance of MPB in mediating properties and processes
This may represent the old sediment bed surface. In some cases, on intertidal flats makes understanding how and when they
this line of weakness is formed (or enhanced) by the formation grow of vital importance. MPB growth on tidal flats can be
and trapping of oxygen bubbles within the biofilm. Underwater, responsible for more than 50% of total estuarine primary
these become buoyant and can lift away whole chunks of production (Underwood and Kromkamp, 1999). MPB primary
 Ecogeomorphology of Tidal Flats 213

production rates range from 30 to 230 g C m2 yr1 (Krom- attached or free-floating benthic macroalgae, and benthic
kamp and Forester, 2006). Such high productivity helps sustain microalgae (‘MPBs’) all influence sediment suspension and
a robust and diverse food web that provides critical food for transport. Seagrasses are often referred to as ‘ecosystem engin-
migrating shorebirds and fish. eers’ (Jones et al., 1994) because they can alter their physical
MPB inhabit the uppermost surface of sediments at the environment by affecting currents, waves, and turbulence. The
sediment–water interface. Sediment type explains much of the presence of dense vegetation in the benthic boundary layer
observed variability in MPB. Typically, muddy sediments are alters the roughness of the bottom and affects the vertical vel-
dominated by diatoms, whereas sandy sediments have more ocity profiles, reducing flows within the canopy as currents are
diversity that includes cyanobacteria and euglenids (Jesus deflected over the canopy. Gambi et al. (1990) distinguished
et al., 2009). The depth distribution of benthic algae is also between two hydrodynamic environments related to seagrass
determined, in part, by the sediment type. Generally, in the canopies, a canopy–water interface region characterized by
top 2 mm of muddy sediment, a strong exponential decay of high shear stress and high turbulence intensity, and a below-
chlorophyll (a proxy for algal biomass) is seen, but in sandy canopy region characterized by low shear stress and reduction
sediments the chlorophyll is more evenly distributed and of turbulence. The slower current velocities within the canopy
generally down to a few centimeters (Jesus et al., 2006). These result in decreased sediment suspension and increased particle
differences are attributed to light attenuation patterns, as deposition (e.g., Heiss et al., 2000; Peterson et al., 2004).
photosynthetically active radiation (PAR) is restricted to the
top 2 mm (or less) in muddy sediments and several centi-
meters deep in sandy sediments (Jesus et al., 2009).
12.13.3.1 Modification of Near-Bed Hydrodynamics
It has also been commonly observed that many MPB ex-
hibit vertical migration often associated with water level and/ Seagrass shoot density and architecture influences how the
or light levels (Consalvey et al., 2004; Kromkamp et al., 1998). aboveground vegetation slows flow velocities, traps sediment,
In response to these daily migrations and to the subsequent and attenuates waves and turbulence (Fonseca and Fisher,
diurnal cycle of production and respiration, the entire sedi- 1986; Gambi et al., 1990; Gacia et al., 1999); and the extensive
ment microbial community and chemical zonations also belowground roots and rhizomes help to stabilize the sedi-
show strong diurnal vertical migrations (Fenchel, 1969). In ment and increase resistance to storm and wave disturbance.
fact, vertical sediment profiles of pH, oxygen, hydrogen sul- Most studies on the effect of seagrasses on hydrodynamics and
fide, and nutrients exhibit sharp contrasts between illuminated sediment transport have focused on larger subtidal species,
and aerobic surface sediments to dark anaerobic subsurface such as Zostera marina, Thalassia testudinum, Syringodium fili-
sediments (Macintyre et al., 1996 and references therein). forme, and Posidonia oceanica (Fonseca and Fisher, 1986;
MPB communities have been described as a ‘filter’ of nu- Gambi et al., 1990; Gacia et al., 1999). However, smaller
trients at the sediment–water interface (Henriksen et al., 1980; intertidal species (e.g., Zostera novazelandica) also can have a
Sundback and Miles, 2000; Sundback et al., 2000). They have significant effect on sediment suspension (Heiss et al., 2000).
been shown to alter nutrient fluxes by oxygenating surface Seagrass densities vary from patchy or low-density meadows
sediments through photosynthesis and nutrient assimilation (o100 shoots per square meter) in areas that are physically
(An and Joye, 2001; Eyre and Fergusan, 2002; Sundback and disturbed, have low water quality, or have been recently re-
Miles, 2000; Sundback et al., 2000; Tobias et al., 2003). MPB stored, to very dense meadows (41000 shoots per square
are dependent on water column nutrients and light and are thus meter). In flume studies, Peterson et al. (2004) showed that
sensitive to the daily changes in surface irradiance and tidal there were greater flow reductions inside the canopy with in-
height. However, in contrast to other estuarine primary pro- creasing vegetation density, and that there were significant
ducers, they can remain active throughout the year (Kristensen, differences in flow on the edges of meadows. Their model
1993). On an annual scale, MPB incorporate large amounts of showed that the ‘edge effect,’ or the zone in which flow de-
nitrogen (N) and thus provide an important nitrogen retention celerates, is a declining function of vegetation density, and this
mechanism in systems. In some cases, the photosynthetic pro- influences the variation in mean current flow between seagrass
duction of oxygen by MPB promotes coupled nitrification– patches of different sizes. Flow speeds within seagrass canopies
denitrification making the sediment a nitrogen sink. In this typically are 2–10 times slower than outside the meadow
scenario, ammonium is oxidized to nitrate, while nitrate is re- (Gambi et al., 1990; Gacia et al., 1999), and also less variable
duced to dinitrogen (N2) gas and diffuses out of the system (Heiss et al., 2000). Specific water velocities are often o10
(Sundback et al., 2004; Sundback and Miles, 2000). However, cm s1, but can be as high as 100 cm s1 (Koch, 2001).
in systems with low nitrogen availability, MPB can effectively Although there is general agreement that the slowing of
outcompete denitrifying bacteria for nitrate and in doing so current velocities enhances particle deposition within seagrass
decrease denitrification (Risgaard-Petersen, 2003; Sundback meadows, there are few direct measurements of sediment de-
et al., 2004; Sundback and Miles, 2000). position and retention. Gacia et al. (1999) used sediment
traps in Mediterranean P. oceanica meadows and showed that
seagrass canopies slowed current velocities with intensities
12.13.3 Tidal Flats Vegetation and Sediment proportional to the canopy height, and that particle retention
Transport Interactions was up to 15 times higher than unvegetated sediment. They
found that the particle trapping capacity of the meadow was
Benthic vegetation on subtidal flats plays an important role in related to the surface area of the leaves, leaf bending, and
regulating near-bed flow and particle dynamics. Seagrasses, particle attachment to the leaf surface.
214 Ecogeomorphology of Tidal Flats

The effect of the seagrass canopy on wave attenuation is suspension (Friedrichs et al., 2000). For an individual seagrass
still not well understood (Koch et al., 2006). Wave attenuation shoot, vertical differences in velocities between the canopy
is likely highest when the canopy occupies a large proportion and sediment surface, and horizontal differences in upstream
of the water column (Ward et al., 1984; Fonseca and Cahalan, and downstream velocities result in a pressure gradient
1992), but reduction in wave energy also has been docu- downstream of the seagrass shoot. High pressure near the
mented in deep meadows (Verduin and Backhaus, 2000; sediment surface and low pressure near the top of canopy
Granata et al., 2001). In wave-swept environments, seagrasses where the currents are strongest leads to ascending flows on
are exposed to more complex flows than when unidirectional the downstream side of the shoot (Figure 9; Koch et al.,
flows are dominant. A long flexible shape is advantageous, as 2006). This can result in bed scouring and enhanced sediment
the leaves can sway back and forth with the water motion and suspension, as well as pore water upwelling in permeable
thus minimize forces on the root/rhizome systems that anchor sediments (Nepf and Koch, 1999; Koch and Huettel, 2000).
the plants in the sediment (Koch et al., 2006). On tidal flats where benthic macroalgae are dominant,
Benthic algal populations also can stabilize tidal flat sedi- flow causes the thalli to move and scour the sediment when
ments at times of year when the algae are growing actively and densities are low, increasing sediment suspension relative to
populations are physically stable. Dense populations of mat- bare sediments. Macroalgal transport as bedload has been
forming species such as Ulva rigida and Enteromorpha intesti- documented in a number of systems, including the Lagoon of
nalis stabilize sediments by decreasing shear flow at the sedi- Venice, where Flindt et al. (1997) showed that measured
ment surface (Escartin and Aubrey, 1995) and reducing transport rates of Ulva reached 300 gDW m2 h1 during peak
sediment suspension (Sfriso and Marcomini, 1997; Romano tidal flows. This mechanism for sediment destabilization is
et al., 2003). Thick mats (equivalent to 3.5–6.2 kg wet weight akin to the well-documented phenomenon of saltating or
per square meter) displace velocities vertically and can deflect abrading particles increasing erosion in cohesive sediments
as much as 90% of the flow over the mat, with only 10% of the (e.g., Houser and Nickling, 2001; Thompson and Amos, 2002,
flow traveling through the mat (Escartin and Aubrey, 1995). 2004). Macroalgae that scrape the bed while moving across it
Benthic microalgal mats also stabilize surface sediments can dislodge particles and increase sediment suspension and
by producing and extruding EPS (e.g., Paterson, 1989; de erosion. In cohesive beds, the critical stress required to initiate
Brouwer et al., 2006). EPS bind sediment particles together so erosion is often greater than the stress required to maintain
that the shear stress needed for erosion of the sediment is the sediment in suspension.
increased. This sediment stabilization is a seasonal phenom-
enon at least in temperate systems, and the deposited material
may be resuspended at times of the year when the microalgae Z U
are less productive (Widdows et al., 2002).

12.13.3.2 Vegetation Density Effects

Most previous work on vegetation effects on sediment stabil-
ization has been done on high-density populations, and less is
known about the effects of lower densities on sediment fluxes,
although it is common for seagrass and macroalgae to occur in
low densities or to be patchy in distribution. Lawson (2008)
found that low densities of both seagrass and macroalgae in-
creased sediment suspension by as much as 97% compared
with bare sediments, and that the threshold between destabil-
izing and stabilizing effects of these benthic communities is
density dependent and likely requires the generation of skim-
ming flow. Dense seagrass meadows typically displace velocity
vertically creating a protected area next to the sediment surface
(Gambi et al., 1990). At low densities, flow is deflected hori-
zontally around individual shoots and localized areas of high Figure 9 Vertical differences in velocities between the canopy and
shear stress can develop around the shoots. sediment surface and horizontal differences in upstream and
The threshold between destabilizing and stabilizing effects downstream velocities result in a pressure gradient downstream of
on sediments in seagrass-vegetated tidal flats is likely the seagrass shoot. High pressure near the sediment surface and low
dependent on the ratio of the distance between individual pressure near the top of canopy where the currents are strongest
leads to ascending flows on the downstream side of the shoot. This
shoots and their height, as suggested by Vogel (1994) for
can result in bed scouring and enhanced sediment suspension, as
emergent features. When the distance-to-height ratio is rela-
well as pore water upwelling in permeable sediments. Reproduced
tively large, the localized flow that develops around each with permission from Koch E.W., Ackerman J., van Keulen M.,
feature is independent, whereas when the ratio is 1 or smaller, Verduin, J., 2006. Fluid dynamics in seagrass ecology: from
a vertically displaced skimming flow develops. This mech- molecules to ecosystems. In: Larkum, A.W.D., Orth, R.J., Duarte,
anism has been used to explain the effects of other emergent C.M. (Eds.), Seagrasses: Biology, Ecology and Conservation.
features, such as polycheate worm tubes, on sediment Springer-Verlag, Dordrecht, The Netherlands, pp. 193–225.
 Ecogeomorphology of Tidal Flats 215

Overall, for both seagrass and macroalgae, sediment sus- phytoplankton growth (Folkard, 2005; McGlathery et al.,
pension is influenced largely by whether the flow interacts 2007). The absence of seagrass would increase sediment
with the primary producers as one solid feature (i.e., the resuspension and water column turbidity, thereby making
meadow or mat) or as isolated, individual structures (shoots conditions less hospitable for the seagrass growth. These
or fronds), despite their great differences in morphology feedbacks may lead to alternative states for the tidal flat eco-
(Lawson, 2008). For seagrass, canopy closure over the sedi- system – either seagrass meadows with high water column
ment as the leaves bend with the current depends on the clarity and enough light to support seagrass growth, or bare
morphology of the blades, shoot density, and flow rates, and sediment beds with high levels of suspended sediments in the
this influences sediment suspension (Koch and Gust, 1999). water column and light conditions unsuitable for the seagrass
Similarly, suppression of sediment suspension for sediments growth (van der Heide et al., 2007; Carr et al., 2010).
overlain by macroalgae occurs when flow is deflected around Carr et al. (2010) developed a one-dimensional hydro-
dense algal populations, whereas scouring of the sediment bed dynamic model of vegetation–sediment–water flow inter-
occurs when flow is through the macroalgae, causing move- actions and used it to investigate the strengths of the positive
ment of the algae (Lawson, 2008). feedbacks between seagrass cover, stabilization of bed sedi-
ments, turbidity of the water column, and the existence of a
favorable light environment for seagrasses. The model in-
cludes attenuation of wave orbital velocities due to the sea-
12.13.3.3 Feedbacks and Bistability
grass canopy, combined wave and current stresses, and an
On subtidal flats in temperate environments, the growth of active-layer bed formulation. Modifications to the velocity
seagrass is often limited by light availability, which is affected profile due to drag of seagrass stems and stem deflection are
by turbidity from suspended sediments and phytoplankton accounted for in the model, and an empirical relationship
concentrations, and by colored organic matter (Gallegos, defining the light attenuation coefficient from total suspended
2001). In shallow systems, turbidity is influenced strongly by solids, chlorophyll a, and gilvin is used to calculate PAR at the
internal sediment resuspension caused by waves and tides, canopy surface. The results indicated that under typical con-
and this is the dominant controlling factor in lagoonal and ditions in a temperate shallow coastal lagoon, seagrass was
coastal bays systems that lack riverine inputs (Lawson et al., stable in water depths o2.2 m (51% of the bay bottom deep
2007). Rooted vegetation exerts positive feedbacks on water enough for seagrass growth) and bistable conditions existed
clarity by reducing the susceptibility of sediments to resus- for depths of 2.2–3.6 m (23% of bay), where the preferred
pension, enhancing deposition of fine sediment, and immo- state depended on initial seagrass cover. The remaining 26%
bilizing nutrients produced in the sediment through of the bay is too deep to sustain seagrass (Figure 10).
remineralization that would otherwise be available to support Decreases in the sediment size and increases in water

2
0 shoots
100
250
1
1000
Net specific growth rate (year−1)

Growth
0

Loss
−1

−2
Bistable

−3

−4
1 2 3 4 5 6 7 8 9 10
Depth (m)
Figure 10 Specific growth rate as a function of depth for varying shoot densities. Conditions suitable/unsuitable for seagrass establishment and
growth were evaluated by looking at the sign of the net annual growth rate. Positive (negative) values were associated with favorable
(unfavorable) conditions for seagrass establishment and persistence. Reproduced from Carr, J., D’Odorico, P., McGlathery, K., Wiberg, P.L.,
2010. Stability and bistability of seagrass ecosystems in shallow coastal lagoons. Journal of Geophysical Research – Biogeosciences 115,
G03011, doi:10.1029/2009JG001103, with permission from AGU.
216 Ecogeomorphology of Tidal Flats

temperature and degree of eutrophication shifted the Escartin, J., Aubrey, D.G., 1995. Flow structure and dispersion within algal mats.
bistable range to shallower depths, with more of the bay Estuarine Coastal and Shelf Science 40(4), 451–472.
Evans, G., 1965. Intertidal flat sediments and their environnment of deposition
bottom unable to sustain seagrass (Carr et al., 2010).
in the Wash. Quaternary Journal Geological Society of London 121,
Further research is needed on the effects of vegetation 209–245.
patchiness (size of patches and proximity) on sediment sta- Eyre, B., Fergusan, A., 2002. Comparison of carbon production and decomposition,
bilization of tidal flats. Also, more work is needed on the role benthic nutrient fluxes and denitrfication in seagrass, phytoplankton, benthic
of tidal flats, and the feedback between benthic vegetation and microalgae, and macroalgae dominated warm temperatre Australian lagoons.
Marine Ecology-Progress Series 229, 43–59.
sediment stabilization, on the supply of sediments for marsh Fenchel, T., 1969. The ecology of marine microbenthos 1V. Structure and
accretion. function of the benthic ecosystem, its chemical and physical factors and the
microfauna communities with special reference to the ciliated protozoa. Ophelia
6, 1–182.
Acknowledgments Flindt, M., Salomonsen, J., Carrer, M., Bocci, M., Kamp-Nielsen, L., 1997. Loss,
growth and transport dynamics of Chaetomorpha aerea and Ulva rigida in the
This research was supported by the Department of Energy, Lagoon of Venice during an early summer field campaign. Ecological Modelling
NICCR program award # TUL-538-06/07; by the National 102, 133–141.
Folkard, A.M., 2005. Hydrodynamics of model Posidonia oceanica patches in
Science Foundation, LENS project award # OCE 0924287 and shallow water. Limnology and Oceanography 50, 1592–1600.
OCE 0923689, TIDE project award # DEB 0816963, the VCR- Fonseca, M.S., Cahalan, J.S., 1992. A preliminary evaluation of wave attenuation by
LTER program award # DEB 0621014, the PIE-LTER program 4 species of seagrass. Estuarine Coastal Shelf Science 35, 565–576.
award # OCE 9726921; and by the Office of Naval Research Fonseca, M.S., Fisher, J.S., 1986. A comparison of canopy friction and sediment
movement between four species of seagrass with reference to their ecology and
award # N00014-10-1-0269.
restoration. Marine Ecology Progress Series 29, 15–22.
Friedrichs, M., Graf, G., Springer, B., 2000. Skimming flow induced over a
simulated polycheate tube lawn at low population densities. Marine Ecology
References Progress Series 192, 219–228.
Friend, P.L., Ciavola, P., Cappucci, S., Santos, R., 2003. Bio-dependent bed
parameters as a proxy tool for sediment stability in mixed habitat intertidal
An, S., Joye, S., 2001. Enhancement of coupled nitrification–denitrificaiton by areas. Continental Shelf Research 23, 1899–1917.
benthic photosynthesis in shallow estuarine sediments. Limnology and Friend, P.L., Lucas, C.H., Holligan, P.M., Collins, M.B., 2008. Microalgal mediation
Oceanography 46, 61–74. of ripple mobility. Geobiology 6, 70–82.
Black, K.S., Tolhurst, T.J., Paterson, D.M., Hagerthey, S.E., 2002. Working with Friend, P.L., Lucas, C.H., Rossington, S.K., 2005. Day-night variation of cohesive
natural cohesive sediments. Journal of Hydraulic Engineering 128, 2–8. sediment stability. Estuarine Coastal and Shelf Science 64, 407–418.
de Brouwer, J.F.C., de Neu, T.R., Stal, L.J., 2006. On the function of secretion of Gacia, E.T., Granata, C., Duarte, C.M., 1999. An approach to measurement of
extracellular polymeric substances by benthic diatoms and their role intertidal particle flux and sediment retention with seagrass (Posidonia oceanica)
mudflats. In: Kromkamp, J.C., de Brouwer, J.F.C., Blanchard, G.F., Forster, R.M., meadows. Aquatic Botany 65, 255–268.
Creach, V. (Eds.), Functioning of Microphytobenthos in Estuaries. Royal Gallegos, C.L., 2001. Calculating optical water quality targets to restore and protect
Netherlands Academy of Arts and Sciences, Amsterdam, pp. 45–61. submersed aquatic vegetation: overcoming problems in partitioning the diffuse
de Brouwer, J.F.C., Ruddy, G.K., Jones, T.E.R., Stal, L.J., 2002b. Sorption of EPS to attenuation coefficient for photosynthetically active radiation. Estuaries 24,
sediment particles and the effect on the rheology of sediment slurries. 381–397.
Biogeochemistry 61, 57–71. Gambi, M.C., Nowell, A.R.M., Jumars, P.A., 1990. Flume observations on flow
de Brouwer, J.F.C., Stal, L.J., 2002a. Daily fluctuations of exopolymers in cultures dynamics in Zostera marina (eelgrass) bed. Marine Ecology Progress Series 61,
of the benthic diatoms Cylindrotheca closterium and Nitzschia sp 159–169.
(Bacillariophyceae). Journal of Phycology 38, 464–472. Ginsburg, R.N. (Ed.), 1975. Tidal Deposits. Springer, Berlin, 428 pp.
de Brouwer, J.F.C., Wolfstein, K., Ruddy, G.K., Jones, T.E.R., Stal, L.J., 2005. Biogenic Granata, T.C., Serra, T., Colomer, J., Casamitjana, X., Duarte, C.M., Garcia, E.,
stabilization of intertidal sediments: the importance of extracellular polymeric 2001. Flow and particle distributions in a nearshore meadow before and after a
substances produced by benthic diatoms. Microbial Ecology 49, 501–512. storm. Marine Ecology Progress Series 218, 96–106.
Carr, J., D’Odorico, P., McGlathery, K., Wiberg, P.L., 2010. Stability and bistability Grant, J., Bathmann, U.V., Mills, E.L., 1986. The Interaction between Benthic Diatom
of seagrass ecosystems in shallow coastal lagoons. Journal of Geophysical Films and Sediment Transport. Estuarine Coastal and Shelf Science 23, 225–238.
Research – Biogeosciences 115, G03011. http://dx.doi.org/10.1029/ Grant, J., Gust, G., 1987. Prediction of coastal sediment stability from photopigment
2009JG001103. content of mats of purple sulfur bacteria. Nature 330, 244–246.
Consalvey, M., Jesus, B., Perkins, R.G., Brotas, V., Underwood, G.J.C., Paterson, van der Heide, T., van Nes, E.H., Geerling, G.W., et al., 2007. Positive feedbacks in
D.M., 2004. Monitoring migration and measuring biomass in benthic biofilms: seagrass ecosystems: implications for success in conservation and restoration.
the effects of dark/far-red adaptation and vertical migration on fluorescence Ecosystems 10, 1311–1322.
measurements. Photosynthesis Research 81, 91–101. Heiss, W.M., Smith, A.M., Probert, P.K., 2000. Influence of the small intertidal
Dade, W.B., Nowell, A.R.M., Jumars, P.A., 1992. Predicting erosion resistance of seagrass Zostera novazelandica on linear water flow and sediment texture. New
muds. Marine Geology 105, 285–297. Zealand Journal of Marine and Freshwater Research 34, 689–694.
Dalrymple, R.W., Zaitlin, B.A., 1989. Tidal deposits in the macrotidal Cobequid Bay- Henriksen, K., Hansen, J., Blackburn, T.H., 1980. The influence of benthic infauna
Salmon River estuary, Bay of Fundy. Canadian Society of Petroleum Geologists, on exchange rates of inorganic nitrogen between sediment and water. Ophelia 1,
Field Guide. Second International Research Symposium on Clastic Deposits. 249–256.
Calgary, Alberta, 84 pp. Houser, C.A., Nickling, W.G., 2001. The factors influencing the abrasion efficiency
Daniel, G.F., Chamberlain, A.H.L., Jones, E.B.G., 1987. Cytochemical and electron- of saltating grains on a clay-crusted playa. Earth Surface Processes and
microscopic observations on the adhesive materials of marine fouling diatoms. Landforms 26, 491–505.
British Phycology Journal 22, 101–118. Jesus, B., Brotas, V., Ribeiro, L., Mendes, C.R., Cartaxana, P., Paterson, D.M.,
Decho, A., 1990. Microbial exopolymer secretions in ocean environments: their 2009. Adaptations of microphytobenthos assemblages to sediment type and tidal
role(s) in food webs and marine processes. Oceanography and Marine Biology: position. Continental Shelf Research 29, 1624–1634.
Annual Review 28, 73–153. Jesus, B., Mendes, C.R., Brotas, V., Paterson, D.M., 2006. Effect of sediment type
Decho, A.W., 2000. Microbial biofilms in intertidal systems: an overview. on microphytobenthos vertical distribution: modelling the productive biomass
Continental Shelf Research 20, 1257–1273. and improving ground truth measurements. Journal of Experimental Marine
Edgar, L.A., Pickett-Heaps, J.D., 1984. Diatom locomotion. In: Round, F.E., Biology and Ecology 332, 60–74.
Chapman, D.J. (Eds.), Progress in Phycological Research. Biopress, Bristol, 3, Jones, C.G., Lawton, J.H., Shachak, M., 1994. Organisms as ecosystem engineers.
pp. 47–88. Oikos 69, 373–386.
 Ecogeomorphology of Tidal Flats 217

Klein, G. de V., 1967. Paleocurrent analysis in relation to modern marine sediment Mylkestad, S., 1974. Production of carbohydrates by marine planktonic diatoms. 1.
dispersal patterns. American Association of Petroleum Geologists Bulletin 51, Comparison of nine different species in culture. Journal of Experimental Marine
366–382. Biology and Ecology 15, 261–274.
Klein, G. de V., 1970. Depositional and dispersal dynamics of intertidal sand bars. Nepf, H.M., Koch, E.W., 1999. Vertical secondary flows in stem arrays. Limnology
Journal of Sedimentary Petrology 40, 1095–1127. and Oceanography 44, 1072–1080.
Klein, G. de V., 1977. Clastic Tidal Facies. CEPCO, Champaign, IL, 149 pp. Netto, S.A., Lana, P.C., 1997. Influence of Spartina alterniflora on superficial
Knight, R.J., Dalrymple, R.W., 1975. Intertidal sediments from the South Shore of sediment characteristics of tidal flats in Paranagua Bay in Southeastern Brazil.
Cobequid Bay, Bay of Fundy, Nova Scotia, Canada. In: Ginsburg, R.N. (Ed.), Estuarine Coastal and Shelf Science 44, 641–648.
Tidal Deposits: A Casebook of Recent Examples and Fossil Counterparts. Neu, T.R., 1994. Biofilms and microbial mats. In: Krumbein, W.E., Paterson, D.M.,
Springer-Verlag, pp. 47–55. Stal, L.J. (Eds.) Biostabilisation of Sediments. Bibliotheks und
Koch, E.W., 2001. Beyond light: physical, geological and geochemical parameters Informationssystems der Carl von Ossietzky, Univsitat Oldenburg, Oldenburg.
as possible submersed aquatic vegetation habitat requirements. Estuaries 24, Neumeier, U., Amos, C.L., 2006. The influence of vegetation on turbulence and flow
1–17. velocities in European salt-marshes. Sedimentology 53, 259–277.
Koch, E.W., Ackerman, J., van Keulen, M., Verduin, J., 2006. Fluid dynamics in Paterson, D.M., 1989. Short-term changes in erodibility of intertidal cohesive
seagrass ecology: from molecules to ecosystems. In: Larkum, A.W.D., Orth, R.J., sediments related to the migratory behaviour of epipelic diatoms. Limnology and
Duarte, C.M. (Eds.), Seagrasses: Biology, Ecology and Conservation. Springer Oceanography 34, 223–234.
Verlag, Dordrecht, The Netherlands, pp. 193–225. Paterson, D.M., 1997. Biological mediation of sediment erodibility: ecology and
Koch, E.W., Gust, G., 1999. Water flow in tide and wave dominated beds of the physical dynamics. In: Burt, N., Parker, R., Watts, J. (Eds.), Cohesive Sediments:
seagrass Thalassia testudinum. Marine Ecology Progress Series 184, 63–72. 4th Nearshore and Estuarine Cohesive Sediment Transport Conference
Koch, E.W., Huettel, M., 2000. The impact of single seagrass shoots on solute INTERCOH ’94. John Wiley and Sons, Wallingford, UK, pp. 197–211.
fluxes between the water column and permeable sediments. Biologia Marina Peterson, C.H., Luettich, R.A., Micheli, F., Skilleter, G.A., 2004. Attenuation of water
Mediterranea 7, 235–239. flow inside seagrass canopies of differing structure. Marine Ecology Progress
Kornman, B.A., de Deckere, E.M.G.T., 1998. Temporal variation in sediment Series 268, 81–92.
erodibility and suspended sediment dynamics in the Dollard estuary. In: Black, Pilditch, C.A., Widdows, J., Kuhn, N.J., Pope, N.D., Brinsley, M.D., 2008. Effects of
K.S., Paterson, D.M., Cramp, A. (Eds.), Sedimentary Processes in the Intertidal low tide rainfall on the erodibility of intertidal cohesive sediments. Continental
Zone. Geological Society, London, Special Publications 139 pp. Shelf Research 28, 1854–1865.
Kristensen, E., 1993. Seasonal-variations in benthic community metabolism and Postma, H., 1961. Transport and accumulation of suspended matter in the Dutch
nitrogen dynamics in a shallow, organic-poor Danish Lagoon. Estuarine Coastal Wadden Sea. Netherlands Journal of Sea Research 1, 148–190.
and Shelf Science 36, 565–586. Postma, H., 1967. Sediment transport and sedimentation in the estuarine
Kromkamp, J., Barranguet, C., Peene, J., 1998. Determination of microphytobenthos environment. In: Lauff, G.H. (Ed.), Estuaries. American Association of Advanced
PSII quantum efficiency and photosynthetic activity by means of variable Scholarly Publication, Washington, DC, vol. 83, pp. 158–179.
chlorophyll fluorescence. Marine Ecology-Progress Series 162, 45–55. Reineck, H., Wunderlich, F., 1968. Classification and origin of flaser and lenticular
Kromkamp, J., Forester, R.M., 2006. Developments in microphytobenthos primary bedding. Sedimentology 11, 99–104.
productivity studies. In: Kromkamp, J.d.B., J.F.C., et al. (Eds.), Functioning of Reineck, H.E., 1972. Tidal flats. In: Rigby, J.K., Hamblin, W.K. (Eds.), Recognition
microphytobenthos in estuaries: Proceedings of the Colloquium, Amsterdam, of Ancient Sedimentary Environments. SEPM, pp. 146–159 Spec. Publ. 17.
21–23 August 2003, pp. 9–30. Koninklijke Nederlandse Akademie van Reineck, H.E., Singh, I.B., 1980. Depositional Sedimentary Environments. Springer-
Wetenschappen Verhandelingen, Afd. Verlag, Berlin, 439 pp.
Lanuru, M., Riethmuller, R., Van Bernem, C., Heymann, K., 2007. The effect of Risgaard-Petersen, N., 2003. Coupled nitrification-denitrification in autotrophic and
bedforms (crest and trough systems) on sediment erodibility on a back-barrier heterotrophic estuarine sediments: on the influence of benthic microalgae.
tidal flat of the East Frisian Wadden Sea, Germany. Estuarine Coastal and Shelf Limnology and Oceanography 48, 93–105.
Science 72, 603–614. Romano, C., Widdows, J., Brinsley, M.D., Staff, F.J., 2003. Impact of Enteromorpha
Larson, F., Lubarsky, H., Gerbersdorf, S.U., Paterson, D.M., 2009. Surface adhesion intestinalis mats on near-bed currents and sediment dynamics: flume studies.
measurements in aquatic biofilms using magnetic particle induction: MagPI. Marine Ecology Progress Series 256, 63–74.
Limnology and Oceanography: Methods 7, 490–497. Ruz, M., Allard, M., Michaud, Y., Héquette, A., 1998. Sedimentology and evolution
Lawson, S.E., 2008. Physical and biological controls on sediment and nutrient of subarctic tidal flats along a rapidly emerging coast, Eastern Hudson Bay,
fluxes in a temperate lagoon. Ph.D. Dissertation, University of Virginia, Canada. Journal of Coastal Research 14, 1242–1254.
Charlottesville, VA, 187 pp. Savage, D.C., Fletcher, M., 1985. Bacterial Adhesion. Plenum Publication
Lawson, S.E., Wiberg, P.L., McGlathery, K.J., Fugate, D.C., 2007. Wind-driven Corporation, New York.
sediment suspension controls light availability in a shallow coastal lagoon. Semeniuk, V., 1981. Long-term erosion of the tidal flats King Sound. North Western
Estuaries and Coasts 30, 102–112. Australia Marine Geology 43, 21–48.
Lee, H.J., Jo, H.J., Chu, Y.S., Bahk, K.S., 2004. Sediment transport on macrotidal Sfriso, A., Marcomini, A., 1997. Macrophyte production in a shallow coastal lagoon
flats in Garolim Bay, west Korea: significance of wind waves and asymmetry of 1. Coupling with chemico-physical parameters and nutrient concentrations in
tidal currents. Continental Shelf Research 24, 821–832. waters. Marine Environmental Research 44(4), 351–375.
Lundkvist, M., Gangelhof, U., Lunding, J., Flindt, M.R., 2007. Production and fate Shinn, E.A., Lloyd, R.M., Ginsburg, R.N., 1969. Anatomy of a modern carbonate
of extracellular polymeric substances produced by benthic diatoms and bacteria: tidal-flat, Andros Island, Bahamas. Journal of Sedimentary Petrology 39, 1202–1228.
a laboratory study. Estuarine Coastal and Shelf Science 75, 337–346. Smith, D.J., Underwood, G.J.C., 1998. Exopolymer production by intertidal epipelic
Macintyre, H.L., Geider, R.J., Miller, D.C., 1996. Microphytobenthos: the ecological diatoms. Limnology and Oceanography 43, 1578–1591.
role of the ‘secret garden’ of unvegetated, shallow-water marine habitats. 1. Stal, L.J., 2010. Microphytobenthos as a biogeomorphological force in intertidal
Distribution, abundance and primary production. Estuaries 19, 186–201. sediment stabilization. Ecological engineering 36, 236–245.
Madsen, K.N., Nilsson, P., Sundback, K., 1993. The influence of benthic microalgae Stal, L.J., De Brouwer, J.F.C., 2003. Biofilm formation by benthic diatoms and their
on the stability of a subtidal sediment. Journal of Experimental Marine Biology influence on the stabilization of intertidal mudflats. Berichte –
and Ecology 170, 159–177. Forschungszentrum TERRAMARE 12, 109–111.
McGlathery, K.J., Sundbeck, K., Anderson, I.C., 2007. Eutrophication in shallow Van Straaten, L.M.J.U., Kuenen, Ph.H., 1957. Tidal action as a cause of clay
coastal bays and lagoons: the role of the coastal plant filter. Marine Ecology accumulation. Journal of Sedimentary Petrology 28, 406–413.
Progress Series 348, 1–18. Sundback, K., Linares, F., Larson, F., Wulff, A., Engelsen, A., 2004. Benthic nitrogen
Mouget, J.L., Perkins, R., Consalvey, M., Lefebvre, S., 2008. Migration or fluxes along a depth gradient in a microtidal fjord: the role of denitrification and
photoacclimation to prevent high irradiance and UV-B damage in marine microphytobenthos. Limnology and Oceanography 49, 1095–1107.
microphytobenthic communities. Aquatic Microbial Ecology 52, 223–232. Sundback, K., Miles, A., 2000. Balance between denitrification and microalgal
Murphy, R.J., Tolhurst, T.J., 2009. Effects of experimental manipulation of algae and incorporation of nitrogen in microtidal sediments, NE Kattegat. Aquatic Microbial
fauna on the properties of intertidal soft sediments. Journal of Experimental Ecology 22, 291–300.
Marine Biology and Ecology 379, 77–84. Sundback, K., Miles, A., Goransson, E., 2000. Nitrogen fluxes, denitrification and
Mwamba, M.J., Torres, R., 2002. Rainfall effects on marsh sediment redistribution, the role of microphytobenthos in microtidal shallow-water sediments: an annual
North Inlet, South Carolina, USA. Marine Geology 189, 267–287. study. Marine Ecology-Progress Series 200, 59–76.
218 Ecogeomorphology of Tidal Flats

Sutherland, T.F., Amos, C.L., Grant, J., 1998a. The effect of buoyant biofilms on the Tolhurst, T.J., Watts, C.W., Vardy, S., Saunders, J.E., Consalvey, M.C., Paterson,
erodobility of sublittoral sediments of temperate mictrotidal estuary. Limnology D.M., 2008. The effects of simulated rain on the erosion threshold and
and Oceanography 43, 225–235. biogeochemical properties of intertidal sediments. Continental Shelf Research
Sutherland, T.F., Grant, J., Amos, C.L., 1998b. The effect of carbohydrate production 28, 1217–1230.
by the diatom Nitzschia curvilineata on the erodibility of sediment. Limnology Underwood, G.J.C., Kromkamp, J., 1999. Primary production by the pnhytoplankton
and Oceanography 43, 65–72. and microphytobenthos in estuaries. Advanced Ecology Research 29, 93–153.
Thompson, C.E.L., Amos, C.L., 2002. The impact of mobile disarticulated shells of Underwood, G.J.C., Paterson, D.M., 1993. Seasonal-changes in diatom biomass,
Cerastoderma edulis on the abrasion of a cohesive substrate. Estuaries 25, sediment stability and biogenic stabilization in the Severn Estuary. Journal of the
204–214. Marine Biological Association of the United Kingdom 73, 871–887.
Thompson, C.E.L., Amos, C.L., 2004. Effect of sand movement on a cohesive Underwood, G.J.C., Paterson, D.M., Parkes, R.J., 1995. The measurement of
substrate. Journal of Hydraulic Engineering-ASCE 130, 1123–1125. microbial carbohydrate exopolymers from intertidal sediments. Limnology and
Tobias, C., Giblin, A., Mcclelland, J., Tucker, J., Peterson, B., 2003. Sediment DIN Oceanography 40, 1243–1253.
fluxes and preferential recycling of benthic microalgal nitrogen in a shallow Underwood, G.J.C., Smith, D.J., 1998. Predicting epipelic diatom exopolymer
macrotidal estuary. Marine Ecology-Progress Series 257, 25–36. concentrations in intertidal sediments from sediment chlorophyll a. Microbial
Tolhurst, T.J., Defew, E.C., de Brouwer, J.F.C., Wolfstein, K., Stal, L.J., Paterson, Ecology 35, 116–125.
D.M., 2006a. Small-scale temporal and spatial variability in the erosion Verduin, J.J., Backhaus, J.O., 2000. Dynamics of plant-flow interactions for the
threshold and properties of cohesive intertidal sediments. Continental Shelf seagrass Amphibolis antarctica: field observations and model simulations.
Research 26, 351–362. Estuarine Coastal and Shelf Science 50, 185–204.
Tolhurst, T.J., Defew, E.C., Perkins, R.G., Sharples, A., Paterson, D.M., 2006b. The Vogel, S., 1994. Life in Moving Fluids. Pergamon, New York.
effects of tidally driven temporal variation on measuring intertidal cohesive Ward, L.G., Kemp, W.M., Boynton, W.R., 1984. The influence of waves and
sediment erosion threshold. Aquatic Ecology 40, 521–531. seagrass communities on suspended particulates in an estuarine embayment.
Tolhurst, T.J., Friend, P.L., Watts, C., Wakefield, R., Black, K.S., Paterson, D.M., Marine Geology 59, 85–103.
2006c. The effects of rain on the erosion threshold of intertidal cohesive Widdows, J., Lucas, J.S., Brinsley, M.D., Salkeld, P.N., Staff, F.J., 2002.
sediments. Aquatic Ecology 40, 533–541. Investigation of the effects of current velocity on mussel feeding and mussel bed
Tolhurst, T.J., Gust, G., Paterson, D.M., 2002. The influence of an extracellular stability using an annular flume. Helgoland Marine Research 56, 3–12.
polymeric substance (EPS) on cohesive sediment stability. In: Winterwerp, J.C., Willows, R.I., Widdows, J., Wood, R.G., 1998. Influence of an infaunal bivalve on
Kranenburg, C. (Eds.), Fine Sediment Dynamics in the Marine Environment. the erosion of an intertidal cohesive sediment: a flume and modeling study.
Elsevier Science Bv, Amsterdam, pp. 409–425. Limnology and Oceanography 43, 1332–1343.
Tolhurst, T.J., Jesus, B., Brotas, V., Paterson, D.M., 2003. Diatom migration and Yallop, M.L., Paterson, D.M., Wellsbury, P., 2000. Interrelationships between rates of
sediment armouring - an example from the Tagus Estuary, Portugal. microbial production, exopolymer production, microbial biomass, and sediment
Hydrobiologia 503, 183–193. stability in biofilms of intertidal sediments. Microbial Ecology 39, 116–127.

Biographical Sketch

Sergio Fagherazzi is an associate professor of surface processes and marine sciences at the Department of Earth
Sciences, Boston University. He obtained a Doctoral Degree in Hydrodynamics in 1999 at the Department of
Environmental, Maritime, Geotechnical, and Hydraulic Engineering, University of Padua, Italy. His primary study
interest concerns the evolution of coastal environments. His research activities cover all aspects of coastal mor-
phodynamics, including wetlands erosion, hurricanes impact on sandy beaches and mitigation, tsunami effects
on shorelines, and long-term evolution caused by sea-level rise and climate change. He is a member of the
editorial board for Earth Surface Processes and Landforms.

Duncan M FitzGerald is a professor in the Department of Earth Sciences at Boston University, with an expertise in
coastal processes, stratigraphy, and morphodynamics. He received his Masters in Geological Oceanography at
Texas A & M and his PhD in Geology at the University of South Carolina. He is a marine geologist who studies
sediment exchange; hydrodynamics; and coastal evolution of marshes, estuaries, river deltas, barrier islands, and
tidal inlets. His three major research themes are presently focused on coastal response to accelerating sea-level rise,
impact of major storms along the Louisiana coast, and climatic and oceanographic controls on strand plain
development in Brazil. His textbook on coastal geology is used both nationally and internationally. He is a Fellow
of the Geological Society of America and has received numerous teaching awards at Boston University.
 Ecogeomorphology of Tidal Flats 219

Robinson W Fulweiler is assistant professor of Earth Sciences at Boston University. She is a biogeochemist and
coastal ecosystem ecologist. Her research included coastal watershed mass balances of major biogenic elements in
New England (C, N, P, and Si), the biogeochemistry of nitrogen in coastal marine ecosystems, especially sedi-
ments, and wetland ecology in coastal Louisiana. Her recent focus has been on how climate change may influence
nitrogen fixation and denitrification in estuarine and shelf systems and anthropogenic impacts on the coastal
silica cycle.

Zoe Hughes is a senior postdoctoral research associate in the Department of Earth Sciences, Boston University. She
received her PhD from the University of Southampton working at the National Oceanography Centre, South-
ampton. Hughes uses a combination of modeling and field investigations to examine the impacts of waves, tidal
currents and sea-level rise on coastal sediment transport, and geomorphology. Her recent work involves col-
laborations with ecologists, undertaking interdisciplinary studies of feedbacks between hydrodynamics, sediment
movement, and both flora and fauna. She is involved in a range of studies within coastal barrier and saltmarsh
systems along the East and Gulf coasts of USA.

Patricia L Wiberg is Professor and Chair of Environmental Sciences at the University of Virginia, where she has
taught since 1990. Wiberg received her PhD in oceanography from the University of Washington in 1987. Her
research focuses on the mechanics of sediment erosion, transport, and deposition, as well as associated evolution
of sediment bed properties and morphology. Her research topics include mechanisms and rates of saltmarsh
erosion and accretion, sediment dynamics on tidal flats, effects of seagrass on turbidity in coastal lagoons, and the
impacts of sea-level rise and climate change on the evolution of coastal landscapes. She currently serves as the
chair of the Marine Working Group of the Community Surface Dynamics Modeling System.

Karen J McGlathery is Professor of Environmental Sciences at the University of Virginia. Her research focuses on
the effects of long-term change, including climate, sea-level rise, land-use and species invasions in coastal marine
ecosystems. She received her BS from Connecticut College, her PhD from Cornell University, and she did her
postdoctoral work at the University of Copenhagen and the National Environmental Research Institute in Den-
mark. She is the Program Director of the Virginia Coast Reserve Long-Term Ecological Research site, which is one
of 25 sites in the nation funded by the National Science Foundation to study long-term environmental changes in
marine and terrestrial ecosystems. Karen serves on the editorial board of the journal Ecosystems.
220 Ecogeomorphology of Tidal Flats

James T Morris is the Director of the Belle Baruch Institute for Marine and Coastal Sciences, Professor of Biological
Sciences, Distinguished Professor of Marine Studies at the University of South Carolina, and is an American
Association for the Advancement of Science Fellow. He served as a Program Officer at the National Science
Foundation in the Division of Environmental Biology from 2003 to 2005 and was a visiting professor at Aarhus
University, Denmark in 1990. His academic background includes degrees in environmental sciences (BA, Uni-
versity of Virginia), biology (MA, Yale), and forestry and environmental studies (PhD, Yale). He held a post-
doctoral fellowship at the Marine Biological Laboratory, Woods Hole before taking a faculty position at the
University of South Carolina in 1981. He currently serves on the Wetland Carbon Modeling working group at the
National Center for Ecological Synthesis, the Special Science, Engineering, and Technology panel that is making
recommendations on the restoration of the Mississippi River Delta, and at National Research Council Gulf Oil
Spill committee.

Trevor J Tolhurst is a lecturer in coastal processes at the School of Environmental Sciences, University of East
Anglia, UK. He obtained a Doctoral Degree in Marine Biology in 2000 at the Gatty Marine Laboratory, University
of St. Andrews, Scotland. His primary interests concern the sedimentology and ecology of intertidal sedimentary
habitats. His research is multidisciplinary and field based, investigating the nature of direct and indirect inter-
actions between sediments and biota, particularly how biota affects the erosion of sediments. He has helped
develop various devices for measuring erosion, as well as remote sensing techniques for investigating microbial
biofilms.

Linda A Deegan is a Senior Scientist at The Ecosystems Center, Marine Biological Laboratory at Woods Hole. She
obtained a doctorate in Marine Science in 1985 from Louisiana State University. Her primary interest is in how
animals through their behavior and interactions shape ecosystem processes, including geomorphic dynamics and
nutrient cycling. She works in many ecosystems, from temperate coastal wetlands to arctic and tropical streams.
She has been a member of the editorial boards of Ecological Applications and Estuaries and Coasts. She currently
serves on the Board of The Nature Conservancy and is a member of the Science and Engineering Special Team
advising nongovernmental organizations, especially Environmental Defense Fund, National Audubon Society,
and National Wildlife Federation, on issues underlying the restoration of the Mississippi River delta in the Gulf of
Mexico.

David S Johnson is a research associate in the Ecosystems Center at the Marine Biological Laboratory in Woods
Hole, MA, USA. He is a marine ecologist who is interested in species interactions and ecosystem processes. David
is particularly fond of invertebrates. He received his PhD from Louisiana State University, Baton Rouge, LA, USA in
2008.
 12.14 Valley Plugs, Land Use, and Phytogeomorphic Response
AR Pierce, Nicholls State University, Thibodaux, LA, USA
SL King, School of Renewable Natural Resources, Baton Rouge, LA, USA
r 2013 Elsevier Inc. All rights reserved.

12.14.1 Introduction 221

12.14.1.1 Fluvial Processes 221
12.14.1.2 Channelization 223
12.14.2 Valley-Plug Formation 224
12.14.2.1 Geology 224
12.14.2.2 Land-Use Practices 226
12.14.3 Fluvial-Geomorphic Responses 227
12.14.3.1 Hydrologic Regimes 227
12.14.3.2 Sedimentation 228
12.14.4 Vegetative Responses 229
12.14.4.1 Germination 229
12.14.4.2 Species Composition and Structure 229
12.14.5 Restoration 230
12.14.6 Summary 231
References 233

Glossary Ephemeral Generally lasting only a brief time.

Alluvium Fine-grained sediments consisting of silt, clay, Fluvial Hydrologic action produced by flowing water.
and sand deposited by flowing water on floodplains, Stream power Rate of energy expenditure of flowing water
riverbeds, and in estuaries. against channel banks and bed.
Anthropogenic Caused by or relating to humans. Valley plug Area where the stream channel becomes filled
Channelization Changing of stream channel with sediment (sometimes mixed with debris), which can
morphology, including straightening, widening, deepening, exceed channel banks and forces stream flow and sediment
and shortening of stream channel and increasing stream onto floodplain.
gradient.

Abstract

Anthropogenic alteration of fluvial systems can disrupt functional processes that provide valuable ecosystem services.
Channelization alters fluvial parameters and the connectivity of river channels to their floodplains which is critical for
productivity, nutrient cycling, flood control, and biodiversity. The effects of channelization can be exacerbated by local
geology and land-use activities, resulting in dramatic geomorphic readjustments including the formation of valley plugs.
Considerable variation in the response of abiotic processes, including surface hydrology, subsurface hydrology, and sedi-
mentation dynamics, to channelization and the formation of valley plugs. Altered abiotic processes associated with these
geomorphic features and readjustments influence biotic processes including species composition, abundance, and suc-
cessional processes. Considerable interest exists for restoring altered fluvial systems and their floodplains because of their
social and ecological importance. Understanding abiotic and biotic responses of channelization and valley-plug formation
within the context of the watershed is essential to successful restoration. This chapter focuses on the primary causes of
valley-plug formation, resulting fluvial-geomorphic responses, vegetation responses, and restoration and research needs for
these systems.

12.14.1 Introduction

12.14.1.1 Fluvial Processes

Pierce, A.R., King, S.L., 2013. Valley plugs, land use, and phytogeomorphic
Floodplains provide numerous valuable ecosystem services,
response. In: Shroder, J. (Editor in chief), Butler, D.R., Hupp, C.R. (Eds.),
Treatise on Geomorphology. Academic Press, San Diego, CA, vol. 12, such as flood control, sediment and nutrient retention, timber
Ecogeomorphology, pp. 221–235. production, and wildlife habitat. In the light of the social and

Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00330-4 221

222 Valley Plugs, Land Use, and Phytogeomorphic Response

ecological importance of these critical functions, floodplains hydroperiod and floodplain communities. Fluvial systems
have received considerable attention in recent years in regard transport sediment downstream, from erosional areas, in the
to conservation and restoration efforts. In floodplain systems, form of suspended sediment within the water column (fine-
the primary driving process responsible for the existence, grain particles) and bedload (coarse-grain particles) moving in
productivity, and interactions of the major biota is periodic a wave-like motion across the channel bed. Movement of the
overbank flooding, also known as the flood pulse (Junk et al., sediment through the system is not continuous but is a result
1989). The fertility of floodplain soils depends on nutrient of discrete climatic events. The largest portion of the total
inputs from the main channel and the quality of deposited sediment load is usually carried by one or two high flows per
sediments from overbank flooding (Wharton et al., 1982; Junk year (Knighton, 1998). A dynamic equilibrium exists between
et al., 1989; Stanturf and Schoenholtz, 1998). Contemporary sediment load and stream-flow velocity, where increases in
sedimentation dynamics have been the focus of much of this sediment load create corresponding decreases in flow velocity.
attention because of their role in river basin management and The dissipation of energy results in within-channel de-
their ecological significance in determining the composition positions of alluvium that is reworked by fluvial-geomorphic
and structure of vegetation communities and overall bio- processes into stream meanders. Stream energy is therefore
diversity of floodplain systems (Hodges, 1997; Walling and decreased by lateral meandering that widens the floodplain
He, 1998; Hupp, 2000; Ward and Tockner, 2001). (Knighton, 1998). Streams continually adjust their slopes by
Sediments within fluvial systems originate from the con- meandering until a steady state is reached within a dynamic
tinuous erosion of the landscape through time. Sheet-wash and equilibrium among sediment load, water velocity, and water
gully erosion can transport sands, silts, and clays into streams volume.
especially during precipitation events. Gullies are unstable Floodplain development primarily occurs through within-
landforms that are constantly changing within the developing channel (point-bar) and overbank depositional (vertical ac-
drainage network (Figure 1). Gully erosion typically occurs in cretion) processes (Knighton, 1998; Walling and He, 1998).
steep ephemeral stream valleys that are actively eroding and Point-bar deposition occurs on the convex banks of river
over time create relatively deep channels on valley sides and meanders by lateral accretion (Figure 2). This process is cou-
valley floors (Schumm et al., 1984). Gullies have been shown pled with erosion of the opposite concave banks of the me-
to develop rapidly as a result of agricultural practices by in- anders resulting in the bulk of sediment being stored within
corporating adjacent rills into larger and deeper gullies (Bloom, the floodplain (Leopold and Wolman, 1957; Hupp, 2000).
1978). Land-use activities in the uplands and basin areas have Deposition on the floodplain (vertical accretion) occurs dur-
been shown to have a major influence on the quantities ing flood events where water flows are forced outward into the
of sediment entering fluvial systems. Erosion from forested floodplain causing deposition and forming the primary
uplands has been estimated to be relatively small, 2.5 cm ha–1 floodplain soils (Figure 3).
per 16 000 years (Soil Conservation Service, 1977) compared During overbank flows or flooding, sediment can be de-
to the erosion from croplands which can be more than posited on the floodplain by different mechanisms based on
87 000 kg ha–1 yr–1 as reported for the Obion–Forked Deer the quantity and texture of the sediment resulting in general-
River in Tennessee (Soil Conservation Service, 1977). ized spatial patterns of deposition (Middelkoop and Van Der
Hydrological energy that creates areas of erosion and de- Perk, 1998; Nicholas and Walling, 1998; Walling and He,
position of sediment is responsible for the morphology of 1998) that create broad, flat floodplains. Greatest deposition
floodplains and establishes complex relationships between the rates typically occur near the channel as bedload material is

Figure 1 Photograph of a massive gully in an agricultural setting in western Tennessee that contributes sediment into a fluvial system (upper
right corner). Photo by A.R. Pierce.
 Valley Plugs, Land Use, and Phytogeomorphic Response 223

due to the roughness of the floodplain causing lower de-

position rates of fine sediments in flats and backswamps. In
these areas water may remain stagnant for relatively long
periods which allow fine particles to settle out of the water
column (Happ et al., 1940; Hodges, 1997; Hupp, 2000).
High-velocity flows through sheet flow and scour channels can
also transport coarse bedload deep into the floodplain (Leo-
pold and Wolman, 1957; Pierce and King, 2008). Average
floodplain sedimentation rates can vary greatly but typical
ranges estimate between 0.3 and 0.6 m every 200–400 years
(Leopold and Wolman, 1957; Hupp and Bazemore, 1993;
Kleiss, 1996; Heimann and Roell, 2000).
Local factors, such as flood frequency and magnitude,
vegetation cover, stream gradient, and floodplain morphology
can create variations in the quantity and quality of sediment
that is deposited and stored on floodplains (Knighton, 1998;
Figure 2 Photograph illustrating point-bar deposition on the convex Hupp, 2000). Basin level factors, including basin size, position
bank of a river meander in western Tennessee. Photo by A.R. Pierce. within the system, geology, climatic conditions, and land use
can also influence overbank sedimentation dynamics (Steiger
et al., 2003). The dynamic nature of alluvial systems enables
sediment deposited on floodplains to be reworked over time,
with storage times being extremely variable, which can facili-
tate future river management problems (Walling and He,
1998). Re-activation of stored sediment and continual geo-
morphic adjustments in conjunction with local and basin
level factors result in spatial and temporal variation in de-
positional areas and the interrelated processes of hydrological
regimes and vegetation succession (Brinson, 1990; Sharitz and
Mitsch, 1993; Hodges, 1997).

(a)
12.14.1.2 Channelization
Channelization of fluvial systems includes widening and
deepening the stream channel, which increases the channel
capacity, shortening the stream channel length, and increasing
the stream gradient (Figure 4). These factors combine to
typically move greater volumes of water through the system at
a much more rapid rate compared to pre-channelization
conditions. As a result of increased channel capacities and
increased transport efficiency, channelization of streams cau-
ses the channels to be hydrologically disconnected from the
adjacent floodplain and alters functional processes of fluvial
systems (Kroes and Hupp, 2010). There has been no official
assessment of the number of streams that have been chan-
nelized, but this type of alteration has been recognized as
(b) being more extensive than damming (National Research
Figure 3 (a) Photograph of an inundated floodplain during an overbank Council, 1992), which has occurred on most of the world’s
flooding event that contributes to the development of floodplain soils. largest rivers (Nilsson et al., 2005). Throughout the south-
(b) Photograph illustrating the productivity of floodplain systems during eastern United States, stream channelization has been used
low-flow periods occurring in the summer. Photos by A.R. Pierce. extensively to reduce flooding and facilitate agriculture pro-
duction on floodplain areas (Simon and Hupp, 1992; Hupp
transported by traction from the main channel to the flood- and Bazemore, 1993; Shankman, 1993).
plain. As overbank flows move out of the channel, flow vel- Channelization can typically reduce flooding in upstream
ocities are reduced facilitating high deposition rates that create reaches of a system; meanwhile, lower reaches usually ex-
the natural levee of alluvial systems (Happ et al., 1940; Pierce perience an increase in peak flood levels and have a higher
and King, 2008). These deposits generally consist of heavier frequency of flooding (Shankman and Pugh, 1992). Chan-
and coarser particles, but layers of coarse and fine particles can nelization, along with dredging activities used in maintenance
occur in areas where drainage is poor (Allen, 1965). As flows of channelized systems, has also been found to influence
penetrate further into the floodplain, additional velocity is lost water-table levels within the adjacent floodplain (Tucci and
224 Valley Plugs, Land Use, and Phytogeomorphic Response

(a)

(b)

Figure 4 Example of a channelized stream in western Tennessee during high-flow (a) and low-flow (b) periods. Photos by A.R. Pierce.

Hileman, 1992). Stream channelization can also produce 12.14.2 Valley-Plug Formation
conditions that initiate continued degradation of the stream
channel, including headcutting and channel erosion that can Valley plugs are areas where the entire channel becomes filled
produce extensive bank failures (Robbins and Simon, 1983; with sediment, which can even exceed stream bank levels,
Simon and Hupp, 1987; Simon, 1994). resulting in stream flows and bedload material to be forced
The increased stream power that characterizes channelized out into the floodplain (Happ, 1975). Valley plugs usually
streams also facilitates increased rates of sediment transport form in the lower reaches of alluvial systems, where the stream
within these altered systems. Typically, channelized systems gradient is reduced or debris jams form, causing reduced flow
will accumulate large amounts of sediment in the lower velocities and facilitating deposition of sediment (Happ et al.,
reaches of the system because of decreased stream gradients 1940). Valley plugs force stream flow and sediment to spread
and channel obstructions that occur in these areas, which re- across floodplains and typically create anastomosing streams
duce stream velocities and initiate sediment deposition throughout the floodplain that have the potential to distribute
(Schumm et al., 1984; Simon and Hupp, 1987). Greater de- great quantities of sediment into floodplain areas (Figure 5).
position rates in these lower reaches reduce the channel cap- Geology of the region, land-use practices, and anthropogenic
acity and cause a widening of the stream channel; this is alterations to fluvial systems can contribute to the formation
recognized as a critical stage of geomorphic recovery of of valley plugs.
channelized streams (Schumm et al., 1984; Simon and Hupp,
1987). However, in some cases, where stream channelization
combines with unique geological conditions and land-use
12.14.2.1 Geology
practices, severe accumulation occurs that completely blocks
the channels with sediment, forming valley plugs (Happ et al., The geology of the region encompassing a drainage basin can
1940; Fryirs and Brierley, 1998; Brierley and Fryirs, 1999; characterize the quantity and quality of sediment supplied to
Pierce and King, 2007a, 2008). fluvial systems. Areas with highly erodible soils have a greater
 Valley Plugs, Land Use, and Phytogeomorphic Response 225

Water flow
Flow
Ponded timber

Channelized stream

Valley plug Plug expansion

Anastomosing
Anastomosing channel
channel
(a) Sediment accumulation

(b)

(c)

Figure 5 (a) Illustration of a channelized fluvial system affected by valley-plug formation and the associated impacts. (b) Aerial view of a
large sand splay created as a result of valley-plug formation. (c) Excessive sedimentation of coarse sand bury trees in floodplains adjacent to
valley plugs. Photos by A.R. Pierce.
226 Valley Plugs, Land Use, and Phytogeomorphic Response

potential for the formation of valley plugs due to the increased settlement, were estimated to range from 0.02 to 0.09 cm yr–1
amount of sediment that can enter a system and more rapid with post-settlement rates increasing to approximately 3 cm yr–1
geomorphic responses within the system. For example, west- (Wolfe and Diehl, 1993). Channel-flood capacity decreased in
ern Tennessee, western Kentucky, and northern Mississippi, the region as increased erosion rates resulted in sediment de-
USA, encompass both the Gulf Coastal Plain and Lower position in stream channels and floodplains. Streams were said
Mississippi Alluvial Valley. The parent material of the region to be ‘stifled with sediment and debris’ that caused frequent and
consists of unconsolidated coarse sand that was deposited prolonged flooding in lower basin stretches (Morgan and
during the Quaternary period (Saucier, 1994). There is little McCrory, 1910). Stream-channel alterations were then proposed
bedrock in the region to control the base level of streams throughout the region to alleviate the flooding problems and
flowing into the Mississippi River basin. Therefore, streams are facilitate more extensive use of floodplain areas for agricultural
free to adjust their profiles in response to disturbances such as purposes (Hidinger and Morgan, 1912).
dredging, channelization, or sediment supply (Simon, 1994). In addition to these US streams, some alluvial and semi-
The unconsolidated alluvial sands of western Tennessee, alluvial river systems in southeastern Australia have also
western Kentucky, and northern Mississippi are mostly cov- undergone channel metamorphosis following European settle-
ered by a thin layer of loess deposits of silt and clay particles. ment and associated land clearing (Brierley and Murn, 1997;
In western Tennessee, these deposits range in depth from 1 to Brierley and Fryirs, 1999; Brooks and Brierley, 2004). By the
30 m and are dissected by most of the tributary systems within 1890s, much of the vegetation on valley flats and hillslopes were
the region (Saucier, 1994; Simon, 1994). A few tributaries in removed, wetlands were drained, and the landscape was dom-
the area originate on block clay deposits known as Porters inated by agricultural fields and pasture (Brierley and Fryirs,
Creek Clay of the Midway Group, deposited during the Ter- 1999). Changes in land-use practices since European settlement
tiary period (Miller et al., 1966). However, the main fluvial in southeastern Australia have resulted in dramatic channel re-
systems in the region including the Obion River, the Forked adjustments. For example, the Cann River has experienced a
Deer River, and the Hatchie River cut through the alluvial 360% increase in channel depth, a 240% increase in channel
sands deposited during the Quaternary age (Saucier, 1994). slope, a 700% increase in channel capacity and an 860-fold
Both the loess and alluvial sand deposits are highly erod- increase in annual sediment load (Brooks et al., 2003; Brooks
ible. Most of the erosion in western Tennessee, western Ken- and Brierley, 2004). Although many upper reaches of fluvial
tucky, and northern Mississippi is thought to occur in the systems in southeastern Australia are experiencing severe chan-
loess-capped uplands. This erosion consists of the loess cap nel incision, downstream reaches are accumulating extensive
and the alluvial sands beneath, producing erosion rates that volumes of sediment and debris creating valley fills (Brierley
are consistently acknowledged as among the highest in the and Fryirs, 1999) similar to the valley plugs described in
United States (Langdale et al., 1985). The past 150 years of Northern Mississippi and western Tennessee (Happ et al., 1940;
erosion in the uplands of western Tennessee exceeds the ero- Pierce and King, 2007a, 2008; Bennett and Rhoton, 2009).
sion that occurred in the last several thousand or tens of Western Tennessee (Figure 6), where valley plugs and
thousands of years (Saucier, 1994). These geological con- channelization have been extensively investigated, has ex-
ditions set the stage for the numerous valley-plug formations perienced widespread hydrologic and land-use changes. Most
that have occurred in western Tennessee, western Kentucky, streams in western Tennessee were channelized by 1926 and
and northern Mississippi (Happ et al., 1940; Pierce and King, included 132 km of stream alterations (Speer et al., 1965;
2007a, 2008; Bennett and Rhoton, 2009) including over 30 Simon and Robbins, 1987). Channelized systems in western
valley plugs that have been identified in the Hatchie River Tennessee experience many of the disturbances associated
watershed (Diehl, 2000). with channelized streams, including sediment aggradation
and debris accumulation. To maintain channelized features,
persistent maintenance of streams is needed to clear debris
and sediments. Maintenance dredging and clearing of debris
12.14.2.2 Land-Use Practices
were widespread in western Tennessee from the 1930s through
Land-use practices, particularly clearing of upland, floodplain, the 1950s (Simon, 1994). By 1970, the sediment accumu-
and riparian vegetation for timber production or agricultural lation within the tributary systems was so extensive that the US
purposes can also contribute to the formation of valley plugs. Department of Agriculture (1970) initiated a channelization
Changes in vegetation type and cover have a direct impact on project called the West Tennessee Tributaries Project (WTTP).
runoff and erosion rates. For example, the western Tennessee Only 35% of the project was actually completed, but the
region was rapidly colonized by European settlers in the early project resulted in 128 km of channel alterations (Robbins
1800s and the region experienced dramatic changes in land use. and Simon, 1983). The mainstems of the Obion–Forked Deer
Upland areas were quickly cleared for the valuable timber and and Wolf Rivers (i.e., Obion–Forked Deer River) were chan-
replaced with agriculture fields to take advantage of the fertile nelized, whereas alterations to the Hatchie River were re-
soils (Wilder, 1998). Vegetation cover changed from forest to stricted to its tributary system, as 33 of its 36 major tributaries
being dominated by corn, cotton, and tobacco fields. Before were channelized (USDA, 1986; Simon, 1994).
deforestation of the region occurred, the rivers were described as Overall, channelization projects in western Tennessee re-
flowing at ‘consistently good depths’ (Ashley, 1910). Meanwhile, sulted in streams being shortened by 44%, lowered by 170%,
clearing of the upland areas and upper drainage basins resulted and steepened by 600% (Simon and Hupp, 1992). Bed levels
in erosion and gullying of the loess and sandy soils (Simon, in altered channel reaches were lowered by as much as
1994). Sediment deposition rates in the region, previous to 5 m (Simon and Hupp, 1987; Simon, 1994). Transition slopes
 Valley Plugs, Land Use, and Phytogeomorphic Response 227

Ob io n
h
ort Fo r k

N
Fork Obion
Middle
ion
Ob

y
So

s s ee
od
ut
h

Sa
No Fork Fork Fo
rk Obion

nne
rt h ed

Big
Mi
D

Te
dd
i
ipp

ee
le
Fo

r
iss
r
rk Dee

ss
So o r k ed

F
Mi
uth
Fo
rk F Beech
ork
Hatchi ed
e

De
e r
L o osahat chie

Tennessee
Wolf

Figure 6 Location of major watersheds in western Tennessee affected by channelization and valley-plug formation.

were constructed to offset differences in bed elevations at the systems (Schumm et al., 1984; Simon and Hupp, 1992;
junction of the upper channelized reaches and lower Brierley and Murn, 1997; Brooks and Brierley, 2004; Hupp
unchannelized reaches (Robbins and Simon, 1983). These et al., 2009). Fluvial-geomorphic systems maintain a dynamic
transition areas had steeper slopes than both the channelized equilibrium (Hack, 1960) through adjustments in sediment
and unchannelized reaches resulting in headcutting and deg- load, discharge rates, and channel geometry (Osterkamp and
radation of upstream reaches. This erosion moved upstream at Hedman, 1977). Human alterations (channelization, levees,
a rate of 2.6 km yr–1 on the South Fork Forked Deer River and land use, and others) that affect fluvial attributes may result in
resulted in approximately 2.6 m of incision from 1966 to disequilibrium and dramatic shifts in fluvial-geomorphic
1967 (Simon, 1994). Downstream areas accumulated vast parameters (Hupp et al., 2009). Responses to channelization
amounts of sediment resulting from the erosion occurring in are induced by increases in channel gradient that affect dis-
the upstream reaches. Deposition rates on downstream charge and erosion rates upstream, whereas downstream de-
reaches of the Obion River ranged from 0.03 to 0.12 m yr–1, position (valley plugs) may occur from reduced gradients and
whereas the South Fork Forked Deer River filled in with 2.2 m high sediment loads. The fluvial-geomorphic adjustments re-
of sediment over a 12-year period (Simon, 1994). sult in reduced topographic relief and relatively high flood-
Channelization projects throughout western Tennessee fa- plains or the active floodplain may be restricted within incised
cilitated greater agriculture use of floodplain areas. By 1971, banks. The homogenized nature of floodplains or reduced
forest habitat in floodplains along channelized streams was floodplain area may affect nutrient accumulation and cycling
reduced by 60% (Barstow, 1971). Systems not directly im- process that impact the productivity and diversity of flood-
pacted by channelization projects experienced losses of plains (Hupp et al., 2009).
floodplain forest due to clearing of the floodplain and con- Channelization has been shown to disrupt system func-
struction of drainage ditches by individual private landowners tions in the following ways: altered surface and subsurface
(Barstow, 1971). During this period of drastic hydrologic al- hydrology (Shankman and Pugh, 1992; Tucci and Hileman,
teration, floodplain forests in western Tennessee were reduced 1992), altered sedimentation rates (Happ et al., 1940; Hupp
from 404 000 ha in 1940 to 291 000 ha in 1970 (Turner et al., and Bazemore, 1993), reduced lateral channel migration re-
1981). Land-use changes in western Tennessee during the sponsible for creating sloughs and oxbow lakes (Shankman,
1800s to the mid-1900s were mainly driven by agricultural 1993), loss of aquatic habitat (Hohensinner et al., 2004),
needs or desires, but were realized through extensive stream reduced growth and increased mortality of floodplain tree
channelization. species (USDA, 1986), loss of plant-species diversity (Miller,
1990), changes in plant-species composition (Oswalt and
King, 2005; Pierce, 2005), and negative effects on fish and
wildlife communities (Heitmeyer and Fredrickson, 1981;
12.14.3 Fluvial-Geomorphic Responses
Risotto and Turner, 1985; Hunter et al., 1993; Hoover and
Killgore, 1997). Many of these effects are magnified by
12.14.3.1 Hydrologic Regimes
the formation of valley plugs within channelized systems
The fluvial-geomorphic effects of channelization can disrupt (Happ et al., 1940; Diehl, 2000; Pierce and King, 2007a,
functional processes and ecosystem services of floodplain 2008).
228 Valley Plugs, Land Use, and Phytogeomorphic Response

Valley plugs exacerbate the higher peak flood levels and unchannelized sites. These sites may represent older valley-
higher frequency of flooding that is typically associated with plug sites, where anastomosing channels and floodplain sur-
the lower reaches of a channelized system (Shankman and face elevations have stabilized, creating dryer sites that are
Pugh, 1992). The valley-plug channel obstruction of sand and only inundated during major flood events.
woody debris forces stream flow on to the floodplain, inun- Channelization and dredging activities have also been
dating large portions of the floodplain (Figure 7). In com- shown to influence groundwater levels in the floodplain.
bination with increased sediment accumulation in these Typically, these activities result in a lowering of the ground-
reaches, the drainage capacity of the inundated floodplains is water in the floodplain due to entrenchment of the river and
commonly reduced, resulting in swamping of floodplain sur- the connectivity of stream flow to groundwater levels in the
faces for extended periods of time (Happ et al., 1940; Miller, adjacent floodplain (Tucci and Hileman, 1992); as the river
1990; Diehl, 2000; Oswalt and King, 2005). However, the age bed lowers so does the groundwater level. At valley-plug sites,
of the valley plug and stage of development along with specific it has been hypothesized that floodplain groundwater levels
site conditions such as position in the basin, basin size, stream may be higher than unchannelized sites due to increased rates
gradient, and floodplain morphology can create variability in of channel filling (Pierce, 2005). As predicted, floodplains
the hydrologic response. Pierce (2005) found that several associated with channelized streams without valley plugs in
valley-plug sites within the Hatchie River watershed actually the Hatchie River watershed had lower groundwater levels
had lower flood depths and duration of flooding compared to than unchannelized sites (Pierce, 2005). By contrast,
groundwater levels were also lower at valley-plug sites com-
pared to unchannelized sites. This may be due to the higher
elevation of floodplain surfaces due to increased deposition
rates and the flashiness of the channelized systems. Little
study of groundwater levels in floodplains affected by valley
plugs has been done, and as with surface hydrology there may
be considerable variation depending on the age of the valley
plug, its stage of development, and the specific site con-
ditions. Burt et al. (2002) showed that during low stream-
flow periods, there can be an influx of groundwater to the
stream. Valley-plug sites investigated by Pierce (2005) typi-
cally had little or no flow during the summer months and
therefore may have been discharging groundwater from the
floodplain to the channel. Other valley-plug sites that have
created open water and marsh communities, such as those
investigated by Miller (1990) and Oswalt and King (2005),
(a)
may experience very different sub-surface hydrological
regimes; however, they have not been investigated in this
regard.

12.14.3.2 Sedimentation
Channels filling with sediment and debris that form valley
plugs also force bedload sediment out of the channel and
across the floodplain (Figure 7). Channel filling, sand splays,
and vertical accretion occur in greater quantities in floodplains
associated with valley plugs compared to unaltered systems
(Happ et al., 1940; Brierley and Fryirs, 1999; Pierce and King,
2007a, 2008). The type of sediment deposited at valley-plug
sites is also different, generally consisting mostly of sand while
nonvalley plugs are composed more of silt and clay (Pierce
and King, 2008). Although channelized streams tend to carry
heavier sediment loads (Simon, 1994; Hopkinson and Val-
lino, 1995), the reduced lateral connectivity between the
channel and floodplain in upper reaches can prevent increased
deposition rates (Hupp and Bazemore, 1993). The formation
(b) of valley plugs, however, reconnects the channelized system to
Figure 7 (a) Example of a debris jam that contributed to the the floodplain and facilitates the episodic but substantial de-
formation of a valley plug in the tributary system of the Hatchie River posits of coarse sediment within adjacent floodplains. In the
in western Tennessee. (b) Photograph of coarse sand deposited in Hatchie River watershed, deposition rates at valley plugs ran-
the floodplain as a result of a downstream valley plug in western ged from 0 to 42 cm yr–1, with a mean of 4.67 cm yr–1 (Pierce
Tennessee. and King, 2008).
 Valley Plugs, Land Use, and Phytogeomorphic Response 229

Valley plugs can alter the spatial dynamics of floodplain growth, and survival of floodplain tree species (Harms et al.,
sedimentation; however, responses vary depending on site 1980; Huffman, 1980; Streng et al., 1989; Johnson, 1994,
conditions (Pierce and King, 2008). Classical fluvial geo- 2000; Jones et al., 1994). Although the effects of sedimen-
morphology theory suggests that the quantity and texture of tation on vegetation have received limited study, burial of
sediment result in generalizable spatial patterns of deposition freshwater lowland plant species by sediment can reduce
on the floodplain (Middelkoop and Asselman, 1998; Mid- shoot density (van der Valk et al., 1983), and, in riparian
delkoop and Van Der Perk, 1998; Nicholas and Walling, 1998; systems, tree recruitment may be controlled by stream-flow
Walling and He, 1998). Coarse particles are deposited at the pulses that erode or bury newly germinated individuals
greatest rates near the channel as bedload material is trans- (Johnson, 2000).
ported by traction from the stream channel into the flood-
plain. Floodplain surfaces farther away from the main channel
may retain stagnant water, allowing for the deposition of fine 12.14.4.1 Germination
sediments. Although there can be variability in spatial patterns
Past germination studies on floodplain tree species have fo-
of deposition as a result of local factors including frequency
cused mainly on flooding regimes (Hosner, 1957; Briscoe,
and magnitude of flooding, sediment load, water velocity,
1961; Larsen, 1963; Guo et al., 2002). These studies have
floodplain morphology, and vegetation cover (Knighton,
demonstrated that flood duration can influence germination
1998; Hupp, 2000), unaltered floodplains demonstrate a high
rates of some floodplain tree species, but the response varies
degree of spatial correlation among deposition rates within
by species. Sedimentation rates have also been shown to affect
the floodplain with a direction of spatial dependence parallel
germination and establishment of floodplain tree species, but
with stream flow corresponding to the development of natural
there has been little investigation of these relationships (Walls
levees (Pierce and King, 2008). Deposition rates at valley-plug
et al., 2005; Pierce and King, 2007b).
sites have also been shown to have a high degree of spatial
Walls et al. (2005) found that flooding and sedimentation
dependence but the direction of spatial dependence can be
can have an additive effect to delay germination and germin-
sensitive to the channel blockage (valley plug), changing the
ation rates, but tolerance to these synergistic processes varies
direction of spatial dependence from parallel relative to
by species. Pierce and King (2007b) also noted that both
stream flow to perpendicular to stream flow and extending out
flooding and sedimentation can influence germination and
great lengths onto the floodplain (Pierce and King, 2008).
growth of some floodplain tree species, but flooding seemed
The high rates of deposition and spatial patterns of
to be the overriding factor. However, greater sedimentation
deposition at valley-plug sites may be short-lived at specific
rates may reduce seedling’s competitive ability and tolerance
locations. High deposition rates, which influence surface and
for flooding. As all studies have suggested, the response of
sub-surface hydrology, can dramatically alter floodplain
each species to these abiotic processes depends on their life-
morphology over short time periods and thus decrease de-
history characteristics, but they clearly influence the succes-
position rates at those specific locations. However, valley plugs
sional patterns and processes within floodplain systems.
can rapidly expand the channel blockage upstream and in-
fluence the hydrological regime and sedimentation rates in
new floodplain areas. Pierce and King (2008) documented the
12.14.4.2 Species Composition and Structure
expansion on one valley plug 80 m upstream in less than 1
year. The rates of plug expansion, and therefore the influence Species composition and structure of floodplain forests can be
of valley plugs on the adjacent floodplain, depend mainly on moderately to strongly affected by the formation of valley
the upstream sources of sediment and the stream’s ability to plugs and the resulting alteration to geomorphic and hydro-
transport the sediment. logic processes, including decreased productivity, increased
tree mortality, and altered germination potential. Rapid and
dramatic changes in floodplain forests correspond with ex-
12.14.4 Vegetative Responses cessive sedimentation and the associated inundation, whereas
more subtle changes in species composition and structure can
Floodplains support distinct assemblages of plants that are occur in response to altered abiotic processes in concert with
related to particular floodplain geomorphic features, soils, and the dispersal and germination abilities of floodplain tree
hydrologic regimes (Wharton et al., 1982; Brinson, 1990; species.
Sharitz and Mitsch, 1993). Hydroperiod and light availability Prolonged inundation or swamping of floodplain forests
are typically recognized as the primary forces that determine has been documented at numerous valley-plug sites (Happ
plant-species composition (Hall and Harcombe, 1998); the et al., 1940; Miller, 1990; Oswalt and King, 2005; Figure 8),
type and rate of sedimentation can also have a significant due to high sedimentation rates that prevent drainage of the
influence on plant-species composition as well as successional floodplain, a rise in the groundwater table as a result of
patterns within floodplains. Community-level changes are the channel aggradation, and altered hydrologic regimes (Happ
result of differential germination rates and survival of seed- et al., 1940). Swamping in the floodplain has been shown to
lings and overstory trees and the abiotic processes of flooding cause total mortality of original timber stands that are replaced
and sedimentation that can be greatly altered by the formation by willow (Salix spp.), maple (Acer spp.), and other disturb-
of valley plugs. ance-tolerant species (Happ et al., 1940). Frequent flooding,
Flooding is widely recognized as a major abiotic process high water, and/or long periods of inundation can kill seed-
that influences dispersal, germination, seedling establishment, lings, saplings, and mature trees (Hosner, 1960). Miller (1990)
230 Valley Plugs, Land Use, and Phytogeomorphic Response

Figure 8 Photograph illustrating tree mortality associated with prolonged inundation.

showed that historically, seasonally flooded, bottomland- composition and structure of floodplain forests (Oswalt and
hardwood forests in the Middle Fork–Forked Deer watershed King, 2005; Pierce, 2005).
in Tennessee had been replaced by shallow-water marsh Floodplain trees can become buried by as much as 1 m of
habitats as a result of valley-plug formation. This study also sediment at valley-plug locations (Pierce, 2005). Although the
showed that plant-species and microhabitat diversity were direct impact of this process on the stress and mortality of
reduced as a result of valley-plug formation. Oswalt and King mature trees is unknown, this burial can interact with
(2005) also documented a reduction in plant-species diversity groundwater levels to inundate root systems for extended
as altered abiotic processes due to valley-plug formation and periods of time. Pierce (2005) found that groundwater levels
channelization caused a shift in tree-species composition from were lower at valley-plug sites compared to unchannelized
typical floodplain tree species, such as sweetgum (Liquidambar sites; however, due to the burial of trees and their root systems,
styriciflua) and oak (Quercus spp.), to disturbance-tolerant the number of days that groundwater inundated root systems
species, such as maple (Acer spp.). during the growing season at valley plugs was more than 50%
Studies have also shown the loss of baldcypress-tupelo greater than the number of days root systems were inundated
(Taxodium distichum – Nyssa spp.) swamps in floodplains af- at unchannelized sites. Prolonged inundation of root systems
fected by channelization and valley plugs, even though they can have similar effects as prolonged surface flooding, in-
are considered flood tolerant (Oswalt and King, 2005; Pierce, cluding causing stress, low production of seeds, reduced
2005). High deposition rates of sediment can homogenize growth, and mortality (Happ et al., 1940; Hosner and Boyce,
and elevate floodplain surfaces over short-time periods (Pierce 1962; Kozlowski, 2002).
and King, 2008; Hupp et al., 2009), creating sites that are
unsuitable for flood-tolerant species. Meanwhile, the reduced
lateral migration of the channel associated with channelized 12.14.5 Restoration
streams may affect baldcypress-tupelo swamps in the long
term. Shankman (1993) noted that the development of Restoration of fluvial systems and their floodplains requires
baldcypress-tupelo swamps is the result of channel migrations watershed-based approaches that use geomorphic, hydrologic,
laterally through the floodplain over long periods of time. As a and biological processes to restore floodplain systems into
result of channelization, stabilized channels preclude lateral acceptable alternative states. The actual restoration target
migration and the formation of baldcypress-tupelo swamps, should be defined based on broad public input including that
thus limiting the occurrence of this forest community in of scientific experts and, as importantly, that of stakeholders.
floodplains of channelized streams. The lack of such a collaborative effort can result in confusion
High sedimentation rates can also increase stress and and concern among landowners, conservation organizations,
mortality of floodplain trees species but probably to a lesser state and federal agencies, over the various restoration options
extent than flooding. Accelerated deposition of coarse material and outcomes, potentially undermining restoration efforts.
that is typically associated with valley plugs can cause a de- One restoration approach that has been discussed for the
crease in the productivity of the soil, thereby affecting channelized tributary system of the Hatchie River that con-
germination, seedling establishment, and tree growth (Happ tains numerous valley plugs includes leaving valley plugs in
et al., 1940; Cavalcanti and Lockaby, 2005). Quality of place and creating artificial valley plugs in channelized tribu-
site factors, such as soil properties and elevation, is affected taries that currently lack valley plugs to prevent tributaries
by accelerated deposition and therefore influence the from transporting sediment downstream into the mainstem of
 Valley Plugs, Land Use, and Phytogeomorphic Response 231

the Hatchie River. In this case, much of the restoration focus is 1997). At valley-plug sites, high deposition rates over a large
on protecting the mainstem of the Hatchie River because it extent of the floodplain may have reduced the micro-
remains the longest unchannelized stretch of river in the topography of the floodplain. Greenhouse experiments and
Lower Mississippi River Valley. Short-term deposition analysis previous research (Hosner, 1957; Briscoe, 1961; Larsen, 1963;
(Pierce and King, 2008) indicated that valley plugs can protect Guo et al., 2002; Walls et al., 2005; Pierce and King, 2007b)
downstream sections from excessive sedimentation. However, have shown that floodplain tree species respond differently to
it is essential to note that this effect of downstream protection flood duration depending on life-history characteristics.
is only short term and is at the expense of the surrounding Seed availability may also be a limiting factor in the res-
floodplain. Previous research (Happ et al., 1940; Pierce and toration of the floodplain. Studies (Oswalt and King, 2005;
King, 2007a, 2008) indicated that eventually the plug will be Pierce, 2005) indicate that typically common floodplain tree
circumvented by the formation of new channels. This may species including oaks, baldcypress, and tupelo are absent at
result in formation of other plugs either upstream or down- valley-plug sites. Dispersal of seeds from upstream locations
stream of the previous plug. For example, along a 5-km stretch may also be minimal because of intensive agricultural prac-
of Clover Creek, TN, four former valley plugs occurred in tices that have reduced most of the floodplain forests up-
which the stream had created new channels through and stream. Particular attention may also be warranted regarding
around the valley plugs (Pierce, 2005), resulting in a swamped the impact of channel-restoration projects on water tables in
floodplain similar to those described by Miller (1990) and the floodplain. It has been demonstrated that channel alter-
Oswalt and King (2005), along the Middle Fork–Forked Deer ations can have a significant impact on water tables (Tucci and
River. Hileman, 1992; Pierce, 2005).
In the Hatchie River watershed, tributary floodplains that The probability of successful restoration of both the fluvial
have been degraded by channelization and valley plugs have system and the floodplain will be increased by considering the
been reduced in economic value, mainly timber value, by $5 above-mentioned factors as well as effective communication
438 ha–1 (Wells, 2004). Changes in forest composition and with landowners. Restoration efforts are complicated by the
structure may also affect wildlife communities. Research on fact that the land that is the source of the problems (e.g.,
the bird communities supported by the floodplain forests upstream gullies and upstream channelized reaches) and the
along the tributary systems of the Hatchie River suggests that land where the problem is expressed (e.g., the valley plug) are
tributary systems may have significant ecological importance commonly owned by different landowners. Yet, the cooper-
as undisturbed sites that support a diverse array of Neotropical ation of private landowners will be a critical factor in any
species (Pierce and King, 2011). Valley plugs and the associated restoration effort in watersheds affected by channelization and
hydrologic alterations may also impact the migration patterns valley-plug formation.
and productivity of fish and other aquatic organisms (Risotto
and Turner, 1985; Junk et al., 1989; Ward and Tockner, 2001).
Currently, most fluvial restoration approaches that address 12.14.6 Summary
channelization and valley-plug formation include a hydro-
logic and geomorphic assessment (Brierley and Fryirs, 2000; Valley plugs can become locally abundant in specific circum-
Brooks and Brierley, 2004). This approach can characterize the stances, which typically include highly erodible soils, removal
condition of various river sections to provide a biophysical of vegetation cover due to shifting land uses, and alterations of
basis for prioritizing restoration sites and strategies. Reach- the fluvial system that increase its transport capabilities. Once
scale conditions are considered within the watershed-scale established, valley plugs alter geomorphic, hydrologic, and
context to assess the fluvial and geomorphic recovery potential biological processes in floodplain systems. The types of impacts
(Brierley and Fryirs, 2000; Brooks and Brierley, 2004). The are highly variable and depend upon local and watershed
integrated watershed-scale framework allows this approach to characteristics, stage of recovery from fluvial disturbances, and
address system-wide problems, including: (1) stabilizing the spatial location of the plugs within the system.
sediment sources, including gully erosion and channel ero- Hydrological responses to valley-plug formation can in-
sion/bank failure, and (2) changes in stream gradient, stream clude increased flooding as a result of decreased capacity for
power, and channel morphology to prevent headcutting. Ad- channel storage and reduced drainage capacity of the flood-
vancement in our understanding of recovery processes asso- plain (Happ et al., 1940; Miller, 1990; Shankman and Sam-
ciated with human disturbances affecting fluvial systems has son, 1991; Diehl, 2000) or reduced surface flooding due to
enabled the geomorphic assessment approach to effectively accelerated sedimentation on floodplain surface (Pierce,
prioritize restoration sites and the appropriate strategies 2005). Sub-surface hydrologic response to valley plugs can be
(Schumm et al., 1984; Simon and Hupp, 1992; Brierley and complex due to influences of bed-level lowering through
Fryirs, 2000). channelization, within channel-filling from high sediment
Although restoration of the fluvial system is a necessary loads, and high rates of floodplain accretion.
first step in restoring the floodplain system, there are other Floodplain sedimentation at valley-plug sites can be sig-
processes involved that could prevent the establishment of nificantly greater, up to 10 times greater than at unchannelized
typical floodplain tree species in the floodplain. Additional sites, and occur over a large extent of the adjacent floodplain
concerns include the loss of microtopography as a result of (Pierce and King, 2007a, 2008). Sediment deposited on
excessive deposition and seed availability/dispersal. Micro- floodplains adjacent to valley plugs also has a greater pro-
topography has a direct effect on flooding, which can deter- portion of coarse sand that suggests that high-velocity over-
mine the distribution of floodplain tree species (Hodges, bank flows are created as a result of valley-plug formation.
232 Valley Plugs, Land Use, and Phytogeomorphic Response

Geospatial analysis of short-term deposition rates shows that indicate the complexity of these systems. For floodplain con-
valley plugs can strongly affect the spatial dynamics of de- servation and restoration efforts to be successful, a clearer
position rates by changing the direction of spatial dependence understanding of overbank sedimentation associated with
from parallel to stream flow to perpendicular to the stream valley plugs and other potential influential factors is needed.
flow (Pierce and King, 2008). However, the rates of de- Recovery processes of channelization also seem to be
position, types of sediment being deposited, and spatial pat- changing the hydrological conditions at valley-plug sites, but
terns can be variable because of other factors such as processes our understanding of these relationships is still rudimentary.
of channel recovery, local climatic variability, sediment sources The mechanisms involved in the creation of permanently
upstream, and anthropogenic disturbances. flooded areas and different developmental stages of valley-
Evaluation of environmental variables indicated that sur- plug formation are still poorly understood. Further research is
face and sub-surface hydrology, rates and types of sediment needed to test hypotheses of surface and sub-surface hydro-
deposited, and macronutrient concentrations of floodplain logical response to valley-plug formation related to stage of
soils are affected by valley-plug formation (Oswalt and King, development and specific site conditions in order to under-
2005; Pierce, 2005). These alterations have created environ- stand the factors influencing the variability of hydrological
mental gradients that are strongly affecting species com- responses. This information will also be useful for under-
position and stand structure of associated floodplain forests. standing the implications of valley-plug formation on flood-
Floodplain forests impacted by valley plugs can have lower plain forests and will enhance management and restoration
tree species diversity and no longer contain the typical asso- efforts.
ciations of oak species and baldcypress-tupelo swamps (Miller, The variability of abiotic processes associated with valley
1990; Oswalt and King, 2005; Pierce, 2005). Forest associ- plugs, particularly plug expansion, makes the future of the
ations consisting of disturbance-tolerant tree species domin- adjacent forests, especially in upstream sections, uncertain.
ated floodplains adjacent to valley plugs; seedling densities Simon and Hupp (1992) suggested that at least 65 years may
suggest that these associations will continue at least in the near be required for streams to recover from channelization, but
future (Oswalt and King, 2005; Pierce, 2005). Nevertheless, currently no data exist on the time period needed for flood-
the considerable variability associated with abiotic processes plain forests to recover from valley-plug formation. In add-
influenced by valley plugs has created temporal and spatial ition, the recovery process may be reset as private landowners
variability in the forest changes. The considerable variability of frequently remove valley plugs, through dredging, to protect
abiotic processes associated with valley plugs, particularly with their floodplain resources. This private activity allows sedi-
plug expansion, makes the future of these forests, especially in ment to travel downstream and eventually form plugs in other
upstream sections, uncertain. areas that reset or prolong channel recovery. The floodplain-
The hydrologic and sedimentation responses to valley-plug recovery process may depend on channel recovery, but may
formation may have dramatic impacts on floodplain processes also be complicated by anthropogenic disturbances and
and functions. The reduced flood pulse and increased sand limitations of seed availability and dispersal. Further research
deposits that can occur at valley-plug sites may reduce the is needed to determine if and when floodplain forests recover
fertility of the floodplain soils and directly influence the es- from valley plugs and the processes involved in their recovery.
tablishment and growth of floodplain tree species (Happ Flooding velocity overbank has been less studied in
et al., 1940). Although the effects of flood reduction on floodplain systems, but flow velocity can have a major influ-
floodplain forests have been less studied (Bedinger, 1978), ence on plant communities. Johnson (2000) showed that tree
decreased flooding has been shown to reduce tree growth and recruitment and seedling mortality were mainly influenced by
seed production (Burgess et al., 1973). Reduced flooding at stream-flow pulses that either eroded or buried seedlings. The
valley-plug sites may also influence seed dispersal of many deposition rates measured at valley-plug sites (Pierce and
floodplain tree species that have adapted seed production King, 2008) suggest that flooding velocities are much greater at
cycles to the timing of flood pulses for dispersal by water and valley-plug sites than at unchannelized sites. High-velocity
fish (Junk et al., 1989). flows within the floodplain not only allow for the transport of
Although recent research (Oswalt and King, 2005; Pierce bedload material into the floodplain but also contribute to
and King, 2007a, 2008) has been able to further our under- scouring and channel formation. Multiple valley plugs have
standing of floodplain systems and the effects of valley been observed over relatively short river stretches because
plugs on critical processes of these systems, many questions high-velocity flows have cut new channels through the
still remain. Expansion of valley plugs seems to be floodplain that allow transport of sediment downstream
unpredictable as a result of climate variability and geomorphic (A. Pierce, personal observation). High-velocity flows within
thresholds. Further study is necessary to understand the rates the floodplain may also influence the composition and
and processes involved in upstream expansion of valley plugs; structure of floodplain plant communities. High-flow vel-
this would be useful in prioritizing restoration efforts and ocities, sediment loads, and turbidity also reduce primary
predicting potential impacts to upstream forest communities. production affecting the biological processes within the sys-
Further investigation is also needed to understand the in- tems (Junk et al., 1989). Additional research is needed to gain
fluence that factors, such as channel-recovery processes and a better understanding of the variability in hydrological re-
anthropogenic disturbances, have on overbank sedimentation sponses to valley plugs and the influence and dynamic nature
dynamics in conjunction with valley plugs. Recent studies of channelization recovery processes.
(Pierce and King, 2007a, 2008) clearly show the variability in Restoration efforts would also benefit by research that
responses of sedimentation to valley-plug formation and identifies and categorizes the amount of degradation that has
 Valley Plugs, Land Use, and Phytogeomorphic Response 233

occurred within watersheds affected by channelization and Research Conference. Gen. Tech. Rep. SRS-48. Asheville, NC. US Department of
valley-plug formation. Sites could then be prioritized based on Agriculture, Forest Service, Southern Research Station, pp. 55–58.
Hack, J.T., 1960. Interpretation of erosional topography in humid temperate regions.
level of degradation, landowner desires, and potential to im-
American Journal of Science 258-A, 80–97.
prove the ecological conditions of the site and contribution to Hall, R.B.W., Harcombe, P.A., 1998. Flooding alters apparent position of floodplain
the entire system. In addition, if sacrificing some tributary sys- saplings on a light gradient. Ecology 79, 847–855.
tems to protect main-stem channels is going to be considered a Happ, S., Rittenhouse, G., Dobson, G., 1940. Some principles of accelerated stream
viable option, then research that determines the ecological costs and valley sedimentation. Technical Bulletin 695, US Department of Agriculture.
Happ, S.G., 1975. Genetic classification of valley sediment deposits. In: Vanoni,
of losing these systems and potential impacts on the watershed V.A. (Ed.), Sedimentation Engineering. American Society of Civil Engineers, New
system would be valuable. More research is needed to deter- York, NY, pp. 286–292.
mine the overall significance of these tributary systems to the Harms, W., Schreuder, H., Hook, D., Brown, C., 1980. The effects of flooding on
entire watershed, including their role in the conservation of the Swamp Forest in Lake Ocklawaha, Florida. Ecology 61, 1412–1421.
Heimann, D.C., Roell, M.J., 2000. Sediment loads and accumulation in a small
floodplain fish and Neotropical migrant songbirds.
riparian wetland system in Northern Missouri. Wetlands 20(2), 219–231.
Heitmeyer, M., Fredrickson, L., 1981. Do wetland conditions in the Mississippi
Delta hardwoods influence mallard recruitment? Transactions of the North
References American Wildlife and Natural Resources Conference 46, 44–57.
Hidinger, L.L., Morgan, A.E., 1912. Drainage problems of Wolf, Hatchie, and South
Fork of Forked Deer Rivers, in West Tennessee. Resources of Tennessee 2(6),
Allen, J.R.L., 1965. A review of the origin and characteristics of recent alluvial
231–249.
sediments. Sedimentology 5(2), 89–191.
Hodges, J.D., 1997. Development and ecology of bottomland hardwood sites.
Ashley, G.H., 1910. Drainage Law of Tennessee. State of Tennessee, State
Forest Ecology and Management 90, 125–177.
Geological Survey, Nashville, TN. Bulletin: Drainage reclamation in Tennessee.
Hohensinner, S., Habersack, H., Jungwirth, M., Zauner, G., 2004. Reconstruction of
Barstow, C.J., 1971. Impact of channelization on wetland habitat in the
the characteristics of a natural alluvial river–floodplain system and
Obion–Forked Deer Basin, Tennessee. In: Trefethen, J.B. (Ed.), Transactions of
hydromorphological changes following human modifications: the Danube River
the Thirty-Sixth North American Wildlife and Natural Resources Conference.
(1812–1991). River Research and Applications 20, 25–41.
Wildlife Management Institute, Washington, DC, pp. 362–376. Hoover, J.J., Killgore, K.J., 1997. Fish communities. In: Messina, M.G., Conner,
Bedinger, M.S., 1978. Relation between forest species and flooding. In: Greeson, W.H. (Eds.), Southern Forested Wetlands: Ecology and Management. Lewis
P.E., Clark, J.R., Clark, J.E. (Eds.), Wetland Functions and Values: The State of Publishers, Boca Raton, FL, pp. 237–260.
Our Understanding. Proceedings of the National Symposium on Wetlands, Hopkinson, C.S., Vallino, J.J., 1995. The relationships among man’s activities in
American Water Resources Association, Lake Buena Vista, FL, pp. 427–435. watersheds and estuaries. Estuaries 18, 598–621.
Bennett, S.J., Rhoton, F.E., 2009. Linking upstream channel instability to Hosner, J., 1957. Effects of water upon the seed germination of bottomland trees.
downstream degradation: Grenada Lake and the Skuna and Yalobusha River Forest Science 3, 67–70.
Basins, Mississippi. Ecohydrology 2, 235–247. Hosner, J., 1960. Relative tolerance to complete inundation fourteen bottomland tree
Bloom, A.L., 1978. Geomorphology: A Systematic Analysis of Late Cenozoic species. Forest Science 6, 246–251.
Landforms. Prentice-Hall Publishers, Englewood Cliffs, NJ, 510 pp. Hosner, J., Boyce, S., 1962. Tolerance to water saturated soil of various bottomland
Brierley, G.J., Fryirs, K., 1999. Tributary–trunk stream relations in a cut-and-fill hardwoods. Forest Science 8, 180–186.
landscape: a case study from Wolumla catchment, New South Wales, Australia. Huffman, R.T., 1980. The relation of flood timing and duration to variation in
Geomorphology 28, 61–73. selected bottomland hardwood communities of southern Arkansas: Final report.
Brierley, G.J., Fryirs, K., 2000. River styles, a geomorphic approach to catchment US Army Corps of Engineers miscellaneous paper EL-80-4. Washington, DC.
characterization: implications for river rehabilitation in Bega Catchment, Hunter, W., Carter, M., Pashley, D., Barker, K., 1993. The partners in flight species
New South Wales, Australia. Environmental Management 25, 661–679. prioritization scheme. General Technical Report RM-227, USDA Forest Service,
Brierley, G.J., Murn, C.P., 1997. European impacts on downstream sediment transfer Fort Collins, CO.
and bank erosion in Cobargo catchment, New South Wales, Australia. Catena Hupp, C., 2000. Hydrology, geomorphology, and vegetation of Coastal Plain Rivers
31, 119–136. in the southeastern United States. Hydrological Processes 14, 2991–3010.
Brinson, M., 1990. Riverine forests. In: Brown, S. (Ed.), Forested Wetlands. Elsevier, Hupp, C., Bazemore, D., 1993. Temporal and spatial patterns of wetland
New York, NY, pp. 87–141. sedimentation. West Tennessee. Journal of Hydrology 141, 179–196.
Briscoe, C.B., 1961. Germination of cherrybark and nuttall oak acorns following Hupp, C.R., Pierce, A.R., Noe, G.B., 2009. Floodplain geomorphic processes and
flooding. Ecology 42, 430–431. environmental impacts of human alteration along coastal plain rivers. USA.
Brooks, A.P., Brierley, G.J., 2004. Framing realistic river rehabilitation targets in Wetlands 29, 413–429.
light of altered sediment supply and transport relationships: lessons from East Hupp, C.R., Schenk, E.R., Richter, J.M., Peet, R.K., Townsend, P.A., 2009.
Gippsland, Australia. Geomorphology 58, 107–123. Bank erosion along the regulated lower Roanoke River, North Carolina.
Brooks, A.P., Brierley, G.J., Millar, R.G., 2003. The long-term control of vegetation Special Publication 451. Geological Society of America, Washington, DC,
and woody debris on channel and flood-plain evolution: insights from a paired pp. 97–108.
catchment study in southeastern Australia. Geomorphology 51, 7–29. Johnson, W., 1994. Woodland expansions in the Platte River, Nebraska: patterns
Burgess, R.L., Johnson, W.C., Keammerer, W.R., 1973. Vegetation of the Missouri and causes. Ecological Monographs 64, 45–84.
River floodplain in North Dakota. North Dakota Water Resources Research Johnson, W., 2000. Tree recruitment and survival in rivers: influence of hydrological
Institute Report WI-221-018-73, p. 162. processes. Hydrological Processes 14, 3051–3074.
Burt, T.P., Bates, P.D., Stewart, M.D., Claxton, A.J., Anderson, M.G., Price, D.A., Jones, R., Sharitz, R., Dixon, P., Segal, D., Schneider, R., 1994. Woody plant
2002. Water table fluctuations within the floodplain of the River Severn, regeneration in four floodplain forests. Ecological Monographs 64, 345–367.
England. Journal of Hydrology 262, 1–20. Junk, W.J., Bayley, P.B., Sparks, R.E., 1989. The flood pulse concept in
Cavalcanti, G.G., Lockaby, B.G., 2005. Effects of sediment deposition on fine root river–floodplain systems. In: Dodge, D.P. (Ed.), Proceedings of the International
dynamics in riparian forests. Soil Science Society of America Journal 69, Large River Symposium. Canadian Journal of Fisheries and Aquatic Sciences,
729–737. Special Publication 106, pp. 110–127.
Diehl, T., 2000. Shoals and valley plugs in the Hatchie River Watershed. 00-4279. Kleiss, B.A., 1996. Sediment retention in a bottomland hardwood wetland in Eastern
USGS, Nashville, TN. Arkansas. Wetlands 16(3), 321–333.
Fryirs, K., Brierley, G.J., 1998. The character and age structure of valley fills in Knighton, D., 1998. Fluvial Forms and Processes. Arnold, London, 383 pp.
upper Wolumla Creek, South Coast, New South Wales, Australia. Earth Surface Kozlowski, T.T., 2002. Physiological–ecological impacts of flooding on riparian
Processes and Landforms 23, 271–287. forest ecosystem. Wetlands 22(3), 550–561.
Guo, Y., Shelton, M.G., Heitzman, E., 2002. Effects of flood duration and depth on Kroes, D.E., Hupp, C.R., 2010. The effect of channelization on floodplain sediment
germination of cherrybark, post, southern, white, and willow oak acorns. In: deposition and subsidence along the Pocomoke River, Maryland. Journal of the
Outcalt, K.W. (Ed.), Proceedings of the Eleventh Biennial Southern Silvicultural American Water Resources Association 46, 686–699.
234 Valley Plugs, Land Use, and Phytogeomorphic Response

Langdale, G.W., Denton, H.P., White, Jr. A.W., Gilliam, J.W., Frye, W.W., 1985. Sharitz, R., Mitsch, W., 1993. Southern floodplain forests. In: Martin, W., Boyce, S.,
Effects of soil erosion on crop productivity of southern soils. In: Follett, R.F., Echternacht, A. (Eds.), Biodiversity of Southeastern United States/Lowland
Stewart, B.A. (Eds.), Soil Erosion and Crop Productivity. ASA-CSA-SSSA, pp. Terrestrial Communities. Wiley, New York, NY, pp. 311–371.
251–270. Simon, A., 1994. Gradation processes and channel evolution in modified West
Larsen, H.S., 1963. Effects of soaking in water on acorn germination of four Tennessee streams: process, response, and form. US Geological Survey
southern oaks. Forest Science 9, 236–241. Professional Paper 1470. US Geological Survey, Washington, DC.
Leopold, L.B., Wolman, G.M., 1957. River channel patterns: braided, meandering, Simon, A., Hupp, C.R., 1987. Geomorphic and vegetative recovery processes along
and straight. US Geological Survey Professional Paper 282-B. US Geological modified Tennessee streams: an interdisciplinary approach to disturbed fluvial
Survey, Washington, DC. systems. Proceedings of the Forest Hydrology and Watershed Management
Middelkoop, H., Asselman, N.E.M., 1998. Spatial variability of floodplain Symposium, Vancouver, August 1987. International Association of Scientific
sedimentation at the even scale in the Rhine-Meuse Delta, The Netherlands. Hydrology, Publication No. 167, pp. 251–261.
Earth Surface Processes and Landforms 23, 561–573. Simon, A., Hupp, C.R., 1992. Geomorphic and vegetative recovery processes along
Middelkoop, H., Van Der Perk, M., 1998. Modelling spatial patterns of overbank modified stream channels of West Tennessee. US Geological Survey, Nashville, TN.
sedimentation on embanked floodplains. Physical Geography 80, 95–109. Simon, A., Robbins, C.H., 1987. Man-induced gradient adjustment of the South
Miller, N.A., 1990. Effects of permanent flooding on bottomland hardwoods and Fork Forked Deer River. West Tennessee. Environmental Geology and Water
implications for water management in the Forked Deer River Floodplain. Sciences 9(2), 108–118.
Castanea 55, 106–112. Soil Conservation Service, 1977. Land treatment plan for erosion control and water
Miller, R.A., Hardeman, W.D., Fullerton, D.S., 1966. Geologic Map of Tennessee. quality improvement in the Obion–Forked Deer Basin River. USDA.
Tennessee Division of Geology, TN. Speer, P.R., Perry, W.J., McCabe, J.A., et al., 1965. Low-flow characteristics of
Morgan, A.E., McCrory, S.H., 1910. Preliminary report upon the drainage of the streams in the Mississippi embayment in Tennessee, Kentucky, and Illinois, with
lands overflowed by the North and Middle Forks of the Forked Deer River and a section of quality of water, by H.G. Jeffery: USGS professional paper 448-H.
the Rutherford Fork of the Obion River in Gibson County, TN. State of Stanturf, J.A., Schoenholtz, S.H., 1998. Soils and landforms. In: Messina, M.G.,
Tennessee, State Geological Survey, Nashville, TN. Conner, W.H. (Eds.), Southern Forested Wetlands: Ecology and Management.
National Research Council, 1992. Restoration of Aquatic Ecosystems. National Lewis Publishers, Chelsea, MI, pp. 123–148.
Academy Press, Washington, DC. Steiger, J., Gurnell, A.M., Goodson, J.M., 2003. Quantifying and characterizing
Nicholas, A.P., Walling, D.E., 1998. Numerical modeling of floodplain hydraulics contemporary riparian sedimentation. River Research and Applications 19,
and suspended sediment transport and deposition. Hydrological Processes 12, 335–352.
1339–1355. Streng, D., Glitzenstein, J., Harcombe, P., 1989. Woody seedling dynamics in an
Nilsson, C., Reidy, C.A., Dynesius, M., Revenga, C., 2005. Fragmentation and flow east Texas floodplain forest. Ecological Monographs 59, 177–204.
regulation of the world’s large river systems. Science 308, 405–408. Tucci, P., Hileman, G.E., 1992. Potential effects of dredging the South Fork Obion
Osterkamp, W.R., Hedman, E.R., 1977. Variation of width and discharge River on ground-water levels near Sidonia, Weakley County, Tennessee. US
for natural high-gradient stream channels. Water Resources Research 13, Geological Survey, Water-Resources Investigations Report 90-4041.
256–258. Turner, R.E., Forsythe, S.W., Craig, N.J., 1981. Bottomland hardwood forest land
Oswalt, S.N., King, S.L., 2005. Channelization and floodplain forests: impacts of resources of the southeastern US. In: Clark, J.R., Benforado, J. (Eds.), Wetlands
accelerated sedimentation and valley plug formation on floodplain forests of the of Bottomland Hardwood Forests. Proceedings of a Workshop on Bottomland
Middle Fork Forked Deer River, Tennessee, USA. Forest Ecology and Hardwood Forest Wetlands of the Southeastern US. Elsevier, New York, NY, pp.
Management 215, 69–83. 13–43.
Pierce, A.R., 2005. Sedimentation, hydrology, and bottomland hardwood forest US Department of Agriculture Soil Conservation Service, 1970. Hatchie River basin
succession in altered and unaltered tributaries of the Hatch River, TN. PhD survey report, Tennessee and Mississippi. US Department of Agriculuture Soil
dissertation. University of Tennessee, Knoxville, TN. Conservation Service.
Pierce, A.R., King, S.L., 2007a. The influence of valley plugs in channelized US Department of Agriculture Soil Conservation Service, 1986. Sediment transport
streams on floodplain sedimentation dynamics over the last century. Wetlands analysis report, Hatchie River Basin special study. US Department of Agriculture
27, 631–643. Soil Conservation Service, Tennessee and Mississippi, Nashville, TN, 17 pp.
Pierce, A.R., King, S.L., 2007b. The effects of flooding and sedimentation on van der Valk, A.G., Swanson, S.D., Nuss, R.F., 1983. The response of plant species
seed germination of two bottomland hardwood tree species. Wetlands 27, to burial in three types of Alaskan wetlands. Canadian Journal of Botany 61,
588–594. 1150–1164.
Pierce, A.R., King, S.L., 2008. Spatial dynamics of overbank sedimentation in Walling, D.E., He, Q., 1998. The spatial variability of overbank sedimentation on
floodplain systems. Geomorphology 100, 256–268. river floodplains. Geomorphology 24, 209–223.
Pierce, A.R., King, S.L., 2011. A comparison of avian communities and habitat Walls, R.L., Wardrop, D.H., Brooks, R.P., 2005. The impact of experimental
characteristics in floodplain forests associated with valley plugs and sedimentation and flooding on the growth and germination of floodplain trees.
unchannelized stream. River Research and Applications 27(10), 1315–1324. Plant Ecology 176, 203–213.
Risotto, S., Turner, R., 1985. Annual fluctuation in abundance of the commercial Ward, J.V., Tockner, K., 2001. Biodiversity: towards a unifying theme for river
fisheries of the Mississippi River and tributaries. North American Journal of ecology. Freshwater Biology 46, 807–819.
Fisheries Management 5, 557–574. Wells, A.R., 2004. Integrating geographic information systems and remote sensing
Robbins, C.H., Simon, A., 1983. Man-induced channel adjustment in Tennessee with spatial econometric and mixed logit models for environmental evaluation.
streams. USGS Water Resources Investigations Report, 82-4098. Doctoral dissertation, University of Tennessee, Knoxville, TN.
Saucier, R.T., 1994. Geomorphology and quaternary geologic history of the Lower Wharton, C.H., Kitchens, W.M., Pendleton, E.C., Sipe, T.W., 1982. The ecology of
Mississippi Valley Volume 1. US Army Corps of Engineers, Vicksburg, MS. bottomland hardwood swamps of the southeast: a community profile. Publication
Schumm, S.A., Harvey, M.D., Watson, C.C., 1984. Incised Channels: Morphology, No. FWS/OBS-81/37. US Fish and Wildlife Service, Washington, DC.
Dynamics, and Control. Water Resources Publications, Littleton, CO. Wilder, T.C., 1998. A comparison of mature bottomland hardwood forests in natural
Shankman, D., 1993. Channel migration and vegetation patterns in the Southeastern and altered settings in West Tennessee. MS thesis. Tennessee Technological
Coastal Plain. Conservation Biology 7(1), 176–183. University, Cookeville, TN.
Shankman, D., Pugh, T.B., 1992. Discharge response to channelization for a Coastal Wolfe, W.J., Diehl, T.H., 1993. Recent sedimentation and surface-water flow patterns
Plain stream. Wetlands 12(3), 157–162. on the flood plain of the North Fork Forked Deer River, Dyer County, Tennessee.
Shankman, D., Samson, S.A., 1991. Channelization effects on Obion River flooding, USGS, in cooperation with TWRA, Nashville, TN. Water Resources Investigations
western Tennessee. Water Resources Bulletin 27, 247–254. Report 92-4082.
 Valley Plugs, Land Use, and Phytogeomorphic Response 235

Biographical Sketch

Dr. Aaron R. Pierce received his undergraduate degree in biology from Hartwick College in New York. He earned a
master’s of science degree in ecology from Purdue University and a PhD in natural resources from the University
of Tennessee. He is currently an assistant professor of biological sciences at Nicholls State University in Thibo-
daux, Louisiana. His research investigates wetland ecosystems to improve management and conservation efforts.

Dr. Sammy King received his undergraduate degree in biology from Nicholls State University, in Thibodaux
Louisiana. He earned a master’s of science degree in zoology and wildlife from Auburn University and a PhD in
wildlife and fisheries sciences from Texas A&M University. He is currently a research wildlife biologist with US
Geological Survey and serves as the leader of the Louisiana Cooperative Fish and Wildlife Research Unit. His
research is focused on an integrated understanding of freshwater wetland ecosystems, particularly floodplain
ecosystems.
 12.15 Fire as a Geomorphic Agent
MB Stine, Texas State University – San Marcos, San Marcos, TX, USA
r 2013 Elsevier Inc. All rights reserved.

12.15.1 Introduction 236

12.15.2 Soil 237
12.15.2.1 Hydrophobicity 238
12.15.2.2 Infiltration 238
12.15.2.3 Nutrients 239
12.15.2.4 Organic Matter and Litter 239
12.15.2.5 Microbial and Faunal Activity 240
12.15.2.6 Soil Temperature and Moisture 240
12.15.3 Weathering 240
12.15.4 Erosion 241
12.15.4.1 Surface Erosion 241
12.15.4.2 Gully/Rill Formation 244
12.15.4.3 Mass Movements 244
12.15.4.4 Wind Erosion 244
12.15.5 Hydrology 245
12.15.5.1 Runoff 245
12.15.5.2 Streamflow 245
12.15.5.3 Sediment Loads and Channel Morphology 246
12.15.5.4 Large Woody Debris and Riparian Zones 246
12.15.6 Prehistoric Fire 246
12.15.7 Geomorphic and Topographic Influences on Fire 247
12.15.8 Conclusion 247
References 247

Glossary Hydrophobicity The property of being water repellent.

Bioturbation The action of plants or animals displacing Sediment bulking The accumulation of sediments in
or mixing sediments. overland flow, resulting in a debris flow.
Dry ravel The sliding or bouncing downslope movement Spalling The breaking of lensoid-shapes rock fragments
of sediment or organic matter. off of boulders.

Abstract

Intense heat produced by fire and the burning of vegetation can significantly alter geomorphic features and processes. The
effects of fire on geomorphology include increased erosion, runoff, and weathering rates, and changes to soil properties and
hydrologic processes. Fire severity and extent commonly determine fire’s influence on geomorphology. Widespread, high-
severity fires may alter sediment loads, erosion rates, and streamflow for an entire drainage basin; however, small, low-
intensity fires may not produce any noticeable geomorphic changes. Location and environmental conditions also exert
strong influence on the degree of fire’s effects. Vegetation type and density, topographic features, and soil conditions may
buffer, enhance, or influence the pattern of geomorphic disturbance. Many geomorphic alterations are a result of loss of
vegetation, and the duration of the changes to geomorphic processes is generally related to the re-establishment of
vegetation. Inversely, damage to geomorphology may influence vegetation regrowth rates, pattern, and composition. The
effects of fire on geomorphology are well addressed in the literature, particularly for fire-prone regions and for common
and highly visible geomorphic changes. This chapter provides an overview and a synthesis of major topics of
fire–geomorphology interactions as they are presented in the literature. Topics addressed include soil hydrophobicity,
infiltration, mineralization, and nutrients; surface erosion and mass movements; weathering; runoff, streamflow, and
morphology; prehistorical fire and geomorphology; and topographic influence on fire behavior.

Stine, M.B., 2013. Fire as a geomorphic agent. In: Shroder, J. (Editor in

Chief), Butler, D.R., Hupp, C.R. (Eds.), Treatise on Geomorphology.
Academic Press, San Diego, CA, vol. 12, Ecogeomorphology, pp. 236–251.

236 Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00331-6

 Fire as a Geomorphic Agent 237

12.15.1 Introduction ecosystem function. These fires prevented a buildup of fuel

and, in the process, hampered the occurrence of high-intensity
Fire exerts a significant influence on geomorphic processes, at fires.
numerous temporal and spatial scales and in varying inten- Currently, however, wildfires are increasing in intensity and
sities. The effects of fire, directly from heat and indirectly frequency and are projected to continue to become larger in
through ecological changes, can result in geomorphic dis- size in the future (Moreno et al., 1998; Pausas and Vallejo,
turbances, including increased erosion and overland flow rates 1999; Fagre, 2003; Westerling et al., 2006; Marlon et al.,
and changes to soil conditions (e.g., Christensen et al., 1989; 2009). The occurrence of high-severity fires is commonly
Brown, 1990; Moody and Martin, 2001b; Shakesby and Doerr, attributed to management decisions and climate change (Price
2006; Cannon et al., 2010). The interactions among fire, and Rind, 1994; Lenihan et al., 1998; McKenzie et al., 2004;
geomorphology, and organisms are extensive, and the effects Pausas, 2004; Running, 2006; Marlon et al., 2009; van Man-
that fire has on organisms can lead to significant changes to tgem et al., 2009). Marlon et al. (2009) and Westerling et al.
geomorphic features and processes. Likewise, the geomorphic (2006) concluded that fire events are increasing because of
processes affected by fire can influence post-fire plant re- changes in moisture, temperatures, and storm-tract regimes
establishment and growth. Similar to the varying effects of fire associated with shifts in climate. In the western United States,
on vegetation, fire’s influence on geomorphology varies widely snow is important to the hydrologic cycle and provides
as a result of differences in fire severity and size and the lo- moisture in characteristically otherwise arid regions (Bales
cation of the fire event. Fire can contribute significantly to et al., 2006). However, decreases in snowfall and earlier
landscape denudation (Jackson and Roering, 2009; Sass et al., melting times associated with climate fluctuations may result
2010) and is one of the primary causes of erosion in some in drier conditions and longer fire seasons between snow
areas (Shakesby and Doerr, 2006). cover. In addition to climate changes, an increase in fuel is also
Fire severity and frequency are significant factors in the responsible for greater fire intensities (Schoennagel et al.,
resulting extent of geomorphic disturbance. Wildfire severity 2004). Fire is a natural disturbance in many ecosystems;
may be categorized as ground (low), surface (moderate), and however, for decades it was a practice and a policy of the
crown (high, very high, or extreme), a classification that is United States to suppress wildfires (Busenberg, 2004). This
generally based on the amount of vegetation consumed and action led to a buildup of fuel that would have otherwise been
the behavior of the fire (Byram, 1959; Neary et al., 1999; consumed in more numerous but lower-intensity fire events.
Shakesby et al., 2003; Shakesby and Doerr, 2006). Ground The extensive influence that fire has on geomorphology;
or low-intensity fires burn ground cover and small (o2 m) the numerous interactions among fire, geomorphology, and
shrubs. Surface or moderate fires consume ground fuel, organisms; and on the increase in high-intensity fires em-
shrubs, and possibly small woody vegetation. Crown or high- phasizes the importance of understanding the effects of fire on
intensity fires burn all ground fuel, shrubs, and at least part geomorphology. The purpose of this chapter is to present a
of tree crowns (Figure 1). Extreme intensities burn much of current overview of the role of fire as a geomorphic agent, as it
the tree crowns. Fire frequencies vary from every 2–5 years is presented in the literature. The interactions among fire,
to more than several thousand years, depending on location, vegetation, and geomorphology are many, and geomorphic
environmental factors, and human influences (Baisan and disturbances as a result of fire may be found around the world
Swetnam, 1990; Grissino-Mayer, 1995; Wheelan, 1995; Fulé in locations where fires occur. Fire affects soil conditions,
et al., 1997; Shakesby and Doerr, 2006). In most fire- erosion rates, and hydrologic processes, both directly from
prone areas, low-intensity fires were historically a common intense heat and indirectly from the burning of vegetation.
Location and intensity of fire significantly influence the re-
sulting extent of geomorphic disturbance. Much of the litera-
ture on fire and geomorphology focuses on fire-prone areas,
such as the western United States, Mediterranean Spain, and
Eucalyptus forests of Australia (e.g., Christensen et al., 1989;
Doerr et al., 1994; DeBano, 2000; Cerdà and Doerr, 2005;
Shakesby and Doerr, 2006) and, therefore, many of the studies
presented in this chapter were conducted in those locations.
This chapter provides a general overview and select examples
from the literature on fire’s effects on geomorphology.

12.15.2 Soil

Many of the effects that fire has on geomorphology relate to

the changes that fire produces in soil. Fire directly influences
the soil by subjecting it to high temperatures. After a fire event,
distinct temperature gradients commonly occur in soils.
Temperatures on the ground surface can reach 500–700 1C
Figure 1 A crown fire immediately south of Glacier National Park, and possibly even 850 1C (DeBano, 2000). Temperatures up
MT, USA in 2007. Photo by David R. Butler. to 150 1C may occur 5 cm below the surface of mineral soil,
238 Fire as a Geomorphic Agent

but rarely does heat penetrate to 20–30 cm below the surface. between 175 and 270 1C but obliterated above 270–400 1C
Fire duration is generally the strongest overall factor in heating (Doerr et al., 2004). In agreement with the laboratory studies
depth. Fire-related heat can remain in the soil from a few on heat and hydrophobicity, Woods et al. (2006) found that
minutes up to several days following a fire. Fire also influences hydrophobic patches were greatest in severity and spatial ex-
soil indirectly by damaging vegetation, which can result in a tent with intermediate-intensity fire. Lewis et al. (2006) also
decrease in soil stabilization, nutrient alteration, and loss of found a greater extent of hydrophobic soils after moderate-
an insulating layer. Soil hydrophobicity (water repellency), intensity fires; however, they determined that high-severity
porosity, nutrients, infiltration, organic matter, microbial ac- fires resulted in hydrophobic soils at greater depths. Low-
tivity, erosion, pH, moisture, leaching, particle size, and color intensity fires resulted in the least amount of hydrophobicity
may be affected by fire, and the extent of soil damage is gen- because they did not generate enough heat to form large areas
erally related to the duration of the fire, temperature of the fire, of hydrophobic soils. Hydrophobic compounds commonly
and location characteristics (Certini, 2005). Changes to soil occur in undisturbed soils, and hydrophobicity will return
may be short term, long term, or permanent (Certini, 2005). to background levels generally within months to several years
after the fire event, once the soil reaches a certain moisture
threshold (Doerr et al., 2003; MacDonald and Huffman,
2004).
12.15.2.1 Hydrophobicity
The duration of hydrophobicity is highly variable and has
Intense heat from fire can cause soil to become hydrophobic been found to exist from a few seconds to 8 years (DeBano
(water repellant), which has numerous ramifications to geo- et al., 1976; Dyrness, 1976; King, 1981; Doerr and Thomas,
morphic processes (e.g., Munns, 1920; Rice et al., 1969; 2000). However, Shakesby and Doerr (2006) noted that little
Jackson and Roering, 2009). Hydrophobic soils may lead to is known concerning the length of time water repellency exists
increased erosion and overland flow, enhanced streamflow, because of a lack of long-term studies on post-fire conditions,
uneven wetting patterns, and decreased infiltration. The ex- and comparisons between studies are difficult because of dif-
tent, severity, and spatial pattern of the hydrophobic soils fering methods and environmental conditions among sites
determine the effects that it has on geomorphology (Woods and studies. The degree of hydrophobicity also varies over
et al., 2006). The degree of hydrophobicity varies greatly, with time and generally decreases in repellency with increasing
some soils being hydrophobic for years, whereas others are time after fire. Huffman et al. (2001) found evidence of re-
unaffected by fire (Leitch et al., 1983; Doerr et al., 2004; pellent soil 19 months after fire, though they had recorded a
Sheridan et al., 2007). Water-repellent soils are formed from significant decrease in hydrophobicity 3 months after the fire.
heat-concentrating hydrophobic compounds near the soils’ Hydrophobic soils are generally found in very patchy mosaics
surface (Imeson et al., 1992; Woods et al., 2006), and heat and vary extensively in spatial pattern and site location
may also bind these compounds to soil particles, further en- (Cannon and Reneau, 2000; Woods et al., 2006).
hancing soils’ water repellency. Some soils are naturally Soil hydrophobicity influences several geomorphic and
hydrophobic, and the hydrophobic compounds may be un- hydrologic processes, most of which are discussed in greater
affected by fire. However, the hydrophobicity of the soil may detail in following sections of this chapter. Where soils are
be destroyed if the temperature reaches the threshold in which hydrophobic, overland flow and infiltration are reduced
hydrophobic compounds are volatilized. Other soils contain (Burch et al., 1989; Letey, 2001). Water flow will be increased
hydrophobic compounds but do not display any water- in the hydrophobic patches, but generally will infiltrate into
repellent characteristics until triggered by fire (Doerr et al., the soil where it reaches hydrophilic soils or burrows
2000; Varela et al., 2005). Sheridan et al. (2007) also found (Shakesby et al., 2000; Woods et al., 2006). Therefore, the
that unburned soil hydrophobicity fluctuated throughout the spatial extent and position of the hydrophobic patches will
year, with high water repellency in the summer and non- exert a great influence on the amount of runoff that will occur
repellency in the winter, in their study on Eucalyptus forests in as a result of soil hydrophobicity. The patchy nature of
Australia. Marston and Haire (1990) found water-repellent hydrophobic soils results in differential patterns of soil
soils in both burned and unburned locations following the moisture. Hydrophobicity also varies in depth. Spigel and
1988 Yellowstone fires, but the hydrophobic soils extended to Robichaud (2007) found hydrophobic soils from 9 to 13 mm
a greater depth in the burned sites. Various organisms and within their study sites on hillslopes in Montana. Johansen
decaying plant matter, including bacteria, fungus, algae, and et al. (2001) reported hydrophobicity 1–2 cm from the ground
trees, particularly evergreen trees that commonly contain surface in a semiarid forest in New Mexico. Hydrophobicity, in
waxes and resins, have been associated with hydrophobicity summary, results in numerous geomorphic effects; however,
(Doerr et al., 2000). Hydrophobicity is most generally asso- the spatial pattern, presence, and degree of hydrophobicity
ciated with coarser-grained soils, such as sands; however, soils vary extensively within burned areas.
with high clay content may have high water repellency as well
(Crockford et al., 1991).
Fire intensity and the amount of time after a fire affect the
12.15.2.2 Infiltration
degree of hydrophobicity (Huffman et al., 2001; Shakesby
et al., 2003). Moderate intensity fires tend to produce more Decreased infiltration rates may be caused by water-repellent
water-repellant soils than low- or high-severity fires (Lewis soils, decreased interception from plants, loss of litter, in-
et al., 2006; Woods et al., 2006; Glenn and Finley, 2010). creased bulk density, and a cover of ash as a result of fire
Laboratory studies found that hydrophobicity was enhanced (Cerdà, 1998a; Robichaud, 2000; Martin and Moody, 2001).
 Fire as a Geomorphic Agent 239

Infiltration after a fire is generally lower than pre-fire rates, severe surface fires, though nitrate was significantly (Po0.05)
which results in an increase in erosion and overland flow higher in crown fires. Net nitrogen mineralization (ammoni-
(Cerdà, 1998a; Robichaud, 2000). Various factors brought on fication and nitrification) varied with time after the fire, with a
by fire affect infiltration rates, including bulk density, frozen negative net rate 1 year following fire, but with highest
ground, loss of vegetation, and decreased animal activity. Bulk amounts found 2–3 years after stand-replacing fire, and de-
density may increase following a fire as a result of changes to creasing in subsequent years.
soil aggregates and the action of voids being filled with ash Turner et al.’s (2007) results also supported drastic changes
and clay (Certini, 2005). These changes to fine-scale soil to nitrogen cycling following fire. Vegetation succession
properties result in a decrease in porosity and permeability is recognized to strongly influence nitrogen amounts. Herb-
and, therefore, a decrease in infiltration. Infiltration may also aceous vegetation in the Northern Rocky Mountains generally
be impeded by frozen ground, which could form as a result of increased dramatically in the 1–4 years following fire. These
a loss of an insulating vegetation layer (Campbell et al., 1977). plants used mineralized nitrogen and added nitrogen-rich lit-
A decrease of subsurface faunal activity in the burned area may ter to the soil. About 10 years after the fire, trees, which do not
also decrease infiltration rates. Intense heat and changes to soil add nitrogen to the soil as herbaceous vegetation does, begin
properties may deter animal activity in the soil, decreasing overtaking herbaceous activity, resulting in a decrease in ni-
their bioturbating effects, such as burrowing, actions that trogen input compared to the several years after fire when
often increase the porosity and permeability of the soil. herbaceous vegetation dominated (Chapman et al., 2006;
Infiltration rates are highly dependent upon location and Turner et al., 2007). In a chaparral ecosystem, DeBano and
environmental factors, as well as time since the fire event. Conrad (1978) found less nitrogen on the ground surface
Cerdà (1998a) studied the infiltration rates over the course of after a prescribed burn compared to pre-fire amounts. About
several years within a Mediterranean scrubland following fire. 110 kg ha1 of nitrogen was lost from plants and litter,
He used a rainfall simulator and measured infiltration rates 21.7 kg ha1 from 0–1 cm below the soil surface, and
beginning 6 months after a rangeland fire and continued to 14.7 kg ha1 from 1–2 cm below the surface during the fire.
record measurements up to 5.5 years following the fire. The They attributed this result to the likelihood that most of the
lowest infiltration rates were found soon after the fire and nitrogen, which was located in vegetation before the fire, was
increased throughout his study. The timing of runoff was volatilized in the high temperatures generated by the fire. In a
3 min, 47 s (StdDev ¼ 1.460 ) within the year following the fire different system, Lewis (1974) found an increase in soluble
and increased throughout the following 5 years to an average nitrate after fire in a South Carolina pine forest, but no
time of 25 min, 30 s (StdDev ¼ 7.290 ). Overland flow dropped notable difference in biologically available nitrogen. He at-
from 45% to o6% over the course of the 5.5 years after the tributed the increase in soluble nitrate to microbial activity and
fire. Ash provided a continuous ground cover of 3–5 cm fol- postulated that some nitrogen may have been lost through
lowing the fire. The increase in infiltration rates corresponded volatilization.
with vegetation growth. Other indirect effects of fire may in- Phosphorus is not volatilized by fire as is nitrogen (Wells,
fluence infiltration as it relates to vegetation growth. Erosion 1971; DeBano and Conrad, 1978). DeBano and Conrad
and other changes to soil may impede vegetation regrowth, (1978) found low pre-fire phosphorus amounts in their
which may subsequently decrease infiltration rates. chaparral study and similar amounts following fire. However,
before fire, phosphorus was located primarily in vegetation,
and after the fire significant amounts occurred in the ash layer
on the ground surface. Lewis (1974) had similar results, with
12.15.2.3 Nutrients
an increase in soluble phosphorus after fire but no significant
Fire’s effects on nutrients may indirectly influence geomorphic change to biological phosphorus. Several studies reported
process because of the importance of nutrients to plant es- greater phosphorus concentrations in runoff and streamflow
tablishment. Vital nutrients for vegetation growth, such as following fire, likely from increased runoff and available
nitrogen and phosphorus, may be altered in fires (DeBano phosphorus (Tiedemann et al., 1978; Schindler et al., 1980;
and Conrad, 1978; Cook, 1994; Knicker, 2007; Turner Leitch et al., 1983; Belillas and Roda, 1993).
et al., 2007 and references). A loss, gain, or redistribution of
nutrients may influence vegetation succession patterns, which
subsequently affects several geomorphic processes, such as
12.15.2.4 Organic Matter and Litter
erosion and runoff rates. Nitrogen receives particular attention
because it is especially important for vegetation establishment Vegetation litter and organic matter (OM), which are factors in
and growth, is commonly a limiting nutrient for vegetation, the amount of nitrogen present after a burn and which in-
and may be used as an indicator for ecosystem function. Fire, fluence erosion rates and soil temperature and moisture, may
however, may burn nitrogen sources, cause mineralization of be reduced or volatilized during a fire (DeBano and Conrad,
the nitrogen that does remain, and increase ammonium 1978; Marston and Haire, 1990; Turner et al., 2007). A loss of
concentrations. Turner et al. (2007), in their study on fire in surface litter may result in increased soil temperature, greater
the Greater Yellowstone Ecosystem, found that ammonium erosion rates, and decreased soil moisture because of the loss
was higher following fire; however, the concentration de- of an insulating layer. In the Greater Yellowstone fires that
creased with time, whereas nitrate increased in the first several occurred in 1988, Marston and Haire (1990) found the
years following fire. With regard to fire severity, ammonium amount of litter cover to be the greatest factor in lessening
resulted in no significant differences between crown and sediment loss in burned sites. They found that the amount of
240 Fire as a Geomorphic Agent

ground surface litter dropped from 72.5% in the unburned (e.g., 1–2 years) following a fire may result from a loss in
sites to 17.1% in burned areas. Similarly, percent OM de- vegetation cover. The resulting warmer soils can affect vege-
creased from 14.8% in unburned, unlogged areas to 5.3% in tation establishment, decrease soil moisture, and increase
burned, unlogged sites (Marston and Haire, 1990). DeBano thawing depth in areas that contain permafrost or seasonally
and Conrad (1978) found an average of 46% loss of OM on frozen ground. In high mountain areas, snow melt may be an
the ground surface after a fire compared to pre-fire levels in the important source of soil moisture; however, snow accumu-
same sites with a range of 0–70% in their six plots. Their study lation may be decreased with a loss of vegetation and ground
was performed before and after a prescribed burn in a chap- cover, and evaporation rates may be higher without litter and
arral ecosystem. The extent of OM loss depends heavily on the vegetation (e.g., Hiemstra et al., 2002).
intensity of the fire and has numerous ramifications regarding Other soil characteristics, such as color and pH, may also
vegetation succession and soil erosion. A study in a Pinus be altered. Color change following fire is generally a result of
pinaster forest determined that organic carbon was unaffected the increased temperatures to which the soil was exposed.
by temperatures of 150 1C, but completely oxidized at 490 1C Low- and moderate-severity fires commonly lead to a covering
(Fernández et al., 1999). At 220 1C, 37% of the OM was lost. of black or gray ash. High-severity fires may result in a redder
Combustion of OM in higher-severity fires may result in a loss soil if the organic matter is highly combusted and iron oxide
of nitrogen storage but increases nitrogen mineralization, and forms (Ulery and Graham, 1993; Certini, 2005). Ground color
therefore availability (Turner et al., 2007). Nutrients previ- may affect albedo, which can result in changes to the soil
ously stored within vegetation matter and roots may be re- temperature and, therefore, possibly soil moisture and plant
leased during and after fire as litter is burned and roots decay, growth. The burning of vegetation and organic matter may
resulting in an initial spike in nutrients several years following increase pH (Ulery et al., 1993; Pyne, 2001; Arocena and Opio,
fire. Though OM may be reduced following fire, it actually may 2003). Changes to pH levels may enhance vegetation regrowth
be increased beyond pre-fire levels in the long term. This result (Baath and Arnebrant, 1994; Chambers and Attiwill, 1994);
may be attributed to the type of plants that re-establish or a however, soil type, vegetation, and heat intensity will influence
decrease in mineralization following fire (Fernández et al., pH values, and some soil pH will not be affected by fire.
1999; Johnson and Curtis, 2001).

12.15.2.5 Microbial and Faunal Activity 12.15.3 Weathering

A loss of microbial activity in the top layer of soil influences Relative to the extensive literature on some of the other aspects
nutrient cycling and geomorphic processes (Klopatek, 1987; of fire, less attention has been devoted to fire and weathering
Shakesby and Doerr, 2006). Microbial, fungal, and faunal (Dorn, 2003; Shakesby and Doerr, 2006). Several studies have
activities directly affect water infiltration and erodibility assessed boulder and bedrock weathering rates and found that
through their burrowing activities and secretion of soil co- fire exerts notable influence on the breakdown and erosion of
hesion compounds, and indirectly by their influence on rock, in both field studies and lab experiments (Blackwelder,
vegetation succession. Microbes within the top few centi- 1927; Zimmerman et al., 1994; Allison and Bristow, 1999;
meters of soil are commonly destroyed in a fire, but some may Dorn, 2003). Blackwelder (1927) was one of the first geolo-
return in greater densities than pre-fire populations. Faunal gists to recognize the importance of fire to rock weathering,
organisms may also move deeper into the soil during a fire and through the use of laboratory experiments and by field
event and return near the ground surface after the fire. These observations concluded that fire was a significant factor of
organisms may increase infiltration and reduce hydro- rock weathering in the Rocky Mountains of the United States.
phobicity by burrowing and bioturbating the soil (Dragovich Blackwelder (1927) and others (Goudie et al., 1992; Allison
and Morris, 2002; Shakesby and Doerr, 2006). However, and Bristow, 1999; Dorn, 2003), however, recognized that
microorganisms may vacate an area if the fire has destroyed various factors influence rock-weathering rates, including fire
their food source or habitat. intensities, rock composition, air temperature, and moisture
content within the rock.
Spalling, the breaking of lensoid-shaped rock fragments off
12.15.2.6 Soil Temperature and Moisture
boulders, is the most common rock-weathering process that
Direct heat from fire and a loss of vegetation, ground cover, occurs during a fire (Humphreys et al., 2003; Shakesby and
and soil litter lead to an increase in soil temperature and a Doerr, 2006; Figure 2). Dorn (2003) collected quantitative
decrease in soil moisture (Tiedemann et al., 1978; Loáiciga field data on boulder weathering after a fire in Arizona. He
et al., 2001; Shakesby and Doerr, 2006). High temperatures found that spalling commonly occurred in a distinctly bi-
generated during fire can change the structure of soil and modal pattern, in which the spalling depth was usually either
decrease soil stability and increase erodibilty (DeBano et al., zero/very low or caused erosion over 76 cm in depth.
1998; Neary et al., 1999). Odion and Davis (2000) found Different rock types also resulted in different erosion
spatial variability in regrowth due to variations in soil heating. amounts. About 60% of sandstone boulders had a spalling
Patches that burned more intensely from falling canopy fuel depth of 0 cm, whereas 20% contained a spalling depth that
resulted in primarily bare ground, and areas that had been was greater than 76 cm in depth. The remaining 20% experi-
spared from the hotter soil heating contained an abundance enced a spalling depth 40, but o76 cm. Diorite, however,
of regrowth, largely a result of less damage to the seedbank in displayed about 60% spalling 476 cm and about 30% with
the soil. Increases in soil temperature over longer timescales no erosion. Similar to Dorn’s findings, other studies have
 Fire as a Geomorphic Agent 241

Figure 2 Rock spalling on a boulder after a fire. Figure 3 Rock fragments being transported downslope.

documented the importance of rock type to the amount of Kutiel and Inbar, 1993; Moody and Martin, 2001b; Cerdà and
weathering (Adamson et al., 1983; Ballis and Bosc, 1994). Not Lasanta, 2005; Shakesby et al., 2007; Sheridan et al., 2007;
only does rock type influence the amount of weathering, but Spigel and Robichaud, 2007; Smith and Dragovich, 2008;
also the pattern of rock breakdown. Dorn (2003) noted that Cannon et al., 2010). Cerdà (1998b) noted that fire is one of
diorite boulders tended to break off in blocks and stated that the primary factors of erosion in mountains of the Medi-
this feature may be because of laminar calcrete occurring in terranean, comparable to erosion rates caused by animal
the rock fractures, forming the breaking points for the boul- grazing. Loss of vegetation, ground cover, and litter from fire
ders during intense heating during a fire. result in increased overland flow and rain splash (Spigel and
Season and climate also influence rock weathering. Dorn Robichaud, 2007). In most cases, hillslopes experience in-
(2003) determined that sandstone eroded more in the winter creased erosion rates after a fire compared to unburned areas,
than in the monsoon season, which may be a result of fire- and the amount of sediment erosion generally corresponds to
induced fractures that captured moisture and subsequently fire intensity, climate, and environmental factors (Inbar et al.,
froze during the winter. Though fire has been attributed to 1998; Moody and Martin, 2001b; Spigel and Robichaud, 2007;
significant rock weathering in a range of environments, some Smith and Dragovich, 2008). Erosion rates vary extensively
environments are more conducive to rock-weathering pro- though and depend on numerous variables, including soil
cesses than others. Dragovich (1993) attributed fire as a sig- type, slope, and post-fire rain events, and rates generally de-
nificant agent in rock weathering in semi-arid regions where crease as vegetation cover begins growing and acts to stabilize
precipitation is not an important factor in weathering. In soil. Rates ranging from twofold to 2240-fold have been re-
warm climates, such as those in the southwestern United corded at the plot scale (Ronan, 1986; Sheridan et al., 2007).
States, fire may be an important agent of boulder weathering,
analogous to cryogenic weathering processes in colder regions.
12.15.4.1 Surface Erosion
van der Beek et al. (2001), however, noted that fire contributed
to rock erosion in humid temperate regions as well. Even in Numerous studies have found a significant increase in surface
areas where frost is a significant weathering agent, fire may erosion rates following fire in burned areas compared to un-
exert an important influence on rock-weathering rates. Ballais burned locations (e.g., Morris and Moses, 1987; Inbar et al.,
and Bosc (1994) found that fire may result in 10–100 times 1998; Mayor et al., 2007; Smith and Dragovich, 2008;
more erosion than frost in mountain environments of Figure 4). Surface erosion may result from a decreased infil-
southern France. Both bedrock fracture and boulder wea- tration capacity of the soil, saturation of the soil to the ground
thering lead to increased overall erosion and additional sedi- surface, or a decrease in surface roughness (Lavee et al., 1995;
ment and rock particles to the ground surface (Dorn, 2003). Shakesby and Doerr, 2006). Smith and Dragovich (2008)
The breakdown of boulders and bedrock results in particles determined a significant difference between unburned slopes
that are more easily transported by erosional processes and burned slopes in regard to erosion on their study of
(Abrahams et al., 1988; Allison and Higgitt, 1988; Figure 3). hillslope erosion following a moderate-severity fire in a sub-
alpine environment in the Snowy Mountains of Australia.
Unburned areas underwent a net gain of 2.6 mm of soil
12.15.4 Erosion and a 4.9 mm change in surface level, whereas the burned
areas had a net loss of 3.8 mm and a 6.7 mm surface
Erosion is one of the most common and obvious geomorphic level change. Morris and Moses (1987) also measured sig-
effects after a fire and is extensively addressed in the literature nificantly increased erosion rates between burned and un-
(e.g., Morris and Moses, 1987; Cerdà, 1998b; Brown, 1990; burned sites in their study in the Colorado Front Range.
242 Fire as a Geomorphic Agent

erodibility occurring 2 years after fire primarily to soil con-

ditions, such as ash cover and a lack of dry, loose soil, rather
than the regrowth of vegetation cover. Inbar et al. (1998)
found that erosion rates were highest in the first year following
the fire, lowest in the second year, and then increased again
above second-year rates during the third year as a result of
heavy rainfall. Moody and Martin (2001b) witnessed in-
creased erosion rates for the first two years following a fire, but
subsequently a return to pre-fire erosion rates. Sheridan et al.
(2007), in their study on post-fire erosion in a Eucalyptus forest
in Australia, found that highest erosion rates occurred during
the first winter following the fire. However, Mayor et al. (2007)
found erosion rates to increase during the third year after the
burn and then decrease with time.
As the previous examples indicated, erosion rates and
timing of erosion following fire vary greatly. These differences
may be largely attributed to the various environmental con-
Figure 4 Soil erosion on a slope that experienced a fire in 2006.
ditions associated with individual site locations (Inbar et al.,
1998; Bracken and Kirkby, 2005; Smith and Dragovich, 2008).
Similarly, Mayor et al. (2007) found erosion rates of 4563 kg Smith and Dragovich (2008) found that erosion rates de-
ha1 for burned areas compared to 0.12 kg ha1 in unburned creased with increasing elevation in the subalpine environ-
areas in the Mediterranean area of Alicante, Spain. Lavee et al. ment of Australia. They attributed this result to longer snow
(1995) determined significant variations in erosion rates cover at mid- to upper elevations and more organic matter,
in their study on Mt. Carmel, Israel. They concluded that such as fibrous roots, in the soils at the higher elevations.
surface roughness accounted for the spatial mosaic in erosion Inbar et al. (1998) determined that rainfall intensity exerted
differences that occurred in their study location. Areas that the greatest influence on erosion amounts in their study on
had experienced high-intensity fires, where only an ash layer the Mediterranean forest hillslopes of Mt. Carmel, Israel. Most
covered the ground, resulted in higher erosion rates. Lower- of the sediment yield from their plot occurred on days of high-
intensity burn areas, though, contained greater surface intensity rainfall. Vegetation cover, fire intensity, and slope
roughness because of the presence of vegetation, including exposure followed as influencing factors to sediment erosion
partially burned plants and an uneven ground surface from in their study. Slope aspect has been found to be a strong
fallen trees. Cerdà and Lasanta (2005) found that erosion variable of erosion rates in several studies. Moody and Martin
rates were not significantly different in one burned study (2001b) found more interrill sediment loss on north-facing
plot compared to the control plot, but their other burned plot slopes compared to south-facing slopes within the first
experienced about 10 times more erosion than the control year after fire in the Colorado Front Range. They estimated an
during the second years after the fire. Average rates for the erosion rate of at least 0.048 kg m1 d1 from north-facing
8-year study were 89.11 kg ha1 yr1 for the control plot, slopes and 0.0070 kg m1 d1 from south-facing slopes
143.18 kg ha1 yr1 for one of the burned plots, and during the summer following the fire. However, south-facing
566.44 kg ha1 yr1 for the other burned plot. The two slopes had a greater flux of sediment during the winter
burned plots were chosen to be replicates of each other on (B0.9 kg m1 d1) as a result of freeze–thaw patterns com-
their study on post-fire erosion and runoff in an abandoned pared to north-facing slopes (B0.09 kg m1 d1). Cerdà et al.
field in the Central Spanish Pyrenees. (1995), however, also found greater erosion rates on south-
The rate of erosion generally decreases with increasing time facing slopes on their study on a Mediterranean scrubland in
since the fire event (e.g., Diaz-Fierros et al., 1987; Morris and La Costera, Valencia, southeast Spain. They compared erosion
Moses, 1987; Marques and Mora, 1992; Inbar et al., 1998; rates of north-facing and south-facing slopes 10 years after a
Moody and Martin, 2001b; Cerdà and Lasanta, 2005: Smith fire. They attributed this result to slower vegetation regrowth
and Dragovich, 2008), with the greatest rates generally oc- and greater hydrologic stress on the southern slope.
curring within a year after fire (e.g., Helvey, 1980; Inbar et al., Vegetation type, regrowth, densities, and the spatial vari-
1998). However, different environments, even different sites ability of plants can have significant effects on erosion rates
within a study in one general location (Cerdà and Lasanta, (Figure 5), and these variables are well acknowledged in the
2005), may exhibit varying relaxation times (Mayor et al., literature for numerous ecosystems and over many timescales
2007). Smith and Dragovich (2008) found that the highest (e.g., Swanson, 1978; Brown, 1990; Shakesby et al., 1993;
amount of erosion occurred within the first 59 days following Moody and Martin, 2001a; Wondzell and King, 2003; Cerdà
fire, but continued throughout the study that spanned several and Doerr, 2005; Roering and Gerber, 2005). Johansen et al.
years. Morris and Moses (1987) found that rates declined (2001) and DeBano et al. (1998) suggested that ground cover
rapidly following fire but were still significant 4 years later. The is a greater factor in erosion rates than soil conditions. Vege-
decrease in erosion over time may be attributed to a loss of tation and erosion share a negative, exponential association
water-repellant soils, an increase in vegetation regrowth, and (Thornes, 1990; Cerdà, 1998b). As vegetation increasingly
detachment-limited soil (Morris and Moses, 1987; Inbar et al., re-establishes to pre-fire densities or greater, erosion rates cor-
1998). Sheridan et al. (2007) attributed decreased soil respondingly decrease to pre-fire levels. However, numerous
 Fire as a Geomorphic Agent 243

Figure 5 Herbaceous vegetation regrowth 4 years after a fire. Areas Figure 6 A root apparently acting to stabilize the soil.
that contained vegetation retained several centimeters more soil
compared to areas that lacked vegetation.

Underground biomass may also have significant impli-

factors influence the time between the fire event and sub- cations to erosion rates (Jackson and Roering, 2009). Roots
sequent vegetation regrowth and erosion rates. Fire intensity within the soil act as a stabilizing agent; however, they may
generally corresponds directly to the amount of available be weakened after a fire from death or stress to the plant
fuelwood provided by the vegetation, and subsequent geo- (Agee, 1993; Figure 6). Fire can directly damage tree roots by
morphic impacts generally relate to the intensity of the fire scorching root tissue near the ground surface. Tree death and
(Swanson, 1978). In low-intensity fires, herbaceous vegetation the loss of root systems would contribute to destabilizing the
and shrubs may be incinerated; however, much of the trees soil and increasing the chance of slide formations (Wondzell
and larger vegetation will likely still be living. These plants and King, 2003). Soil destabilization from loss of tree roots as
would provide soil stabilization and only low amounts of a result of tree death would likely not be a significant factor
erosion may take place in response to a loss of the ground immediately after a fire, but rather once the root systems have
cover. Recovery time would also likely be minimal. Moderate- had time to begin weakening and decaying. Jackson and
to high-severity fires, however, can result in much loss of Roering (2009), however, found a significant loss of root
vegetation and root biomass. This result would potentially strength only 1 month after a fire. They attribute this result
lead to greater amounts of surface erosion, changes to at least partially to damage the trees may have experienced
soil, debris flows, and overland flow. However, many com- before the fire (e.g., logging and insects).
pounding factors may be involved in erosion rates (Swanson, Regional and seasonal differences in rainfall, frequency
1978), such as fallen trees acting as sediment traps. and patterns of storm events, and freeze–thaw processes exert
Not only does the presence or absence of vegetation affect considerable influence on erosion rates and their temporal
erosion, but vegetation composition and plant type as well variability (Cerdà and Doerr, 2005). Fires that occur soon
exert a significant influence on the rate of erosion and changes before the rainy season in wet–dry climates, before vegetation
to erodibility over time (Brown, 1990; Cerdà and Doerr, becomes seasonally dormant, or preceding high-intensity
2005). Brown (1990) determined that even in areas of the storm events, commonly result in increased erosion and
same vegetation type (Mediterranean type in the case of his overland flow rates than would otherwise occur (Cannon
study), different plant species lead to various erosion rates. He et al., 1998; Cannon and Reneau, 2000; Cerdà and Doerr,
concluded that a plant’s life history, phenology, structure, and 2005; Vermeire et al., 2005; Gabet and Bookter, 2008). Cerdà
litter would influence the spatial pattern of surface erosion. and Doerr (2005) found average erosion rates of 79.75 g
Cerdà and Doerr (2005) studied erosion rates and corres- m2 h1 (StdDev. ¼ 42.47) within the first year after a fire in
ponding plant cover and type over a span of 11 years following the wet season and 30.05 g m2 h1 (StdDev. ¼ 12.96) in the
a fire in the eastern Mediterranean. They found that erosion dry season. Cannon et al. (1998) assessed the post-fire debris
rates in plots dominated by herbs decreased to below neg- and concentric flows that resulted from torrential rains after
ligible (1.1 g m2 h1) 2 years after the fire and, similarly, shrub fire in an area near Glenwood Springs, CO, USA. The resulting
plots fell below negligible erosion rates 2.5 years following the flows covered Highway 70 with 70 000 m3 of dry ravel, soil,
fire. Dwarf-shrub and tree plots, however, only reached these and other materials from the hillslopes and ravines where
levels, or close to them, by 6 and 11 years, respectively, after an 800 ha fire burned prior to the storm event. Spigel
the fire. They found, however, complex compounding factors and Robichaud (2007) found that high-intensity, short-
associated with the results between the tree plots and the duration thunderstorms resulted in the most erosion on
erosion rates. Hydrophobicity was found in increasing steep Montana hillslopes. Rainfall at a rate of 75 mm h1 for
amounts up to 8 years after the fire in the tree plots, which 15 min resulted in an erosion rate greater than 20 t ha1. Low-
would have contributed to greater erosion. intensity, long-duration rain events produced only negligible
244 Fire as a Geomorphic Agent

amounts (o0.01 t ha1). Though heavy rainfall after a fire can Creek site, which had relatively moderate slope angles (about
result in an increase in erosion, drought may also lead to long- 25% slope). The Capulin Creek basin was also larger in area
term increased erosion rates. Mayor et al. (2007) attributed (39.9 km2) than the North Tributary basin (6 km2). This result
their finding of increasing erosion to drought that impeded supported Meyer and Wells (1997) finding that smaller basins
vegetation regrowth following fire. tend to have more debris flows than larger basins, which are
more apt to flood (Cannon and Reneau, 2000).
Dry ravel (the sliding and bouncing downslope movement
12.15.4.2 Gully/Rill Formation of sediment or organic matter) can occur after fire as a result of
a loss in stabilizing vegetation (Jackson and Roering, 2009).
Erosion can cut channels into hillslopes in the form of rills
Jackson and Roering (2009) determined dry ravel to be a
and gullies, which can transport significant amounts of sedi-
significant source of erosion and sediment movement fol-
ment and create landscape features that persist for years
lowing fire on the steep slopes of the Oregon Coast Range,
(Wells, 1987; Moody and Martin, 2001b). These features may
USA. They estimated that dry ravel after a fire accounts for
become enlarged over time as water flows down through the
about 5–20% of the long-term erosion in parts of Oregon. In
channels and frost action works to continue to loosen and
some areas of their study, they found that ravel was likely
remove sediment. Moody and Martin (2001b) estimated rill
responsible for removing large patches of the soil mantel and
erosion account for 40 m2 of sediment on the north-facing
exposing bedrock. Ravel quickly diminished with time after
slope and 100 m2 of sediment on the south-facing slope
fire (most of the ravel deposits occurred within 2 weeks fol-
during the summer following fire. The average distance be-
lowing fire) because of a loss of available sediment material.
tween rills was about 10 m and the average cross-sectional area
A less commonly studied type of debris flow results from
of rills was 0.02 m2 (n ¼ 681). From 2–4 years following the
sediment bulking, in which overland flow accumulates sedi-
fire, rills widened because of freeze–thaw action and heavy
ment until it becomes a debris flow. Gabet and Bookter (2008)
rainfall events. Unlike some locations (Schumm, 1956), rills
found bulked debris flows started as surface runoff accumu-
remained in place for the 4 years of the study in the semi-arid
lating sediment, which subsequently became a rill, increased
mountains of the Colorado Front Range. They hypothesized
in size to a gully, and then formed into a bulked debris flow.
that over long time spans, rills in semi-arid areas may form
Their study sites were located on steep, low-order hillslopes
into gullies and possibly larger drainage systems, especially on
in southwestern Montana, an area that had experienced a fire
north-facing hillslopes. Cannon and Reneau (2000) found rill
a year prior to the flows. This type of debris flow is indicated
formations in moderate-severity burn sites up to 5 cm wide
by a lack of landslides scarps and may have landform impli-
and 4 cm deep and extensive rill erosion in high-severity fire
cations (Cannon, 1997, 2001; Cannon et al., 2001b; Gabet
sites in their study location of Capulin Canyon, New Mexico.
and Bookter, 2008). Gabet and Bookter (2008) discussed
However, erosion response varied widely among their sites
that gullies and debris flows likely formed and filled in
with some experiencing little or no rill erosion.
throughout time, and suggested that sediments from talus
slides, tree–throw, and bank failure supply sediments that fill
in the excavated gullies. They noted the importance of this
12.15.4.3 Mass Movements
fire–geomorphic relationship to hillslope processes and how it
Fire can result in increased overland flow and sediment influences landforms over geologic timescales.
movement, decreased infiltration, the formation of water- Soil creep and snow avalanches may also form as a result of
repellent soils, and loss of vegetation and litter cover – factors fire and the loss of vegetation on hillslopes (Swanson, 1978;
that increase the chance for debris flows (Mersereau and Sass et al., 2010). Soil creep, the slow movement of soil
Dyrness, 1972; Wohl and Pearthree, 1991; Meyer and Wells, downslope, may occur or increase after fire due to the loss of
1997; Cannon, 2001; Cannon et al., 2001a, 2001b; Cannon stabilizing vegetation and roots, and may be a significant long-
and Gartner, 2005). Three primary factors that contribute to term factor resulting from fire. Relative to a fire event, soil
debris flows following fire are a loss of root structures, which creep generally occurs before root systems have re-established,
leads to sediment destabilization; the accumulation and but after the soil has become wettable and potentially satur-
release of dry ravel deposits; and the accumulation of sedi- ated. Vegetation loss in high elevation areas will alter snow
ments in overland flow, resulting in a bulked debris flow accumulation and melting patterns. Vegetation also acts to
(Wells, 1987; Reneau et al., 1990; Meyer and Wells, 1997; stabilize snow, and the loss of vegetation in avalanche initi-
Cannon et al., 2001b; Gabet and Bookter, 2008; Jackson and ation areas may result in increased avalanche events (Sass
Roering, 2009). Topography, soil type, lithology, burn mosaic, et al., 2010).
past debris flows, root matter, and vegetation cover are char-
acteristically more important indicators for debris flow risk
than just the event of a fire however (Wohl and Pearthree,
12.15.4.4 Wind Erosion
1991; Spittler, 1995; Cannon and Reneau, 2000; Larsen et al.,
2006). Loose, fine-grained, cohesionless materials and steep Though erosion from water is a more common occurrence and
slopes are more prone to debris flows than other materials and well addressed in the literature, wind-induced erosion can also
topography (Spittler, 1995). Cannon and Reneau (2000) occur after fire, especially in nonforested areas (Zobeck et al.,
found that debris flows occurred on the steep-sloped sites of 1989; Cannon et al., 1998; Whicker et al., 2002; Vermeire
the North Tributary Basin (up to 60% slope), even though the et al., 2005). Wind is a particularly significant agent of erosion
area had experienced less extensive fire than their Capulin after fire in grasslands, prairies, and arid and semi-arid lands
 Fire as a Geomorphic Agent 245

(Whicker et al., 2002; Ravi et al., 2006, 2009), though even intensities and regimes (Rubio et al., 1997; Gonzàlez-Pelayo
winds at 20 m s1 may only result in windblown sediment et al., 2006). High-intensity rainfall generated more runoff in
20% of the time (Stout, 2001). Atkinson (1984) found 8 cm of burned soils than in unburned soils and lower-intensity
sediment and ash buildup on the windward side of objects in rainfall events (Gonzàlez-Pelayo et al., 2006). Cerdà and
their study located south of Sydney, Australia. Soil type, soil Lasanta’s (2005) results on runoff in the Central Spanish
moisture, hydrophobicity, ground cover, and time of year are Pyrenees indicted that rates were also significantly dependent
factors that influence the amount of windblown erosion on vegetation and soil characteristics in addition to rainfall
(Fryrear, 1995; Vermeire et al., 2005; Ravi et al., 2009). Ravi events and intensities. Their plots averaged 48.49 mm of
et al. (2009) found that wind erosion was enhanced following runoff per year for the control plot compared to two burned
fire and determined that hydrophobic soils resulted in in- plot runoff values of 78.69 mm year–1 and 94.3 mm year–1.
creased soil loss from wind erosion following fire in Cimarron Time of year also factors into runoff values because of the
National Grasslands in Kansas, USA. Hydrophobic soils are pattern of storm events and vegetation growth. In areas that
drier than normal, and lack of soil moisture is a significant experience intense summer thunderstorms or high rainfall,
factor in wind-blown erosion. The implications of wind ero- runoff may be greater during the summer than the winter,
sion include the loss of soils as well as nutrients from some even though less vegetation growth would be present during
areas and accumulation of soils and nutrients elsewhere. the winter (Gonzàlez-Pelayo et al., 2006). If rainfall amounts
Wind-blown sediments result in redistribution of nutrients, and intensities are comparable in winter and summer, higher
generally from areas of little to no vegetation to vegetation runoff rates will likely occur during the winter because of less
patches, forming a mosaic of soil-nutrient patches and influ- vegetation growth. Rainfall regimes also influence soil hydro-
encing vegetation growth (Ravi et al., 2009). phobicity, which affects runoff rates (Gonzàlez-Pelayo et al.,
2006). Consistent wetting of the soil may reduce soil hydro-
phobicity, leading to increased infiltration and lower runoff
values. Gonzàlez-Pelayo et al. (2006) found an increase
12.15.5 Hydrology
in runoff of 77.15% in burn plots in the year following
fire compared to control plots in their study in a Mediterra-
Fire can result in significant changes to hydrology and stream
nean region of Llı́ria-Valencia, Spain. They determined that
morphology (Brown, 1972; Prosser and Williams, 1998;
the spring/autumn average maximum-intensity rainfall was
Dwire and Kauffman, 2003; Shakesby and Doerr, 2006; Bart
18.49 mm h1 within a 30 min time span, and was of short
and Hope, 2010). Many of the processes that occur after a fire,
duration, compared to 5.51 mm h1 within a rain event of
such as increased erosion, decreased infiltration, mass move-
longer duration during the winter. This difference in precipi-
ments, and loss of vegetation, affect streams. The effects of fire
tation regimes resulted in 57% less runoff in the winter for
may include a loss of riparian vegetation; sediment influx
the high-severity fire and 65% in moderate-intensity fire plots.
from debris flows and erosion; the reduction of log jams
A reduction of 18% was found in the control plots.
during fire and the resulting release of stored sediment
(Cannon and Reneau, 2000; Benda et al., 2003; Larsen et al.,
2006); changes to in-channel erosional and depositional fea-
tures (Moody and Martin, 2001b); increased hillslope runoff, 12.15.5.2 Streamflow
stream flow, and storm flow (Keller et al., 1997; Scott, 1997;
Effects of fire on streamflow vary widely, with some systems
Ferreira et al., 2000; Moody and Martin, 2001a); the release of
undergoing significant increases to streamflow, as others hav-
dry ravel into the stream channel (Florsheim et al., 1991;
ing no noticeable difference between pre- and post-fire flows
Shakesby and Doerr, 2006); turbation (Gresswell, 1999); and
(Chandler et al., 1983; Bosch et al., 1984; Lindley et al., 1988;
increased amounts of woody debris within several years fol-
Britton, 1991; Lavabre et al., 1993; Scott, 1993; Hessling,
lowing fire (Bendix and Cowell, 2010). Hydrologic changes
1999; Loáiciga et al., 2001; Aronica et al., 2002). Streamflow is
after fire have been studied in small and large catchment
commonly a factor of numerous terrestrial variables, which
basins in numerous ecosystems and results have varied
themselves respond to fire in different ways. Hessling (1999)
extensively.
and Loáiciga et al. (2001) found increased streamflow of
as much as 30–50% after fire; however, Aronica et al. (2002)
did not find any significant post-fire changes in streamflow in
12.15.5.1 Runoff
a similar-sized watershed. Several studies have found that
Hydrophobicity, decreased infiltration, and vegetation and streamflow increases in small (o5 km2) and large (450 km2)
litter loss lead to greater overland flow after a fire, which can watersheds (Lavabre et al., 1993; Scott, 1993; Hessling, 1999).
result in increased base and storm flows in streams (Lindley Bart and Hope (2010) assessed six paired catchment basins
et al., 1988; Lavabre et al., 1993; Scott, 1993; Hessling, 1999; greater than 50 km2 in area in a Mediterranean-climate region
Loáiciga et al., 2001). Johansen et al. (2001) used a rainfall of California and found that streamflow varied among their
simulator to determine runoff rates after a fire in a semi-arid sites. They determined that soil wetness was the most signifi-
forest in New Mexico. They found that 45% of the 120 mm of cant factor in post-fire streamflow changes and that catchment
applied precipitation resulted in runoff in the severely burned and burn area did not exert differences to streamflow, either
areas compared to the 23% in the unburned sites. This result monthly or annually. Comparatively, Beeson et al. (2001)
correlated with greater areas of bare mineral soil. Several created a hydrologic model to predict and map overland flow
studies found that runoff was directly dependent on rainfall in response to fire and determined that post-fire response
246 Fire as a Geomorphic Agent

would lead to significantly greater overland flow and therefore small watersheds in the southern California chaparral region.
influence stream morphology and increase flow rates. The After 2 years, 16.8% of the dead trees had fallen. Their results
great variations in results in post-fire streamflow indicated that indicated that tree species was an important factor in deter-
streamflow reactions depend greatly on spatial and temporal mining which trees had fallen within the 2 years since their
variability, including site- or region-specific characteristics, initial sampling following the fire and, therefore, which spe-
weather patterns, vegetation, burn severity, season, ground- cies would likely serve as woody debris in stream channels.
water and surface water dynamics, and topography (Townsend
and Douglas, 2000; Beeson et al., 2001; Jung et al., 2009).
12.15.6 Prehistoric Fire

12.15.5.3 Sediment Loads and Channel Morphology Fire histories have attracted much interest because of the in-
formation they can provide on long-term climate–fire rela-
Higher runoff rates and the resulting increase in streamflow
tionships and fire frequencies and events over the span of
following fire may result in added sediment to streams and
thousands or millions of years (e.g., Meyer et al., 1995, 2001;
influence stream channel morphology, especially in areas with
Moody and Martin, 2001b; Gavin et al., 2003; Roering and
steep slopes (Laird and Harvey, 1986; Morris and Moses, 1987;
Gerber, 2005). Fire events in the early Holocene were frequent
DeBano et al., 1998; Inbar et al., 1998; Benda et al., 2003).
and likely responsible for significant erosion and hillslope
Morris and Moses (1987) concluded that fire events may
aggradation (Personius et al., 1993; Roering and Gerber,
contribute to much of the long-term sediment yield in the
2005). An assessment of past fires may be made by analyzing
Colorado Front Range. Cerdà and Lasanta (2005) found an
sediment stratigraphies in alluvial fans (Meyer et al., 1995),
erosion rate of 143 kg ha1 year1 in burn plots of the Central
tree-ring chronologies (Gavin et al., 2003), and with the use
Spanish Pyrenees compared to 89 kg ha1 year1 for the con-
of charcoal studies (Long et al., 1998; Gavin et al., 2003). The
trol plots. Johansen et al. (2001) found sediment yields of
presence of charcoal and/or ash within a layer of sediment
76 kg ha1 mm1 in burn plots compared to 3 kg ha1 mm1
may indicate that the layer was deposited as a result of
in unburned plots on their study in the semi-arid forest of New
post-fire erosion. Though documented histories of fire extend
Mexico, USA. They found that sediment yields were directly
back 6000 years for parts of Europe, those of North America
proportional to loss of ground cover; however, runoff and
are much shorter, and even 6000 years does not provide
sediment loads were not proportional. Sediment yields were
much insight into long-term, climate–fire and fire–erosion
about 25 times greater in burned plots compared to unburned
relationships.
plots, whereas runoff rates were only about 2 times greater in
Meyer et al. (1995) and Meyer et al. (2001) analyzed
burned areas. They attributed this result of significant sediment
sediment deposits using radiocarbon dating to determine
yields to a loss of ground cover, which would lead to the ero-
erosion amounts in prehistorical time. Meyer et al. (1995)
sive forces of increased rainsplash, shear erosion, and overland
found that 30% of an alluvial fan deposit was comprised of
flow. Similar to some other studies (Campbell et al., 1977;
fire-related sediment in their study on an alluvial fan in
Davenport et al., 1998), they found that sediment yield in-
northeastern Yellowstone National Park. They determined fire-
creased significantly when bare soil exceeded 60–70%. Sedi-
induced sediment pulses from 7500, 5500, 4600–4000, and
ments may reach stream channels and subsequently be
950–800 years BP, and found that major sediment deposition
transported downstream or be deposited within the channel
activity occurred from fire in conjunction with the Medieval
(Bart and Hope, 2010). This result will lead to increased stream
Warm Period. Their results regarding prehistoric fire activity as
turbidity, changes to channel form, and enhanced erosion as
well as more recent events indicated that increased fire activity
sediments are transported out of a system.
corresponded to warmer climate cycles and conditions. Meyer
et al. (2001) found that prehistoric sedimentation rates, as a
result of large sediment fluxes following fire events, in the
12.15.5.4 Large Woody Debris and Riparian Zones
Holocene were much greater in two tributary basins of the
Large woody debris (LWD) has numerous implications South Fork Payette River in west–central Idaho. Climate, they
to stream channel geomorphology, including sediment re- concluded, was an important factor to sediment yields because
tention, bank stability, increased roughness, decreased vel- fire events corresponded with climate, and storms following a
ocity, and changes to channel and bank morphology (Keller fire event are likely to produce greater erosion rates compared
and Swanson, 1979; Gurnell et al., 2002; Faustini and Jones, to burned areas with no post-fire rain events.
2003; May and Gresswell, 2003; Montgomery and Piégay, Roering and Gerber (2005) used measurements of geo-
2003). Woody debris, or potentially future woody debris, may morphic response rates after a fire with a high-resolution
be lost during a fire from the incineration of wood in the ALSM topographic dataset and built a physically based model
channel and trees located within the riparian zone. Conversely to determine erosion rates of natural landscapes on steep
however, tree mortality, as a result of fire, can lead to an in- soil-mantled slopes in the Oregon Coast Range over long
creased influx of woody debris in stream channels following timescales. Their results indicated that steep slopes were an
fire (Bragg, 2000; Bendix and Cowell, 2010). Bendix and important factor in post-fire sediment yield. However, most
Cowell (2010) presented an overview of LWD and riparian 10 m or 30 m digital elevation models do not contain the
vegetation after fire in their recent publication on riparian tree accuracy necessary for determining the distribution of steep
mortality. They found post-fire tree mortality of 94% within hillslopes and their specific gradient, which is crucial for cal-
the riparian zone of their study, which was located in two culating more accurate erosion rates. They found that fire may
 Fire as a Geomorphic Agent 247

account for about 50% of long-term sediment erosion on are well addressed in the literature, such as erosion (e.g.,
steep slopes and concluded that fire could be responsible for Morris and Moses, 1987; Moody and Martin, 2001b; Cerdà
high aggradation rates in the early Holocene. and Lasanta, 2005; Shakesby et al., 2007; Smith and Drago-
vich, 2008; Cannon et al., 2010), whereas other topics are
lacking, including fire’s influence on weathering (Zimmerman
12.15.7 Geomorphic and Topographic Influences on et al., 1994; Allison and Bristow, 1999; Dorn, 2003; Shakesby
Fire and Doerr, 2006) and LWD (Bendix and Cowell, 2010).
Though literature on fire-prone areas is extensive, many fewer
Geomorphic and topographic features and processes may also studies are available on less fire-frequented areas (Gavin et al.,
influence fire behavior, especially with low- to moderate-inten- 2003). The literature presented in this chapter represents
sity fires. Debris flows and snow avalanches form linear land- a broad overview on available fire–geomorphic literature, but
scape features that extend perpendicular to the contour is by no means all inclusive. However, this sampling of the
of slopes and serve as breaks in vegetation (Butler, 2001). The literature reveals that fire–geomorphic interactions are quite
lack of fuel in these paths may act to stop, slowdown, or alter fire important to many fire-affected landscapes around the world.
movement. Unvegetated stream channels, which lack fuel, serve It also indicates that many gaps still remain in the under-
a similar function (Pettit and Naiman, 2007). Riparian zones standing of fire–geomorphic–organism interactions. The re-
may also form fire breaks due to their humid, moist conditions; sults of fire vary to such great extents, depending on local
however, in dry conditions, they may also function as fire cor- environmental conditions and fire severity, that studies are
ridors. The orientation of the breaks in regard to fire movement commonly required for many different locations under vari-
will strongly determine its effectiveness in stopping a fire. Debris ous conditions and fire intensities before one can begin to
flows, avalanche paths, or stream channels that extend per- understand fire’s effects on particular sites. Changes to fire
pendicular to the fire will be more effective compared to linear regimes and the potential effects that climate change may have
breaks that run parallel to the fire advancement. These breaks, on fire emphasize the importance of understanding fire’s ef-
however, will generally not stop high-severity or crown fires, fects on the landscape, and how changes to fire may sub-
which are sufficiently intense to jump breaks. sequently alter geomorphic dynamics.
Topographic features may also influence fire behavior.
Slope aspect and angle, mountains, valleys, leeward and
windward aspects, and how these features interact with wind
References
direction, moisture patterns, and vegetation growth may in-
fluence fire movement on a landscape. In the Northern
Abrahams, A.D., Parsons, A.J., Luk, S., 1988. Hydrologic and sediment responses
Hemisphere north of the Tropic of Cancer, southern and to simulated rainfall on desert hillslopes in southern Arizona. Catena 15,
western slopes are generally drier and therefore facilitate fire 103–117.
ignitions and more intense burning (Jackson and Roering, Adamson, D., Selkirk, P.M., Mitchell, P., 1983. The role of fire and lyre-birds in the
2009). North-facing slopes commonly contain more vege- landscape of the Sydney basin. In: Young, R.W., Nanson, G.L. (Eds.), Aspects of
the Australian Sandstone Landscapes. University of Wollongong, NSW, pp. 81–93.
tation and facilitate vegetation regrowth (Rebertus et al., 1991;
Agee, J.K., 1993. Fire Ecology of Pacific Northwest Forests. Island Press,
Stueve et al., 2009), whereas south-facing slopes have drier, Washington, DC.
less cohesive soils (Marques and Mora, 1992; Jackson and Allison, R.J., Bristow, G.E., 1999. The effects of fire on rock weathering: some
Roering, 2009). These factors can result in significantly higher further considerations of laboratory experimental simulation. Earth Surface
erosion rates on south- and west-facing slopes. Winds associ- Processes and Landforms 24, 707–713.
Allison, R.J., Higgitt, D.L., 1988. Slope form and associations with ground boulder
ated with mountains and valleys may carry fire up- or down- cover in arid environments, northeast Jordan. Catena 33, 47–74.
slope depending on the time of day, season, and location. Arocena, J.M., Opio, C., 2003. Prescribed fire-induced changes in properties of
sub-boreal forest soils. Geoderma 113, 1–16.
Aronica, G., Candela, A., Santoro, M., 2002. Changes in the hydrological response
of two Sicilian basins affected by fire. In: van Lanen, H.A.J., Demuth, S. (Eds.),
12.15.8 Conclusion FRIEND 2002 – Regional Hydrology: Bridging the Gap between Research and
Practice, IAHS Publication 274. IAHS Press, Wallingford, pp. 163–169.
Fire unequivocally serves as a geomorphic agent in many Atkinson, G., 1984. Erosion damage following brushfires. Journal of Soil
locations, particularly in fire-prone areas. Changes to soil Conservation Service NSW 40, 4–9.
properties, increases in erosion and weathering rates, and in- Baath, E., Arnebrant, K., 1994. Growth rate and response of bacterial communities
to pH in limed and ash treated forest soils. Soil Biology and Biochemistry 26,
fluences on hydrology may all be affected by fire. The rela- 995–1001.
tionship between geomorphology and vegetation acts as a Baisan, C.H., Swetnam, T.W., 1990. Fire history on a desert mountain range: Rincon
strong factor in the influence that fire exerts on geomorph- Mountain Wilderness, Arizona, USA. Canadian Journal of Forest Research 20,
ology. Vegetation serves as fuel for fire, which affects geo- 1559–1569.
Bales, R.C., Molotch, N.P., Painter, T.H., Dettinger, M.D., Rice, R., Dozier, J., 2006.
morphic processes and features with direct heat and
Mountain hydrology of the Western United States. Water Resources Research
vegetation loss. Disturbances to geomorphology subsequently 42, W08432. http://dx.doi.org/10.1029/2005WR004387.
influence vegetation regrowth, and inversely, vegetation re- Ballais, J.-L., Bosc, M.-C., 1994. The ignifracts of the Sainte-Victoire Mountain
establishment affects soil conditions, erosion processes, and (Lower Provence, France). In: Sala, M., Rubio, J.L. (Eds.), Soil Erosion and
hydraulic factors. The interactions among fire, geomorph- Degradation as a Consequence of Forest Fires. Geoforma Ediciones, Logroño,
pp. 217–227.
ology, and organisms are extensive, and literature on fire re- Bart, R., Hope, A., 2010. Streamflow response to fire in large catchments of a
flects the great influence that fire can exert on a location. With Mediterranean-climate region using paired-catchment experiments. Journal of
particular regard to fire–geomorphic interactions, some topics Hydrology 388, 370–378.
248 Fire as a Geomorphic Agent

Beeson, P.C., Martens, S.N., Breshears, D.D., 2001. Simulating overland flow Cerdà, A., Imeson, A.C., Calvo, A., 1995. Fire and aspect induced differences on
following wildfire: mapping vulnerability to landscape disturbance. Hydrological the erodibilty and hydrology of soils at La Costera, Valencia, southeast Spain.
Processes 15, 2917–2930. Catena 24, 289–304.
Belillas, C.M., Roda, F., 1993. The effects of fire on water quality, dissolved nutrient Cerdà, A., Lasanta, T., 2005. Long-term erosional responses after fire in the Central
losses and the export of particulate matter from dry heathland catchments. Spanish Pyrenees: 1. Water and sediment yield. Catena 60, 59–80.
Journal of Hydrology 150, 1–17. Certini, G., 2005. Effects of fire on properties of forest soils: a review. Oecologia
Benda, L., Miller, D., Bigelow, P., Andras, K., 2003. Effects of post-wildfire erosion 143, 1–10.
on channel environments, Boise River, Idaho. Forest Ecology and Management Chambers, D.P., Attiwill, P.M., 1994. The ash-bed effect in Eucalyptus regnans
178, 105–119. forest: chemical, physical and microbiological changes in soil after heating or
Bendix, J., Cowell, C.M., 2010. Fire, floods and woody debris: interactions between partial sterilization. Australian Journal of Botany 42, 739–749.
biotic and geomorphic processes. Geomorphology 116, 297–304. Chandler, C., Cheney, P., Thomas, P., Trabaud, L., Williams, D., 1983. Fire in
Blackwelder, E., 1927. Fire as an agent in rock weathering. Journal of Geology 35, Forestry. Volume 1, Forest Fire Behavior and Effects, Wiley, New York, NY.
134–140. Chapman, S.K., Langley, J.A., Hart, S.C., Koch, G.W., 2006. Plants actively control
Bosch, J.M., Schulze, R.E., Kruger, F.J., 1984. The effect of fire on water yield. nitrogen cycling: uncorking the microbial bottleneck. New Phytologist 169, 27–34.
In: Booysen, P.D., Tainton, N.M. (Eds.), Ecological Studies 48: Ecological Effects Christensen, N.L., Agee, J.K., Brussard, P.F., Hughes, J., Knight, D.H., Minshall,
of Fire in South African Ecosystems. Springer, New York, pp. 327–348. G.W., Peek, J.M., Pyne, S.J., Swanson, F.J., Thomas, J.W., Wells, S., Williams,
Bracken, L.J., Kirkby, M.J., 2005. Differences in hillslope runoff and sediment S.E., Wright, H.A., 1989. Interpreting the Yellowstone Fires of 1988. BioScience
transport rates within two semi-arid catchments in southeast Spain. 39(10), 678–685.
Geomorphology 68, 183–200. Cook, G.D., 1994. The fate of nutrients during fires in a tropical savanna. Australian
Bragg, D.C., 2000. Simulating catastrophic and individual large woody debris Journal of Ecology 19, 359–365.
recruitment for a small riparian system. Ecology 81, 1383–1394. Crockford, S., Topalidis, S., Richardson, D.P., 1991. Water repellency in a dry
Britton, D.L., 1991. Fire and the chemistry of a South African mountain stream. sclerophyll forest – measurements and processes. Hydrological Processes 5,
Hydrobiologia 218, 177–192. 405–420.
Brown, A.G., 1990. Soil erosion and fire in areas of Mediterranean type vegetation: Davenport, D.W., Breshears, D.D., Wilcox, B.P., Allen, C.D., 1998. Viewpoint:
results from chaparral in southern California, USA and Matorral in Andalucia, sustainability of piñon–juniper ecosystems: a unifying perspective of soil
southern Spain. In: Thornes, J.B. (Ed.), Vegetation and Erosion, Processes and erosion thresholds. Journal of Range Management 51, 231–240.
Environments. Wiley, West Sussex, pp. 269–287. DeBano, L.F., 2000. The role of fire and soil heating on water repellency in
Brown, J.A.H., 1972. Hydrologic effects of a bushfire in a catchment in wildland environments: a review. Journal of Hydrology 231–232, 195–206.
southeastern New South Wales. Journal of Hydrology 15, 77–96. DeBano, L.F., Conrad, C.E., 1978. The effects of fire on nutrients in a chaparral
Burch, G.S., Doerr, S.H., Burns, J., 1989. Soil hydrophobic effects on infiltration ecosystem. Ecology 59(3), 489–497.
and catchment runoff. Hydrological Processes 3, 211–222. DeBano, L.F., Neary, D.G., Ffolliott, P.F., 1998. Fire’s Effects on Ecosystems. Wiley,
Busenberg, G., 2004. Wildfire management in the United States: the evolution of a New York, NY.
policy failure. Review of Policy Research 21, 145–156. DeBano, L.F., Savage, S.M., Hamilton, D.A., 1976. The transfer of heat and
Butler, D.R., 2001. Geomorphic process-disturbance corridors: a variation on a hydrophobic substances during burning. Proceedings – Soil Science Society of
principle of landscape ecology. Progress in Physical Geography 25, 237–248. America 40, 779–782.
Byram, G.M., 1959. Combustion of forest fuels. In: Davis, K.P. (Ed.), Forest Fire: Diaz-Fierros, F., Benito Rueda, E., Perez Moreira, R., 1987. Evaluation of the
Control and Use. McGraw-Hill, New York, pp. 61–89. U.S.L.E. for the prediction of erosion in burnt forest areas in Galicia – N.W.
Campbell, R.E., Baker, M.B., Jr., Ffolliott, P.F., Larson, F.R., Avery, C.C., 1977. Spain. Catena 14, 189–199.
Wildfire effects on a ponderosa pine ecosystem: an Arizona case study. RM-191. Doerr, S.H., Blake, W.H., Humphreys, G.S., Shakesby, R.A., Stagnitti, F., Vuurens,
USDA Forest Service Rocky Mountain Range Experimental Station, Fort Collins, S.H., Wallbrink, P., 1994. Heating effects on water repellency in Australian
CO, 12 pp. eucalypt forest soils and their value in estimating wildfire soil temperatures.
Cannon, S.H., 1997. Evaluation of the potential for debris and hyperconcentrated International Journal of Wildland Fire 13, 157–163.
flows in Capulin Canyon as a result of the 1996 Dome Fire, Bandelier National Doerr, S.H., Blake, W.H., Humphreys, G.S., Shakesby, R.A., Stagnitti, F., Vuurens,
Monument, New Mexico. U.S. Geological Survey, Denver, CO, pp. 97–136. S.H., Wallbrink, P., 2004. Heating effects on water repellency in Australian
Cannon, S.H., 2001. Debris-flow generation from recently burned watersheds. eucalypt forest soils and their value in estimating wildfire soil temperatures.
Environmental and Engineering Geoscience 7(4), 321–341. International Journal of Wildland Fire 13, 157–163.
Cannon, S.H., Bigio, E.R., Mine, E., 2001a. A process for fire-related debris flow Doerr, S.H., Ferreira, A.J.D., Walsh, R.P.D., Shakesby, R.A., Leighton-Boyce, G.,
initiation, Cerro Grande fire, New Mexico. Hydrological Processes 15, Coelho, C.O.A., 2003. Soil and water repellency as a potential parameter in
3011–3023. rainfall-runoff modeling: experimental evidence at a point to catchment scales
Cannon, S.H., Gartner, J.E., Rupert, M.G., Michael, J.A., Rea, A.H., Parrett, C., from Portugal. Hydrological Processes 17, 363–377.
2010. Predicting the probability and volume of postwildfire debris flows in the Doerr, S.H., Shakesby, R.A., Walsh, R.P.D., 2000. Soil and water repellency, its
intermountain western United States. Geological Society of America Bulletin 122, characteristics, causes and hydro-geomorphological consequences. Earth-
127–144. http://dx.doi.org/10.1130/B26459.1. Science Reviews 51, 33–65.
Cannon, S.H., Gartner, M.G., 2005. Wildfire-related debris flow from a hazards Doerr, S.H., Thomas, A.D., 2000. The role of soil moisture in controlling water
perspective. In: Hungr, O., Jacob, M. (Eds.), Debris Flows and Debris repellency: new evidence from forest soils in Portugal. Journal of Hydrology
Avalanches – The State of the Art. Springer-Praxis Books in Geophysical 231–232, 134–147.
Sciences, New York, pp. 321–344. Dorn, R.I., 2003. Boulder weathering and erosion associated with a wildfire, Sierra
Cannon, S.H., Kirkham, R.M., Parise, M., 2001b. Wildfire-related debris-flow initiation Ancha Mountains, Arizona. Geomorphology 55, 155–171.
processes, Storm King Mountain, Colorado. Geomorphology 39, 171–188. Dragovich, D., 1993. Fire-accelerated weathering in the Pilbara, Western Australia.
Cannon, S.H., Powers, P.S., Savage, W.Z., 1998. Fire-related hyperconcentrated and Zeitschrift für Geomorphologie 37, 295–307.
debris flows on Storm King Mountain, Glenwood Springs, Colorado, USA. Dragovich, D., Morris, R., 2002. Fire intensity, slopewash and biotransfer of
Environmental Geology 35, 210–218. sediment in eucalypt forest. Australia. Earth Surface Processes and Landforms
Cannon, S.H., Reneau, S.L., 2000. Conditions for generation of fire-related debris 27, 1309–1319.
flows, Capulin Canyon, New Mexico. Earth Surface Processes and Landforms Dwire, K.A., Kauffman, J.B., 2003. Fire and riparian ecosystems in landscapes of
25, 1103–1121. the western USA. Forest Ecology and Management 178, 61–74.
Cerdà, A., 1998a. Changes in overland flow and infiltration after a rangeland fire in Dyrness, C.T., 1976. Effect of wildfire on soil wettability in the High Cascades of
a Mediterranean schrubland. Hydrological Processes 12, 1031–1042. Oregon. Research Paper PNW-202. USDA Forest Service, Pacific Northwest
Cerdà, A., 1998b. Relationships between climate and soil hydrological and Forest and Range Exp. Sta. Portland, OR, 18 pp.
erosional characteristics along climate gradients in Mediterranean limestone Fagre, D.B., 2003. Taking the pulse of the mountains: ecosystem responses to
areas. Geomorphology 25, 123–134. climatic variability. Climatic Change 59, 263–282.
Cerdà, A., Doerr, S.H., 2005. The influence of vegetation recovery on soil hydrology Faustini, J.M., Jones, J.A., 2003. Influence of large woody debris on channel
and erodibility following fire: an eleven-year investigation. International Journal morphology and dynamics in steep, boulder-rich mountain streams, western
of Wildland Fire 14, 423–437. Cascades, Oregon. Geomorphology 51, 187–205.
 Fire as a Geomorphic Agent 249

Fernández, I., Cabaneiro, A., Carballas, T., 1999. Carbon mineralization dynamics in King, P.M., 1981. Comparison of methods for measuring severity of water
soils after wildfires in two Galician forests. Soil Biology and Biochemistry 31, repellence of sandy soils and assessment of some factors that affect its
1853–1865. measurement. Australian Journal of Soil Research 19, 275–285.
Ferreira, A.J.D., Coelho, C.O.A., Walsh, R.P.D., Shakesby, R.A., Ceballos, A., Klopatek, J.M., 1987. Nitrogen mineralization and nitrification in mineral soils of
Doerr, S.H., 2000. Hydrological implications of soil water repellency in pinyon-juniper ecosystems. Soil Science Society of America Journal 51,
Eucalyptus globulus forests, north-central Portugal. Journal of Hydrology 453–457.
231–232, 165–177. Knicker, H., 2007. How does fire affect the nature and stability of soil organic
Florsheim, J.L., Keller, E.A., Best, D.W., 1991. Fluvial sediment transport in nitrogen and carbon? A review. Biogeochemistry 85, 91–118.
response to moderate storm flows following chaparral wildfire, Ventura County, Kutiel, P., Inbar, M., 1993. Fire impacts on soil nutrients and soil erosion in a
southern California. Geological Society of America Bulletin 103, 504–511. Mediterranean pine forest plantation. Catena 20, 129–139.
Fryrear, D.W., 1995. Soil losses by wind erosion. Soil Science Society of America Laird, J.R., Harvey, M.D., 1986. Complex response of a Chaparral Drainage basin
Journal 59, 668–672. to fire. In: Hadley, R.F. (Ed.), Drainage Basin Sediment Delivery, Volume 159.
Fulé, P.Z., Covington, W.W., Moore, M.M., 1997. Determining reference conditions IAHS Publication, IAHS Press, Wallingford, pp. 165–183.
for ecosystem management of southwestern ponderosa pine. Ecological Larsen, I.J., Pederson, J.L., Schmidt, J.C., 2006. Geologic versus wildfire controls
Applications 7, 895–908. on hillslope processes and debris flow initiation in the Green River canyons of
Gabet, E.J., Bookter, A., 2008. A morphometric analysis of gullies scoured by post- Dinosaur National Monument. Geomorphology 81, 114–127.
fire progressively bulked debris flows in southwest Montana, USA. Lavabre, J., Torres, S., Cernesson, F., 1993. Changes in the hydrological response
Geomorphology 96, 298–309. of a small Mediterranean basin a year after a wildfire. Journal of Hydrology 142,
Gavin, D.G., Brubaker, L.B., Lertzman, K.P., 2003. Holocene fire history of a coastal 273–299.
temperate rain forest based on soil charcoal radiocarbon dates. Ecology 84, Lavee, H., Kutiel, P., Segev, M., Benyamini, Y., 1995. Effect of surface roughness on
186–201. runoff and erosion in a Mediterranean ecosystem: the role of fire.
Glenn, N.F., Finley, C.D., 2010. Fire and vegetation type effects on soil Geomorphology 11, 227–234.
hydrophobicity and infiltration in the sagebrush-steppe: I. Field analysis. Journal Leitch, C.J., Flinn, D.W., van de Graaff, R.H.M., 1983. Erosion and nutrient loss
of Arid Environments 74, 653–659. resulting from Ash Wednesday (February 1983) wildfires: a case study.
Gonzàlez-Pelayo, O., Andreu, V., Campo, J., Gimeno-Garcı́a, E., Rubio, J.L., 2006. Australian Forestry 46, 173–180.
Hydrological properties of a Mediterranean soil burned with different fire Lenihan, J., Daly, C., Bachelet, D., Neilson, R.P., 1998. Simulating broad-scale fire
intensities. Catena 68, 168–193. severity in a dynamic global vegetation model. Northwest Science 72, 91–103.
Goudie, A.S., Allison, R.J., McLaren, S.J., 1992. The relations between modulus Letey, J., 2001. Causes and consequences of fire-induced soil water repellency.
of elasticity and temperature in the context of the experimental simulation Hydrological Processes 15, 2867–2875.
of rock weathering by fire. Earth Surface Processes and Landforms 17, Lewis, S.A., Wu, J.Q., Robichaud, P.R., 2006. Assessing burn severity and
605–615. comparing soil water repellency, Hayman Fire, Colorado. Hydrological Processes
Gresswell, R.E., 1999. Fire and aquatic ecosystems in forested biomes of North 20, 1–16.
America. Transactions of the American Fisheries Society 128, 193–221. Lewis, W.M., 1974. Effects of fire on nutrient movement in a South Caroline Pine
Grissino-Mayer, H.D., 1995. Tree-ring reconstructions of climate and fire history at Forest. Ecology 55, 1120–1127.
El Malpais National Monument, New Mexico. Ph.D. Dissertation, University of Lindley, A.J., Bosch, J., van Wyk, D.B., 1988. Changes in water yield after fire in
Arizona, Tucson. fynbos catchments. Water SA 14, 7–12.
Gurnell, A.M., Piegay, H., Swanson, F.J., Gregory, S.V., 2002. Large wood and Loáiciga, H.A., Pedreros, D., Roberts, D., 2001. Wildfire-streamflow interactions in a
fluvial processes. Freshwater Biology 47(4), 601–619. chaparral watershed. Advances in Environmental Research 5, 295–305.
Helvey, J.D., 1980. Effects of a north central Washington wildfire on runoff and Long, C.J., Whitlock, C., Bartlein, P.J., Millspaugh, S.H., 1998. A 9000-year fire
sediment production. Water Resources Bulletin 16, 627–634. history from the Oregon Coast Range, based on a high-resolution charcoal
Hessling, M., 1999. Hydrological modelling and a pair basin study of study. Canadian Journal of Forest Research 28, 774–787.
Mediterranean catchments. Physics and Chemistry of the Earth, Part B: MacDonald, L.H., Huffman, E.L., 2004. Post-fire soil water repellency: persistence
Hydrology, Oceans and Atmosphere 24, 59–63. and soil moisture thresholds. Journal of Soil Science Society of America 68,
Hiemstra, C.A., Liston, G.E., Reiners, W.A., 2002. Snow redistribution by wind and 1729–1734.
interactions with vegetation at upper Treeline in the Medicine Bow Mountains, Marlon, J.R., Bartlein, P.J., Walsh, M.K., Harrison, S.P., Brown, K.J., Edwards, M.E.,
Wyoming, U.S.A. Arctic, Antarctic, and Alpine Research 34, 262–273. Higuera, P.E., Power, M.J., Anderson, R.S., Briles, C., Brunelle, A., Carcaillet, C.,
Huffman, E.L., MacDonald, L.H., Stednick, J.D., 2001. Strength and persistence of Daniels, M., Hu, F.S., Lavoie, M., Long, C., Minckley, T., Richard, P.J.H., Scott,
fire-induced soil hydrophobicity under ponderosa and lodgepole pine, Colorado A.C., Shafer, D.S., Tinner, W., Umbanhowa, C.E., Jr., Whitlock, C., 2009. Wildfire
Front Range. Hydrological Processes 15, 2877–2892. responses to abrupt climate change in North America. Proceedings of the
Humphreys, G.S., Shakesby, R.A., Doerr, S.H., Blake, W.H., Wallbrink, P., Hart, National Academy of Sciences 106, 2519–2524.
D.M., 2003. Some effects of fire on the regolith. In: Roach, I.C. (Ed.), Advances Marques, M.A., Mora, E., 1992. The influence of aspect on runoff and soil loss in
in Regolith. CRC, LEME, ACT, pp. 216–220. a Mediterranean burnt forest. Catena 19, 333–344.
Imeson, A.C., Verstraten, J.M., van Mulligen, E.J., Sevink, J., 1992. The effects of Marston, R.A., Haire, D.H., 1990. Runoff and soil loss following the 1988
fire and water repellency on infiltration and runoff under Mediterranean type Yellowstone fires. Great Plains – Rocky Mountain Geographical Journal 18,
forest. Catena 19, 345–361. 1–8.
Inbar, M., Tamir, M., Wittenburg, L., 1998. Runoff and erosion processes after a Martin, D.A., Moody, J.A., 2001. Comparison of soil infiltration rates in burned and
forest fire in Mount Carmel, a Mediterranean area. Geomorphology 24, 17–33. unburned mountainous watersheds. Hydrological Processes 15, 2893–2903.
Jackson, M., Roering, J.J., 2009. Post-fire geomorphic response in steep, forested May, C.L., Gresswell, R.E., 2003. Processes and rates of sediment and wood
landscapes: Oregon Coast Range, USA. Quaternary Science Reviews 28, accumulation in headwater streams of the Oregon Coast Range, USA. Earth
1131–1146. Surface Processes and Landforms 28, 409–424.
Johansen, M.P., Hakonson, T.E., Breshears, D.D., 2001. Post-fire runoff and erosion Mayor, A.G., Bautista, S., Llovet, J., Bellot, J., 2007. Post-fire hydrological and
from rainfall simulation: contrasting forests with shrublands and grasslands. erosional responses of a Mediterranean landscape: seven years of catchment-
Hydrological Processes 15, 2953–2965. scale dynamics. Catena 71, 68–75.
Johnson, D.W., Curtis, P.S., 2001. Effects of forest management on soil C and N McKenzie, D., Gedalof, Z., Peterson, D.L., Mote, P., 2004. Climate change, wildfire,
storage: meta analysis. Forest Ecology and Management 140, 227–238. and conservation. Conservation Biology 18, 890–902.
Jung, H.Y., Hogue, T.S., Rademacher, L.K., Meixner, T., 2009. Impact of wildfire on Mersereau, R.C., Dyrness, C.T., 1972. Accelerated mass wasting after logging and
source water contributions in Devil Creek, CA: evidence from end-member slash burning in western Oregon. Journal of Soil and Water Conservation 27,
mixing analysis. Hydrological Processes 23, 183–200. 112–114.
Keller, E.A., Swanson, F.J., 1979. Effects of large organic material on channel form Meyer, G.A., Pierce, J.L., Wood, S.H., Jull, A.J.T., 2001. Fire, storms, and erosional
and fluvial processes. Earth Surface Processes 4, 361–380. events in the Idaho batholith. Hydrological Processes 15, 3025–3038.
Keller, E.A., Valentine, D.W., Gibbs, D.R., 1997. Hydrological response of small Meyer, G.A., Wells, S.G., 1997. Fire-related sedimentation events on alluvial fans,
watersheds following the southern California painted cave fire of June 1990. Yellowstone National Park, U.S.A. Journal of Sedimentary Research 67,
Hydrological Processes 11(4), 401–414. 776–791.
250 Fire as a Geomorphic Agent

Meyer, G.A., Wells, S.G., Jull, A.J.T., 1995. Fire and alluvial chronology in from forested watersheds. Canadian Journal of Fisheries and Aquatic Sciences
Yellowstone National Park: climatic and intrinsic controls on Holocene 37, 328–334.
geomorphic processes. Geological Society of America Bulletin 107(10), Schoennagel, T., Veblen, T.T., Romme, W.H., 2004. The interaction of fire, fuels, and
1211–1230. climate across Rocky Mountain forests. BioScience 54, 661–676.
Montgomery, D.R., Piégay, H., 2003. Wood in rivers: interactions with channel Schumm, S.A., 1956. Evolution of drainage systems and slopes in badlands at
morphology and processes. Geomorphology 51, 1–5. Perth Amboy, New Jersey. Bulletin of the Geological Society of America 67,
Moody, J.A., Martin, D.A., 2001a. Post-fire, rainfall intensity-peak discharge 597–646.
relations for three mountainous watersheds in the western USA. Hydrological Scott, D.F., 1993. The hydrological effects of fire in South African mountain
Processes 15, 2981–2993. catchments. Journal of Hydrology 150, 409–432.
Moody, J.A., Martin, D.A., 2001b. Initial hydrologic and geomorphic response Scott, D.F., 1997. The contrasting effects of wildfire and clearfelling on the
following a wildfire in the Colorado Front Range. Earth Surface Processes and hydrology of a small catchment. Hydrological Processes 11, 543–555.
Landforms 26, 1049–1070. Shakesby, R.A., Chafer, C.J., Doerr, S.H., Blake, W.H., Humphreys, G.S., Wallbrink,
Moreno, J.M., Vázquez, A., Vélez, R., 1998. Recent history of forest fires in Spain. P., Harrington, B.H., 2003. Fire severity, water repellency characteristics and
In: Moreno, J.M. (Ed.), Large Fires. Backhuys Publishers, Leiden, The hydrogeomorphological changes following the Christmas 2001 forest fires.
Netherlands, pp. 159–185. Australian Geographer 34, 147–175.
Morris, S.E., Moses, T.A., 1987. Forest fire and the natural soil erosion regime in Shakesby, R.A., Coelho, C.de.O.A., Ferreira, A.D., Terry, J.P., Walsh, R.P.D., 1993.
the Colorado Front Range. Annals of the Association of American Geographers Wildfire impacts on soil erosion and hydrology in wet Mediterranean forest,
77, 245–254. Portugal. International Journal of Wildland Fire 3, 95–110.
Munns, E.N., 1920. Chaparral cover, run-off, and erosion. Journal of Forestry 18, Shakesby, R.A., Doerr, S.H., 2006. Wildfire as a hydrological and geomorphogical
806–814. agent. Earth-Science Reviews 74, 269–307.
Neary, D.G., Klopatek, C.C., DeBano, L.F., Ffolliott, P.F., 1999. Fire effects on Shakesby, R.A., Doerr, S.H., Walsh, R.P.D., 2000. The erosional impact of soil
belowground sustainability: a review and synthesis. Forest Ecology and hydrophobicity: current problems and future research directions. Journal of
Management 122, 51–71. Hydrology 231–232, 178–191.
Odion, D.C., Davis, F.W., 2000. Fire, soil heating, and the formation of vegetation Shakesby, R.A., Wallbrink, P.J., Doerr, S.H., et al., 2007. Distinctiveness of wildfire
patterns in chaparral. Ecological Monographs 70, 149–169. effects on soil erosion in south-east Australian eucalypt forests assessed in a
Pausas, J.G., 2004. Changes in fire and climate in the eastern Iberian Peninsula global context. Forest Ecology and Management 238, 347–364.
(Mediterranean Basin). Climate Change 63, 337–350. Sheridan, G.J., Lane, P.N.J., Noske, P.J., 2007. Quantification of hillslope runoff and
Pausas, J.G., Vallejo, R., 1999. The role of fire in European mediterranean erosion processes before and after wildfire in a wet Eucalyptus forest. Journal of
ecosystems. In: Chuvieco, E. (Ed.), Remote Sensing of Large Wildfires in the Hydrology 343, 12–28.
European Mediterranean Basin. Springer, Berlin, pp. 3–16. Smith, H.G., Dragovich, D., 2008. Post-fire hillslope erosion response in a sub-
Personius, S.F., Kelsey, H.M., Grabau, P.C., 1993. Evidence for regional stream alpine environment, south-eastern Australia. Catena 73, 274–285.
aggradation in the central Oregon Coast Range during the Pleistocene–Holocene Spigel, K.M., Robichaud, P.R., 2007. First-year post-fire erosion rates in Bitterroot
transition. Quaternary Research 40, 297–308. National Forest, Montana. Hydrological Processes 21, 998–1005.
Pettit, N.E., Naiman, R.J., 2007. Fire in the riparian zone: characteristics and Spittler, T.E., 1995. Fire and debris-flow potential of winter storms. In: Keeley, J.E.,
ecological consequences. Ecosystems 10, 673–687. Scott, T. (Eds.), Proceedings of the Brush Fires in California Wildlands, Ecology
Price, C., Rind, D., 1994. The impact of a 2  CO2 climate on lightning-caused and Resource Management. International Association of Wildland Fire, Fairfield,
fire. Journal of Climate 7, 1484–1494. Washington, pp. 113–119.
Prosser, I.P., Williams, L., 1998. The effect of wildfire on runoff and erosion in Stout, J.E., 2001. Dust and environment in the Southern High Plains of North
native Eucalyptus forest. Hydrological Processes 12, 251–265. America. Journal of Arid Environments 47, 425–441.
Pyne, S.J., 2001. Fire, a Brief History. University of Washington Press, Seattle, WA. Stueve, K.M., Cerney, D.L., Rochefort, R.M., Kurth, L.L., 2009. Post-fire
Ravi, S., D’Odorico, P., Herbert, B.E., Zobeck, T.M., Over, T.M., 2006. Enhancement tree establishment patterns at the alpine treeline ecotone: Mount Rainer
of wind erosion by fire-induced water repellency. Water Resources Research National Park, Washington, USA. Journal of Vegetation Science 20,
42(11), W11422. http://dx.doi.org/10.1029/2006WR004895. 107–120.
Ravi, S., D’Odorico, P., Zobeck, T.M., Over, T.M., 2009. The effect of fire-induced Swanson, F.J., 1978. Fire and geomorphic processes. Proceedings of the Fire
soil hydrophobicity on wind erosion in a semiarid grassland: experimental Regimes and Ecosystems Conference, 1979 December 11–15. Honolulu, HI.
observations and theoretical framework. Geomorphology 105, 80–86. Gen. Tech. Rep. WO-23. Washington, DC. U.S. Department of Agriculture, Forest
Rebertus, A.J., Burns, B.R., Veblen, T.T., 1991. Stand dynamics of Pinus flexilis Service, June 1981.
dominated subalpine forests in the Colorado Front Range. Journal of Vegetation Thornes, J.B., 1990. The interaction of erosional and vegetation dynamics in land
Science 2, 445–458. degradation: spatial outcomes. In: Thornes, J.B. (Ed.), Vegetation and Erosion,
Reneau, S.L., Dietrich, W.E., Donahue, D.J., Jull, A.J.T., Rubin, M., 1990. Late Processes and Environments. Wiley, West Sussex, pp. 125–144.
Quaternary history of colluvial deposition and erosion in hollows, Central Tiedemann, A.R., Helvey, J.D., Anderson, T.D., 1978. Stream chemistry and
California Coast Ranges. Geological Society of America Bulletin 102, watershed nutrient economy following wildfire and fertilization in eastern
969–982. Washington. Journal of Environmental Quality 7, 580–588.
Rice, R.M., Corbett, E.S., Bailey, R.G., 1969. Soil slips related to vegetation, Townsend, S.A., Douglas, M.M., 2000. The effect of three fire regimes on stream
topography, and soil in Southern California. Water Resources Research 5, water quality, water yield and export coefficients in a tropical savanna (northern
637–659. Australia). Journal of Hydrology 229, 118–137.
Robichaud, P.R., 2000. Fire effects on infiltration rates after prescribed fire in Northern Turner, M.G., Smithwick, E.A.H., Metzger, K.L., Tinker, D.B., Romme, W.H., 2007.
Rocky Mountain forests. USA. Journal of Hydrology 231–232, 220–229. Inorganic nitrogen availability after severe stand-replacing fire in the Greater
Roering, J.J., Gerber, M., 2005. Fire and the evolution of steep, soil-mantled Yellowstone ecosystem. Proceedings of the National Academy of Sciences 104,
landscapes. Geology 33, 349–352. 4782–4789.
Ronan, N.M., 1986. The hydrological effects of fuel reduction burning and wildfire Ulery, A.L., Graham, R.C., 1993. Forest fire effects on soil color and texture. Soil
at wildfire at Wallaby Creek. Report for the Melbourne Metropolitan Board of Science Society of America Journal 57, 135–140.
Works Report Number MMBW-W-0015, pp. 229. Ulery, A.L., Graham, R.C., Amrhein, C., 1993. Wood-ash composition and soil pH
Rubio, J.L., Forteza, J., Andreu, V., Cernı́, R., 1997. Soil profile characteristics following intense burning. Soil Science 156, 358–364.
influencing runoff and soil erosion after forest fire: a case of study (Valencia, van der Beek, P., Pulford, A., Braun, J., 2001. Cenozoic landscape development in
Spain). Soil Technology 11, 67–78. the Blue Mountains (SE Australia): lithological and tectonic controls on rifted
Running, S.W., 2006. Is global warming causing more, larger wildfires? Science margin morphology. Journal of Geology 109, 35–56.
313, 927–928. van Mantgem, P.J., Stephenson, N.L., Byrne, J.C., Daniels, L.D., Franklin, J.F., Fulé,
Sass, O., Heel, M., Hoinkis, R., Wetzel, K.-F., 2010. A six-year record of debris P.Z., Harmon, M.E., Larson, A.J., Smith, J.M., Taylor, A.H., Veblen, T.T., 2009.
transport by avalanches on a wildfire slope (Arnspitze, Tyrol). Zeitschrift für Widespread increase of tree mortality rates in the western United States. Science
Geomorphologie 54, 181–193. 323, 521–524.
Schindler, D.W., Newbury, R.W., Beaty, K.G., Prokopowich, J., Ruszcznski, T., Varela, M.E., Benito, E., de Blas, E., 2005. Impacts of wildfires on surface water
Dalton, J.A., 1980. Effects of a windstorm and forest fire on chemical losses repellency in soils of northwest Spain. Hydrological Processes 19, 3649–3657.
 Fire as a Geomorphic Agent 251

Vermeire, L.T., Wester, D.B., Mitchell, R.B., Fuhlendorf, S.D., 2005. Fire and grazing Wohl, E.E., Pearthree, P.P., 1991. Debris flows as geomorphic agents in the
effects on wind erosion, soil water content, and soil temperature. Journal of Huachuca Mountains of southeastern Arizona. Geomorphology 4, 273–292.
Environmental Quality 34, 1559–1565. Wondzell, S.M., King, J.G., 2003. Postfire erosional processes in the Pacific
Wells, C.G., II, 1971. Effects of prescribed burning on soil chemical properties and Northwest and Rocky Mountain regions. Forest Ecology and Management 178,
nutrient availability. Prescribed burning symposium, Proceedings of the 75–87.
Southeastern Forest Experimental Station, Charleston, South Carolina, USA, Woods, S.W., Birkas, A., Ahl, R., 2006. Spatial variability of soil
April 14–17. pp. 86–99. hydrophobicity after wildfires in Montana and Colorado. Geomorphology 86,
Wells, W.G., II, 1987. The effects of fire on the generation of debris flows in 645–679.
southern California. Reviews in Engineering Geology 2, 105–114. Zimmerman, S.G., Evenson, E.B., Gosse, J.C., Erskine, C.P., 1994. Extensive
Westerling, A.L., Hidalgo, H.G., Cayan, D.R., Swetnam, T.W., 2006. Warming and boulder erosion resulting from a range fire on the type-Pinedale moraines,
earlier spring increase western U.S. forest wildfire activity. Science 313, 940–943. Fremont Lake, Wyoming. Quaternary Research 42, 255–265.
Wheelan, R.J., 1995. The Ecology of Fire. Cambridge University Press, Cambridge. Zobeck, T.M., Fryrear, D.W., Pettit, R.D., 1989. Management effects on wind-eroded
Whicker, J.J., Breshears, D.D., Wasiolek, P.T., Kirschner, T.B., Tavani, R.A., Schoep, sediment and plant nutrients. Journal of Soil and Water Conservation 44,
D.A., Rodgers, J.C., 2002. Temporal and spatial variation of episodic wind 160–163.
erosion in unburned and burned semiarid shrubland. Journal of Environmental
Quality 31, 599–612.

Biographical Sketch

Melanie Stine is a PhD student in geography at Texas State University, San Marcos. Her research interests are
biogeomorphology, mountain environments, vegetation and geomorphic response following disturbance, and
aquatic ecology and stream geomorphology. Melanie received her BS in environmental science from Sweet Briar
College and her MS in geography from Virginia Tech.
 12.16 The Faunal Influence: Geomorphic Form and Process
DR Butler, CJ Whitesides, and SG Tsikalas, Texas State University – San Marcos, San Marcos, TX, USA
r 2013 Elsevier Inc. All rights reserved.

12.16.1 Introduction 252

12.16.2 Categories of Geomorphic Impacts by Animals 253
12.16.2.1 Trampling and Loading 253
12.16.2.2 Digging 255
12.16.2.3 Burrowing 257
12.16.2.4 Beaver Damming 257
12.16.3 Geomorphic Impacts of Domesticated and Feral Animals 257
12.16.4 Zoogeomorphology at Ecotones 258
12.16.5 Conclusion 258
References 259

Glossary Lithophagy The eating/ingestion of rocks.

Ecotone A transition area of vegetation between two Phytogeomorphology The study of the geomorphic
different plant communities, such as forest and grassland, effects of plants.
or forest and tundra. Zoogeomorphology The study of the geomorphic effects
Geophagy The eating/ingestion of soil. of animals.

Abstract

Zoogeomorphology is the study of the geomorphic effects of animals, ranging from small invertebrates to large vertebrates
such as elephants and bison. It was not until the late twentieth century that geomorphologists began examining
the geomorphological activities of, and resultant landform features created by, animals. Animals engage in a variety of
geomorphic activities including trampling, loading, digging/foraging for food, geophagy, lithophagy, nest building, bur-
rowing, mound building, and damming of streams by beaver. These activities have both direct and indirect geomorphic
impacts, as well as impacts on vegetation density, species richness, and distribution that, in turn, also have geomorphic
influences. Feral animals and introduced species impact landscapes in much the same way as do free-ranging, native
populations. We suggest that the zoogeomorphological effects of animals may be fruitfully examined at ecotones, where
animal actions seem particularly notable and measurable.

12.16.1 Introduction coming from the field of wildlife management (e.g., Beyer
et al., 1994).
Zoogeomorphology is the study of the geomorphic effects of Why, however, had the role of animals (the ‘faunal influ-
animals (Butler, 1992), including both invertebrates and ver- ence’) as agents of landscape development and change been
tebrates (both ectothermic, i.e., ‘cold-blooded’, and endo- essentially ignored up to the late twentieth century? Butler
thermic, i.e., ‘warm-blooded’). The field of zoogeomorphology (1995) stated that traditional geomorphology textbooks fo-
has expanded rapidly since the mid-1990s, with the publi- cused on established geomorphic topics such as weathering,
cation of Butler’s (1995) book Zoogeomorphology – Animals as volcanism, and fluvial, glacial, and other surface processes but
Geomorphic Agent and many subsequent studies from around examined these processes in a world largely devoid of life.
the world. At roughly the same time, the study of ecosystem A recent study by Stine and Butler (2011) documented the
engineering was being defined by Jones and associates in the dearth of biogeomorphology in geomorphology textbooks
field of ecology (Jones et al., 1994), with additional efforts in from throughout the twentieth and early twenty-first centuries,
examining the role of animals as agents of landscape change and even within the very limited amount of biogeomorphic
material presented in those books, less (to none) of the ma-
terial was devoted to the role of animals as geomorphic agents.
Viles (1988) stated that the lack of geomorphic studies asso-
Butler, D.R., Whitesides, C.J., Tsikalas, S.G., 2013. The faunal influence:
ciated with biota was due to the influence and persistence of
geomorphic form and process. In: Shroder, J. (Editors in chief), Butler, D.R.,
Hupp, C.R. (Eds.), Treatise on Geomorphology. Academic Press, San Diego, CA, William Morris Davis’s geographical cycle (Davis, 1899), with
vol. 12, Ecogeomorphology, pp. 252–260. its focus on landscape scales of landform development

252 Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00332-8

 The Faunal Influence: Geomorphic Form and Process 253

through time. Another probable reason for the historical ab- influences, the effects of burrowing by animals and the geo-
sence of zoogeomorphological studies was noted by Butler morphic and hydrological role of beavers on the landscape.
(1995), who indicated that many geomorphologists are
trained in the fields of Earth science and have a poor back-
ground in the biological sciences. Butler believed that this 12.16.2 Categories of Geomorphic Impacts by
limited background prevented geomorphologists from study- Animals
ing the effects of animals as geomorphic agents.
With the parallel development of the fields of zoogeo- Figure 1, taken from Butler (2006) and based originally on a
morphology and ecosystem engineering in the early-to-mid- similar version created by Hall and Lamont (2003), illustrates
1990s, conceptual frameworks were established within which the primary geomorphic roles of animals on the landscape of
the geomorphic impacts of fauna on the landscape could be the Earth. Categories include trampling, loading, digging, bur-
assessed. In this chapter, we provide an overview of the major rowing, and beaver damming. Butler (2006) noted definitions
geomorphic influences of, and landforms created or shaped of each of these categories, and readers are referred to that paper.
by, animals. We comment primarily on the effects of naturally
occurring populations of animals in their native ranges;
12.16.2.1 Trampling and Loading
however, we also acknowledge the geomorphic impacts of
domesticated animals as well as feral animals. Two subsequent Each of the processes illustrated in Figure 1 has both direct and
chapters (see Chapters 12.18 and 12.20) in this volume of the indirect effects – trampling, for example, can directly remove
treatise go into greater detail on two of the most geographic- sediment from a location by sediment clinging to, or being
ally widespread and geomorphologically significant of these chiseled out of the ground, by hooves and paws of animals

Animal action

Burrowing Loading Digging Trampling Beaver damming

Hollows Compaction

Ponding Dam
Decrease failure
Direct sediment
slope
removal
stability Fluvial
erosion Reduction of
stream velocity
Precipitation
Slope Disrupt Displaced Sediment
failure vegetation sediment dispersal
Increased
water Mass Exposed Rainsplash Sedimentation
infiltration movement matrix and storage;
elevation of
Deflation water table
Sediment
removal
Change Downslope Needle ice
pedogenesis transport Downslope
sediment Segregation Stream profile
dispersal ice change
Changes
outflow
chemistry

Drainage basin
river/lake
catchments

Figure 1 Flow paths of the geomorphic impacts of animals. Reproduced from Butler, D.R., 2006. Human-induced changes in animal
populations and distributions, and their subsequent effects on fluvial systems. Geomorphology 79(3–4), 448–459, based on Hall and Lamont
(2003).
254 The Faunal Influence: Geomorphic Form and Process

(Bennett, 1999). Trampling by heavy animals such as hippos and Wallowing by large animals such as bison or elephants, which
elephants can create a widespread network of deeply incised can conceptually be considered a form of trampling (Butler,
trails (McCarthy et al., 1998) (Figure 2). Trampling also causes 2006), exerts significant compacting force on the underlying
indirect effects by disrupting vegetation (see Figure 1 for path- sediment that may lead to ponding. Subsequent wallowing in
ways), or compacting the surface and leading to ponding of mud in the same location initiates downward excavation via
water. In spite of the widespread acknowledgement of the im- sediment removal as mud clings to the animal’s coat/surface skin
pacts of trampling, few studies exist examining the actual forces and is carried elsewhere (Figure 3). Wallowing also acts to con-
induced by animals on ground surfaces. Bennett (1999) re- centrate carbon and nitrogen in the floor of wallows, leading to
viewed the scant literature available on the topic, and concluded profound effects on soil biochemistry and local landscape het-
that introduced grazing animals, for example, cattle and sheep, erogeneity (Eldridge and Rath, 2002).
cause greater mechanical disruption of soil surfaces, leading to Loading occurs when the weight of animals on a
increased rates of soil erosion, compared to the indigenous slope induces slope instability, and potential slope failure
grazing fauna of Australia. (Figure 1). Terracettes on hillslopes in areas where wild or

Figure 2 (a) Footprints of African elephant in sediment in Kruger National Park, South Africa, illustrating the broad trampling impact individual
feet have. Note lens cap for scale. (b) Elephant and rhinoceros trail incised into the landscape, Kruger National Park, South Africa.
 The Faunal Influence: Geomorphic Form and Process 255

Figure 3 Elephant wallow, Kruger National Park, South Africa.

(stone ingestion). Geophagy (soil ingestion) has received

some recent attention among bird ecologists. Birds that prac-
tice soil ingestion, or geophagy, include geese, parrots,
cockatoos, pigeons, cracids, passeriforms, hornbills, and
cassuaries (Emmons and Stark, 1979; Wink et al., 1993;
Diamond et al., 1999; Burger and Gochfeld, 2003). Beyer et al.
(1994) reported percentages of soil ingestion in the diets
of three avian species: sandpipers (Calidris spp.), 7–30%;
Canadian geese (Branton canadensis), 8%; and wild turkey
(Meleagris gallopavo), 9%. Munn (1994) studied parrot and
macaw species’ use of a clay-rich riverbank in the Peruvian
Amazon as a nutrition source. Geomorphic impact was not
discussed; however, it was noted that more than a thousand
birds were viewed consuming the clay on the riverbank.
Additional research on parrot and macaw geophagy in this
region of the Peruvian Amazon has been published within the
last decade. Brightsmith and Munoz-Najar (2004) observed 15
Figure 4 Terracettes attributed to elk grazing, Glacier National Park,
Montana, USA. species of birds ingesting clay-rich soils during 191 h along the
Tambopata River in southeastern Peru. Parrot, macaw, and
parakeet species made up the majority of the Psittacidae
domesticated animals graze have been attributed to loading family of birds present at the study site. Other species recorded
(Figure 4), but few empirical studies exist to document this include three pigeons, one guan, and one chachalaca. A total
attribution. Loading in association with the effects of tram- of 4334 birds were observed in a 56 m2 area. The highest
pling were examined by Boelhouwers and Scheepers (2004) in number of species to be seen practicing geophagy at a single
hyper-arid Namibia, and it should be noted that loading and site was reported as 28 species at a clay lick 500-m long and
trampling, although illustrated as separate processes in Fig- 25–30-m high along the western edge of the upper Tambopata
ure 1, will inevitably be operative together. River (Brightsmith, 2004). Powell et al. (2009) noted that up
to 1700 parrots of 17 species were observed in a single day.
Unfortunately, research addressing erosion rates along riv-
12.16.2.2 Digging
erbank geophagy sites is wanting.
Digging is distinct from burrowing in Figure 1, although dig- On the other end of the animal-size spectrum, geophagy by
ging is certainly necessary in order to create a burrow. Digging African elephants in Kenya significantly enlarges and modifies
in a nonburrow context refers to excavations of the surface by cave shape on Mount Elgon (Lundberg and McFarlane, 2006).
animals associated with food excavation, geophagy and Cave sediment is removed by ‘tusking’, an upward motion that
lithophagy, and sediment excavation for habitat construction. carves out geophagous sediment from the cave wall in scars
Butler (1995) reviewed the scant literature on lithophagy 2–4 cm wide and 10–30 cm long. Because of the size of
256 The Faunal Influence: Geomorphic Form and Process

elephants, tusking scars can extend up to 4 m high on cave

walls, and up to 3 m deep on cave floors. Elephant geophagy
also removes cave debris, further modifying the cavern system.
Other examples of ‘salt cave’ excavation by geophagy have
been described by Lundquist and Varnedoe (2006), who not
only attribute some existing caves to geophagous excavation
by modern animals such as bison and deer, but also note that
some salt caves have been created via geophagy by now-extinct
Pleistocene animals such as the milodon, a giant ground sloth.
Another unusual report of geophagy undertaken by large
mammals was reported by Mattson et al. (1999), who de-
scribed sites in Yellowstone National Park, Wyoming, USA,
where grizzly bears (Ursus arctos horribilis) were excavating and
consuming geothermally induced, hydrothermally altered
sediments. Examination of 12 sites shows deep excavations,
claw marks, and bear tracks. A total of 21 excavations at the 12
sites averaged 189.6 dm3 in size, with a total excavated volume
of 362.1 dm3 per site. The ingested material was transported
elsewhere, where it was later discovered in bear scats.
Digging for food (e.g., excavation of roots and tubers,
digging out other animals in burrows/mounds) may cover a
large local area. Foraging for food may result in
notable landform creation and impacts upon surface and
subsurface infiltration rates (Garkaklis et al., 1998, 2005). In
the American western state of Idaho, for example, badgers
(Taxidea taxus) excavate large volumes of soil and surface
sediments that are deposited on the surface as fan-shaped
mounds (Figure 5) while foraging for fossorial rodents
(Eldridge, 2004). Densities of 790 mounds per hectare, cov-
ering 5–8% of the land surface, were recorded by Eldridge
(2004) on the Snake River Plain of western Idaho. He esti-
mated that these excavations and resultant mounded soil in- Figure 5 American badger at burrow mouth with sediment fan.
volved 26 tons of material per hectare. The surface distribution
of mounds and excavations, in turn, produced impacts on the
surface cover of plants, cryptogams, and litter, with attendant of soda mud used by the lesser flamingos (Phoeniconias minor)
effects (Figure 1) on infiltration and surface runoff. The effect of Tanzania, and Rowley (1970) reported that mud is a vital
of American badgers is but one example of an animal with component in nest construction for approximately 5% of all
widespread geomorphic impacts via foraging efforts. Bragg bird species. Few of these species, however, construct nests
et al. (2005) illustrated the significant geomorphic impacts of entirely from mud. The Hirundinidea family, swallows and
foraging porcupines in South Africa. Whitford (1998) de- martins, are among those that do. Swallows are the only birds
scribed widespread foraging pits created by large goannas to build elevated attached nests composed entirely of mud
(Varanus gouldii) in eastern Australia, and illustrated that (Rowley, 1970).
overgrazing by domesticated animals was reducing the spatial The barn swallow is the most widely distributed and
density of the pits. Two useful general reviews of the effects of abundant species of swallow in the world (Brown and Brown,
animal foraging, and burrowing, in Australia (Eldridge and 1999), indicating their potential as geomorphic agents. Barn
James, 2009) and from sites around the world including swallows have a diverse habitat range and can occur in farm-
Australia, South Africa, Israel, and the USA (Garkaklis et al., lands, rural areas, suburban areas, or villages. Colonial nesting
2004) contain quantitative data on foraging densities and sites occur in various structures such as barns or other farm
tons of sediment excavated per site per year, and readers are outbuildings, bridges, wharves, boathouses, or culverts (Har-
directed to those publications for additional information. rison, 1975). The colonies are relatively more abundant in
Additional examples of the geomorphic effects of digging for human-built structures than in natural settings such as caverns
food may also be found in Butler (1995). and cliff sides. The site-specific requirements for their mud
Birds are active geomorphic agents in excavating soil and nests are the presence of a ledge and vertical wall to support
rocks for construction of nests or nesting mounds, but little the nest and a protective roof (Robbins et al., 1966; Link,
quantitative data exist from which conclusions can be drawn 2004). They have an open-cup-shaped nest, compared to the
about the overall significance of these actions (Butler, 1995). gourd-shaped nest of the cliff swallow (Figure 6). Soler et al.
Colonial seabirds excavate and utilize large numbers of peb- (2007) reported an average nest volume of 189 cm3 of sedi-
bles in ground-nest construction (reviewed in Butler, 1995), ment, although, the methods for this calculation were not
but fewer studies have analyzed the impact of mud-nest con- described. Extrapolating that volume to the many thousands,
struction. Brown and Root (1971) reported over 20 000 tons if not millions, of nests around the world illustrates the
 The Faunal Influence: Geomorphic Form and Process 257

Figure 6 Barn swallow nests at Versailles, France.

potential geomorphic significance of nest-building activities, have also seen a rise in interest in the impacts of European
and offers an avenue for additional quantitative zoogeo- beaver. This recent interest has been associated with
morphic research. deliberate and inadvertent introductions and recovery of
beaver populations in western and northern Europe. As one
of the most significant ecosystem engineers (Jones et al.,
12.16.2.3 Burrowing 1994) after humans, Chapter 12.20 in this volume of
Burrowing is a common process engaged in by a diversity the treatise by Westbrook et al. is devoted to the geo-
of invertebrate and vertebrate species, ranging across a suite morphological and hydrological impacts of beavers and
of insects and arthropods, worms, fish, amphibians, reptiles, beaver landforms, and readers are referred to that subsequent
birds, and mammals. Burrowing serves a variety of functions chapter.
(Butler, 1995), including denning and rearing of young,
protection of eggs, shelter from predators and protection from
climatic stress, socialization, access to below-ground food 12.16.3 Geomorphic Impacts of Domesticated and
sources, food caching, and sites for seasonal hibernation Feral Animals
or estivation. In the broadest sense, burrowing encompasses
features ranging from extensive underground tunnel com- Domesticated animals are easily included in the conceptual
plexes to simple shallow daybeds. Figure 1 illustrates the framework of Figure 1. The findings of Trimble and Mendel
myriad ways in which burrowing directly affects geomorphic (1995) quantitatively demonstrated that heavy grazing by
processes. Additionally, through its influences on surface cattle, for example, increased soil compaction, reduced mois-
vegetation and soils, burrowing has indirect geomorphic ture infiltration, and enhanced surface runoff and erosion
ramifications via its influences on soil structure and texture, (Figure 7). Cattle trampling along river banks was also found
soil fertility, infiltration capacity, and the resulting changes to directly increase slope failure (loading) and result in in-
wrought both in surface runoff and erosion and in production creased erosion and sediment transport (Trimble, 1994;
of vegetation cover (Butler, 1995). Because of the widespread Trimble and Mendel, 1995).
nature and geomorphic significance of burrowing, Chapter The impacts of several domesticated animals as geo-
12.18 by Butler et al. in this volume of the treatise is devoted morphic agents has been well documented (Evans, 1998), and
in detail to the direct and indirect geomorphic effects of continues to be studied in various aspects (Isselin-Nondedeu
burrowing and the related resultant landforms. et al., 2006; Isselin-Nondedeu and Bédécarrats, 2007). The
impact of feral animals on the landscape, however, is generally
difficult to recognize, and even more difficult to quantify. Feral
12.16.2.4 Beaver Damming
hogs, whose geomorphic effects were briefly summarized by
The genus Castor is comprised of two beaver species, the Butler (2006), disrupt vegetation via trampling, rooting for
North American beaver (Castor canadensis) and the European food, and creating hog wallows. Trampling and overgrazing by
beaver (Castor fiber). The former has been more intensively feral horses and burros, and the geomorphic actions of feral
examined for its widespread geomorphic impacts across rabbits in Australia, have also been the focus of research
much of the continent of North America, but recent years summarized elsewhere (Butler, 2006).
258 The Faunal Influence: Geomorphic Form and Process

Figure 7 Field overgrazed and heavily trampled by cattle, western Kansas, USA. Overgrazing allows invasion of field by ants. Ant mound is 18-
cm high and 35-cm in diameter.

12.16.4 Zoogeomorphology at Ecotones the marmots was not included as an explanation of the dis-
tribution of vegetation. English and Bowers (1994) estab-
In establishing an agenda for additional and future research, lished that woodchuck (a species of the genus Marmota)
we suggest that ecotones are ideal locations to study direct burrows in fields were likely to alter vegetation characteristics
zoogeomorphological impacts and their attendant indirect of the entire field. They discovered that vegetation near bur-
effects on surface vegetation density and pattern. Alpine en- rows contained lower species richness but richness increased
vironments contain an abundant amount of wild animals and with distance from the burrow. Although actual burrows ac-
exhibit more readily identifiable disturbances than those that counted for a small proportion of the entire field, the influ-
occur within the heart of an ecosystem (Butler, 1992; Hall ence was evident over a much wider area. Although this study
et al., 1999; Hall and Lamont, 2003). Butler (1992) found that again identified the importance of marmots in vegetation
grizzly bears in Glacier National Park, Montana, USA, were distribution, geomorphology was not mentioned. The distri-
likely to aid in the annual erosion of more sediment than a bution of vegetation around marmot burrows was finally
100-year snow avalanche. Hall et al. (1999) found similar linked to the microscale geomorphic influences of marmots by
results for grizzly bear excavations in Canada. Doak and Loso Semenov et al. (2001), who found that plant species and
(2003) emphasized the effects that excavation of dens by richness surrounding primary marmot burrows in northern
grizzly bears has on local vegetation in alpine regions, but Siberia were greatly reduced. They also found that marmots
focused primarily on the impacts to alpine vegetation density, modified the microtopography near the burrows and caused
richness, and pattern produced by bears removing vegetation changes in soil properties that had a direct effect on the
via digging. vegetation of the region (Figure 8). The connection between
Research along alpine ecotones has focused primarily zoogeomorphology and the direct impact on vegetation has
on the removal of soil and subsequent erosion of unconsoli- expanded to zokors in Tibet (Wang et al., 2008), pikas in
dated material. Insufficient research has been conducted on Mongolia (Wesche et al., 2007), gophers in the American
factors beyond excavation, trampling, and erosion. A logical Cascade Range (Jones et al., 2008), and ungulates in Scandi-
avenue for future research should include not only the geo- navia (Cairns et al., 2007). This linkage between the impacts
morphic impact that animals have on the environment, but of phytogeomorphology and zoogeomorphology indicates
also the secondary impact of vegetation on geomorphic substantial advancement of both subdisciplines and the
features created by animals. In essence, the interaction of foundation needed for continued study in the larger realm of
phytogeomorphology with zoogeomorphology is a likely biogeomorphology.
worthwhile step for advancement of the discipline. Several
early findings came close to this connection but failed to
capture the heart of the connection. Moral (1984) was quick 12.16.5 Conclusion
to identify the impact of Olympic Marmots on subalpine
vegetation. He found that palatable vegetation near the center Research in zoogeomorphology has shown that animals are
of marmot colonies was reduced and nonpalatable vegetation significant geomorphic agents, in a variety of environments,
was greatly increased. The microgeomorphology created by and need to be considered as important landscape drivers in
 The Faunal Influence: Geomorphic Form and Process 259

Davis, W.M., 1899. The geographical cycle. Geography Journal 14, 481–504.
Diamond, J., Bishop, K.D., Gilardi, J.D., 1999. Geophagy in New Guinea birds.
Ibis 141, 181–193.
Doak, D.F., Loso, M.G., 2003. Effects of grizzly bear digging on alpine plant
community structure. Arctic, Antarctic, and Alpine Research 35(4), 421–428.
Eldridge, D.J., 2004. Mounds of the American badger (Taxidea taxus): significant
features of North American shrub-steppe ecosystems. Journal of Mammalogy
85, 1060–1067.
Eldridge, D.J., James, A.I., 2009. Soil-disturbance by native animals plays a critical
role in maintaining healthy Australian landscapes. Ecological Management and
Restoration 10, S27–S34.
Eldridge, D.J., Rath, D., 2002. Hip holes: kangaroo (Macropus spp.) resting sites
modify the physical and chemical environment of woodland soils. Austral
Ecology 27, 527–536.
Emmons, L.H., Stark, N.M., 1979. Elemental composition of a natural mineral lick
in Amazonia. Biotropica 11, 311–313.
English, E.I., Bowers, M.A., 1994. Vegetational gradients and proximity to
woodchuck (Marmota monax) burrows in an old field. Journal of Mammalogy
75(3), 775–780.
Evans, R., 1998. The erosional impacts of grazing animals. Progress in Physical
Figure 8 Marmot burrow and stony debris fan beneath the mouth Geography 22, 251–268.
of the burrow negatively impacts tundra vegetation, Olympic National Garkaklis, M.J., Bradley, J.S., Wooller, R.D., 1998. The effects of Woylie (Bettongia
Park, Washington, USA. Note lens cap for scale. penicillata) foraging on soil water repellency and water infiltration in heavy
textured soils in southwestern Australia. Australian Journal of Ecology 23,
492–496.
future geomorphology research. Future investigation of geo- Garkaklis, M.J., Bradley, J.S., Wooller, R.D., 2004. Digging and soil turnover by a
morphic agents should be conducted along ecotones where mycophagous marsupial. Journal of Arid Environments 56, 569–578.
more apparent effects may be observed and quantified. Past Garkaklis, M.J., Bradley, J.S., Wooller, R.D., 2005. Digging by vertebrates as an
activity promoting the development of water-repellent patches in sub-surface
research has poorly integrated the geomorphic effects of ani-
soil. Journal of Arid Environments 45, 35–42.
mals on plant distributions, and future research should at- Hall, K., Boelhouwers, J., Driscoll, K., 1999. Animals as erosion agents in the
tempt to identify linkages between these factors. These areas of alpine zone: some data and observations from Canada, Lesotho, and Tibet.
cross-disciplinary research are ripe for study and exploitation Arctic, Antarctic, and Alpine Research 31, 436–446.
by those willing to shed traditional research dogmas and Hall, K., Lamont, N., 2003. Zoogeomorphology in the Alpine: some observations on
abiotic–biotic interactions. Geomorphology 55, 219–234.
conduct research beyond their home discipline. Harrison, H.H., 1975. A Field Guide to Birds’ Nests of 285 Species Found Breeding
in the United States East of the Mississippi River. Houghton Mifflin Company,
Boston, MA.
References Isselin-Nondedeu, F., Bédécarrats, A., 2007. Influence of alpine plants growing on
steep slopes on sediment trapping and transport by runoff. CATENA 71,
330–339.
Bennett, M.B., 1999. Foot areas, ground reaction forces and pressures beneath the Isselin-Nondedeu, F., Rey, F., Bédécarrats, A., 2006. Contributions of vegetation
feet of kangaroos, wallabies and rat-kangaroos (Marsupialia: Macropodoidea). cover and cattle hoof prints towards seed runoff control on ski pistes. Ecological
Journal of Zoology (London) 247, 365–369. Engineering 27, 193–201.
Beyer, W.N., Conner, E.E., Gerould, E., 1994. Estimates of soil ingestion by wildlife. Jones, C.C., Halpern, C.B., Niederer, J., 2008. Plant succession on gopher mounds
Journal of Wildlife Management 58, 375–382. in western Cascade meadows: consequences for species diversity and
Boelhouwers, J., Scheepers, T., 2004. The role of antelope trampling on scarp heterogeneity. The American Midland Naturalist 159(2), 275–286.
erosion in a hyper-arid environment, Skeleton Coast, Namibia. Journal of Arid Jones, C.G., Lawton, J.H., Shachak, M., 1994. Organisms as ecosystem engineers.
Environments 58, 545–557. Oikos 69, 373–386.
Bragg, C.J., Donaldson, J.D., Ryan, P.G., 2005. Density of Cape porcupines in a Link, R., 2004. Living With Wildlife in the Pacific Northwest. University of
semi-arid environment and their impact on soil turnover and related ecosystem Washington Press, Washington, DC.
processes. Journal of Arid Environments 61, 261–275. Lundberg, J., McFarlane, D.A., 2006. Speleogenesis of the Mount Elgon elephant
Brightsmith, D.J., 2004. Effects of weather on parrot geophagy in Tambopata, Peru. caves, Kenya. Geological Society of America Special Paper 404, 51–63.
The Wilson Bulletin 116(2), 134–145. Lundquist, C.A., Varnedoe, W.W., Jr., 2006. Salt ingestion caves. International
Brightsmith, D.J., Muñoz-Najar, A., 2004. Avian geophagy and soil characteristics in Journal of Speleology 35, 13–18.
Southeastern Peru. Biotropica 36(4), 534–543. Mattson, D.J., Green, G.I., Swalley, R., 1999. Geophagy by Yellowstone grizzly
Brown, C.R., Brown, M.B., 1999. Barn swallow (Hirundo rustica). In: Poole, A., Gill, bears. Ursus 11, 109–116.
F. (Eds.), The Birds of North America. The Birds of North America, Philadelphia, McCarthy, T.S., Ellery, W.N., Bloem, A., 1998. Some observations on the
PA, vol. 452. geomorphological impact of hippopotamus (Hippopotamus amphibius L.) in the
Brown, L.J., Root, A., 1971. The breeding behaviour of the lesser flamingo Okavango Delta, Botswana. African Journal of Ecology 36, 44–56.
Phoeniconaias minor. Ibis 113, 147–172. Moral, R.D., 1984. The impact of the Olympic marmot on subalpine vegetation
Burger, J., Gochfeld, M., 2003. Parrot behavior at a Rio Manu (Peru) clay lick: temporal structure. American Journal of Botany 71(9), 1228–1236.
patterns, associations, and antipredator responses. Acta Ethologica 6, 23–34. Munn, C.A., 1994. Macaws: winged rainbows. National Geographic 184,
Butler, D.R., 1992. The grizzly bear as an erosional agent in mountainous terrain. 118–140.
Zeitschrift für Geomorphologie 36, 179–189. Powell, L.L., Powell, T.U., Powell, G.V.N., Brightsmith, D.J., 2009. Parrots take it
Butler, D.R., 1995. Zoogeomorphology: Animals as Geomorphic Agents. Cambridge with a grain of salt: available sodium content may drive collpa (clay lick)
University Press, Cambridge. selection in Southeastern Peru. Biotropica 41(3), 279–282.
Butler, D.R., 2006. Human-induced changes in animal populations and Robbins, C.S., Bruun, B., Zim, H.S., 1966. Birds of North America. Western
distributions, and their subsequent effects on fluvial systems. Geomorphology Publishing Company, New York, NY.
79(3–4), 448–459. Rowley, I., 1970. The use of mud in nest-building: a review of the incidence and
Cairns, D.M., Lafon, C., Moen, J., Young, A., 2007. Influences of animal activity on taxonomic importance. Ostrich Supplement 8, 139–148.
treeline position and pattern implications for treeline responses to climate Semenov, Y., Ramousse, R., Berre, M.L., Tutukarov, Y., 2001. Impact of the Black-
change. Physical Geography 28(5), 419–433. Capped Marmot (Marmota camtschatica bungei) on Floristic diversity of Arctic
260 The Faunal Influence: Geomorphic Form and Process

Tundra in Northern Siberia. Arctic, Antarctic, and Alpine Research 33(2), meadow of the Qinghai–Tibet Plateau. Journal of Arid Environments
204–210. 72, 84–96.
Soler, J.J., Martin-Vivaldi, M., Haussy, C., Møller, A.P., 2007. Intra and interspecific Wesche, K., Nadrowski, K., Retzer, V., 2007. Habitat engineering under dry
relationships between nest size and immunity. Behavior Ecology 18, 781–791. conditions: the impact of pikas (Ochotona pallasi) on vegetation and site
Stine, M.B., Butler, D.R., 2011. A content analysis of biogeomorphology within conditions in southern Mongolian steppes. Journal of Vegetation Science 18,
geomorphology textbooks. Geomorphology 125, 336–342. 665–674.
Trimble, S.W., 1994. Erosional effects of cattle on streambanks in Tennessee, USA. Whitford, W.G., 1998. Contribution of pits dug by goannas (Varanus gouldii) to the
Earth Surface Processes and Landforms 19, 451–464. dynamics of banded mulga landscapes in eastern Australia. Journal of Arid
Trimble, S.W., Mendel, A.C., 1995. The cow as a geomorphic agent – a critical Environments 40, 453–457.
review. Geomorphology 13, 233–253. Wink, M., Hofer, A., Bilfinger, M., Englert, E., Martin, M., Schneider, D., 1993.
Viles, H.A. (Ed.), 1988. Biogeomorphology. Basil Blackwell, New York, NY. Geese and dietary allelochemicals – food palatability and geophagy.
Wang, T.C., Xiong, Y.C., Ge, J.P., et al., 2008. Four-year dynamic of Chemoecology 4, 93–107.
vegetation on mounds created by zokors (Myospalax baileyi) in a subalpine

Biographical Sketch

Dr. David R. Butler is the Texas State University System Regents’ professor of geography, and a University Dis-
tinguished professor at Texas State University-San Marcos where he has been on faculty since 1997. His research
interests are in the areas of mountain geomorphology, zoogeomorphology, biogeomorphology, and den-
drogeomorphology, focusing his work in the Glacier National Park region of Montana, USA. He has published
over 150 refereed papers in journals and conference volumes, and more than 35 book chapters. He is the author
of Zoogeomorphology – Animals as Geomorphic Agents (Cambridge University Press, 1995 and 2007), and co-editor
of Tree Rings and Natural Hazards (Springer, 2010), The Changing Alpine Treeline – The Example of Glacier National
Park, Montana (Elsevier, 2009), and Mountain Geomorphology – Integrating Earth Systems (Elsevier, 2003). He has
received the G.K. Gilbert Award for Excellence in Geomorphological Research from the Geomorphology Specialty
Group of the Association of American Geographers (AAG), and the Distinguished Career Award and the Out-
standing Recent Accomplishment Award from the AAG’s Mountain Geography Specialty Group. David holds a BA
and MSc in geography from the University of Nebraska at Omaha, and the PhD in geography from the University
of Kansas.

Clayton J Whitesides is a PhD student in the Department of Geography at Texas State University-San Marcos. His
research interests are in the fields of mountain geomorphology, biogeomorphology, and mountain biogeography.
He has published refereed papers in the international journal Progress in Physical Geography and in the Papers of the
Applied Geography Conferences. Clayton holds the Bachelor of Science degree from Utah State University, and the
Master of Science in geography from Brigham Young University.

Stephen G Tsikalas is a PhD student in the Department of Geography at Texas State University-San Marcos. His
interests are in zoogeomorphology, geomorphology, and physical geography. Stephen holds a Bachelor of Arts
from the University of Pittsburgh at Johnstown, and a Master of Arts in geography from the Indiana University of
Pennsylvania.
 12.17 Microbioerosion and Bioconstruction
HA Viles, University of Oxford, Oxford, UK
r 2013 Elsevier Inc. All rights reserved.

12.17.1 Introduction 261

12.17.2 What Are Microbes and Why Are They Important to Geomorphology? 261
12.17.3 What Do We Know about Microbial Contributions to Geomorphology? – a Brief Historical Review 264
12.17.4 State-of-the-Art of Microbial Contributions to Geomorphology – Case Study Environments 265
12.17.4.1 Arctic, Antarctic, and High Mountain Environments 265
12.17.4.2 Rocky Coasts and Coral Reefs 266
12.17.4.3 Hot Desert Environments 267
12.17.4.4 Ruiniform Landscapes 267
12.17.5 Current Key Questions in Microbial Geomorphology 268
References 268

Glossary Prokaryote Organism whose cells do not have a nucleus.

Endolithic Growing within the surface of a rock. Ruiniform Landscape resembling ruins.
Epilithic Growing on top of a rock surface. Tsingy Madagascan name for pinnacle karst relief as
Eukaryote Organism whose cells have a nucleus. found in the Tsingy de Bemaraha Strict Nature Reserve in
Geosites A protected site or area of geomorphological western Madagascar.
interest.

Abstract

Microbes are effective contributors to biogeochemical processes, make up a large part of global biomass and biodiversity,
are found in virtually all geomorphic environments, and, as a consequence, are of great biogeomorphological significance.
Rapid developments in culture-independent molecular methods of identifying microbes and improved microscopy now
permit fuller assessments of the nature and function of microbial communities. Microbes are known to contribute to earth
surface processes through their bioerosive and bioprotective roles, as demonstrated by a number of researchers from the
eighteenth century onward. Microbial contributions to geomorphology have been particularly well studied in a number of
extreme environments characterized by large swathes of bare rock, such as cold and hot arid environments, rocky coasts and
coral reefs, and ruiniform landscapes. Current key questions are: What are the net impacts of microbial communities on
geomorphic processes? How do microbial processes interact with other geomorphic processes? How do microbial processes
contribute to larger-scale landform and landscape development? How does geomorphology influence microbial com-
munities and their activities?

12.17.1 Introduction communities. This chapter outlines the state of the art of
knowledge on microbial contributions to geomorphology. It
The world of microbes may seem to be far removed from the begins with an introduction to what microbes are and why
world of geomorphology, but in reality the linkages can be they are important, and then reviews some of the classic re-
very close and important to both sides (Viles, 2012). Although search on their geomorphological roles. The third section
by definition small, microbes are also highly diverse and provides a series of case studies focusing on major geo-
adaptable, contribute importantly to global biomass and morphic environments where microbial activities are known
biodiversity, and engage in a very wide array of biogeo- to be important, that is, rocky coasts and coral reefs, arctic
chemical processes, often with substantial geomorphic effects. and Antarctic environments, hot arid environments, and rui-
In turn, geomorphological characteristics and processes can niform landscapes, whereas the final section discusses some
impact upon the growth forms and function of microbial of the key issues in today’s research on microbial contri-
butions to geomorphology. Microbial influences on geo-
morphology (and vice versa) are a potentially hot topic over
coming years, as technological developments now make it
Viles, H.A., 2013. Microbioerosion and bioconstruction. In: Shroder, J.
possible to study the nature and roles of microbial com-
(Editor in Chief), Butler, D.R., Hupp, C.R. (Eds.), Treatise on
Geomorphology. Academic Press, San Diego, CA, vol. 12, munities and link them to geomorphological processes and
Ecogeomorphology, pp. 261–270. landscape development.

Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00333-X 261

262 Microbioerosion and Bioconstruction

12.17.2 What Are Microbes and Why Are They In situ microscopical studies of microbes in the natural
Important to Geomorphology? environment have revealed that microbial communities are
characteristically complexly structured. Particularly common
Microbes (also known as microorganisms) are usually defined microbial forms noted to play geomorphological roles include
as being microscopic life forms, such as bacteria, fungi, and species of cyanobacteria (formerly called ‘blue-green algae’),
algae. Perhaps more importantly, the term ‘microbe’ covers a fungi, green algae, and lichens. However, there are many mi-
wide array of organisms that are able to live in the form of crobes also of importance which have received much less
single cells or small clusters of cells. This creates a clear dis- study, such as other bacteria. Although early studies focused
tinction between microbes and higher plants and animals, on the geomorphological roles of individual types of micro-
whose cells can only live as part of multicellular organisms. organism, or even individual species, it is now clear that the
Microbes, either as single cells or as small groups of cells, can influence of the community as a whole needs to be studied.
carry out all the processes necessary for life on their own, al- The term ‘biofilm’ is often applied to microbial communities
though they may also come together symbiotically to form inhabiting the interface between atmosphere or water and
other organisms such as lichens. A fundamental distinction can rock or soil surfaces and particles. Biofilms are generally made
be made between prokaryotic and eukaryotic microbes. The up of a range of microbes, occupying a number of niches. For
former do not possess a true nucleus and are divided into example, epilithic microbes are those that grow on the surface
bacteria (or eubacteria) and archaea (or achaeobacteria). The of a rock, whereas endolithic ones are grown into the rock,
latter do contain a true nucleus, and include fungi and algae. either by boring (euendolithic types) or by inhabiting pre-
Microbial taxonomy has been a dynamic and controversial existing cracks and fissures (chasmoendolithic types) or cav-
subject ever since the first microscopic observations were made, ities (cryptoendolithic). A full explanation of these terms is
and structural and phylogenetic approaches to microbial clas- given in Golubic et al. (1981). Endolithic growth forms are
sification produce different and sometimes conflicting infor- thought to be partly an adaptation to hostile conditions at the
mation. For example, bacteria and archaea are structurally surface, enabling the microbes to search for nutrients, shelter,
similar (both prokaryotic), but show very different evolutionary and water within the surface and avoid high ultra violet B
histories. Traditional taxonomy requires laboratory culture of (UVB) and temperature swings at the surface. Microbes
microbes followed by observation under the microscope to growing under stones and rocks are called hypolithic. In
look for structural differences. The real drawback of this ap- addition to these diverse microbial types, biofilms are also
proach is that culturing only captures a very small percentage of characterized by extra-polymeric substances (EPSs) secreted by
the entire microbial communities collected from the field, thus the microbes which form a slime matrix that protects and
giving a very biased view of what is present. In recent years, helps anchor the microbial community to the surface. Many
culture-independent methods of identifying microbes (mo- components of EPS can be highly reactive and participate in
lecular genetics) have facilitated significant progress in our many weathering processes.
understanding of the variability of microorganisms and the Microbes occur in a wide range of geomorphological
evolution of microbial diversity (Pace, 1997). situations and are particularly important in interface en-
Molecular taxonomic approaches describe the phylogenetic vironments – for example, at the rock–air interface or the
diversity of microorganisms without need for culturing sam- sediment–water interface. Microbes live on and within rock
ples. Such methods detect and identify microorganisms using surfaces in a very wide range of environments, from the poles
species-specific DNA sequences of ribosomal RNA genes to the deep sea, and inhabit similar niches on many sedi-
(rDNA). A chain of analyses is carried out, from extraction of mentary surfaces as well. For example, microbial biofilms
the DNA from an environmental sample, amplification of that are commonly important components of the biomass on
DNA (using polymerase chain reaction or PCR), cloning to the surface of soils within arid and semi-arid areas. Many
separate out the different DNA types, and then sequencing to microbial life forms are extremely well adapted to such hostile
characterize the organisms present. Identification of organisms interface environments, and many of these adaptations
present is done, where possible, by comparing the sequences translate into contributions to geomorphic processes
found to databases of known sequences (see Walker and Pace (Gorbushina, 2007). Microbes are widely distributed within
(2007) for a more detailed overview). Rapid improvements are water bodies and also make up a key part of many soils be-
being made in these sorts of techniques, allowing faster sides forming biofilms around soil particles. In these en-
throughput of samples and rapid profiling of the community vironments, their geomorphological roles may be less directly
metagenome, for example. Next-generation techniques such as apparent than when they grow on surfaces, but they can still
pyrosequencing, although initially prohibitively expensive, are be highly important. For example, microbial biomass within
now coming down in price and allow examination of phylo- soils can be a major contributor to high soil carbon dioxide
genetic and metabolic profiles of microbial communities. To concentrations, which play a key role in limestone dissolution
handle and analyze the vast amounts of data produced, a within the epikarstic zone (at the base of the soil profile), and
whole series of new bioinformatics tools is also being de- can also directly weather mineral grains (Uroz et al., 2009).
veloped. In addition to identifying the microorganisms pre- Microbes can also coexist with higher plants, as for example in
sent, molecular techniques are also available to ascertain the the rhizosphere where mycorrhizae involve a symbiotic asso-
function of different taxa. For example, fluorescent in situ hy- ciation of fungi and plant roots. In such cases, their geo-
bridization (FISH) can be used to identify the location of morphological influences become bound up with those of
microbes on mineral and rock surfaces, from which one can higher plants (Bonafante and Anca, 2009). Microbes are also
infer any key linkages. important sources of food for microinvertebrates and other
 Microbioerosion and Bioconstruction 263

animals, and grazing on microbial biofilms can cause erosion working within the coastal and oceanic arena, view them as
of the underlying rock surface. contributing to erosion. Thus, in the terrestrial realm the term
Microbes play many important geomorphological roles ‘biological weathering’ is dominantly used, whereas in coastal
despite, or often because of, their small size and simple nature. and marine studies, the term ‘bioerosion’ or ‘microbioerosion’
Such roles can be categorized into two main types, bioero- is preferred. Such a difference may relate to the surrounding
sional and bioprotective. Looking firstly at bioerosional ef- environmental conditions in coastal and oceanic locations,
fects, these involve the weathering and removal in solution or where frequent tidal, wave and current action and grazing of
as particulate debris of rocks, minerals, and sediments. As higher organisms prevent the buildup of weathered material;
Figure 1(a) demonstrates, microbial boring can produce holes thus, all weathering leads directly to removal of material and
within a rock surface and cause erosion. Much terminological thus, by definition, to erosion. In other, more benign, en-
confusion and inconsistency surrounds such roles, largely vironments such as most terrestrial surfaces, microbial pro-
because workers in different environments have focused on cesses may cause weathering but the products are not so
different consequences of bioerosional processes. In essence, effectively eroded away and so the weathering effect is the
microbes engage in bioerosion through the production of more noticeable end result. As Figure 1(b) illustrates, the net
extracellular substances that can dissolve and decompose effect of lichen weathering in some places can be the pro-
minerals, as well as through physical changes (e.g., growth duction of small depressions and micro relief.
leading to volume increases, and swelling and contracting on Microbes also play bioprotective roles, whereby directly or
wetting and drying). Some workers regard these processes as indirectly they contribute to the reduction of erosion and, in
contributing to weathering, whereas others, notably those some cases, the production of resilient bioconstructions.
Figure 2(a) shows the involvement of diatoms and EPS in the
precipitation and trapping of calcium carbonate within a
freshwater tufa deposit, for example. As with the bioerosional
effects of microbes reviewed above, both chemical and phys-
ical processes can be involved. Physically, microbial com-
munities can be highly effective at trapping and binding
sediment, through the large active surface areas produced by
complex associations of single cells and filaments on a surface.
These can reduce air and water velocities over the surface, thus
aiding deposition, as well as preventing erosion through their
physical presence (covering and anchoring grains and thereby
reducing the erosivity of the surface). Microbes can also make
chemical contributions to surface stabilization through the
production of sticky EPS compounds which encourage ma-
terial to adhere to cell surfaces, and through biochemical
precipitation of minerals (e.g., calcite and calcium oxalates),
which can build up on a rock or soil surface. Microbial bio-
constructions take this bioprotective role further, through
(a)
the production of visible and commonly large, resilient
constructions that store sediment on the short to medium
term. For example, stromatolites occur in some coastal en-
vironments, which are layered deposits (characteristically rich
in calcium carbonate) whose growth is mediated by cyano-
bacteria. Famous examples include those in Shark Bay, Aus-
tralia. Within freshwater environments, tufa deposits can also
be produced by microbial precipitation of minerals (generally
alongside abiotic precipitation and that associated with higher
plants) as shown in Figure 2(b). Many similar examples of
microbial contributions to the production of constructive
landforms exist, such as many cave deposits. A smaller and
more enigmatic bioconstruction is desert varnish, a very thin
Mn- and Fe-rich layer forming a polished surface to many
rocks within arid environments (Figure 3). Various workers
have found microbial processes to play a key role in the
production of desert varnish (e.g., Dorn and Oberlander,
(b) 1981), although inorganic methods of desert varnish for-
Figure 1 (a) Fungal hyphae at the base of a lichen growing
mation have also been proposed, and debate rages about the
euendolithically in marble from the hyperarid Namib desert. Scale contribution microbes may make (Perry et al., 2006).
bar ¼ 20 mm. (b) Pits in sandstone rocky outcrop at Golden Gate Few studies have been made of the contributions microbial
Highlands National Park, South Africa, created and inhabited by communities as a whole make to bioerosion and bio-
endolithic lichens. Sunglasses for scale. protection, as most studies focus on one particular process.
264 Microbioerosion and Bioconstruction

However, Schneider and Le Campion-Alsumard (1999) pro-

vided a good review of the contributions that one group of
microbes, the cyanobacteria, makes to bioerosion and bio-
construction in marine and freshwater environments. More
such assessments, especially of the role of whole biofilm
communities in comparison with other weathering and ero-
sive processes, would be very helpful in providing a fuller
picture of microbial contributions to geomorphology.

12.17.3 What Do We Know about Microbial

Contributions to Geomorphology? – a Brief
Historical Review

Studies of the interplay between microbes, rocks, and sedi-

ments have a long history within the earth and environmental
(a) sciences starting with the advent of microscopes powerful
enough to observe microorganisms. As Gorbushina (2007)
noted, detailed studies of epilithic algae, fungi, and bacteria in
the eighteenth century were carried out by famous natural
historians such as von Humboldt and Ehrenberg. By the end of
the nineteenth century, many detailed observations had been
made of microorganisms growing endolithically in rocks,
shells, and coral substrates (e.g., Duncan, 1876; Bornet and
Flahault, 1888; Huber and Jadin, 1892; Duerden, 1902). Al-
though most studies were quite small scale and focused on
identification of the microorganisms and elucidation of how
they produced boreholes, a few workers speculated about the
larger-scale impacts of microbial communities. For example, in
1907 the algologist, F.E. Fritsch described the many and various
ways in which green algae and blue green algae (now called
cyanobacteria) cover the land surface in temperate and tropical
areas, and how they might contribute to geomorphological
(b) processes. He noted that green algae were more common on
bare rock surfaces in the temperate zone, whereas blue-green
Figure 2 (a) Diatoms and EPS forming part of a tufa deposit. Scale
algae dominated in the tropics. He observed a range of differ-
bar ¼ 50 mm. (b) Actively forming tufa deposit in an ephemeral
stream system in the Naukluft Park, Namibia. Width of the barrage at ent algal growth forms on rock surfaces in Sri Lanka (adhesive,
the base is approximately 30 m. tangled, tufted, and stratified) and noted that a sequence of
these growth forms may occur or, as Fritsch put it ‘‘Here we
have, therefore, an interesting succession of the different types
of growth, and these diverse forms undoubtedly have a slow
but sure disintegrating action on the rock’’ (Fritsch, 1907, p.
539). Nadson (1927) discussed the distribution of endolithic
algae and the impacts their boring processes might have on the
environment.
During the twentieth century, botanists, microbiologists,
geologists, and geomorphologists carried out a wide range of
studies on microbial impacts on earth surface processes.
E. Jennie Fry in the 1920s, for example, carried out some in-
novative and important experimental and observational studies
on the impacts of lichens on weathering (Fry, 1922, 1927).
Several key hypotheses were proposed to link a wide range of
microbial processes to diagnostic ‘biosignatures’ or larger-scale
contributions to geological and geomorphological phenom-
ena. For example, Bathurst (1966) proposed that micrite en-
velopes, commonly occurring in skeletal remains in calcareous
rocks and sediments, could be produced by the boring activity
of endolithic algae. He suggested that the micrite envelopes
Figure 3 Desert varnish on sandstone in hyper-arid conditions in enhanced resistance to erosion, and were crucial in the survival
the Libyan Sahara Desert. of some skeletal remains, such as mollusk shells, in limestones.
 Microbioerosion and Bioconstruction 265

Various authors, including Purdy and Kornicker (1958), ascri- their roles in the larger-scale evolution of landforms and
bed the notch found at around sea level in tropical limestone landscapes.
coasts and the erosion of limestone outcrops in general to the
action of algae and other boring organisms. In their words ‘‘it is
becoming increasingly apparent that boring blue-green algae
12.17.4 State-of-the-Art of Microbial Contributions
are among the most important agents of destruction of coastal
to Geomorphology – Case Study
limestone’’ (Purdy and Kornicker, 1958, p. 96). Following this
Environments
early work, Schneider (1976) illustrated by painstaking studies
on limestone coasts in the Adriatic that cyanobacterial boring
Although microbes have been demonstrated to contribute to
weakened the rock surfaces, allowing gastropod grazing activ-
geomorphological processes in virtually all environments
ities to cause erosion. The observation that various boring
(Viles, 2012), their role has been particularly well studied in
microbial species occurred at specific depths with respect to sea
some environments, particularly ones that are thought to be
level, and that they left recognizable boring ‘biosignatures’, led
extreme such as rocky coasts, arctic and Antarctic, and arid and
workers such as Swinchatt (1969) to propose that boring could
ruiniform landscapes. By ruiniform landscapes, we mean
be a useful bathymetric indicator for paleoenvironmental
landscapes such as the tower karst of Guilin, southern China,
studies in areas of sea-level change. Swinchatt (1969) noticed
the Danxia landscapes of China, the Bungle Bungles in
that algal boring in coral reef environments was generally
Australia, and other dramatic landscapes characterized by
concentrated in the upper 20 m or so, whereas fungal bore-
complex arrangements of towers, and large expanses of bare
holes occurred at much greater depths.
rock and inhospitable terrain. Such sites are generally pro-
On terrestrial surfaces, Folk et al. (1973) proposed the
tected because of their geomorphological beauty and im-
hypothesis that algal boring on limestone could produce
portance, with many listed as World Heritage Sites. The term
weirdly sculpted pinnacles or ‘phytokarst’, thus making a bold
‘ruiniform’ (probably first introduced into the geomorpho-
and controversial link between microscale boring and mac-
logical literature by Mainguet, 1972) nicely describes their
roscale landforms. Krumbein (1969) hypothesized, from work
appearance – like a series of ruined buildings and structures,
in the Negev desert, that a ‘biological solution front’ was
or the ruins of a younger, less dissected landscape.
formed from a mixed community of epitihic and endolithic
Microbial biofilms are known to be adept at surviving in
lichens, algae, and fungi that produced complex weathering
extreme environments, especially those that suffer from vari-
effects. Friedmann (1982), building on the pioneering obser-
able stresses (sometimes the term ‘poikilo-tolerant’ is applied
vations of Friedmann and Ocampo (1976), developed similar
to subaerial biofilms growing on rocks to summarize this
ideas for endolithic microbial communities in Antarctica.
ability; see Gorbushina, 2007). Environments classed as
Endolithic lichens and cyanobacteria in Antarctic sandstones
‘extreme’ are those that encounter frequent, severe, and vari-
were noted to contribute to exfoliation of the rock surface
able conditions of physical stress, such as scarcity of liquid
(Friedmann, 1982). Other workers also proposed important
water, high and/or low temperatures, high UVB levels, high
contributions of microbes to the development of crusts. Kahle
wave energies, and low nutrient levels. Under these hostile
(1977), for example, showed that surface calcareous crusts
conditions, microbial biofilms commonly adopt, wholly or
could be formed by microbial weathering and precipitation,
partly, an endolithic growth form and thus become in close
and suggested that they might be recognizable in the geo-
and active association with their host surfaces. Such extreme
logical record, whereas Klappa (1979) illustrated how lichens
environments are interesting to geomorphologists and other
could form laminated calcretes. Many subsequent studies have
earth and environmental scientists for many reasons. First,
evaluated, tested, and refuted these hypotheses.
they are thought to contain relatively simple microbiological
During the latter part of the twentieth century, a number of
communities and thus they are probably some of the easiest
technological advances led to improved observations of mi-
environments in which to characterize the structure and
crobial impacts on Earth surface processes, allowing better
function of microbial communities. Second, they contain
testing of the types of hypotheses reviewed above. For ex-
many examples of the early developmental stages of eco-
ample, Golubic et al. (1970), Le Campion Alsumard (1979),
systems (e.g., on moraines as glaciers retreat or on newly
and Lucas (1979) developed and used novel scanning electron
erupted volcanic surfaces), which provide useful analogs for
microscopy (SEM) and casting/embedding techniques to
past conditions more generally on the Earth. Third, they
visualize cyanobacterial and fungal boring. The advent of
provide analogs for extra-terrestrial environments that might
environmental scanning electron microscopy (ESEM), con-
host life. Finally, the lack of ecological community complexity
focal scanning laser microscopy (CSLM), atomic force mi-
(and the lack of extensive soil cover) makes the geomorpho-
croscopy (AFM), and other techniques has permitted ever
logical process regimes relatively simple.
more detailed observations of microbes, and their interactions
with rocks and mineral grains (as reviewed by Hutchens,
2009). Field experimental studies, using substrates of
12.17.4.1 Arctic, Antarctic, and High Mountain
glass, calcite, or coral, also allowed estimations of the rate and
Environments
nature of endolithic and epilithic colonization and their im-
pacts over timespans of a few years in marine, coastal, and The cold extreme poses particular challenges to microbial
terrestrial environments (e.g., Le Campion Alsumard, 1975; communities, with very hostile conditions inhibiting growth.
Tudhope and Risk, 1985; Viles, 1987). Measuring rates of Remarkably, such environments have been found to host a
microbial processes allows better testing of hypotheses about wide range of microbial communities, as demonstrated in the
266 Microbioerosion and Bioconstruction

pioneering studies of E.I. Friedmann and colleagues in the very different geologies. Although the communities had some
McMurdo Dry Valleys, Antarctica (Friedmann and Ocampo, similarities, they also had clear differences, which Skidmore
1976; Friedmann, 1982). Previously thought to be too hostile et al. (2005) hypothesized to result from the different bio-
for life of any sort, sandstones within the Dry Valleys are now geochemical roles played (sulfide oxidation vs. gypsum dis-
known to support mixed endolithic biofilm communities with solution). Microbial communities are also of key importance
lichens, fungi, cyanobacteria, and a range of other bacteria in the development of soils and the development of more
(de la Torre et al., 2003). Although studies have shown these complex ecosystems in cold environments (Borin et al., 2010),
communities to be very slow-growing (Johnston and Vestal, and have also been shown to contribute to the development of
1991), they have also been hypothesized to play crucial roles small-scale calcite crusts.
in weathering and mineral transformation (Friedmann, 1982).
Similarly, diverse microbial communities on rock surfaces
have also been observed from other cold environments, in-
12.17.4.2 Rocky Coasts and Coral Reefs
cluding granite in Arctic Norway (Chaput, 2011), High Arctic
Canada (Omelon et al., 2007), Tibetan plateau (Wong et al., Hard coastal substrates have been found to be another major
2010), and associated with weathering activity (Etienne, 2002; habitat for microbial biofilms, with endolithic growths being
Matthews and Owen, 2008; Brunner et al., 2011). particularly important here, especially within carbonate sub-
Friedmann and Ocampo (1976) noted cyanobacteria strates. Life along the coast and within the shallow marine
growing cryptoendolithically in pore spaces within ortho- environment is hostile because of high wave forces, low nu-
quartzite rocks in the Dry Valleys, Antarctica in layers of trient availability, and dramatic wetting and drying regimes,
around 1.5 mm thick, stretching from 1 to 2 mm beneath the encouraging microbes to grow endolithically. Through active
rock surface. Similar cyanobacterial communities had previ- boring into susceptible rocks and substrates (e.g., shells, live
ously been identified in hot desert situations. Subsequent and dead corals, and limestones), these euendolithic mi-
studies showed another key, lichen-dominated endolithic crobial biofilms leave their imprint, and cause microbioero-
community to also be present in the Dry Valleys, forming a sion. This process of microbioerosion underpins a whole
thick band zoned into white, black, and green layers chain of erosional processes, involving a mixture of biotic and
(Friedmann, 1982). Lichen-dominated endolithic com- abiotic agents, which ultimately results in the shaping of entire
munities are more abundant than cyanobacterial ones in the coastal profiles. Bioerosion on coastal surfaces is commonly
Dry Valleys. Culture-independent methods have shown that divided into three types, that is, grazing, macroboring (with
both endolithic communities are relatively low in microbial borehole diameters of 4100 mm), and microboring (borehole
diversity, but that the cyanobacterial community is dominated diameters o100 mm). The fact that microbioeroding taxa are
by alpha-Proteobacteria and Thermus-Deinococcus taxa as distributed according to depth and light levels in conjunction
well as cyanobacteria (de la Torre et al., 2003). Persistent cold with the observation that individual taxa produce specific and
and dry conditions render these endolithic communities slow unique boreholes (Golubic et al., 1975) makes them powerful
growing, but such conditions also limit the rate of other po- tools for paleoenvironmental reconstruction (see, e.g., Hein-
tential weathering processes, and the long-term persistence of del et al., 2009).
microorganisms in this near-surface band make them a very Early investigations demonstrated that cyanobacteria,
crucial player in weathering and surface protection. High chlorophytes, and fungi were dominant components of shal-
growth of cryptoendolithic fungal communities in Iceland has low marine and coastal biofilms on limestone coasts and coral
been hypothesized to encourage flaking of rock surfaces reefs and contributed importantly to overall bioerosion
(Etienne, 2002), but such a growth strategy ultimately leads to (Schneider, 1976; Kobluk and Risk, 1977). Euendolithic
the loss of protective habitat for the microbes. Bioprotective growth forms dominate, producing boreholes that penetrate
rather than bioerosive strategies might lead to competitive up to a few millimeters into the substrates. A whole host of
advantages for microbial communities living in such cold micro- and macro-grazing species have been observed to eat
conditions, or bioerosive events such as exfoliation may act to these endolithic microbes, removing the weakened outer sur-
stimulate the growth of other species, which may facilitate the faces of the substrate at the same time. Within coral reef en-
long-term development of the microbial community. vironments, for example, parrot fish and urchins have been
Microbial biofilms also occur in a range of other geomor- shown to be important macro-bioeroding species (Mumby
phologically important situations in cold environments, such et al., 2007). Microbioerosion also weakens the substrate and
as under periglacially sorted stones (in the hypolithic niche, as makes it more susceptible to inorganic erosion by waves and
observed by Cockell and Stokes, 2004), in permafrost (Steven other agents, and to inorganic weathering processes (such as
et al., 2007), in evaporites (Cockell et al., 2010); within wetting and drying within the intertidal one). On the other
cryoconite hollows in glacier surfaces (Anesio et al., 2009) and hand, microbioerosion can be followed by micritization (the
in subglacial environments (Skidmore et al., 2005). Little is formation of fine-grained calcite grains as cements within and
known about the precise roles played by microbial com- around the borings) and the production of protective micrite
munities in these environments, but evidence shows that they envelopes (a layer of micrite which covers an entire clast or
are involved in the dissolution and transformation of a range fossil).
of minerals (Hodson et al., 2008). Skidmore et al. (2005), for A number of well-designed field experiments have clarified
example, used culture-independent methods to characterize the rate of different erosive processes on reefs and rocky coasts,
the subglacial microbial communities in two glaciers (one in and the nature and importance of their interactions over time
Alaska and one on Ellesmere Island, Canada) that flow over and space. For example, field experiments on Moorea, French
 Microbioerosion and Bioconstruction 267

Polynesia, showed that microbioerosion dominated early on

(responsible for 60% of total bioerosion within the first 2
months), with grazing becoming more important later.
However, other studies have demonstrated a longer-term
dominance of microbioerosion, with Tribollet and Golubic
(2005) finding, from a 3-year study of an inshore–offshore
transect in the Great Barrier Reef, that microbioeroders con-
tinue to be the major cause of carbonate loss after 3 years,
because of their facilitation of grazing. Tribollet and Golubic
(2005) also noted that grazing and microbioerosion domin-
ated within the offshore parts of the transect, where total
bioerosion rates were also highest, whereas macroborers
dominated within inshore portions. Along limestone coasts,
more complex relationships between microbioerosion, graz-
ing, macroboring, and inorganic processes are commonly
observed, especially where wave energies are high and where
high tidal ranges lead to large areas subjected to weathering Figure 4 Lichens growing on granitoid rocks in the hyper arid
northern Namib desert.
during emersion within the intertidal zone. Clear spatial pat-
terning characteristically develops on such coastlines, with
microboerosive, grazing, and macroboring activity concen- Danin and Garty (1983), and others in the Negev and Sinai
trated in particular areas such as around the intertidal notch, Deserts. These workers found that epilithic and endolithic li-
pools, and joints (Schneider, 1976). Microbial biofilms can chens and cyanobacteria were common inhabitants and
also contribute to the accumulation and lithification of sedi- caused recognizable imprints on limestone surfaces (e.g.,
ment within these coastal and shallow marine environments, micropitting). The distribution of epilithic versus endolithic
illustrating their bioprotective role. growth forms was shown to be controlled by moisture avail-
ability, with drier microenvironments dominated by endo-
12.17.4.3 Hot Desert Environments lithic forms. Although hot deserts are by definition largely
lacking in water, rock surface biofilms can survive by extracting
The hot arid extreme is also home to interesting and geo- water from precipitating dew, fog, condensation, and high
morphologically important microbial communities. Within levels of atmospheric humidity. Although growth rates are
these environments, microbial communities need to be cap- probably slow, the conditions are more favorable than those
able of surviving desiccation, extreme heat, large temperature in cold desert environments for all but hyper-arid locations. As
swings and high levels of salinity (encouraged by high evap- with the endolithic communities occurring in Antarctica and
oration rates). Early observations of microbial communities other cold, dry environments, microbial biofilms in hot des-
and their effects came from arid areas such as the Negev erts are thought to play key roles in both weathering and
Desert. Here, for example, Shachak et al. (1987) noted snails surface stabilization, but little work has been done on quan-
grazing on lichens growing endolithically within limestone, tifying their net effects in comparison with those of inorganic
causing appreciable erosion. More recent work has focused on processes. It has also not yet been established whether grazing
hyper-arid environments, such as the Atacama and Namib on endolithic biofilms plays a general role in surface lowering
Deserts which have large tracts of land without higher plants (as in coral reef and rocky coastal environments) or if it is a
and where lichens and microbial biofilms are important rare occurrence in the hot arid realm.
components of the ecosystem (Figure 4). Parts of the hyper-
arid Atacama Desert in Chile and Peru have long been thought
to be biologically dead, but detailed observations have re-
12.17.4.4 Ruiniform Landscapes
vealed the presence of bacterial communities growing in soils,
and endolithically on rocks and evaporitic deposits (Drees The term ‘ruiniform’ (meaning ‘ruin-like’) was applied by
et al., 2006). Wierzchos et al. (2011), for example, noted that Mainguet (1972) to sandstone landscapes dominated by rough
gypsum crusts in the hyper-arid core of the desert are colon- towers, arches, and other outcrops, and has since been applied
ized by epilithic lichens, endolithic algae, fungi, cyanobacteria more widely to complexly dissected topographies with many
and non-photosynthetic bacteria. Gorbushina (2007) noted bare rock surfaces. For example, ruiniform relief occurs within
that desert rocks are commonly colonized by microcolonial many karst areas (e.g., tower karst in Southern China, the
fungi, which seem to be particularly well adapted to life in tsingy of Madagascar), on sandstones, quartzites, and con-
such a parched and variable environment. Microbial biofilms glomerates (e.g., sandstone landscapes of Vila Velha, southern
also play a key role in surface stabilization in many desert Brazil, Zhangjiajie in southern China as shown in Figure 5,
areas, with biological soil crusts known to protect soils from and the Cedarberg mountains in South Africa; the quartzite
wind and water erosion (Belnap and Lange, 2003). A simple tepuis of Venezuela, and the continental redbeds of the Danxia
review of the roles of microbial communities in biological landform sites, China), as well as on granite and basalt (e.g.,
crusts in arid areas is provided by Viles (2008). the tors, inselbergs and boulder piles of the granite landscape
Much of our understanding of microbial influences on wea- in the West Sudetes in Poland; Migon and Latocha, 2008).
thering in hot deserts comes from the work of Krumbein (1969), Such ruiniform landscapes occur within all climatic zones
268 Microbioerosion and Bioconstruction

12.17.5 Current Key Questions in Microbial

Geomorphology

The research reviewed in the preceding sections demonstrates

that the study of microbial influences on geomorphology (and
vice versa) is flourishing as a result of a range of new techno-
logical developments, which build upon earlier explorations of
the geomorphic roles microbes can play. Several key questions
remain to be answered that require the joint efforts of micro-
biologists, geochemists, and geomorphologists over the com-
ing years. First, although we know a great deal about the nature
and roles of many microbes within some landscapes, we still
need to find out more about the overall composition of mi-
crobial communities and their combined bioerosive and bio-
protective roles. This requires widespread sampling of
communities, the application of molecular taxonomic meth-
ods, carefully designed laboratory experiments, and careful
microscopic observation of field material. Are all microbial
communities in landscapes across the world much the same?
Or are there key biogeographical patterns? If so, what might
explain them and how might they influence the geomorphic
activity of microbial communities in different environments?
Second, we need to investigate how microbiological processes
link with the activities of larger organisms and inorganic geo-
Figure 5 Ruiniform sandstone landscape at Zhangjiajie, southern morphic processes to produce the net erosive and constructive
China. Are microbial communities contributing to landscape processes observed in the landscape. This requires well-de-
development here and in similar landscapes? signed field experiments, for example, with the use of exclusion
plots, which can quantify the individual and combined con-
tributions of different processes. Third, we need to tackle the
and owe their origin to long-term denudation. Ruiniform important question of upscaling in order to establish the extent
landscapes are commonly protected geosites (several figure on to which microbioerosion and bioprotection contribute to the
the World Heritage List, for example) as the complex relief is development of small-, medium, and large-scale landform and
both beautiful and intriguing. landscape evolution. For example, although we can now be
Many observations have been made of the presence of confident that microbial communities cause small-scale fea-
microbial biofilms within such ruiniform landscapes, and they tures such as pitting, and may be involved in exfoliation, it is
have been hypothesized to play key roles in weathering and still unclear how such involvement translates into larger-scale
surface protection. For example, Budel et al. (2004) illustrated persistence and erosion. In order to address this question we
that chasmoendolithic cyanobacteria in sandstones are cap- need to be able to add microbial influences to numerical
able of raising the local pH to c. 9, enhancing the solubility of landscape evolution models. Finally, we need to address the
quartz. On the other hand, Gorbushina et al. (2001) observed reverse side of the relationship, that is, does geomorphology
both biological weathering and biological precipitation of influence microbial communities and their activities? Relief at
minerals from quartzite tepuis in Venezuela, noting that most all scales and inorganic earth surface processes may play key
surfaces were covered with a black biofilm dominated by li- roles in the establishment and growth of microbial com-
chens and cyanobacteria, with biogenic precipitation of for- munities, but little is as yet known about this.
sterite microstromatolites common. In turn, following the
hypothesis of Folk et al. (1973), such roles may be crucial in
the evolution of the complex relief forms even at large scales.
Thus, Budel et al. (2004) suggested that bioalkalization of References
sandstones produces small-scale surface flaking, which over
time might produce rock shelters or small caves. Similarly, the Anesio, A.M., Hodson, A.J., Fritz, A., Psenner, R., Sattler, B., 2009. High microbial
development of protective microstromatolitic coatings on rock activity on glaciers: importance to the global carbon cycle. Global Change
Biology 15, 955–960.
surfaces may enhance their strength and encourage the de-
Bathurst, R.G.C., 1966. Boring algae, micrite envelopes and lithification of
velopment of towers. However, all such links to larger-scale molluscan biosparites. Geological Journal 5, 15–32.
landscape evolution are as yet speculative. Thus, as with the Belnap, J., Lange, O. (Eds.), 2003. Biological soil crusts: structure, function, and
other extreme environments reviewed in this chapter, con- management. Springer, Berlin.
vincing evidence has been proposed for both bioerosive and Bonafante, P., Anca, I-U., 2009. Plants, mycorrhizal fungi, and bacteria: a network of
interactions. Annual Review of Microbiology 63, 363–383.
bioprotective roles of microbial communities in ruiniform
Borin, S., Ventura, A., Tambone., F., et al., 2010. Rock weathering creates oases of
landscapes, but more work needs to be done to quantify their life in a high Arctic desert. Environmental Microbiology 12, 293–303.
impact in comparison with inorganic processes and relate it to Bornet, M.E., Flahault, C., 1888. Note sur deux nouveaux genres d’algues
larger-scale landscape evolution. perforantes. Journal de Botanique 10, 161–165.
 Microbioerosion and Bioconstruction 269

Brunner, I., Plotze, M., Rieder, S., Zumsteg, A., Furrer, G., Frey, B., 2011. Klappa, C.F., 1979. Lichen stromatolites: criterion for subaerial exposure and a
Pioneering fungi from the Damma glacier foreland in the Swiss Alps can mechanism for the formation of laminar calcretes (caliche). Journal of
promote granite weathering. Geobiology 9, 266–279. Sedimentary Research 49, 387–400.
Budel, B., Weber, B., Kuhl, M., Pfanz, H., Sultemeyer, D., Wessels, D., 2004. Kobluk, D.R., Risk, M., 1977. Calcification of exposed filaments of endolithic algae,
Reshaping of sandstone surfaces by cryptoendolithic cyanobacteria: micrite envelope formation and sediment production. Journal of Sedimentary
bioalkalization causes chemical weathering in arid landscapes. Geobiology 2, Petrology 47, 517–528.
261–268. Krumbein, W.E., 1969. Uber den Einfluss der Microflora auf die exogene Dynamik.
Chaput, D., 2011. Structure, diversity and metabolic profile of lithobiontic (Verwitterung und Krustenbildung). Geologische Rundschau 58, 333–363.
communities inhabiting exposed Arctic granite: a molecular microbial ecology Le Campion Alsumard, T., 1975. Etude experimentale de la colonisations d’eclats de
approach. Unpublished D.Phil thesis, University of Oxford, UK. calcite par les cyanophycees endoliths marines. Lea Cahiers de Biologie Marine
Cockell, C.S., Osinski, G.R., Banerjee, N.R., Howards, K.T., Gilmour, I., Watson, 16, 177–185.
J.S., 2010. The microbe-mineral environment and gypsum neogenesis in a Le Campion Alsumard, T., 1979. Les cyanophycees endoliths marins: systematique,
weathered polar evaporate. Geobiology 8, 293–308. ultrastructure, ecologie et biodestruction. Oceanologica Acta 2, 143–156.
Cockell, C.S., Stokes, M.D., 2004. Widespread colonization by polar hypoliths. Lucas, K., 1979. The effects of marine microphytes on carbonate substrata.
Nature 431, 414. Scanning Electron Microscopy 11, 447–456.
Danin, A., Garty, J., 1983. Distribution of cyanobacteria and lichens on hillsides of Mainguet, M., 1972. Le modele de gres. Institut Geographique Nationale, Paris.
Negev highlands and their impact on biogenic weathering. Zeitschrift für Matthews, J.A., Owen, G., 2008. Endolithic lichens, rapid biological weathering and
Geomorphologie 27, 423–444. Schmidt hammer R-values on recently exposed rock surfaces: storbreen glacier
Dorn, R.I., Oberlander, T.M., 1981. Microbial origin of desert varnish. Science 213, foreland, Jotunheim, Norway. Geografiska Annaler 90A, 287–297.
1245–1247. Migon, P., Latocha, A., 2008. Enhancement of cultural landscape by
Drees, K.P., Neilson, J.W., Betancourt, J.L., et al., 2006. Bacterial community geomorphology. A study of granite parklands in the west Sudetes, SW Poland.
structure in the hyperarid core of the Atacama Desert, Chile. Applied and Geografia Fisica et Dinamica Quaternaria 31, 195–203.
Environmental Microbiology 72, 7902–7908. Mumby, P.J., Hastings, A., Edwards, H.J., 2007. Thresholds and the resilience of
Duerden, J.E., 1902. Boring algae as agents in the disintegration of corals. Bulletin, Caribbean coral reefs. Nature 450, 98–101.
American Museum of Natural History 14, 323–333. Nadson, G.A., 1927. Les algues perforantes, leur distribution et leur role dans la
Duncan, P.M., 1876. On some thallophytes parasitic within recent Madreporia. nature. Compte Rendus de l’Academie des Sciences 184, 1015–1017.
Proceedings of the Royal Society 25, 238–256. Omelon, C.R., Pollard, W.H., Ferris, F.G., 2007. Inorganic species distribution and
Etienne, S., 2002. The role of biological weathering in periglacial areas: a study of microbial diversity within high Arctic cryptoendolithic habitats. Microbial
weathering rinds in south Iceland. Geomorphology 47, 75–86. Ecology 54, 740–752.
Folk, R.L., Roberts, H.H., Moore, C.H., 1973. Black phytokarst from Hell, Cayman Pace, N.R., 1997. A molecular view of microbial diversity and the biosphere.
Islands, British West Indies. Bulletin, Geological Society of America 84, Science 276, 734–740.
2351–2360. Perry, R.S., Lynne, B.Y., Sephton, M.A., Kolb, V.M., Perry, C.C., Staley, J.T., 2006.
Friedmann, E.I., 1982. Endolithic microorganisms in the Antarctic cold desert. Baking black opal in the desert sun: the importance of silica in desert varnish.
Science 215, 1045–1053. Geology 34, 537–540.
Friedmann, E.I., Ocampo, R., 1976. Endolithic blue-green algae in the Dry Valleys: Purdy, E.G., Kornicker, L.S., 1958. Algal disintegration of Bahamian limestone
primary producers in the Antarctic desert ecosystem. Science 193, 1247–1249. coasts. Journal of Geology 66, 96–99.
Fritsch, F.E., 1907. The role of algal growth in the colonization of new ground and Schneider, J., 1976. Biological and inorganic factors in the destruction of limestone
in the determination of scenery. Geographical Journal 30, 531–548. coasts. Contributions in Sedimentology 6, 112.
Fry, E.J., 1922. Some types of endolithic limestone lichens. Annals of Botany 36, Schneider, J., Le Campion-Alsumard, T., 1999. Construction and destruction of
541–562. carbonates by marine and freshwater cyanobacteria. European Journal of
Fry, E.J., 1927. The mechanical action of crustaceous lichens on substrata of shale, Phycology 34, 417–426.
schist, gneiss, limestone and obsidian. Annals of Botany 41, 437–460. Shachak, M., Jones, C.G., Granot, Y., 1987. Herbivory in rocks and the weathering
Golubic, S., Brent, G., Le Campion, T., 1970. Scanning electron microscopy of of a desert. Science 236, 1098–1099.
endolithic algae and fungi using a multipurpose casting-embedding technique. Skidmore, M., Anderson, S., Sharp, M., Foght, J., Lanoil, B.D., 2005. Comparison
Lethaia 3, 203–209. of microbial community compositions of two subglacial environments reveals a
Golubic, S., Friedmann, E.I., Schneider, J., 1981. The lithobiontic ecological niche, possible role for microbes in chemical weathering processes. Applied and
with special reference to microorganisms. Journal of Sedimentary Petrology 51, Environmental Microbiology 71, 6986–6997.
475–478. Steven, B., Briggs, G., McKay, C.P., Pollard, W.H., Greer, C.W., Whyte, L.G., 2007.
Golubic, S., Perkins, R.D., Lukas, K.J., 1975. Boring microorganisms and Characterization of the microbial diversity in a permafrost sample from the
microborings in carbonate substrates. In: Frey, R.W. (Ed.), The Study of Trace Canadian high Arctic using culture-dependent and culture-independent methods.
Fossils. Springer, Heidelberg, pp. 229–259. FEMS Microbiology Ecology 59, 513–523.
Gorbushina, A.A., 2007. Life on the rocks. Environmental Microbiology 9, Swinchatt, J.P., 1969. Algal boring: a possible depth indicator in carbonate rocks
1613–1631. and sediments. Bulletin, Geological Society of America 80, 1391–1396.
Gorbushina, A.A., Boettcher, M., Brumsach, H.-J., Krumbein, W.E., Vendrell-Saz, M., de la Torre, J.R., Goebel, B.M., Friedmann, E.I., Pace, N.R., 2003. Microbial
2001. Biogenic forsterite and opal as a product of biodeterioration and lichen diversity of cryptoendolithic communities from the McMurdo Dry Valleys,
stromatolite formation in table mountain systems (tepuis) in Venezuela. Antarctica. Applied and Environmental Microbiology 69, 3858–3867.
Geomicrobiology Journal 18, 117–132. Tribollet, A., Golubic, S., 2005. Cross-shelf differences in the pattern and pace of
Heindel, K., Wisshak, M., Westphal, H., 2009. Microbioerosion in Tahitian reefs: a bioerosion of experimental carbonate substrates exposed for 3 years on the
record of environmental change during the last deglacial sea level rise (IODP northern Great Barrier Reef, Australia. Coral Reefs 24, 422–434.
310). Lethaia 42, 322–340. Tudhope, A.W., Risk, M.J., 1985. Rate of dissolution of carbonate sediments by
Hodson, A.J., Aneiso, A.M., Tranter, M., et al., 2008. Glacial ecosystems. Ecological microboring organisms, Davies Reef, Australia. Journal of Sedimentary Petrology
Monographs 78, 41–67. 55, 440–447.
Huber, M.J., Jadin, M.F., 1892. Sur une nouvelle algae perforante d’eau douce. Uroz, S., Calvaruso, C., Turpault, M.-P., Frey-Klett, P., 2009. Mineral weathering
Journal of Botany 6, 278–286. by bacteria: ecology, actors and mechanisms. Trends in Microbiology 17,
Hutchens, E., 2009. Microbial selectivity on mineral surfaces: possible implications for 378–387.
weathering processes. Fungal Biology Reviews (http://dx.doi.org/10.1016/ Viles, H.A., 1987. Blue-green algae and terrestrial limestone weathering on Aldabra
j.fbr.2009.10.002)?. Atoll: an SEM and light microscope study. Earth Surface Processes and
Johnston, C.G., Vestal, J.R., 1991. Photosynthetic carbon incorporation and Landforms 12, 319–330.
turnover in Antarctic cryptoendolithic microbial communities: are they the Viles, H.A., 2008. Understanding dryland landscape dynamics: do biological crusts
slowest growing communities on Earth? Applied and Environmental hold the key? Geography Compass 3. (http://dx.doi.org/10.1111/j.1749-
Microbiology 57, 2308–2311. 8198.2008.00099.x)?.
Kahle, C.F., 1977. Origin of subaerial Holocene calcareous crusts: role of algae, Viles, H.A., 2012. Microbial geomorphology: a neglected link between life and
fungi and spar-micritization. Sedimentology 24, 413–435. landscape. Geomorphology 157–158, 6–16.
270 Microbioerosion and Bioconstruction

Walker, J.J., Pace, N.R., 2007. Endolithic microbial ecosystems. Annual Review of Wong, F.K.Y., Lau, M.C.Y., Lacap, D.C., Aitchison, J.C., Cavan, D.A., Pointing, S.B.,
Microbiology 61, 331–347. 2010. Endolithic microbial colonization of limestone in a high altitude arid
Wierzchos, J., Camara, B., de los Rios, A., et al., 2011. Microbial colonization of environment. Microbial Ecology 59, 689–699.
Ca-sulfate crusts in the hyperarid core of the Atacama Desert: implications for
the search for life on Mars. Geobiology 9, 44–60.

Biographical Sketch

Heather Viles is professor of biogeomorphology and Heritage Conservation in the School of Geography and the
Environment, University of Oxford, UK, and visiting research fellow in the School of Geography, Archaeology and
Environmental Sciences, University of the Witwatersrand, South Africa. Her research focuses on biogeomor-
phology, weathering (mainly in extreme environments), and the deterioration and conservation of building stone.
She has carried out field research in many places, including Essex, UK, NW Australia, Germany, Namibia, South
Africa, Libya, Washington State, USA, and Aladabra Atoll, Seychelles. Her publications include over 115 papers in
refereed journals and edited volumes, and she has edited nine books and collections of papers, including Bio-
geomorphology (Blackwell, 1988). She is the author of a number of books including Salt Weathering Hazards
(John Wiley, 1997), and A Very Short Introduction to Landscapes and Geomorphology (Oxford University Press,
2010), both with Andrew Goudie. She is currently vice-president (Expeditions and Fieldwork) of the Royal
Geographical Society (with IBG), and vice chair of the British Society for Geomorphology. She holds an MA in
geography from the University of Cambridge, and a DPhil in geography from the University of Oxford.
 12.18 The Geomorphic Impacts of Animal Burrowing and Denning
DR Butler, CJ Whitesides, JM Wamsley, and SG Tsikalas, Texas State University – San Marcos, San Marcos, TX, USA
r 2013 Elsevier Inc. All rights reserved.

12.18.1 Introduction 271

12.18.2 Haplotaxida – Earthworms 272
12.18.3 Isoptera and Hymenoptera 272
12.18.3.1 Termites and Ants 272
12.18.3.2 Bees 272
12.18.4 Salmoniformes – Salmon and Trout 273
12.18.5 Testudines – Gopher Tortoises and Related Species 273
12.18.6 Procellariiformes – Wedge-tailed and Sooty Shearwaters 274
12.18.7 Lagomorphs (Lagomorpha) – Rabbits and Pikas 274
12.18.8 Rodents (Rodentia) 274
12.18.8.1 Gophers 274
12.18.8.2 Ground Squirrels 275
12.18.8.3 Marmots 275
12.18.8.4 Voles and Zokors 276
12.18.9 Carnivores (Carnivora) 276
12.18.9.1 Badgers 276
12.18.9.2 Grizzly Bears 277
12.18.10 Soricomorpha – Moles 277
12.18.11 Conclusions 277
References 277

Glossary Tumuli A mound of earth, in this instance deposited on

Fossorial Adapted for or used in burrowing or digging. the surface by the burrowing efforts of bees.
Redd A spawning nest made by a salmonid (salmon or Zoogeomorphology The study of the geomorphic effects
trout) fish. of animals.
Sett A badger’s underground den, usually consisting of a
network of tunnels.

Abstract

Fossorial animals have a substantial impact on landforms and landform processes in both a direct and indirect manner.
Perhaps the first notable research on fossorial animals in zoogeomorphology can be attributed to Charles Darwin’s study of
earthworms and their impact on soil characteristics. In this chapter, we examine a sample of animals and provide an
overview of their impacts. The outline of this review categorizes animals based on their taxonomic order and includes:
Haplotaxida, Isoptera and Hymenoptera, Salmoniformes, Testudines, Procellariformes, Lagomorphs, Rodentia, Carnivora,
and Soricomorphia. Several examples of direct soil displacement via burrowing and denning are provided and discussed. A
majority of the zoogeomorphic research on fossorial animals has focused on impacts such as soil erosion, slope failure or
mass wasting events, and biogeochemical alteration of soil. The transition from quantifiable studies of animal geo-
morphology to the secondary impacts on ecosystems is a logical advance of the study of zoogeomorphology; however,
before doing so, additional quantitative data on direct geomorphic impact is required.

12.18.1 Introduction

Fossorial, or burrowing, animals exist around the world. Some

Butler, D.R., Whitesides, C.J., Wamsley, J.M., Tsikalas, S.G., 2013. The are frequently seen on the surface, are interesting to watch,
geomorphic impacts of animal burrowing and denning. In: Shroder, J.
and are known to many. Others reside in remote locations of
(Editor in Chief), Butler, D.R., Hupp, C.R. (Eds.), Treatise on
Geomorphology. Academic Press, San Diego, CA, vol. 12, the Earth where they are known to relatively few and spend
Ecogeomorphology, pp. 271–280. very little time outside their subterranean homes (the Plateau

Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00334-1 271

272 The Geomorphic Impacts of Animal Burrowing and Denning

Habits (cf. Meysman et al., 2006, for an examination of the

impact his book has had). Since that time, earthworm studies
have burgeoned to include all aspects of earthworm ecology
(see the edited volumes by Edwards and Lofty (1972); Edwards
(1998), and Hendrix (1995) for more information on bio-
logical, ecological, and distributional information of earth-
worms). The primary geomorphic effect of earthworms is the
formation of casts, which are surficial, fine-grained material
deposits. In Luxembourg, Hazelhoff et al. (1981) quantified the
rate of earthworm cast production and determined that earth-
worms were responsible for the surficial deposition of 1.5 kg
m 2 yr 1 of soil. In the tropics, Goudie (1988) found earth-
worm cast production to be much higher and concluded that
the typical range was 2–30 kg m 2 yr 1 of soil. Henrot and
Brussard (1997) concluded that earthworms were capable of
depositing between 260 and 570 kg ha 1 of casts annually, and
also found that cast production was not constant over the
course of a year but rather tended to have a period of high cast
production followed by a period of low production. The large
variety of earthworms and their broad distribution has resulted
in some scenarios where earthworms represent the highest in-
dividual contributor to overall soil biomass (James, 1991). It
therefore comes as no surprise that, in addition to direct geo-
morphic effects of cast production, earthworms strongly influ-
ence the secondary geomorphic processes of infiltration, surface
wash, rain-splash detachment, and soil creep. Anderson (1988)
noted that earthworms were capable of producing up to 220
tunnel openings (3–5 mm in diameter) per square meter,
which increased a soil’s infiltration capacity (Goudie, 1988;
Schrader and Joschko, 1991). Although surface wash tends to
Figure 1 Two examples of termite mounds from the Southern
be the most logical method of cast removal, rain-splash de-
Hemisphere. (a) Mound in subtropical forest in New South Wales,
tachment has been shown to be the catalyst required to break
Australia. 50-mm lens cap on crest of mound for scale. (b) Mound in
Kruger National Park, South Africa. down casts and prepare them for transport (Le Bayon and
Binet, 1999). In fact, Le Bayon and Binet (2001) further sug-
zokor of Tibet, for example). Regardless of location or popu- gested that the presence of earthworm casts retarded surface
larity, all of these creatures create burrows, setts, or dens that wash because casts increased surface roughness and slowed
attest to their often unseen, yet productive, lives. In doing water flow, and the presence of earthworm activity increased
so, they have widespread and characteristically significant infiltration and percolation and resulted in less surface flow
geomorphic impacts upon the landscape (see Figure 1, see (Jouquet et al., 2008).
Chapter 12.16). The purpose of this review is to highlight a
variety of fossorial animals that have substantial impact on
12.18.3 Isoptera and Hymenoptera
landforms and landform processes. This review does not dis-
cuss all burrowing species known to have an impact on geo-
12.18.3.1 Termites and Ants
morphology, but provides a sampling of animals and an
overview of their impacts. For a thorough review of burrowing It is well established that ants (order Hymenoptera) and ter-
and denning by animals, readers are referred to Meadows and mites (order Isoptera) can increase surface infiltration by
Meadows (1991). Our purpose is to utilize examples that il- forming soil macropores during nest establishment and sub-
lustrate the geomorphic potential of fossorial animals. The sequent growth (Cerd`a and Doerr, 2010). Ant mounds and
outline of this review categorizes animals based on their termiteria (termite mounds) (Figure 1) are surface manifest-
taxonomic order. Each category contains a brief sampling of ations of much larger underground tunnel systems. The geo-
the species that comprise each order. Common order names morphic impacts of ant and termite tunneling and associated
are supplied when appropriate. mound building are sufficiently profound that a separate
chapter in this treatise volume (see Chapter 12.19) is devoted
to those impacts, and readers are referred to that chapter for
12.18.2 Haplotaxida – Earthworms additional details.

The idea that earthworms are capable of influencing soil ero-

12.18.3.2 Bees
sion and soil properties was described in detail by Charles
Darwin (1881), in his final major work The Formation of Vege- Bees, wasps, and sawflies are also members of the order
table Mould through the Action of Worms With Observation of Their Hymenoptera, in addition to ants. Thousands of bee species
 The Geomorphic Impacts of Animal Burrowing and Denning 273

are ground nesters, digging central underground tunnels includes three species in North America, all of which have at least
whose excavated soil is pushed to the surface and deposited as localized geomorphic significance. The gopher tortoise (Gopherus
tumuli (analogous to the castings of earthworms) (Cane, polyphemus) has a geomorphic impact of considerable magnitude.
2003). The quantities of soil pushed to the surface by ground- It is considered ‘‘a key species in Florida’s xeric communities’’
nesting bees can be impressive. Cane (2003) estimated that (Moler, 1992). Its tunnel-building habits create underground
alkali bees in the 155 km2 Touchet Valley of southeastern habitat for a number of other species, and the soil displaced by
Washington State, USA, annually brought 87 500 kg (96 tons) their tunneling efforts alters surficial habitat. The size and density
of soil to the surface in their tumuli, much of which was of the tunnels affect the hydrology of the sandy soils of the
subsequently eroded by wind and rain. Although this is, ac- coastal plain, and the frequently traveled tortoise paths compact
cording to Cane (2003), an extreme example, it is clear that the soil in the area around the burrow entrance. Gopher tortoises
under favorable circumstances with high concentrations, bees spend approximately 75% of their lives within the burrow. The
can be significant zoogeomorphic agents of bioturbation and typical burrow has a semi-circular entrance, with the bottom
subsequent erosion. being the flat side and the upper part arched to accommodate the
domed shell of the occupant. Gopher tortoise burrows have an
average length of 5 m and typically extend more than 2 m below
12.18.4 Salmoniformes – Salmon and Trout the surface of the ground (Lips, 1991; Butler, 1995). Burrows may
be as long as 13 m and as deep as 3 m below the surface, with the
Salmon and trout (salmonids) are recognized nest builders deepest known burrow extending more than 6 m below ground
and have a geomorphic impact on the streams in which they level (Gopher Tortoise Council, 1998). The burrow is always
spawn (for non-nesting geomorphic effects caused by fish, see wider than its occupant’s length, allowing the tortoise to turn
Butler, 1995). Female fish create nests by lateral body flexures, around inside the burrow. The burrow terminates in an end
which create a vortex by which sediment is lifted from the river chamber, which is larger than the burrow itself.
bed and transported downstream (van den Berghe and Gross, An individual tortoise may dig several burrows during its
1984). Salmon nests, or redds, are a rudimentary form of lifetime. Females may dig nest tunnels in which they do not
burrow that typically consist of a shallow pit, 10–20 cm in live, and as resource availability changes tunnels may be
depth and 1–2 m in length, that are oriented parallel to the abandoned. Tortoises have been observed returning to tunnels
direction of the current (van den Berghe and Gross, 1984; that had been abandoned for at least 5 months (Brode, 1959),
Crisp and Carling, 1989; Kondolf et al., 1993). The action of suggesting seasonal usage of individual tunnels. In a study of
nest building results in coarse gravel nests and the removal of burrow abandonment conducted over a 5-year period, 20% of
fine sediment. In the Colorado River, Kondolf et al. (1989) the burrows vacated for more than 1 year were eventually
found that spawning rainbow trout (Oncorhynchus mykiss) reoccupied (Aresco and Guyer, 1999).
created redds that had only 2% fine sediment (o0.85 mm) Gopher tortoises play important roles in conferring het-
compared to 6% fine sediment in areas without spawning erogeneity on soil resources and vegetation through their
trout. In Alaska, sockeye salmon (Oncorhynchus nerka) were tunnel- and burrow-building activities. The large quantity of
excluded from river stretches normally used for spawning in a soil excavated from the tunnel is spread around the tunnel
study by Moore et al. (2004). The authors reported accumu- entrance, resulting in a mound of displaced soil known as an
lations of fine sediment during spawning season in the ex- apron (Butler, 1995). With significant rainfall, additional
cluded reaches. Additionally, redd construction had secondary bioturbation occurs as mound materials wash back into bur-
impacts on algal biomass and invertebrate assemblages rows or off onto adjacent surfaces (Butler, 1995).
(Moore et al., 2004). The desert tortoise (Gopherus agassizii) fills a similar niche
Two separate studies in British Columbia, Canada, place the in the Mojave and Sonoran deserts of the southwestern USA
geomorphic impact of redd construction in proper perspective. and northern Mexico. The size and type of burrow depend
Gottesfeld et al. (2004) used magnetically tagged particles to largely on the geographic location of the individual tortoise
investigate the effects of sockeye salmon redd construction. of this species. Tortoises in the Mojave Desert, representing
They found that up to 100% of streambed particles were mo- the northern portion of its range, dig burrows up to 12 m in
bilized by salmon, and depth of burial of particles averaged up length. These burrows may be occupied by more than one
to 14 cm. They noted that the amount of vertical mixing of bed tortoise. Desert tortoises in the Sonoran Desert construct
sediments was on the same order of magnitude as flood events. much smaller burrows, usually less than 2 m long (Lawler,
Hassan et al. (2008) illustrated on four separate streams in 2000). Many tortoises in the Sonoran Desert portion of their
British Columbia that the activity of salmon engaged in redd range do not burrow at all, instead seeking refuge under rocky
construction moved an average of almost half of the annual outcrops.
bed load yield, substantially influenced total sediment trans- The Texas tortoise (Gopherus berlandieri) utilizes very short
port rates, and altered alluvial-reach morphology. burrows, also called pallets, during the active season. In some
cases, burrows are not even constructed. Although active
season pallets are rarely more than a meter in length, hiber-
12.18.5 Testudines – Gopher Tortoises and Related nation takes place in deeper burrows, sometimes made by
Species enlarging the existing burrows of other animals, including
wood rat middens (Rose and Judd, 1975). Little other quan-
The order Testudines includes turtles and tortoises, many of titative data on burrow size and numbers are available for this
which have only limited geomorphic impact. The genus Gopherus species.
274 The Geomorphic Impacts of Animal Burrowing and Denning

12.18.6 Procellariiformes – Wedge-tailed and Sooty different coastal environment in the Netherlands, Rutin
Shearwaters (1992) documented extensive rabbit caves and noted both
direct geomorphic effects caused by slope failure as well as
Among avian species, burrowing, or nest-cavity excavating, is secondary effects associated with increased potential for fluvial
uncommon (Terres, 1980); however, it is more common in and aeolian transport. Additional influences of rabbits on soil
colonial seabirds, such as petrels, shearwaters, storm petrels, and vegetation characteristics can be found in Gálvez et al.
diving petrels, and some auks and penguins (Furness, 1991). (2008) and references therein.
Colonial seabirds nest in extremely high numbers along Plateau pikas (Ochotona curzoniae) on the Tibetan Plateau
coastal regions and on islands. Butler (1995) reviewed ex- have been regarded as a keystone species that is important in
amples of burrowing and attendant geomorphic impacts by maintaining biodiversity on the plateau (Smith and Foggin,
several colonial seabird species, including the destruction of 1999). Pika burrows are commonly used as breeding habitat
Grassholm Island by the burrowing and excavating actions of for a variety of reptiles and birds (Smith and Foggin, 1999; Lai
puffins in the early twentieth century, and readers are referred and Smith, 2003), and both Daurian (Ochotona daurica) and
to that volume. Here we describe recent research focused on plateau pika (Ochotona curzoniae) have been shown to have an
the burrowing activities of the wedge-tailed shearwater impact on vegetation richness and distribution (Smith and
(Puffinus pacificus) and sooty shearwater (Puffinus griseus) that Foggin, 1999; Komonen et al., 2003; Wesche et al., 2007;
post-dates the release of Butler’s volume. Xinghu et al., 2007), which has obvious geomorphic ramifi-
The wedge-tailed shearwater is a burrowing bird that nests cations in terms of soil infiltration, runoff, and surface ero-
in dense colonies on islands throughout the tropical and sion. Unfortunately, until recently, studies involving pika
subtropical extents of the Pacific and Indian Oceans (Bancroft burrows have focused on secondary geomorphic effects with
et al., 2004a). Wedge-tailed shearwaters excavate ground- little commentary on direct zoogeomorphology.
nesting burrows in densities of 0.01–0.93 burrows m 2 In 2007, Xinghu et al. examined pika burrow entrances in
(Bancroft et al., 2004a). The soil bioturbated by the birds is Tibet and recorded 2100 entrances per hectare. The authors
enriched by their guano, such that the bioturbated soils are also recorded pika mounds, debris piles of burrow ejecta, at
drier, warmer, and possess greater bulk density, wetting cap- densities of 330 per hectare. Local herdsmen in Tibet reported
acity, and hydraulic conductivity than undisturbed soils that soil excavated by pika was highly susceptible to surface
(Bancroft et al., 2005a). The shearwaters select burrow sites to runoff and aeolian erosion, both secondary geomorphic pro-
minimize the risk of burrow collapse, and are known to clear cesses of mound creation, and mound sediment was com-
and re-excavate damaged burrows (Bancroft et al., 2005b). pletely eroded within 2–3 years (Xinghu et al., 2007). As with
Roughly, two-thirds of the burrows in Bancroft et al.’s (2005b) the majority of fossorial mammals descussed below, add-
study area collapsed over the course of 1 year, and three- itional research is required to identify the specifics of pika
quarters of those burrows were subsequently re-excavated the geomorphology.
following year. In doing so, at least 11.5 t ha 1 yr 1 (8.5 m3
ha 1 yr 1) of soil was displaced through this process (Bancroft
et al., 2004b, 2005b).
12.18.8 Rodents (Rodentia)
McKechnie (2006) reported on the burrow excavations of a
related bird, the sooty shearwater of New Zealand. Burrow
12.18.8.1 Gophers
depths up to 90 cm were measured on three different islands
on the southern coast of New Zealand. Burrow lengths at in- The term ‘gopher’ is generic and is commonly applied to a
dividual study plots ranged between 2257 and 16 617 m ha 1, variety of ground-dwelling animals. In this brief synopsis of
and burrow volumes ranged from about 88 to 650 m3 ha 1. gopher-related geomorphology, we will focus solely on the
Between 18% and 34% of the ground surface of the three Geomyidae family that consists of a variety of species. Gophers
islands was undermined by burrow space. In a separate study of the Geomyidae family are widely distributed in the western
on a different island in the same area, Newman et al. (2008) United States, Canada, and northern Mexico as well as por-
estimated a population of over 5 million sooty shearwaters on tions of the southeastern United States (Teipner et al., 1983).
an island of only 930 ha. They extrapolated that the island Extensive research on gopher activity has revealed that gophers
contained over 1.67 million burrow entrances, effectively are capable of altering the chemical soil and vegetative prop-
honeycombing the entire island. erties surrounding their burrows (Grant et al., 1980; Reichman
and Smith, 1985; Carlson and Crist, 1999; Rogers et al., 2001;
Sherrod and Seastedt, 2001). These alterations may produce
12.18.7 Lagomorphs (Lagomorpha) – Rabbits and secondary geomorphic effects and result in increased erosion
Pikas due to altered plant communities and distributions.
In comparison to studies of the effects of gophers upon soil
The characteristics of rabbit dens and warrens have been well properties, fewer studies exist that examine the direct geo-
described and documented (Parker et al., 1976; Kolb, 1985, morphic impacts of gophers. However, compared to many
1991a, b) but few studies have expanded burrow character- other animals of the Rodentia order, much is known about
istics to direct geomorphic effects caused by rabbits. Voslam- gopher-induced geomorphology. An early report by Ellison
ber and Veen (1985) estimated that rabbit burrows were (1946) reported that northern pocket gophers (Thomomys
responsible for excavating between 475 and 71 308 kg of soil talpoides) in subalpine environments of Utah were responsible
per hectare on forested hillslopes in Belgium. In a very for bringing 11.0–15.5 t ha 1 yr 1 of soil to the surface. Burns
 The Geomorphic Impacts of Animal Burrowing and Denning 275

(1979) stated that northern pocket gophers in the Front Range

of Colorado were capable of lowering the average surface
elevation of the area by 0.0037 cm yr 1. By comparison, Burns
(1979) stated that the average wind and water erosion within
the study site of the alpine tundra accounted for an average
surface lowering of 0.000 1 cm yr 1. Extensive gopher impacts
are not restricted to northern pocket gophers at high ele-
vations. In coastal California, Black and Montgomery (1991)
calculated that gopher (Thomomys bottae) spoil mounds aver-
aged 1100 cm3 and determined that gopher activity was the
dominant cause of sediment transport in the study area. Re-
cently, both Butler and Butler (2009) and Knight (2009) de-
scribed gopher eskers, the sinuous, anastomosing networks of
infilled tunnels that are exposed by spring snowmelt in high
mountain areas of the American west (Figure 2). Both papers
noted that the sediment comprising the gopher eskers breaks
down over the course of several weeks, with the sediment
subsequently distributed by surface wash, runoff, and aeolian
processes to adjacent areas of the alpine environment.
Butler and Butler (2009) also noted that the sediment com-
prising the gopher eskers was less compacted than that of the
surrounding undisturbed landscape, suggesting that de-
composing eskers would be amenable sites for plant estab-
lishment that would increase local site heterogeneity.

12.18.8.2 Ground Squirrels

Butler (1995) identified three important studies highlighting the
effects of three different ground squirrel species on local geo-
morphology; they include Arctic ground squirrels (Citellus un-
dulates; Price, 1971), Columbian ground squirrels (Spermophilus
columbianus; Smith and Gardner, 1985), and the Townsend
ground squirrel (Spermophilus townsendii; Laundré, 1993). In
these studies, the authors recognized direct geomorphic impacts Figure 2 Two examples of decomposing gopher eskers and
of ground squirrels that included downslope transport of ma- associated sediment aprons, Sangre de Cristo Mountains, southern
Colorado, USA. 50-mm lens cap for scale in both photos.
terial during the burrowing process (Price, 1971; Smith and
Gardner, 1985), weakening of soil structure (Price, 1971), and
increased water infiltration associated with burrows (Laundré,
year and species abundance. During winter months, marmots
1993, 1998). Specific calculations estimated that Arctic ground
generally hibernate in burrows comprising a burrow network
squirrels were capable of excavating and depositing spoils in
in which the animals spend approximately 80% of their lives
mounds at the rate of 19.5 t ha 1 yr 1 (Price, 1971), and
(Barash, 1989).
Columbian ground squirrels were capable of bringing between
The subterranean lifestyle of marmots and their extensive
1.12 and 1.35 t ha 1 yr 1 of soil to the Earth’s surface (Smith
burrow system results in the excavation of soil and the creation
and Gardner, 1985). Although these rates may be orders of
of new topographic features that modify hill slopes and may
magnitude lower than other mass-wasting processes (Smith and
result in slope failure and other mass wasting events (Figure 3).
Gardner, 1985), the geomorphic activity of ground squirrels has
Past research has examined the intricacies of marmot burrow
been suggested as a controlling factor of valley asymmetry in
location and structure (Beltz and Booth, 1952; Svendsen, 1976;
Arctic environments (Price, 1971).
Barash, 1989; English and Bowers, 1994; Bassano and Peracino,
1997) but few studies have examined the amount of material
discharged by the animal and the geomorphic effect of the dis-
12.18.8.3 Marmots
charged material. In the eastern Pamirs, Tadzhiyev and Odi-
Marmots are large rodents that comprise the genus Marmota noshoyev (1987) stated that the amount of sediment removed
and consist of six North American species and eight Eurasian by red marmot (Marmota caudate) excavations ranged from 5 to
species (Barash, 1989). Marmots are typically steppe dwelling 28 m3 ha 1. In Colorado, USA, Plaster (2003) examined the
animals in Eurasia and subalpine dwellers in North America amount of soil volume displaced by yellow-bellied marmots
(Butler, 1995) with one species, Marmota monax, residing in (Marmota flaviventris) and determined that the potential geo-
agricultural areas of eastern North America (Barash, 1989). morphic effect of marmots in the study area ranged between 59.7
Marmots are primarily vegetarians that consume a variety of and 109.5 m3 km 2. Hall et al. (1999) estimated the amount of
meadow vegetation. However, forage is dependent on time of sediment removal by all rodent activity (including marmots) in
276 The Geomorphic Impacts of Animal Burrowing and Denning

(Edelman, 2003). Yoshihara et al. (2010) examined the re-

sponses of vegetation to soil disturbances by Siberian marmots
in Mongolia. Additional secondary geomorphic impacts cre-
ated by marmots throughout North America and Eurasia have
been reported by Semenov et al. (2001, 2003).

12.18.8.4 Voles and Zokors

Unlike gophers mentioned above, voles (members of the
Cricetidae family and consisting of a variety of species) do not
move large volumes of soil and the vole burrow structure is
commonly inhabited and maintained by generations of voles
that allow the burrow network to become a semi-permanent
feature in the soil profile (Gervais et al., 2010). Despite the
longevity of burrow networks, vole populations can fluctuate
rapidly due to high birthrates and short gestational periods
(Elton, 1942; Schaffer and Tamarin, 1973), which may result
in burrow entrance densities in excess of 10 per m2 (Gervais
et al., 2010). Several studies have examined the subterranean
nature of burrow networks (Kalisz and Davis, 1992; Laundré
and Reynolds, 1993; Harper and Batzli, 1996; Gervais et al.,
2010) but unfortunately, despite the high density of vole
populations and the knowledge of their burrow systems, no
known studies have exclusively examined the primary geo-
morphic effects of vole diggings. Research has instead focused
on the secondary geomorphic effects, with particular interest
in vegetation change surrounding burrows (Gibson, 1989;
Kalisz and Davis, 1992; Milton et al., 1997; Fehmi and Bar-
tolome, 2002; Questad and Foster, 2007), biogeochemical
alteration of soil near vole disturbances (Gervais et al., 2010),
and the impact of soil texture on burrow location (Laundré
and Reynolds, 1993).
Zokors (Myospalax spp.) are small (approximately 220 g for
Figure 3 (a) Material excavated from adjacent marmot burrow, female and 270 g for male) Asiatic rodents that spend most of
Olympic National Park, Washington. (b) Olympic marmots on lookout their lives in subterranean burrow systems (Zhang and Liu,
on porches of sediment adjacent to burrows, Olympic National Park, 2003; Zhang, 2007b). Average densities of animals are 15 per
Washington. hectare, with estimates ranging from 5 to 70 per hectare (Wang
and Fan, 1987). Daily digging results in the formation of
surface spoil piles known as mounds that consist of an average
the Canadian Rocky Mountains to be between 0.6408 and volume of 0.007 m3 per zokor per day (Wang and Fan, 1987).
600 m3 km 2 yr 1. Despite these few studies quantifying mar- The majority of zokor studies examines secondary effects
mot excavations, little is documented about direct, marmot- related to vegetation and soil characteristics (Zhang and Liu,
induced geomorphology. Additionally, further research is needed 2003; Zhang, 2007a; Wang et al., 2008; Li et al., 2009).
to evaluate how excavated soil affects mass wasting.
Secondary geomorphic impacts, commonly termed bio-
turbation or biopedoturbation (see Whitford and Kay, 1999),
12.18.9 Carnivores (Carnivora)
of marmot activity have been studied more intensively than
direct geomorphic impacts, with the majority of secondary
12.18.9.1 Badgers
studies emphasizing soil characteristics and vegetation. Choler
(2005) established that marmot (Marmota marmota) activity in Badgers are members of the Mustelidae family that consists of
the southwestern Alps of France modified microtopography subfamilies and many species. The large badger family is dis-
and soil properties, creating vegetation gaps that provided tributed throughout the world but the European/Eurasian (Meles
opportunity for new species recruitment and ultimately meles) and the North American (Taxidea taxus) badgers are most
maintained species diversity. In the Olympic Mountains of the heavily studied (Butler, 1995). Unlike the rodents or lago-
State of Washington, USA, del Moral (1984) noted that morphs, badgers are carnivores and much of their digging is in
Olympic marmots (Marmota olympus) reduced palatable sub- pursuit of other fossorial mammals (Verbeek, 1965; Murie, 1992;
alpine vegetation surrounding burrows, whereas less palatable Armitage, 2004). In addition to predation, badgers also construct
vegetation increased in density. Olympic marmots (Marmota resting burrows and natal dens; the latter of which are typically
olympus) also grazed subalpine meadows and resulted in twice the size of resting burrows and are associated with larger
higher plant community diversity in subalpine environments spoil mounds (Lindzey, 1976). Past research has highlighted
 The Geomorphic Impacts of Animal Burrowing and Denning 277

several secondary geomorphic effects of North American badger in humid deciduous forests. In Imeson’s study area, moles
burrows, including the potential to increase water infiltration were responsible for the surface deposition of 1940 m3
due to soil aeration and thus reduce surface runoff (Butler, km 2 yr 1 of soil. Imeson and Kwaad (1976) suggested that
1995), modifying soil characteristics via mixing (Long and the geomorphic influence caused by the amount of surficial
Killingley, 1983), and the addition of organic matter (Lindzey, material deposited by moles was secondary to the indirect
1976). Long and Killingley (1983) measured the spatial di- geomorphic effect of rain-splash erosion on mole-exposed
mensions of resting burrows and Lindzey (1976) documented sediment. Additional research is needed to address both the
the size of natal burrows, but few studies have explicitly ad- primary and secondary geomorphic impacts of the American
dressed the direct geomorphic impact of North American bad- garden mole (Scalopus aquaticus) as well as secondary impacts
gers. For an examination of the geomorphic effects of foraging by on erosion and vegetation by the European mole (Talpa
North American badgers, readers are referred to (see Chapter europaea).
12.16).
European badgers typically construct main burrows, or
setts, and outlier burrows (Roper et al., 2001) that remain as 12.18.11 Conclusions
permanent features on the landscape for long periods of time
and can consist of over 100 entrances per hectare for a single As shown by this review, the concept of animals as geo-
sett (Roper, 1992). Despite the longevity and abundance morphic agents is not a new one. Many dated examples testify
of burrows on the landscape, quantifiable rates of badger that geomorphologists and scientists from a variety of discip-
excavations and insights into local long-term impacts are rare. lines have observed and attempted to document and quantify
Voslamber and Veen (1985) reported that badger mounds in the nature of geomorphic disturbances created by animals
the Belgian Ardennes transported 15.05 g cm 1 yr 1 but, (Darwin, 1881; Beltz and Booth, 1952; Verbeek, 1965). Much
when averaged over the entire study site, soil transport asso- of the recent research in this field has focused on indirect
ciated with badgers was 0.0192 g cm 1 yr 1, which the impacts such as biogeochemical change in soils and vegetation
authors deemed as insignificant. dynamics associated with zoogeomorphology. A limited
number of studies have directly addressed the geomorphic
impacts of burrowing animals ranging from bees to salmon to
12.18.9.2 Grizzly Bears shearwater birds to a variety of mammals such as gophers. The
transition from quantifiable studies of animal geomorphology
Bears are among the largest denning animals on the Earth, with
to the secondary impacts that animal geomorphology has on
both grizzly bears and black bears (Ursus americanus) active den
the ecosystem is a logical advance of the study of zoogeo-
excavators (Butler, 1995). The grizzly bear (Ursus arctos) dens
morphology and should be continued. However, in too many
for a period of 5–6.5 months per year, with the majority of dens
cases, very little is known about the primary zoogeomorpho-
being created by excavation into mountain sideslopes or on
logical effects and much more is known about the secondary
broad upland plateaus (Vroom et al., 1980; Butler, 1992;
effects. Although studies examining the indirect effects are
Haroldson et al., 2002; Podruzny et al., 2002; Ciarniello et al.,
needed and will undoubtedly persist, we as scientists have
2005). Many excavated dens collapse during the period of
‘jumped the gun’ and advanced our knowledge of indirect
spring snowmelt and summer rain (Butler, 1995). In a study in
zoogeomorphology before acquiring a sufficiently firm
British Columbia, Canada, Ciarniello et al. (2005) found that
understanding of direct zoogeomorphology. Additional re-
69% of studied dens were excavated annually, with 31% of the
search in this area will undoubtedly provide us with new in-
dens experiencing re-use a second year. Butler (1992) calculated
sights and strengthen our understandings of secondary effects
the volume of sediment displaced downslope for a typical
throughout the ecosystem.
grizzly bear den, using previously published values of den size.
He showed that each den displaces roughly 4.3 m3 of sediment
onto a spoil heap downslope of the tunnel mouth leading to
the interior hibernation chamber. This sediment enters the
References
slope debris cascade where it is dispersed below the mouth of
Anderson, J.M., 1988. Spatiotemporal effects of invertebrates on soil processes.
each individual den.
Biology and Fertility of Soils 6, 216–227.
Aresco, M.J., Guyer, C., 1999. Burrow abandonment by gopher tortoises in slash
pine plantations of the Conecuh National Forest. Journal of Wildlife
Management 63, 26–35.
12.18.10 Soricomorpha – Moles Armitage, K.B., 2004. Badger predation on yellow-bellied marmots. American
Midland Naturalist 151, 378–387.
Moles are similar to burrowing rodents described above except Bancroft, W.J., Garkaklis, M.J., Roberts, J.D., 2004a. Continued expansion of the
Wedge-tailed Shearwater, Puffinus pacificus, nesting colonies on Rottnest Island,
that moles are insectivores. They are productive burrowers and
Western Australia. Emu 104, 79–82.
deposit burrow spoils on the surface as molehills. Molehill Bancroft, W.J., Garkaklis, M.J., Roberts, J.D., 2005a. Burrow building in seabird
density varies greatly. For example, Jońca (1972) documented colonies: a soil-forming process in island ecosystems. Pedobiologia 49, 149–165.
a range of less than 20 to several thousand per hectare. Bancroft, W.J., Hill, D., Roberts, J.D., 2004b. A new method for calculating volume
The size of molehills also varies, but commonly ranges of excavated burrows: the geomorphic impact of Wedge-tailed Shearwater
burrows on Rottnest Island. Functional Ecology 18, 752–759.
between 30–40 cm in diameter and 15–20 cm in height Bancroft, W.J., Roberts, J.D., Garkaklis, M.J., 2005b. Burrow entrance attrition rate
(Jońca, 1972). In the Ardennes of Luxembourg, Imeson in Wedge-tailed Shearwater (Puffinus pacificus) colonies on Rottnest Island,
(1976) found moles to be one of the major slope processes Western Australia. Marine Ornithology 33, 23–26.
278 The Geomorphic Impacts of Animal Burrowing and Denning

Barash, D.P., 1989. Marmots: Social Behavior and Ecology. Stanford University Grant, W.E., French, N.R., Folse, Jr. L.T., 1980. Effects of pocket gopher mounds on
Press, Palo Alto, CA. plant production in shortgrass prairie ecosystems. The Southwestern Naturalist
Bassano, B., Peracino, V., 1997. A robotized system for exploring mammal burrows. 25, 215–224.
Wildlife Society Bulletin 25, 98–100. Hall, K., Boelhouwers, J., Driscoll, K., 1999. Animals as erosion agents in alpine
Beltz, A., Booth, E.S., 1952. Notes on burrowing and food habits of the Olympic zone: some data and observations from Canada, Lesotho, and Tibet. Arctic,
marmot. Journal of Mammalogy 33, 495–496. Antarctic, and Alpine Research 31, 436–456.
van den Berghe, E.P., Gross, M.R., 1984. Female size and nest depth in coho Haroldson, M.A., Ternent, M.A., Gunther, K.A., Schwartz, C.C., 2002. Grizzly bear
salmon (Oncorhynchus kisutch). Canadian Journal of Fisheries and Aquatic denning chronology and movements in the Greater Yellowstone Ecosystem.
Sciences 41, 204–206. Ursus 13, 29–37.
Black, T.A., Montgomery, D.R., 1991. Sediment transport by burrowing Harper, S.J., Batzli, G.O., 1996. Effects of predators on structure of the burrows of
mammals, Marin County, California. Earth Surface Processes and Landforms voles. Journal of Mammalogy 77, 1114–1121.
16, 163–172. Hassan, M.A., Gottesfeld, A.S., Montgomery, D.R., et al., 2008. Salmon-driven bed
Brode, W.E., 1959. Notes on behavior of Gopherus polyphemus. Herpetologica 15, load transport and bed morphology in mountain streams. Geophysical Research
101–102. Letters, 35. http://dx.doi.org/10.1029/2007GL032997.
Burns, S.F., 1979. The northern pocket gopher (Thomomys talpoides): major Hazelhoff, I., Van Hoof, P., Imerson, A.C., Kwaad, F.J.P.M., 1981. The exposure of
geomorphic agent on the alpine tundra. Journal of Colorado-Wyoming Academy forest soil to erosion by earthworms. Earth Surface Processes and Landforms 6,
of Science 2, 86. 235–250.
Butler, D.R., 1992. The grizzly bear as an erosional agent in mountainous terrain. Hendrix, P.F., 1995. Earthworm Ecology and Biogeography in North America. Lewis
Zeitschrift für Geomorphologie 35, 179–189. Publishers, Boca Raton, FL.
Butler, D.R., 1995. Zoogeomorphology: Animals as Geomorphic Agents. Cambridge Henrot, J., Brussard, L., 1997. Abundance, casting activity, and cast quality of
University Press, Cambridge. earthworms in an acid Ultisol under alley-cropping in the humid tropics.
Butler, D.R., Butler, W.D., 2009. The geomorphic effects of gophers on soil Applied Soil Ecology 6, 169–179.
characteristics and sediment compaction: a case study from alpine treeline, Imeson, A.C., 1976. Some effects of burrowing animals on slope processes in the
Sangre de Cristo Mountains, Colorado, USA. Open Geology Journal 3, Luxembourg Ardennes, part 1: excavation of animals burrows in experimental
82–89. plots. Geografiska Annaler 58A, 115–125.
Cane, J.H., 2003. Annual displacement of soil in nest tumuli of alkali bees (Nomia Imeson, A.C., Kwaad, F.J.P.M., 1976. Some effects of burrowing animals on slope
melanderi) (Hymenoptera: Apiformes: Halictidae) across an agricultural processes in the Luxembourg Ardennes, part 2, the eorsion of animal mounds
landscape. Journal of the Kansas Entomological Society 76, 172–176. by splash under forest. Geografiska Annaler 58A, 317–328.
Carlson, J.M., Crist, T.O., 1999. Plant responses to pocket-gopher disturbances James, S.W., 1991. Soil, nitrogen, phosphorous and organic matter processing by
across pastures and topography. Journal of Range Management 52, 637–645. earthworms in tallgrass prairie. Ecology 72, 2102–2109.
Cerd`a, A., Doerr, S.H., 2010. The effect of ant mounds on overland flow and soil Jońca, E., 1972. Winter denudation of molehills in mountainous areas. Acta
erodibility following a wildfire in eastern Spain. Ecohydrology 3, 392–401. Theriologica 17, 407–417.
Choler, P., 2005. Consistent shifts in alpine plant traits along a mesotopographical Jouquet, P., Podwojewski, P., Bottinelli, N., et al., 2008. Above-ground earthworm
gradient. Arctic, Antarctic, and Alpine Research 37, 444–453. casts affect water runoff and soil erosion in northern Vietnam. Catena 74,
Ciarniello, L.M., Boyce, M.S., Heard, D.C., Seip, D.R., 2005. Denning behavior and 13–21.
den site selection of grizzly bears along the Parsnip River, British Columbia, Kalisz, P.J., Davis, W.H., 1992. Effect of prairie voles on vegetation and soils in
Canada. Ursus 16, 47–58. central Kentucky. American Midland Naturalist 127, 392–399.
Crisp, D.T., Carling, P.A., 1989. Observations on siting, dimension and structure of Knight, J., 2009. Infilled pocket gopher tunnels: seasonal features of high alpine
salmonid redds. Journal of Fish Biology 34, 119–134. plateaux. Earth Surface Processes and Landforms 34, 590–595.
Darwin, C., 1881. The Formation of Vegetable Mould, through the Action of Kolb, H.H., 1985. The burrow structure of the European rabbit (Oryctolagus
Worms, with Observations on Their Habitats. John Murray, London. cuniculus L.). Journal of Zoology, London 206, 253–262.
Edelman, A.J., 2003. Marmota Olympus. Mammalian Species 736, 1–5. Kolb, H.H., 1991a. Use of burrows and movements by wild rabbits (Oryctolagus
Edwards, C.A., 1998. Earthworm Ecology. St. Lucie Press, Boca Raton, FL. cuniculus) on an area of sand dunes. Journal of Applied Ecology 28,
Edwards, C.A., Lofty, J.R., 1972. Biology of Earthworms. Chapman and Hall, 879–891.
London. Kolb, H.H., 1991b. Use of burrows and movements by wild rabbits (Oryctolagus
Ellison, L., 1946. The pocket gopher in relation to soil erosion on mountain range. cuniculus) in an area of hill grazing and forestry. Journal of Applied Ecology
Ecology 27, 101–104. 28, 892–905.
Elton, C.S., 1942. Voles. Mice and Lemmings: Problems in Population Dynamics. Komonen, M., Komonen, A., Otgonsuren, A., 2003. Daurian pikas (Ochotona
Clarendon Press, Oxford. daurica) and grassland condition in eastern Mongolia. Journal of Zoology,
English, E.I., Bowers, M.A., 1994. Vegetational gradients and proximity to London 259, 281–288.
woodchuck (Marmota monax) burrows in an old field. Journal of Mammalogy Kondolf, G.M., Cook, S.S., Maddux, H.R., Persons, W.R., 1989. Spawning gravels
75, 775–780. of rainbow trout in Glen and Grand Canyons, Arizona. Journal of the Arizona-
Fehmi, J.S., Bartolome, J.W., 2002. Species richness and California voles in an Nevada Academy of Science 23, 19–28.
annual and a perennial grassland. Western North American Naturalist 62, 73–81. Kondolf, G.M., Sale, M.J., Wolman, M.G., 1993. Modification of fluvial gravel size
Furness, R.W., 1991. The occurrence of burrow-nesting among birds and its by spawning salmonids. Water Resources Research 29, 2265–2274.
influence on soil fertility and stability. Symposium of the Zoological Society of Lai, C.H., Smith, A.T., 2003. Keystone status of plateau pikas (Ochotona curzoniae):
London 63, 53–67. effect of control on biodiversity of native birds. Biodiversity and Conservation
Gálvez, L., López-Pintor, A., Miguel, J.M.D., et al., 2008. Ecosystem engineering 12, 1901–1912.
effects of European rabbits in a Mediterranean habitat. Lagomorph Biology 2, Laundré, J.W., 1993. Effects of small mammal burrows on water infiltration in a
125–139. cool desert environment. Oecologia 94, 43–48.
Gervais, J.A., Griffith, S.M., Davis, J.H., Cassidy, J.R., Dragila, M.I., 2010. Effects of Laundré, J.W., 1998. Effect of ground squirrel burrows on plant productivity in a
grey-tailed vole activity on soil properties. Northwest Science 84, 159–169. cool desert environment. Journal of Range Management 51, 638–643.
Gibson, D.J., 1989. Effects of animal disturbance on tallgrass prairie vegetation. Laundré, J.W., Reynolds, T.D., 1993. Effects of soil structure on burrow
American Midland Naturalist 121, 144–154. characteristics of five small mammal species. Great Basin Naturalist 53,
Gopher Tortoise Council, 1998. http://www.gophertortoisecouncil.org/index.php 358–366.
(accessed March 2011). Lawler, H., 2000. A natural history of the Desert Tortoise, Gopherus [Xerobates]
Gottesfeld, A.S., Hassan, M.A., Tunnicliffe, J.F., Poirier, R.W., 2004. Sediment agassizii. http://www.biopark.org/destort1.html (accessed March 2011).
dispersion in salmon spawning streams: the influence of floods and salmon Le Bayon, R.-C., Binet, F., 1999. Rainfall effects on erosion of earthworm
redd construction. Journal of the American Water Resources Association 40, casts and phosphorus transfers by water runoff. Biology and Fertility of
1071–1086. Soils 30, 7–13.
Goudie, A.S., 1988. The geomorphical role of termites and earthworms in the Le Bayon, R.-C., Binet, F., 2001. Earthworm surface casts affect soil erosion by
tropics. In: Viles, H.A. (Ed.), Biogeomorphology. Basil Blackwell, New York, NY, runoff water and phosphorus transfer in a temperate maize crop. Pedobiologia
pp. 166–192. 45, 430–442.
 The Geomorphic Impacts of Animal Burrowing and Denning 279

Li, X.G., Zhang, M.L., Li, Z.T., Shi, X.M., Ma, Q., Long, R.J., 2009. Dynamics of Schrader, S., Joschko, M., 1991. A method for studying the morphology of
soil properties and organic carbon pool in topsoil of zokor-made mounds at an earthworm burrows and their function in respect to water movement.
alpine site of the Qinghai–Tibetan Plateau. Biology and Fertility of Soils 45, Pedobiologia 35, 185–190.
865–872. Semenov, Y., Ramousse, R., Le Berre, M., 2003. Alpine marmot (Marmota marmota)
Lindzey, F.G., 1976. Characteristics of the natal den of the badger. Northwest impact on floristic diversity of alpine meadows. In: Ramousse, R., Allaine, D., Le
Science 50, 178–180. Berre, M. (Eds.), Adaptive Strategies and Diversity in Marmots. International
Lips, K.R., 1991. Vertebrates associated with tortoise (Gopherus polyphemus) Network on Marmots, Lyon, France, pp. 269–274.
burrows in four habitats in south-central Florida. Journal of Herpetology 25, Semenov, Y., Ramousse, R., Le Berre, M., Tutukarov, Y., 2001. Impact of the black-
477–481. capped marmot (Marmota camtschatica bungei) on floristic diversity of Arctic
Long, C.A., Killingley, C.A., 1983. The Badgers of the World. Charles C. Thomas tundra in northern Siberia. Arctic, Antarctic, and Alpine Research 33, 204–210.
Publisher, Springfield, IL. Sherrod, S.K., Seastedt, T.R., 2001. Effects of the northern pocket gopher
McKechnie, S., 2006. Biopedoturbation by an island ecosystem engineer: burrowing (Thomomys talpoides) on alpine soil characteristics. Niwot Ridge, CO.
volumes and litter deposition by sooty shearwaters (Puffinus griseus). New Biogeochemistry 55, 195–218.
Zealand Journal of Zoology 33, 259–265. Smith, A.T., Foggin, J.M., 1999. The plateau pika (Ochotona curzoniae) is a
Meadows, P.S., Meadows, A. (Eds.), 1991. The Environmental Impact of Burrowing keystone species for biodiversity on the Tibetan plateau. Animal Conservation 2,
Animals and Animal Burrows. Clarendon Press, Oxford. 235–240.
Meysman, F.J.R., Middelburg, J.J., Heip, C.H.R., 2006. Bioturbation: a fresh look at Smith, D.J., Gardner, J.S., 1985. Geomorphic effects of ground squirrels in the
Darwin’s last idea. Trends in Ecology and Evolution 21, 688–695. Mount Rae areas, Canadian Rocky Mountains. Arctic and Alpine Research 17,
Milton, S.J., Dean, W.R.J., Klotz, S., 1997. Effects of small-scale animal 205–210.
distributions on plant assemblages of set-aside land in central Germany. Journal Svendsen, G.E., 1976. Structure and location of burrows of yellow-bellied marmot.
of Vegetation Science 8, 45–54. The Southwestern Naturalist 20, 487–494.
Moler, P.E., 1992. Rare and Endangered Biota of Florida, III, Amphibians and Tadzhiyev, U., Odinoshoyev, A., 1987. Influences of marmots on soil cover of the
Reptiles. University Press of Florida, Gainesville. eastern Pamirs. Soviet Soil Science 2, 22–30.
Moore, J.W., Schindler, D.E., Scheuerell, M.D., 2004. Disturbance of freshwater Teipner, C.L., Garton, E.O., Nelson, Jr. L.T., 1983. Pocket gophers in forest
habitats by anadromous salmon in Alaska. Oecologia 139, 298–308. ecosystems. U.S. Forest Service General Technical Report no. INT-154.
del Moral, R., 1984. The impact of the Olympic marmot on subalpine vegetation Terres, J.K., 1980. The Audubon Society Encyclopedia of North American Birds.
structure. American Journal of Botany 71, 1228–1236. Alfred A. Knopf, New York, NY.
Murie, J.O., 1992. Predation by badgers on Columbian ground squirrels. Journal of Verbeek, N.A.M., 1965. Predation by badger on yellow-bellied marmot in Wyoming.
Mammalogy 73, 385–394. Journal of Mammalogy 46, 506–507.
Newman, J., Scott, D., Moller, H., Fletcher, D., 2008. A population and harvest Voslamber, B., Veen, A.W.L., 1985. Digging by badgers and rabbits on some wooded
intensity estimate for sooty shearwater, Puffinus griseus, on Taukihepa (Big slopes in Belgium. Earth Surface Processes and Landforms 10, 782–799.
South Cape), New Zealand. Papers and Proceedings of the Royal Society of Vroom, G.W., Herrero, S., Ogilvie, R.T., 1980. The ecology of winter den sites of
Tasmania 142, 177–184. grizzly bears in Banff National Park. Alberta. International Conference on Bear
Parker, B.S., Meyers, K., Caskey, R.L., 1976. An attempt at rabbit control by warren Research and Management 4, 321–330.
ripping in semi-arid western New South Wales. Journal of Applied Ecology 13, Wang, Q.Y., Fan, N.C., 1987. Studies on the digging activities and exploration
353–367. about the method of number estimation of plateau zokor. Acta Theriologica
Plaster, B.R., 2003. The geomorphic impacts of marmots in Gothic, Colorado. MS Sinica 7, 283–290.
Thesis, Texas State University-San Marcos, San Marcos, 100 pp. Wang, T.-C., Xiong, Y.-C., Ge, J.-P., et al., 2008. Four-year dynamic of vegetation
Podruzny, S.R., Cherry, S., Schwartz, C.C., Landenburger, L.A., 2002. Grizzly bear on mounds created by zokors (Myospalax baileyi) in a subalpine meadow of the
denning and potential conflict areas in the Greater Yellowstone Ecosystem. Qinghai–Tibet Plateau. Journal of Arid Environments 72, 84–96.
Ursus 13, 19–28. Wesche, K., Nadrowski, K., Retzer, V., 2007. Habitat engineering under dry
Price, L.W., 1971. Geomorphic effect of the Arctic ground squirrel in an alpine conditions: the impact of pikas (Ochotona pallasi) on vegetation and site
environment. Geografiska Annaler 53A, 100–106. conditions in southern Mongolian steppes. Journal of Vegetation Science 18,
Questad, E.J., Foster, B.L., 2007. Vole disturbance and plant diversity in a 665–674.
grassland metacommunity. Oecologia 153, 341–351. Whitford, W.G., Kay, F.R., 1999. Biopedturbation by mammals in deserts: a review.
Reichman, O.J., Smith, S.C., 1985. Impact of pocket gopher burrows on overlying Journal of Arid Environments 41, 203–320.
vegetation. Journal of Mammalogy 66, 720–725. Xinghu, W.E.I., Sen, L.I., Ping, Y., Huaishun, C., 2007. Soil erosion and vegetation
Rogers, W.E., Hartnett, D.C., Elder, B., 2001. Effects of plains pocket gopher succession in alpine Kobresia steppe meadow caused by plateau pika – a case
(Geomys bursarius) disturbances on tallgrass-prairie plant community structure. study of Nagqu County, Tibet. Chinese Geographical Science 17, 75–81.
American Midland Naturalist 145, 344–357. Yoshihara, Y., Okuro, T., Buuveibaatar, B., Undarmaa, J., Takeuchi, K., 2010.
Roper, T.J., 1992. Badger Meles meles setts – architecture, internal environment Responses of vegetation to soil disturbance by Siberian marmots within a
and function. Mammal Review 22, 43–53. landscape and between landscape positions in Hustai National Park, Mongolia.
Roper, T.J., Ostler, J.R., Schmid, T.K., Christian, S.F., 2001. Sett use in European Japanese Society of Grassland Science 56, 42–50.
badgers Meles meles. Behaviour 138(2), 173–187. Zhang, Y., 2007a. Influence of plateau zokors (Eospalax fontanierii) on alpine
Rose, F.L., Judd, F.W., 1975. Activity and home range size of the Texas tortoise, meadows. In: Begall, S., Burda, H., Schleich, C.E. (Eds.), Subterranean Rodents:
Gopherus berlandieri, in south Texas. Herpetologica 31, 448–456. News from the Underground. Springer, Berlin, pp. 301–307.
Rutin, J., 1992. Geomorphic activity of rabbits on a coastal sand dune, Zhang, Y., 2007b. The biology and ecology of plateau zokors (Eospalax fontanierii).
De Blink dunes, the Netherlands. Earth Surface Processes and Landforms In: Begall, S., Burda, H., Schleich, C.E. (Eds.), Subterranean Rodents: News
17, 85–94. from the Underground. Springer, Berlin, pp. 237–247.
Schaffer, W.M., Tamarin, R.H., 1973. Changing reproductive rates and population Zhang, Y., Liu, J., 2003. Effects of plateau zokors (Myospalax fontanierii) on plant
cycles in lemmings and voles. Evolution 27(1), 111–124. community and soil in an alpine meadow. Journal of Mammalogy 84, 644–651.
280 The Geomorphic Impacts of Animal Burrowing and Denning

Biographical Sketch

Dr. David R. Butler is the Texas State University System Regents’ Professor of geography, and a University Dis-
tinguished Professor at Texas State University-San Marcos where he has been on faculty since 1997. His research
interests are in the areas of mountain geomorphology, zoogeomorphology, biogeomorphology, and den-
drogeomorphology, focusing his work in the Glacier National Park region of Montana, USA. He has published
over 150 refereed papers in journals and conference volumes, and more than 35 book chapters. He is the author
of Zoogeomorphology – Animals as Geomorphic Agents (Cambridge University Press, 1995 and 2007), and co-editor
of Tree Rings and Natural Hazards (Springer, 2010), The Changing Alpine Treeline – The Example of Glacier National
Park, Montana (Elsevier, 2009), and Mountain Geomorphology – Integrating Earth Systems (Elsevier, 2003). He has
received the G.K. Gilbert Award for Excellence in Geomorphological Research from the Geomorphology Specialty
Group of the Association of American Geographers (AAG), and the Distinguished Career Award and the Out-
standing Recent Accomplishment Award from the AAG’s Mountain Geography Specialty Group. David holds a BA
and MSc in geography from the University of Nebraska at Omaha, and the PhD in geography from the University
of Kansas.

Clayton J. Whitesides is a PhD student in the Department of Geography at Texas State University-San Marcos. His
research interests are in the fields of mountain geomorphology, biogeomorphology, and mountain biogeography.
He has published refereed papers in the international journal Progress in Physical Geography and in the Papers of the
Applied Geography Conferences. Clayton holds the bachelor of science degree from Utah State University, and the
master of science in geography from Brigham Young University.

John M. Marty Wamsley is a PhD student in the Department of Geography at Texas State University-San Marcos.
His interests are in zoogeomorphology, physical geography, and herpetofaunal ecology. He is particularly inter-
ested in the geographic distributions of snakes, and the ranges of Texas reptiles. Marty holds a bachelor of arts in
biology from St. Edwards University, and a master of applied geography degree from Texas State University-San
Marcos.

Stephen G. Tsikalas is a PhD student in the Department of Geography at Texas State University-San Marcos. His
interests are in zoogeomorphology, geomorphology, and physical geography. Stephen holds a bachelor of arts
from the University of Pittsburgh at Johnstown, and a master of arts in geography from the Indiana University of
Pennsylvania.
 12.19 Effects of Ants and Termites on Soil and Geomorphological Processes
WG Whitford, New Mexico State University, Las Cruces, NM, USA
DJ Eldridge, University of New South Wales, Sydney, NSW, Australia
r 2013 Elsevier Inc. All rights reserved.

12.19.1 Introduction 281

12.19.2 Geographic Distribution and Diversity 282
12.19.3 Effects of Ants and Termites on Soil Physical Properties 282
12.19.3.1 Development of Surface and Sub-Surface Structures 282
12.19.3.1.1 Elevated ant nests in seasonally flooded environments 283
12.19.3.1.2 Termitaria: Structures of mound-building termites 283
12.19.3.1.3 Stone lines 284
12.19.3.2 Soil Turnover and Soil Development 284
12.19.3.2.1 Soil turnover by ants and termites 284
12.19.3.2.2 Effects on soil particle size and clay mineralogy 285
12.19.3.2.3 Soil profile development or profile homogenization? 286
12.19.3.3 Soil Porosity, Infiltration and Water Storage 286
12.19.3.4 Soil Erosion 287
12.19.4 Effects of Ants and Termites on Soil Chemical Processes 288
12.19.4.1 Effects on Movement of Organic Matter 288
12.19.4.2 Effects on Soil Chemistry 288
12.19.5 Impacts of Alien Species: The Imported Fire Ant (Solenopsis invicta) as an Example 289
12.19.6 Conclusions 290
Acknowledgments 290
References 290

Glossary Microflora Microbes – bacteria, yeasts, and fungi.

Calcrete Hard soil layer composed of calcium carbonate. Smectite clay Family of clays that swell when immersed in
Chaff Dry scaly protective coverings of seeds or similar water.
fine, dry, scaly plant materials. Soil aggregate Group of soil particles adhering together in
Epigeic Aboveground. a cluster.
Ferricrete Hard soil layer composed of iron compounds. Suspended sediment Very fine soil particles that remain
Illite clay Nonexpanding clay composed of silicates. in suspension for a long time without contact with the
Microfauna Primarily protozoans such as amoebae and substrate.
flagellates.

Abstract

Ants and termites are one of the most widespread insect groups that occur on all continents except Antarctica. Soil-dwelling
ants and termites disturb surface and subsurface soils while constructing their nests and interconnecting tunnels. This soil
movement has major effects on soil turnover and development, clay mineralogy, the retention and infiltration of water, and
soil chemical properties. The effects of ant and termite activity appear to be greatest for structures that persist for many
years. Ants and termites can therefore be regarded as important moderators of soil and geomorphic processes.

12.19.1 Introduction nonreproductive workers. These workers construct the nests,

gather food, care for the developing young, and, most
Ants and termites are social colonial insects whose colonies importantly, care for the queen. Nests are the central focus of
are typically formed by fertilized queens producing the lives of ant and termite colonies. They house the central
food stores and, in some instances, the gardens that provide
the energy and nutrient requirements of the colony. Nurseries
Whitford, W.G., Eldridge, D.J., 2013. Effects of ants and termites on soil and
for the larvae and pupae are an essential component of most
geomorphological processes. In: Shroder, J. (Editor in Chief), Butler, D.R.,
Hupp, C.R. (Eds.), Treatise on Geomorphology. Academic Press, San Diego, nests. Soil-dwelling ant colonies vary in size from as few as
CA, vol. 12, Ecogeomorphology, pp. 281–292. 20 to more than 100 000 individual workers. Numbers of

Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00335-3 281

282 Effects of Ants and Termites on Soil and Geomorphological Processes

individuals in termite colonies range from 1400 to several respectively. Diversity decreases with increase in latitude to a
million in some of the mound-building species in the tropics single species inhabiting the Austrocedrus chilensis forests of the
(Holldobler and Wilson, 1990). sub-Antarctic region at 45–48o S (Torales et al., 2009). In the
Workers forage for food in the environment around the Northern Hemisphere, the number of termite species increases
nest, a process known as ‘central-place foraging’ (Bailey and exponentially from the Tropic of Cancer (about 20 species) to
Polis, 1987). This central-place foraging concentrates organic the Equator (460 species) (Goudie, 1988). Diversity even
materials in the nest or its immediate surroundings and results changes markedly over smaller climatic regions. For example,
in changes in the spatial distribution of organic materials. The in Israel, the highest ant-species richness occurs at the inter-
distance over which they forage is strongly species dependent section between two biogeographic regions where resource
and varies with the size of the species and their feeding habits. heterogeneity is likely high (Segev, 2010). In Africa and
For example, harvester ants have been known to forage for South America, numerous species occur in tropical rain forest,
seeds at distances of up to 20 m from the nest (Gordon, but in Australia, only four species are restricted to tropical rain
1992), while other species such as Pheidole spp. in Australia forest. In Australia, the termite fauna is more diverse and
typically forage over shorter distances (Hughes, 1991). numerous in dry sclerophyll forests, woodlands, and savannas
The effects of ants and termites on soil and geomorpho- (Robert, 2007).
logical processes result from their tendency to produce In general, ant communities nesting in soils are dominated
resource-rich patterns around the nests, a consequence of by small to very small species. For example, the ant fauna on a
their foraging patterns. They are also agents of biopedo- Chihuahuan Desert watershed are dominated by very small
turbation (soil disturbance or pedogenesis; Wilkinson et al., (Forelius spp. and Solenopsis spp.) and small (Dorymyrmex spp.
2009), and this disturbance takes many forms. The most direct and Pheidole spp.) ants. Ant colony abundance depends on
and widespread form of soil disturbance occurs where they many complex and interacting factors such as topographic
excavate subterranean nests and associated interconnecting position relative to the redistribution of water (e.g., runoff
tunnels and chambers. The large aboveground structures of James et al., 2008), soil texture, and vegetation structure
many ants and termites are dominant and have long-lived and composition. The abundance of soil-dwelling ants may
features of some landscapes. The materials used in building be as high as 7000 colonies ha 1 in relatively flat, sandy
these structures have long-term consequences for the devel- desert grassland but as few as three colonies ha 1 on period-
opment of soil profiles and the creation of soil heterogeneity. ically inundated, fine-textured soils (Whitford et al., 1995).
The importance of ants and termites as geomorphic The average density of nests of large seed harvesters that
agents depends on their abundance, diversity and species produce ephemeral mounds (Pogonomyrmex californicus and
composition, and the density of their nest structures. Most of P. desertorum) was 130 ha 1 and the average density of the
the studies of bioturbation by ants have focused on species relatively permanent nests of five large species (Trachymyrmex
that construct large nests that are occupied by the colonies for smithii, Aphaenogaster cockerelli, Myrmecocystus spp., Formica
more than a decade. Many of these studies have reported ef- perpilosa) was 40 ha 1 (Schumacher and Whitford, 1976). The
fects of these ants on soil chemistry. The bioturbatory effects of average density of large (5 m tall by 2 m diameter) Macrotermes
ants that build ephemeral nest structures have, however, goliath mounds in tropical rainforests in West Africa is about
largely been ignored. The nest cones and mounds of these ants 17 ha 1. Smaller termite mounds, however, may reach dens-
are short lived because they lack structural stability and are ities of up to 1000 ha 1 (Orhue et al., 2007).
easily disintegrated by wind or rain.
In this chapter, we describe the main effects of ants and
termites on geomorphology globally, with an emphasis on
12.19.3 Effects of Ants and Termites on Soil
their effects on soil physical and chemical processes. A sub-
Physical Properties
stantial amount of our knowledge is drawn from arid and
semi-arid landscapes where much of the work on the geo-
12.19.3.1 Development of Surface and Sub-Surface
morphic effects of these organisms has been carried out.
Structures
However, we supplement this with examples from ecosystems
worldwide to reinforce the notion that their overwhelming Most colonial ants and temperate-zone termites build sub-
effect is to lead to the formation of nutrient-rich soil profiles. terranean nests with either small (o1 m3) aboveground nest
structures or no structures at all. The construction and main-
tenance of these nests and tunnels of soil-dwelling ants and
12.19.2 Geographic Distribution and Diversity termites modify soil properties and affect geomorphic pro-
cesses. Subterranean ant nests generally have a number of
Ants are abundant in most of the world’s terrestrial environ- descending tunnels that connect to horizontal chambers
ments from the tropics to the subarctic and from arid to ranging in shape from oval to hemispherical. The horizontal
extremely mesic areas (Holldobler and Wilson, 1990). Ter- dimensions of subterranean ant nests commonly exceed more
mites live equatorward of 45–48o N and 45–48o S latitudes. than 2 m (Tschinkel, 2003; Figure 1). Similarly, chambers and
The highest species diversity of ants and termites is in the tunnels of honey-pot ants (Myrmecocystus spp.) have been
tropics. For example, in Argentina, the number of termite observed at a depth of 4 m, whereas galleries and tunnels
species ranges from 78 in the wet forests of the Chacoan of the subterranean termite Gnathamitermes tubiformans are
Province at the Tropic of Capricorn, to 41 to 7 species in known to reach depths of up to 9 m (W. Whitford, personal
the forests, shrublands, and grasslands of the mid-latitudes observation). Many tropical and subtropical termite species
 Effects of Ants and Termites on Soil and Geomorphological Processes 283

Figure 2 Large nest of an Irydomyrmex sp. ant on a seasonally

inundated floodplain in north-eastern Australia. Photo by Melissa
Dryden.

In arid areas, the densities of ant nests are known to be

greatest in higher elevation positions and lowest at the bottom
of the slopes. Lower landscape positions are prone to episodic
flooding from upland runoff resulting in fewer species and
lower abundances. For example, in the periodically inundated
Chaco region of subtropical South America, the dry, vegetated
islands are the result of the large (185 cm high, 120 cm across)
nest mounds of Camponotus punctulatus that occur at densities
Figure 1 Cast showing the nest architecture of the Florida harvester of between 200–1000 mounds ha 1 (Pire et al., 1991). Large
ant (Pogonomyrmex badius). Photo by Charles F. Badland. mima-type earth mounds (150 cm high, 20 m across) that are
prominent topographic features of grasslands of Buenos Aires
Province, Argentina, are the result of horizontal transport
build aboveground termitaria composed of soil and organic of soil to the nest sites by black fire ants, Solenopsis richteri
materials. Mounds of tropical termites range in size from a few (Cox et al., 1992).
centimeters in height to as much as 9 m tall and from 6.4 to
9.5 m in diameter (Lal, 1987).
12.19.3.1.2 Termitaria: Structures of mound-building
termites
12.19.3.1.1 Elevated ant nests in seasonally flooded Mound-building termites are the dominant arthropods of many
environments tropical and subtropical soils. In addition to their nest mounds,
The construction of large, tall nest mounds is an important many species build foraging galleries or construct sheeting,
adaptation of ants to seasonally flooded or inundated soils which is used to cover vegetation and protect the termites
(Figure 2). Nest mounds are produced from soils transported against predation while they are harvesting plant material. Most
from areas around the mound site. The production of large of the nest mounds and galleries are constructed with repacked
hummocks and mounds is a characteristic of several species of soil particles collected from the B horizon and mixed with
ants that inhabit wet meadows, fens, peat lands, and tropical salivary secretions and, commonly, feces. Mounds can be clas-
wet savannas. For example, the hummock topography of a sified according to whether they use orally transported material
Montana fen is attributed to the nest mounds of Formica mixed with secretions or excreta, or whether the material is
podzolica (Lesica and Kannowski, 1998). Subarctic peatland mainly feces or plant-derived excreta in the construction
hummock topography is largely formed by the nests of For- of their mounds (Orhue et al., 2007) (see Figure 3).
mica spp. and Myrmica spp. ants (Lesica and Kannowski, In Botswana, two topographic features (ridges and
1998). The hummock topography of some European grass- mounds) have been attributed to the activities of termites
lands and salt marshes are attributed to nest mounds of Lasius (Odontotermes spp.). These features occur in cultivated areas
flavus. These are relatively large ant species that are widely and range from 0.2 to 0.8 m above the ground surface (Miller
distributed at mid- to high latitudes. In fens, particularly sedge et al., 1994). In tropical semi-deciduous forest, termites build
meadows, ant mounds are significant topographic features large mounds (1.6 m high, 7–8 m across) around the trunks of
that provide an important refuge for plant species adapted to dead trees. The volume of these mounds can be up to 50 m3
drier conditions (Barton et al., 2009). They are also important (Awadzi et al., 2004). The mounds remain after the tree trunk
because they provide habitat for plant species that require has disappeared and the termites have left and remain as
well-drained soils. topographic features for many years.
284 Effects of Ants and Termites on Soil and Geomorphological Processes

The most widely studied topographic features produced by Stoops, 1989). Since termites and ants mix and transfer small
termites are the heuweltjies (pronounced hew-wel-tees). These soil particles to the surface in the construction of nest
mima-like mounds from South Africa are raised areas of finer- mounds, stones and prehistoric stone tools and artifacts are
textured alkaline soil with enhanced plant nutrients. In the commonly displaced downward and ultimately join these
succulent Karoo of South Africa, heuweltjies are thought to stone layers (Johnson et al., 2005). By this mechanism, the
have resulted from the foraging and nest-building activities of biomantle becomes sorted texturally into two layers, an upper
the termite Microhodotermes viator. At a landscape level, heu- layer of fine material over a basal stone layer of coarser ma-
weltjies resemble a pock-marked pattern that rests above the terial. Although ants and termites are probably important, the
level of inundation during the wet season (Palmer et al., 1999). origin of stone lines in some areas is probably the result of
They are highly persistent, even after a century of cultivation. persistent activities of fossorial vertebrates such as pocket
gophers (Johnson, 1989).

12.19.3.1.3 Stone lines

Layers of stone and coarse material in the soil are known as
12.19.3.2 Soil Turnover and Soil Development
‘stone lines’, three-dimensional (carpet-like) subsurface layers
of stones (Johnson, 1989). In Africa, stone line complexes In the process of nest construction and reconstruction, ants
are interpreted by some as being formed entirely or partially pick up soil aggregates, bite off fragments with their man-
by termites, based on the mineralogy similarities of for- dibles, and relocate the fragments to the walls or tunnels or
mations above and below the layer, and by the volume of fine their subterranean chambers or around the nest entrance. The
particles transported to the surface by termites (Soyer, 1987; size of these particles is a function of the size of the mandibles
(Sudd, 1969). Thus, the size of particles directly varies with the
body size of the ant species (Cammeraat and Risch, 2008).

12.19.3.2.1 Soil turnover by ants and termites

Most estimates of the quantities of soil deposited on the sur-
face by ants come from studies in arid and semi-arid eco-
systems. Estimates vary widely, with annual rates ranging from
o1 kg ha 1 (Whitford et al., 1995) in an arid system to more
than 10 000 kg ha 1 yr 1 in moist tropical ecosystems (Lock-
aby and Adams, 1985; Table 1). Average global rates of soil
turnover by ants have been estimated at 5000 kg ha 1 yr 1
(Folgarait, 1998). The large quantities of soil moved by ants
may be of equal or greater importance than earthworms in soil
turnover because of their wider geographical distribution,
particularly in arid and semi-arid areas. Some ant species
move nest entrances frequently (e.g., Aphaenogaster; Eldridge
Figure 3 Termitaria from semi-arid woodland in eastern Australia. and Pickard, 1994; Wilkinson et al., 2009) or build very large
Photo by D. Eldridge. mounds (e.g., Atta spp.; Bucher and Zuccardi, 1967). Ants

Table 1 Annual rates of subsoil accumulation on the soil surface around the nest entrances of ants

Community Location Soil movement References

(kg soil ha 1 yr 1)

Temperate forest Massachusetts, USA 60 Lyford (1963)

Rural environments Berkshire, England 8240 Waloff and Blackith (1962)
Rural environments New York, USA 948 Levan and Stone (1983)
Rural environments Wisconsin, USA 11 360 Salem and Hole (1968)
Rural environments Louisiana, USA 1600a Lockaby and Adams (1985)
Humid savanna Africa 300 Wood and Sands (1978)
Humid savanna Africa 4000 Aloni et al. (1983)
Semi-arid woodland Argentina 1100 Bucher and Zuccardi (1967)
Sown pasture Argentina 2100 Folgarait (1998)
Sclerophyll woodland Cataract, Australia 841 Humphreys (1981)
Semi-arid woodland Cobar, Australia 336 Eldridge and Pickard (1994)
Semi-arid shrubland Deniliquin, Australia 350–420 Briese (1982)
Arid grassland New Mexico, USA 0.1–255 Whitford et al. (1995)
Arid shrubland New Mexico, USA 6.3–35 Whitford unpublished data
Arid dunefield New Mexico, USA 61.1–187 Whitford unpublished data
a
One-off measurement only.
 Effects of Ants and Termites on Soil and Geomorphological Processes 285

with these behavioral characteristics may be responsible for Termite mounds, galleries, and sheeting are constructed
high rates of soil turnover. with clay, sand, and some silt collected from subsurface soil
In arid environments, the rate of soil deposition also layers, sometimes cemented using chemical secretions and
varies with soil texture and bulk density, which affect the en- up to 15% water. Sheeting and mounds therefore comprise
ergy costs of digging. For example, studies in the Chihuahuan predominantly finer-textured materials (e.g., G. tubiformans,
Desert in southern New Mexico have shown that the greatest MacKay et al. (1985); Macrotermes, Petts et al. (2009)).
nest densities and soil deposition rates are on sandy-loam Tall-thin Macrotermes sp. mounds have clay to sand ratios
to loamy-sand soils. Similarly, the lowest nest densities and between 1:1 and 3:1, whereas large-domed Macrotermes sp.
deposition rates by ants were in clay-loam and clay soils that mounds have clay to sand ratios between 2:1 and 18:1 (Orhue
are frequently inundated (Whitford et al., 1995). et al., 2007). Termite nests tend to have more clay (generally
Termites also move large quantities of soil from subsoil about 20% more than surrounding soils, e.g., Odontotermes
horizons where constructing sheeting over foraging runways in northern Kenya, Bagine (1984); Amitermes, Cornitermes,
or foraging galleries (Table 2). Rates of gallery sheeting for- Nasutitermes in Brazil; Sarcinelli et al., 2009), but sometimes
mation from a shrubland in New Mexico ranged from less silt (MacKay et al., 1985). For example, the epigeic nests of
1000 kg ha 1 yr 1 on coarse-textured upper slopes to 5650 kg mound-building termites, Cubitermes sp. mounds, comprised
ha 1 yr 1 on sandy-loam soils in the lower slopes (MacKay 67% clay and 27% sand compared with adjacent soils, which
and Whitford, 1988). This movement has important impli- contained only 31% clay but twice the amount of sand (63%;
cations for the development of soil profiles. Rates of soil Orhue et al., 2007).
deposition in the form of sheeting material range from Construction of aboveground termitaria of Drepanotermes,
2.5 mm per century in seasonally arid tropical grasslands in Amitermes, and Nasutitermes in the Tanami region of
West Africa to 10 mm per century in tropical woodlands in north–central Australia results in the transport of fine material
Northern Australia (Lee and Wood, 1971). into clay-depleted upper soil horizons. These fines are derived
from underlying ferricrete and calcrete duricrusts. Soil organic
matter and clay particles are primary ingredients of soil ag-
12.19.3.2.2 Effects on soil particle size and clay gregates. Thus, anything that affects clay and organic matter
mineralogy will have a major influence on the formation and character-
Material deposited around the entrances of ant nests tends istics of soil aggregates and the movement and retention of
to reflect the texture of the soil layers in which the nests are ions in upper soil layers. In general, the particle size distri-
constructed. In general, there are few general predictions bution of termite nest and sheeting material tends to differ
about the textural patterns of the soils deposited in ant from surrounding surface source material. Termites influence
mounds. Soils of Pogonomyrmex rugosus mounds in southern the distribution of particle sizes through the differential
Nevada sometimes had higher but sometimes lower clay movement of particular particle sizes, by altering soil aggre-
contents than the surrounding soils. Similarly, although gation levels and modifying clay mineralogy.
mound soils of two seed-harvesting ants (Messor spp.) were Opinions vary about the origin of the clay present in ter-
reported to have higher clay contents than surrounding soils, mite mounds and galleries. Some reports conclude that ter-
nests of a functionally similar species Iridomyrmex greensladei mites select specific particles from soil horizons below the
from Australia had less clay and more sand than the sur- mound whereas others conclude that soil particles are phys-
rounding soils (Nkem et al., 2000). Nest mounds of the ically fractured in the guts of the termites. Recent studies (e.g.,
European ant (Lasius niger) generally have a lower proportion Jouquet et al., 2002; 2005; 2007) have provided evidence that
of coarse particles (0.1–2 mm diameter), a lower bulk density termites have a direct effect on clay mineralogy. The fungus-
(Dostál et al., 2005), and a higher silt content (Cammeraat growing termite, Pseudacanthotermes spiniger, for example,
and Risch, 2008) than surrounding soils. Given that the builds subterranean nest mounds to support the fungus-comb
density of these long-lived ant mounds are up to 5200 structures. These nest structures, foraging galleries, and sheet-
nests ha 1 (Dostál et al., 2005), their effect on subsoil trans- ing are clay enriched. The termites modify these silicate clays
port is considerable. through a weathering process, increasing the proportion of

Table 2 Quantities of soil incorporated into the foraging galleries and sheeting over foraging runways by a range of subterranean termites
(Gnathamitermes, Odontotermes, Dermatotermes, Odontotermes, and Pseudacanthotermes)
1
Community Location Soil (kg ha yr 1) References

Desert shrubland Chihuahuan desert, New Mexico 272–498 Whitford et al. (1982) and Whitford (1999)
Desert grassland Chihuahuan desert, New Mexico 624–2958 MacKay and Whitford (1988)
Humid highlands Kenya 1059 Bagine (1984)
Semi-arid woodland Cobar, Australia 656 Whitford et al. (1992)
Savanna Southern Nigeria 300 Goudie (1988)
Savanna Senegal 675–900 Goudie (1988)
Tropical grassland India 5804 Gupta et al. (1981)
Rain forest Rio Negro, Venezuela 780 Salick et al. (1983)
Savanna Kenya 1300 Jouquet et al. (2007)
286 Effects of Ants and Termites on Soil and Geomorphological Processes

smectite layers and/or reducing the number of clay layers

in the crystallites (Jouquet et al., 2005; 2007). This increase
in smectites (and corresponding reduction in illite clays) in-
creases the clay’s chemically active surface area, resulting in a
high cation-exchange capacity. This change in clay ratios
affects soil fertility and hydrology (Jouquet et al., 2007).

12.19.3.2.3 Soil profile development or profile

homogenization?
The size and aggregate characteristics of the selected soil par-
ticles deposited around nest entrances can have significant
effects on soil profile development and stability. The effects of
ants on the degree of development of soil profiles depend on
temperature and rainfall characteristics of nest sites, the type
and abundance of ant colonies, and the longevity of their Figure 4 Soil removed by the ant Aphaenogaster barbigula covers
aboveground structures. Many soil-nesting ants modify the pine needles that have been moved by overland flow. The nest
soil profile by transporting organic matter to the subsoil or entrances are about 4 cm in diameter. Photo by D. Eldridge.
transporting fine particles from deeper layers to the surface
during the construction and maintenance of nest galleries, Australia (Figure 4) leads to a complete mixing of surface
tunnels, and chambers. Their activity, therefore, either can lead horizons to a depth of about 30 cm over a period of about 200
to the creation of new horizons from the B horizon material, years (Eldridge and Pickard, 1994). In more mesic areas,
or can result in the creation of relatively uniform surface erosion of the mounds of its conspecific Aphaenogaster long-
horizons in some places (e.g., Wilkinson et al., 2009). iceps, and sediment transport by rainwash and overland flow,
Ongoing transport of finer material to the surface by ter- leads to the development of a granular topsoil above a finer-
mites and ants may lead to the creation of distinct organic textured subsoil (Wilkinson et al., 2009).
layers. By removing particles o2 mm in diameter, some
tropical mound-building termites such as Cryptotermes sp.,
Coptotermes intermedius, and Microtermes subhyalinus contribute
12.19.3.3 Soil Porosity, Infiltration and Water Storage
to development of an ironstone gravel layer below the surface
below a gravel-free layer of termite-transported soil (Breuning- Ants and termites directly influence the hydrological prop-
Madsen et al., 2007). Whereas the mound and top 15 cm of erties of soils by producing macropores; continuous tubes
the soil comprised similar levels of sand, silt, and clay, soils or voids (spaces) in the soil body that are greater than about
below 15 cm were dominated by 70% gravel (Breuning- 0.75 mm in diameter (Bouma, 1992). In soils with numerous
Madsen et al., 2007). Similarly, termites in semi-deciduous macropores, water is transported to the deeper parts of the soil
tropical forest in Ghana construct soil heaps around dead tree profile more rapidly than predicted by infiltration models
trunks (Awadzi et al., 2004). Whereas surface soils in the dead based on water movement through the matrix pores (micro-
tree mounds contained mainly fine (o1.25 mm) material pores). Matrix pores occur between individual mineral grains
from the B horizon, material at depth (B170 cm) comprised and between soil particles, but are not biogenic. The network
predominantly gravel, markedly different from surrounding of tunnels and chambers constructed by ants and termites
soil at the same depth (Awadzi et al., 2004). Thus, the termites results in lower soil bulk density (or increased porosity)
produced a soil that was markedly different in structure compared with adjacent soils (Lobry de Bruyn and Conacher,
and nutrient content than the surrounding soils. Ants can alter 1990; Dostál et al., 2005).
the size range of particles to make them easier to transport to Most studies of the effects of ants and termites on infil-
the surface. For example, leaf cutter Attine ants create pellets tration have been conducted in arid or semi-arid regions, on
ranging from 0.5 to 4 mm in diameter, transport them to the single species, and at single locations. The magnitude of the
surface, and deposit them in a nest-mound pile. These pellets effect of ants and termites on infiltration depends on topo-
contain plant fragments, silt, and clay particles cemented with graphic position, soil texture and structure, the type of ant and
clay, producing an organic-rich surface layer (Cosarinsky and termite species, and nest morphology. For example, a land-
Roces, 2007). The rearrangement, sorting, and mixing of soils scape-scale study examined differences in infiltration between
by termites result in a locally thickened biomantle (Obi and nests of Aphaenogaster cockerelli (a generalist omnivorous ant)
Ogunkunle, 2009). The development of latosols (soils high in and P. rugosus (a seed-harvesting ant) in a Chihuahuan Desert
iron, silicon oxide, and aluminum oxide) in Brazilian tropical watershed (James et al., 2008). Ponded steady-state infiltration
rainforests has been attributed to the long-term action of ter- rates were 8 times greater on ant-modified soils than on non-
mites and some secondary movement by ants (Reatto et al., nest soils, but the relative effect of ants differed with position
2009). in the landscape (James et al., 2008). This study reinforced the
The effect of mixing of surface and subsurface materials by notion that the effect of ants on infiltration is site and land-
ants can lead to homogenization of the soil profile, impeding scape specific. Topography and species type are also important
the formation of soil horizons (Folgarait, 1998; Eldridge and determinants of ant effects on hydrology. The network of
Pickard, 1994). For example, construction of nest entrances by tunnels and chambers that make up the subterranean nests of
the funnel ant (Aphaenogaster barbigula) in semi-arid eastern ants provides a method of bulk flow percolation of water to
 Effects of Ants and Termites on Soil and Geomorphological Processes 287

deep soil layers. Water moving to deep soil layers is isolated hydrologically active and 30% had been sealed by some
from the water stored in the upper layers, which is commonly means (Leonard et al., 2004).
lost by evaporation. Soil moisture in soils of harvester-ant
(Pogonomyrmex occidentalis) nests was consistently greater at
depths of 60–100 cm compared with the surface (o10 cm)
12.19.3.4 Soil Erosion
soils over a period of 12 months (Laundre, 2008).
Subterranean termites effect infiltration by producing Material from ant nests and termite sheeting consists of fine
tunnels at the soil surface that allow foragers to carry plant sand, silt, clay, and small soil aggregates. It is therefore po-
fragments or organic matter into the nest. Most studies on tentially susceptible to removal by wind and water. The rela-
termites have reported generally increased water infiltration. tive importance of ant nests to suspended sediments in runoff
Mound-building termites in a Columbian forest have also water likely varies with nest density, volume of nest mounds,
been shown to increase porosity by up to 100% and ant-nest and the texture and composition of the mounds (Aalders
mounds by up to 300%, compared with the area surrounding et al., 1989). Sediment concentrations are generally higher on
their mounds (Decaëns et al., 2001). However, the density of plots with ant mounds than those without mounds, mainly
functional termite-produced macropores is known to vary because ant-removed soil has low levels of aggregation (Cerda
both spatially and temporally. Studies of the effects of termites and Jurgensen, 2008). For example, erosion rates on plots with
on infiltration have been conducted on plots where termites ants in an orchard in eastern Spain were 41 kg ha 1 yr 1
were eliminated with termiticides (Elkins et al., 1986; Mando compared with 13 kg ha 1 yr 1 on plots without ant mounds
et al., 1996; Mando and Miedema, 1997), in areas adjacent to (Cerda et al., 2009). Similarly, in eastern Australia, Richards
functional termite pavements (Eldridge, 1994), or sites col- (2009) demonstrated relatively high rates of erosion of ant-
onized by termites (Tongway et al., 1989). Elkins et al. (1986) mounded soil from A. barbigula ant-nest entrances.
demonstrated substantially higher infiltration rates in the Sheeting and gallery material of subterranean termites are
Chihuahuan Desert in the presence of termites (88.4 mm h 1) relatively fragile and susceptible to raindrop impact and
than sites where termites had been removed with termaticides therefore splash erosion. The proportion of gallery sheeting
(51.3 mm h 1). Similar results were reported for termite and dispersed as sediment by splash erosion increases with rainfall
termite-free sites in the wet–dry tropics of Burkina Faso, West erosivity and the quantity of sheeting. Gallery material and
Africa (Mando et al., 1996). In the semi-arid mulga (Acacia surface sheeting connecting the aboreal nests of Nasutitermes
aneura) woodland in eastern Australia, mounds formed by spp. are relatively fragile and easily eroded during high rain-
termites around decaying down logs had infiltration rates al- fall. This material contributes significantly to soil erosion
most 10 times higher than the surrounding termite-free soils and sediment transport in African savannas, with rates of
(Tongway et al., 1989). Finally, Eldridge (1994) showed that 300–1059 kg ha 1 yr 1 reported for southern Nigeria, Senegal,
surface cementation on the pavements of subterranean ter- and Kenya (Goudie, 1988). Termite sheeting may be substantial
mites (Drepanotermes sp.) resulted in depressed infiltration in arid areas. For example, sheeting of subterranean termites in
rates (6.7 mm h 1), which effectively results in water being eastern Australia has been shown to cover up to 6.6% of the
shed from the pavements to the annular zone surrounding the timbered groves and 2.3% of unvegetated inter-groves in semi-
pavement (113.2 mm h 1; Figure 5). A field experiment and arid mulga (Acacia aneura) woodlands (Whitford et al., 1992).
modeling study provided details on variation in abundance Erosion rates of 0.025–0.05 mm yr 1 have been reported
of termite-produced macropores and the bulk flow dynamics from aboveground termite mounds in semi-arid woodland in
of those macropores (Leonard et al., 2004). In the Sahelian northern Australia, where there were 283 mounds ha 1 (Holt
region of Niger, the area covered by termite sheeting was lin- et al., 1980). Erosion rates of abandoned termitaria (400 kg
early correlated to the number of foraging holes, which aver- ha 1 yr 1) were reported to be half that of occupied mounds
aged 123 holes m 2 of termite sheeting. After a 17 mm (800 kg ha 1 yr 1) in tropical Upper Volta, Africa (Lal, 1987).
rainfall, 70% of the foraging hole macropores had been The main erosion process was identified as soil creep, with
rates of 1163 kg ha 1 yr 1 from the mound to the corona
around the mound (Lal, 1987). Erosional fans commonly
surround Odontotermes spp. mounds for distances of 10–30 m
from the base of the mounds (Goudie, 1988). Eroded mounds
are continually repaired by termites as long as the colony is
active. When colonies expire, however, the abandoned nest
mounds erode rapidly, and the original B horizon material
making up the mounds is deposited on the soil surface around
the base of the mounds.
Rates of mound erosion are known to be accelera-
ted by foraging by domestic and native animals. Aardvarks
(Orycteropus afer) dig into termitaria, exposing the interior
structures that are less resistant to splash erosion than the clay-
rich surface. Elephants destroy termitaria in search of mineral-
Figure 5 The aboveground nest capping of a Drepanotermes sp. in rich soil, accelerating erosion of the mounds (Goudie, 1988).
semi-arid eastern Australia. The cap is about 1 m in diameter. Photo In Central and southern America, foraging for termites
by James Val. by giant anteaters (Mymecophagus tridactyla) and armadillos
288 Effects of Ants and Termites on Soil and Geomorphological Processes

(Dasypus spp.) destroys the hydrophobic surface crust and termites use less organic carbon and nitrogen to construct
accelerates rates of splash erosion (Anacleto, 2007). their sheeting and rely more on clay particles. Organic matter
levels in the mounds 3–5 times greater than that in the sur-
rounding soils are not uncommon (Decaëns et al., 2001).
12.19.4 Effects of Ants and Termites on Soil In arid, semi-arid, and some tropical ecosystems, more
Chemical Processes than half of the potential inputs to the soil organic pool are
consumed by termites (Whitford et al., 1992). Organic matter
The chemical composition of mound, gallery, and sheeting is metabolized more rapidly and effectively by termite gut
soils of soil-nesting ants and termites is largely determined by microflora and microfauna than by free-living microflora
the chemical composition of the soil horizons from which the and micro- and mesofauna in the soil. Termite feces contain
material has been derived. Many of the studies on the effects very little humic material. They, therefore, return little of the
of ants on soil chemistry have been conducted on seed-har- organic matter that they consume back to the soil. The absence
vesting ant species in arid, semi-arid environments, and tem- of any natural organic matter pattern in organic matter in arid
perate woodlands and grasslands. These studies have focused shrublands and the significant negative correlation between
on large body size, long-lived species that build nest structures termite abundance and soil organic matter indicate that sub-
which persist for several decades (MacMahon et al., 2000). terranean termites influence variation in soil organic matter
Seed-harvesting ants are central-place foragers and deposit across the watershed (Nash and Whitford, 1995).
organic debris and unused seed parts in refuse or chaff Only half of the studies reporting changes in organic car-
piles around the perimeter of the nest or mound. When bon concentrations in mound-building termites have reported
decomposed, this material is a source of essential elements higher concentrations adjacent to the mounds (Tables 3
for plant growth (Folgarait, 1998; Whitford et al., 2008). In and 4). Thus, species such as Macrotermes bellicosus that do
tropical and subtropical areas, fungus-culturing Attine ants not use feces in the construction of their mounds produce
collect leaves or leaf fragments to use as a growth medium for mounds with lower organic carbon content than the sur-
the fungi that provide food for the colony. These Attine ants rounding soil (Abe et al., 2009). This suggests that the origin
build large nest mounds that are maintained by the regular of higher concentrations of organic carbon in termite-mound
addition of plant and soil materials (Leal and Oliveira, 2000). soils is via the use of their feces as a cementing agent.

12.19.4.1 Effects on Movement of Organic Matter 12.19.4.2 Effects on Soil Chemistry

Ants and termites have their greatest effect on soil chemistry Soils modified by seed-harvester ants generally have higher
by increasing the organic matter and organic carbon contents concentrations of total nitrogen, nitrate and ammonium, total
of the soils surrounding their mounds (Table 3). Organic carbon, phosphorus, calcium, and magnesium (e.g., Salem
matter increase is a result of the use of clay particles by ter- and Hole, 1968; Czerwinski et al., 1969; Mandel and Sor-
mites in addition to the salivary secretions used by the termites enson, 1982; Petal, 1978, Table 3). Soil pH is, however, gen-
to cement the material together to form their structures erally unaffected by ants (Wiken et al., 1976), though some
(Jouquet et al., 2007). In soils with a higher content of clay, activity may result in reduced pH in calcareous soils or

Table 3 The numbers of studies reporting effects of different groupsa of ants on soil chemical properties under or at the edges of the nest
compared with adjacent, unmodified soils

Chemical attribute Seed harvesters Fungus culturing Temperate scavengers

I NC D I NC D I NC D

Soil organic matter 9 2 0 2 0 0 8 1 0

Total N 9 0 0 1 0 1 5 2 6
Nitrate and ammonium 4 0 0 – – – 2 0 0
Total C 5 0 0 0 0 4 2 3 0
pH 1 1 4 0 1 1 8 8 1
Total phosphorus 6 1 0 2 2 0 4 0 2
Exchangeable sodium – – – 2 0 1 4 0 2
Exchangeable potassium 2 0 0 5 3 0 6 1 1
Exchangeable calcium 3 0 0 2 2 2 3 3 2
Exchangeable magnesium 3 1 0 1 0 1 4 1 1
Total 42 5 4 15 8 10 46 19 15

I, increased concentration; NC, no change; D, decrease concentration.

a
Seed harvester ants: Chelaner spp., Messor bouvieri, M. Andrei, Pogonomyrmex barbatus, P. badius, P. occidentalis, P. rugosus. Fungus culturing ants: Atta colombica, A.
laevigata, Trachymyrmex spp. Temperate zone scavangers: Camponotus spp., C. punctulatus, Formica cunicularia, F. montana, F. rufa, F. rufibarbus, Lasius flavius, L. niger,
Solenopsis invicta (an alien fire ant that has invaded the southern United States).
 Effects of Ants and Termites on Soil and Geomorphological Processes 289

Table 4 Comparison of concentrations of selected soil chemicals in termite-mound soils compared with the concentrations in soils of adjacent
pedons

Community Location Genus OC N Ca Mg K P pH

Savanna Nigeria Macrotermes & odontotermes I I I I I

Savanna Brazil Several species D I I I I D
Woodlands Australia Cubitermes & trinervitermes I I I
Woodlands Burkina Faso Macrotermes D D
Woodlands Australia Cubitermes & macrotermes I I I I I I I
Desert Israel Anacanthotermes ubachi I
Savanna Africa Macrotermitidae I I
Savanna Guyana Several species I I
Savanna South Africa Microhodotermes I I I
Savanna India Odontotermes & microcerotermes I I D I
Savanna Nigeria Macrotermes D I I I I D

OC, organic carbon; N, total nitrogen; Ca, calcium; Mg, magnesium; K, potassium; P, available phosphorus; I, increased concentration; D, decreased concentration.
Source: Data from Orhue et al. (2007), Abe et al. (2009), Zaady et al. (2003).

increases it in acid soils (Folgarait, 1998). Because ants have a 12.19.5 Impacts of Alien Species: The Imported Fire
marked effect on organic matter around their nests (Table 3), Ant (Solenopsis invicta) as an Example
they also influence soluble and exchangeable cations that are
adsorbed to the surface of the clay-rich particles (Sarcinelli Invasive alien species of social insects have been documented
et al., 2009). The stability of these cations may be influenced to reduce the diversity of native insects. Some invasive alien
by biospheric interactions promoting their release by wea- species produce impacts that cascade through ecosystems,
thering (Paton et al., 1995; Folgarait, 1998). Soil nutrients also altering basic ecosystem properties, changing hydrology, nu-
vary with position in the landscape. For example, Sarcinelli trient cycles, soil chemistry, and many other biotic and abiotic
et al. (2009) showed that the relative differences in calcium, parameters (Wagner and Van Driesche, 2010). The red imported
magnesium, aluminum, and phosphorus between termite- fire ant, S. invicta, is a species that is known to produce dra-
mound and intermound soils increased with distance down matic cascading effects on ecosystems in which it has invaded.
the slope. S. invicta is a native of South America that was inadvert-
By far, the greatest effect on soil chemistry results from ently released in the southern United States in the late 1930s.
the activity of harvester ants (Table 3), though the activities S. invicta has now spread to 15 states. As S. invicta spread
of temperate scavenging ants such as Campanotus, Solenopsis, across the southern USA, it virtually eliminated the native fire
Formica, and Lasius generally produce increases in soil cations ant, Solenopsis xyloni, and markedly reduced the populations
and soil organic matter concentrations compared with ad- of S. geminata and the seed harvester, Pogonomyrmex badius.
jacent soils. The effects of fungus-culturing ants such as Atta In areas with high densities of S. invicta colonies, most of the
and Trachymyrmex, however, are less clear. Temperate-zone native resident ant species are eliminated (Holldobler and
ants from mesic environments provide no consistent patterns Wilson, 1990). S. invicta colonies occur at average densities
of effects on soil chemistry. There are more reports of of 200 ha 1 but have been reported to reach densities of 400
ants increasing concentrations of soil chemicals than those colonies ha 1 in east Texas (Porter et al., 1991).
reporting reductions in concentrations. Some ants may Most, but not all, of the ant species reduced or eliminated
affect soil chemistry by accessing layers of cemented calcrete by S. invicta have small, ephemeral, epigeic nests. The main
(CaCO3) to access deeper soil layers (Whitford, 2003; Liu effect of those native resident ant species is soil profile mixing.
et al., 2007) and bringing this to the surface. For example, the By contrast, the invasive fire ants construct large mounds that
large nest mounds of Atta vollenweidere are formed from cal- alter the biogeochemical and physical properties of the soils in
cium carbonate-rich soils derived from up to 2 m below the the areas where they establish colonies (DeFauw et al., 2008).
surface (Bucher and Zuccardi, 1967). In the Chihuahuan Mound construction by imported fire ant workers results in a
Desert, fragments of dislodged calcrete are used by seed- dense network of narrow tunnels made of soil pellets trans-
harvester ants (Pogonomyrmex spp.) and fungus-culturing ported from deeper soil horizons. The aggregates on the sur-
ants (Trachymyrmex spp.) to construct solid caps on the nest face of the mounds are less dense and less stable than the
mounds. These caps increase the resistance of nest mounds adjacent soil aggregates. The fragile crust allows greater infil-
to wind and water erosion, but the calcium may be distributed tration than surrounding soils. Infiltrating water rapidly drains
to other areas through erosion of the nest caps, potentially to the deeper tunnels where free water was present between
affecting soil chemistry. 48 cm–120 cm while the adjacent pedons were dry at those
Like the ants, the mound soils of various termite species depths (Green et al., 1999). Although imported fire ant nests
generally have higher concentrations of organic carbon, total alter chemical and physical properties of soils, the specific
nitrogen and nitrates, available phosphorus and exchangeable cations and anions concentrated in the nest soils vary with soil
calcium, magnesium, and potassium than adjacent soils type. For example, nest soils in some locations are phosphorus
(Table 4). The effect, however, is generally dependent on soil enriched, but in others are not. Calcium and magnesium are
textural characteristics. concentrated in S. invicta mounds in some soils but not in
290 Effects of Ants and Termites on Soil and Geomorphological Processes

other locations. In some forest soils, S. invicta mounds were aspects of the paper, and Walter Tsinkel for allowing us to use
higher in organic matter, phosphorus, potassium, and calcium the figure of the Florida harvester ant (Pogonomyrmex badius).
compared to surrounding soils (DeFauw et al., 2008).
Since S. invicta reduces the abundance and diversity of
native ant species and constructs high densities of large nest
mounds, they are a major driver of change in geomorphic References
processes in areas in which they are dominant. Other ant
Aalders, I.H., Augustinus, P.G.E.F., Nobbe, J.M., 1989. The contribution of ants to
species that are alien invaders from distant geographic areas
soil erosion: a reconnaissance survey. Catena 16, 449–459.
have the potential of producing changes in geomorphic pro- Abe, S.S., Yamamoto, S., Wakatsuki, T., 2009. Physiochemical and morphological
cesses similar to those that have been documented for the red properties of termite (Macrotermes bellicosus) mounds and surrounding pedons
imported fire ant. on a toposequence of an inland valley in the southern Guinea savanna zone of
Nigeria. Soil Science and Plant Nutrition 55, 514–522.
Aloni, K., Malaisse, F., Kapinga, I., 1983. Roles des termites dans la decomposition
dubois et le transfert de terre dans une fort claire zambezienne. In: Lebrun, P.,
12.19.6 Conclusions Andre, H., de Medts, A., Gregoire-Wibo, C., Wanthy, G. (Eds.), New Trends in
Soil Biology. Dieu Brichart, Louvain-la-Neuve, pp. 600–602.
Most of the ecosystems of the planet are subjected to some Anacleto, T.C.D.S., 2007. Food habits of four armadillo species in the Cerrado area,
Mato grosso, Brazil. Zoological Studies 46, 529–537.
degree of management by humans, which can have cascading
Arshad, M.A., 1981. Physical and chemical properties of termite mounds of two
effects on biota and ecosystem processes. There remains, how- species of Macrotermes (Isoptera, Termitidae) and the surrounding soils of the
ever, a paucity of information on the effects of land manage- semiarid savanna of Kenya. Soil Science 132, 161–174.
ment practices on the diversity of ants and termites. Increasing Awadzi, T.W., Cobblah, M.A., Breuning-Madsen, H., 2004. The role of termites in
use of intensive agriculture and the use of chemical pesticides soil formation in the tropical semi-deciduous forest zone, Ghana. Geografisk
Tidsskrift 104, 27–34.
are likely to substantially affect populations of ants and termites Bagine, R.K.N., 1984. Soil translocation by termites of the genus Odontotermes
thereby affecting soil properties, nutrient cycling, and hydrol- (Holmgren)(Isoptera: Macrotermitinaae) in an arid area of northern Kenya.
ogy. Clearing of forest and brush for cultivation results in loss of Oecologia 64, 263–266.
wood-feeding termites, though fungus-feeding and humus- Bailey, K.H., Polis, G.A., 1987. Optimal and central-place foraging theory applied
to a desert harvester ant, Pogonomyrex-Californicus. Oecologia 72, 440–448.
feeding species are able to survive by foraging in the adjacent
Barton, B.J., Kirschbaum, C.D., Bach, C.E., 2009. The impacts of ant mounds on
agricultural fields (Orhue et al., 2007). Management practices sedge meadow and shrub. Natural Areas Journal 29, 293–300.
associated with pastoralism appear to have only short-term Bouma, J., 1992. Influence of soil macroporosity on environmental quality.
effects on ants and termites, while fire may have an effect, de- In: Sparks, D.L. (Ed.), Advances in Agronomy. Academic Press, New York, vol.
pending on the community. Fire in some desert grasslands, 46, pp. 1–37.
Brener, A.G.F., Silva, J.F., 1995. Leaf-cutting ants and forest groves in a tropical
which is used to increase forage production for livestock or parkland savanna of Venezuela: facilitated succession. Journal of Tropical
reduce undesirable shrubs, is known to substantially reduce soil Ecology 11, 651–669.
deposition by ants and the volume of termite-feeding galleries Breuning-Madsen, H., Awadzi, T.W., Kock, C.B., Borggaard, O.K., 2007.
and sheeting (Killgore et al., 2009). Fire in lichen-spruce Characteristics and genesis of pisolitic soil layers in a tropical moist semi-
deciduous forest of Ghana. Geoderma 141, 130–138.
woodland in eastern Canada, however, had no appreciable
Briese, D.T., 1982. The effect of ants on the soil of a semi-arid saltbush habitat.
effect on ant colonies (LaFleur et al., 2002). More importantly, Insectes Sociaux 29, 375–382.
projected changes in climate are likely to have a range of effects Bucher, E.H., Zuccardi, R.B., 1967. Signification de los hormigueros de Atta
on ants and termites from changes in their distribution, to Vollenweideri Forel como alteradores del suelo en la Provinces de Tucuman.
altered efficiency of foraging. Increasing temperatures resulting Acta Zoologica Lilloana 23, 83–96.
Cammeraat, E.L.H., Risch, A.C., 2008. The impact of ants on mineral soil properties
from changing climates may favor the invasion of exotic ant and processes at different spatial scales. Journal of Applied Entomology 132,
species such as the fire ant (S. invicta) at the expense of local 285–294.
ants, with unknown effects on a range of ecosystem services as Carlson, S.R., Whitford, W.G., 1991. Ant mound influence on vegetation and soils
diverse as soil hydrology and seed dispersal. Despite their small in a semiarid mountain ecosystem. American Midland Naturalist 126, 125–139.
Cerda, A., Jurgensen, M.F., 2008. The influence of ants on soil and water losses
size, a large number of studies have shown that both ants and
from an orange orchard in eastern Spain. Journal of Applied Entomology 132,
termites have substantial beneficial effects on ecosystem prop- 306–314.
erties and processes that would be almost impossible to value. Cerda, A., Jurgensen, M.F., Bodi, M.B., 2009. Effects of ants on water and soil
Although much has been written about ants and termites, we losses from organically-managed citrus orchards in eastern Spain. Biologia 64,
still know relatively little about their precise roles in soil and 527–531.
Cosarinsky, M.I., Roces, F., 2007. Neighbor leaf-cutting ants and mound-building
ecological processes, their linkages with other biota, and their termites: comparative nest micromorphology. Geoderma 141, 224–234.
importance in maintaining healthy and resilient ecosystems Cox, G.W., Mills, J.N., Ellis, B.A., 1992. Fire ants (Hymenoptera: Formicidae)
(Folgarait, 1998). There is, therefore, an urgent need to quantify as major agents of landscape development. Environmental Entomology 21,
some of these unknown effects and to include the study of ants 281–286.
Czerwinski, Z., Jakubczyk, H., Petal, J., 1969. The influence of ants of the genus
and termites in future studies to develop an integrated under-
Myrmica on the physico-chemical and microbiological properties of soil within
standing of their importance in ecosystem functioning. the compass of ant hills in Strzeleckie Meadows. Polish Journal of Soil Science
3, 51–58.
Decaëns, T., Galvis, J.H., Amézquita, E., 2001. Propriétés des structures produites
Acknowledgments par les ingénieures écologiques á la surface du sol d’une savane colombienne.
Comptes Rendus Academie Sciences Paris, Sciences de la vie 324, 465–478.
DeFauw, S.I., Vogt, J.T., Boykin, D.I., 2008. Influence of mound construction by red
The authors thank Hanna Lee, Dan Dugas, Erin Roger, David and hybrid imported fire ants on soil chemical properties and turfgrass in a sod
Butler, Alister Spain and Jack Shroder for comments on various production agroecosystem. Insectes Sociaux 55, 301–312.
 Effects of Ants and Termites on Soil and Geomorphological Processes 291

Dostál, P., Březnová, M., Kozličková, V., Herben, T., Kovár, P., 2005. Ant induced Lane, D.R., Bassirirad, H., 2005. Diminishing effects of ant mounds on soil
soil modification and its effect on plant below-ground biomass. Pedobiologia heterogeneity across a chronosequence of prairie restoration sites. Pedobiologia
49, 127–137. 49, 359–366.
Eldridge, D.J., 1994. Nests of ants and termites influence infiltration in a semiarid Laundre, J.W., 2008. Soil moisture patterns below mounds of harvester ants.
woodland. Pedobiologia 38, 481–492. Journal of Range Management 43, 10–12.
Eldridge, D.J., Pickard, J., 1994. Effects of ants on sandy soils in semiarid eastern Leal, I.R., Oliveira, P.S., 2000. Foraging ecology of attine ants in a Neotropical
Australia. 2. Relocation of nest entrances and consequences for biodurbation. savanna: seasonal use of fungal substrate in the cerrado vegetation of Brazil.
Australian Journal of Soil Research 32, 555–570. Insectes Sociaux 47, 376–382.
Elkins, N.Z., Sabol, G.V., Ward, T.J., Whitford, W.G., 1986. The influence of Lee, K.E., Wood, T.G., 1971. Physical and chemical effects on soils of some
subterranean termites on the hydrological characteristics of a Chihuahuan desert Australian termites and their pedological significance. Pedobiologia 11, 376–409.
ecosystem. Oecologia 68, 521–528. Leonard, J., Perrier, E., Rajot, J.L., 2004. Biological macropores effect on runoff
Folgarait, P.J., 1998. Ant biodiversity and its relationship to ecosystem functioning: and infiltration: a combined experimental and modeling approach. Agriculture,
a review. Biodiversity and Conservation 7, 1221–1244. Ecosystems and Environment 104, 277–285.
Frouz, J., Holec, M., Kalcik, J., 2003. The effect of Lasius niger (Hymenoptera, Lesica, P., Kannowski, P.B., 1998. Ants create hummocks and alter structure and
Formicidae) ant nests on selected soil chemical properties. Pedobiologia 47, vegetation of a Montana Fen. American Midland Naturalist 139, 58–68.
205–212. Levan, M.A., Stone, E.L., 1983. Soil modification by colonies of Black Meadow ants in
Gordon, D.M., 1992. How colony growth affects forager intrusion between neighboring a New York old field. Soil science society of America Proceedings 47, 1192–1195.
harvester ant colonies. Behavioral Ecology and Sociobiology 31, 417–427. Liu, X., Monger, H.C., Whitford, W.G., 2007. Calcium carbonate in termite galleries:
Goudie, A.S., 1988. The geomorphological role of termites and earthworms in the biomineralization or upward transport? Biogeochemistry 82, 241–250.
tropics. In: Viles, H.A. (Ed.), Biogeomorphology. Blackwell, Oxford, pp. 166–192. Lobry de Bruyn, L.A., Conacher, A.J., 1990. The role of ants and termites in soil
Green, W.P., Pettry, D.E., Switzer, R.E., 1998. Impact of imported fire ants on the modification: a review. Australian Journal of Soil Research 28, 55–95.
texture and fertility of Mississippi soils. Communications in Soil Science and Lockaby, B.G., Adams, J.C., 1985. Pedoturbation of a forest soil by fire ants. Soil
Plant Analysis 29, 447–457. Science Society of America Proceedings 46, 785–788.
Green, W.P., Pettry, D.E., Switzer, R.E., 1999. Structure and hydrology of mounds of Lyford, W.H., 1963. Importance of ants to brown podzolic soil genesis in New
the imported fire ants in the southeastern United States. Geoderma 93, 1–17. England. Harvard Forest Paper, Petersham MA, No. 7, 18 pp.
Gupta, S.R., Rajvanshi, R., Singh, J.S., 1981. The role of the termite Odontotermes MacKay, W.P., Blizzard, J.H., Miller, J.J., Whitford, W.G., 1985. Analysis of above-
gurdaspurensis (Isoptera: Termitidae) in plant decomposition in a tropical ground gallery construction by the subterranean termite Gnathamitermes
grassland. Pedobiologia 22, 254–261. tubiformans (Isoptera: Termitidae). Environmental Entomology 14, 470–474.
Holec, M., Frouz, J., 2006. The effect of two ant species Lasius niger and Lasius MacKay, W.P., Whitford, W.G., 1988. Spatial variability of termite gallery production
flavus on soil properties in two contrasting habitats. European Journal of Soil in Chihuahuan Desert plant communities. Sociobiology 14, 281–289.
Biology 42(Supplement 1), S213–S217. MacMahon, J.A., Mull, J.F., Crist, T.O., 2000. Harvester ants (Pogonomyrmex spp.):
Holldobler, B., Wilson, E.O., 1990. The Ants. The Belknap Press of Harvard their community and ecosystem influences. Annual Review in Ecology and
University Press, Cambridge, Massachusetts, 732 pp. Systematics 31, 265–291.
Holt, J.A., Coventry, R.J., Sinclair, D.F., 1980. Some aspects of the biology and Mandel, R.D., Sorenson, C.J., 1982. The role of harvester ant (Pogonomyrmex
pedological significance of mound-building termites in a red and yellow earth occidentalis) in soil formation. Soil Science Society of America Journal 46,
landscape near Charters Towers, North Queensland. Australian. Journal of Soil 785–788.
Research 18, 97–109. Mando, A., Miedema, R., 1997. Termite-induced change in soil structure after
Hughes, L., 1991. The relocation of ant nest entrances: potential consequences for mulching degraded (crusted) soil in the Sahel. Applied Soil Ecology 6, 261–263.
ant-dispersed seeds. Australian Journal of Ecology 16, 207–214. Mando, A., Stroosnijder, L., Brussaard, L., 1996. Effects of termites on infiltration
Humphreys, G.S., 1981. The rate of ant mounding and earthworm casting near into crusted soil. Geoderma 74, 107–113.
Sydney, New South Wales. Search 12, 129–131. Miller, S.T., Brinn, P.J., Fry, G.J., Harris, D., 1994. Microtopography and agriculture
James, A.I., Eldridge, D.J., Koen, T.B., Whitford, W.G., 2008. Landscape position in semiarid Botswana. 1. Soil variability. Agricultural Water Management 26,
moderates how ant nests affect hydrology and soil chemistry across a 107–131.
Chihuahuan Desert watershed. Landscape Ecology 23, 961–975. Nash, M.H., Whitford, W.G., 1995. Subterranean termites: regulators of soil organic
Jiménez, J.J., Decaëns, T., Lavelle, P., 2008. C and N concentrations in biogenic matter in the Chihuahuan Desert. Biology and Fertility of Soils 19, 15–18.
structures of a soil-feeding termite and a fungus-growing ant in the Colombian Nkem, J.N., Lobry de Bruyn, L.A., Grant, C.D., Hulugalle, N.R., 2000. The impact of
savannas. Applied Soil Ecology 40, 120–128. ant bioturbation and foraging activities on surrounding soil properties.
Johnson, D.L., 1989. Subsurface stone lines, stone zones, artifact – manuport Pedobiologia 33, 609–621.
layers and biomantles produced by bioturbation via pocket gophers (Thomomys Obi, J.C., Ogunkunle, A.O., 2009. Influence of termite infestation on the spatial
bottae). American Antiquity 54, 370–389. variability of soil properties in the Guinea savanna region of Nigeria. Geoderma
Johnson, D.L., Domier, J.E.J., Johnson, D.N., 2005. Reflections on the nature of soil 148, 357–363.
and its biomantle. Annals of the Association of American Geographers 93, 11–31. Orhue, E.R., Uzu, O.F., Osaigbovo, U.A., 2007. Influence of activities of termites on
Jouquet, P., Barré, P., Lepage, M., Velde, B., 2005. Impact of subterranean fungus- some physical and chemical properties of soils under different land use
growing termites (Isoptera, Macrotermitinae) on chosen soil properties in a West patterns: a review. International Journal of Soil Science 2, 1–14.
African savanna. Biology and Fertility of Soils 41, 365–370. Palmer, A.R., Novellie, P.A., Lloyd, J.W., 1999. Community patterns and dynamics.
Jouquet, P., Bottinelli, N., Lata, J.-C., Mora, P., Caquineau, S., 2007. Role of the In: Dean, W.R.J., Milton, S.J. (Eds.), The Karoo. Cambridge University Press,
fungus-growing termite Pseudacanthotermes spiniger (Isoptera, Macrotermitinae) Cambridge, pp. 208–223.
in the dynamic of clay and soil organic matter content. An experimental Paton, T.R., Humphreys, G.S., Mitchell, P.B., 1995. Soils: A New Global View. Yale
analysis. Geoderma 139, 127–133. University Press, New Haven, 213 pp.
Jouquet, P., Mamou, L., Lepage, M., Velde, B., 2002. Effect of termites on clay Petal, J., 1978. The role of ants in Ecosystems. In: Brain, M.V. (Ed.), Production
minerals in tropical soils: fungus-growing termites as weathering agents. Ecology of Ants and Termites. Cambridge University Press, Cambridge,
European Journal of Soil Science 53, 521–527. pp. 293–325.
Killgore, A., Jackson, E., Whitford, W.G., 2009. Fire in Chihuahuan Desert Petts, A.E., Hill, S.M., Worrall, L., 2009. Termite species variations and their
grassland: short-term effects on vegetation, small mammal populations, and importance for termitaria biogeochemistry: towards a robust media approach for
faunal pedoturbation. Journal of Arid Environments 73, 1029–1034. mineral exploration. Geochemistry: Exploration, Environment Analysis 9, 257–266.
Kilpelainen, J., Finer, L., Niemela, P., Domisch, T., Neuvonen, S., Ohashi, M., Risch, Pire, E.F., Torres, P.F., Romagnoli, O.D., Lewis, J.P., 1991. The significance of ant-hills
A.C., Sundstrom, L., 2007. Carbon, nitrogen and phosphorus dynamics of ant in depressed areas of the Great Chaco. Revista de Biologı́a Tropical 39, 71–76.
mounds (Formica rufa group) in managed boreal forests of different Porter, S.D., Bhatkar, A., Mulder, R., Vinson, S.B., Clair, D.J., 1991. Distribution and
successional stages. Applied Soil Ecology 36, 156–163. density of polygyne fire ants (Hymenoptera: Formicidae) in Texas. Journal of
Lafleur, B., Bradley, R.L., Francoeur, A., 2002. Soil modifications created by ants along Economic Entomology 84, 866–874.
a post-fire chronosequence in lichen-spruce woodland. Ecoscience 9, 63–73. Reatto, A., deSouza Martins, E., da Silva, E.M., et al., 2009. Development and origin
Lal, R., 1987. Termites. In: Lal, R. (Ed.), Tropical Ecology and Physical Edaphology. of the microgranular structure in latosols of the Brazilian Central Plateau:
Wileyand, New York, pp. 337–422. significance of texture, mineralogy, and biological activity. Catena 76, 122–134.
292 Effects of Ants and Termites on Soil and Geomorphological Processes

Richards, P.J., 2009. Aphaenogaster ants as bioturbators: impacts on soil and slope Waloff, N., Blackith, R.E., 1962. The growth and distribution of the mounds of
processes. Earth Science Reviews 96, 92–106. Lasius flavus (F.) (Hymenoptera: Formicidae) in Silwood Park, Berkshire. Journal
Robert, O.E., 2007. Influence of activities of termites on some physical and of Animal Ecology 31, 421–437.
chemical properties of soils under different land use patterns: a review. Whitford, W.G., 1999. Effects of habitat characteristics on the abundance and
International Journal of Soil Science 2, 1–14. activity of subterranean termites in arid southeastern New Mexico (Isoptera).
Salem, M., Hole, F.D., 1968. Ant (Formica exsectoides) pedoturbation in a forest Sociobiology 34, 493–504.
soil. Soil Science Society of America Proceedings 32, 563–567. Whitford, W.G., 2003. The functional significance of cemented nest caps of the
Salick, J., Herrera, R., Jordan, C.F., 1983. Termitaria: nutrient patchiness in nutrient harvester ant, Pogonomyrmex maricopa. Journal of Arid Environments 53,
deficient rain forests. Biotropica 15, 1–7. 281–284.
Sarcinelli, T.S., Schaefer, C.E.G.R., de Souza Lynch, L., Arato, H.D., Viana, J.H.M., Whitford, W.G., Barness, G., Steinberger, Y., 2008. Effects of three species of
de Albuquerque Filho, M.R., Goncalves, T.T., 2009. Chemical, physical and Chihuahuan Desert ants on annual plants and soil properties. Journal of Arid
micromorphological properties and adjacent soils along a toposequence in Zona Environments 72, 392–400.
da Mata, Minas Gerais State, Brazil. Catena 76, 107–113. Whitford, W.G., DiMarco, R., 1995. Variability in soils and vegetation associated
Schumacher, A., Whitford, W.G., 1976. Spatial and temporal variation in with harvester ant (Pogonomyrmex rugosus) nests on a Chihuahuan Desert
Chihuahuan Desert ant faunas. The Southwestern Naturalist 21, 1–8. watershed. Biology and Fertility of Soils 20, 169–173.
Segev, U., 2010. Regional patterns of ant-species richness in an arid region: the Whitford, W.G., Forbes, G.S., Kerley, G.I., 1995. Diversity, spatial variability, and
importance of climate and biogeography. Journal of Arid Environments 74, functional roles of invertebrates in desert grassland ecosystems. In: McClaran,
646–652. M., Van Devender, T. (Eds.), Desert Grasslands. University of Arizona Press,
Soyer, J., 1987. Role des termites dans la formation du complexe de la stone-line. Tucson, AZ, pp. 152–195.
Geo-Eco-Trop 11, 97–108. Whitford, W.G., Ludwig, J.A., Noble, J.C., 1992. The importance of subterranean
Stoops, G., 1989. Contribution of in situ transformations to the formation of stone- termites in semi-arid ecosystems in south-eastern Australia. Journal of Arid
layer complexes in central Africa. Geo-Eco-Trop 11, 139–149. Environments 22, 87–92.
Sudd, J.H., 1969. The excavation of soil by ants. Zeitschrift fur Tierpsychologie 26, Whitford, W.G., Steinberger, Y., Ettershank, G., 1982. Contributions of subterranean
257–276. termites to the ‘‘economy’’ of Chihuahuan Desert ecosystems. Oecologia 55,
Tongway, D.J., Ludwig, J.A., Whitford, W.G., 1989. Mulga log mounds: fertile 298–302.
patches in the semi-arid woodlands of eastern Australia. Australian Journal of Wiken, E.B., Boersma, L.M., Lavkulich, L.M., Farstead, L., 1976. Biosynthetic
Ecology 14, 263–268. alteration in a British Columbia soil by ants (Formica fusca, Linne). Soil
Torales, G.J., Coronel, J.M., Laffont, E.R., Fontana, J.L., Godoy, M.C., 2009. Termite Science Society of America Journal 40, 422–426.
associations (Insecta, Isoptera) in natural or semi-natural plant communities in Wilkinson, M.T., Richards, P.J., Humphreys, G.S., 2009. Breaking ground:
Argentina. Sociobiology 54, 383–437. pedological, geological, and ecological implications of soil bioturbation.
Tschinkel, W.R., 2003. Subterranean ant nests: trace fossils past and future? Earth-Science Reviews 97, 257–272.
Palaeogeography, Palaeoclimatology. Palaeoecology 192, 321–333. Wood, T.G., Sands, W.A., 1978. The role of termites in ecosystems. In: Brian, M.V.
Wagner, D., Brown, M.J.F., Gordon, D.M., 2004. Development of harvester ant (Ed.), Production Ecology of Ants and Termites. Cambridge University Press,
colonies alters soil chemistry. Soil Biology and Biochemistry 36, 797–804. Cambridge, pp. 245–292.
Wagner, D.L., Van Driesche, R.G., 2010. Threats posed to rare or endangered Zaady, E., Groffman, P.M., Shachak, M., Wilby, A., 2003. Consumption and
insects by invasions of nonnative species. Annual Review of Entomology 55, release of nitrogen by the harvester termite Anacanthotermes ubachi Navas in the
547–568. northern Negev Desert, Israel. Soil Biology and Biochemistry 35, 1299–1303.

Biographical Sketch

Walter G. Whitford is currently a collaborating scientist with the USDA-ARS Jornada Experimental Range. He is a
senior research ecologist emeritus of the US Environmental Protection Agency and a professor emeritus at New
Mexico State University. His research for the past 47 years has focused on arid lands ecology with an emphasis on
soil ecology.

Dr. David Eldridge is a research scientist based at the University of NSW, Sydney, Australia. His research aims to
understand how arid and semi-arid ecosystems function, specifically the relationships between plants, animals,
and soil processes. He is currently examining the effects of vertebrates and invertebrates on soil processes, the
ecology of desert soil crusts and shrub encroachment. The focus of his research is on the semi-arid woodlands of
eastern Australia, and he has long-term research interests in the western United States.
 12.20 Beaver Hydrology and Geomorphology
CJ Westbrook, University of Saskatchewan, Saskatoon, SK, Canada
DJ Cooper, Colorado State University, Fort Collins, CO, USA
DR Butler, Texas State University-San Marcos, San Marcos, TX, USA
r 2013 Elsevier Inc. All rights reserved.

12.20.1 Introduction 293

12.20.2 History and Geographic Distribution of Beaver 294
12.20.3 Main Hydrologic Signatures of Beaver 294
12.20.4 Influence of Beaver Activities on the Water Cycle 295
12.20.4.1 Surface-Water Storage 296
12.20.4.2 Streamflow 297
12.20.4.3 Evapotranspiration 298
12.20.4.4 Surface Water–Groundwater Exchange 299
12.20.5 Beaver Geomorphology – Landforms and Sedimentation 300
12.20.6 Conclusions and Future Challenges 302
References 302

Glossary Phytogeomorphology The study of the geomorphic

Ecotone A transition area of vegetation between two effects of plants, and the relation of different plant species
different plant communities, such as forest and grassland, to landforms.
or forest and tundra. Zoogeomorphology The study of the geomorphic effects
Geophagy The eating/ingestion of soil. of animals.
Lithophagy The eating/ingestion of rocks.

Abstract

Hydrological and geomorphological processes are influenced by beaver (Castor canadensis and C. fiber) activities in aquatic
and semi-aquatic environments throughout much of North America, Eurasia, and the austral archipelago of Chile and
Argentina. The main hydrologic signature of beaver activities varies with hydrogeomorphic setting – along confined streams
it is the pond formed upstream of dams, along unconfined streams it is downstream flooding on floodplains and terraces,
and in preexisting wetlands it is the formation of open-water bodies. A review of the existing literature shows that it is rich
with descriptions of how beaver activities influence specific hydrologic and geomorphic processes. The main findings are
that beaver dams moderate stream flows, increase surface water and riparian groundwater storage, regulate hyporheic flows,
and enhance evapotranspiration rates. Beavers also excavate canals on the margins of beaver ponds and create extensive
burrow systems in riverbanks where damming is not possible. Bank burrows are also common in beaver ponds. Missing in
the beaver hydrogeomorphological literature, however, are clear linkages between affected hydrological processes and
ecosystem functioning, especially at larger spatial and temporal scales. In addition, knowledge of effects of beaver activities
on the form and function of the expansive peatlands that span northern latitudes is lacking.

12.20.1 Introduction Examples include alligators (Alligator mississippiensis) that

provide refuge for a variety of bird species by creating wallows
Organisms that can create, modify, and maintain their habitat in Everglades National Park (Craighead, 1968), prairie dogs
directly or indirectly by changing the availability of biotic or (Cynomys spp.) that alter soil nutrient cycling in short
abiotic resources are commonly called ecosystem engineers and mixed grass prairie by burrowing and creating soil
(Jones et al., 1994; Brown, 1995). Ecosystem engineers have a mounds (Whicker and Detling, 1988), Sphagnum spp. mosses
large influence on the flow of materials and energy in eco- in northern or boreal peatland ecosystems that control
systems throughout North America (Butler, 1995; Jones et al., peat formation and the nutrient status in soil water (van
1997) and control the presence and abundance of other biota. Breemen, 1995), and phytoplankton that regulate light
interception and temperature regimes in freshwater lakes
(Mazumder et al., 1990). In aquatic and semi-aquatic
Westbrook, C.J., Cooper, D.J., Butler, D.R., 2013. Beaver hydrology and
environments of North America and Eurasia, beaver (Castor
geomorphology. In: Shroder, J. (Editor in Chief), Butler, D.R., Hupp, C.R.
(Eds.), Treatise on Geomorphology. Academic Press, San Diego, CA, vol. 12, canadensis and C. fiber) are the key ecosystem engineers
Ecogeomorphology, pp. 293–306. (Gurney and Lawton, 1996) that affect landscape structure

Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00337-7 293

294 Beaver Hydrology and Geomorphology

and dynamics at a level rivaled only by humans (Butler, 1995; mid-1700s to mid-1800s, which helped encourage Caucasian
Gurnell, 1998). settlement of the West (Cline, 1974 as cited in Naiman et al.,
Descriptions of beaver natural history and effects on eco- 1988). Excessive trapping led to a near extirpation of
logical processes were plentiful in the late 1800s and early C. canadensis by 1900 throughout North America. A similar
1900s (e.g., Mills, 1913; Dugmore, 1914). Some of the con- situation occurred throughout Eurasia. Nolet and Rosell
cepts – such as beaver create and maintain riparian wetlands, (1998) estimated that only B1200 C. fiber in eight relict
beaver ponds function as efficient sediment traps that even- populations remained in Eurasia by 1900.
tually become meadows, and below beaver dams there is re- At the beginning of the twentieth century, the direction of
duced erosion because the dam reduces stream velocity – were fashion moved away from furs and toward silk (Marston,
generally accepted and became grounded in the ecological 1994). Public concerns about the large declines in beaver and
literature without rigorous quantitative testing (Gurnell, 1998; other fur-bearing animals led to the implementation of trap-
Meentemeyer and Butler, 1999). Only in the last few decades ping regulations. The population recovery that ensued was
researchers have begun to test concepts of how beaver affect facilitated by reintroduction efforts that peaked in the 1950s
physical and biological processes in ecosystems (e.g., Wilde in North America (Baker and Hill, 2003) and were first initi-
et al., 1950; Barnes and Dibble, 1986; Naiman et al., 1986; ated in Sweden in the 1920s (Nolet, 1997). Although beaver
Butler and Malanson, 1995). Research has focused on the are now present throughout most of their former North
influence of beaver on ecosystem processes, primarily on first- American range, they are estimated to have repopulated at
and second-order rivers, and the effects of beaver dams and approximately 10% of their historic density (6–12 million
the upstream ponds on local hydrogeomorphology (Naiman individuals) due to a concomitant loss of habitat through
et al., 1988; Butler, 1995; Gurnell, 1998; Butler and Malanson, conversion of wetlands to drylands (Naiman et al., 1988).
2005). Beaver can also have significant impacts on landscape Beaver populations have not thrived where excessive com-
processes in rivers of any size or in preexisting wetlands petition exists with other herbivores for riparian shrubs and
through felling of trees for food and building of dams, dens, trees, such as some parks in the Rocky Mountains (Meente-
lodges, canals, and food caches. Synthesized in this article is meyer and Butler, 1995; Nietvelt, 2001; Baker et al., 2005).
the existing body of literature describing the influence of Because C. fiber and C. canadensis were initially believed to be
beaver activities on hydrologic and geomorphic processes. the same species, both species have seen a revival in Eurasia
since the 1920s and are currently expanding their range
(Rosell et al., 2005). At least 639 000 beaver are currently
12.20.2 History and Geographic Distribution of present in Eurasia, with the greatest populations residing in
Beaver Russia, Latvia, Sweden, Norway, and Lithuania (Halley and
Rosell, 2003).
The genus Castor originated in the Pleistocene (1.8 million A population of beavers (C. canadensis) also exists in South
years BP) or the late Tertiary (2.4 million years BP). The for- America. Twenty-five mating pairs were introduced into Tierra
mation and melting of Pleistocene ice sheets in North America del Fuego, Argentina by the Army in 1946 to promote the fur
probably affected beaver evolution and abundance during this industry (Lizarralde, 1993). Due to the absence of natural
period (Baker and Hill, 2003). The paleo-Indians who mi- predators, the population expanded to between 60 000 and
grated to the continent approximately 18 000–8000 years BP 90 000 individuals at the start of the twenty-first century
likely interacted with beaver (C. canadensis); however, their (Skewes et al., 2006) that occupy nearly every watershed in the
effects on beaver populations or distribution are not known. area (Anderson et al., 2006).
Prior to the settlement of Europeans in North America, the
beaver population was estimated to be between 60 and 400
million (Seton, 1929). Their range was estimated to cover 12.20.3 Main Hydrologic Signatures of Beaver
B15 million km2 and included all of North America other
than the North Slope of Alaska, northern Nunavut, southern The biggest influence of beaver on hydrogeomorphic processes
California, south central Florida, central to southern Nevada, is through dam construction. Dams can grow over time from
and the Texas Panhandle (Jenkins and Busher, 1979; Baker being small and contained wholly within the channel to being
and Hill, 2003). The European beaver (C. fiber) originally long and extending across an entire floodplain (Richard, 1967
ranged from the tundra through the steppe zones of Europe as cited in Gurnell, 1998). The main hydrologic effect of
and Asia (Nolet, 1997). beaver that inhabit stream courses depends on the geo-
The arrival of Europeans initiated the systematic elimin- morphic constraints of the channel and watershed relief. The
ation of beaver from much of its North American range, pond is the main hydrologic signature of beaver dams built
largely to supply pelts for felt hats in Europe and to a lesser within relatively confined riverine systems (Figure 1(a);
extent, food and medicine derived from castoreum (Bryce, Naiman et al., 1988). By contrast, downstream inundation of
1904). Large fur-trading companies, such as the Hudson Bay floodplains and terraces (Figure 1(b); Westbrook et al., 2011)
and the Rocky Mountain Fur Trading Companies, employed and conversion of a single-thread channel to one that is
both European and First Nations people to trap beaver in multithread are the main hydrologic signatures of beaver
northeastern Canada and the eastern United States starting in dams built within unconstrained riverine systems. On large
the early 1600s through the 1700s (Baker and Hill, 2003). rivers (greater than fifth order), the main hydrologic signature
Continued demand for felt hats and the decline of beaver of beaver is riparian tree cutting (Figure 1(c); Johnston and
in the East persuaded trappers to move west during the Naiman, 1990a; Breck et al., 2003). In areas of low relief,
 Beaver Hydrology and Geomorphology 295

Floodplain

Channel

(a) (b) (c)

Figure 1 The main hydrologic signatures of beaver inhabiting streams: (a) a confined stream where storage of surface water is enhanced
behind a series of beaver dams; (b) inundation of floodplains and terraces downstream of a beaver dam on an unconfined stream; and
(c) clear felling of riparian trees by beaver. (a) From Denver Public Library, Western History Collection, Rose & Hopkins, H-168.

Figure 2 A beaver pond in northern Ontario showing how a stream course was diverted across a watershed boundary. Reproduced from
Dugmore, A.R., 1914. The Romance of the Beaver: Being the History of the Beaver in the Western Hemisphere. Lippincott, Philadelphia, 330 pp.

beaver dams can pond water high enough to divert it over the to the main vector of groundwater flow (J. Thie, personal
watershed divide and into another basin (Figure 2; Dugmore, observation; C. Westbrook and D. Cooper, unpublished data).
1914; Woo and Waddington, 1990). The loss of beaver from The open-water bodies formed by beaver dams are commonly
streams can initiate channel incision (Marston, 1994; Persico colonized by emergent and floating macrophytes (Turetsky
and Meyer, 2009) and move the channel from multithread to and St. Louis, 2006).
single thread (Green and Westbrook, 2009).
The main hydrologic signature of beaver that inhabit pre-
existing wetlands such as peatlands is shallow open-water 12.20.4 Influence of Beaver Activities on the Water
bodies (Figure 3; Cunningham et al., 2006). Beaver form Cycle
open-water bodies by building dams in the moat areas of fens
(Racine and Walters, 1994) and bogs (Heinselman, 1965; Beaver were historically ubiquitous on almost every water
Dinsmore et al., 2009). Also, beaver can dam seepage in course in North America and much of Europe. Due to exten-
peatlands formerly lacking open water (Rebertus, 1986; Ray sive trapping/removals, the effect of beaver in many areas has
et al., 2004); these dams are generally oriented perpendicular been diminished, and some consider riparian contraction of
296 Beaver Hydrology and Geomorphology

activities affect hydrologic processes is discussed in the fol-

lowing sections.

12.20.4.1 Surface-Water Storage

The areal extent of flooding created by beaver dams is con-
trolled by a combination of dam height relative to river bank
height, width of the dam, elevation of the riparian area relative
to the height of the dam, and degree of valley constraint.
Beaver dams are unlike debris dams in that they are generally
characterized by the tight arrangement of small branches such
that they collect fine sediment across their upstream surface
(Bisson et al., 2006). This architecture promotes greater up-
stream pooling of water. Beaver further expand their habitats
and, therefore, surface-water storage, by digging canals (see
below) and building dams on floodplains and terraces
(Naiman et al., 1988; Baker and Hill, 2003).
The surface-water storage capacity upstream of dams in
confined valleys may be large (Neff, 1957; Naiman et al.,
1988). This is especially true where beaver, either one family
or generations of a family, build a network of dams along a
stream. For example, Zurowski (1989) as cited in Rosell et al.
(2005) reported 24 beaver dams along a 1300-m stream reach,
varying in length from 1.5 to 60 m. Both beaver habitat
abandonment and dam removal from stream courses can
drastically reduce surface-water storage (Neff, 1957; Hillman,
1998; Green and Westbrook, 2009).
The volume of water stored is partly a function of dam
water tightness, as well as dam size, stream depth, flow vol-
ume, and other factors. Woo and Waddington (1990) pre-
sented a system for classifying dams based on their capacity to
Figure 3 The main hydrologic signature of beaver inhabiting impound water as a function of degree of decay. They found
preexisting wetlands is shallow open-water bodies. The top photo is that the overflow class produced the largest water-level dif-
of a beaver dam and open-water pond formed in a peatland in Rocky ferences upstream and downstream of the dam. Although not
Mountain National Park, USA. The bottom photo is open water measured, presumably overflow dams also create the largest
created by beaver in a marshy depression near Dubingiai, Lithuania. upstream surface-water storage potential.
In unconfined valleys, storage of water upstream of dams
may be relatively small compared to downstream flooding if
large proportions to have occurred (Naiman and Rogers,
dams do not extend across the entire valley width (Westbrook
1997; Naiman et al., 1988). However, surprisingly little data-
et al., 2006). A large proportion of the volume of the stream
rich documentation exists on the influence of their activities
may instead be diverted onto the floodplain. For example,
on the hydrologic (water) cycle. For the most part, hydrolo-
Woods (2000) showed how 70% of the volume of the
gists have tended to steer away from examining hydrological
Colorado River near its headwaters was diverted off channel
processes in watersheds where beaver are abundant. Some
by one beaver dam. This overbank flow was further spread
reasons for this are explained well in the following quote.
across the floodplain and terraces in a braided river fashion by
a series of off-channel dams, ponds, and canals (Westbrook
The beaver dams create the largest obstacle to attaining a water et al., 2006). Some of the redirected stream water eventually
balance at present. Besides reducing the peaks on the hydrograph
and preventing development of meaningful short-term precipi-
returns to the stream tens to hundreds of meters downstream
tation-runoff relationships, the beavers also release water at inter- (Lowry and Beschta, 1994; Westbrook et al., 2006), but some
vals during the winter to maintain an air space under the ice, thus water may recharge the underlying aquifer (Winter, 1998) or
creating problems in establishing the base flow rate. Also the size of be lost to evaporation (Woo and Waddington, 1990).
the ponds behind the dams varies with the degree of beaver activity,
Ecosystems such as the extensive boreal forest consist of
creating more problems in determining the surface below-ground
relationships and estimating evapotranspiration. Removing the many drainageways that are not mapped as streams. Several
beaver and destroying the dams is an obvious answer to the researchers have documented rates at which beaver have
problem, but this changes the natural balance (Storr, 1970: 323). converted boreal forest to aquatic patches. For example,
Johnston and Naiman (1990b) used a geographic information
Documented hydrogeomorphic processes affected by bea- system (GIS) to show that beaver caused a 12% increase in
ver activities include streamflow regulation, valley flooding, total area impounded within Voyageurs National Park, USA.
and groundwater–surface water interactions. Existing literature Martell et al. (2006) found considerable expansion of lentic
describing how the main hydrologic signatures of beaver habitat along riparian systems in northern Alberta in the past
 Beaver Hydrology and Geomorphology 297

50 years. Newly created surface-water storage can persist in a influence of beaver on streams is through dam construction,
stable state for decades due to the low energy associated with which can change the volume and timing of water flowing
the shallow-gradient drainageways. Ulevičius (1999) con- downstream. Flows are spread across dam crests that are
firmed how the reintroduction of C. fiber greatly enhanced generally wider than the original stream channel, thereby
surface-water storage in the boreal forests of Lithuania. Similar dissipating stream energy (Gurnell, 1998). Most commonly
expansion of surface-water storage has been noted throughout reported in the literature is that dams reduce downstream
boreal forests in Eurasia as beavers reconquer their previous annual discharge and attenuate seasonal-flow fluctuations
territories (Halley and Rosell, 2002). Beaver may help mitigate (Figure 4). This is due to the large storage capacity of beaver
the effects of changing climate on open-water storage in the ponds (Duncan, 1984). By slowing or diverting streamflow,
boreal forest. Hood and Bayley (2008) showed that the pres- beaver significantly affect the sediment budget of rivers and
ence of beaver was associated with a ninefold increase in the floodplains (Butler, 1995; Butler and Malanson, 1995; Pollock
presence of open water in Elk Island National Park, Canada, et al., 2007; Westbrook et al., 2011).
which persisted throughout both wet and drought periods. Only a few researchers have actually documented effects of
However, boreal forest geomorphology will eventually limit beaver dams on streamflows. Reports include that of Woo and
locations available for aquatic patch creation after beaver re- Waddington (1990) who found reduced total annual runoff
occupy an area, resulting in finite volumes in beaver-created from a subarctic basin that was dammed by beaver compared
surface-water storage. Interestingly, exploitation of natural re- to a controlled basin. Reid (1952) had similar findings in the
sources in the boreal forest has allowed beaver to expand their Adirondacks. Vining (2004) documented reductions in in-
habitat. Flynn (2006) showed that beaver ponds are colocated stantaneous flows along a beaver-dammed prairie stream.
with culverts under road crossings in the Alberta, Canada. Green and Westbrook (2009) reported a fivefold increase in
Colonization of previously marginal habitat by beaver led to mean discharge following beaver dam removal in a mountain
an increase in the spatial distribution of open-water wetlands. stream in British Columbia. Filling of ponds during peak flows
Beaver also build dams in peatlands that lack open water, can lead to higher base flows (Parker, 1986). Face-Collins and
termed closed peatlands (Rebertus, 1986; Racine and Walters, Johnston (2007) reported that beaver dams maintained flow
1994; Walbridge, 1994). In fact, the longest beaver dam in the in streams in South Dakota despite extreme drought. Storage
world (B850-m long) is reported to be located in a peatland of water in a series of beaver dams along a stream course could
in northern Alberta, Canada that lacks channelized surface result in continuous flow in what was previously an inter-
drainage (J. Thie, personal communication). Beaver are mittent stream (Yeager and Hill, 1954; Rutherford, 1955).
thought to create dams in peatlands by dredging canals and Storage of water in beaver ponds and on floodplains causes
damming seepage outlets (Rebertus, 1986; Mitchell and a large increase in the probability of overbank flooding for a
Niering, 1993). Damming these causes anchored mat vege- given stream discharge (Westbrook et al., 2006) that leads to
tation to be eliminated or partially sheared, resulting in the an increase in the annual residence time of stream water
formation of open water (Ray et al., 2001). A few researchers (Devito and Dillon, 1993). Multiple dams in sequence can
have investigated temporal trends in open-water formation in magnify these effects (Grasse, 1951). Further, extensive chan-
peatland by beaver (e.g., Rebertus, 1986; Härkönen, 1999), ges to the drainage network that can occur because of a beaver
but no studies to date have estimated how much water in dam, such as the subdivision of the main stream channel into
stored in these beaver ponds. Estimation of water storage in
the large number of beaver dams in remote peatlands, for
example, across the Canadian subarctic, likely requires the
use of remotely sensed imagery. The simplified volume–
area–depth method developed specifically to estimate water
storage in prairie pothole wetlands (Minke et al., 2010) can be
parametrized via Light Detection And Ranging (LiDAR) im-
agery (Minke et al., submitted). Preliminary field tests where
the method was applied to beaver ponds yielded promising
results (C. Westbrook and K. Allen, unpublished data) and
widespread testing is currently underway.
Beaver inhabit many environments but do not build dams.
Surface-water storage in these environments is, thus, unlikely
to be affected. For example, lentic environments such as large
lakes can support beaver. Even if the beaver were to dam lake
outlets, these dams would have little effect on storage capacity
of lake water. Beaver also live in rivers too large to dam, for
example, the Danube. They tend to dwell in bank dens and,
thus, have little to no influence on stream-water storage.
Figure 4 Beaver dams generally reduce downstream discharge,
especially during low flows, in relation to the storage capacity of the
12.20.4.2 Streamflow upstream pond as influenced by dam size, age, and condition. The
2.5-m-high dam in the photo (located near Hudson Bay,
Perhaps the most commonly reported influence of beaver Saskatchewan, Canada) has a high degree of water tightness,
on the hydrological cycle is that of streamflow. The main allowing very little downstream flow.
298 Beaver Hydrology and Geomorphology

several distributaries that flow across floodplains (Townsend, forested environment, which enhances short-wave radiation
1953; Westbrook et al., 2006), have been shown to reduce absorption (Vowinckel and Orvig, 1973). Season and pond
downstream discharge by up to 70% (Woods, 2001). Beaver age have small effects on radiation absorption. Pond age is
dam-driven flow diversions across watershed boundaries, as considered a proxy for water depth, where older and, thus,
illustrated by Dugmore (1914), are hypothesized to increase deeper ponds are able to store more heat. Although radiation
runoff volumes in receiving watersheds, but no reports exist in adsorption changes may be small, the resulting changes in
the literature. evaporation can be large. Vowinckel and Orvig (1973) re-
The magnitude of beaver dam effects on streamflow is a ported that the small changes in absorbed solar radiation by
function of storage capacity of the upstream pond as influ- beaver ponds compared to riparian forests are equivalent to a
enced by dam size, age, and condition (Meentemeyer and southward shift of a site by 71 latitude in spring and 1.51
Butler, 1999). Younger dams are less efficient in reducing annually.
streamflow velocity than older dams because dams are porous Beaver dam building and foraging activities commonly
when first built and then leak less as sediment accumulates affect riparian plant community composition and plant bio-
upstream (Woo and Waddington, 1990). Streamflow magni- mass along small streams up to large rivers. Foraging activities
tude is also important in determining whether beaver dams though are only not limited to riparian areas. For example,
will influence downstream discharge. Burns and McDonnell Crawford et al. (1976) showed that beaver can climb up Ap-
(1998) used isotopes to show that beaver dams delayed palachian mountain slopes with grades of up to 80% to cut
downstream water transmission during small but not large- trees. Oddly, no researchers have attempted to draw quanti-
runoff events. Tracer injection experiments conducted by tative links between changes in plant community composition
Lautz et al. (2006) yielded similar results. On regulated and evapotranspiration rates in beaver habitats. Reviewed
streams, Andersen and Shafroth (2010) showed that releases below are the main plant community changes resulting from
from man-made dams can significantly damage or breach beaver activities that have been documented in the literature,
beaver dams, thereby reducing their influence on downstream and hypotheses about how these might influence evapo-
flow transmission. Although beaver generally reduce the transpiration rates.
severity of flooding events, several reports describe situations Beaver can reverse the process of succession by harvesting
where beaver dam failure enhanced stream flows and early and mid-successional species (as reviewed in Rosell et al.,
caused catastrophic downstream flooding (e.g., Butler, 1989; 2005), meaning that deciduous stands become replaced with
Hillman, 1998; Butler and Malanson, 2005). stands of shrubs or herbaceous plants (e.g., Walbridge, 1994).
Beaver dams do not significantly influence streamflow in Typically, evapotranspiration rates are scaled to the area
all environments. In the western United States where many indices of plant leafs (Running et al., 1989). Ecological
streams are deeply incised, streamflow regulation may not modeling efforts of Sturtevant (1998) indicate that beaver
occur to a great extent as beaver may not be able to build a dam-driven disruption to wetland vegetation succession pat-
dam high enough to extensively pond water (Apple et al., terns may be long term. Any resulting changes in evapo-
1985). However, beaver dams may enhance water storage in transpiration rates are, thus, likely to persist for decades or
floodplain soils, that is, ‘bank storage’, along losing stream centuries.
reaches (Gurnell, 1998; Hillman, 1998). Wright et al. (2002) showed that beaver dams increase
plant species richness at the landscape level, but by creating
floristically distinct, relatively homogenous patches within old
12.20.4.3 Evapotranspiration
ponds. Where beaver build dams in unconstrained river val-
Very few studies have directly examined the influence of beaver leys, heterogeneous patches can result (Westbrook et al.,
activities such as tree cutting and dam building on evapo- 2011). Both types of shifts in plant species composition could
transpiration rates. Mostly, the existing body of literature raise or lower transpiration rates, depending on if the new
highlights the principle that an area with no water deficit will hydrochemical conditions of the beaver meadows were con-
have actual evaporation rates near or at potential. By contrast, ducive to supporting plants with greater or smaller water and
areas with some resistance to the upward movement of water, nutrient demands. Invasive species abundance may also in-
that is, high soil water tension, will have lowered actual to crease when beaver inhabit a site. Webb and Leake (2006)
potential evaporation ratios. In regard to beaver, this means partially attributed an increase in native riparian vegetation
that enhanced surface storage of water, either in beaver ponds along rivers in the American southwest to trapping of beaver.
or in floodplain depressions, should enhance evaporat- Further, Lesica and Miles (2004) showed that beaver indirectly
ion rates. Woo and Waddington (1990) and Burns and enhanced the growth of invasive species such as Russian olive
McDonnell (1998) demonstrated enhanced evaporation from and tamarisk within 50 m of stream channels in the American
beaver-dammed sites in subarctic and temperate environ- west. These plants consume more water than native cotton-
ments, respectively. Newton et al. (1996) reported 25% greater wood trees (Di Tomasco, 1998). Anderson et al. (2006)
evapotranspiration rates in an Adirondack watershed that showed much of the increased herbaceous richness in sub-
contained a beaver pond and associated wetlands than in a Antarctic beaver dammed sites resulted from invasion of exotic
nearby, hydrologically similar watershed lacking beaver. plants. Sometimes beaver activities could result in loss of rare
Coupled water and energy budgets can be useful in ex- plants (Bonner et al., 2009), but because of their low abun-
plaining how beaver, through converting vegetated areas to dance, negligible effects on evapotranspiration rates are likely.
open surface waters, affect evaporation rates. Such a landscape Plant biomass can be either reduced or enhanced along
change lowers the albedo of the pond surface compared to a stream systems, depending on the ability of the forage species
 Beaver Hydrology and Geomorphology 299

targeted to rebound. For example, the forests in southern zone. Further, Shaw and Westbrook (submitted) provided
Argentina, where beaver are considered an invasive species, are evidence of nested hyporheic systems occurring in association
unable to adapt to support long-term foraging (Anderson with the presence of multiple beaver dams of varying shapes
et al., 2006; Guillermo et al., 2006). By contrast, many eco- and sizes.
systems in North America have evolved to positively respond Water fluxes along beaver-driven hyporheic flow pathways
to beaver foraging. For example, beaver cutting has been are not commonly quantified, even though they influence the
shown to stimulate vigorous sprouting below the cut or pro- flux of dissolved solutes across the stream–groundwater
duce higher regrowth the following year (Kindschy, 1985; interface, and ultimately downstream water quality. Recent
Baker, 2003). work along a stream draining a peat-capped Canadian Rocky
Quasiquantitative links have been made between beaver- Mountain valley peatland shows that despite strong hydraulic
enhanced evapotranspiration rates and changes to different gradients for flow, it seems that flow was not actually looping
water balance terms. Reid (1952) reported that increased underneath beaver dams as very little water actually moved
evaporation from beaver ponds in the Adirondacks reduced along these flowpaths (Janzen and Westbrook, submitted).
streamflow volumes. Sharps (1996) used a modeling ap- Instead, the researchers suggested there may be vertical re-
proach to demonstrate how high groundwater levels near the charge of underlying aquifers upstream of beaver dams. By
pond and stream, which intersected the plant rooting zone, contrast, Lautz and Siegel (2005) used a modeling approach
resulted in higher evapotranspiration rates compared to ri- to show that small debris dams (constructed to mimic beaver
parian areas not adjacent to beaver ponds. She also showed dams) caused the largest hyporheic flux rates along a stream
that evapotranspiration demands were greater than water flux reach in Wyoming. Due to varying findings, it would be useful
rates from the channel into the riparian area such that the if more researchers reported actual flow volumes rather than
beaver pond area shrank over the summer. Others have also simply describing flow pathways. This would provide greater
reported that greater water use by the riparian forest fringing insight into the relative importance of beaver dam-driven
beaver ponds, due to greater rooting zone moisture avail- hyporheic flows as a proportion of total stream discharge.
ability, may be responsible for enhanced evapotranspiration Ponding of stream water behind beaver dams also increases
rates (Peterjohn and Correll, 1986; Correll et al., 2000). Fur- the pressure differences, that is, hydraulic gradient, between
ther, increasing groundwater availability in the riparian area the stream and the connected riparian water table. This en-
adjacent to beaver ponds (Cole et al., 2008) enhances vege- hances lateral infiltration of stream water into the near-stream
tation growth (Olson and Hubert, 1994), which could result aquifer (Lowry and Beschta, 1994; Sharps, 1996; Shaw and
in enhanced evapotranspiration. Westbrook, submitted) such that beaver dam-driven hypor-
heic flowpaths become comparable to those at severe me-
anders along gaining stream reaches and result in water loss to
the groundwater system along static to losing stream reaches
12.20.4.4 Surface Water–Groundwater Exchange
(Lautz et al., 2006; Hill and Duval, 2009). Beaver dam-created
Hyporheic flow is the exchange of groundwater and surface flowpaths tend to be stable throughout baseflow and storm-
water across the stream bed and banks (Triska et al., 1989). flow conditions (Shaw and Westbrook, submitted) as long as
Many researchers have investigated hyporheic flows because of dam structural integrity is not compromised by stream dis-
their influence on the biogeochemical and ecological func- charge. Enhanced bank infiltration can increase groundwater
tioning of stream ecosystems (e.g., Triska et al., 1993; Findlay, levels in the riparian area near the pond (Wilde et al., 1950;
1995; Jones and Holmes, 1996). The zone across which Johnston and Naiman, 1987; Zav’yalov and Zueva, 1998;
stream and groundwaters mix is temporally and spatially dy- Czerepko et al., 2009). If the pond has a large enough per-
namic (Harvey and Wagner, 2000), and is shaped by local and imeter, enhanced bank infiltration can lead to substantially
regional geomorphology and hydrology (Jones et al., 2008). A enhanced groundwater storage in stream valleys (Woo and
few recent studies have focused on how beaver dams and Waddington, 1990; Westbrook et al., 2006). Conversely, re-
other debris dams restructure hyporheic flow directions. moval of beavers from stream systems has been shown to
Beaver create step-pool stream profiles and decrease stream dewater riparian areas (Pollock et al., 2007; Wolf et al., 2007;
water velocity through building dams (Naiman et al., 1986), Bilyeu et al., 2008).
which are two physical stream characteristics known to en- The orientation of a dam in relation to the principal dir-
hance hyporheic exchange (Wondzell and Swanson, 1996; ection of groundwater flow and valley confinement influences
Gooseff et al., 2006). Ponding increases the water pressure whether beaver dams affect patterns of groundwater flow. In
behind a beaver dam, which can increase the hydrologic situations where dams are built parallel to the principal dir-
interaction of the stream with its bed. Similar to man-made ection of groundwater flow, enhanced infiltration into river
dams, surface water is driven into the stream bed behind the banks can cause a local change in flow patterns, where infil-
beaver dam, flows as hyporheic water under the dam, and trated stream water flows underground toward the valley
reemerges as surface water downstream (White, 1990; Triska center and ultimately loops around the dam (Lowry and
et al., 2000; Stofleth et al., 2008). Where multiple dams are Beschta, 1994; Lautz et al., 2006; Westbrook et al., 2006). The
present along a stream reach, the impact of beaver on scale of this subsurface loop can vary from tens to several
groundwater–surface water interactions is magnified. For ex- hundreds of meters, depending on dam size, stream geometry,
ample, Jin et al. (2009) used a series of tracer experiments to and valley constraint. By contrast, dams built across a channel
demonstrate how a greater number of beaver dams enhanced with a geometry that is parallel to the direction of ground-
the potential for transient storage of water in the hyporheic water flow will have little effect on flow patterns. However,
300 Beaver Hydrology and Geomorphology

dams in such orientations have been shown to enhance the

magnitude of flow because the dam increases the hydraulic
gradient between the stream and riparian area (Westbrook
et al., 2006). Valley gradient is also important in determining
whether a beaver dam will affect groundwater flow patterns.
For example, Woo and Waddington (1990) showed no effect
of beaver dams on subsurface flow patterns in a flat, subarctic
landscape.
Most work relating to examining the role of beaver in
shaping groundwater–surface water interactions has been dir-
ected toward examining dams built along streams. However,
water-level differences and, thus, pressure differences exist
above and below beaver dams located in oxbows, abandoned
channels, and peatlands (Jeglum, 1975; Westbrook and Cooper,
unpublished data). We know very little about how dams in
these types of landscape settings affect groundwater–surface
water interactions.

12.20.5 Beaver Geomorphology – Landforms and

Sedimentation

Beaver ponds themselves are the primary landform created by

beaver damming (Butler and Malanson, 1994). Their size and
number on the landscape shift with changing beaver popu- Figure 5 Sediment core extracted from a beaver pond in Glacier
lations and environmental conditions. Few studies exist, National Park, Montana, USA.
however, that provide information on the longevity of indi-
vidual beaver ponds. Several studies have illustrated beaver-
pond expansion or contraction over broad areas across The efficiency of beaver dams and their attendant ponds as
several decades using aerial photographs or satellite imagery sediment traps has been well documented in the literature.
(Remillard et al., 1987; Johnston and Naiman, 1990b; Naiman et al. (1988) cited examples where small dams con-
Townsend and Butler, 1996; Green and Westbrook, 2009), but tain not more than 4–18 m3 of wood retained 2000–6500 m3
those papers did not provide age-specific information for of sediment. Meentemeyer and Butler (1999) described small
individual beaver ponds. The most specific description of the beaver ponds in the Rocky Mountains that retained from
longevity of a single beaver pond in a mountain environment about 10 to more than 250 m3 of sediment. In the same
comes from Neff (1959), who used historical published general study area (Glacier National Park, Montana, USA),
descriptions, aerial photographs, and fieldwork to document a Butler and Malanson (1995) studied eight beaver ponds that
70-year history of a beaver pond in the Rockies near Long’s ranged from 10 to almost 5100 m3 of sediment per pond.
Peak in Rocky Mountain National Park, Colorado, USA. Butler Naiman et al. (1986) examined the issue of the totality of
and Malanson (1995) described two beaver ponds in Glacier sediment entrapped in beaver ponds at the landscape scale in
National Park, Montana, USA, which underwent creation via the Matamek River watershed in Quebec, Canada. They ex-
damming through draining as a result of dam breaching over a trapolated that beaver were directly responsible for the re-
5–7-year period, and subsequently (Butler and Malanson, tention of 3.2 106 m3 of sediment in small-order streams in
2005) described a small beaver pond that was created in 1991 their 673 km2 study area, enough to cover the bottom of every
and completely infilled by 2002. stream in the drainage with an additional 42 cm of sediment.
Beaver damming and subsequent ponding of streams in- Butler and Malanson (2005) took published data on
duce significant changes to the geomorphology of the dam- amount of sediment volume per beaver pond, and created a
med stream channels. In addition to the hydrological conservative range of 200–500 m3 of sediment per pond. They
implications discussed in Section 12.20.3, damming of then multiplied those minimum and maximum numbers
streams by beaver produces a stair-step channel gradient that times a minimum and maximum estimate of the number of
would otherwise not exist (Naiman et al., 1988; Gurnell, beaver ponds in pre-Contact North America, as well as times a
1998; Meentemeyer and Butler, 1999; Bigler et al., 2001). This minimum and maximum estimate of present-day beaver
altered gradient, in turn, reduces stream velocity (Naiman ponds in North America. Their calculated values for present-
et al., 1988; Gurnell, 1998; Meentemeyer and Butler, 1999; day North America were 750 million to 3.85 billion m3 of
Collen and Gibson, 2001; Hering et al., 2001) and induces sediment in storage in beaver ponds. For pre-Contact North
retention of substantial amounts of sediments in beaver ponds America, the values reached staggering levels of 3–50 billion
(Naiman et al., 1986, 1988; Butler and Malanson, 1995, 2005; m3 of sediment for a minimum range, and a maximum range
Gurnell, 1998; Meentemeyer and Butler, 1999; Bigler et al., of 7.5–125 billion m3 of sediment. They concluded their ex-
2001; Pollock et al., 2007; Persico and Meyer, 2009; B"˛edzki ercise by speculating on what values might be achieved if the
et al., 2010; Burchsted et al., 2010) (Figure 5). high sedimentation rates of 6500 m3 of sediment per beaver
 Beaver Hydrology and Geomorphology 301

pond from Naiman et al. (1986) were used in such an undermined trees and caused bank slumps and bank failures,
extrapolation. introducing additional, unquantified amounts of sediment
In addition to dam building, beaver are also excellent into the stream. Similar undermined trees along beaver pond
diggers. As noted above, beaver excavate canals along the edges introduce additional, unquantified amounts of sedi-
flanks of beaver ponds (Figure 6). Beaver canals are excavated ment into beaver ponds (Figure 8).
primarily to transport logs from the surrounding forest to the The presence of beaver bank burrows along streams and
beaver pond (Butler and Malanson, 1994), but canals may rivers extends the geomorphic influences of beaver beyond the
also be constructed to divert water in order to maintain suf- beaver pond environment. Beaver affect the very geographic
ficient pond depth. Cowell (1984) described a canal excavated location of streams in some instances by diverting the dir-
in a subarctic peatland with the purpose of maintaining water ection of flow with dams (see Section 12.20.3). Beaver can
level. Instead, however, the additional water that was supplied also profoundly affect riverine system floodplains and ad-
via the canal induced increased karstification in the underlying jacent fluvial terraces, as first suggested but not documented
limestone terrain that led to pond drainage underground. by Ruedemann and Schoonmaker (1938). John and Klein
Beaver canals range in size from under a meter to well over 100 (2004) illustrated that the dams of European beaver can divert
m in length, and are about 35 cm to 41 m in width (Butler, water onto floodplains, resulting in a multichanneled, anas-
1991). tomosing, drainage network with widespread avulsion de-
Beaver also burrow into the banks of ponds, as well as in posits attributable to beaver-dam placement. John and Klein
riverbanks where damming is unable to be successful (Butler (2004) further suggested that, as a result of these potentially
and Malanson, 1994) (Figure 7). Bank burrowing, as well as widespread efforts of beavers in redirecting sediments onto
canal excavation, introduces sediment into beaver ponds, but floodplains, beaver-damming should be considered as a for-
the amount of such beaver-induced sedimentation is un- cing factor of Quaternary floodplain development along
quantified. Meentemeyer et al. (1998) provided the only lower-order rivers in central Europe.
known quantitative values associated with stream-bank bur- More recently, Westbrook et al. (2011) described an in-
rowing by beaver, from a stream in the Piedmont of North channel beaver dam on a fourth-order segment of the head-
Carolina, USA. Along an 817-m stretch of stream, almost waters of the Colorado River in Rocky Mountain National
22 m3 of sediment from over 200 bank-burrow landforms was Park, Colorado, USA. A 1.7-m-high dam triggered overbank
estimated to have entered the stream system over a 5-year flooding that deposited sediment in a spatially heterogeneous
period (Meentemeyer et al., 1998). The bank burrows also pattern on the floodplain and adjacent terrace; soil was
also scoured in other areas by the flooding. Subsequent dam
breaching during high water led to rapid dewatering of
the terrace and flooding, protecting the deposited sediment
from remobilization by future floods. The nutrient-rich sedi-
ment and scoured areas were both rapidly colonized by plants,
forming a beaver meadow. Westbrook et al.’s (2011) results
illustrate that in riverine environments where downstream
flooding on floodplains and terraces is the main hydrologic
effect of beavers (sensu Westbrook et al., 2006), sediment
deposition and scouring on terraces is the main geomorphic
signature of beaver meadow formation, and may help drive
the formation of alluvial valleys and their complex vegetation

Figure 6 The person stands with one foot on the floor of a beaver
pond drained by catastrophic flooding, and the other foot in a
beaver-excavated canal on the floor of the drained pond. Such canals Figure 7 Beaver bank burrow, right of center, in drained beaver
are used to carry pieces of wood to food caches and burrows in the pond in Glacier National Park, Montana, USA. Note also the quantity
banks of the pond; one such burrow is beneath the exposed roots in of entrapped sediment on the floor of the pond. This is the same
the edge of the drained pond behind and to the right of the person. pond as illustrated in Figure 6.
302 Beaver Hydrology and Geomorphology

(a) (b)

Figure 8 Tree-tip root mass on bank of beaver pond in Glacier National Park, Montana, USA, in 1994 (a) and 2002 (b). Saturated conditions
along the pond margins caused bank failure and the tree complex to fail, probably as a result of strong downvalley winds.

patterns as first postulated by Ruedemann and Schoonmaker underway to eradicate beaver from the austral archipelago of
(1938). Chile and Argentina where they are an invasive species. What
will be beaver’s legacy on hydrological and ecological pro-
cesses of the region’s rivers and floodplains? Will the legacy be
different from the one in North America? Needed are field-
12.20.6 Conclusions and Future Challenges based experimental and observational data that can be used in
new ecohydrological models capable of explicitly dealing with
The first step to understanding how beaver influence hydro- the stochastic nature of beaver activities in time and space to
geomorphic processes is documenting population distri- answer these types of research questions.
butions. What were the historic distribution, population, and Last, most of the hydrologic and geomorphic beaver lit-
hydrologic and ecological influence of beavers on steams and erature is focused on stream courses. Knowledge of effects of
other water bodies and wetlands? Where have beaver not yet beaver activities on the form and function of the expansive
recolonized? Many researchers in Europe are currently focused peatlands that span northern latitudes is lacking. Answers to
on these questions as Eurasian beaver populations grow, but important questions, such as whether or not beaver activities
such work lost momentum in North America in the early can create the hydrologic conditions suitable for peat accu-
1990s. mulation?, could prove useful for restoring peatlands across
The literature is rich with descriptions of how beaver in- areas heavily exploited for subsurface natural resources such as
fluence specific hydrogeomorphic processes such as moder- the oil sands region in Canada. Collection of coupled hydro-
ating stream flows and increasing surface water and logic and ecological data in places where beaver are currently
groundwater storage. Lacking, however, are clear linkages be- being reintroduced in climates suitable for peatland formation,
tween these effects and ecosystem functioning, especially at such as Scotland, could be used to address a variety of questions.
larger spatial and temporal scales. For example, we know little
about the overall changes to rivers throughout North America
that were caused by the regional removal of beavers. What References
have been the long-term effects on stream, watershed and
landscape scale hydrological processes, floodplain landforms, Anderson, C.B., Griffith, C.R., Rosemond, A.D., Rozzi, R., Dollenz, O., 2006. The
and vegetation from beaver extirpation? Have rivers and their effects of invasive North American beavers on riparian plant communities in
floodplains recovered? Are they permanently changed? And Cape Horn Chile: do exotic beavers engineer differently in sub-Antarctic
ecosystems? Biological Conservation 128, 467–474.
were many of the alluvial valleys of North America and central Andersen, D.C., Shafroth, P.B., 2010. Beaver dams, hydrological thresholds, and
Europe heavily influenced by riverine overbank flooding, controlled floods as a management tool in a desert riverine ecosystem, Bill
scouring, and sedimentation? Further, efforts are currently Williams River, Arizona. Eoohydrology 3(3), 325–338. doi:10.1002/eco.113.
 Beaver Hydrology and Geomorphology 303

Apple, L.L., Smith, B.H., Dunder, J.D., Baker, B.W., 1985. The use of beavers for Cunningham, J.M., Calhoun, A.J.K., Glanz, W.E., 2006. Patterns of beaver
riparian/aquatic habitat restoration of cold desert, gully-cut stream systems in colonization and wetland change in Acadia National Park. Northeastern Naturalist
southwestern Wyoming. In: Pilleri, G. (Ed.), Investigations on Beavers. Brain 13(4), 583–596.
Anatomy Institute. Brain Anatomy Institute, Berne, pp. 123–130. Czerepko, J., Wróbel, M., Boczoń, A., Sokołowski, K., 2009. The response of ash-
Baker, B.W., 2003. Beaver (Castor canadensis) in heavily browsed environments. alder swamp forest to increasing stream water level caused by damming by the
Lutra 46(2), 173–181. European beaver (Castor fiber L.). Journal of Water and Land Development 13,
Baker, B.W., Ducharme, H.C., Mitchell, D.C.S., Stanley, T.R., Peinetti, H.R., 2005. 249–262.
Interaction of beaver and elk herbivory reduces standing crop of willow. Devito, K.J., Dillon, P.J., 1993. The importance of runoff and winter anoxia to the P
Ecological Applications 15(1), 110–118. and N dynamics of a beaver pond. Canadian Journal of Fisheries and Aquatic
Baker, B.W., Hill, E.P., 2003. Beaver (Castor canadensis). In: Feldhamer, G.A., Sciences 50, 2222–2234.
Thompson, B.C., Chapman, J.A. (Eds.), Wild Mammals of North America, Dinsmore, K.J., Billett, M.F., Moore, T.R., 2009. Transfer of carbon dioxide and
Second ed. John Hopkins University Press, Baltimore, MD, pp. 288–310. methane through the soil–water–atmosphere system at Mer Bleue peatland
Barnes, W.J., Dibble, E., 1986. The effects of beaver in riverbank forest succession. Canada. Hydrological Processes 23(2), 330–341.
Canadian Journal of Botany 66, 40–44. Di Tomasco, J.M., 1998. Impact, biology, and ecology of salt cedar (tamarisk spp.)
Bigler, W., Butler, D.R., Dixon, R.W., 2001. Beaver-pond sequence morphology and in the southwestern United States. Weed Technology 12, 326–336.
sedimentation in northwestern Montana. Physical Geography 22, 531–540. Dugmore, A.R., 1914. The Romance of the Beaver: Being the History of the Beaver
Bilyeu, D.M., Cooper, D.J., Hobbs, N.T., 2008. Water tables constrain height in the Western Hemisphere. Lippincott, Philadelphia, PA, 330 pp.
recovery of willow on Yellowstone’s northern range. Ecological Applications Duncan, S.L., 1984. Leaving it to beaver. Environment 26, 41–45.
18(1), 80–92. Face-Collins, M.S., Johnston, C.A., 2007. Rangeland drought mitigation by beaver
Bisson, P.A., Buffington, J.M., Montgomery, D.R., 2006. Valley segments, stream (Castor canadensis) impounded water. Proceedings of the South Dakota
reaches, and channel units. In: Hauer, F.R., Lamberti, G.A. (Eds.), Methods in Academy of Science 86, 228.
Stream Ecology. Elsevier, San Diego, CA, pp. 23–49. Flynn, N.J., 2006. Spatial associations of beaver ponds and culverts in boreal
Bł˛edzki, L.A., Bubier, J.L., Moulton, L.A., Kyker-Snowman, T.D., 2010. Downstream headwater streams. MSc Thesis, University of Alberta, Edmonton, Canada,
effects of beaver ponds on the water quality of New England first- and second- 108 pp.
order streams. Ecohydrology, http://dx.doi.org/10.1002/eco.163. Findlay, S., 1995. Importance of surface-subsurface exchange in stream ecosystems:
Bonner, J.L., Anderson, J.T., Rentch, J.S., Grafton, W.N., 2009. Vegetation the hyporheic zone. Limnology and Oceanography 40, 159–164.
composition and community structure associated with beaver ponds in Gooseff, M.N., Anderson, J.K., Wondzell, S.M., LaNier, J., Haggerty, R., 2006. A
Canaan valley, West Virginia, USA. Wetlands Ecology and Management 17, modelling study of hyporheic exchange pattern and the sequence, size, and
543–554. spacing of stream bedforms in mountain stream networks, Oregon, USA.
Breck, S.W., Wilson, K.R., Andersen, D.C., 2003. Beaver herbivory and its effects on Hydrological Processes 20, 2443–2457. http://dx.doi.org/10.1002/hyp.6349.
cottonwood trees: influence of flooding along matched regulated and Grasse, J.E., 1951. Beaver ecology and management in the Rockies. Journal of
unregulated rivers. River Research and Applications 19(1), 43–58. Forestry 49, 3–6.
van Breemen, N., 1995. How Sphagnum bogs down other plants. Trends in Ecology Green, K., Westbrook, C.J., 2009. Changes in sediment yield, channel hydraulics
and Evolution 10(7), 270–275. and riparian area structure following loss of beaver dams. B.C. Journal of
Brown, J.H., 1995. Organisms as engineers: a useful framework for studying effects Ecosystems and Management 10(1), 68–78.
on ecosystems? Trends in Ecology and Evolution 10(2), 51–52. Gurnell, A.M., 1998. The hydrogeomorphological effects of beaver dam-building
Bryce, G., 1904. The remarkable history of the Hudson Bay Company. Reprint activity. Progress in Physical Geography 22(2), 167–189.
1968. Burt Franklin, New York, NY. Gurney, W.S.C., Lawton, J.H., 1996. The population dynamics of ecosystem
Burchsted, D., Daniels, M., Thorson, R., Vokoun, J., 2010. The river discontinuum: engineers. Oikos 76, 273–283.
applying beaver modifications to baseline conditions for restoration of forested Halley, D.J., Rosell, F., 2002. The beaver’s reconquest of Eurasia: status, population
headwaters. BioScience 60, 908–922. development and management of a conservation success. Mammal Review
Burns, D.A., McDonnell, J.J., 1998. Effects of a beaver pond on runoff processes: 32(3), 153–178.
comparison of two headwater catchments. Journal of Hydrology 205, 248–264. Halley, D.J., Rosell, F., 2003. Population and distribution of European beavers
Butler, D.R., 1989. The failure of beaver dams and resulting outburst flooding: a (Castor fiber). Lutra 46, 91–101.
geomorphic hazard of the southeastern Piedmont. Geographical Bulletin 31, Härkönen, S., 1999. Forest damage caused by the Canadian beaver (Castor
29–38. canadensis) in south Savo, Finland. Silva Fennica 33(4), 247–259.
Butler, D.R., 1991. Beavers as agents of biogeomorphic change: a review and Harvey, J.W., Wagner, B.J., 2000. Quantifying hydrologic interactions between
suggestions for teaching exercises. Journal of Geography 90, 210–217. streams and their subsurface hyporheic zones. In: Jones, J.B., Mulholland, P.J.
Butler, D.R., 1995. Zoogeomorphology: Animals as Geomorphic Agents. Cambridge (Eds.), Streams and Ground Waters. Academic Press, San Diego, CA, pp. 3–44.
University Press, New York, NY, 244 pp. Heinselman, M.L., 1965. String bogs and other patterned organic terrain near
Butler, D.R., Malanson, G.P., 1994. Beaver landforms. Canadian Geographer 38, Seney, upper Michigan. Ecology 46(1/2), 185–188.
76–79. Hering, D., Gerhard, M., Kiel, E., Ehlert, T., Pottgiesser, T., 2001. Review study on
Butler, D.R., Malanson, G.P., 1995. Sedimentation rates and patterns in beaver near-natural conditions of central European mountain streams, with particular
ponds in a mountain environment. Geomorphology 13, 255–269. reference to debris dams and beaver dams: results of the ‘‘REG Meeting’’ 2000.
Butler, D.R., Malanson, G.P., 2005. The geomorphic influences of beaver dams and Limnologica 31, 81–92.
failures of beaver dams. Geomorphology 71(1–2), 48–60. Hill, A.R., Duval, T.P., 2009. Beaver dams along an agricultural stream in southern
Cole, C.A., Cirmo, C.P., Wardrop, D.H., Brooks, R.P., Peterson-Smith, J., 2008. Ontario, Canada: their impact on riparian zone hydrology and nitrogen
Transferability of an HGM wetland classification scheme to a longitudinal chemistry. Hydrological Processes 23, 1324–1336. http://dx.doi.org/10.1002/
gradient of the central Appalachian Mountains: initial hydrological results. hyp.7249.
Wetlands 28(2), 439–449. Hillman, G.R., 1998. Flood wave attenuation by a wetland following a beaver-dam
Collen, P., Gibson, R.J., 2001. The general ecology of beavers (Castor spp.), as failure on a second order boreal stream. Wetlands 18, 21–34.
related to their influence on stream ecosystems and riparian habitats, and the Hood, G.A., Bayley, S.E., 2008. Beaver (Castor canadensis) mitigate the effects of
subsequent effects on fish – a review. Reviews in Fish Biology and Fisheries climate on the area of open water in boreal wetlands in western Canada.
10, 439–461. Biological Conservation 141, 556–567.
Correll, D.L., Jordan, T.E., Weller, D.E., 2000. Beaver pond biogeochemical effects Janzen, K.F., Westbrook, C.J., submitted. Hyporheic flows along a beaver damned
in the Maryland Coastal Plain. Biogeochemistry 49, 239–317. stream draining a peat capped mountain valley. Canadian Water Resources
Cowell, D.W., 1984. The Canadian beaver, Castor canadensis, as a geomorphic Journal.
agent in karst terrain. The Canadian Field-Naturalist 98, 227–230. Jeglum, J.K., 1975. Vegetation-habitat changes caused by damming a peatland
Craighead, Sr. F.C., 1968. The role of the alligator in shaping plant communities and drainageway in northern Ontario. Canadian Field-Naturalist 89, 400–412.
maintaining wildlife in the Southern Everglades. Florida Naturalist 41, 69–74. Jenkins, S.H., Busher, P.E., 1979. Castor canadensis. Mammalian Species 120, 1–8.
Crawford, H.S., Hooper, R.G., Harlow, R.F., 1976. Woody plants selected by beavers Jin, L., Siegel, D.I., Lautz, L.K., Otz, M.H., 2009. Transient storage and downstream
in the Appalachian Ridge and Valley Province. USDA Forest Service Research solute transport in nested stream reaches affected by beaver dams. Hydrological
Paper NE-346, 10 pp. Processes 23, 2438–2449.
304 Beaver Hydrology and Geomorphology

John, S., Klein, A., 2004. Hydrogeomorphic effects of beaver dams on floodplain Nolet, B.A., 1997. Management of the Beaver (Castor fiber): Towards Restoration of
morphology: avulsion processes and sediment fluxes in upland valley floors Its Former Distribution And Ecological Function in Europe. Council of Europe
(Spessart, Germany). Quaternaire 15, 219–231. Publishing, Strasbourg, 37 pp.
Johnston, C.A., Naiman, R.J., 1987. Boundary dynamics at the aquatic–terrestrial Nolet, B.A., Rosell, F., 1998. Come back of the beaver Castor fiber: an overview of
interface: the influence of beaver and geomorphology. Landscape Ecology 1, old and new conservation problems. Biological Conservation 83, 165–173.
47–57. Olson, R., Hubert, W.A., 1994. Beaver: Water Resources And Riparian Habitat
Johnston, C.A., Naiman, R.J., 1990a. Browse selection by beaver: effects on riparian Manager. University of Wyoming, Laramie, 48 pp.
forest composition. Canadian Journal of Forest Resources 20, 1036–1043. Parker, M., 1986. Beaver, water quality and riparian systems. Proceedings of the
Johnston, C.A., Naiman, R.J., 1990b. The use of a geographic information system to Wyoming Water and Streamside Zone Conference. Wyoming Water Research
analyze long-term landscape alteration by beaver. Landscape Ecology 4(1), 5–19. Centre, University of Wyoming, Laramie 1, 88–94.
Jones, J.B., Holmes, R.M., 1996. Surface–subsurface interactions in stream Persico, L., Meyer, G., 2009. Holocene beaver damming, fluvial geomorphology,
ecosystems. Trends in Ecology and Evolution 11, 239–242. and climate in Yellowstone National Park, Wyoming. Quaternary Research 71,
Jones, C.G., Lawton, J.H., Shachak, M., 1994. Organisms as ecosystem engineers. 340–353. http://dx.doi.org/10.1016/j.yqres.2008.09.007.
Oikos 69, 373–386. Peterjohn, W.T., Correll, D.L., 1986. The effect of riparian forest on the volume and
Jones, C.G., Lawton, J.H., Shachak, M., 1997. Positive and negative effects of chemical composition of baseflow in an agricultural watershed. In: Correll, D.L.
organisms as physical ecosystem engineers. Ecology 78(7), 1946–1957. (Ed.), Watershed Research Perspectives. Smithsonian Press, Washington, DC,
Jones, K.L., Poole, G.C., Woessner, W.W., et al., 2008. Geomorphology, hydrology, pp. 244–262.
and aquatic vegetation drive seasonal hyporheic flow patterns across a gravel- Pollock, M.M., Beechie, T.J., Jordan, C.E., 2007. Geomorphic changes upstream of
dominated floodplain. Hydrological Processes 22, 2105–2113. http://dx.doi.org/ beaver dams in Bridge Creek, an incised stream channel in the interior
10.1002/hyp.6810. Columbia River basin, eastern Oregon. Earth Surface Processes and Landforms
Kindschy, R.R., 1985. Response of red willow to beaver use in southeastern 32(8), 1174–1185.
Oregon. Journal of Wildlife Management 49, 26–28. Racine, C.H., Walters, J.C., 1994. Groundwater-discharge fens in the tanana
Lautz, L.K., Siegel, D.I., 2005. Modeling surface and ground water mixing in the lowlands, Interior Alaska, U.S.A. Arctic and Alpine Research 26(4), 418–426.
hyporheic zone using MODFLOW and MT3D. Advances in Water Resources 29, Ray, H.L, Ray, A.M., Rebertus, A.J., 2004. Rapid establishment of fish in isolated
1618–1633. peatland beaver ponds. Wetlands 24(2), 399–405.
Lautz, L.K., Siegel, D.I., Bauer, R.L., 2006. Impact of debris dams on hyporheic Ray, H.L., Rebertus, A.J., Ray, A.M., 2001. Macrophyte succession in Minnesota
interaction along a semi-arid stream. Hydrological Processes 20, 183–196. http: beaver ponds. Canadian Journal of Botany 79, 487–499.
//dx.doi.org/10.1002/hyp.5910. Rebertus, A.J., 1986. Bogs as beaver habitat in north-central Minnesota. American
Lesica, P., Miles, S., 2004. Beavers indirectly enhance the growth of Russian olive Midland Naturalist 116(2), 240–245.
and tamarisk along eastern Montana river. Western North American Naturalist Reid, K.A., 1952. Effects of beaver on trout waters. Maryland Conservationist 29,
64(1), 93–100. 21–23.
Lizarralde, M.S., 1993. Current status of the introduced beaver (Castor canadensis) Remillard, M.M., Gruendling, G.K., Bogucki, D.J., 1987. Disturbance by beaver
population in Tierra del Fuego, Argentina. Ambio 22(6), 351–358. (Castor canadensis Kuhl) and increased landscape heterogeneity. In: Turner,
Lowry, M.M., Beschta, R.L., 1994. Effects of a Beaver Pond on Groundwater M.G. (Ed.), Landscape Heterogeneity and Disturbance. Springer, New York, NY,
Elevation and Temperatures in a Recovering Stream System. American Water pp. 103–123.
Resources Association Special Publication, Bethesda, MD, pp. 503–513. Richard, P.B., 1967. Le determinisme de la construction des barrages chez le castor
Marston, R.A., 1994. River entrenchment in small mountain valleys of the western du Rhone. La Terre et la Vie 4, 339–470.
USA: influence of beaver, grazing and clearcut logging. Revue de Géographie de Rosell, F., Bozsér, O., Collen, P., Parker, H., 2005. Ecological impact of beavers
Lyon 69, 11–15. Castor fiber and Castor canadensis and their ability to modify ecosystems.
Martell, K.A., Foote, A.L., Cumming, S.G., 2006. Riparian disturbance due to Mammal Review 35(3–4), 248–276.
beavers (Castor canadensis) in Alberta’s boreal mixedwood forests: Implications Ruedemann, R., Schoonmaker, W.J., 1938. Beaver-dams as geologic agents. Science
for forest management. Ecoscience 13(2), 164–171. 88, 523–525.
Mazumder, A., Taylor, W.D., McQueen, D.J., Lean, D.R., 1990. Effects of fish and Running, S.W., Nemani, R.R., Peterson, D.L., Band, L.E., Potts, D.F., Pierce, L.L.,
plankton on lake temperature and mixing depths. Science 247, 312–315. Spanner, M.A., 1989. Mapping regional forest evapotranspiration and
Meentemeyer, R.K., Butler, D.R., 1999. Hydrogeomorphic effects of beaver dams in photosynthesis by coupling satellite data with ecosystem simulation. Ecology
Glacier National Park, Montana. Physical Geography 20(5), 436–446. 70(4), 1090–1101.
Meentemeyer, R.K., Vogler, J.B., Butler, D.R., 1998. The geomorphic influences of Rutherford, W.H., 1955. Wildlife and environmental relationships of beavers in
burrowing beavers on streambanks, Bolin Creek, North Carolina. Zeitschrift für Colorado forests. Journal of Forestry 53(11), 803–806.
Geomorphologie 42, 453–468. Seton, J.R., 1929. Lives of Game Animals. Doubleday, Doran. Garden City, New
Mills, E.A., 1913. In Beaver World. Houghton Mifflin Company, New York, NY, 229 pp. York, NY, vol. 2, part 2, pp. 441–949.
Minke, A.G., Westbrook, C.J., van der Kamp, G., 2010. Simplified volume-area- Sharps, D.E., 1996. Spatial and temporal characteristics of groundwater levels
depth method for estimating water storage of prairie potholes. Wetlands 30(3), adjacent to beaver ponds in Oregon. MS Thesis, Oregon State University,
541–551. http://dx.doi.org/10.1007/s13157-010-0044-8. Corvallis.
Minke, A.G., Westbrook, C.J., Pomeroy, J.W., Guo, X., submitted. Estimating water Shaw, E.L., Westbrook, C.J., submitted. Lateral hyporheic exchange of water, nitrogen
storage in prairie wetlands from a LiDAR DEM. Water Resources Research. and DOC along a beaver-dammed Rocky Mountain stream. Ecohydrology.
Mitchell, C.C., Niering, W.A., 1993. Vegetation change in a topogenic bog following Skewes, O., Gonzalez, F., Olave, R., Ávila, A., Vargas, V., Paulsen, P., Konig, H.E.,
beaver flooding. Bulletin of the Torrey Botanical Club 120(2), 136–147. 2006. Abundance and distribution of American beaver, Castor canadensis (Kuhl
Naiman, R.J., Johnston, C.A., Kelley, J.C., 1988. Alteration of North American 1820), in Tierra del Fuego and Navarino Islands. European Journal of Wildlife
streams by beaver. BioScience 38(11), 753–762. Resources 52, 292–296.
Naiman, R.J., Melillo, J.M., Hobbie, J.E., 1986. Ecosystem alteration of boreal Stofleth, J.M., Shield, F.J., Fox, G.A., 2008. Hyporheic and total transient storage in
forest streams by beaver (Castor canadensis). Ecology 67, 1254–1269. small, sand-bed streams. Hydrological Processes 22, 1885–1894.
Naiman, R.A., Rogers, K.H., 1997. Large animals and system-level characteristics in Storr, D., 1970. A Survey of Water Balance Problems, Projects and Progress in
river corridors: implications for river management. BioScience 47(8), 521–529. Alberta, Symposium on World Water Balance, International Association of
Neff, D.J., 1957. Ecological effects of beaver habitat abandonment in the Colorado Scientific Hydrology, Publication No. 93, vol. 2, pp. 314–324.
Rockies. Journal of Wildlife Management 21(1), 80–84. Sturtevant, B.R., 1998. A model of wetland vegetation dynamics in simulated beaver
Neff, D.J., 1959. A seventy-year history of a Colorado beaver colony. Journal of impoundments. Ecological Modelling 112, 195–225.
Mammalogy 40, 381–387. Townsend, J.E., 1953. Beaver ecology in western Montana with special reference to
Newton, R.M., Burns, D.A., Blette, V.L., Driscoll, C.T., 1996. Effect of whole movements. Journal of Mammalogy 34, 459–479.
catchment liming on the episodic acidification of two Adirondack streams. Townsend, P.A., Butler, D.R., 1996. Patterns of landscape use by beaver on the lower
Biogeochemistry 32, 299–322. Roanoke River floodplain North Carolina. Physical Geography 17, 253–269.
Nietvelt, C.N., 2001. Herbivory interactions between beaver (Castor canadensis) and Triska, F.J., Duff, J.H., Avanzino, R.J., 1993. The role of water exchange between a
elk (Cervus elaphus) on willow (Salix spp.) in Banff National Park, Alberta. MSc stream channel and its hyporheic zone in nitrogen cycling at the
Thesis, University of Alberta, Edmonton. terrestrial–aquatic interface. Hydrobiologia 251, 167–184.
 Beaver Hydrology and Geomorphology 305

Triska, F.J., Jackman, A.P., Duff, J.H., Avanzino, R.J., 2000. Subchannel flow White, D.S., 1990. Biological relationships to convective flow patterns within
velocity and dispersion beneath a relict beaver dam, Shingobee River, streambeds. Hydrobiologia 196, 149–158.
Minnesota, USA: implications for nutrient cycling. Verhandlungen Internationale Wilde, S.A., Youngberg, C.T., Hovind, J.H., 1950. Changes in composition of
Vereinigung für Theoretische und Angewandte Limnologie 27, 463–467. ground water, soil fertility, and forest growth produced by the construction and
Triska, F.J., Kennedy, V.C., Avanzino, R.J., Zellweger, G.W., Bencala, K.E., 1989. removal of beaver dams. Journal of Wildlife Management 14, 123–128.
Retention and transport of nutrients in a third-order stream in northwestern Winter, T.C., 1998. Relation of streams, lakes, and wetlands to groundwater flow
California: hyporheic processes. Ecology 70, 1893–1905. systems. Hydrogeology Journal 7, 28–45.
Turetsky, M.R., St. Louis, V., 2006. Disturbance in boreal peatlands. In: Wieder, R.K., Wolf, E.C., Cooper, D.J., Hobbs, N.T., 2007. Beaver, streamflow and elk influence
Vitt, D.H. (Eds.), Boreal Peatland Ecosystems. Springer, Berlin, pp. 359–380. willow establishment and floodplain stability on Yellowstone’s northern range.
Ulevičius, A., 1999. Density and habitats of the beaver (Castor fiber) in Lithuania. Ecological Applications 17, 1572–1587.
Proceedings of the Latvian Academy of Sciences, Section B. 53(2), 101–106. Wondzell, S.M., Swanson, F.J., 1996. Seasonal and storm dynamics of the
Vining, K.C., 2004. Simulation of runoff and wetland storage in the Hamden and hyporheic zone of a 4th-order mountain stream. I: hydrologic processes. Journal
Lonetree Watershed sites within the Red River of the North basin, North Dakota of the North American Benthological Society 15, 3–19.
and Minnesota. U.S. Geological Survey Scientific Investigations Report Woo, M.K., Waddington, J.M., 1990. Effects of beaver dams on sub-Arctic wetland
2004–5168. U.S. Department of the Interior, Washington, DC. hydrology. Arctic 43, 223–230.
Vowinckel, E., Orvig, S., 1973. The heat and water budgets of a beaver pond. Woods, S.W., 2000. Hydrologic effects of the Grand Ditch on streams and wetlands
Atmosphere 11, 166–178. in Rocky Mountain National Park, Colorado. MS Thesis, Colorado State
Walbridge, M.R., 1994. Plant community composition and surface water chemistry University, Fort Collins, 219 pp.
of fen peatlands in West Virginia’s Appalachian plateau. Water, Air and Soil Woods, S.W., 2001. Ecohydrology of subalpine wetlands in the Kawuneeche Valley,
Pollution 77, 247–269. Rocky Mountain National Park, Colorado. Ph.D. dissertation, Colorado State
Webb, R.H., Leake, S.A., 2006. Ground-water surface-water interactions and long- University, Fort Collins, 166 pp.
term change in riverine riparian vegetation in the southwestern United States. Wright, J.P., Jones, C.G., Flecker, A.S., 2002. An ecosystem engineer, the beaver,
Journal of Hydrology 320, 302–323. increases species richness at the landscape scale. Oecologia 132, 96–101.
Westbrook, C.J., Cooper, D.J., Baker, B.W., 2006. Beaver dams and overbank floods Yeager, L.E., Hill, R.R., 1954. Beaver management problems on western public
influence groundwater – surface water interactions of a Rocky Mountain riparian lands. Transactions of the North American Wildlife Conference 19, 462–480.
area. Water Resources Research, 42. http://dx.doi.org/10.1029/2005WR004560. Zav’yalov, N.A., Zueva, S.S., 1998. Influence of beaver dams on soil cover (at the
Westbrook, C.J., Cooper, D.J., Baker, B.W., 2011. Beaver assisted river valley Darwin Reserve). Lesovedenie 5, 38–47.
formation. River Research and Applications 27(2), 247–256. doi:10.1002.rra/1359. Zurowski, W., 1989. Dam building activity of beavers on the mountainous streams.
Whicker, A.D., Detling, J.K., 1988. Ecological consequences of prairie dog Proceedings of the Fifth International Theriological Congress, Rome, 1,
disturbances. BioScience 38, 778–785. 316–317.

Biographical Sketch

Cherie J Westbrook is an associate professor in the Centre for Hydrology in the Department of Geography and
Planning at the University of Saskatchewan, in Saskatoon, Saskatchewan, Canada. Her research interests are in the
fields of ecohydrology of mountain valleys, prairie hydrology and water quality, wetlands and riparian areas, river
system cumulative impact assessment, and beavers as hydrologic agents. She has published more than a dozen
refereed papers in a variety of international journals, as well as several published reports to governmental agencies
for the province of Alberta. She holds the BSc (Honors) and MSc degrees in biological sciences from the Uni-
versity of Alberta, and the PhD from the Department of Forest, Rangeland, and Watershed Stewardship at
Colorado State University.

David J Cooper is a senior research scientist and associate professor in the Department of Forest, Rangeland and
Watershed Stewardship at Colorado State University, in Fort Collins, Colorado, USA, where he has held positions
since 1992. His research interests are in the fields of wetland and riparian ecology, especially mountain wetland
hydrology, peatland ecology, ecosystem monitoring, and restoration ecology. His work has been honored with
several awards from the National Park Service and other branches of the U.S. federal government, as well as by
The Nature Conservancy. He has published more than 70 refereed papers, numerous book chapters, and U.S.
government reports. He holds the B.A. degree in environmental biology from the University of Colorado, and the
PhD degree in biology from the University of Colorado.
306 Beaver Hydrology and Geomorphology

Dr. David R Butler is the Texas State University System Regents’ professor of geography and a University Dis-
tinguished Professor at Texas State University-San Marcos where he has been on faculty since 1997. His research
interests are in the areas of mountain geomorphology, zoogeomorphology, biogeomorphology, and den-
drogeomorphology, focusing his work in the Glacier National Park region of Montana, USA. He has published
over 150 refereed papers in journals and conference volumes, and more than 35 book chapters. He is the author
of Zoogeomorphology – Animals as Geomorphic Agents (Cambridge University Press, 1995 and 2007), and coeditor of
Tree Rings and Natural Hazards (Springer, 2010), The Changing Alpine Treeline – The Example of Glacier National Park,
Montana (Elsevier, 2009), and Mountain Geomorphology – Integrating Earth Systems (Elsevier, 2003). He has received
the G.K. Gilbert Award for Excellence in Geomorphological Research from the Geomorphology Specialty Group
of the Association of American Geographers (AAG), and the Distinguished Career Award and the Outstanding
Recent Accomplishment Award from the AAG’s Mountain Geography Specialty Group. David holds a BA and MSc
in geography from the University of Nebraska at Omaha, and the PhD in geography from the University of
Kansas.
 12.21 Interactions among Hydrogeomorphology, Vegetation, and Nutrient
Biogeochemistry in Floodplain Ecosystems
GB Noe, US Geological Survey, Reston, VA, USA
Published by Elsevier Inc.

12.21.1 Floodplains and Their Essential Interactive Processes 308

12.21.2 The Template of Hydrogeomorphology in Floodplains 308
12.21.2.1 History of Hydrogeomorphic Concepts in Rivers 308
12.21.2.2 Hydrogeomorphic Controls on Floodplain Ecosystems 309
12.21.3 Controls of Vegetation in Floodplains 310
12.21.3.1 Hydrogeomorphic Controls of Vegetation 310
12.21.3.2 Influence of Vegetation on Hydrogeomorphology 311
12.21.3.3 Biogeochemical Controls on Vegetation 311
12.21.3.4 Other Biota 311
12.21.4 Controls of Nutrient Biogeochemistry in Floodplains 312
12.21.4.1 Hydrogeomorphic Controls of Nutrient Biogeochemistry 312
12.21.4.2 Influence of Vegetation on Nutrient Biogeochemistry 314
12.21.4.3 Biogeochemical Controls on Hydrogeomorphology? 314
12.21.5 Case Studies 314
12.21.5.1 Hummock/Hollow Geomorphology in Peatlands and Floodplains 314
12.21.5.2 Coastal Plain Floodplains 315
12.21.5.3 Montane Floodplains 316
12.21.5.4 Desert Streams 316
12.21.6 Conclusions 317
References 317

Glossary Hydrogeomorphology The integrated study of hydrology

Ecogeomorphology The integration of hydrology, fluvial and geomorphology.
geomorphology and ecology to study riverine systems Hydrologic connectivity: The exchange of water between
(Thoms and Parsons 2002). floodplain and river channel.
Floodplain A unit of riparian systems that occur on Riparian It is related to the interface of aquatic and
alluvial surfaces and are sometimes inundated with upland ecosystems.
flooding surface water.

Abstract

Hydrogeomorphic, vegetative, and biogeochemical processes interact in floodplains resulting in great complexity that provides
opportunities to better understand linkages among physical and biological processes in ecosystems. Floodplains and their
associated river systems are structured by four-dimensional gradients of hydrogeomorphology: longitudinal, lateral, vertical,
and temporal components. These four dimensions create dynamic hydrologic and geomorphologic mosaics that have a large
imprint on the vegetation and nutrient biogeochemistry of floodplains. Plant physiology, population dynamics, community
structure, and productivity are all very responsive to floodplain hydrogeomorphology. The strength of this relationship
between vegetation and hydrogeomorphology is evident in the use of vegetation as an indicator of hydrogeomorphic
processes. However, vegetation also influences hydrogeomorphology by modifying hydraulics and sediment entrainment and
deposition that typically stabilize geomorphic patterns. Nitrogen and phosphorus biogeochemistry commonly influence plant
productivity and community composition, although productivity is not limited by nutrient availability in all floodplains.
Conversely, vegetation influences nutrient biogeochemistry through direct uptake and storage as well as production of organic
matter that regulates microbial biogeochemical processes. The biogeochemistries of nitrogen and phosphorus cycling are very
sensitive to spatial and temporal variation in hydrogeomorphology, in particular floodplain wetness and sedimentation. The
least-studied interaction is the direct effect of biogeochemistry on hydrogeomorphology, but the control of nutrient avail-
ability over organic matter decomposition and thus soil permeability and elevation is likely important. Biogeochemistry also

Noe, G.B., 2013. Interactions among hydrogeomorphology, vegetation, and

nutrient biogeochemistry in floodplain ecosystems. In: Shroder, J. (Editor in
Chief), Butler, D.R., Hupp, C.R. (Eds.), Treatise on Geomorphology.
Academic Press, San Diego, CA, vol. 12, Ecogeomorphology, pp. 307–321.

Treatise on Geomorphology, Volume 12 http://dx.doi.org/10.1016/B978-0-12-374739-6.00338-9 307

308 Interactions among Hydrogeomorphology, Vegetation, and Nutrient Biogeochemistry in Floodplain Ecosystems

has the more documented but indirect control of hydrogeomorphology through regulation of plant biomass. In summary, the
defining characteristics of floodplain ecosystems are determined by the many interactions among physical and biological
processes. Conservation and restoration of the valuable ecosystem services that floodplains provide depend on improved
understanding and predictive models of interactive system controls and behavior.

12.21.1 Floodplains and Their Essential Interactive feedbacks define riparian systems (Naiman and Décamps,
Processes 1997) and have been termed ‘functional ecogeomorphology’
The value of floodplains to society highlights the importance of (Fisher et al., 2007). The direct and indirect feedbacks between
understanding the controls on their functions. Frequently cited and among these three broadly aggregated processes are, for
ecosystem services provided by floodplains include flood re- the most part, well documented but make it difficult to study
duction, water-quality improvement, habitat, food production, and to predict pattern and process in floodplains. These pro-
and recreation. Costanza et al. (1997) found that the value of cesses and feedbacks also are at the core of the valuable eco-
the ecosystem services provided by floodplains was among the system services that floodplains provide to society. Successful
highest of any ecosystem on the Earth, providing about 10% of river restoration depends on restoring the geomorphic com-
total global ecosystem services that occur on a mere 0.3% of the plexity of floodplains (Petts et al., 1992) through re-
Earth’s surface. However, most floodplains are impacted by introduction of hydrogeomorphic processes (Beechie et al.,
human modifications to hydrology and geomorphology (Hupp 2010) – including the direct effects as well as the indirect
et al., 2009) and the rest are sensitive to future human impacts interactions of hydrogeomorphic, vegetative, and biogeo-
(Tockner and Stanford, 2002). Ecological restoration of flood- chemical processes. The direct effect of nutrient biogeochem-
plains through the reintroduction of flooding has the potential istry on the hydrogeomorphology of floodplains is perhaps
to increase the function, resiliency, and values of impacted less obvious than the other interactions, but specific examples
floodplains (Sparks, 1995; Galat et al., 1998). Future impacts of and concepts exist to support this feedback. To begin
altered precipitation and sea level due to climate change, as well discussion of these fundamental processes and interactions,
as human water use and land-cover change, on river discharge, the four-dimensional (4D) nature of riverine hydrogeo-
flood characteristics, and salinity make an improved under- morphology will first be described.
standing and predictive ability even more essential for suc-
cessful management of floodplain ecosystems. Effective
management of floodplain ecosystems to conserve, restore, or
12.21.2 The Template of Hydrogeomorphology in
augment their ecosystem services requires thorough under-
Floodplains
standing of the processes that control system dynamics.
Hydrologic and geomorphic (‘hydrogeomorphic’ for
12.21.2.1 History of Hydrogeomorphic Concepts in Rivers
short), vegetative, and nutrient biogeochemical processes
interact to control the dynamics of floodplain ecosystems Hydrogeomorphic concepts in river science have progressed
(Figure 1). These essential interactive processes and their through time to include the four dimensions of riverine
hydrogeomorphology, namely longitudinal, lateral, vertical,
and temporal gradients (Figure 2) that are the fundamental
Floodplain ecosystems controls of ecosystem structure and processes in streams and
rivers (Ward, 1989a; Poole, 2010). Leopold et al. (1964) pro-
vided the foundational conceptual and quantitative synthesis
of the interaction of hydrology and geomorphology in fluvial
Hydro- systems. Subsequent concepts in river hydrogeomorphology
geomorphology identified upland control (Hynes, 1975), then added longi-
tudinal (Vannote et al., 1980), lateral (Junk et al., 1989), ver-
tical (Stanford and Ward, 1993), and temporal dimensionality
? (Amoros et al., 1987). The dendritic network approach has also
been applied to stream landscapes (Fisher, 1997).
The longitudinal model of ecological structure and function
in rivers, the river continuum concept (RCC), predicts con-
tinuous gradients in physical conditions, biotic communities,
and ecological processes from headwaters to the mouth
Vegetation Biogeochemistry (Vannote et al., 1980). By constrast, the flood pulse concept is a
lateral model that predicts that the hydrogeomorphic charac-
teristics of flood pulses control both vegetative and animal
communities and biogeochemical cycles in river–floodplain
Figure 1 The interaction of hydrogeomorphic, vegetative, and systems (Junk et al., 1989). The RCC does not include the
biogeochemical processes in floodplain ecosystems. Of these lateral dimension of rivers that is critically important to eco-
interactions, the direct effects of biogeochemistry on system dynamics particularly in large rivers (Ward, 1989b).
hydrogeomorphology are the least understood. Thus, the longitudinal RCC does not suffice to explain biotic
 Interactions among Hydrogeomorphology, Vegetation, and Nutrient Biogeochemistry in Floodplain Ecosystems 309

Bank full
channel

e
lop
ills ces
H ra
r n Temporal
Te lai
o dp
e flo
tiv

al
Ac

k
luvi

ro c
d
Paraf

Be
Lateral

Streambed
Hy Vertical
Pa

le
oc por
ha
nn
h e ic Fissure
el
tic
Phrea
Longitudinal
Figure 2 Example of an alluvial river–floodplain corridor showing the longitudinal, vertical, and lateral dimensions that change through time to
influence the interactions of hydrogeomorphology, vegetation, and nutrient biogeochemistry. Modified with permission from National Research
Council, 2002. Riparian Areas. National Academy Press, Washington, DC.

assemblages or biogeochemistry of rivers with lateral floodplain with vertical and lateral subsystems (Fisher et al., 1998). The
connections (Sedell et al., 1989). The importance of lateral four dimensions also control river-system biodiversity (Ward,
connectivity is demonstrated by human modifications to rivers, 1998). For any dimension, hierarchy theory and consideration
including dams, flood protection levees, and channelization, of scale have been used to understand river-system dynamics
that minimize lateral connectivity and, as a result, diminish the and show how hydrogeomorphic processes that occur at a
ecosystem services of river–floodplain systems (Ward and larger scale influence ecologic processes at the focal scale
Stanford, 1995; Hupp et al., 2009). (ecogeomorphology; Thoms and Parsons, 2002).
Vertical hyporheic exchange is an important hydrologic
dimension affecting river–floodplain systems. Hyporheic ex-
change can affect nutrient transformations and subsidies in
12.21.2.2 Hydrogeomorphic Controls on Floodplain
rivers and floodplains (Holmes et al., 1994). This groundwater
Ecosystems
exchange can also influence the transport of water, heat, and
other material and structures riverine vegetation and bio- Riparian systems are defined as the interface of aquatic eco-
geochemistry gradients (Stanford and Ward, 1993; Valett et al., systems and uplands (Gregory et al., 1991; National Research
1994; Stanford et al., 2005). In addition, lateral hyporheic Council, 2002; Naiman et al., 2005). Floodplains are a specific
exchange between channel and floodplain is important to unit of riparian systems that occur on alluvial surfaces and are
system hydrology and biogeochemistry (Triska et al., 1993). sometimes inundated with flooding surface water. By defin-
Temporal variation in longitudinal, lateral, and vertical ition, riparian and floodplain systems are functions of their
transport processes defines the fourth dimension in river sys- geomorphic position in landscapes and the temporal and
tems. The timing of hydrologic variation, known as the flow spatial variation in the presence and actions of water (Hupp
regime, is a characteristic that distinguishes among river eco- and Osterkamp, 1996). Consequentially, floodplains are ideal
systems (Poff et al., 1997). Natural temporal variation can systems to study how hydrogeomorphic processes control and
generate large changes in the extent and ecological functions interact with ecological processes (Steiger et al., 2005; Tockner
of rivers (Stanley et al., 1997). Human modifications to the et al., 2010). Not all riparian systems, including floodplains,
flow regime impact river ecosystem dynamics with the re- are wetlands; wetlands are characterized by the presence of
sponse depending on local geomorphic setting (Poff et al., surface water or shallow subsurface water, hydric soils, and
2010). Bornette et al. (2008) added the temporal aspect of vegetation adapted to flooding (Mitsch and Gosselink, 1993).
hydrogeomorphic regimes to a model of fluvial hydrosystems Some floodplain locations may lack these traits and be either
in order to emphasize patch dynamics and biotic diversity wetter (e.g., oxbow lakes) or drier (e.g., floodplains in urban-
following physical disturbances. ized catchments) than a wetland. However, the influence of
Thus, river ecology is controlled by the 4D nature of the water is fundamental to wetland ecosystems; therefore, wet-
hydrogeomorphic characteristics of the system with its inherent lands will be used as an analog for understanding floodplains.
spatial complexity and patchiness (Thorp et al., 2006). For It is very well known that hydrology and geomorphology are
example, the telescoping ecosystem model is a 4D concept that responsible for regulating many ecosystem functions of wet-
was developed to describe temporal variation in longitudinal lands in general (Brinson, 1993) and floodplains in particular
processes, such as nutrient spiraling, due to stream interaction (Hupp, 2000). Specifically, geomorphic setting, water source,
310 Interactions among Hydrogeomorphology, Vegetation, and Nutrient Biogeochemistry in Floodplain Ecosystems

and hydrodynamics classify wetlands into different functional Physiologic and metabolic stresses from flooding and soil
groups (Brinson, 1993) and determine their functions in saturation restrict plant species and communities to specific
watersheds (Brinson, 1988; Whigham et al., 1988). locations along the moisture gradient within the floodplain
Flow regime, sediment size and supply, and flow com- (Wharton et al., 1982). The elevation of the soil relative to
petence (the ability of flow to move sediment of different water level creates strong spatial gradients of soil oxygen de-
sizes) determine the hydrogeomorphology of floodplains pletion in floodplain wetlands (Dwire et al., 2006). Plant
(Church, 2002). Hydrogeomorphology, specifically, the tim- species differentially tolerate the physiologic stresses associ-
ing, duration, and velocity of flooding all influence com- ated with the lack of oxygen, which include anaerobic me-
munity structure, productivity, and nutrient cycling in tabolism and the toxicity of reduced metals (Ernst, 1990). In
floodplains (Conner and Day, 1982). Floodplains, in turn, the short term, acclimation processes include changes in
influence watershed hydrology through transient water storage physiology, anatomy, and morphology in response to the lack
and high evapotranspiration rates (Conner and Day, 1982). of oxygen (McKelvin et al., 1998).
Water origin also determines the physical and chemical attri- Wetland plants also respond to flooding stresses at the
butes of floodplains and influences the structure and function longer timescale of population dynamics through acclimation,
of the ecosystem (Amoros and Bornette, 2002). The area of mortality, or avoidance by seed dispersal and selective ger-
floodplain that is directly influenced by river water is deter- mination cues, all of which lead to species-distribution pat-
mined by system geomorphology and antecedent hydrology terns. Flood energy is an important mechanism for moving
but does not encompass the entire extent of a floodplain plant propagules and influencing plant population dynamics in
(Mertes, 1997). In addition to river water, important sources of riparian landscapes. Flood magnitude and seasonality affect the
flood water at some times and locations in floodplains include dispersal of propagules in floodplains (Gurnell et al., 2004).
precipitation, groundwater, and upland runoff that may also Soil surface elevation relative to flood magnitude determines
limit direct inputs of river water to some floodplain locations. the number, diversity, and floristic composition of propagules
Hydrogeomorphology influences ecosystem dynamics delivered to floodplains (Moggridge and Gurnell, 2009).
through the provision of flood-derived resources, known as Hydrochory, the dispersal of seeds by water, can be a more
flood subsidy (Odum et al., 1979), which is broader than the important vector for seed dispersal than wind, particularly for
flood pulse concept. Open wetland ecosystems, whose geo- delivery of allochthonous propagules (Moggridge and Gurnell,
morphology supports rapid hydrologic exchange with streams 2009). After dispersal to a floodplain location, floodplain
or estuaries, have greater vegetative productivity and faster microtopography influences plant germination and establish-
biogeochemical cycling due to greater nutrient inputs than ment (Blood and Titus, 2010).
closed wetland ecosystems (Hopkinson, 1992). Floodplains Hydrogeomorphology also influences vegetative com-
on large rivers experience predictable flood inputs of water, munity composition (Wharton et al., 1982; Hupp, 2000;
sediment, carbon, and nutrients that, together with floodplain Naiman et al., 2010). Gradients in soil surface elevation rela-
spatial geomorphic configuration, determine many ecological tive to local hydrology determine species composition and
processes (Lewis et al., 2000). Thus, the physical configuration cause turnover in species assemblages across the geomorphic
of river–floodplain systems controls their biogeochemical features of floodplains. Floodplains with intermediate flood
functions and biotic communities. More details on how frequency and greater microtopographic variation have great-
floodplain hydrogeomorphology influences floodplain vege- est plant species richness (Pollock et al., 1998). Spatial patches
tation and biogeochemistry are presented below. with less flow energy and greater discharge of hyporheic
groundwater are also associated with greater plant species
richness within floodplains (Mouw et al., 2009). The relative
roles of plant competition and facilitation as structuring forces
12.21.3 Controls of Vegetation in Floodplains
in plant communities are a function of abiotic stress gradients
(Callaway and Walker, 1997; Bornette et al., 2008). Thus,
12.21.3.1 Hydrogeomorphic Controls of Vegetation
plant competition for resources is more likely to occur at
Many chapters in this volume provide extensive reviews of higher elevations in floodplains where stress from anoxia,
vegetation in floodplain ecosystems. In this chapter, a brief water velocity, and debris impacts occurs less frequently and
review of the interaction between hydrogeomorphology and with less magnitude.
vegetation and the influence of biogeochemistry on vegetation Finally, the hydrogeomorphology of floodplains influences
are presented. their rates of primary production. Through history, flood-
Vegetation is responsive to hydrogeomorphic setting and plains have been renowned for their fertility resulting in large-
processes in floodplains. This relationship is the basis for using scale conversions to agricultural production (Brown, 1997).
vegetation as an indicator of current and historic hydrologic The subsidy of water and nutrients to floodplains results in
processes on the different geomorphic surfaces of floodplains natural plant communities with high net primary productivity
(Brinson, 1993; Gurnell, 1997; Osterkamp and Hupp, 2010). (Conner and Day, 1982), although not always greater than
For example, dendrogeomorphic analyses and interpretation of adjacent upland communities (Megonigal et al., 1997). In
species distribution can identify hydrogeomorphic processes support of the flood pulse concept, Conner and Day (1982)
and estimate rates of geomorphic change in floodplains (Hupp reported higher net primary production in forested flood-
and Bornette, 2003). Hydrogeomorphology influences flood- plains with flowing seasonal flooding than those with stag-
plain vegetation through multiple processes at physiologic, nant flooding. Greater plant growth rates occur in areas of
population, community, and ecosystem levels. floodplains with hyporheic discharge likely due to both
 Interactions among Hydrogeomorphology, Vegetation, and Nutrient Biogeochemistry in Floodplain Ecosystems 311

greater water and nutrient availability (Mouw et al., 2009). nutrient availability. The productivity of wetland vegetation is
Total production (litterfall, wood, and root production) is generally determined by nutrient availability as well as hy-
greatest with wetter conditions along a floodplain moisture drology (Brinson et al., 1981). Nutrient availability on some
gradient due mostly to enhanced wood production (Burke floodplains can be high enough to result in net primary pro-
et al., 1999; Clawson et al., 2001). However, belowground duction that is not limited by nutrients (Spink et al., 1998;
production can also decrease with greater wetness (Day and Wassen et al., 2003). Yet some floodplains still have nutrient
Megonigal, 1993; Clawson et al., 2001), and Megonigal et al. limitation of productivity (Lockaby and Walbridge, 1998).
(1997) documented a decrease in aboveground production Some large wetland and floodplain landscapes with low
with prolonged inundation due to plant stress. allochthonous nutrient and mineral sediment loading, like
limnogenous peatlands, are oligotrophic, and nutrient avail-
12.21.3.2 Influence of Vegetation on Hydrogeomorphology ability highly constrains plant productivity, community com-
position, and diversity (Bridgham et al., 1996; Noe et al.,
Hydrogeomorphology clearly influences vegetation pattern 2001; King et al., 2004).
and process on floodplains, but vegetation also influences The sensitivity of primary production to nutrient avail-
hydrologic and geomorphic processes in these interactive ability is supported by the observation that maximum litter-
systems (Simon et al., 2004). The establishment of vegetation fall production on floodplains occurs at a nitrogen (N) to
on bare surfaces generally stabilizes geomorphic surfaces and phosphorus (P) mass ratio of 12 in litterfall, indicating
thereby influences hydrology (Osterkamp and Hupp, 2010). sufficient supply of both N and P, and decreases with changes
The fossil record demonstrates that the evolution of land away from this optimal ratio (Lockaby and Walbridge,
plants, and in particular rooting, changed the hydro- 1998). Plant productivity in pristine European floodplains is
geomorphic patterns of rivers likely due to increased geo- N limited, but nutrient limitation is generally absent in
morphic stability (Davies and Gibling, 2010). The importance regulated and impacted floodplains due to high anthropo-
of vegetation to hydrogeomorphic processes has resulted in genic nutrient inputs (Antheunisse et al., 2006). In addition,
the labeling of vegetation as the bioengineer of floodplains discharge of nutrient-rich groundwater to floodplains in-
(Gurnell and Petts, 2006). creases basal area and decreases leaf carbon (C) to N ratios
Vegetation generally impacts geomorphology by influ- (Harner and Stanford, 2003) and also increases woody
encing the balance between deposition and erosion in flood- vegetation growth rates (Mouw et al., 2009). Furthermore,
plains, as well as by directly contributing to soil surface macrophyte and phytoplankton production in flood-
elevation in organic soil systems. Riparian and floodplain plain water bodies are enhanced by river nutrient import
vegetation affects bank erosion by modifying roughness and (Hamilton and Lewis, 1987; Heiler et al., 1995; Knowlton
hydraulics, including turbulence, flow diversion, infiltration, and Jones, 1997; Maine et al., 2004). In a forested floodplain,
and evapotranspiration (Tabacchi et al., 2000; Simon et al., the lowest soil N availability and the lowest plant product-
2004). Vegetation also directly impacts bank erosion by im- ivity were both found with fluctuating inundation compared
pacting water balance and mechanical reinforcement (Gurnell, to wetter or drier floodplain zones (Burke et al., 1999).
1997; Gray and Barker, 2004). On floodplains, dense vege- Despite the larger number of studies of the effects of nutrient
tation increases roughness and decreases water velocity and availability on aboveground production in floodplains, in
the shear stress on soils and sediments to levels that may be general little knowledge exists of how belowground pro-
below the critical entrainment threshold resulting in reduced duction in wetlands responds to nutrient enrichment. In
erosion (Smith, 2004; Larsen et al., 2011) and enhanced terrestrial forests, fine root production increases with N
sediment deposition (Ross et al., 2004). Vegetation on islands availability (Nadelhoffer, 2000).
in rivers also increases geomorphic stability and influences
hydrology (Schumm, 1985; Osterkamp, 1998). Similarly, large
woody debris influences hydrogeomorphic stability as well as
vegetative succession (Naiman et al., 2010). By contrast, less 12.21.3.4 Other Biota
vegetated floodplains can be subject to large-scale geomorphic It should be noted that other biota in addition to vegetation
change following large, eroding floods (Griffin and Smith, are important to floodplain dynamics and support complex
2004). The removal of riparian vegetation also dramatically interactions with biogeochemistry and hydrogeomorphology.
increases erosion rates (Vincent et al., 2009). Finally, flood- For example, beaver are widely regarded as ecosystem engin-
plain soils are typically thought of as mineral dominated but eers and keystone species that can impact the physical,
some floodplains have high soil organic content (median of chemical, and biotic environment of aquatic and terrestrial
swamps ¼ 43%, Appendix B in Bedford et al., 1999). Above- ecosystems (Johnston and Naiman, 1987; Naiman et al.,
ground and belowground production by floodplain vege- 1988). Marine-derived nutrients from salmon have been
tation, along with allochthonous inputs, directly supplies the shown to fuel aquatic production, food webs, and riparian
high organic content of soils in some floodplains and thus vegetation productivity (Naiman et al., 2002). Grazing of
influences their soil elevation and hydrogeomorphology. riparian vegetation can extensively change channel geo-
morphology, floodplain connectivity, and floodplain plant
communities (Beschta and Ripple, 2006). Of course, soil mi-
12.21.3.3 Biogeochemical Controls on Vegetation
crobes also are critical to ecosystem processes through their
The characteristics of vegetation can vary with the nutrient participation in biogeochemical cycles (Megonigal et al.,
biogeochemistry of floodplain ecosystems, depending on 2004).
312 Interactions among Hydrogeomorphology, Vegetation, and Nutrient Biogeochemistry in Floodplain Ecosystems

12.21.4 Controls of Nutrient Biogeochemistry in Survey, unpublished data). Ammonification rates are greater
Floodplains with the water table near the soil surface (o10 cm below),
nitrification rates are greater at the lowest water table (430 cm
12.21.4.1 Hydrogeomorphic Controls of Nutrient below), and denitrification rates are greatest in the middle
Biogeochemistry where greater nitrate availability and lower soil redox coincide
(water table 10–30 cm below the soil surface; Hefting et al.,
Hydrology and geomorphology create circumstances for aug- 2004). Johnston et al. (2001) found that hydrology was more
mented biogeochemical reactions by regulating when and important than soil characteristics in determining nutrient
where transport pathways converge and mix organic matter cycling in floodplains, with hot spots of nitrate production and
(electron donors) and electron acceptors (McClain et al., denitrification on natural levees. Flooding also deposits re-
2003; Vidon et al., 2010). One such hot spot is floodplains, fractory P and releases labile P from riparian soils (Fabre et al.,
where hot moments also occur. Floodplains are located at 1996). Finally, floodplains may vary from being sinks, sources,
depositional and low-velocity geomorphic positions in river- or transformers of nutrients depending on the seasonality of
ine landscapes where high primary productivity and large inundation (Noe and Hupp, 2007).
pools of organic matter (hot spot) mix with transport path- Hydrologic connectivity sets the template for river–-
ways of reactive and particulate N and P (often during flood floodplain interaction and the capacity for floodplains to retain
pulses; hot moments). These hydrogeomorphic properties are material (Heiler et al., 1995). The exchange of water between
the ultimate reason why floodplains are important regulators river and floodplain is an essential aspect of floodplains that
of water quality. For example, denitrification in river corridors regulates internal biogeochemical cycling as well as their cap-
is enhanced where nitrate is delivered to carbon-rich and acity to influence riverine water quality. Only material loaded to
oxygen-poor environments along lateral floodplain and a floodplain has the possibility of being retained or processed
vertical hyporheic gradients (Pinay et al., 1994). in the floodplain. In the case of surface water, flood pulses and
River–floodplain hydrology, as influenced by floodplain water exchange between channel and floodplain are critical
geomorphology, is generally accepted as the dominant control (Junk et al., 1989). Floodplains along more incised reaches
of biogeochemical process rates in floodplains (Lockaby (Stubblefield et al., 2006) or channelized and artificially leveed
and Walbridge, 1998; Hill, 2000). Hydrogeomorphology de- reaches (Noe and Hupp, 2005) have less connectivity between
termines key aspects of floodplains: inundation characteristics the main channel and floodplains and, therefore, trap less
and water velocity. These, in turn, determine the hydraulic sediment and associated P, N, and C. Riverine floodplains,
residence time of transport pathways in the floodplain and the having connection to rivers with larger watershed areas and
loading rate of material to the floodplain, together with the material loading, have greater long-term P accumulation rates
amount of material in source waters. All of these determine than nonconnected depressional wetlands (Craft and Casey,
the deposition rates of material and biotic processing rates. 2000). It should be noted that not only overbank flooding is
Thus, hydrogeomorphology determines biogeochemical sinks, important – flow pulses below bank full also have important
sources, and transformations in floodplains. influence on ecosystem productivity and spatial heterogeneity
In particular, inundation dynamics are one basic de- in river–floodplain systems (Tockner et al., 2000).
terminant of biogeochemical processes in floodplains. For ex- Surficial flowpaths route water and associated materials
ample, the biogeochemistry of floodplain soils is highly into, through, and out of floodplains. Retention of particulate
responsive to drying and wetting events (Baldwin and Mitchell, N and P is strongly influenced by hydrogeomorphology and
2000). Decomposition rates, and associated mineralization or vegetation through regulation of sedimentation rates in
immobilization of N and P, are a function of inundation in floodplain flowpaths (Walbridge and Struthers, 1993; Steiger
wetlands (Brinson et al., 1981). Lockaby et al. (1996) experi- and Gurnell, 2002; Noe and Hupp, 2005; Hupp et al., 2008;
mentally manipulated inundation on a floodplain and found Larsen et al., 2011). Greater sedimentation and associated C,
that decomposition rates were highest with temporary, aerobic N, and P accumulation occur where water rich in sediment
flooding compared to nonflooded and longer duration flood- and particulates and dissolved nutrients enter floodplains
ing. Floodplain wetness also affects N and P retranslocation, (Johnston et al., 1984; Sánchez-Carrillo et al., 2001; Ross et al.,
the removal of nutrients from tissues prior to senescence, in 2004; Knösche, 2006; Fink and Mitsch, 2007; Kronvang et al.,
deciduous floodplain trees (Clawson et al., 2001). Because the 2007). Lower elevation soil surfaces in floodplains also sup-
nutrient content and lability of organic matter can influence port greater sedimentation and C, N, and P accumulation rates
microbial decomposition of organic matter and N and P cyc- (Olde Venterink et al., 2006; Stoeckel and Miller-Goodman,
ling (Aerts, 1997), the effects of inundation patterns can 2001). For example, near-channel riparian areas that were
propagate through other floodplain ecosystem processes. lower in elevation trapped more sediment, C, and N than
Water-table depth and fluctuation also are influential in de- more extensive but higher floodplains (Brunet et al., 1994).
termining redox and N cycling. Denitrification rates are greater These spatial patterns of nutrient accumulation in floodplains
in restored floodplains that have re-established hydrology than propagate through other ecosystem processes, exemplifying
those that do not have re-established hydrology and have in- the interactions among hydrogeomorphology, nutrient bio-
frequent surface-water inundation (Hunter et al., 2008). geochemistry, and vegetation. Wassen et al. (2003) found that
Similarly, fluctuations of surface water level increase with floodplain plant biomass, plant N and P storage, and soil N
connectivity to streams and support greater rates of N sedi- and P mineralization rates all increased with proximity to the
mentation and ammonification in restored and natural channel. Increased deposition of sediment N and P on
floodplains (Wolf, K.L., Noe, G.B., Ahn, C., U.S. Geological floodplain soils leads to enhanced rates of soil N and P
 Interactions among Hydrogeomorphology, Vegetation, and Nutrient Biogeochemistry in Floodplain Ecosystems 313

mineralization and also greater litterfall N fluxes (Noe, G.B., Specifically, concentrations of amorphous (noncrystalline) Al
et al., U.S. Geological Survey, unpublished data). and Fe best predict the P adsorption capacity of soils in wet-
In wetlands in general, water residence time in flowpaths is lands in general (Richardson, 1985). The large spatial variation
a master controller of biogeochemical reaction rates (Kadlec in these parameters within and among floodplain wetlands
and Wallace, 2009). In floodplains, water residence time is results in patchy P dynamics (Bridgham et al., 2001).
critical to nutrient retention and transformation rates once Amorphous Fe and Al have complex but different distribution
streamflow occurs over a connected floodplain flowpath. patterns and associations; amorphous Fe was more abundant at
Floodplain backwaters with shorter residence times have less lower elevations in floodplains and inputs were predominantly
percent retention of nitrate compared to longer residence time from flooding, whereas amorphous Al was associated with or-
flowpaths; nitrate uptake rates are maximal at intermediate ganic matter and inputs were predominantly from uplands
nitrate-loading rates because moderate residence time permits (Darke and Walbridge, 2000). However, Fe–P and Al–P did not
some uptake in the upper reaches and transports nitrate to change with experimental floodplain flooding, although inun-
downstream reaches for further uptake (James et al., 2008). dation did release microbial P that increased soil P availability
Nitrate concentrations are heterogeneous across the flood- (Wright et al., 2001). Chacon et al. (2005) also found greater
plain during periods of low flow, when limited hydraulic labile P in drier zones associated with microbes, but less
connectivity and longer residence time permit depletion of amorphous Fe and Al in flooded floodplains. Riparian wetlands
nitrate pools in water bodies away from the main channel, with differing clay and noncrystalline Al and Fe content have
and more homogenous during high flow, when nitrate-rich differing P sorption capacity (Bruland and Richardson, 2004).
main channel waters may be distributed across floodplains, Watershed urbanization increases the crystallinity of riparian Fe,
and shorter residence times restrict nitrate uptake (Heiler and decreases soil P sorption capacity, likely due to increased
et al., 1995; Tockner et al., 1999; Richardson et al., 2004). In deposition of crystalline mineral sediment eroded from terres-
addition, slow-moving groundwater associated with shallower trial soils (Hogan and Walbridge, 2007). At the other extreme,
hydraulic gradients is also associated with greater nitrate re- riparian soils with very high organic content but little Al and Fe
moval than faster-moving groundwater in riparian ecosystems can have less P sorption capacity (Richardson, 1985; Lyons
(Sabater et al., 2003). et al., 1998). In conclusion, it is necessary to consider Fe and Al
Inorganic nutrients transported from the river channel inputs, fractionation, and transformations to predict floodplain
to floodplain lakes fuel primary productivity and are converted P dynamics and retention.
into organic nutrient forms (Hamilton and Lewis, 1987; Other soil characteristics influence nutrient biogeochemistry
Heiler et al., 1995; Knowlton and Jones, 1997; Tockner et al., in floodplains. The proximity to current or previous channels in
1999; Hein et al., 2003; Maine et al., 2004). During low-flow river–floodplain ecosystems determines the texture and ele-
periods, a side arm of the Danube River was dominated by vation of alluvial floodplain soils (Stanturf and Schoenholtz,
autochthonous primary production, whereas during flooding, 1998). In addition to determining metal biogeochemistry, soil
autochthonous carbon was exported to the main channel and variability associated with floodplain microtopography directly
allochthonous carbon was imported to the floodplain (Preiner influences rates of N and P fluxes. For example, potential rates
et al., 2008). However, short-hydroperiod floodplains can ex- of soil N mineralization flux are greater in finer texture soils of
hibit the opposite pattern of organic nutrient retention and in- riparian geomorphic zones due to greater N pools compared to
organic nutrient export to the river during flooding (Noe and coarser-textured soils (Bechtold and Naiman, 2006). Hot spots
Hupp, 2007). of nitrate production and denitrification have been found on
Groundwater flowpaths in river–floodplain corridors have a natural levees with coarser-textured soils (Johnston et al., 2001).
large role in regulating nutrient transport, redistribution, and However, flooding or rain after dry periods can flush large
retention (Stanford et al., 2005.) Discharge of hyporheic amounts of soil nitrate that accumulated from mineralization
groundwater on the floodplain can be enriched in nutrients and nitrification in oxidized soils (Bechtold et al., 2003). The
(Harner and Stanford, 2003). Groundwater discharge increases larger particle size and lower organic content of soils in flood-
tree growth and increases lability of tree leaves (decreases C:N; plain natural levees compared to finer-textured backswamp
Harner and Stanford, 2003), which should lead to enhanced N deposits result in less P sorption in sandier levees (Axt and
mineralization rates. Hydrology, geomorphology, and soil Walbridge, 1999; Bridgham et al., 2001).
characteristics determine the retention of groundwater nitrate Finally, the geologic setting of catchments also influences
transported from uplands through floodplains, with shallow the biogeochemistry of floodplains. Floodplains on alluvial
groundwater transported slowly through wet and organic versus blackwater rivers differ in chemistry and sediment supply
floodplain soils having high capacity for nitrate removal (Gold (Wharton et al., 1982; Junk and Furch, 1993; Hupp, 2000;
et al., 2001). Denitrification can also occur in hyporheic flow- Schilling and Lockaby, 2006; Anderson and Lockaby, 2007).
paths of the riparian zone (Triska et al., 1993). Blackwater rivers draining Coastal Plains are characterized by
Hydrogeomorphology has an additional influence on P lower suspended sediment load, ionic strength, pH, less in-
biogeochemistry through metal biogeochemistry. Trace-metal organic material, and more organic matter than alluvial rivers
transformations (including those involving P) are generally a draining Piedmont or montane physiographic provinces (Beck
function of topography, flooding, and vegetation in floodplains et al., 1974). As a result, sedimentation rates on floodplains of
(Du Laing et al., 2009). The retention of dissolved P, in par- blackwater rivers are among the lowest recorded (Hupp, 2000).
ticular, is determined by the geochemistry of Fe, Al, and Ca, as Floodplain soils of blackwater rivers are more organic, acidic,
well as organic matter, which, in turn, are influenced by hy- and N and P deficient, and have lower cellulose and lignin
drology and redox (Walbridge and Struthers, 1993). decomposition than those of alluvial rivers (Stanturf and
314 Interactions among Hydrogeomorphology, Vegetation, and Nutrient Biogeochemistry in Floodplain Ecosystems

Schoenholtz, 1998; Entry, 2000). Alluvial rivers are character- permeability. Other mechanisms for direct control are pos-
ized by more P and a lower N:P in floodplain litterfall that is sible, but in general very little information exists on this topic.
closer to optimum for plant productivity (Schilling and Lock- One possible mechanism for direct control of bio-
aby, 2006) and, as a result, have greater vegetative productivity geochemistry on hydrogeomorphology is through changes in
(Junk and Furch, 1993; Lockaby and Walbridge, 1998; Schilling soil macropores and permeability. The amount and degree of
and Lockaby, 2006) than blackwater rivers. decomposition of soil organic matter influence soil physical
properties and processes, including bulk density, water-hold-
12.21.4.2 Influence of Vegetation on Nutrient ing capacity, and hydraulic conductivity, dominantly (Stanturf
Biogeochemistry and Schoenholtz, 1998). Macropores increase water and sol-
ute fluxes through wetland soils (Harvey and Nuttle, 1995).
The vegetation of floodplains directly affects nutrient bio- Many soil macropores are formed in channels left by de-
geochemistry by storing nutrients in biomass as well as in- composing roots (Beven and Germann, 1982). Changes in the
directly by producing organic matter that supports microbial decomposition rate of the refractory organic matter that sta-
processing of nutrients. Although not permanently retained, bilizes macropores and smaller pores due to increased nutri-
plant uptake of nutrients into biomass can be a large magni- ent availability could reduce permeability and have large
tude flux. Riparian vegetative uptake can range from 13% to impacts on the water balance of floodplain wetlands. Changes
99% of total measured system N retention in riparian buffers in root production and subsequent macropore formation by
across Europe (uptake þ denitrification; Hefting et al., 2005). altered nutrient availability could indirectly influence hydrol-
In some floodplains, similar quantities of P enter the flood- ogy as well, especially during the rising and falling limbs of
plain and are recycled through the vegetation of the floodplain inundation events.
(Mitsch et al., 1979; Yarbro, 1983). Similarly, standing stocks The elevation of the soil surface relative to site hydrology is
of N and P stored in herbaceous biomass are similar to the an essential trait in wetland ecosystems. For example, suf-
annual inputs of N and P in wetlands (Olde Venterink et al., ficient accretion of the soil surface is necessary for coastal
2002). However, storage in plant tissues can be a small pro- wetlands to remain stable during rising sea levels (Callaway
portion of experimental N and P additions to floodplains et al., 1996; Morris et al., 2002). Soil organic matter pro-
(Brinson et al., 1984). duction is a critical component of soil accretion in wetlands
Vegetation also indirectly influences biogeochemistry and may be enhanced by increased nutrient availability (Craft
through the production of food and electron receptors for and Richardson, 1993; Nyman et al., 2006). Importantly, de-
microbes. The production of roots, exudates, and detritus in- composition of soil organic matter in wetlands also is
fluences the populations, communities, and activities of soil generally limited by nutrient availability, either through
microbes including denitrifiers and symbiotic root microbes bioavailable N or P inputs to soil microbes or through regu-
(Tabacchi et al., 2000). The lability of detritus, due in part to lation of the nutrient content, stoichiometry, and lability of
species differences in lignin content as well as nutrient stoi- detritus (Brinson et al., 1981). Nutrient enrichment increases
chiometry, influences nutrient cycling (Melillo et al., 1982; respiration in wetland soils (Amador and Jones, 1995) as well
Lockaby and Walbridge, 1998; Burke et al., 1999). Portions of as decreases the pool of C in some wetland soils (Morris and
the N and P stored in floodplain plant litter are mineralized by Bradley, 1999; Wigand et al., 2009) but increases C storage in
microbes into inorganic nutrient forms that are recycled others (Craft and Richardson, 1993). Most research on this
through plant uptake or exported from the floodplain (Noe, topic has not occurred in floodplains. However, these pro-
G.B., et al., U.S. Geological Survey, unpublished data). cesses are relevant to floodplains because floodplain soils can
have a large organic component as previously described.
Clearly, more research is needed to determine the generalized
12.21.4.3 Biogeochemical Controls on
effects of nutrient availability on organic matter fluxes and
Hydrogeomorphology?
thus soil elevation in floodplains.
Nutrient biogeochemistry indirectly influences floodplain
hydrogeomorphology through nutrient control of vegetation
(see the previous section). Because vegetation in many
12.21.5 Case Studies
floodplains is nutrient limited, increased nutrient availability
will increase plant productivity. This increase in productivity
The following case studies describe some well-known examples
results in greater aboveground biomass. Aboveground bio-
of floodplain ecosystems with strong interactions among
mass has a large influence on water velocity, sediment
hydrogeomorphology, vegetation, and nutrient biogeochemistry.
entrainment thresholds, and deposition rates in floodplains
(Harvey et al., 2009). Belowground biomass influences
soil surface elevation by modifying erosion rates (Wynn and
12.21.5.1 Hummock/Hollow Geomorphology in Peatlands
Mostaghimi, 2006), although little is known about nutrient
and Floodplains
controls on belowground production.
Does nutrient biogeochemistry only indirectly influence Scale-dependent feedback processes among hydro-
the hydrogeomorphology of floodplain ecosystems geomorphology, vegetation, and biogeochemistry can generate
through control of vegetation (Figure 1)? No – nutrient con- pattern in ecosystems (Rietkerk and van de Koppel, 2008). The
trols on decomposition rates could directly influence hydro- patterned hummock and hollow microtopography in some
geomorphology through controls on soil elevation and wetlands and floodplains is a clear example of these interactive
 Interactions among Hydrogeomorphology, Vegetation, and Nutrient Biogeochemistry in Floodplain Ecosystems 315

processes. The hummock and hollow pattern in peatlands is 2000; Naiman et al., 2010). Forested floodplain wetlands were
controlled by scale-dependent feedback processes: greater abundant along the small and large rivers of the Coastal Plain
evapotranspiration on hummocks creates groundwater trans- in the Southeastern United States (Conner and Day, 1982;
port toward the hummock from hollows, transporting nutri- Wharton et al., 1982; Sharitz and Mitsch, 1993; Messina and
ents to hummocks that increase vegetative production, and Conner, 1998). However, conversion to agricultural land use
soil surface elevation, and further increase evapotranspiration and logging has impacted a large portion of their areal extent
on hummocks (Eppinga et al., 2008). Alternatively, greater (Conner and Day, 1982; Sharitz and Mitsch, 1993).
plant production on hummocks increases the deposition of The fluvial dynamics of Coastal Plain river–floodplain sys-
sediment and associated nutrients which then increase local tems generate complex geomorphic patterns, including natural
plant production and soil elevation (Larsen et al., 2007). levees, crevasses and their splays, backswamps, ridges, swales,
Hummock and hollow microtopography has large impli- sloughs, abandoned channels, oxbow lakes, and point bars
cations for wetland ecosystem processes. The greater hetero- (Wharton et al., 1982). Coastal Plain floodplains also have
geneity occurring with hummocks and hollows increases distinct hydrology. Sweet and Geratz (2003) found that Coastal
diversity of wet-meadow plants (Werner and Zedler, 2002) Plain rivers reach bank full discharge on average every 0.19 year,
and soil microbes in floodplains (Ahn et al., 2009). Hum- 8 times more frequently than other rivers. As a result, Coastal
mocks have different pore-water chemistry than hollows Plain floodplains are frequently inundated. Over 50% of the
(Stribling et al., 2007). Root production is greater on flood- floodplain of a Coastal Plain floodplain is inundated more
plain hummocks (Jones et al., 2000), suggesting that higher than 15% of the average year (Benke et al., 2000). Floodplains
elevations are hot spots for nutrient and water uptake. Add- along Coastal Plain reaches of rivers typically have very high
itions of hummock and hollow microtopography to created sediment-accumulation rates where rivers originate outside the
wetlands increase plant diversity and spatial heterogeneity in Coastal Plain or have disturbed catchments with high sediment
soil chemistry (Bruland and Richardson 2005; Moser et al., loads (Hupp, 2000; Noe and Hupp, 2005; Phillips and Slattery,
2007; Moser et al., 2009). 2006; Kronvang et al., 2007; Hupp et al., 2008). By contrast,
Hummock and hollow microtopography is common in floodplain ecosystems along blackwater rivers that originate
tidal freshwater to brackish forested floodplains (Day et al., within the Coastal Plain have different structure and function
2007). This microtopography creates large differences in compared to those along alluvial rivers that originate outside
water-table elevation relative to the soil surface in tidal the Coastal Plain, as previously discussed.
freshwater floodplains and is the dominant gradient control- Plant communities change with geomorphic position on
ling plant distribution and species richness in these frequently Coastal Plain floodplains due to differential species tolerance to
inundated wetlands (Rheinhardt, 2007). Tree species in tidal inundation and anoxia (Wharton et al., 1982; Sharitz and
freshwater floodplains occur preferentially on hummocks Mitsch, 1993; Conner, 1994) and successional age of dynamic
(and likely help build and maintain hummocks) compared to geomorphic surfaces (Shankman, 1993). Annual productivity of
hollows in the more stressful, wetter areas of the floodplain forested wetlands in the Southeastern Coastal Plain of the
(Duberstein and Conner, 2009). However, little is known United States varies greatly depending on site hydrology and
about the nutrient biogeochemistry of hummocks compared disturbance (Conner, 1994). Although early research found high
to hollows in tidal freshwater floodplains (Anderson and production in Coastal Plain floodplain forests that were sub-
Lockaby, 2007). sidized by fluctuating hydrology, it is now thought that prod-
Microtopographic heterogeneity in floodplains, and wet- uctivity in floodplains is similar to uplands and decreases with
lands in general, clearly has importance to the structure and flooding depth during the growing season (Megonigal et al.,
function of these ecosystems. A critical research need is to better 1997). However, it was found no differences in aboveground
understand the interacting physical and biological processes production among wetter and drier sites on two floodplains. A
that create, maintain, and destroy this geomorphology, par- large amount of the high litterfall production may be de-
ticularly in response to disturbances. For example, the peatlands composed, especially in wetlands with fluctuating hydrology
of the Florida Everglades historically supported but have lost (Brinson et al., 1981), compared to terrestrial ecosystems in
much of their microtopographic patterning (mimicking riv- temperate and arid regions (Aerts, 1997). Nonetheless, a large
er–floodplain hydrogeomorphology) following human modi- quantity and proportion of the detritus production on forested
fication of system hydrology; identification of the processes that floodplains of the Coastal Plain is exported to the main channel
will restore this pattern is a priority (Larsen et al., 2011). of the river during floods (Cuffney, 1988).
The hydrogeomorphic complexity of Coastal Plain
floodplains makes them difficult to study but determines
12.21.5.2 Coastal Plain Floodplains
their essential biogeochemical characteristics (Lockaby and
The extensive floodplains in the Coastal Plain physiographic Walbridge, 1998). Thus, geologic, geographic, geomorphic,
province also demonstrate the importance of feedbacks and hydrologic characteristics of these floodplains imprint
to ecosystem pattern and process. Coastal Plains are pre- biogeochemical processes. The blackwater rivers that originate
dominantly flat, depositional landscapes with wide alluvial in the Coastal Plain have low pH and abundant organic matter
valleys located downstream from sloping erosional landscapes with Al and Fe complexes (Beck et al., 1974). Coastal Plain
(Markewich et al., 1990). The low gradient of Coastal Plain floodplains are therefore generally rich in organic matter
rivers promotes high depositional rates of transported river complexed with Al and Fe which enhances their ability to
sediment load, creating meandering channels with extensive retain inorganic P (Darke and Walbridge, 2000). The crystal-
areas of alluvial floodplains (Wharton et al., 1982; Hupp, linity, complexation with organic matter, or oxidation state of
316 Interactions among Hydrogeomorphology, Vegetation, and Nutrient Biogeochemistry in Floodplain Ecosystems

Fe and Al is sensitive to inundation and anoxia, which should heterogeneity in water sources create river channel expansion
decrease P retention in wetter locations (Darke and Walbridge, and contraction cycles that control spatiotemporal hetero-
2000). Flooding increases soil P availability, but the mech- geneity, water quality, primary productivity, and food webs
anism was loss of microbial P instead of solubilization of Fe–P (Stanford et al., 2005; Malard et al., 2006).
or Al–P fractions with waterlogging (Wright et al., 2001). The frequency and intensity of scouring critically influence
Within Coastal Plain floodplains, lower elevations accu- the distribution of vegetation in montane floodplains
mulate finer sediments and are enriched in soil P, N, and C (Fetherston et al., 1995). Scour holes provide lentic habitat in
(Stoeckel and Miller-Goodman, 2001). Soil N availability is a lotic environment (Stanford et al., 2005). Scouring at lower
also greater in wetter portions of these floodplains (Burke elevations limits the survival of seedlings, whereas coarser
et al., 1999). The fine texture of Coastal Plain floodplain soils sediment at higher elevations also limits the survival of seed-
results in much higher P sorption capacities than the sandier lings by decreasing water availability (Gage and Cooper,
regional upland soils (Walbridge and Struthers, 1993). The 2004). Flood damage to riparian vegetation is greater in
overall high sedimentation rates on Coastal Plain floodplains steeper gradient reaches of montane floodplains and increases
correspond to high accumulation and retention rates of the plant species diversity (Hupp, 1982).
C, N, and P associated with sedimentation (Noe and Hupp,
2009; Hupp et al., 2009). Riverine Coastal Plain floodplain
12.21.5.4 Desert Streams
wetlands have greater soil P accumulation rates compared to
nonriverine depressional wetlands, although sedimentation Desert stream ecosystems also are illustrative examples of the
mass, and C and N accumulation rates were similar (Craft and interaction of hydrogeomorphology, vegetation, and bio-
Casey, 2000). However, the percent of river nitrogen retained geochemistry. In fact, they have been used to describe the
by floodplains may be insignificant in Coastal Plain rivers with functional ecomorphology of streams (Fisher et al., 2007).
high dissolved nutrient loads (van der Lee et al., 2004). Less infiltration in upland desert soils generates more overland
flow and less groundwater flow to desert riparian zones (Martı́
et al., 2000). As a result, desert rivers are more hydro-
12.21.5.3 Montane Floodplains
geomorphically active and dynamic than their temperate
Montane rivers are characterized by high channel gradients, counterparts with higher rates of upland and bank erosion
which lead to high hydraulic energy, dynamic channels, large and episodic sediment transport (Ffolliott et al., 2004).
sediment size, and high bedload/suspended load ratios Marked temporal variation in the absence of surface water
(Schumm, 1985). Channel geomorphic pattern within mon- highlights the importance of drying processes in intermittent
tane alluvial floodplain reaches is a function of catchment slope channels (Stanley et al., 1997). At the other end of hydrologic
and channel constriction (Beechie et al., 2006). Higher gradient extremes, frequent flash flooding resets ecosystem succession
reaches of montane floodplain rivers are generally braided, (Fisher et al., 1982). The distribution of riparian vegetation is
transitioning to island braided and then meandering with de- closely linked to groundwater depths and the frequency of
creasing gradient at lower elevations (Schumm, 1985). This discharge events (Hupp and Osterkamp, 1996). Stream en-
change in pattern is associated with increasing turnover of trenchment lowers riparian groundwater tables resulting in
floodplain surfaces at higher elevations due to more frequent desertification of wetland plant communities in arid regions
channel migration (Beechie et al., 2006). Within the island- (Stromberg et al., 1996; Chambers et al., 2004). Longitudinal
braided domain of channel pattern in montane river– gradients in riparian soil moisture are typically opposite of
floodplain systems, Ward et al. (1999) proposed a model of most temperate climates, with greater soil moisture at higher
island development that incorporates feedbacks between vege- watershed elevations due to greater evapotranspiration and
tation and hydrogeomorphology. Island persistence is a func- less precipitation at lower elevations (Ffolliott et al., 2004).
tion of vegetation density, organic matter accumulation, and the Research in arid stream ecosystems has shown that
frequency of high discharge (Ward et al., 1999). River regulation hydrologic exchange between surface and subsurface waters
can change channel geomorphology from multi-channel brai- has important implications for ecological and biogeochemical
ded to single-channel anastomosing (Ward et al., 1999). processes. Algal productivity is commonly nitrogenlimited in
The habitat template for biotic communities in montane desert streams (Grimm and Fisher, 1986). Hyporheic flow-
floodplains is a function of their highly dynamic geomorph- paths are hot spots of nitrate production in desert stream
ology that generates high geomorphic and hydrologic diversity ecosystems due to high nitrification rates coupled with min-
(Poole et al., 2002). The extensive hyporheic zone of montane eralization of surficial dissolved or particulate organic N
river–floodplain systems supports large flows, extensive and (Holmes et al., 1994) in microbially active subsurface sedi-
biodiverse subterranean foodwebs, and important biogeo- ments (Grimm and Fisher, 1984). Reaches with hyporheic
chemical transformations (Stanford and Ward, 1993). discharge, which is determined by stream geomorphology,
Groundwater discharge locations on montane floodplains therefore have greater primary production due to greater
increase nutrient availability and periphyton and macrophyte nitrate loading (Valett et al., 1994).
production and lower plant C:N (Harner and Stanford, 2003; Riparian vegetation stabilizes the geomorphology of desert
Stanford et al., 2005; Mouw et al., 2009). Surface flow events streams. Desert riparian zones (both in-channel and flood-
at bankfull or below bankfull are important for creating eco- plain) exhibit alternative stable states where abundant vege-
system heterogeneity in montane floodplains, particularly in tation reduces geomorphic change during flooding, resulting
braided rivers with larger amounts of channel edge (Tockner in increased productivity and further stabilization from cata-
et al., 2000). Seasonal variation in discharge and spatial strophic erosional events that remove vegetation (Heffernan,
 Interactions among Hydrogeomorphology, Vegetation, and Nutrient Biogeochemistry in Floodplain Ecosystems 317

2008). Periods of high macrophyte abundance lead to shal- Amador, J.A., Jones, R.D., 1995. Carbon mineralization in pristine and phosphorus-
lower and wider channels with greater hydraulic retention enriched peat soils of the Florida Everglades. Soil Science 159, 129–141.
Amoros, C., Bornette, G., 2002. Connectivity and biocomplexity in waterbodies of
times in the channel (Harvey et al., 2003). Stream algal
riverine floodplains. Freshwater Biology 47, 761–776.
productivity also is more resilient to hydrologic disturbance in Amoros, C., Roux, A.L., Reygrobellet, J.L., Bravard, J.P., Pautou, G., 1987. A
areas of higher primary production associated with hyporheic method for applied ecological studies of fluvial hydrosystems. Regulated Rivers
upwelling (Valett et al., 1994). 1, 17–36.
Anderson, C.J., Lockaby, B.G., 2007. Soils and biogeochemistry of tidal freshwater
forested wetlands. In: Conner, W.H., Doyle, T.W., Krauss, K.W. (Eds.), Ecology of
Tidal Freshwater Forested Wetlands of the Southeastern United States. Springer,
12.21.6 Conclusions Dordrecht, pp. 65–88.
Antheunisse, A.M., Loeb, R., Lamers, L.P.M., Verhoeven, J.T.A., 2006. Regional
The many and potentially nonlinear interactions among differences in nutrient limitation in floodplains of selected European rivers:
hydrogeomorphology, vegetation, and nutrient biogeochemistry, implications for rehabilitation of characteristic floodplain vegetation. River
Research and Applications 22, 1039–1055.
nested in the four dimensional framework of river–floodplain Axt, J.R., Walbridge, M.R., 1999. Phosphate removal capacity of palustrine forested
systems, make floodplains difficult to study and gain insight. wetlands and adjacent uplands in Virginia. Soil Science Society of America
This complexity also makes quantitative predictions difficult. Journal 63, 1019–1031.
Yet this complexity provides an opportunity to understand the Baldwin, D.S., Mitchell, A.M., 2000. The effects of drying and re-flooding on the
many linkages between physical and biotic processes in eco- sediment and soil nutrient dynamics of lowland river–floodplain systems: a
synthesis. Regulated Rivers: Research and Management 16, 457–467.
systems. Current river–floodplain system concepts and Bechtold, J.S., Edwards, R.T., Naiman, R.J., 2003. Biotic versus hydrologic control
understandings have clearly advanced to include the sub- over seasonal nitrate leaching in a floodplain forest. Biogeochemistry 63, 53–72.
stantial spatial and temporal heterogeneity and interactive Bechtold, J.S., Naiman, R.J., 2006. Soil texture and nitrogen mineralization potential
processes inherent to floodplains. across a riparian toposequence in a semi-arid savanna. Soil Biology and
Biochemistry 38, 1325–1333.
Mechanistic models of floodplains are needed to
Beck, K.C., Reuter, J.H., Perdue, E.M., 1974. Organic and inorganic geochemistry of
better manage floodplains and should include hierarchical, some coastal plain rivers of the southeastern United States. Geochimica et
nonequilibrium, physical, and biological processes (Johnson Cosmochimica Acta 38, 341–364.
et al., 1995). Including interactions among hydrogeo- Bedford, B.L., Walbridge, M.R., Aldous, A., 1999. Patterns in nutrient availability and
morphology, vegetation, and nutrient biogeochemistry has plant diversity of temperate North American wetlands. Ecology 80, 2151–2169.
Beechie, T.J., Liermann, M., Pollock, M.M., Baker, S., Davies, J., 2006. Channel
improved predictive modeling of floodplain systems – a con- pattern and river–floodplain dynamics in forested mountain river systems.
crete demonstration of the utility of considering these inter- Geomorphology 78, 124–141.
actions. However, quantitative models of floodplain ecosystem Beechie, T.J., Sear, D.A., Olden, J.D., et al., 2010. Process-based principles for
dynamics are rare. The conceptual advances described in this restoring river ecosystems. BioScience 60, 209–222.
Benke, A.C., Chaubey, I., Ward, G.M., Dunn, E.L., 2000. Flood pulse dynamics of
chapter are integral to the few quantitative floodplain eco-
an unregulated river floodplain in the southeastern US Coastal Plain. Ecology
system models. For example, integrated and interactive models 81, 2730–2741.
of hydrology, geomorphology, vegetation, and N and P cycling Beschta, R.L., Ripple, W.J., 2006. River channel dynamics following extirpation of
have successfully described ecosystem dynamics in natural, wolves in northwestern Yellowstone National Park. USA. Earth Surface Processes
drained, and created floodplain wetlands (van der Peijl and and Landforms 31, 1525–1539.
Beven, K., Germann, P., 1982. Macropores and water flow in soils. Water Resources
Verhoeven, 2000; Wang and Mitsch, 2000). A model with Research 18, 1311–1325.
floodplain geomorphology, hydrology, and N cycling was Blood, L.E., Titus, J.H., 2010. Microsite effects on forest regeneration in a
used to demonstrate the effects of dams and artificial levees on bottomland swamp in western New York. Journal of the Torrey Botanical Society
floodplain nitrate removal (Gergel et al., 2005). Hydrologic 137, 88–102.
Bornette, G., Tabacchi, E., Hupp, C., Puijalon, S., Rostan, J.C., 2008. A model
exchange between shallow and deep habitats in a modeled
of plant strategies in fluvial hydrosystems. Freshwater Biology 53,
floodplain complex enhances system food-web production 1692–1705.
by coupling N regeneration and phytoplankton growth Bridgham, S., Pastor, J., Janssens, J., Chapin, C., Malterer, T., 1996. Multiple
(Cloern, 2007). Finally, Helton et al. (2010) concluded limiting gradients in peatlands: a call for a new paradigm. Wetlands 16,
that river biogeochemical models need to include hydrologic 45–65.
Bridgham, S.D., Johnston, C.A., Schubauer-Berigan, J.P., Weishampel, P., 2001.
exchange between main channel and floodplain and the Phosphorus sorption dynamics in soils and coupling with surface and pore water
resulting biogeochemical processes to predict N transport in riverine wetlands. Soil Science Society of America Journal 65, 577–588.
and cycling. Further development of holistic quantitative Brinson, M.M., 1988. Strategies for assessing the cumulative effects of wetland
models is needed to predict the future state of floodplain–river alteration on water quality. Environmental Management 12, 655–662.
Brinson, M.M., 1993. Changes in the functioning of wetlands along environmental
systems and their ecosystem services under altered hydro-
gradients. Wetlands 13, 65–74.
geomorphic regimes. Brinson, M.M., Bradshaw, H.D., Kane, E.S., 1984. Nutrient assimilative capacity of
an alluvial floodplain swamp. Journal of Applied Ecology 21, 1041–1057.
Brinson, M.M., Lugo, A.E., Brown, S., 1981. Primary productivity, decomposition
and consumer activity in freshwater wetlands. Annual Review of Ecology and
References Systematics 12, 123–161.
Brown, A.G., 1997. Alluvial Geoarchaeology: Floodplain Archaeology and
Environmental Change. Cambridge University Press, Cambridge.
Aerts, R., 1997. Climate, leaf litter chemistry and leaf litter decomposition in Bruland, G.L., Richardson, C.J., 2004. A spatially explicit investigation of
terrestrial ecosystems: a triangular relationship. Oikos 79, 439–449. phosphorus sorption and related soil properties in two riparian wetlands.
Ahn, C., Gillevet, P., Sikaroodi, M., Wolf, K., 2009. An assessment of soil bacterial Journal of Environmental Quality 33, 785–794.
community structure and physicochemistry in two microtopographic locations Bruland, G.L., Richardson, C.J., 2005. Hydrologic, edaphic, and vegetative
of a palustrine forested wetland. Wetlands Ecology and Management 17, responses to microtopographic reestablishment in a restored wetland.
397–407. Restoration Ecology 13, 515–523.
318 Interactions among Hydrogeomorphology, Vegetation, and Nutrient Biogeochemistry in Floodplain Ecosystems

Brunet, R.C., Pinay, G., Gazelle, F., Roques, L., 1994. Role of the floodplain and Ernst, W.H.O., 1990. Ecophysiology of plants in waterlogged and flooded
the riparian zone in the suspended matter and nitrogen retention in the Adour environments. Aquatic Botany 38, 73–90.
River, south-west France. Regulated Rivers Research and Management 9, Fabre, A., Pinay, G., Ruffinoni, C., 1996. Seasonal changes in inorganic and
55–63. organic phosphorus in the soil of a riparian forest. Biogeochemistry 35,
Burke, M.K., Lockaby, B.G., Conner, W.H., 1999. Aboveground production and 419–432.
nutrient circulation along a flooding gradient in a South Carolina Coastal Plain Fetherston, K.L., Naiman, R.J., Bilby, R.E., 1995. Large woody debris, physical
forest. Canadian Journal of Forest Research 29, 1402–1418. process, and riparian forest development in montane river networks of the
Callaway, J.C., DeLaune, R.D., Patrick, W.H., 1996. Chernobyl 137Cs used to Pacific Northwest. Geomorphology 13, 133–144.
determine sediment accretion rates at selected Northern European coastal Ffolliott, P.F., DeBano, L.F., Baker, M.B., Neary, D.G., Brooks, K.N., 2004. Hydrology
wetlands. Limnology and Oceanography 41, 444–450. and impacts of disturbance on hydrologic function. In: Baker, M.B., Ffolliott,
Callaway, R.M., Walker, L.R., 1997. Competition and facilitation: a synthetic P.F., DeBano, L.F., Neary, D.G. (Eds.), Riparian Areas of the Southwestern United
approach to interactions in plant communities. Ecology 78, 1958–1965. States: Hydrology, Ecology and Management. CRC Press, Boca Raton, FL,
Chacon, N., Dezzeo, N., Muñoz, B., Rodrı́guez, J.M., 2005. Implications of soil pp. 1–9.
organic carbon and the biogeochemistry of iron and aluminum on soil Fink, D.F., Mitsch, W.J., 2007. Hydrology and nutrient biogeochemistry in a created
phosphorus distribution in flooded forests of the lower Orinoco River, Venezuela. river diversion oxbow wetland. Ecological Engineering 30, 93–102.
Biogeochemistry 73, 555–566. Fisher, S.G., 1997. Creativity, idea generation, and the functional morphology of
Chambers, J.C., Tausch, R.J., Korfmacher, J.L., Germanoski, D., Miller, J.R., Jewett, streams. Journal of the North American Benthological Society 16, 305–318.
D., 2004. Effects of geomorphic processes and hydrologic regimes on riparian Fisher, S.G., Gray, L.J., Grimm, N.B., Busch, D.E., 1982. Temporal succession in a
vegetation. In: Chambers, J.C., Miller, J.R. (Eds.), Great Basin Riparian desert stream ecosystem following flash flooding. Ecological Monographs 52,
Ecosystems: Ecology, Management, and Restoration. Island Press, Washington, 93–110.
DC, pp. 196–231. Fisher, S.G., Grimm, N.B., Martı́, E., Holmes, R.M., Jones, J.B.J., 1998. Material
Church, M., 2002. Geomorphic thresholds in riverine landscapes. Freshwater spiraling in stream corridors: a telescoping ecosystem model. Ecosystems 1,
Biology 47, 541–557. 19–34.
Clawson, R.G., Lockaby, B.G., Rummer, B., 2001. Changes in production and Fisher, S.G., Heffernan, J.B., Sponseller, R.A., Welter, J.R., 2007. Functional
nutrient cycling across a wetness gradient within a floodplain forest. Ecosystems ecomorphology: feedbacks between form and function in fluvial landscape
4, 126–138. ecosystems. Geomorphology 89, 84–96.
Cloern, J.E., 2007. Habitat connectivity and ecosystem productivity: implications Gage, E., Cooper, D., 2004. Constraints on willow seedling survival in a Rocky
from a simple model. American Naturalist 169, E21–E33. Mountain montane floodplain. Wetlands 24, 908–911.
Conner, W., 1994. Effect of forest management practices on southern forested Galat, D.L., Fredrickson, L.H., Humburg, D.D., et al., 1998. Flooding to restore
wetland productivity. Wetlands 14, 27–40. connectivity of regulated, large-river wetlands. BioScience 48, 721–733.
Conner, W.H., Day, J.W., 1982. The ecology of forested wetlands in the Gergel, S.E., Carpenter, S.R., Stanley, E.H., 2005. Do dams and levees impact
southeastern United States. In: Gopal, B., Turner, R.E., Wetzel, R.G., Whigham, nitrogen cycling? Simulating the effects of flood alterations on floodplain
D.F. (Eds.), Wetlands Ecology and Management. National Institute of Ecology denitrification. Global Change Biology 11, 1352–1367.
and International Scientific Publications, New Delhi, pp. 69–87. Gold, A.J., Groffman, P.M., Addy, K., Kellogg, D.Q., Stolt, M., Rosenblatt, A.E.,
Costanza, R., d’Arge, R., de Groot, R., et al., 1997. The value of the world’s 2001. Landscape attributes as controls on ground water nitrate removal capacity
ecosystem services and natural capital. Nature 387, 253–260. of riparian zones. Journal of the American Water Resources Association 37,
Craft, C.B., Casey, W.P., 2000. Sediment and nutrient accumulation in 1457–1464.
floodplain and depressional freshwater wetlands of Georgia. USA. Wetlands 20, Gray, D.H., Barker, D., 2004. Root-soil mechanics and interactions. In: Bennett, S.J.,
323–332. Simon, A. (Eds.), Riparian Vegetation and Fluvial Geomorphology. American
Craft, C.B., Richardson, C.J., 1993. Peat accretion and phosphorus accumulation Geophysical Union, Washington, DC, pp. 113–124.
along a eutrophication gradient in the northern Everglades. Biogeochemistry 22, Gregory, S.V., Swanson, F.J., McKee, W., Cummins, K.W., 1991. An ecosystem
133–156. perspective of riparian zones. BioScience 41, 540–551.
Cuffney, T.F., 1988. Input, movement and exchange of organic matter within a Griffin, E.R., Smith, J.D., 2004. Floodplain stabilization by woody riparian
subtropical coastal blackwater river–floodplain system. Freshwater Biology 19, vegetation during an extreme flood. In: Bennett, S.J., Simon, A. (Eds.), Riparian
305–320. Vegetation and Fluvial Geomorphology. American Geophysical Union,
Darke, A.K., Walbridge, M.R., 2000. Al and Fe biogeochemistry in a floodplain Washington, DC, pp. 221–236.
forest: implications for P retention. Biogeochemistry 51, 1–32. Grimm, N.B., Fisher, S.G., 1984. Exchange between interstitial and surface water:
Davies, N.S., Gibling, M.R., 2010. Cambrian to Devonian evolution of alluvial implications for stream metabolism and nutrient cycling. Hydrobiologia 111,
systems: the sedimentological impact of the earliest land plants. Earth-Science 219–228.
Reviews 98, 171–200. Grimm, N.B., Fisher, S.G., 1986. Nitrogen limitation in a Sonoran Desert stream.
Day, F., Megonigal, J., 1993. The relationship between variable hydroperiod, Journal of the North American Benthological Society 5, 2–15.
production allocation, and belowground organic turnover in forested wetlands. Gurnell, A., 1997. The hydrological and geomorphological significance of forested
Wetlands 13, 115–121. floodplains. Global Ecology and Biogeography 6, 219–229.
Day, R.H., Williams, T.M., Swarzenski, C.M., 2007. Hydrology of tidal Gurnell, A., Petts, G., 2006. Trees as riparian engineers: the Tagliamento River, Italy.
freshwater forested wetlands of the southeastern United States. In: Conner, W.H., Earth Surface Processes and Landforms 31, 1558–1574.
Doyle, T.W., Krauss, K.W. (Eds.), Ecology of Tidal Freshwater Forested Gurnell, A.M., Angold, P.G., Goodson, J.M., Morrissey, I.P., Petts, G.E., Steiger, J.,
Wetlands of the Southeastern United States. Springer, New York, NY, 2004. Vegetation propagule dynamics and fluvial geomorphology. In: Bennett,
pp. 29–63. S.J., Simon, A. (Eds.), Riparian Vegetation and Fluvial Geomorphology.
Du Laing, G., Rinklebe, J., Vandecasteele, B., Meers, E., Tack, F.M.G., 2009. Trace American Geophysical Union, Washington, DC, pp. 209–210.
metal behaviour in estuarine and riverine floodplain soils and sediments: a Hamilton, S.K., Lewis, W.M., 1987. Causes of seasonality in the chemistry of a lake
review. Science of the Total Environment 407, 3972–3985. on the Orinoco River floodplain, Venezuela. Limnology and Oceanography 32,
Duberstein, J.A., Conner, W.H., 2009. Use of hummocks and hollows by trees in 1277–1290.
tidal freshwater forested wetlands along the Savannah River. Forest Ecology and Harner, M.J., Stanford, J.A., 2003. Differences in cottonwood growth between a
Management 258, 1613–1618. losing and a gaining reach of an alluvial floodplain. Ecology 84, 1453–1458.
Dwire, K., Kauffman, J., Baham, J., 2006. Plant species distribution in relation to Harvey, J.W., Conklin, M.H., Koelsch, R.S., 2003. Predicting changes in hydrologic
water-table depth and soil redox potential in montane riparian meadows. retention in an evolving semi-arid alluvial stream. Advances in Water Resources
Wetlands 26, 131–146. 26, 939–950.
Entry, J.A., 2000. Influence of nitrogen on cellulose and lignin mineralization in Harvey, J.W., Nuttle, W.K., 1995. Fluxes of water and solute in a coastal wetland
blackwater and redwater forested wetland soils. Biology and Fertility of Soils 31, sediment. 2. Effect of macropores on solute exchange with surface water.
436–440. Journal of Hydrology 164, 109–125.
Eppinga, M.B., Rietkerk, M., Borren, W., Lapshina, E.D., Bleuten, W., Wassen, M.J., Harvey, J.W., Schaffranek, R.W., Noe, G.B., Larsen, L.G., Nowacki, D.J., O’Connor,
2008. Regular surface patterning of peatlands: confronting theory with field data. B.L., 2009. Hydroecological factors governing surface water flow on a low-
Ecosystems 11, 520–536. gradient floodplain. Water Resources Research 45, W03421.
 Interactions among Hydrogeomorphology, Vegetation, and Nutrient Biogeochemistry in Floodplain Ecosystems 319

Heffernan, J.B., 2008. Wetlands as an alternative stable state in desert streams. King, R.S., Richardson, C.J., Urban, D.L., Romanowicz, E.A., 2004. Spatial
Ecology 89, 1261–1271. dependency of vegetation–environment linkages in an anthropogenically
Hefting, M., Clément, J.C., Dowrick, D., et al., 2004. Water table elevation controls influenced wetland ecosystem. Ecosystems 7, 75–97.
on soil nitrogen cycling in riparian wetlands along a European climatic gradient. Knösche, R., 2006. Organic sediment nutrient concentrations and their relationship
Biogeochemistry 67, 113–134. with the hydrological connectivity of floodplain waters (River Havel, NE
Hefting, M.M., Clement, J.-C., Bienkowski, P., et al., 2005. The role of vegetation Germany). Hydrobiologia 560, 63–76.
and litter in the nitrogen dynamics of riparian buffer zones in Europe. Ecological Knowlton, M.F., Jones, J.R., 1997. Trophic status of Missouri River floodplain lakes
Engineering 24, 465–482. in relation to basin type and connectivity. Wetlands 17, 468–475.
Heiler, G., Hein, T., Schiemer, F., Bornette, G., 1995. Hydrological connectivity and Kronvang, B., Andersen, I., Hoffmann, C., Pedersen, M., Ovesen, N., Andersen, H.,
flood pulses as the central aspects for the integrity of a river–floodplain system. 2007. Water exchange and deposition of sediment and phosphorus during
Regulated Rivers: Research and Management 11, 351–361. inundation of natural and restored lowland floodplains. Water, Air, and Soil
Hein, T., Baranyi, C., Herndl, G.J., Wanek, W., Schiemer, F., 2003. Allochthonous Pollution 181, 115–121.
and autochthonous particulate organic matter in floodplains of the River Larsen, L.G., Aumen, N., Bernhardt, C., 2011. Recent and historic drivers
Danube: the importance of hydrological connectivity. Freshwater Biology 48, of landscape change in the Everglades ridge, slough, and tree island
220–232. mosaic. Critical Reviews in Environmental Science and Technology 41(S1),
Helton, A.M., Poole, G.C., Meyer, J.L., et al., 2010. Thinking outside the channel: 344–381.
modeling nitrogen cycling in networked river ecosystems. Frontiers in Ecology Larsen, L.G., Harvey, J.W., Crimaldi, J.P., 2007. A delicate balance: Ecohydrological
and the Environment. http://dx.doi.org/10.1890/080211. feedbacks governing landscape morphology in a lotic peatland. Ecological
Hill, A.R., 2000. Stream chemistry and riparian zones. In: Jones, J.B., Monographs 77, 591–614.
Mulholland, P.J. (Eds.), Streams and Groundwaters. Academic Press, London, van der Lee, G.E.M., Olde Venterink, H., Asselman, N.E.M., 2004. Nutrient retention
pp. 83–107. in floodplains of the Rhine distributaries in The Netherlands. River Research and
Hogan, D.M., Walbridge, M.R., 2007. Urbanization and nutrient retention in Applications 20, 315–325.
freshwater riparian wetlands. Ecological Applications 17, 1142–1155. Leopold, L.B., Wolman, M.G., Miller, J.P., 1964. Fluvial Processes in
Holmes, R.M., Fisher, S.G., Grimm, N.B., 1994. Parafluvial nitrogen dynamics in a Geomorphology. W.H. Freeman and Company, San Francisco.
desert stream ecosystem. Journal of the North American Benthological Society Lewis, W.M., Hamilton, S.K., Lasi, M.A., Rodriguez, M., Saunders, J.F., 2000.
13, 468–478. Ecological determinism on the Orinoco floodplain. BioScience 50, 681–692.
Hopkinson, C.S., 1992. A comparison of ecosystem dynamics in freshwater Lockaby, B.G., Murphy, A.L., Somers, G.L., 1996. Hydroperiod influences on
wetlands. Estuaries and Coasts 15, 549–562. nutrient dynamics in decomposing litter of a floodplain forest. Soil Science
Hunter, R.G., Faulkner, S.P., Gibson, K.A., 2008. The importance of hydrology Society of America 60, 1267–1272.
in restoration of bottomland hardwood wetland functions. Wetlands 28, Lockaby, B.G., Walbridge, M.R., 1998. Biogeochemistry. In: Messina, M.G., Conner,
605–615. W.H. (Eds.), Southern Forested Wetlands: Ecology and Management. Lewis
Hupp, C.R., 1982. Stream-grade variation and riparian-forest ecology along Passage Publishers, Boca Raton, FL, pp. 149–172.
Creek, Virginia. Bulletin of the Torrey Botanical Club 109, 488–499. Lyons, J.B., Gorres, J.H., Amador, J.A., 1998. Spatial and temporal variability of
Hupp, C.R., 2000. Hydrology, geomorphology and vegetation of Coastal phosphorus retention in a riparian forest soil. Journal of Environmental Quality
Plain rivers in the south-eastern USA. Hydrological Processes 14, 27, 895–903.
2991–3010. Maine, M.A., Sune, N.L., Bonetto, C., 2004. Nutrient concentrations in the Middle
Hupp, C.R., Bornette, G., 2003. Vegetation as a tool in the interpretation of fluvial Parana River: effect of the floodplain lakes. Archiv fur Hydrobiologie 160,
geomorphic processes and landforms in humid temperate areas. In: Kondolf, M., 85–103.
Piegay, H. (Eds.), Tools in Fluvial Geomorphology. Wiley, Chichester, pp. Malard, F., Uehlinger, U., Zah, R., Tockner, K., 2006. Flood-pulse and riverscape
269–288. dynamics in a braided glacial river. Ecology 87, 704–716.
Hupp, C.R., Demas, C.R., Kroes, D.E., Day, R.H., Doyle, T.W., 2008. Recent Markewich, H.W., Pavich, M.J., Buell, G.R., 1990. Contrasting soils and landscapes
sedimentation patterns within the central Atchafalaya Basin, Louisiana. Wetlands of the Piedmont and Coastal Plain, eastern United States. Geomorphology 3,
28, 125–140. 417–447.
Hupp, C.R., Osterkamp, W.R., 1996. Riparian vegetation and fluvial geomorphic Martı́, E., Fisher, S.G., Schade, J.D., Grimm, N.B., 2000. Flood frequency and
processes. Geomorphology 14, 277–295. stream-riparian linkages in arid lands. In: Jones, J.B., Mulholland, P.J. (Eds.),
Hupp, C.R., Pierce, A.R., Noe, G.B., 2009. Floodplain geomorphic processes and Streams and Ground Waters. Academic Press, San Diego, CA, pp. 111–136.
environmental impacts of human alteration along Coastal Plain rivers. USA. McClain, M.E., Boyer, E.W., Dent, C.L., et al., 2003. Biogeochemical hot spots and
Wetlands 29, 413–429. hot moments at the interface of terrestrial and aquatic ecosystems. Ecosystems
Hynes, H.B.N., 1975. The stream and its valley. Verh. Internat. Verein Limnol. 19, 6, 301–312.
1–15. McKelvin, M.R., Hook, D.D., Rozelle, A.A., 1998. Adaptation of plants to flooding
James, W.F., Richardson, W.B., Soballe, D.M., 2008. Effects of residence time on and soil waterlogging. In: Messina, M., Conner, W. (Eds.), Southern Forested
summer nitrate uptake in Mississippi River flow-regulated backwaters. River Wetlands. Lewis Publishers, Boca Raton, FL, pp. 173–204.
Research and Applications 24, 1206–1217. Megonigal, J.P., Conner, W.H., Kroeger, S., Sharitz, R.R., 1997. Aboveground
Johnson, B.L., Richardson, W.B., Naimo, T.J., 1995. Past, present, and future production in southeastern floodplain forests: a test of the subsidy-stress
concepts in large river ecology. BioScience 45, 134–141. hypothesis. Ecology 78, 370–384.
Johnston, C.A., Bridgham, S.D., Schubauer-Berigan, J.P., 2001. Nutrient dynamics Megonigal, J.P., Hines, M.E., Visscher, P.T., 2004. Anaerobic metabolism: linkages
in relation to geomorphology of riverine wetlands. Soil Science Society of to trace gases and aerobic processes. In: Schlesinger, W.H. (Ed.),
America Journal 65, 557–577. Biogeochemistry. Elsevier-Pergamon, Oxford, pp. 317–424.
Johnston, C.A., Bubenzer, G.D., Lee, G.B., Madison, F.W., McHenry, J.R., 1984. Melillo, J.M., Aber, J.D., Muratore, J.F., 1982. Nitrogen and lignin control of
Nutrient trapping by sediment deposition in a seasonally flooded lakeside hardwood leaf litter decomposition dynamics. Ecology 63, 621–626.
wetland. Journal of Environmental Quality 13, 283–290. Mertes, L.A.K., 1997. Documentation and significance of the perirheic zone on
Johnston, C.A., Naiman, R.J., 1987. Boundary dynamics at the aquatic-terrestrial inundated floodplains. Water Resources Research 33, 1749–1762.
interface: the influence of beaver and geomorphology. Landscape Ecology 1, Messina, M.G., Conner, W.H., 1998. Southern Forested Wetlands: Ecology and
47–57. Management. CRC Press-Lewis Publishers, Boca Raton, FL.
Jones, R.H., Henson, K.O., Somers, G.L., 2000. Spatial, seasonal, and annual Mitsch, M.J., Dorge, C.L., Wiemhoff, J.R., 1979. Ecosystem dynamics and a
variation of fine root mass in a forested wetland. Journal of the Torrey Botanical phosphorus budget of an alluvial cypress swamp in southern Illinois. Ecology
Society 127, 107–114. 60, 1116–1124.
Junk, W.J., Bayley, P.B., Sparks, R.E., 1989. The flood pulse concept in Mitsch, M.J., Gosselink, J.G., 1993. Wetlands. Van Nostrand Reinhold, New York, NY.
river–floodplain systems. Canadian Special Publication of Fisheries and Aquatic Moggridge, H.L., Gurnell, A.M., 2009. Hydrological controls on the transport and
Sciences 106, 110–127. deposition of plant propagules within riparian zones. River Research and
Junk, W.J., Furch, K., 1993. A general review of tropical South American Applications 26, 512–527.
floodplains. Wetlands Ecology and Management 2, 231–238. Morris, J.T., Bradley, P.M., 1999. Effects of nutrient loading on the carbon balance
Kadlec, R.H., Wallace, S., 2009. Treatment Wetlands. Lewis Publishers, Boca Raton, FL. of coastal wetland sediments. Limnology and Oceanography 44, 699–702.
320 Interactions among Hydrogeomorphology, Vegetation, and Nutrient Biogeochemistry in Floodplain Ecosystems

Morris, J.T., Sundareshwar, P.V., Nietch, C.T., Kjerfve, B., Cahoon, D.R., 2002. Pollock, M.M., Naiman, R.J., Hanley, T.A., 1998. Plant species richness in riparian
Responses of coastal wetlands to rising sea level. Ecology 83, 2869–2877. wetlands – a test of biodiversity theory. Ecology 79, 94–105.
Moser, K., Ahn, C., Noe, G.B., 2007. Characterization of microtopography Poole, G.C., 2010. Stream hydrogeomorphology as a physical science basis for
and its influence on vegetation patterns in created wetlands. Wetlands 27, advances in stream ecology. Journal of the North American Benthological
1081–1097. Society 29, 12–25.
Moser, K.F., Ahn, C., Noe, G.B., 2009. The influence of microtopography on soil Poole, G.C., Stanford, J.A., Frissell, C.A., Running, S.W., 2002. Three-dimensional
nutrients in created mitigation wetlands. Restoration Ecology 17, 641–651. mapping of geomorphic controls on flood-plain hydrology and connectivity from
Mouw, J.E.B., Stanford, J.A., Alaback, P.B., 2009. Influences of flooding and aerial photos. Geomorphology 48, 329–347.
hyporheic exchange on floodplain plant richness and productivity. River Preiner, S., Drozdowski, I., Schagerl, M., Schiemer, F., Hein, T., 2008. The
Research and Applications 25, 929–945. significance of side-arm connectivity for carbon dynamics of the River Danube,
Nadelhoffer, K.J., 2000. Research review: the potential effects of nitrogen Austria. Freshwater Biology 53, 238–252.
deposition on fine-root production in forest ecosystems. New Phytologist 147, Rheinhardt, R.D., 2007. Tidal freshwater swamps of a lower Chesapeake Bay
131–139. subestuary. In: Conner, W.H., Doyle, T.W., Krauss, K.W. (Eds.), Ecology of Tidal
Naiman, R., Bechtold, J., Beechie, T., Latterell, J., Van Pelt, R., 2010. A process- Freshwater Forested Wetlands of the Southeastern United States. Springer, New
based view of floodplain forest patterns in coastal river valleys of the Pacific York, NY, pp. 161–182.
Northwest. Ecosystems 13, 1–31. Richardson, C.J., 1985. Mechanisms controlling phosphorous retention capacity in
Naiman, R.J., Bilby, R.E., Schindler, D.E., Helfield, J.M., 2002. Pacific salmon, freshwater wetlands. Science 228, 1424–1426.
nutrients, and the dynamics of freshwater and riparian ecosystems. Ecosystems Richardson, W.B., Strauss, E.A., Bartsch, L.A., Monroe, E.M., Cavanaugh, J.C.,
5, 399–417. Vingum, L., Soballe, D.M., 2004. Denitrification in the Upper Mississippi River:
Naiman, R.J., Décamps, H., 1997. The ecology of interfaces: riparian zones. Annual rates, controls, and contribution to nitrate flux. Canadian Journal of Fisheries
Review of Ecology and Systematics 28, 621–658. and Aquatic Science 61, 1102–1112.
Naiman, R.J., Décamps, H., McClain, M.E., 2005. Riparia. Academic Press, San Rietkerk, M., van de Koppel, J., 2008. Regular pattern formation in real ecosystems.
Diego, CA. Trends in Ecology and Evolution 23, 169–175.
Naiman, R.J., Johnston, C.A., Kelley, J.C., 1988. Alteration of North American Ross, K.M., Hupp, C.R., Howard, A.D., 2004. Sedimentation in floodplains of
streams by beaver. BioScience 38, 753–762. selected tributaries of the Chesapeake Bay. In: Bennett, S.J., Simon, A. (Eds.),
National Research Council, 2002. Riparian Areas. National Academy Press, Riparian Vegetation and Fluvial Geomorphology. American Geophysical Union,
Washington, DC. Washington, DC, pp. 187–208.
Noe, G.B., Childers, D.L., Jones, R.D., 2001. Phosphorus biogeochemistry and the Sabater, S., Butturini, A., Clement, J.-C., et al., 2003. Nitrogen removal by riparian
impact of phosphorus enrichment: why is the Everglades so unique? Ecosystems buffers along a European climatic gradient: patterns and factors of variation.
4, 603–624. Ecosystems 6, 20–30.
Noe, G.B., Hupp, C.R., 2005. Carbon, nitrogen, and phosphorus accumulation in Sánchez-Carrillo, S., Álvarez-Cobelas, M., Angeler, D.G., 2001. Sedimentation in the
floodplains of Atlantic Coastal Plain rivers, USA. Ecological Applications 15, semi-arid freshwater wetland Las Tablas de Daimil (Spain). Wetlands 21,
1178–1190. 112–124.
Noe, G.B., Hupp, C.R., 2007. Seasonal variation in nutrient retention during Schilling, E.B., Lockaby, B.G., 2006. Relationships between productivity and nutrient
inundation of a short-hydroperiod floodplain. River Research and Applications circulation within two contrasting southeastern U.S. floodplain forests. Wetlands
23, 1088–1101. 26, 181–192.
Noe, G.B., Hupp, C.R., 2009. Retention of riverine sediment and nutrient loads by Schumm, S.A., 1985. Patterns of alluvial rivers. Annual Review of Earth and
Coastal Plain floodplains. Ecosystems 12, 728–746. Planetary Sciences 13, 5–27.
Nyman, J.A., Walters, R.J., Delaune, R.D., Patrick, W.H., 2006. Marsh vertical Sedell, J.R., Richey, J.E., Swanson, F.J., 1989. The river continuum concept: a
accretion via vegetative growth. Estuarine, Coastal and Shelf Science 69, basis for the expected ecosystem behavior of very large rivers? Canadian
370–380. Special Publication of Fisheries and Aquatic Sciences 106, 49–55.
Odum, E.P., Finn, J.T., Eldon, H.F., 1979. Perturbation theory and the subsidy-stress Shankman, D., 1993. Channel migration and vegetation patterns in the Southeastern
gradient. BioScience 29, 349–352. Coastal Plain. Conservation Biology 7, 176–183.
Olde Venterink, H., Pieterse, N.M., Belgers, J.D.M., Wassen, M.J., de Ruiter, P.C., Sharitz, R.R., Mitsch, W.J., 1993. Southern floodplain forests. In: Martin, W.H.,
2002. N, P, and K budgets along nutrient availability and productivity gradients Boyce, S.G., Esternacht, A.C. (Eds.), Biodiversity of the Southeastern
in wetlands. Ecological Applications 12, 1010–1026. United States: Lowland Terrestrial Communities. Wiley, New York, NY, pp.
Olde Venterink, H., Vermaat, J.E., Pronk, M., et al., 2006. Importance of sediment 311–372.
deposition and denitrification for nutrient retention in floodplain wetlands. Simon, A., Bennett, S.J., Neary, V.S., 2004. Riparian vegetation and fluvial
Applied Vegetation Science 9, 163–174. geomorphology: problems and opportunities. In: Bennett, S.J., Simon, A. (Eds.),
Osterkamp, W.R., 1998. Processes of fluvial island formation, with examples Riparian Vegetation and Fluvial Geomorphology. American Geophysical Union,
from Plum Creek, Colorado and Snake River, Idaho. Wetlands 18, Washington, DC, pp. 1–10.
530–545. Smith, J.D., 2004. The role of riparian shrubs in preventing floodplain
Osterkamp, W.R., Hupp, C.R., 2010. Fluvial processes and vegetation – glimpses unraveling along the Clark Fork of the Columbia River in the Deer Lodge
of the past, the present, and perhaps the future. Geomorphology 116, Valley, Montana. In: Bennett, S.J., Simon, A. (Eds.), Riparian Vegetation and
274–285. Fluvial Geomorphology. American Geophysical Union, Washington, DC, pp.
van der Peijl, M.J., Verhoeven, J.T.A., 2000. Carbon, nitrogen, and phosphorus 71–86.
cycling in river marginal wetlands; a model examination of landscape Sparks, R.E., 1995. Need for ecosystem management of large rivers and their
geochemical flows. Biogeochemistry 50, 45–71. floodplains. BioScience 45, 168–182.
Petts, G.E., Large, A.R.G., Greenwood, M.T., Bickerton, M.A., 1992. Floodplain Spink, A., Sparks, R.E., 1998. van Oorschot, M., Verhoeven, J.T.A. Nutrient
assessment for restoration and conservation: linking hydrogeomorphology and dynamics of large river floodplains. Regulated Rivers: Research and Management
ecology. In: Carling, P.A., Petts, G.E. (Eds.), Lowland Floodplain Rivers: 14, 203–216.
Geomorphological Perspectives. Wiley, West Sussex, pp. 217–234. Stanford, J.A., Lorang, M.S., Hauer, F.R., 2005. The shifting habitat mosaic of
Phillips, J.D., Slattery, M.C., 2006. Sediment storage, sea level, and sediment river ecosystems. Verhandlungen des Internationalen Verein Limnologie 29,
delivery to the ocean by coastal plain rivers. Progress in Physical Geography 123–136.
30, 513–530. Stanford, J.A., Ward, J.V., 1993. An ecosystem perspective of alluvial rivers:
Pinay, G., Haycock, N.E., Ruffinoni, C., Holmes, R.M., 1994. The role of connectivity and the hyporheic corridor. Journal of the North American
denitrification in nitrogen removal in river corridors. In: Mitsch, W.J. (Ed.), Benthological Society 12, 48–60.
Global Wetlands: Old World and New. Elsevier, Dordrecht, pp. 107–116. Stanley, E.H., Fisher, S.G., Grimm, N.B., 1997. Ecosystem expansion and
Poff, N.L., Allan, J.D., Bain, M.B., et al., 1997. The natural flow regime: a paradigm contraction in streams. BioScience 47, 427–435.
for river conservation and restoration. BioScience 47, 769–784. Stanturf, J.A., Schoenholtz, S.H., 1998. Soils and landforms of southern forested
Poff, N.L., Richter, B.D., Arthington, A.H., et al., 2010. The ecological limits of wetlands. In: Messina, M.G., Conner, W.A. (Eds.), Southern Forested Wetlands:
hydrologic alteration (ELOHA): a new framework for developing regional Ecology and Management. CRC Press-Lewis Publishers, Boca Raton, FL, pp.
environmental flow standards. Freshwater Biology 55, 147–170. 123–147.
 Interactions among Hydrogeomorphology, Vegetation, and Nutrient Biogeochemistry in Floodplain Ecosystems 321

Steiger, J., Gurnell, A.M., 2002. Spatial hydrogeomorphological influences on Vannote, R.L., Minsall, G.W., Cummins, K.W., Sedell, J.R., Cushing, C.E., 1980.
sediment and nutrient deposition in riparian zones: observations from the The river continuum concept. Canadian Journal of Fisheries and Aquatic
Garonne River. France. Geomorphology 49, 1–23. Sciences 37, 130–137.
Steiger, J., Tabacchi, E., Dufour, S., Corenblit, D., Peiry, J.L., 2005. Vidon, P., Allan, C., Burns, D., et al., 2010. Hot spots and hot moments in riparian
Hydrogeomorphic processes affecting riparian habitat within alluvial channel- zones: potential for improved water quality management. JAWRA Journal of the
floodplain river systems: a review for the temperate zone. River Research and American Water Resources Association 46, 278–298.
Applications 21, 719–737. Vincent, K., Friedman, J., Griffin, E., 2009. Erosional consequence of saltcedar
Stoeckel, D.M., Miller-Goodman, M.S., 2001. Seasonal nutrient dynamics of control. Environmental Management 44, 218–227.
forested floodplain soil influenced by microtopography and depth. Soil Science Walbridge, M.R., Struthers, J.P., 1993. Phosphorus retention in non-tidal palustrine
Society of America Journal 65, 922–931. forested wetlands of the mid-Atlantic region. Wetlands 13, 84–94.
Stribling, J., Cornwell, J., Glahn, O., 2007. Microtopography in tidal marshes: Wang, N., Mitsch, W.J., 2000. A detailed ecosystem model of phosphorus dynamics
ecosystem engineering by vegetation? Estuaries and Coasts 30, 1007–1015. in created riparian wetlands. Ecological Modelling 126, 101–130.
Stromberg, J.C., Tiller, R., Richter, B., 1996. Effects of groundwater decline on Ward, J.V., 1989a. The four-dimensional nature of lotic ecosystems. Journal of the
riparian vegetation of semiarid regions: the San Pedro, Arizona. Ecological North American Benthological Society 8, 2–8.
Applications 6, 113–131. Ward, J.V., 1989b. Riverine–wetland interactions. In: Sharitz, R.R., Gibbons, J.W.
Stubblefield, A.P., Escobar, M.I., Larsen, E.W., 2006. Retention of suspended (Eds.), Freshwater Wetlands and Wildlife. USDOE Office of Scientific and
sediment and phosphorus on a freshwater delta, South Lake Tahoe, California. Technical Information, Oak Ridge, pp. 385–400.
Wetlands Ecology and Management 14, 287–302. Ward, J.V., 1998. Riverine landscapes: biodiversity patterns, disturbance regimes,
Sweet, W.V., Geratz, J.W., 2003. Bankfull hydraulic geometry relationships and and aquatic conservation. Biological Conservation 83, 269–278.
recurrence intervals for North Carolina’s coastal plain. Journal of the American Ward, J.V., Stanford, J.A., 1995. The serial discontinuity concept: extending the
Water Resources Association 39, 861–871. model to floodplain rivers. Regulated Rivers: Research and Management 10,
Tabacchi, E., Lambs, L., Guilloy, H., Planty-Tabacchi, A.-M., Muller, E., Décamps, 159–168.
H., 2000. Impacts of riparian vegetation on hydrological processes. Hydrological Ward, J.V., Tockner, K., Edwards, P.J., et al., 1999. A reference river system for the
Processes 14, 2959–2976. Alps: the ‘Fiume Tagliamento’. Regulated Rivers: Research and Management 15,
Thoms, M.C., Parsons, M., 2002. Eco-geomorphology: an interdisciplinary 63–75.
approach to river science. In: Dyer, F.J., Thoms, M.C., Olley, J.M. (Eds.), The Wassen, M.J., Peeters, W.H.M., Olde Venterink, H., 2003. Patterns in vegetation,
Structure, Function and Management Implications of Fluvial Sedimentary hydrology, and nutrient availability in an undisturbed river floodplain in Poland.
Systems. International Association of Hydrological Sciences, Wallingford, Plant Ecology 165, 27–43.
pp. 113–119. Werner, K., Zedler, J., 2002. How sedge meadow soils, microtopography, and
Thorp, J.H., Thoms, M.C., Delong, M.D., 2006. The riverine ecosystem synthesis: vegetation respond to sedimentation. Wetlands 22, 451–466.
biocomplexity in river networks across space and time. River Research and Wharton, C.H., Kitchens, W.M., Pendleton, E.C., Sipe, T.W., 1982. The Ecology of
Applications 22, 123–147. Bottomland Hardwood Swamps of the Southeast: A Community Profile. US Fish
Tockner, K., Lorang, M.S., Stanford, J.A., 2010. River flood plains are model and Wildlife Service, Washington, DC.
ecosystems to test general hydrogeomorphic and ecological concepts. River Whigham, D.F., Chitterling, C., Palmer, B., 1988. Impacts of freshwater wetlands on
Research and Applications 26, 76–86. water quality: a landscape perspective. Environmental Management 12, 663–671.
Tockner, K., Malard, F., Ward, J.V., 2000. An extension of the flood pulse concept. Wigand, C., Brennan, P., Stolt, M., Holt, M., Ryba, S., 2009. Soil respiration rates
Hydrological Processes 14, 2861–2883. in coastal marshes subject to increasing watershed nitrogen loads in southern
Tockner, K., Pennetzdorfer, D., Reiner, N., Schiemer, F., Ward, J.V., 1999. New England. USA. Wetlands 29, 952–963.
Hydrological connectivity, and the exchange of organic matter and nutrients in a Wright, R.B., Lockaby, B.G., Walbridge, M.R., 2001. Phosphorus availability in an
dynamic river–floodplain system (Danube, Austria). Freshwater Biology 41, artificially flooded southeastern floodplain forest soil. Soil Science Society of
521–535. America Journal 65, 1293–1302.
Tockner, K., Stanford, J.A., 2002. Riverine flood plains: present state and future Wynn, T., Mostaghimi, S., 2006. The effects of vegetation and soil type on
trends. Environmental Conservation 29, 308–330. streambank erosion, Southwestern Virginia, USA. Journal of the American Water
Triska, F.J., Duff, J.H., Avanzino, R.J., 1993. Patterns of hydrological exchange Resources Association 42, 69–82.
and nutrient transformation in the hyporheic zone of a gravel-bottom Yarbro, L.A., 1983. The influence of hydrologic variations on phosphorus cycling and
stream: examining terrestrial-aquatic linkages. Freshwater Biology 29, 259–274. retention in a swamp stream ecosystem. In: Fontaine, T.D., Bartell, S.M. (Eds.),
Valett, H.M., Fisher, S.G., Grimm, N.B., Camill, P., 1994. Vertical hydrologic Dynamics of Lotic Ecosystems. Ann Arbor Science, Ann Arbor, pp. 223–245.
exchange and ecological stability of a desert stream ecosystem. Ecology 75,
548–560.

Biographical Sketch

Gregory B. Noe obtained his honors BS in biology from Virginia Polytechnic Institute and State University and
PhD in ecology from the Joint Doctoral Program of San Diego State University and University of California,
Davis. He is a research ecologist with the National Research Program of the US Geological Survey. His interests
focus on wetland ecosystem ecology, including the interactive influences of hydrology, geomorphology, climate,
and biology on nitrogen and phosphorus biogeochemistry in fluvial ecosystems, as well as plant community
ecology and restoration ecology.