DIFFERENTIAL RESPONSE OF WIND AND WATER EROSION UNDER
CLIMATIC EXTREMES AND ALTERNATE LAND MANAGEMENT PRACTICES
By
Jason Paul Field
A Dissertation Submitted to the Faculty of the
SCHOOL OF NATURAL RESOURCES AND THE ENVIRONMENT
In Partial Fulfillment of the Requirements
For the Degree of
DOCTOR OF PHILOSOPHY
WITH A MAJOR IN WATERSHED MANAGEMENT
In the Graduate College
THE UNIVERSITY OF AIRZONA
2009
2
THE UNIVERSITY OF ARIZONA
GRADUATE COLLEGE
As members of the Dissertation Committee, we certify that we have read the dissertation
prepared by Jason Field
entitled Differential Responses of Wind and Water Erosion Under Climatic Extremes and
Alternate Land Management Practices
and recommend that it be accepted as fulfilling the dissertation requirement for the
Degree of Doctor of Philosophy
________________________________________________________________________________
Date: 11/05/09
David D. Breshears
________________________________________________________________________________
Date: 11/05/09
Jeffrey J. Whicker
________________________________________________________________________________
Date: 11/05/09
Steven R. Archer
________________________________________________________________________________
Date: 11/05/09
Travis E. Huxman
Final approval and acceptance of this dissertation is contingent upon the candidate’s
submission of the final copies of the dissertation to the Graduate College.
I hereby certify that I have read this dissertation prepared under my direction and
recommend that it be accepted as fulfilling the dissertation requirement.
______________________________________________________ Date: 11/05/09
Dissertation Director: David D. Breshears
3
STATEMENT BY AUTHOR
This dissertation has been submitted in partial fulfillment of requirements for an advance
degree at the University of Arizona and is deposited in the University Library to be made
available to borrowers under rules of the Library.
Brief quotations from this dissertation are allowable without special permission, provided
that accurate acknowledgment of source is made. Requests for permission for extended
quotation from or reproduction of this manuscript in whole or in part may be granted by
the head of the major department of the Dean of the Graduate College when in his or her
judgment the proposed use of the material is in the interests of scholarship. In all other
instances; however, permission must be obtained from the author.
SIGNED: Jason Field
4
ACKNOWLEDGEMENTS
I wish to express my gratitude to those who have aided in the development of this
dissertation. I would like to thank my dissertation director, Dr. David D. Breshears, for
his support and endless encouragement and for providing me the opportunity to be a part
of his research group. Special thanks to all of my committee members, Dr. Jeffrey J
Whicker, Dr. Steven R. Archer, and Dr. Travis E. Huxman, for their invaluable
contributions and persistent guidance throughout my graduate studies. Thanks also to Dr.
Guy McPherson and Chris McDonald for implementing the experimental design and the
burning and grazing treatments. I would also like to thank Chris Zou for his assistance in
the initial development and design of the study and for assisting with the sample
collection. I would also like to express my sincere thanks to my wife, family and friends
for their unwavering support and confidence in me.
This work was primarily supported by the U.S. Department of Agriculture Cooperative
State Research, Education, and Extension Service (CSREES 2005-38420-15809) and the
National Science Foundation (NSF-DEB 0816162).
5
TABLE OF CONTENTS
ABSTRACT.....................................................................................................................
8
INTRODUCTION ...........................................................................................................
9
PRESENT STUDY.......................................................................................................... 14
REFERENCES ................................................................................................................ 20
APPENDIX A: THE ECOLOGY OF DUST: LOCAL- TO GLOBAL- SCALE
TERRESTRIAL PERSPECTIVES.................................................................................. 27
Introduction.................................................................................................................. 28
A Wind Erosion Primer ............................................................................................... 30
Plant-Interspace Scale.................................................................................................. 32
Regional- to Global-Scale Consequences.................................................................... 35
Projections and Implications........................................................................................ 38
References.................................................................................................................... 42
Figures.......................................................................................................................... 49
APPENDIX B: DUST AMPLIFICATION AND CAPTURE AMONG VEGETATION
PATCH TYPES: SUPPORT FOR UNDERLYING DESERTIFICATION
ASSUMPTIONS ............................................................................................................. 56
Introduction.................................................................................................................. 57
Materials and Methods................................................................................................. 59
Results.......................................................................................................................... 64
Discussion .................................................................................................................... 66
References.................................................................................................................... 71
Figures.......................................................................................................................... 78
6
TABLE OF CONTENTS – Continued
APPENDIX C: A CONCEPTUAL FRAMEWORK FOR DRYLAND AEOLIAN
SEDIMENT TRANSPORT ALONG THE GRASSLAND-FOREST CONTINUUM:
EFFECTS OF WOODY PLANT CANOPY COVER AND DISTURBANCE.............. 83
Introduction.................................................................................................................. 84
Compilation of Measurements..................................................................................... 88
Boundary Layer Flows and Land Surface Characteristics........................................... 94
A New Conceptual Framework ...................................................................................100
Conclusions..................................................................................................................103
References....................................................................................................................105
Figures..........................................................................................................................115
Tables...........................................................................................................................122
APPENDIX D: BACKGROUND DUST EMISSION FOLLOWING GRASSLAND
FIRE: A SNAPSHOT ACROSS THE PARTICLE-SIZE SPECTRUM HIGHLIGHTS
HOW HIGH-RESOULUTION MEASUREMENTS ENHANCE DETECTION ..........124
Introduction..................................................................................................................125
Materials and Methods.................................................................................................128
Results..........................................................................................................................131
Discussion ....................................................................................................................133
References....................................................................................................................135
Figures..........................................................................................................................140
APPENDIX E: TOWARD A MORE HOLISTIC PERSPECTIVE OF SOIL EROSION:
WHY AEOLIAN RESEARCH NEEDS TO EXPLICITY CONSIDER FLUVIAL
PROCESSES AND INTERACTIONS............................................................................146
Introduction..................................................................................................................147
Environmental and Scale-Dependencies......................................................................152
7
TABLE OF CONTENTS – Continued
Research Output and Resource Investment .................................................................158
Addressing Emerging Challenges................................................................................162
References....................................................................................................................164
Figures..........................................................................................................................174
Tables...........................................................................................................................179
APPENDIX F: A DUSTIER FUTURE UNDER GLOBAL-CHANGE-TYPE
EROSION ........................................................................................................................180
Introduction..................................................................................................................181
Methods........................................................................................................................183
Results..........................................................................................................................185
Discussion ....................................................................................................................186
References....................................................................................................................190
Figures..........................................................................................................................193
APPENDIX G: HOW GRAZING AND BURNING DIFFERENTIALLY ALTER
WIND AND WATER EROSION: IMPLICATIONS FOR RANGELANDS ................196
Introduction..................................................................................................................197
Materials and Methods.................................................................................................201
Results..........................................................................................................................204
Discussion ....................................................................................................................207
References....................................................................................................................210
Figures..........................................................................................................................219
8
ABSTRACT
Wind erosion and associated dust emissions play a fundamental role in many
ecological processes, yet most ecological studies do not explicitly consider dust-driven
processes despite the growing body of evidence suggesting that wind erosion is a key
driver of land surface dynamics and many other environmentally relevant processes such
as desertification. This study provides explicit support for a pervasive underlying but
untested desertification hypothesis by showing that at the vegetation patch scale shrubs
are significantly more efficient at capturing wind-blown sediment and other resources
such as nutrients than grasses and that this difference is amplified following disturbance.
At the landscape scale, the spacing and shape of woody plants were found to be a major
determinant of dryland aeolian sediment transport processes in grasslands, shrublands,
woodlands and forests, particularly following disturbance. This study also found that
disturbance such as fire can have a significant influence on background dust emissions,
which can have important consequences for many basic ecological and hydrological
processes. Potential interactions between aeolian and fluvial processes were also
evaluated in this study, and a new conceptual framework was developed that highlights
important differences and similarities between the two processes as a function of scaledependencies, mean annual precipitation, and disturbance. This study also explicitly
evaluates the effect of climatic extremes and alternate land management practices on the
absolute and relative magnitudes of wind and water erosion. Notably, results indicate
that wet/dry climatic extremes and grazing can increase the wind-to-water erosion ratio,
whereas burning disproportionally increases water erosion relative to wind erosion.
9
INTRODUCTION
Aeolian and fluvial processes associated with soil erosion present widespread and
substantial ecological challenges in environmental science and management (Pye, 1987;
Toy et al., 2002; Peters et al., 2006; CCSP, 2008). Although the ecological importance of
soil erosion associated with water-driven sediment transport has been widely studied, the
ecological significance of aeolian processes and associated dust emissions has been
studied to a much lesser extent despite its global implications. The consequences of
aeolian transport processes have important global implications (Cooke et al., 1993;
Goudie, 2008) and are perhaps most evident in major dust storms across regionally
degraded landscapes, as experienced throughout much of North America during the
1930s Dust Bowl era (Worster, 1979; Peters et al., 2007, 2008) and in China in
association with degraded northern drylands (Chepil, 1949; Shao and Shao, 2001).
Although dust deposition in some regions can have important beneficial effects, such as
the transport of nitrogen, phosphorous, and other essential nutrients to aquatic and
terrestrial systems (Swap et al., 1992; Chadwick et al., 1999; Neff et al., 2008), the
detachment and removal of wind-blown sediment from source areas can significantly
lower soil fertility and water holding capacity (Lal et al., 2003; Li et al., 2007, 2008),
alter biogeochemical processes (Schlesinger et al., 1990; Jickells et al., 2005), and
increase land surface inputs of dust to the atmosphere (Gillette and Passi, 1988; Tegen
and Fung, 1994; Reynolds et al., 2001).
The catastrophic impacts of the North American Dust Bowl of the 1930s, as well
as the Sirocco dust events of 1901-1903, led to a widespread surge in interest in aeolian
10
processes and a significant increase in the number of publications in aeolian research
(Stout et al., 2009), many of which focused on basic aeolian transport processes or soil
conservation through improved land management. This interest subsequently benefitted
from novel, quantitative advances developed by Bagnold (1941), which have served as a
basic foundation for much of the current understanding of aeolian transport processes.
The aeolian research community has been growing steadily since Bagnold’s (1941)
classic work on aeolian entrainment and founding studies of the geomorphology of dune
fields (Stout et al., 2009). Aeolian transport is now clearly recognized as critical to land
surface dynamics for the environmental and geosciences research community and by
many within the resource management community (Peters et al., 2006; CCSP, 2008).
Although aeolian transport is generally recognized as important, current
understanding and focus on aeolian processes is often in isolation from the other primary
driver of land surface dynamics: fluvial transport (Heathcote, 1983; Baker et al., 1995;
Breshears et al., 2003; Visser et al., 2004). More specifically, researchers and
practitioners in soil conservation generally segregate into one of two disciplines, those
focusing on wind erosion or those focusing on water erosion. Many geomorphological
studies focus on inferring relative importance of aeolian vs. fluvial processes in soil
profiles, but these studies do not directly quantify concurrent, co-located rates of both
wind and water erosion. Although both wind and water erosion have contributed close to
one billion tons of soil loss per year within the United States (NRCS, 2000a, 2000b), and
they operate on many similar fundamentals, there are critical differences between the two
types of processes that drive this separation (Toy et al., 2002; Breshears et al., 2003;
11
Visser et al., 2004). These include major differences in the density of the transport fluid
(water verses air), directionality of sediment and dust transport, temporal scales of the
erosion events, and spatial scales of the impact (from localized to global). Though
research on aeolian transport has generally proceeded in isolation from fluvial transport,
there are numerous reasons to re-evaluate the interrelationships between wind and water
erosion and associated aeolian and fluvial processes (Heathcote, 1983; Baker et al., 1995;
Breshears et al., 2003; Bullard and McTainsh, 2003; Visser et al., 2004) because such
interrelationships may have important environmental and ecological consequences
(Aguiar and Sala 1999; Peters et al., 2006; Ravi et al., 2007b). The degree and manner in
which aeolian and fluvial transport processes are interrelated could also have important
implications for relative investments in research and soil conservation for controlling
erosion of both types. This issue is particularly pressing given the growing
environmental challenges related to maintaining agricultural productivity, preventing
ecosystem degradation, and adapting to the projected impacts of global climate change
(Lal et al., 2003; Nearing et al., 2005; CCSP, 2008).
The potential for soil erosion and land degradation due to synergistic effects of
aeolian and fluvial transport may well far exceed that of either type of process alone
(Bullard and Livingstone, 2002). Aeolian and fluvial transport processes can degrade
ecosystems and accelerate desertification (Schlesinger et al., 1990; Belnap, 1995; Peters
et al., 2006; Okin et al., 2009), and both processes can contribute substantially to total
erosion (Breshers et al., 2003; Bullard and McTainsh, 2003; Visser et al., 2004).
Combined, the effects of aeolian- and fluvial- driven soil loss have resulted in moderate
12
to severe soil degradation throughout much of the world’s arable land (Oldeman et al.,
1990; Pimentel, 1993). Globally, perhaps as much as one-third of all arable land has
experienced accelerated rates of erosion that undermine long-term productivity (Brown,
1981; USDA, 2006). It is clear that a majority of lands, what ever the use pattern, are
subject to both aeolian and fluvial transport processes and that these processes operate
together to redistribute soil and other critical resources, such as nutrients, organic debris,
seeds, and water (Schlesinger et al., 1990; Aguiar and Sala, 1999; Bullard and McTainsh,
2003). Interactions between aeolian and fluvial processes can have a large influence on
the transport and deposition of fine sediment and sand-sized material in dryland
environments. For example, aeolian entrainment from lake beds, river beds, and flood
plains can transport fluvial sediment long distances and subsequently deposit it as aeolian
material, at which point either fluvial or aeolian processes can further redistribute the
sediment, thus increasing the potential for interactions between aeolian and fluvial
processes (Bullard and Livingstone, 2002). Additional examples of aeolian-fluvial
interactions include glaciogenic outwash in major drainage systems supplying silt for
aeolian entrainment to form loess (Sun, 2002; Muhs et al., 2008), raindrop destruction of
soil aggregates to yield particle sizes suitable for deflation (Cornelis et al., 2004;
Chappell et al., 2005; Erpul et al., 2009), micro-topography formation beneath plant
canopies (Schlesinger et al., 1990; Ravi et al., 2007a), and reworking of hillslope loess to
form pedisediment (Ruhe et al., 1967).
Despite the importance of wind and water erosion over vast areas, field studies
comparing the absolute and relative magnitudes of both types of erosion are largely
13
lacking (Breshears et al., 2003; Visser et al., 2004). Although conceptual models and
limited field measurements suggest that both wind and water erosion can be of similar
magnitude in many environments (Kirkby, 1980; Baker et al., 1995; Valentin, 1996;
Breshears et al., 2003), substantial uncertainty remains about the relative magnitudes of
the two types of erosion and how they interrelate with each other because few studies
explicitly evaluate both processes. In addition, an integrated perspective of how these
processes contribute to total erosion and how they vary with scale and the degree to
which they interact is lacking. Given that recent field measurements and erosion models
indicate that both processes contribute substantially to total erosion (NRCS, 2000a,
2000b; Breshears et al., 2003) and that the ways in which they interact are being
considered more directly (Bullard and Livingstone, 2002; Bullard and McTainsh, 2003;
Visser et al., 2004), a key challenge that lies before the aeolian and fluvial research
communities is to develop a more integrated perspective of aeolian-fluvial dynamics.
The uncertainty about the relative magnitudes of aeolian and fluvial transport processes
needs to be addressed to develop more effective land management and could be useful in
guiding future deployment of resources.
14
PRESENT STUDY
The methods and design of this study, along with detailed results and conclusions,
are presented in the papers appended to this dissertation. The most important findings in
this document are presented in the summary below.
Soil erosion is driven by not only aeolian but also fluvial transport processes, yet
these two types of processes are usually studied independently, thereby precluding
effective assessment of overall erosion, potential interactions between the two drivers,
and their relative sensitivities to projected changes in climate and land use. The first part
of this document summarizes the fundamental role that wind erosion and associated dust
emissions play in many ecological processes and provide important biogeochemical
connectivity at scales from individual plants up to the entire globe. Yet, most ecological
studies do not explicitly consider dust-driven processes, perhaps because most relevant
research on aeolian (wind-driven) processes has been presented in a geosciences rather
than an ecological context.
Appendix A provides a general overview of the ecological importance of dust.
This paper examines complex interactions between wind erosion and ecosystem
dynamics from the scale of plants and surrounding space to regional and global scales,
and highlight specific examples of how disturbance affects these interactions and their
consequences. It is likely that changes in climate and intensification of land use will lead
to increased dust production from many drylands. To address these issues, environmental
scientists, land managers, and policy makers need to consider wind erosion and dust
emission more explicitly in resource management decisions.
15
Appendix B provides an overview the underlying assumption associated with
most conceptual models of desertification, which is that when herbaceous cover is
reduced, increased erosion from bare patches is redistributed to shrub canopy patches,
resulting in self-reinforcing “islands of fertility”. Notably, however, this underlying
assumption at the scale of vegetation patches has not been explicitly tested with direct
measurements of aeolian processes. This desertification hypothesis was directly by
measuring aeolian transport into and out of bare-, herbaceous-, and shrub-dominated
patch types in a semiarid ecosystem using BSNE samplers both for simulated and natural
dust events, as well as in response to simulated disturbance. Vegetation patch types
differed as expected, with shrub patches capturing more dust than herbaceous patches,
and importantly, bare patches serving as amplified dust sources with more dust leaving
than entering a patch; simulated disturbances that removed herbaceous vegetation and
disturbed the soil surface confirmed hypothesized transformations associated with
desertification. The results provide explicit support for a pervasive underlying but
untested desertification hypothesis and can aide in more effectively managing vegetation
patches and soil surface stability.
Appendix C evaluates potential trends in aeolian sediment transport as a function
of woody plant cover, estimates of aeolian sediment transport from recently published
studies, in concert with rates from four additional locations (two grassland and two
woodland sites) are reported here. The synthesis of these reports leads to the
development of a new conceptual framework for aeolian sediment transport in dryland
ecosystems along the grassland-forest continuum. The findings suggest that: (1) for
16
relatively undisturbed ecosystems, shrublands have inherently greater aeolian sediment
transport because of wake interference flow associated with intermediate levels of density
and spacing of woody plants; and (2) for disturbed ecosystems, the upper bound for
aeolian sediment transport decreases as a function of increasing amounts of woody plant
cover because of the effects of the height and density of the canopy on airflow patterns
and ground cover associated with woody plant cover. Consequently, aeolian sediment
transport following disturbance spans the largest range of rates in grasslands and
associated systems with no woody plants (e.g., agricultural fields), an intermediate range
in shrublands, and a relatively small range in woodlands and forests.
Appendix D focuses on dust emissions in unburned and burned semiarid
grassland using four different instruments spanning different combinations of temporal
resolution and particle-size spectrum: Big Springs Number Eight (BSNE) and Sensit
instruments for larger saltating particles, DustTrak instruments for smaller suspended
particles, and Total Suspended Particulate (TSP) samplers for measuring the entire range
of particle sizes. Unburned and burned sites differed in vegetation cover and
aerodynamic roughness, yet surprisingly differences in dust emission rates were only
detectable for saltation using BSNE and for smaller aerosols using DustTrak. The
results, surprising in the lack of consistently detected differences, indicate that highresolution DustTrak measurements offered the greatest promise for detecting differences
in background emission rates and that the BSNE samplers, which integrate across height,
were effective for longer intervals. More generally, the results suggest that interplay
between particle size, temporal resolution, and integration across time and height can be
17
complex and may need to be considered more explicitly for effective sampling for
background dust emissions.
Appendix E provides a more integrated perspective on aeolian and fluvial
transport processes and highlights the importance of considering both processes in
concert relative to total erosion and to potential interactions, that relative dominance and
sensitivity to disturbance vary with mean annual precipitation, and that there are
important scale-dependencies associated with aeolian-fluvial interactions. The review
builds on previous literature to present relevant conceptual syntheses highlighting these
issues. The relative investments that have been made in soil erosion and sediment control
are highlighted by comparing the amount of resources allocated to aeolian and fluvial
research using readily available metrics. Literature searches suggest that aeolian
transport may be somewhat understudied relative to fluvial transport and, most
importantly, that only a relatively small number of studies explicitly consider both
aeolian and fluvial transport processes. Numerous environmental issues associated with
intensification of land use and climate change impacts depend on not only overall erosion
rates but also on differences and interactions between aeolian and fluvial processes.
Therefore, a more holistic viewpoint of erosional processes that explicitly considers both
aeolian and fluvial processes and their interactions is needed to optimize management
and deployment of resources to address eminent changes in land use and climate.
Appendix F provides among the first field-based estimates of how changes in
climate are likely to alter land surface dynamics through changes in wind-driven dust
emissions, water-driven sediment production, and soil degradation associated with
18
erosion. Because wind and water erosion are almost exclusively studied independently,
holistic assessment of land surface response to changing climate has been largely
precluded to date. This study shows from co-located erosion measurements in a semiarid
grassland that wind-driven sediment transport exceeded water-driven sediment transport
by a factor of ~40 during a year with wet-dry extremes analogous to those projected to
accompany climate change, corresponding to a more than an order-of-magnitude increase
relative to baseline conditions. The results suggest that climate change that combines
more intense precipitation events and extreme droughts could substantially increase rates
of wind erosion relative to water erosion, portending a dustier future for many drylands.
Appendix G focuses on management for rangeland sustainability and how
alternate grazing and burning practices alter the relative and absolute magnitudes of wind
and water erosion. Because few studies directly measure both wind and water erosion,
lacking is an assessment of how both may differentially respond to grazing and burning.
The study reports co-located, simultaneous estimates of both wind- and water-driven
erosion, estimated as sediment transport, in a semiarid grassland over three years for four
land management treatments: control, grazed, burned, and burned+grazed. For all
treatments, cumulatively wind erosion exceeded water erosion due to a combination of
non-trivial ongoing background rates and elevated rates preceding the monsoon season.
For erosion by both wind and water, rates differed consistently by treatment:
burned+grazed > burned >> grazed > control, with effects immediately apparent after
burning but delayed after grazing until the following growing season. Notably, the windto-water erosion ratio decreased following burning but increased following grazing. The
19
results show how rangeland practices differentially alter wind and water erosion—
differences that could potentially help explain divergence between rangeland
sustainability and degradation.
For Appendices A, B, E, F, and G, Jason Field’s role as lead author was to make
the majority of the contributions to the overall development and design of the study,
conduct all sample collection, processing data analyses, and develop the majority of the
subsequent peer-review manuscripts and publications. For Appendices C and D, Jason
Field’s role as supporting author was to make significant contributions to the overall
design and development of the study, data collection and analysis, and writing portions of
the subsequent peer-review manuscripts and publications. More specifically, for
Appendix C, Jason Field provided the data for the two new grasslands sites, as well as
contributing to the conceptual models developed, and for Appendix D, Jason Field
provided central guidance on experimental design and measurement techniques and had a
major role the key emphasis of the introduction and discussion.
20
REFERENCES
Aguiar, M.R., Sala, O.E., 1999. Patch structure, dynamics, and implications for the
functioning of arid ecosystems. Trends Ecol. Evol. 14, 273-277.
Bagnold, R.A., 1941. The Physics of Blown Sand and Desert Dunes. Chapman and Hall
Ltd., London.
Baker, M.B. Jr., DeBano, L.F., Ffolliott, P.F., 1995. Soil loss in piñon-juniper ecosystems
and its influence on site productivity and desired future condition, in: Shaw, D.W.,
Aldon, E.F., LoSapio, C. (Eds.), Desired Future Conditions for Piñon-Juniper
Ecosystems, Proceedings of Symposium, 8–12 August 1994, Flagstaff, AZ. General
Technical Report. RM-258. US Department of Agriculture, Forest Service, Rocky
Mountain Forest and Range Experiment Station: Ft Collins, CO, pp. 9–15.
Belnap, J., 1995. Surface disturbances: their role in accelerating desertification. Environ.
Monit. Assess. 37, 39-57.
Breshears, D.D., Whicker, J.J., Johansen, M.P., Pinder III, J.E., 2003. Wind and water
erosion and transport in semi-arid shrubland, grassland and forest ecosystems:
quantifying dominance of horizontal wind-drive transport. Earth Surf. Proc. Land. 28,
1189-1209.
Brown, L.R., 1981. World population growth, soil erosion, and food security. Science
214, 995-1000.
Bullard. J.E., Livingstone, I., 2002. Interactions between aeolian and fluvial systems in
dryland environments. Area 34, 8-16.
21
Bullard, J.E., McTainsh, G.H., 2003. Aeolian-fluvial interactions in dryland
environments: examples, concepts and Australia case study. Prog. Phys. Geog. 27,
471-501.
CCSP, 2008. The Effects of Climate Change on Agriculture, Land Resources, Water
Resources, and Biodiversity in the United States. United States Climate Change
Science Program. Washington, D.C.
Chadwick, O.A., Derry, L.A., Vitousek, P.M., Huebert, B.J., Hedin, L.O., 1999.
Changing sources of nutrients during four million years of ecosystem development.
Nature 397, 491-497.
Chappell, A., Zobeck, T.M., Brunner, G., 2005. Using on-nadir spectral reflectance to
detect soil surface changes induced by simulated rainfall and wind tunnel abrasion.
Earth Surf. Proc. Land. 4, 489-511.
Chepil, W.S., 1949. Wind erosion control with shelterbelts in North China. Agron. J. 41,
127-129.
Cooke, R.U., Warren, A., Goudie, A.S., 1993. Desert Geomorphology. UCL Press,
London.
Cornelis, W.M., Oltenfreiter, G., Gabriels, D., Hartmann, R., 2004. Splash-saltation of
sand due to wind-driven rain: vertical deposition flux and sediment transport rate.
Soil Sci. Soc. Am. J. 68, 32-40.
Erpul, G., Gabriels, D., Cornelis, W.M., Sammy, H., Guzelordu, T., 2009. Sand transport
under increased lateral jetting of raindrops induced by wind. Geomorphology 104,
191-202.
22
Gillette, D.A., Passi, R., 1988. Modeling dust emission caused by wind erosion. J.
Geophys. Res.-Atmos. 93, 14233-14242.
Goudie, A.S., 2008. The history and nature of wind erosion in deserts. Annu. Rev. Earth
Planet. Sci. 36, 97-119.
Heathcote, R.L., 1983. The Arid Lands: Their Use and Abuse. Longman, New York.
Jickells, T.D., An, Z.S., Andersen, K.K., Baker, A.R., Bergametti, G., Brooks, N., Cao,
J.J., Boyd, P.W., Duce, R.A., Hunter, K.A., Kawahata, H., Kubilay, N., laRoche, J.,
Liss, P.S., Mahowald, N., Prospero, J.M., Ridgwell, A.J., Tegen, I., Torres, R., 2005.
Global iron connections between desert dust, ocean biogeochemistry, and climate.
Science 308, 67-71.
Kirkby, M.J., 1980. The problem, in: Kirkby, M.J., Morgan, R.P.C. (Eds.), Soil Erosion.
John Wiley and Sons, New York, pp. 1-16.
Lal, R., Sobecki, T.M., Iivari, T., Kimble, J.M., 2003. Soil degradation in the United
States: extent, severity, and trends. Lewis Publ., Boca Raton, Florida.
Li, J., Okin, G.S., Alvarez, L.J., Epstein, H.E., 2008. Effects of wind erosion on the
spatial heterogeneity of soil nutrients in a desert grassland of southern New Mexico.
Biogeochemistry 88, 73-88.
Li, J., Okin, G.S., Hartman, L.J., Epstein, H.E., 2007. Quantitative assessment of wind
erosion and soil nutrient loss in desert grasslands of southern New Mexico, USA.
Biogeochemistry. 85, 317-332.
Muhs, D.R., Bettis, E.A., Aleinikoff, J.N., McGeehin, J.P., Beann, J., Skipp, G.,
Marshall. B.D., Roberts, H.M., Johnson, W.C., Benton, R., 2008. Origin and
23
paleoclimatic significance of late Quaternary loess in Nebraska: evidence from
stratigraphy, chronology, sedimentology, and geochemistry. Geol. Soc. Am. Bull.
120, 1378-1407.
Nearing, M.A., 2005. Soil erosion under climate change: rates, implications and
feedbacks-Introduction. Catena 61, 103-104.
Neff, J.C., Ballantyne, A.P., Famer, G.L., Mahowald, N.M., Conroy, J.L., Landry, C.C.,
Overpeck, J.T., Painter, T.H., Lawrence, C.R., Reynolds, R.L., 2008. Increasing
eolian dust deposition in the western United States linked to human activity. Nature Geosciences. doi:10.1038/ngeo133.
NRCS, 2000a. Average Annual Soil Erosion by Water on Cropland and CRP Land, 1997.
Map ID: m5058. NRCS, Resource Assessment Division, Washington, DC.
NRCS, 2000b. Average Annual Soil Erosion by Wind on Cropland and CRP Land, 1997.
Map ID: m5065. NRCS, Resource Assessment Division, Washington, DC.
Okin, G.S., Parsons, A.J., Wainwright, J., Herrick, J.E., Bestelmeyer, B.T., Peters,
D.P.C., Fredrickson, E.L., 2009. Do changes in connectivity explain desertification?
BioScience 59, 237-244.
Oldeman, L., Hakkeling, R., Sombroek, W., 1990. World Map of the Status of HumanInduced Soil Degradation: An Explanatory Note. International soil reference and
information center, Wageningen, The Netherlands and United Nations Environmental
Program, Nairobi, Kenya.
24
Peters, D.P.C., Bestelmeyer, B.T., Herrick, J.E., 2006. Disentangling complex
landscapes: new insights into arid and semiarid system dynamics. Bioscience 56, 491501.
Peters, D.P.C., Groffman, P.M., Nadelhoffer,K.J., Grimm, N.B., Collins, S.C., Michener,
W.K., Huston, M.A., 2008. Living in an increasingly connected world: a framework
for continental-scale environmental science. Frontiers Ecol. Environ. 6, 229-237.
Peters, D.P.C., Sala, O.E., Allen, C.D., Covich, A., Brunson, M., 2007. Cascading events
in linked ecological and socio-economic systems: predicting change in an uncertain
world. Front. Ecol. Environ. 5, 221-24.
Pimentel, D. (Ed.), 1993. World Soil Erosion and Conservation. Cambridge Univ. Press,
Cambridge, UK.
Pye, K., 1987. Aeolian dust and dust deposits. Academic Press, Boca Raton, Florida.
Ravi, S., D’Odorico, P., Okin, G.S., 2007a. Hydrologic and aeolian controls on
vegetation patterns in arid landscapes. Geophys. Res. Lett. 34, L24S23.
Ravi, S., D'Odorico, P., Zobeck, T.M., Over, T.M., Collins, S., 2007b. Feedbacks
between fires and wind erosion in heterogeneous arid lands. J. Geophys ResBiogeosciences 112, G04007.
Reynolds, R., Belnap, J., Reheis, M., Lamothe, P., Luiszer, F., 2001. Aeolian dust in
Colorado Plateau soils: nutrient inputs and recent change in source. Proc. Nat. Acad.
Sci. USA 98, 7123-7127.
Ruhe, R.V., Daniels, R.B., Cady, J.G., 1967. Landscape evolution and soil formation in
southwestern Iowa. USDA Tech. Bull. 1349, Washington, D.C.
25
Schlesinger, W.H., Reynolds, J.F., Cunningham, G.L., Huenneke, L.F., Jarrell, W.M.,
Virginia, R.A., Whitford, W.G., 1990. Biological feedbacks in global desertification.
Science 247, 1043-1048.
Shao, Y.P., Shao, S.X. 2001. Wind erosion and wind erosion research in China: a
review. Ann. Arid Zone 40, 317-336.
Stout, J.E., Warren, A., Gill, T.E., 2009. Publication trends in aeolian research: an
analysis of the Bibliography of Aeolian Research. Geomorphology 105, 6-17.
Sun, J.M., 2002. Provenance of loess material and formation of loess deposits on the
Chinese Loess Plateau. Earth Planet Sci. Lett. 203, 845-859.
Swap, R., Garstang, M., Greco, S., Talbot, R., Kallberg, P., 1992. Saharan dust in the
Amazon Basin. Tellus B Chem. Phys. Meteorol. 44, 133-149.
Tegen, I., Fung, I., 1994. Modeling of mineral dust in the atmosphere – sources,
transport, and optical-thickness. J. Geophys. Res.-Atmos. 99, 22897-22941.
Toy, T.J., Foster, G.R., Renard, K.G., 2002. Soil erosion: processes, prediction,
measurement and control. John Wiley & Sons, New York.
USDA, 2006. Conservation Resource brief: Soil Erosion. United States Department of
Agriculture, Natural Resources Conservation Service February 2006, Brief Number
0602.
Valentin, C., 1996. Soil erosion under global change, in: Walker, B., Steffen, W. (Eds.),
Global Change and Terrestrial Ecosystems. Cambridge University Press, New York,
pp. 317-338.
26
Visser, S.M., Sterk, G., Ribolzi, O., 2004. Techniques for simultaneous quantification of
wind and water erosion in semi-arid regions. J. Arid Environ. 59, 699-717.
Worster, D., 1979. Dust Bowl: the southern plains in the 1930s. Oxford University Press,
New York.
27
APPENDIX A
THE ECOLOGY OF DUST: LOCAL- TO GLOBAL- SCALE
TERRESTRIAL PERSPECTIVES
Jason P Field1, Jayne Belnap2, David D Breshears1,3, Jason C Neff4, Gregory S Okin5,
Jeffrey J Whicker6, Thomas H Painter7, Sujith Ravi8, Marith C Reheis9, and
Richard L Reynolds9
1
School of Natural Resources, University of Arizona, Tucson, AZ; 2US Geological
Survey, Southwest Biological Science Center, Moab, UT; 3Institute for the Study of
Planet Earth, Department of Ecology and Evolutionary Biology, University of Arizona,
Tucson, AZ; 4Geological Sciences Department, Environmental Studies Program,
University of Colorado at Boulder, Boulder, CO; 5Department of Geography, University
of California, Los Angeles, CA; 6Los Alamos National Laboratory, Environmental
Programs , Los Alamos, NM; 7Department of Geography, Snow Optics Laboratory,
University of Utah, Salt Lake City, UT; 8B2 Earthscience, UA Biosphere 2, University of
Arizona, Tucson, AZ; 9Denver Federal Center, US Geological Survey, Denver, CO
Document Type: Peer-reviewed publication in Frontiers in Ecology and the
Environment.
Abstract: Wind erosion and associated dust emissions play a fundamental role in many
ecological processes and provide important biogeochemical connectivity at scales from
individual plants up to the entire globe. Yet, most ecological studies do not explicitly
consider dust-driven processes, perhaps because most relevant research on aeolian (winddriven) processes has been presented in a geosciences rather than an ecological context.
To bridge this disciplinary gap, we provide a general overview of the ecological
importance of dust, examine complex interactions between wind erosion and ecosystem
dynamics from the scale of plants and surrounding space to regional and global scales,
and highlight specific examples of how disturbance affects these interactions and their
consequences. It is likely that changes in climate and intensification of land use will lead
to increased dust production from many drylands. To address these issues, environmental
28
scientists, land managers, and policy makers need to consider wind erosion and dust
emission more explicitly in resource management decisions.
In a nutshell:
•
Ecologists and other environmental scientists often overlook the importance of
dust and wind-driven processes, yet these processes exert a fundamental influence
on biogeochemical and ecological systems
•
The importance of wind-driven processes crosses scales from individual plants up
to global levels
•
Because changes in climate and intensification of land use are expected to result
in increased dust production, ecologists, land managers, and policy makers need
to more explicitly consider and manage dust emission
Introduction
For many scientists, the only dust they think about is the thin film of material that
accumulates on their computer monitor on a regular basis. However, dust has enormous
relevance to a wide range of ecological processes and environmental management
challenges. Dust is fine particulate material that is removed from the land surface by
wind erosion and is small enough to be suspended in the atmosphere (Bagnold 1941; Toy
et al. 2002). Dust emissions vary with climate, as shown by paleo-studies (eg Clark et al.
2002), and also with land use. Perhaps the most notable example of how ecologically
important dust can be is the Dust Bowl era of the 1930s, in the American Great Plains,
which is considered by many experts to be one of the most severe environmental
29
catastrophes in the history of the US (eg Peters et al. 2007). The widespread cultivation
of marginally arable lands, in conjunction with a severe regional-scale drought during the
1930s, caused substantial increases in wind-erosion rates, resulting in the degradation of
roughly 90 million ha of land (Utz et al. 1938) and the loss of nearly 800 million metric
tons of topsoil in 1935 alone (Johnson 1947; Hansen and Libecap 2004). This large-scale
amplification of wind erosion was the result of small fields becoming more erosive and
interconnected (Hansen and Libecap 2004; thereby triggering a threshold response
(Peters et al. 2007). The devastating effects of the Dust Bowl were felt nationally and
resulted in the formation of the Soil Conservation Service in 1935. However, the
important ecological lessons of the Dust Bowl have faded with time, and most ecological
studies do not explicitly consider the impact of dust flux and wind erosion. Ironically, the
former Soil Conservation Service – now the Natural Resources Conservation Service –
has shifted the vast majority of its focus to water erosion, and has largely abandoned the
problem of wind-caused erosion (Field et al. in press).
Environmental scientists are increasingly recognizing dust as both a major
environmental driver and a source of uncertainty for climate models (Tanaka and Chiba
2006; Neff et al. 2008). Wind erosion and dust emission can cause substantial impacts,
not only on human health (eg through respiratory problems), but also on basic ecosystem
processes, at scales ranging from individual plants or even smaller (Figure 1a) up through
local and regional scales (Figure 1b, c) to a global scale (Figure 1d), representing
biogeochemical connectivity across continents (Peters et al. 2007; Okin et al. in press).
30
Here, we provide a primer on the importance of aeolian (wind-driven) processes
associated with wind erosion, as well as an overview of their ecological relevance at
scales from plant-interspace (the area between as well as beneath individual plants) to
global, and associated aspects of biogeochemical connectivity. An underlying theme that
plays out at many scales is that wind erosion has a highly non-linear response to
disturbances that reduce ground cover below a critical threshold. We discuss how the
effects of climate change (Seager et al. 2007) and land use intensification (Okin et al.
2006) could amplify dust emissions from drylands and pose major environmental
challenges for land managers and policy makers.
A Wind Erosion Primer
Wind transports soil material through three mechanisms (Figure 2) that are
roughly differentiated based on the soil particle diameter (these categories overlap):
surface creep for soil particle diameters > 500 μm, saltation for diameters ranging from
20–500 μm, and suspension for diameters < 20 μm (Bagnold 1941; Toy et al. 2002;
Goudie and Middleton 2006). All three processes redistribute soil and associated
nutrients and organic material at different spatial scales (Field et al. in press). Winddriven surface creep and saltating particles dominate the mass movement of soil on a
local scale (< several m; Stout and Zobeck 1996). In contrast, suspended dust particles
can be transported over long distances and can be moved at regional, continental, and
even global scales (Chadwick et al. 1999; Prospero et al. 2001; Goudie and Middleton
2006). Most of the horizontal aeolian sediment transport occurs close to the soil surface,
decreasing sharply with height (Shao et al. 1993). A small fraction of this flux can
31
become suspended and thus be available for long-distance dust transport, as reflected in a
vertical flux that is linearly related to horizontal flux (Gillette et al. 1997). Because soil
nutrients (eg nitrogen, phosphorus) and organic matter are often associated with smaller
soil particles, soil fertility in dust source areas becomes depleted while sink areas are
concomitantly enriched (Li et al. 2007).
Wind erosion rates and dust emissions at a specific location are influenced by a
variety of factors, such as microscale wind gradients and atmospheric relative humidity
(Toy et al. 2002). Wind speed is related to the amount of energy available to move
sediment, and much aeolian research focuses on the “threshold friction velocity” wind
speed at which particles of a given size under a given set of field conditions begin to
detach from the soil surface (Gillette et al. 1980). Atmospheric relative humidity controls
soil moisture at the soil surface, especially in arid and semiarid regions during rainless
periods (Ravi et al. 2004), because soil moisture in particles at the soil surface is typically
at equilibrium with atmospheric moisture. This is important, because soil moisture
influences the interparticle forces that, in turn, influence the threshold velocity, resulting
in a clear, but complex relationship between atmospheric relative humidity, particle size,
and soil erodibility (Ravi et al. 2004). Wind erosion may be interactive with water
erosion, although few studies have specifically addressed this issue (Field et al. in press).
Collectively, these complex relationships need to be considered in terms of their relative
role in affecting aeolian processes at all scales.
32
Plant-Interspace Scale
At the plant-interspace scale, which includes the area between as well as beneath
individual plants, aeolian transport is a major abiotic mechanism for moving material
both within and out of environments with discontinuous cover. The erosivity of the soil
surface, and thus the potential impacts of aeolian processes at the plant–interspace scale,
depend on both the ability of the soil surface to resist erosion and the ability of the wind
to reach the soil surface. Erosion resistance is determined by the strength of the soil and
the presence of surface protectors, such as rocks, plant litter, and physical and biological
soil crusts (Gillette et al. 1980; Okin et al. 2006). Rocks and plant litter too large to be
moved by wind offer the greatest soil protection. Physical soil crusts – created by the
binding together of silt and clay particles when wetted and then dried – protect soils,
except when crusts are subjected to disturbance. Unless disturbed, these soils have an
inherently higher resistance to erosion than soils dominated by coarser sand particles.
Biological soil crusts, composed of cyanobacteria, lichens, and moss, stabilize soils by
excreting mucilaginous material that binds soil surface particles together, thereby
increasing soil aggregate size and increasing soil resistance to the shearing forces of wind
(Belnap and Gillette 1997).
The type, cover, and arrangement of vegetation have the strongest influence on
the ability of the wind to reach the soil surface. The patchy and dynamic nature of
vegetation in dryland regions results in aeolian transport being highly heterogeneous in
both space and time. The amount of material that is moved depends on the size of
unvegetated gaps upon which the wind can act (generally excluding rocky or gravely
33
areas referred to as desert pavement and areas covered by physical or biological soil
crust) and the height and density of the vegetation, which controls the size of the
protected area downwind of individual plants (Breshears et al. 2009). Although surface
characteristics are important, the amount of horizontal flux depends largely on the
structure of the ecosystem and the degree of connectivity between unvegetated gaps
(Okin et al. in press; Figure 3). Unvegetated areas immediately downwind of vegetation
(within 5–10 times the height of an individual plant) are relatively protected from the
erosive force of the wind by the plant. In contrast, unvegetated areas further downwind
from a plant do not experience the same degree of protection from erosion (Okin 2008).
This disparity leads to heterogeneous erosion and the net movement of soil and litter from
unvegetated gaps, and concentration of these resources beneath plant canopies. Saltationsized particles are concentrated in protected areas beneath plant canopies, giving rise to
coppice dunes in extreme circumstances, such as windy environments that have sparse
vegetation cover and easily erodible soil. Because saltating material carries most of the
mass and momentum, it can have significant physical effects on existing vegetation,
including burial, exposure of below-ground plant tissue (pedastaling), abrasion of plant
tissue, and leaf stripping. This has been shown to indirectly lead to reduced plant growth
and mortality and to contribute to rapid changes in ecosystem structure (ie initiating a
rapid change from grassland to shrubland; Okin et al. 2006).
Finer particles moved by wind – including silt- and clay-sized particles – contain
most of the cation-exchange capacity, water-holding capacity, and fertility of the soil
(Toy et al. 2002). Some of these finer particles are trapped by local vegetation (Raupach
34
et al. 2001) and, combined with a similar mechanism for water erosion, contribute to the
formation of fertile islands found throughout dryland regions (Schlesinger et al. 1990). In
addition, fine particles and associated nutrients are added to soils by infiltration through
gravelly surfaces (Reheis et al. 2009). However, many of these finer soil particles are
eventually lost from the local system due to wind erosion (Gillette 1974), resulting in
local depletion of soil fertility and water-holding capacity (Li et al. 2007, 2008). The
relative depletion of fine particles at the surface may not have immediate impacts on
existing vegetation, because the effect is concentrated above the root zone; the
implications of this depletion for vegetation establishment, however, are striking, due to
the heavy reliance of germinants on soil resources and water in the uppermost soil layers.
Many of the factors that drive wind erosion are, of course, greatly affected by soil
surface disturbances. Grazing cattle crush biological and physical soil crusts and decrease
vegetative cover (Nash et al. 2004), thereby increasing wind erosion (eg Neff et al.
2008). Offroad vehicles and military training also crush vegetation and impact plant–
interspace surface characteristics, particularly biological and physical soil crusts (Belnap
and Gillette 1997; Breshears et al. 2009; Figure 4). Fire can dramatically increase wind
erosion (Whicker et al. 2002; Breshears et al. 2009), although fire may be less spatially
extensive than grazing and recreational use. Burning vegetation (even by typical
rangeland fires) releases different amounts of organic compounds, which, in turn, leading
to different levels of water repellency in the soil, depending on various factors, such as
vegetation type, soil properties, and fire intensity and duration (DeBano 2000). Fireinduced water repellency decreases the strength of interparticle wet-bonding forces by
35
increasing the soil–water contact angle. This repellency enhances soil erodibility by
causing a drop in threshold friction velocity, thereby increasing post-fire erosion
(Whicker et al. 2002; Ravi et al. 2007).
There are important feedbacks between the vegetation and aeolian flux in deserts
(Figure 5). Aeolian flux controls the redistribution of sediment and the loss of dust and
dust-borne nutrients, thus affecting the amount and distribution of vegetation on the land
surface. The amount and distribution of vegetation, in turn, affects the degree and spatial
patterns of aeolian flux. This feedback can occur in most environments, including those
with relatively high vegetation cover, and is responsible for cascading land-degradation
phenomena caused by local or regional disturbance events (Peters et al. 2007). At the
same time, dust emitted by desert regions, particularly those that have experienced
substantial disturbance, can have critical consequences for downwind ecosystems.
Regional- to Global-Scale Consequences
The regional and global transport of dust plays many fundamental roles in the
earth system. Dust has an important, yet relatively poorly quantified, influence on climate
at both regional and global scales. At regional scales, dust can have a large effect on the
atmospheric radiative balance and the concentration of condensation nuclei, both of
which can influence climate variability via effects on surface temperatures and
precipitation patterns (Yoshioka et al. 2007). Dust deposited on mountain snowpack can
have an indirect effect on climate through the snow-albedo feedback (Painter et al. 2007;
Steltzer et al. in press; Figure 6). Dust decreases snow albedo, removing snow cover
earlier and revealing a markedly darker land surface that absorbs solar radiation and
36
reradiates in the infrared to the atmosphere. The triggering of earlier and faster snowmelt
by dust can potentially result in less total and less late-season water supplies in areas
where seasonal water scarcity occurs.
In addition to its effects on climate, dust plays an important role in the control of
regional and global biogeochemical cycles and dispersal of pathogens. At the global
scale, nutrient additions by dust may have stimulated the productivity of oceanic plankton
over glacial time scales, thus accelerating the uptake of atmospheric CO2 (eg Jickells et
al. 2005). At the regional scale, there have been a number of studies examining the
impact of dust deposition on terrestrial and aquatic nutrient cycling. In tropical
ecosystems with a long legacy of chemical weathering and depletion of soil base cations
and phosphorus, dust has been suggested as a major nutrient source. For example, the
transport of Saharan dust to the Amazon basin has played an important role in offsetting
the losses of bedrock-derived nutrients to leaching (Koren et al. 2006). Similar studies in
Hawai΄i suggest that dust is responsible for supplying essential plant elements to heavily
weathered soils (Chadwick et al. 1999). There is mounting evidence that dust transport
and deposition are important to temperate ecosystems as well. Transport of nitrogen,
phosphorus, and other nutrients by dust can be substantial (Neff et al. 2008), and the
subsequent deposition of these nutrients may influence both terrestrial and aquatic
ecosystems. In stable soil surfaces on the Colorado Plateau, dust accumulation in soils
has increased the stocks of all macro- and micronutrients, especially phosphorus and
magnesium (Reynolds et al. 2006). Long-term dust accumulation can lead to the
development of loess soils (unstratified wind-blown silt deposits) in some regions, such
37
as the American Midwest, providing for fertile agriculture (Pye 1987). These diverse
studies illustrate that dust is an important and underappreciated component of
contemporary biogeochemical cycles.
The importance of dust to global biogeochemical cycles raises a number of
questions about the magnitude, distribution, and variation in dust fluxes across the Earth;
this has lead to numerous attempts to quantify the contributions of dust sources around
the globe. Dust from land disturbed by humans or extreme climatic events, such as
drought, may constitute a substantial fraction – perhaps one-third to one-half – of the
total atmospheric dust loading (Tegen and Fung 1995). Model assessments indicate that
global fluxes are dominated by the large deserts of North Africa, Asia, and the Middle
East (Tanaka and Chiba 2006). This global emission flux appears to be dominated by
non-arable, hyperarid regions, with large interannual variation in dust fluxes controlled
primarily by climate (Figure 7).
Dust emissions from some less arid drylands appear to be more heavily influenced
by human land use. In the semiarid regions of China, for instance, there is evidence that
wind erosion of soils is influenced by grazing activities (eg Liu et al. 2007), and in South
and North America, ice and sediment core records reveal that human activity has
increased dust deposition over the past 100–200 years. In the western US, lake sediment
records from the San Juan Mountains of Colorado show that dust loading reached a peak
of ~ 500% of background (late Holocene) deposition circa 1900, when settlement and
widespread livestock grazing dramatically increased (Neff et al. 2008). Abandoned
cotton fields in Texas and Arizona and military training grounds in Texas and California
38
consistently produce large regional dust storms that can be seen on satellite imagery (eg
Prospero et al. 2002). In South America, human land use of semiarid regions for grazing
also appears to have increased dust deposition rates in the 20th century, relative to the
19th century (McConnel et al. 2007). The human role in dust emission and deposition
may be limited at the global scale, but at local to regional scales, dust appears to be
mostly a byproduct of human land-use decisions. In this way, humans may be indirectly
responsible for potentially large, but poorly understood, perturbations to local, regional,
and global hydrologic and biogeochemical cycling.
Projections and Implications
The future will bring many environmental changes to dryland areas; these will act
independently and synergistically to affect dust fluxes at the local, regional, and global
scales. Projected climate changes include a global increase in temperatures (Seager et al.
2007) in concert with a range of future precipitation possibilities for drylands. This could
include green-up in some areas, but most regions are more likely to experience a small
decrease in precipitation. By 2050, increased temperature alone is expected to decrease
average soil moisture conditions in the southwestern US (Seager et al. 2007), to levels
below those experienced during the most severe droughts of this century, including the
Dust Bowl years (Pulwarty et al. 2005). Such declines in soil moisture will probably
result in a reduction in the protective vegetative cover, a slower recovery from
disturbance, and an increase in dust emission from exposed soil. Lower soil moisture will
also mean drier fuels that burn more readily – wildfires in the western US are projected to
increase substantially (Ryan et al. 2008), which will also increase exposed soils and the
39
hydrophobicity of those soils, thus amplifying dust emissions (Whicker et al. 2002; Ravi
et al. 2007).
Worldwide, human use of dryland regions, which comprise almost 41% of the
terrestrial land surface, is increasing dramatically. Currently, over 2 billion people use
drylands for habitation and food (MA 2005), and much of the global population growth is
occurring in these water-limited landscapes (Reynolds et al. 2007). For example, human
populations in southern Arizona and California are expected to grow from 25 million to
38 million over the next 11 years (Pulwarty et al. 2005). An increase in human
settlement/use of these landscapes will be accompanied by a further loss in the protective
covering of plants, plant litter, and physical and biological soil crusts, thereby amplifying
dust emissions from the disturbed surfaces. Offroad recreation activity in southern
California has risen from virtually zero in 1960 to almost 10 million user-days in 2006
(Bureau of Land Management RIMS database). If users drive 32 km per day, this specific
activity alone, in this relatively small region, can generate as much as 2.7 metric tons of
dust per year (Dyck and Stukel 1976; Forman et al. 2003). The now exploding
exploration and development of energy resources (including wind and solar) in dryland
regions is also of concern. All of these activities will result in the loss of vegetation and
soil surface protectors (eg scraping away vegetation for solar farms and oil pads),
increased offroad vehicle traffic, pipelines, transmission lines, and greatly increased
traffic on current and newly established dirt roads. The demand for water is also everincreasing in these regions, demands that result in water diversions or the pumping of
water from shallow lakes, often drying them completely (eg Lake Aibi in China, Aral Sea
40
in Uzbekistan, Owens Lake in US), or pumping water from shallow aquifers, which may
lead to the death of vegetation (Elmore et al. 2008). These activities can leave vast
expanses of soils highly vulnerable to wind erosion. An example of the influence of dust
from dry-lake sediments on vegetation is provided by Blank et al. (1999), who found that
dust from playas (dry or ephemeral lakebeds), which are rich in nitrates and other
nutrients, stimulated the invasion of a playa-margin, dune-mantled landscape by weeds
(Salsola paulsenii).
The conversion of perennial plant communities to those dominated by annual
plants is also increasing globally, mostly as a result of fire, abandonment of agricultural
fields, overgrazing, and other soil surface-disturbing activities (D’Antonio and Vitousek
1992). In wet years, the annual cover is sufficient to stabilize soils and may even exceed
the protection offered by the perennial community. However, in dry years, these annual
grasses do not germinate, or die shortly after germination, leaving soils barren and
vulnerable to erosion. Dominance by annual plants also accelerates fire cycles. In wet
years, these grasses produce sufficient continuous fuels to carry fire in the dry years that
follow, leaving post-fire soils vulnerable to erosion.
Dust responses can become synergistic with changes in climate and land use when
one or more of the above factors coincide in time or space. For instance, offroad vehicle
use leads to decreases in plant biomass and cover, as a result of direct impacts to
vegetation and dusting of nearby plants (Sharifi et al. 1999). When these impacts occur
during times of reduced soil moisture, the reduction in plant cover is even greater,
allowing for increased erosion. Another synergistic series of effects can occur on
41
landscapes where perennial plants have been replaced by annual plants. When these
surfaces are then disturbed by livestock or vehicles, an exponential increase in soil loss
can be observed, compared to an annual-dominated but untrampled landscape (Belnap et
al. in press).
In summary, greater dust emissions, including more frequent and larger dust
storms, are likely to occur from dryland regions as temperatures increase and more
dryland areas are trampled, cleared of vegetation, plowed, and/or converted from
perennial to annual plants. These increasing emissions will result in degraded soils and
plants at the dust source, as well as in impacts to human and ecosystem health during
transport and at deposition points. Avoiding the potentially severe consequences of this
future scenario will require a new approach to the management of dryland regions. We
need to identify the chronic and acute sources of dust that have potentially large impacts
at local, regional, and global scales (Peters et al. 2007). We also need to better understand
how the timing, type, and intensity of different land uses affect dust production. The
overarching challenge for ecologists and other environmental scientists, land managers,
and policy makers will be to work together to manage vulnerable areas in ways that
reduce excess dust production to the fullest extent possible (Okin et al. in press).
Acknowledgements
We thank K.A. Ferguson, D.J. Law, C.J. Parry, and S.L. Phillips for comments on the
manuscript. This work was supported by the U.S. Department of Agriculture
Cooperative State Research, Education, and Extension Service (CSREES 2005-38420-
42
15809), the National Science Foundation (NSF DEB-0816162, DEB-0618210, DEB0823205, EAR-072021), and the Global Change Program of the U.S. Geological Survey.
References
Bagnold RA. 1941. The physics of blown sand and desert dunes. London, UK: Chapman
and Hall Ltd.
Belnap J and Gillette DA. 1997. Disturbance of biological soil crusts: impacts on
potential wind erodibility of sandy desert soils in southeastern Utah. Land Degrad
Dev 8: 355–62.
Belnap J, Reynolds RL, Reheis MC, et al. Sediment losses and gains across a gradient of
livestock grazing and plant invasion in a cool semi-arid grassland, Colorado Plateau,
USA. Aeolian Res. In press.
Blank RR, Young JA, and Allen FL. 1999. Aeolian dust in a saline playa environment,
Nevada, USA. J Arid Environ 4: 365–81.
Breshears DD, Whicker JJ, Zou CB, et al. 2009. A conceptual framework for dryland
aeolian sediment transport along the grassland–forest continuum: effect of woody
plant canopy cover and disturbance. Geomorphology 105: 28–38.
Chadwick OA, Derry LA, Vitousek PM, et al. 1999. Changing sources of nutrients
during four million years of ecosystem development. Nature 397: 491–97.
Clark JS, Grimm EC, Donovan JJ, et al. 2002. Drought cycles and landscape responses to
past aridity on prairies of the northern Great Plains, USA. Ecology 83: 595–601.
D’Antonio CM and Vitousek PM. 1992. Biological invasions by exotic grasses, the
grass/fire cycle, and global change. Annu Rev Ecol Syst 23: 63–87.
43
DeBano LF. 2000. The role of fire and soil heating on water repellence in wildland
environments: a review. J Hydrol 231: 195–206.
Dyck RIJ and Stukel JJ. 1976. Fugitive dust emissions from trucks on unpaved roads.
Environ Sci Technol 10: 1046–48.
Elmore AJ, Kaste JM, Okin GS, and Fantle MS. 2008. Groundwater influences on
atmospheric dust generation in deserts. J Arid Environ 72: 1753–65.
Field JP, Breshears DD, and Whicker JJ. Toward a more holistic perspective of soil
erosion: why aeolian research needs to explicitly consider fluvial processes and
interactions. Aeolian Res. In press.
Forman RTT, Sperling D, Bissonette JA, et al. 2003. Road ecology: science and
solutions. Washington, DC: Island Press.
Gillette DA. 1974. On the production of soil wind erosion aerosols having the potential
for long range transport. Journal des Recherches Atmospheriques 8: 735–44.
Gillette DA, Adams J, Endo A, et al. 1980. Threshold velocities for input of soil particles
into the air by desert soils. J Geophys Res-Oc Atm 85: 5621–30.
Gillette DA, Fryrear DW, Gill TE, et al. 1997. Relation of vertical flux of particles
smaller than 10 μm to aeolian horizontal mass flux at Owens Lake. J Geophys ResAtmos 102: 26009–15.
Goudie AS and Middleton NJ. 2006. Desert dust in the global system. Heidelberg,
Germany: Springer-Verlag.
Hansen ZK and Libecap GD. 2004. Small farms, externalities, and the Dust Bowl of the
1930s. J Polit Econ 112: 665–94.
44
Jickells TD, An ZS, Andersen KK, et al. 2005. Global iron connections between desert
dust, ocean biogeochemistry, and climate. Science 308: 67–71.
Johnson V. 1947. Heaven’s tableland: the Dust Bowl story. New York, NY: Farrar,
Straus.
Koren I, Kaufman YJ, Washington R, et al. 2006. The Bodele depression: a single spot in
the Sahara that provides most of the mineral dust to the Amazon forest. Environ Res
Lett 1: 014005, doi:10.1088/1748-9326/1/1/014005.
Li J, Okin GS, Alvarez LJ, and Epstein HE. 2008. Effects of wind erosion on the spatial
heterogeneity of soil nutrients in a desert grassland of southern New Mexico.
Biogeochemistry 88: 73–88.
Li J, Okin GS, Hartman LJ, and Epstein HE. 2007. Quantitative assessment of wind
erosion and soil nutrient loss in desert grasslands of southern New Mexico, USA.
Biogeochemistry 85: 317–32.
Liu LY, Li XY, Shi PJ, et al. 2007. Wind erodibility of major soils in the farming–
pastoral ecotone of China. J Arid Environ 68: 611–23.
MA (Millennium Ecosystem Assessment). 2005. Ecosystems and human well-being:
desertification synthesis. Washington, DC: World Resources Institute.
McConnell JR, Aristarain AJ, Banta JR, et al. 2007. 20th-century doubling in dust
archived in an Antarctic Peninsula ice core parallels climate change and
desertification in South America. P Natl Acad Sci USA 104: 5743–48.
Nash MS, Jackson E, and Whitford WG. 2004. Effects of intense, short-duration grazing
on microtopography in a Chihuahuan Desert grassland. J Arid Environ 56: 383–93.
45
Neff JC, Ballantyne AP, Farmer GL, et al. 2008. Increasing aeolian dust deposition in the
western United States linked to human activity. Nat Geosci doi:10.1038/ngeo133.
Okin GS. 2008. A new model for wind erosion in the presence of vegetation. J Geophys
Res-Earth 113: F02S10.
Okin GS, Herrick JE, Gillette DA. 2006. Multiscale controls on and consequences of
aeolian processes in landscape change in arid and semiarid environments. J Arid
Environ 65: 253–75.
Okin GS, Parsons AJ, Wainwright J, et al. Do changes in connectivity explain
desertification? BioScience. In press.
Painter TH, Barrett AP, Landry CC, et al. 2007. Impact of disturbed desert soils on
duration of mountain snowcover. Geophys Res Lett 34: L12502,
doi :10.1029/2007GL030208.
Peters DPC, Sala OE, Allen CD, et al. 2007. Cascading events in linked ecological and
socio-economic systems: predicting change in an uncertain world. Front Ecol
Environ 5: 221–24.
Prospero JM, Ginoux P, Torres O, et al. 2002. Environmental characterization of global
sources of atmospheric soil dust derived from Nimbus-7 TOMS absorbing aerosol
product. Rev Geophys 40: 1002, doi:10.1029/2000RG000095.
Pulwarty R, Jacobs K, and Dole R. 2005. The hardest working river: drought and critical
water problems in the Colorado River Basin. In: Wilhite D (Ed). Drought and water
crises: science, technology and management. Boca Raton, FL: CRC Press.
46
Pye K. 1987. Aeolian dust and dust deposits. New York, NY: Cambridge University
Press.
Raupach MR, Woods N, Dorr G, et al. 2001. The entrapment of particles by windbreaks.
Atmos Environ 35: 3373–83.
Ravi S, D’Odorico P, Over TM, and Zobeck TM. 2004. On the effect of air humidity on
soil susceptibility to wind erosion: the case of air-dry soils. Geophys Res Lett 31:
L09501.
Ravi S, D’Odorico P, Zobeck TM, et al. 2007. Feedbacks between fires and wind erosion
in heterogeneous arid lands. J Geophys Res-Biogeo 112: G04007.
Reheis MC, Budahn JR, Lamothe PJ, and Reynolds RL. 2009. Compositions of modern
dust and surface sediments in the desert Southwest, USA. J Geophys Res-Earth 114:
F01028, doi:10.1029/2008JF001009.
Reynolds R, Neff J, Reheis M, and Lamothe P. 2006. Atmospheric dust in modern soil on
aeolian sandstone, Colorado Plateau (USA): variation with landscape position and
contribution to potential plant nutrients. Geoderma 130: 108–23.
Reynolds JF, Smith DMS, Lambin EF, et al. 2007. Global desertification: building a
science for dryland development. Science 316: 847–51.
Ryan MG, Archer SR, Birdsey R, et al. 2008. Land resources. In: Walsh M (Ed). The
effects of climate change on agriculture, land resources, water resources, and
biodiversity in the United States. Washington, DC: US Department of Agriculture.
Schlesinger WH, Reynolds JF, Cunningham GL, et al. 1990. Biological feedbacks in
global desertification. Science 247: 1043–48.
47
Seager R, Ting MF, Held I, et al. 2007. Model projections of an imminent transition to a
more arid climate in southwestern North America. Science 316: 1181–84.
Shao Y, Raupach MR, and Findlater PJ. 1993. Effect of saltation bombardment on the
entrainment of dust by wind. J Geophys Res-Atmos 98: 12719–26.
Sharifi MR, Gibson AC, and Rundel PW. 1999. Phenological and physiological
responses of heavily dusted creosote bush (Larrea tridentata) to summer irrigation in
the Mojave Desert. Flora 194: 369–78.
Steltzer H, Landry C, Painter TH, et al. Biological consequences of earlier snowmelt
from desert dust deposition in alpine landscapes. P Natl Acad Sci USA. In press.
Stout JE and Zobeck TM. 1996. The Wolfforth field experiment: a wind erosion study.
Soil Sci 161: 616–32.
Tanaka TY and Chiba M. 2006. A numerical study of the contributions of dust source
regions to the global dust budget. Global Planet Change 52: 88–104.
Tegen I and Fung I. 1995. Contribution to the atmospheric mineral aerosol load from land
surface modification. J Geophys Res-Atmos 100: 18,707–18,706.
Toy TJ, Foster GR, and Renard KG. 2002. Soil erosion: processes, prediction,
measurement and control. New York, NY: John Wiley & Sons.
Whicker JJ, Breshears DD, Wasiolek PT, et al. 2002. Temporal and spatial variation of
episodic wind erosion in unburned and burned semiarid shrubland. J Environ Qual
31: 599–612.
48
Utz EJ, Kellogg CE, Reed EH, et al. 1938. The problem: the nation as a whole. In:
Bennett HH and Lowdermilk WC (Eds). Soils and men. Yearbook of agriculture.
Washington, DC: US Department of Agriculture.
Yoshioka M, Mahowald N, Conley A, et al. 2007. Impact of desert dust radiative forcing
on Sahel precipitation: relative importance of dust compared to sea surface
temperature variations, vegetation changes and greenhouse gas warming. J Climate
20: 1445–67, doi:10.1175/JCLI4056.1.
49
Figures
Figure 1. (a) Individual plant scale: shrub with exposed roots following the aftermath of a
San Joaquin Valley dust storm in December 1977. (b) Local scale: dust storm
approaching Stratford, Texas, April 18, 1935. (c) Regional scale: dust storm (tan pixels)
on July 23, 2003, originating in the western part of Utah and blowing northeastward over
Great Salt Lake. (d) Global scale: dust transport over the Atlantic Ocean, shown here off
the west coast of northern Africa on March 2, 2003.
50
Figure 2. How wind erosion works. Saltation occurs when the shear velocity of the wind
exceeds the threshold shear velocity of the soil; suspension of dust-size particles occurs
when saltating particles sand-blast the soil surface, overcoming the strong interparticle
forces between fine particles.
51
Figure 3. Horizontal aeolian sediment flux as a function of the ratio of average
unvegetated gap size to plant height. The black line indicates how flux would vary in the
presence of an undisturbed herbaceous layer for each ecosystem; the gray line indicates
how flux might vary in the presence of a disturbance that removed most of the
herbaceous layer. Flux rates are calculated for realistic wind conditions in south-central
New Mexico. Flux rates are calculated for a wind shear velocity of 100 cm/s, and a sandy
loam soil with a threshold shear velocity of 24 cm/s is assumed. Woodland applies to
lands that have a mix of small-stature trees and open spaces. Adapted from Okin (2008)
and Breshears et al. (2009).
52
Sand
Figure 4. Wind tunnel results from undisturbed and disturbed (two passes with a fourwheel drive vehicle) soils of different textures. Top panel: Wind speeds at which
sediment is detached from the soil surface (TFV). Thus, a higher value denotes greater
soil stability. Bottom panel: The horizontal aeolian sediment flux; thus, a higher value
denotes lower soil stability and greater sediment production. Error bars represent
plus/minus one standard error. An * denotes statistical differences (P < 0.05). “Y”
indicates that wind speeds generated by the tunnel were unable to disrupt the soil surface
in the control soils; thus, reported values are conservative.
53
Figure 5. Primary feedbacks between ecosystem function, wind erosion, and ecosystem
structure.
54
Figure 6. Dust-laden snow on Mount Sopris, Colorado, May 2007 (photo: P. Newhard).
55
a
b
Figure 7. Map showing global distribution of (a) drylands and (b) their respective dust
emissions by region.
56
APPENDIX B
DUST AMPLIFICATION AND CAPTURE AMONG VEGETATION PATCH TYPES:
SUPPORT FOR UNDERLYING DESERTIFICATION ASSUMPTIONS
Jason P. Field1, David D. Breshears1,2, Jeffrey J. Whicker3 & Chris B. Zou4
1
School of Natural Resources and the Environment, University of Arizona, Tucson, AZ
85721 USA; 2Department of Ecology and Evolutionary Biology, Institute of the
Environment, University of Arizona, Tucson, AZ 85721; 3Environmental Programs, Los
Alamos National Laboratory, Los Alamos, NM 87545; 4Department of Natural Resource
Ecology and Management, Oklahoma State University, Stillwater, OK 74078.
Document Type: In review as of Dec 4, 2009.
Abstract: Desertification impacts a large proportion of globally extensive drylands and
can be driven by a variety of climate and land use factors in these coupled humanenvironmental systems. Most conceptual models of desertification include the
underlying assumption that when herbaceous cover is reduced, increased erosion from
bare patches is redistributed to shrub canopy patches, resulting in self-reinforcing
“islands of fertility”. Notably, however, this underlying assumption at the scale of
vegetation patches has not been explicitly tested with direct measurements of aeolian
processes. We tested this desertification hypothesis directly by measuring aeolian
transport into and out of bare-, herbaceous-, and shrub-dominated patch types in a
semiarid ecosystem using BSNE samplers both for simulated and natural dust events, as
well as in response to simulated disturbance. Vegetation patch types differed as expected,
with shrub patches capturing more dust than herbaceous patches, and importantly, bare
patches serving as amplified dust sources with more dust leaving than entering a patch.
Simulated disturbances that removed herbaceous vegetation and disturbed the soil surface
57
confirmed hypothesized transformations associated with desertification. Our results
provide explicit support for a pervasive underlying but untested desertification hypothesis
and can aide in more effectively managing vegetation patches and soil surface stability.
Introduction
Arguably the most central land management challenge for many of the globally
extensive drylands is dealing with desertification in these coupled human-environmental
systems (Schlesinger et al 1990, Kassas 1995, Reynolds et al. 2007, Henderson-Sellers et
al. 2008, Hill et al. 2008). Desertification can occur in many ways (Grover and Musick
1990, Archer et al. 1995, Van Auken 2000, Peters et al. 2006, Okin et al. 2009), yet
perhaps the most widely applied concept related to desertification is that of the “islands
of fertility”, in which herbaceous cover is reduced and surface soils are disturbed through
processes such as heavy grazing or off road vehicle use. Vegetation patterns in most
drylands are usually characterized by the size, shape, and spatial distribution of high
plant-cover patches within a matrix of low-cover patches (Aguiar and Sala 1999). The
island of fertility concept is central to earlier conceptual models of desertification
(Charley and West 1975, Schlesinger et al. 1990, Rostagno et al. 1991) and remains
embedded within many more recent and comprehensive ones (Peters et al. 2006, Yang et
al. 2007, Wang et al. 2008, Okin et al. 2009).
The sometimes implicit hypothesis associated with the “island of fertility”
assumption is that resource redistribution by wind-driven processes in drylands is a
central driver of desertification (Schlesinger and Pilmanis 1998, Okin et al. 2009).
58
Interactions between vegetation-patch types and wind-driven processes can gives rise to
increased resource heterogeneity at the landscape scale from the redistribution of soil
from bare intercanopy patches to under shrub canopy patches (Hennessy et al. 1985,
Rostagno and del Valle 1988, Coppinger et al. 1991). Because of the sparse vegetation
cover characteristic of most drylands, these systems are generally highly susceptible to
wind erosion and wind-driven redistribution (Okin et al. 2006, Breshears et al. 2009,
Field et al. 2009). Wind erosion can have significant adverse impacts on soil productivity
by removing the fine soil particles and reducing nutrient concentrations and organic
matter content (Leys and McTainsh 1994, Li et al. 2008). The type and structure of
vegetation patches within these dryland ecosystems, as well as their spatial patterns
across the landscape, has a large impact on the aerodynamic roughness length and the
threshold wind shear velocity and thus controls to a large extent the transport and
redistribution of wind-blown material (Wolfe and Nickling 1993, Landcaster and Baas
1998, Gillette et al. 2006, Breshears et al. 2009). Consequently, soil and nutrients can be
either transported by wind-driven processes offsite or redistributed from the low-cover
patches that serve as sources to high-cover patches that serve as sinks. This wind-driven
redistribution of soil and nutrients from source to sink areas can progressively reinforce
patterns of wind erosion and exacerbate the process of desertification (Gibbens et al.
1983, Li et al. 2005, Okin et al. 2009). Water erosion can also play an interactive role
with wind erosion in the desertification process (Schlesinger and Pilmanis 1998,
Schlesinger et al. 2000, Ravi et al. in review), but aeolian processes remain the primary
59
hypothesized mechanism by which material accumulates beneath shrubs (Coppinger et al.
1991, McAuliffe et al. 2007, Bowker et al. 2008, Ravi et al. 2009).
In short, although most conceptual models of desertification include the
underlying assumption that when herbaceous cover is reduced, erosion increases from
bare patches and is redistributed to shrub canopy patches, this underlying assumption at
the scale of vegetation patches has not been explicitly tested with direct measurements of
aeolian processes. We tested this underlying desertification hypothesis by directly
measuring aeolian sediment transport into and out of bare-, herbaceous-, and shrubdominated patch types in a semiarid ecosystem for both simulated and natural dust
events, as well as in response to simulated disturbance. Our approach was to isolate
inputs and outputs from each patch type using directionally fixed samplers that then
allowed us to assess dust amplification or capture. We discuss the importance of our
findings relative to concepts of desertification and to management challenges in drylands.
Materials and Methods
Study Site
The study was conducted in a semiarid ecosystem at the University of Arizona
Santa Rita Experimental Range (31°50’’N, 110°50’’W) approximately 50 km south of
Tucson, Arizona (McClaran et al. 2002). The study plots were located on Sandy Loam
Upland (SLU) Ecological Sites that occupy recently deposited Holocene alluvial fan and
fan terrace surfaces with ≤8% slopes, sandy loam soils to about 15-cm depth, and 5-25%
gravel at the surface (Breckenfeld and Robinett 2003). The site was located
approximately 1100 m above mean sea level, which is near the low elevation limit of the
60
desert grassland and above the desert shrub (McClaran 2003). Mean annual air
temperature at this location is approximately 16ºC, with daily maximum air temperatures
exceeding 35ºC in summer (McClaran et al. 2002). Long-term (1923-2003) annual
precipitation near the study plots is approximately 350 mm and is bimodally distributed
with more than half of the total annual precipitation occurring during the summer
monsoon (July-September) with drier fall and spring months separating the wetter winter
(January-March) and summer months (Sheppard et al. 2002).
Plot Selection
We established nine plots (3 × 3 m) in three replicated blocks immediately
adjacent to a north-south oriented dirt road, which is perpendicular to the predominant
wind direction at the site. Each of the replicated blocks consisted of three plot types,
including bare-, herbaceous-, and shrub-dominated vegetation patches. All study plots
were equal distance from the leeward edge (east side) of the dirt road to minimize the
effect of distance from the road on rates of aeolian sediment transport. In addition, all
vegetation between the leeward edge of the road and the windward edge of the study
plots was removed to minimize the effect of vegetation structure outside of the plots on
rates of aeolian sediment transport. Herbaceous and woody vegetation was removed by
clipping the base of the vegetative growth at the soil surface with a pair of grass clippers
or loppers until a uniform, low level (< 5%) of cover was achieved between the leeward
edge of the dirt road and the windward edge of the study plots. Herbaceous cover prior to
removal was approximately 60%, with Lehmann lovegrass (Eragrostis lehmanniana)
constituting the majority (> 90%) of the grass cover. Woody plant cover prior to removal
61
was approximately 10%, with low-statured (< 1 m) velvet mesquite (Prosopis velutina)
constituting the majority of the shrub cover.
The study plots consisted of three patch types (bare, herbaceous, and shrub) and
were selected based on the following criteria. Bare patches had < 10% herbaceous cover
and < 10% shrub cover, with total cover (i.e., herbaceous, woody, litter, rocks) ≤ 20%.
Herbaceous patches had a minimum of 60% grass cover, with < 10% shrub cover and <
20% bare soil. Shrub-dominated patches were centered around the trunks of mid-sized
Velvet mesquites (2-3 m tall, 3-4 m canopy diameter) to ensure adequate coverage of the
3 × 3 m plot area. The windward edge of the study plots were located exactly 5 m from
the leeward edge of the dirt road and were separated from neighboring plots by at least 3
m of undisturbed buffer.
Laboratory and Field Measurements
Soil samples were collected from four locations within each of the nine plots at
the end of the study period to avoid disturbing the plot surface. Soil cores (5-cm
diameter, 20-cm deep) were collected near each of the plot corners and aggregated into
single composite samples for each of the nine study plots. Soil samples were oven-dried
at 60ºC to achieve constant weight. Particle-size distribution was determined using the
hydrometer method (Bouyoucos 1962). Vegetation cover was measured within each plot
using the line-point intercept method (Herrick et al. 2005) along two 3-m transects that
extend the full length of the plots. Meteorological data, including wind speed and wind
direction, was collected onsite within 100 m of the study plots (Campbell Scientific,
Logan, UT).
62
Horizontal dust flux was measured using a series of BSNE (Big Spring Number
Eight) dust samplers (Fryrear 1986) at four heights aboveground (0.08, 0.25, 0.50, and
1.0 m). The dust samplers were installed at the windward and leeward edges of the study
plots so that the openings of the BSNEs were directionally-fixed parallel to the plot edge.
Dust samples were collected immediately after each wind event by rinsing the BSNEs
with DI water and collecting the rinsate in 20-ml glass vials, which were then oven-dried
at 60ºC to achieve constant weight.
Types of Dust Events
Horizontal dust flux measurements were made under both natural and simulated
dust events for disturbed and undisturbed field conditions. Eight simulated dust events
were generated during the study period under similar wind conditions by driving a pickup
truck at 65 kph back and forth along the dirt road that runs parallel and adjacent to the
study plots. Each simulated event consisted of 30 passes along the dirt road using the
same technique to minimize variability in the dust source (i.e., the same pickup truck was
used for each event and was driven as close as possible directly down the center of the
road with cruse control to achieve constant speed). Natural dust events were observed on
four separate occasions during the study period and meet the following criteria. Natural
events had a minimum average wind speed of approximately 4 m s-1 with guests ranging
from about 8 to 13 m s-1 and an average wind direction of about 270º ±30º for the
duration of the 2-hour natural events. Four additional simulated dust events were
generated to evaluate changes in dust flux patterns among the three different patch types
following disturbance. All of the 3 × 3 m study plots were moderately disturbed to
63
simulate the effects of livestock grazing. The simulated grazing treatment consisted of
clipping all herbaceous vegetation within the study plots to approximately 2-3 cm above
the soil surface. In addition, the litter cover was removed from the plots, and the soil
surface was lightly disturbed with a gravel rake to simulate surface disturbance that can
occur under low-intensity grazing regimes.
Data Analysis
Field measurements of horizontal dust flux (g m-1 h-1) were made at the leeward
and downwind edges of the study plots to estimate net dust flux and dust amplification /
capture efficiency. Total horizontal dust flux was calculated by integrating flux
measurements from 0 to 1 m above the soil surface using a simple exponential
relationship with height (Gillete et al. 1997, Stout 2001). Net dust flux (g m-2 h-1) is
referred to here as the total horizontal dust flux moving into the leeward edge of the patch
minus the total horizontal dust flux moving out of the downwind edge of the plot divided
by the plot length. Positive net dust flux values indicate deposition of sediment into the
plot; negative net dust flux values indicate sediment production or dust amplification
within the plot. Dust amplification / capture efficiency is defined here as the percentage
of total horizontal dust flux moving into the plot minus the total horizontal dust flux
moving out of the plot divided by the total horizontal dust flux into the plot.
We used a one-way ANOVA to test for significant differences (P < 0.05) in the
mean values of horizontal dust flux moving into the plots, horizontal dust flux moving
out of the plots, net dust flux, and dust amplification / capture efficiency. Analysis of
variance was carried out according to the general linear model procedure of the statistical
64
analysis system and the type III sums of squares using JMP® 8.0 (SAS, Cary, NY).
Results were considered significant at the α level of 0.05. When significant differences
among mean values were detected, either the Student’s t-test or Tukey’s HSD (Honesty
Significant Difference) test were used to separate means.
Results
Simulated dust events
Average horizontal dust flux into the three vegetation patch types was similar for
the eight simulated dust events. No significant differences were detected in the average
amount of wind-blown sediment moving into the bare, herbaceous, and woody patches
for the simulated dust events (P < 0.0001, Tukey’s HSD test; Fig. 1a), suggesting that the
simulated dust events were fairly uniform from the source (dirt road) to the leading edge
of the vegetation patches. Horizontal dust flux out of the three vegetation patch types
was significantly greater in the bare patches (64.1 g m-1 h-1) than in the vegetated patches,
and dust flux out of the herbaceous patches (52.4 g m-1 h-1) was significantly greater than
that out of the woody patches (42.9 g m-1 h-1) (P < 0.0001, Tukey’s HSD test; Fig. 1b).
The difference in the amount of horizontal dust flux out of the three patch types is
strongly dependant on the type and structure of the vegetation within the 3 × 3 m patches
since all other factors (e.g., wind speed, soil texture) were fairly uniform within the study
plots. Net dust flux, referred to here as the horizontal dust flux into the vegetation patch
minus the horizontal dust flux out of the vegetation patch, was significantly different
among the three patch types (P < 0.0001, Tukey’s HSD test; Fig. 1c). Net dust flux into
the herbaceous patches was approximately 2.8 g m-2 h-1, whereas net dust flux into the
65
woody patches was nearly twice as great for the simulated dust event (4.4 g m-2 h-1).
However, net dust flux for the bare patches was negative (-1.9 g m-2 h-1), indicating that
bare patches are likely producing dust rather than capturing it as was observed for the
vegetated patches.
Natural dust events
Four periods of natural dust events were evaluated during the study period to
assess how patterns of aeolian sediment transport varied between simulated and natural
dust events. Overall, the patterns of aeolian sediment transport into and out of the three
patch types were similar between the natural and simulated dust events although the
average amount of material being transported during each event was considerably greater
for the simulated events (Fig. 2). No significant differences were detected in the amount
of horizontal dust flux moving into the three patch types (P > 0.5895; Fig. 2a); however,
unlike the simulated events, there were also no significant differences detected in the
amount of horizontal dust flux moving out of the three patch types (P > 0.0738; Fig. 2b).
Dust flux by height
The efficiency of the different patch types to capture wind-blown sediment during
natural and simulated events varied considerably by height. At 1 m above the soil
surface, both the bare and herbaceous patches behaved similarly and had little effect on
the amount of horizontal dust flux through the patch (P < 0.0001, Tukey’s HSD test; Fig.
3a). The woody patches were more than 5 times more efficient at capturing wind-blown
material at 1 m above the soil surface than the bare and herbaceous patches. Differences
66
between the capture efficiency of the bare and herbaceous patches increases significantly
with decreasing height above the soil surface (Fig 3). The greatest difference in the
capture efficiency between the bare and herbaceous patches was observed at the lowest
sampling height (8 cm), whereas the greatest difference between the herbaceous and
woody patches was observed at the greatest sampling height (1 m).
Effects of disturbance
Disturbance can have a large impact on the soil surface and the spatial patterns of
vegetation, both of which are critical factors that influence wind erosion and aeolian
processes in general. Net dust flux for four simulated dust events following moderate
disturbance (simulated grazing and light surface disturbance) indicates that disturbance
can have a large effect on the ability of vegetation patch types to capture wind-blown
material. Net dust flux in the bare patches did not change significantly following
disturbance; however, net dust flux into the vegetated patches did decrease significantly
following disturbance for herbaceous (P < 0.0001) and shrub-dominated patches (P <
0.0025, Tukey’s HSD test; Fig. 4). The most dramatic difference in net dust flux
following disturbance was observed in the herbaceous patches. These patch types
surprisingly shifted from serving as a sediment sink prior to disturbance to serving as a
sediment source area following moderate disturbance. Woody patches were more
resilient following disturbance and maintained their ability to capture wind-blown
sediment under simulated dust events.
67
Discussion
Our measurements of horizontal dust flux at the vegetation patch scale provide
some of the first field-based estimates of net dust flux moving into or out of bare-,
herbaceous-, and shrub-dominated patches. These patch types are the basic fundamental
components of most dryland ecosystems and play an important role in influencing
patterns of aeolian sediment transport and deposition across the landscape (Landcaster
and Baas 1998, Aguiar and Sala 1999, Li et al. 2005, Li et al. 2007). Our results show
that vegetation patch type has a substantial effect on aeolian sediment transport and the
redistribution of soil and other resources. More specifically, our results indicate that
under most wind events bare patches had no effect on capturing wind-blown material but
instead actually served as a dust source through amplification of the incoming horizontal
dust flux. These results are consistent with both early and more recent conceptual models
of desertification and provide direct support for a key underlying assumption in many of
these models that bare patches serve as sediment sources for the development of shrub
islands or islands of fertility (Schlesinger et al 1990, Van Auken 2000, Okin et al. 2009).
Our results also provide direct support for the greater ability of shrub-dominated patches
to capture wind-blown material relative to bare and herbaceous patches, which is another
key assumption central to many earlier and more recent conceptual models of
desertification that remains largely unsupported by field measurements (Schlesinger et al.
1990, Rostagno et al. 1991, Peters et al. 2006, Yang et al. 2007, Wang et al. 2008).
Disturbance, particularly livestock grazing, is one of the most frequently reported
causes of desertification (Archer et al. 1995, Van Auken 2000), but the effects of
68
disturbance on patterns and rates of aeolian sediment transport at the vegetation patch
scale are largely lacking. We found that moderate disturbance had the greatest effect on
the ability of herbaceous patches to capture wind-blown material relative to bare- and
shrub-dominated patches. Notably, we found that herbaceous patches following
moderate disturbance (i.e., simulated grazing) actually shifted from being a sediment sink
to serving as a sediment source, which can have profound implications for the
amplification of the desertification process (Fig. 5). In a relative undisturbed system with
good herbaceous cover, a large fraction of the dust and other wind-blown material
originating from the bare (source) patches is likely captured to a comparable extent by
both herbaceous- and shrub-dominated patches. The ability of shrub islands to form in
relatively undisturbed drylands with good herbaceous cover is probably constrained by
competition from the herbaceous patches to capture wind-blown material and by the lack
of available sediment since only bare patches serve as dust sources prior to disturbance.
However, once the herbaceous cover is removed due to moderate disturbance such as
grazing, shrub-dominated patches are more likely to develop shrub islands beneath their
woody canopies due to the greater influx of wind-blown material not only from the bare
patches but also from the herbaceous patches as well, which lose their ability to capture
material and actually become dust sources following disturbance (Fig. 5). Our results
provide a measurement-based mechanism by which shrub island development can be
proliferated following moderate disturbance such as grazing. Further, our results
highlight that the ability of shrub-dominated patches to capture wind-blown material is
only slightly diminished following disturbance and suggest that shrub islands, once
69
formed, may be more resilient to disturbance than herbaceous patches – a concept that is
often implicitly assumed in many conceptual models of desertification (e.g., Whitford et
al. 1995).
Desertification is often associated with a decrease in herbaceous cover and an
increase in woody plant cover, which typically results in increased resource heterogeneity
on the landscape (Schlesinger et al. 1990, Peters et al. 2006, Li et al. 2008). Several
studies have evaluated soil redistribution between canopy and intercanopy patches and
have identified several different physical and biological processes involved in the in the
development of shrub islands (Rostagno and del Valle 1988, Bochet et al. 2000). Some
studies have evaluated the effects of single vegetation elements on aeolian sediment
transport through direct measurements or modeling approaches (Leenders et al. 2007,
Bowker et al. 2008); however, our study is unique in the fact that it provides among the
first comparative estimates of the ability of three fundamental patch types (bare,
herbaceous, shrub) to capture wind-blown material under simulated or natural dust
events. Since few measurements of aeolian sediment transport at the scale of vegetation
patches are available, much of our current understanding of aeolian processes at the
vegetation patch scale is based on model approaches rather than field measurements,
often making it difficult to assess the ecological and management implications of dust
flux and wind-driven redistribution at these smaller spatial scales. Field-based
measurement of dust flux and aeolian processes in general have been studied to a greater
degree at larger spatial scales such as plot and landscape scales (Breshears et al. 2009);
70
however, aeolian processes at the scale of individual plants determines to a large extent
patterns of dust flux at larger spatial scales (Gillette et al. 2006, Okin et al. 2009).
In summary, desertification can be driven by a variety of climate and land use
factors in drylands, but a key underlying assumption in many conceptual models of
desertification is that when herbaceous cover is reduced, the redistribution of soil and
other resources from the bare patches to the shrub-dominated patches increases,
amplifying the development of shrub islands and exacerbating the desertification process.
Several studies have shown that the type and structure of vegetation patches in dryland
ecosystems can have a large influence on the redistribution of soil by wind (Landcaster
and Baas 1998, Leenders et al. 2007, Li et al. 2007, Bowker et al. 2008), but this
assumption had not been explicitly tested with direct measurements of aeolian processes
across the three fundamental patch types, including bare-, herbaceous-, and shrubdominated patches. Our findings indicate that aeolian processes are at least partially
responsible for increasing the spatial heterogeneity of resources on a landscape through
wind-driven redistribution of sediment and other resource among bare-, herbaceous-, and
shrub-dominated patches. Differences in the ability of the vegetation patch types to
capture wind-blown material was large, with shrub patches capturing more dust than
herbaceous patches, and bare patches serving as dust amplification sources. Disturbance
had little effect on bare- and shrub-dominated patches but transformed herbaceous
patches from sediment sinks to dust sources. These findings are notable in that they
provide explicit support for a pervasive but untested desertification hypothesis and offer
71
insights for more effectively managing soil resources and controlling erosion at the
vegetation patch scale.
Acknowledgements
This study was supported by the Arizona Agricultural Experiment Station (DDB), the
U.S. Department of Agriculture Cooperative State Research, Education, and Extension
Service (JPF, DDB; CSREES 2005-38420-15809), the National Science Foundation
(DDB, JPF; NSF-DEB 0816162), and the Department of Energy (JJW; DE-AC5206NA25396).
References
Aguiar, M. R., and O. E. Sala. 1999. Patch structure, dynamics and implications for the
functioning of arid ecosystems. Trends in Ecology and Evolution 14(7): 273-277.
Archer, S., D. S. Schimel, and E. A. Holland. 1995. Mechanisms of shrubland expansion:
land use, climate or CO2? Climatic Change 29(1):91-99.
Breshears, D. D., J. J. Whicker, C. B. Zou, J. P. Field, C. D. Allen. 2009. A conceptual
framework for dryland aeolian sediment transport along the grassland–forest
continuum: effects of woody plant canopy cover and disturbance. Geomorphology
105(1-2):28-38.
Bochet, E., J. Poesen, and J. L. Rubio. 2000. Mound development as an interaction of
individual plants with soil, water erosion and sedimentation processes on slopes.
Earth Surface Processes and Landforms 25(8):847-867.
72
Bouyoucos, G.J., 1962. Hydrometer method improved for making particle size analyses
of soils. Agron. J. 54: 464-465.
Bowker, G. E., D. A. Gillette, G. Bergametti, B. Marticorena, and D. K. Heist. 2008.
Fine-scale simulations of aeolian sediment dispersion in a small area in the northern
Chihuahuan Desert. Journal of Geophysical Research-Earth Surface 113(F2):F02S11,
doi:10.1029/2007/JF000748.
Breckenfeld, D.J., and D. Robinett. 2003. Soil and ecological sites on the Santa Rita
Experimental Range. In: McClaran, M.P., Ffolliott, P.F., Edminster, C.B. (tech.
coords.), Santa Rita Experimental Range: 100 years (1903-2003) of
Accomplishments and Contributions; Conference Proceedings. US Department of
Agriculture, Forest Service, Rocky Mountain Research Station Proceedings, Tucson,
AZ, RMRS-P-30, October 30-November 1, pp. 157-165.
Charley, J. L., and N. E. West. 1975. Plant-induced soil chemical patterns in some shrubdominated semi-desert ecosystems of Utah. Journal of Ecology 63(3):945-964.
Coppinger, K. D., W. A. Reiners, I. C. Burke, and R. K. Olson. 1991. Net erosion on a
sagebrush steppe landscape as determined by cesium-127 distribution. Soil Science
Society of America Journal, 55(1):254-258.
Field, J. P., D. D. Breshears, and J. J. Whicker. 2009. Toward a more holistic perspective
of soil erosion: why aeolian research needs to explicitly consider fluvial processes
and interactions. Aeolian Research 1(1-2):9-17.
Fryrear, D. W. 1986. A field dust sampler. Journal of Soil and Water Conservation 41:
117–120.
73
Gillette, D.A., D.W. Fryrear, T.E. Gill, T. Ley, T.A. Cahill, and E.A. Gearhart. 1997.
Relation of vertical flux of particles smaller than 10 µm to total aeolian horizontal
mass flux at Owens Lake. Journal of Geophysics Research 102 (D22): 26009-26015.
Gillette, D. A., J. E. Herrick, G. A. Herbert. 2006. Wind characteristics of mesquite
streets in the northern Chihuahuan Desert, New Mexico, USA. Environmental Fluid
Mechanics 6(3):241-275.
Grover, H. D., and H. B. Musick. 1990. Shrubland encroachment in southern New
Mexico, USA – an analysis of desertification processes in the American Southwest.
Climatic Change 17(2-3):305-330.
Henderson-Sellers, A., P. Irannejad, and K. McGuffie. 2008. Future desertification and
climate change: The need for land-surface system evaluation improvement. Global
and Planetary Change 64(3-4):129-138.
Hennessy, J. T., R. P. Gibbens, J. M. Tromble, and M. Cardenas. 1985. Mesquite
(Prosopis glandulosa Torr.) dunes and interdunes in southern New Mexico: a study
of soil properties and soil water relations. Journal of Arid Environments 9(1): 27-38.
Herrick, J.E., J.W. Van Zee, K.M. Havstad, L.M. Burkett, and W.G. Whitford. 2005.
Monitoring Manual for Grassland, Shrubland and Savanna Ecosystems, Volume I,
Quick Start. The University of Arizona Press, Tucson, Arizona.
Hill, J., M. Stellmes, T. Undelhoven, A. Roder, and S. Sommer. 2008. Mediterranean
desertification and land degradation mapping related land use change syndromes
based on satellite observations. Global and Planetary Change 64(3-4):146-157.
74
Kassas, M. 1995. Desertification – a general review. Journal of Arid Environments
30(2):115-128.
Lancaster, N., and A. Baas. 1998. Influence of vegetation cover on sand transport by
wind: field studies at Owens Lake, California. Earth Surface Processes and
Landforms 23(1):69-82.
Leenders, J. K., and J. H. van Boxel, and G. Sterk. 2007. The effect of single vegetation
elements on wind speed and sediment transport in the Sahelian Zone of Burkina Faso.
Earth Surface Processes and Landforms 32(10):1454-1474.
Leys, J., and G. McTainsh. 1994. Soil loss and nutrient decline by wind erosion: cause
for concern. Australian Journal of Soil and Water Conservation 7(3):30-40.
Li, F. R., L. F. Kang, H. Zhang, L. Y. Zhao, Y. Shirato, and I. Taniyama. 2005. Changes
in intensity of wind erosion at different stages of degradation development in
grasslands of Inner Mongolia, China. Journal of Arid Environments 62(4):567-585.
Li, J., G. S. Okin, J. J. Alvarez, and H. E. Epstein, 2007. Quantitative effects of
vegetation cover on wind erosion and soil nutrient loss in a desert grassland of
southern New Mexico, USA. Biogeochemistry 85(3):317-332.
Li, J., G. S. Okin, J. J. Alvarez, and H. E. Epstein, 2008. Effects of wind erosion on the
spatial heterogeneity of soil nutrients in a desert grassland of southern New Mexico.
Biogeochemistry 88(1):73-88.
McAuliffe, J. R., E. P. Hamerlynck, and M. C. Eppes. 2007. Landscape dynamics
fostering the development and persistence of long-lived creosotebush (Larrea
tridentata) clones in the Mojave Desert. Journal of Arid Environments 69(1):96-126.
75
McClaran, M.P., D.L. Angell, and C. Wissler. 2002. Santa Rita Experimental Range
Digital Database: User’s Guide. US Department of Agriculture, Forest Service,
Rocky Mountain Experiment Station General Technical Report RM GTR-100.
McClaran, M.P., 2003. A Century of vegetation change on the Santa Rita Experimental
Range. In: Santa Rita Experimental Range: 100 Years (1903– 2003) of
accomplishments and contributions. Proceedings RMRS-P-30. U.S. Department of
Agriculture Forest Service, Rocky Mountain Research Station, Fort Collins CO.
Nash, M. S., E. Jackson, and W. G. Whitford. 2004. Effects of intense, short-duration
grazing on microtopography in a Chihuahuan Desert grassland. Journal of Arid
Environments 56(3):383-393.
Okin, G. S., J. E. Herrick, and D. A. Gillette. 2006. Multiscale controls on and
consequences of aeolian processes in landscape change in arid and semiarid
environments. Journal of Arid Environments 65(2):253-75.
Okin, G. S., A. J. Parsons, J. Wainwright, J. E. Herrick, B. T. Bestelmeyer, D. P. C.
Peters, and E. L. Fredrickson. 2009. Do changes in connectivity explain
desertification? BioScience 59(3):237-244.
Parsons, A. J., A. D. Abrahams, and J. R. Simanton. 1992. Microtopography and soilsurface materials on semi-arid piedmont hillslopes, southern Arizona. Journal of Arid
Environments 22(2):107-115.
Peters, D. P. C., B. T. Bestelmeyer, and J. E. Herrick. 2006. Disentangling complex
landscapes: new insights into arid and semiarid system dynamics. Bioscience
56(6):491-501.
76
Ravi, S., D. D. Breshears, T. E. Huxman, and P. D’Odorico. Land degradation in
drylands: interactions among hydrologic-aeolian erosion and vegetation dynamics.
Geomorphology In Review.
Ravi, S., P. D’Odorico, L. X. Wang, C. S. White, G. S. Okin, S. A. Macko, and S. L.
Collins. Post-fire resource redistribution in desert grasslands: a possible negative
feedback on land degradation. Ecosystems 12(3):434-444.
Reynolds, J. F., D. M. Stafford-Smith, E. F. Lambin, B. L. Turner, M. Mortimore, S. P.
J. Batterbury, T. E. Downing, H. Dowlatabadi, R. J. Fernandez, J. E. Herrick, E.
Huber-Sannwald, H. Jiang, R Leemans, T. Lynam, F. T. Maestre, M. Ayarza, B.
Walker. 2007. Global desertification: building a science fro dryland development.
Science 316(5826):847-851.
Rostagno, C. M., and H. F. del Valle. 1988. Mounds associated with shrubs in aridic soils
of North-eastern Patagonia: characteristics and probable genesis. Catena 15(3-4):347359.
Rostagno, C. M., H. F. Del Valle, and L. Videla. 1991. The influence of shrubs on some
chemical and physical properties of an aridic soil in north-eastern Patagonia,
Argentina. Journal of Arid Environments 20(2):179-188.
Schlesinger, W. H., and A. M. Pilmanis. 1998. Plant-soil interactions in deserts.
Biogeochemistry 42(1-2):169-187.
Schlesinger, W. H., J. F. Reynolds, G. L. Cunningham, L. F. Huenneke, W. M. Jarrell, R.
A. Virginia, and W. G. Whitford. 1990. Biological feedbacks in global desertification.
Science 247(4946):1043-1048.
77
Schlesinger, W. H., T. J. Ward, and J. Anderson. 2000. Nutrient losses in runoff from
grassland and shrubland habitats in southern New Mexico: II. Field plots.
Biogeochemistry 49(1):69-86.
Sheppard, P.R., A.C. Comrie, G.D. Packin, K. Augersbach, and M.K. Hughes. 2002. The
Climate of the US Southwest. Climate Research 21: 219-238.
Stout, J.E. 2001. Dust and environment in the Southern High Plains of North America.
Journal of Arid Environments 47: 425-441.
Van Auken, O. W. 2000. Shrub invasion of North American semiarid grassland. Annual
Reviews of Ecology and Systematics 31:197-215
Wang, X. M. F. Chen, E. Hasi, and J. C. Li. 2008. Desertification in China: An
assessment. Earth-Science Reviews 88(3-4):188-206.
Whitford, W. G., G. Martinez-Turanzas, and E. Martinez-Meza. 1995. Persistence of
desertified ecosystems: explanations and implications. Environmental Monitoring and
Assessment 37(1-3):319-332.
Wolfe, S. A., and W. G. Nickling. 1993. The protective role of sparse vegetation in wind
erosion. Progress in Physical Geography 17(1):50-68.
Yang, X., Z. Ding, X. Fan, Z. Zhou, and N. Ma. 2007. Processes and mechanisms of
desertification in northern China during the last 30 years, with a special reference to
the Hunshandake Sandy Land, eastern Inner Mongolia. Catena 71(1):2-12.
78
Figures
Figure 1. Mean horizontal dust flux (a) into and (b) out of bare-, herbaceous-, and shrub
dominated patches and (c) net dust flux into (deposition) and out of (production) the three
patch types for eight simulated dust events under relatively undisturbed conditions. Error
bars represent the standard error of the mean; means with the same letter do not differ
significantly (P < 0.05).
79
Figure 2. Mean horizontal dust flux (a) into and (b) out of bare-, herbaceous-, and shrub
dominated patches and (c) net dust flux into (deposition) and out of (production) the three
patch types for four natural dust events under relatively undisturbed conditions. Error
bars represent the standard error of the mean; means with the same letter do not differ
significantly (P < 0.05).
80
Figure 3. Dust amplification / capture efficiency for bare-, herbaceous-, and shrubdominated patches at (a) 100 cm, (b) 50 cm, (c) 25 cm, and (d) 8 cm above the soil
surface for eight simulated dust events under relatively undisturbed conditions. Error
bars represent the standard error of the mean; means with the same letter do not differ
significantly (P < 0.05).
81
Figure 4. Net dust flux (height integrated values from 0 to 1 m above the soil surface) for
bare-, herbaceous-, and shrub-dominated patches at for eight simulated dust events under
relatively undisturbed conditions and for four simulated dust events following moderate
disturbance. Error bars represent the standard error of the mean; means within each patch
type that have the same letter do not differ significantly (P < 0.05).
82
Figure 5. Conceptual framework for dryland aeolian sediment transport dust
amplification / capture processes following moderate disturbance and the removal of
herbaceous cover.
83
APPENDIX C
A CONCEPTUAL FRAMEWORK FOR DRYLAND AEOLIAN SEDIMENT
TRANSPORT ALONG THE GRASSLAND-FOREST CONTINUUM: EFFECTS OF
WOODY PLANT CANOPY COVER AND DISTURBANCE
David D. Breshears1,2,3, Jeffrey J. Whicker4, Chris B. Zou1, Jason P. Field1, and
Craig D. Allen5
1
School of Natural Resources, 2Institute for the Study of Planet Earth, and 3Department
of Ecology and Evolutionary Biology, University of Arizona, Tucson, AZ, 85721-0043
USA. 4Health Physics Measurements Group, Los Alamos National Laboratory, Los
Alamos, NM 87545 USA. 5USGS Jemez Mountains Research Center, Los Alamos, NM
87544 USA.
Document Type: Peer-reviewed publication in Geomorphology.
Abstract: Aeolian processes are of particular importance in dryland ecosystems where
ground cover is inherently sparse because of limited precipitation. Dryland ecosystems
include grassland, shrubland, savanna, woodland, and forest, and can be viewed
collectively as a continuum of woody plant cover spanning from grasslands with no
woody plant cover up to forests with nearly complete woody plant cover. Along this
continuum, the spacing and shape of woody plant determine the spatial density of
roughness elements, which directly affects aeolian sediment transport. Despite the
extensiveness of dryland ecosystems, studies of aeolian sediment transport have generally
focused on agricultural fields, deserts, or highly disturbed sites where rates of transport
are likely to be greatest. Until recently, few measurements have been made of aeolian
sediment transport over multiple wind events and across a variety of types of dryland
ecosystems. To evaluate potential trends in aeolian sediment transport as a function of
woody plant cover, estimates of aeolian sediment transport from recently published
84
studies, in concert with rates from four additional locations (two grassland and two
woodland sites) are reported here. The synthesis of these reports leads to the
development of a new conceptual framework for aeolian sediment transport in dryland
ecosystems along the grassland-forest continuum.
The findings suggest that: (1) for relatively undisturbed ecosystems, shrublands
have inherently greater aeolian sediment transport because of wake interference flow
associated with intermediate levels of density and spacing of woody plants; and (2) for
disturbed ecosystems, the upper bound for aeolian sediment transport decreases as a
function of increasing amounts of woody plant cover because of the effects of the height
and density of the canopy on airflow patterns and ground cover associated with woody
plant cover. Consequently, aeolian sediment transport following disturbance spans the
largest range of rates in grasslands and associated systems with no woody plants (e.g.,
agricultural fields), an intermediate range in shrublands, and a relatively small range in
woodlands and forests. These trends are consistent with previous observations relating
large rates of wind erosion to intermediate values for spatial density of roughness
elements. The framework for aeolian sediment transport, which is also relevant to dust
fluxes, wind erosion, and related aeolian processes, is applicable to a diverse suite of
environmental challenges, including land degradation and desertification, dust storms,
contaminant transport, and alterations of the hydrological cycle.
Introduction
Dryland ecosystems cover a substantial portion of the terrestrial biosphere (House
et al., 2003). Vegetation cover is often relatively sparse and soils are often dry in these
85
ecosystems relative to more mesic ecosystems because of less precipitation and greater
evaporative demand (McPherson, 1997; Anderson et al., 1999; Whitford, 2002; Loik et
al., 2004). A key consequence of sparser vegetation cover and drier soils is the potential
for an increase in aeolian sediment transport and related processes of wind erosion and
dust flux (Toy et al., 2002). Further, many dryland ecosystems are undergoing
accelerated land degradation which affects sediment redistribution and erosional loss
through aeolian and fluvial processes (Schlesinger et al., 1990; Aguiar and Sala, 1999;
Breshears et al., 2003; Peters et al., 2006). An improved understanding of aeolian
processes is required to assess atmospheric, hydrologic, and biogeochemical processes
(Miller and Tegen, 1998; Reynolds et al., 2001, 2006a,b,c; Peters et al., 2006), as well as
to address a diverse suite of environmental challenges related to soil and water quality
(Toy et al., 2002; Lal, 2003), land quality and productivity (Lal, 1996; Toy et al., 2002),
and human health (Griffin et al., 2001; Whicker et al., 2006b).
Despite the fundamental importance of aeolian sediment transport and erosion
processes in dryland systems, few studies have estimated aeolian sediment transport in
the field, particularly for periods spanning multiple events over several months to a few
years. Even fewer measurements have been reported for disturbed and relatively
undisturbed ecosystems. Notably, previous debates about rates of erosion in dryland
ecosystems (and associated policy implications) have not directly addressed the large
knowledge gap associated with rates of wind erosion, focusing instead primarily on water
erosion (Crosson, 1995; Pimentel, 1995a,b; Nearing et al., 2000; Trimble and Crosson
2000a,b).
86
Dryland ecosystems are extensive globally and encompass a diverse set of major
types of vegetation that include grassland, shrubland, savanna, woodland, and forest
(House et al., 2003). A first-order descriptor for these types of vegetation is the amount
and stature of woody plant cover (Breshears, 2006). These dryland types of vegetation
can be viewed collectively as a continuum of woody plant cover spanning from
grasslands with no woody plant cover to forests with nearly complete woody plant
cover—referred to as the grassland-forest continuum. The height of woody plants often
increases with canopy cover along this conceptual gradient (Martens et al., 2000).
Woody plants affect many ecosystem properties beneath the canopies and around them,
and these effects can translate into trends along the grassland-forest continuum (Fig. 1a)
for such properties as near-ground solar radiation (Martens et al., 2000) or soil water
content (Breshears and Barnes, 1999). This perspective is relevant to numerous, diverse
environmental issues such as desertification and forest restoration (Breshears, 2006).
Notably, the height and spacing of woody plants are key determinants of surface
roughness and associated factors that are fundamental to aeolian processes. The mean
and variance of numerous ecosystem properties related to energy, water, and
biogeochemistry (e.g., carbon) are hypothesized to exhibit trends along the continuum.
Of particular relevance to aeolian sediment transport and associated wind erosion is that
the variance of many ecosystem properties is hypothesized to be greatest at an
intermediate amount of canopy cover—at a value that is often less than 50% woody
canopy cover because of the influence that woody plants have on adjacent intercanopy
patches (also referred to as the degree of “connectivity”; Fig. 1b; Breshears, 2006).
87
Aeolian sediment transport and associated wind erosion are likely related to the amount
of woody plant cover in an ecosystem because woody plants are generally the
predominant “roughness elements”. The height, width and spacing of the main roughness
elements in an ecosystem determine the predominant way in which vegetation influences
the vertical wind profile to produce one of three types of flow: isolated roughness, wake
interference, or skimming (Lee and Soliman, 1977; Lee, 1991a,b; Wolfe and Nickling,
1993; Fig. 2). In addition to serving as roughness elements, woody plants often increase
ground cover via large inputs of litter to the ground beneath the plant canopy.
Consequently, the effects of woody plants on aeolian processes (Aguiar and Sala, 1999;
Okin and Gillette, 2001) represent an important, but perhaps somewhat underappreciated, linkage between ecology and geomorphology (Urban and Daniels, 2006).
Despite the overall importance of aeolian processes in dryland ecosystems,
variability in aeolian sediment transport as a function of woody plant cover along the
grassland-forest continuum has yet to be systematically evaluated. The goal of this paper
is to address this issue. The specific objectives are: (1) compile measurements of aeolian
sediment transport as a function of woody plant canopy cover for undisturbed and
disturbed sites (from published studies and other measurements reported here), (2)
evaluate relevant interrelationships between boundary layer flows and land surface
characteristics associated with woody plant cover, and (3) propose a general framework
for aeolian sediment transport that builds on three key findings that emerged from
objectives 1 and 2—(i) among relatively undisturbed ecosystems, shrublands have
inherently greater aeolian sediment transport; (ii) for disturbed ecosystems, the upper
88
bound for aeolian sediment transport decreases with increasing amounts of woody plant
cover because of the height and density of the canopy effects on airflow patterns and
because the total amount of ground cover generally increases with woody plant cover;
and, (iii) grasslands and other systems with few woody plants span the largest range of
rates for aeolian sediment transport and have the largest values after disturbance,
shrublands have inherently large rates for aeolian sediment transport that can be increased
moderately by disturbance, and woodlands and forests have smaller rates, even following
disturbance. The implications of the conceptual framework are discussed relative to
major environmental challenges that include land degradation and desertification, dust
storms, contaminant transport, and alterations of the hydrological cycle.
Compilation of Measurements
Methods and measurements
Estimates of aeolian sediment transport, expressed as mass flux transported in the
horizontal direction (g m-2 d-1), were compiled from published studies and from new
measurements (detailed below) for ecosystems with variable coverage of woody plants.
Estimates were categorized as either relatively undisturbed sites or sites that had severe
recent disturbance, including agricultural sites, scraped sites, burned sites, sites with
extensive military tracks, and sites that were so degraded through historical grazing and
other land use that essentially no herbaceous cover remained between woody plants.
Estimates were included if field measurements were obtained using Big Spring Number
Eight samplers (BSNE; see Fryrear, 1986). One study that used a Bagnold-type sampler
was also included because a comparison with BSNE samplers indicated that these two
89
types of samplers provided estimates of aeolian sediment transport that were within 10%
of each other (Breshears et al., 2003). Measurement intervals spanned from 3 to 30
months. Studies that focused on individual events or were conducted using wind tunnels
were not included. In several cases, estimates of sediment transport were calculated from
data provided in the original studies and converted to a daily basis. Spatial density of
roughness elements (see 3.1 below) was calculated using percent woody plant cover,
height, and mean plant diameter, all of which were available for most sites. For some
sites woody plant cover was estimated from photos of the site or inferred from the text.
In two cases where plant height was not reported, an approximate height associated with
the woody plant species was assumed. Height and mean plant diameter were available
for several of the sites that spanned from small (~5%) to large (72%) amounts of canopy
cover. A predictive equation, derived from sites with available data, was applied to sites
where mean canopy diameter was missing: diameter = 1.44(height)0.44 (r2 = 0.97). Where
sediment transport was reported for multiple measurement heights, the rates were
averaged over the available heights to provide an approximate estimate of mean sediment
transport up through 1 m above ground. Disturbed sites were evaluated with respect to
pre-disturbance canopy cover. In particular, canopy cover for burned sites included
standing, burned, and sometimes dead trees as part of canopy cover. Although some
woody plants had lost foliage, the woody branching structure can still be considered a
roughness element. For one type of less common disturbance, tree thinning with heavy
machinery in which tree cover and ground cover were substantially reduced, pre- and
post-disturbance values of canopy cover were evaluated.
90
Aggregated estimates of sediment transport from ongoing studies were included
to supplement the overall assessment given the few such measurements in grasslands and
woodlands (Breshears et al., 2003; Vermeire et al., 2005; but see Baker and Jemison,
1991; Baker et al., 1995). These studies focus on estimating aeolian sediment transport
in the context of land disturbances and contaminant transport (more detailed evaluations
of these data sets will be reported elsewhere). Sediment transport was measured for
periods spanning 10-12 months for three of the sites and 30 months for the fourth site
(Frijolito; see below). Measurements for all four sites were obtained using BSNE
samplers (Zobeck et al., 2003) at various heights up to 1 m and collected roughly every 12 weeks. Samples were collected, dried, and weighed. Sampling intervals that were
confounded by rainsplash as described in Whicker et al. (2006a) were eliminated from
the analysis. All four sets of new measurements were obtained during a period of intense
drought, which was severe enough to trigger extensive mortality of piñon trees at or near
the northern New Mexico, USA sites (Breshears et al., 2005a; Rich et al., in press).
Estimates of sediment transport ranged from 0.17 to 4.85 g m-2 d-1 (Table 1) across the
four study sites:
•
Santa Rita Experimental Range (SRER), Sonoran Desert grassland, Arizona,
USA. The first grassland site was a Sonoran desert grassland dominated by the
non-native annual grass Eragrostis lehmanniana (Lehman lovegrass) with a small
amount of shrub cover from Prosopis velutina (velvet mesquite). The field plots
were located within the University of Arizona pasture cell at SRER, where longterm vegetation change and grazing and management impacts have been studied
91
for more than a century (McClaran et al., 2003). Sediment transport at SRER was
the greatest of the four new estimates, even though ground cover was substantial
(60%): 4.85 g m-2 d-1 (Field et al., manuscript in preparation).
•
Area J, semiarid grassland on a landfill at Los Alamos National Laboratory New
Mexico, USA. The second grassland site had no woody plant cover; much of the
herbaceous cover was Bouteloua gracilis (blue grama), which had been
established in the topsoil with irrigation in previous years and then equilibrated
with ongoing climate. Vegetation on the landfill cover included native and nonnative herbaceous species. Early successional vegetation cover on landfills in this
area (Breshears et al., 2005b) is similar to herbaceous vegetation in neighboring
woodlands (Martens et al., 2001) and shares the same dominant herbaceous
species as shortgrass steppe ecosystems within the region (Lauenroth and Sala,
1992). Additional details on the site and measurements are available in Whicker
and Breshears (2004). Mean daily sediment transport at this site was less than a
third of that at the SRER grassland: 1.4 g m-2 d-1.
•
Mesita del Buey, piñon-juniper woodland at Los Alamos National Laboratory,
New Mexico, USA. The Mesita del Buey woodland site is an intensively studied
site dominated by piñon (Pinus edulis) and juniper (Juniperus monosperma; see
Breshears, 2006, 2007 and references therein). The site is located at Technical
Area 54 within Los Alamos National Laboratory Environmental Research Park.
Additional details on site and measurements are reported in Whicker and
92
Breshears (2004). This woodland site had the smallest sediment transport of the
four new estimates: 1.1 g m-2 d-1.
•
Frijolito, piñon-juniper woodland site within Bandelier National Monument, New
Mexico, USA. The Frijolito woodland site is also an intensively studied site that
was originally forest dominated by Pinus ponderosa (ponderosa pine) but was
transformed into piñon-juniper woodland (Pinus edulis and Juniperus
monosperma) when nearly all P. ponderosa trees died during a drought in the
1950s (Allen and Breshears, 1998). As noted above, this site was subsequently
impacted by a second severe drought beginning around 2000 (as was Area J and
Mesita del Buey) that resulted in extensive P. edulis mortality during the
measurement interval (Breshears et al., 2005a). The mean sediment transport at
Frijolito was 3.0 g m-2 d-1.
Relatively undisturbed sites
For relatively undisturbed dryland sites along the grassland-forest continuum,
mean sediment transport ranged from as little as 0.17 g m-2 d-1 to as much as 27.4 g m-2 d1
(Table 1). The sites ranged from 0 to 75% woody canopy cover and from 38 to 98%
ground cover. The maximum measured flux was from a shrubland site that had 28%
woody canopy cover and overall ground cover that exceeded 65% (Fig. 3). The rate of
transport in relatively undisturbed grasslands and woodlands were less than those for
shrubland sites and exceeded those for forests. These results suggest an overall trend in
sediment transport related to variation in woody plant cover for grasslands, shrublands,
woodlands, and forests and are similar to those hypothesized by Breshears et al. (2003)
93
based on a subset of this data set that included single estimates each from a grassland, a
shrubland, and a forest.
Disturbed sites
For disturbed sites along the grassland-forest continuum, sediment transport
ranged from 1.1 g m-2 d-1 to 6002 g m-2 d-1 with sites ranging from 0% to 72% canopy
cover (Table 2). Sediment transport at disturbed sites varied substantially among sites
with different amounts of woody plant cover and generally exceeded fluxes for
undisturbed sites with comparable amounts of woody plant cover. Notably, observations
from disturbed sites suggest that an upper bound exists to sediment transport and this
upper bound decreases with increasing canopy cover (Fig. 3b). Grasslands and other
systems with little or no woody canopy cover (e.g., bare areas or agricultural fields) have
the widest range of potential sediment transport, with the variation within this range
being associated with variation in the amount of intercanopy ground cover from
herbaceous plants (Okin et al. 2006) and biological and non-biological soil crusts
(Belnap, 2003). Shrublands have inherently larger sediment transport that can be
increased by disturbance. In contrast, woodlands and forests have smaller rates of
sediment transport, even following disturbance. Sediment transport at the thinned
ponderosa pine forest site (diamonds in Fig. 3) could be associated with either of two
values of cover, both of which are near the lower limit trend line for undisturbed systems.
Overall, the trends are consistent with previous observations that large sediment
transports are associated with a specific range of values for the spatial density of
roughness elements (woody plants in this case), as discussed below.
94
Surprisingly, no clear trends appeared related to soil texture within the limited
available data sets (Tables 1 and 2). Rates of wind erosion from wind tunnel studies
demonstrate relationships with soil texture (Pye, 1987; Toy et al. 2002), and soil texture
differences have been discussed relative to rates of wind and water erosion in grassland,
shrubland and forest ecosystems (Breshears et al., 2003). The synthesis here did not
detect any clear effects of soil texture.
Boundary Layer Flows and Land Surface Characteristics
Variation in land surface characteristics and associated effects on flow regimes
influences aeolian sediment transport. Of particular relevance are trends in flow regimes,
shear stress, and surface erodibility.
Flow regimes
The woody plant mosaics associated with sites along the grassland-forest
continuum fundamentally influence airflow regimes (Fig. 2). Woody plants interact with
wind flow in the atmospheric boundary layer altering speed, direction, and turbulence
structure. A single, isolated plant alters wind flow by imparting frictional force against
the wind that retards the flow and transfers energy and momentum to branches and
leaves. In addition, the fluid is accelerated around the sides and top of the plant, and
turbulent eddies are created in the wake of the plant (this influence provides potential
“connectivity” between woody plants and the areas around them, as defined above and in
Fig. 1; Breshears, 2006). Wind flow through stands of vegetation at larger ecosystem or
landscape scales reflects the combined effects of individual plants and generally has been
95
grouped into one of three flow regimes: 1) isolated roughness flow, 2) wake interference
flow, or 3) skimming flow (Wolfe and Nickling, 1993; Fig. 2). Isolated roughness flow
is characterized as having flows and wakes that act in isolation of one another and occurs
in ecosystems with small plant densities (< 16% canopy cover). Wake interference flow
is associated with the wakes that form when the wind and plant interactions begin to
interfere with one another, occurring at plant densities of approximately 16 to 40%
canopy cover. Skimming flow is associated with conditions in which winds skim across
the top portions of plants, with little wind energy diverted to the soil surface and,
therefore, resulting in little wind erosion. This type of flow generally occurs when plant
canopy cover exceeds 40%.
In addition to woody plant cover, the shape and spacing of plants are important
characteristics of canopy architecture that influence wind flow through vegetation.
Roughness density has been used extensively to relate canopy architecture and surface
roughness (Raupach et al., 1993). Roughness density, λ, is defined as:
λ=
nbh
,
A
(1)
where n is the number of plants, b is the base diameter, h is the plant height, bh is the
frontal area of the plant, and A is total area Assuming uniformly and isotropically
distributed vegetation, the fraction of the area with canopy cover (Fc), as viewed from
above, is exponentially related to λ (Fryrear, 1985; Findlater et al., 1990):
Fc = 1 − e − λ .
(2)
96
Another parameter that describes the interaction of wind and vegetation is the
concentration of roughness elements, or the dimensionless C value (Raupach et al., 1980;
Warner, 2004). The C value is slightly different from λ in that it normalizes the frontal
area of the plants to the square of the distance (d) between the plants rather than to the
size of the area (as shown in Eqn. 1):
C=
hb
.
d2
(3)
As with λ, the C value is related to canopy cover. Again, by assuming an even and
isotropic distribution of woody plants, the C value can be rewritten in terms of Fc:
h
C= 2
b
2
⎛ Fc ⎞
⎜⎜
⎟⎟ ,
⎝ 1 − Fc ⎠
(4)
λ and C are related to Fc, and also to shear stress, as shown below.
3.2. Shear stress partitioning
Vegetation cover and the soil surface extract momentum from near-surface wind,
resulting in shear stress (τ), which is defined as a tangential force applied per unit surface
area. Total shear stress can be partitioned between the vegetation and the surface soil as
(Raupauch et al., 1993):
τ = ρu* = τv + τs,
(5)
where ρ is the density of air, u* is the friction (or shear) velocity, τv is the shear stress
apportioned to vegetation, and τs is the shear stress apportioned to the surface, or in this
case the soil surface. The relationship between λ and C with respect to τ and τv is such
97
that as λ and C increase, the ratio of τv to τ goes towards one. That is, as vegetation
roughness increases, a greater portion of total shear stress is caused by the vegetation
relative to the soil surface.
The parameter τs in Equation 5 can be rewritten in terms of shear stress
normalized to stress on bare surface areas only (Raupach et al., 1993):
τ s = τ s′ (1 − Fc ) ,
(6)
where τs’ is the shear stress on bare ground only. Further, the total shear stress for
partially vegetated areas can be portioned into two surface components, woody (τwoody)
and herbaceous vegetation (τgrass):
τ v = τ woody + τ grass ,
(7)
or in terms of stress only on plant surfaces,
′
τ v = Fc (τ woody
+ τ ′grass ) .
(8)
3.3. Relationship between flow regime and erodibility
The interaction between airflow, vegetation and soil surface drives the aeolian
erodibility of a landscape, and can be described in terms of the metrics λ, C, and τ, all of
which vary along the grassland-forest continuum. In “ideal” grasslands with only
′
+ τ ′grass ) to
herbaceous cover, Equation 8 can be simplified from τ v = Fc (τ woody
τ v = Fcτ ′grass because woody plants are largely absent. In undisturbed grasslands, Fc can
be quite large although the shear stress from individual herbaceous plants is expected to
be small compared to that from woody vegetation, which is typically taller and wider. In
98
grasslands, λ and C values have small h and b values that are counterbalanced by the
relatively large Fc yielding larger τv values. The large amount of herbaceous cover
present in undisturbed grasslands with no woody canopy cover likely produces skimming
type airflow (Lee, 1991a,b).
Shrublands and woodlands are characterized by substantial components of woody
and herbaceous plants (House et al., 2003). The woody canopy cover, Fc, in shrublands
and woodlands can vary significantly, which in turn affects the partitioning of the shear
stress between the woody and herbaceous components. If woody plant cover is sparser,
as occurs in many shrublands, the flow regime will likely correspond to either isolated
wake flow or wake interference flow, depending on the canopy cover and the height and
width of the woody vegetation. Dryland sites with wake interference flow are expected
to have a greater amount of shear stress transferred to the intercanopy surface, potentially
causing greater aeolian sediment transport. Woodlands generally have greater amounts
of woody plant cover and taller canopy heights than shrublands, such that the shear stress
′ , resulting in skimming flow.
from vegetation generally approaches τ v = Fcτ woody
′ from additional increases
For forests, shear stress further approaches τ v = Fcτ woody
in height and canopy cover, such that only a few locations are not beneath woody plants.
Taller tree heights and larger base diameters (including the branches and leaves/needles)
result in relatively large values of roughness, λ, and spatial density, C . Relatively large
values of basal diameters for forest trees (spanning the woody canopy patch, not just the
trunk diameter) also result in a greater Fc and indicate that airflow above the tree canopy
should be skimming type flow. Consequently, thickly forested areas would be expected
99
to have lesser aeolian transport because of the tall and thick vegetation associated with
the greater amount of canopy cover.
Spatial density, C, varied with canopy cover for the undisturbed (Table 1) and
disturbed (Table 2) sites. The variation in C, focusing solely on the woody plants as
roughness elements (i.e., ignoring the roughness effects of herbaceous plants) for
relatively undisturbed sites along the grassland-forest continuum, highlights how C
increases with canopy cover (Fig. 4a). Previous studies of wind erosion have suggested
that values of C near 0.1 are associated with wake interference flow and large rates of
wind erosion (Warner, 2004). Results presented herein are consistent with this
observation (Fig. 4b). Note that C is near 0.1 for shrublands but generally not for the
other types of ecosystems along the grassland-forest continuum. In addition, note that C
is affected by woody canopy cover, spacing, and height and recall that the height of
woody plants often increases with woody plant cover along the grassland-forest
continuum (Fig. 1; e.g., Martens et al., 2000). Consequently, only shrublands, which
often have relatively small amounts of woody canopy cover (~20-30%) and are
composed of relatively low stature woody plants, intersect the values of C near 0.1 (Fig.
5), where aeolian sediment transport and associated wind erosion are thought to be
greatest.
100
A New Conceptual Framework
Model description
The available data suggest two trends in how aeolian sediment transport varies
along the grassland-forest continuum (Fig. 6). Firstly, for relatively undisturbed
ecosystems, shrublands have inherently greater sediment transport potential, likely
because of wake interference flow associated with the spatial concentration of roughness
elements. Secondly, for disturbed ecosystems, the upper bound for sediment transport
decreases with increasing amount of woody plant cover, probably because of effects of
canopy height and density on airflow patterns and because the minimum amount of
ground cover at a site tends to increase as woody plant cover increases. That is, ground
cover is interrelated with, and directly proportional to, woody plant cover. Because
woody plants represent a substantial concentration of biomass and generate large
amounts of litter, the ground cover directly beneath woody plants generally is completely
covered with litter. Consequently, woody canopy cover also often determines a
minimum amount of ground cover for an ecosystem. That is, as woody canopy cover
increases along the grassland-forest continuum, so does minimum amount of ground
cover. For example, a site with 80% tree cover, should have at least 80% ground cover,
and so intercanopy cover can range only from 0-20%. In addition, woodlands and forests
also are likely to have skimming flow because they are associated with > 40% cover.
Under unusual circumstances, more ground cover may be associated with the undisturbed
trend line, in which case aeolian sediment transport can be less than that associated with
that trend line. In addition, although soil particle sizes certainly affect aeolian sediment
101
transport (Pye, 1987; Toy et al., 2002), within the broad perspective of the framework
presented here, trends associated with woody plant cover are evident whereas expected
trends with soil texture (e.g., Breshears et al., 2003) are not. This suggests that woody
plant cover may be an equally important factor in controlling rates of aeolian transport at
the regional to global scales for which woody plant cover varies greatly among drylands.
Soil texture likely will explain additional variance in sediment transport potential within
the proposed framework. Additional studies are needed to further test the trends
hypothesized within the framework and for modifications because of soil texture.
Implications, applications and future directions
The proposed conceptual framework requires additional testing but offers promise
for addressing numerous key environmental issues related to aeolian processes. For
example, the framework may be applicable to large-scale efforts to manage ground cover
and provide shelter belts to reduce dust storms, as is currently being done in China (Shao
and Dong, 2006; Zhao et al., 2006). More specifically, the framework suggests that
increasing shrub cover could reduce wind erosion at bare sites, but that adding shrub
cover to sites with existing herbaceous cover could, in some cases, result in increased
aeolian sediment transport and associated wind erosion. The framework also suggests
that establishing larger trees with sufficient density, if feasible, could simultaneously
decrease aeolian sediment transport, increase rates of dust deposition and improve soil
quality (Shirato et al., 2004; McGowan and Ledgard, 2005).
The proposed framework also has important implications for atmospheric
modeling. One of the central uncertainties in atmospheric models is dust flux, yet few
102
field data exist on atmospheric dust flux. The reported synthesis is insufficient to address
this issue in a robust way but can serve as a starting point for linking sediment transport
near the ground to vertical dust flux inputs for atmospheric models. The relationship
between sediment transport (horizontal flux) and vertical dust flux varies with site
conditions (e.g., vegetation cover) and has particle size dependencies (e.g., PM10 as in
Gillette et al., 1997 and Whicker et al., 2006b). At the risk of oversimplifying these
relationships, estimates from Whicker et al. (2006a) suggest that vertical dust flux may be
~5% of sediment transport. Building on this assumption, mean daily sediment transport
for general land surface categories are proposed that may be relevant to large-scale
atmospheric models: undisturbed forests, disturbed forests, undisturbed woodlands,
disturbed woodlands, relatively undisturbed shrublands, disturbed shrublands,
undisturbed grasslands, moderately disturbed grasslands, essentially bare areas (sites with
no herbaceous or woody plant cover), and areas that have unusually large amounts of
ground cover within semiarid settings (Fig. 7).
The proposed conceptual framework (Fig. 6) and its extension (Fig. 7) also has
implications that span from applied issues, related to contaminant transport and
associated risks, to more fundamental issues related to biogeochemical, hydrological and
ecological processes that underlie challenges related to global climate change and land
use. The need to assess potential contaminant transport has motivated many of the recent
studies of aeolian sediment transport (Whicker et al., 2002, 2006a,b, 2007a,b; Breshears
et al., 2003). Findings relevant to contaminant transport that affect respirable fractions of
airborne contaminants and associated doses, in addition to the estimated sediment
103
transport themselves, include documenting a direct relationship between sediment
transport and PM10 concentrations (Whicker et al., 2006a) and quantifying site-specific
and particle-size-dependent variation in partition coefficients (Kds) for U in surface soils
(Whicker et al., 2007b). The framework also has biogeochemical implications relevant to
how resources are redistributed within and among ecosystems (Reynolds et al., 2001,
2006a,b,c; Neff et al., 2005). Indeed, nutrient losses and redistribution by aeolian
processes are viewed as key components of desertification and degradation processes
(Schlesinger et al., 1990; Aguiar and Sala, 1999; Peters et al., 2006). In addition, the
potential relationships between sediment transport and vertical dust flux, described above
in the context of atmospheric processes, has fundamental hydrological implications. For
example, vertical dust flux and subsequent transport at regional scales can lead to
inhibition of precipitation (Rosenfeld et al., 2001) and to hastening of snowmelt (Painter
et al., 2007).
Conclusions
A more comprehensive understanding of how aeolian sediment transport and
associated wind erosion and dust flux vary along the grassland-forest continuum is
needed to better assess trends across a broad spectrum of ecosystem types and in
associated responses to disturbance. Estimates of sediment transport along the grasslandforest continuum, which have been largely lacking until recently, seem to be bounded
within an envelope related to woody plant cover and site disturbance. New data in
conjunction with previously reported estimates for sites along the grassland-forest
continuum that have not been recently and/or drastically disturbed suggest that aeolian
104
sediment transport may be inherently greater for shrublands. This result is consistent
with recent evaluations of roughness elements focusing on roughness density and height
to width ratios of roughness elements, which are associated with woody plants in this
case. Notably, because height tends to increase with woody plant cover along the
grassland forest continuum, it is primarily low density shrublands that have maximal
effects on the wind profile and on aeolian sediment transport; greater densities or taller
woody plants tend to be associated with skimming flow over the woody plant canopy. In
addition, the synthesis presented here highlights that for disturbed ecosystems, the upper
bound for sediment transport potential decreases with increasing amount of woody plant
cover. Canopy height and density affects the airflow patterns and woody plant cover
influences the minimum amount of ground cover. Consequently, grasslands or associated
systems with no woody plants can have the largest potential for sediment transport and
can span the largest range of values. Shrublands have inherently large sediment transport
potential that can be increased by disturbance, and woodlands and forests have the
smallest sediment transport potential, even following disturbance. These trends are
consistent with previous observations that wind erosion is greatest for systems with a
specific range of values of spatial density of roughness elements, woody plants in this
case. The proposed framework has implications for predictions of land surface inputs of
dust for atmospheric models, land management planning for reducing dust storms and
associated wind erosion, and biogeochemical and contaminant transport within and
across ecosystems along the grassland-forest continuum that span much of the terrestrial
biosphere.
105
Acknowledgements
We thank current and previous sponsors for wind erosion research, including Los Alamos
National Laboratory, Department of Energy Environmental Management Science
Program, USDA CSREES training grant in Ecohydrology (2005-02335; to DDB and
supporting JPF), U.S. Geological Survey, U.S. D.I. National Park Service, Arizona
Agricultural Experiment Station. We thank Rebecca J. Oertel and Kay L. Beeley of NPS
at Bandelier National Monument for collection of data from Frijolito, Paul Dimmerling
of LANL for data analysis, and Jayne Belnap, Dale Gillette, Greg Okin, Ted Zobeck, and
Xuben Zhang for related discussions.
References
Aguiar, M.R., Sala, O.E., 1999. Patch structure, dynamics and implications for the
functioning of arid ecosystems. Trends in Ecology & Evolution 14, 273-277.
Allen, C.D., Breshears, D.D., 1998. Drought-induced shift of a forest-woodland ecotone:
Rapid landscape response to climate variation. Proceedings of the National Academy
of Sciences of the United States of America 95, 14839-14842.
Anderson, R.S., Fralish, J.S., Baskin, J.M., 1999. Savannas, barrens, and rock outcrop
plant communities of North America. Cambridge University Press, Cambridge, UK.
Baker Jr., M.B., Jemison, R.L., 1991. Soil loss—key to understanding site productivity.
In: Proceedings of the 36th annual meeting of the Agencies and Science Working
Together for the Future, New Mexico Water Research Institute, pp 71-76.
Baker, M.B., DeBano, L.F., Ffolliott, P.F., 1995. Soil loss in piñon-juniper ecosystems
and its influence on site productivity and desired future condition. In: Desired Future
106
Conditions for Piñon-Juniper Ecosystems, D. W. Shaw, E. F. Aldon, and C. LoSapio
(eds.). Proceedings of Symposium, 8-12 August 1994, Flagstaff, AZ. General
Technical Report, RM-258. USDA Forest service, Fort Collins, CO; 9-15.
Belnap, J., 2003. The world at your feet: desert biological soil crusts. Frontier in Ecology
and the Environment 4, 181-189.
Bielders, C.L., Rajot, J.L., Amadou, M., 2002. Transport of soil and nutrients by wind in
bush fallow land and traditionally managed cultivated fields in the Sahel. Geoderma
109, 19-39.
Breshears, D.D., 2006. The grassland-forest continuum: trends in ecosystem properties
for woody plant mosaics? Frontiers in Ecology and the Environment 4, 96-104.
Breshears, D.D., 2007. The structure and function of woodland mosaics: consequences of
patch-scale heterogeneity and connectivity along the grassland-forest continuum. In:
Western North American Juniperus Woodlands. O. W. Van Auken, ed. Ecological
Studies. Springer, New York, pp. 58-92.
Breshears, D.D., Barnes, F.J., 1999. Interrelationships between plant functional types and
soil moisture heterogeneity for semiarid landscapes within the grassland/forest
continuum: a unified conceptual model. Landscape Ecology 14, 465-478.
Breshears, D.D., Whicker, J.J., Johansen, M.P., Pinder, J.E., 2003. Wind and water
erosion and transport in semiarid shrubland, grassland, and forest ecosystems:
Quantifying dominance of horizontal wind-driven transport. Special Issue: Wind
Erosion Field Studies: Contributions to a Joint Meeting of ICAR5 and GTCE-SEN.
Earth Surface Processes and Landforms 28, 1189-1209.
107
Breshears, D.D., Cobb, N.S., Rich, P.M., Price, K.P., Allen, C.D., Balice, R.G., Romme,
W.H., Kastens, J.H., Floyd, M.L., Belnap, J., Anderson, J.J., Myers, O.B., Myers,
C.W., 2005a. Regional vegetation die-off in response to global-change-type drought.
Proceedings of the National Academy of Sciences of the United States of America
102, 15144-15148.
Breshears, D.D., Nyhan, J.W., Davenport, D.W., 2005b. Ecohydrology monitoring and
excavation of semiarid landfill covers a decade after installation. Vadose Zone
Journal 4, 798-810.
Crosson, P., 1995. Soil erosion estimates and costs. Science 269, 461-464.
Findlater, P.A., Carter, D.J., Scott, W.D., 1990. A model to predict the effects of prostrate
ground cover on wind erosion. Australian Journal of Soil Research 28, 609-622.
Fryrear, D.W., 1985. Soil cover and wind erosion. Transactions of the ASAE 28, 781784.
Fryrear, D.W., 1986. A field dust sampler. Journal of Soil and Water Conservation 41,
117-120.
Gillette, D.A., Pitchford, A.M., 2004. Sand flux in the northern Chihuahuan desert, New
Mexico, USA, and the influence of mesquite-dominated landscapes. Journal of
Geophysical Research-Earth Surface 109, F04003.
Gillette, D., 1997. Soil derived dust as a source of silica: Aerosol properties, emissions,
deposition, and transport. Journal of Exposure Analysis and Environmental
Epidemiology 7, 303-311.
108
Griffin, D.W., Kellogg, C.A., Shinn, E.A., 2001. Dust in the wind: Long range transport
of dust in the atmosphere and its implications for global public and ecosystem health.
Global Change and Human Health 2, 20-33.
House, J.I., Archer, S., Breshears, D.D., Scholes, R.J., 2003. Conundrums in mixed
woody-herbaceous plant systems. Journal of Biogeography 30, 1763-1777.
Lal, R. (Ed.), 1996. Soil Quality and Soil Erosion. CRC Press: Boca Raton, FL, USA.
Lal, R., Sobecki, T.M., Iivari, T., Kimble, J.M., 2003. Soil degradation in the United
States. CRC Press: Boca Raton, FL, USA.
Lauenroth, W.K., O.E. Sala., 1992. Long-term forage production of North American
shortgrass steppe. Ecological Applications 2, 397-403.
Lee, B.E., Soliman, B.F., 1977. Investigation of forces on 3 dimensional bluff bodies in
rough wall turbulent boundary-layers. Journal of Fluids Engineering-Transactions of
the ASME 99, 503-510.
Lee, J.A., 1991a. Near-surface wind flow around desert shrubs. Physical Geography 12,
140-146.
Lee, J.A., 1991b. The role of desert shrub size and spacing on wind-profile parameters.
Physical Geography 12, 72-89.
Loik, M.E., Breshears, D.D., Lauenroth, W.K., Belnap, J., 2004. A multi-scale
perspective of water pulses in dryland ecosystems: climatology and ecohydrology of
the western USA. Oecologia 14, 269-281.
109
Martens, S.N., Breshears, D.D., Meyer, C.W., 2000. Spatial distributions of understory
light along the grassland/forest continuum: effects of cover, height, and spatial pattern
of tree canopies. Ecological Modeling 126, 79-93.
Martens, S.N., Breshears, D.D., Barnes, F.J., 2001. Development of species dominance
along an elevational gradient: population dynamics of Pinus edulis and Juniperus
monosperma. International Journal of Plant Sciences 162, 777-783.
Masri, Z., Zobisch, M., Bruggeman, A., Hayek, P., Kardous, M., 2003. Wind erosion in a
marginal Mediterranean dryland area: A case study from the Khanasser Valley, Syria.
Earth Surface Processes and Landforms 28, 1211-1222.
McClaran, M.P. 2003. A century of vegetation change on the Santa Rita Experimental
Range. Pages 16–33. In: McClaran, M. P., Ffolliott, P. F., Edminster, C. B. (Eds.).
Santa Rita Experimental Range: 100 years (1903-2003) of accomplishments and
contributions. RMRS-P-30. U.S. Department of Agriculture, Forest Service, Rocky
Mountain Research Station, Ogden UT.
McGowan, H., Ledgard, N., 2005. Enhanced dust deposition by trees recently established
on degraded rangeland. Journal of the Royal Society of New Zealand 35: 269-277.
McPherson, G.R., 1997. Ecology and management of North American Savannas.
University of Arizona Press, Tucson, AZ, USA.
Miller, R.L., Tegen, I., 1998. Climate response to soil dust aerosols. Journal of Climate
11, 3247-3267.
Nearing, M.A., Romkens, M.J.M., Norton, L.D., Stott, D.E.,, Rhoton, F.E., Laflen, J.M.,
Flanagan, D.C., Alonso, C.V., Binger, R.L., Dabney, S.M., Doering, O.C., Huang,
110
C.H., McGregor, K.C., Simon, A., 2000. Measurements and models of soil loss rates.
Science 290, 1300-1301.
Neff, J.C., Reynolds, R.L., Belnap, J., Lamothe, P., 2005. Multi-decadal impacts of
grazing on soil physical and biogeochemical properties in southeast Utah. Ecological
Applications 15, 87-95.
Okin, G.S., Gillette, D.A., 2001. Distribution of vegetation in wind-dominated
landscapes: Implications for wind erosion modeling and landscape processes. Journal
of Geophysical Research-Atmospheres 106, 9673-9683.
Okin, G.S., Gillette, D.A., Herrick, J.E., 2006. Multi-scale controls on and consequences
of aeolian processes in landscape change in arid and semi-arid environments. Journal
of Arid Environments 65, 253-275.
Painter, T.H., Barrett, A.P., Landry, C.C., Neff, J.C., Cassidy, M.P., Lawrence, C.R.,
McBride, K.E., Farmer, G.L., 2007. Impact of disturbed desert soils on duration of
mountain snow cover. Geophysical Research Letters 34, L12502,
doi:10.1029/2007GL030284.
Peters, D.P.C., Bestelmeyer, B.T., Herrick, J.E., Fredrickson, E.L., Monger, H.C.,
Havstad, K.M., 2006. Disentangling complex landscapes: New insights into arid and
semiarid system dynamics. Bioscience 56, 491-501.
Pimentel, D., Harvey, C., Resosudarmo, P., Sinclair, K., Kurz, D., Mcnair, M., Crist, S.,
Shpritz, L., Fitton, L., Saffouri, R., Blair, R., 1995a. Environmental and economic
costs of soil erosion and conservation benefits. Science 267, 1117-1123.
111
Pimentel, D., Harvey, C., Resosudarmo, P., Sinclair, K., Kurz, D., McNair, M., Crist, S.,
Shpritz, L., Fitton, L., Saffouri, R., Blair, R., 1995b. Soil-erosion estimates and costs Reply. Science 269, 464-465.
Pye, K., 1987. Aolian Dust and Dust Deposits. London: Academic Press.
Raupach, M.R., Thom, A.S., Edwards, I., 1980. A wind tunnel study of turbulent flow
close to regularly arrayed rough surfaces. Boundary Layer Meteorology 18: 373-397.
Raupach, M.R., Gillette, D.A., Leys, J.F., 1993. The effect of roughness elements on
wind erosion threshold. Journal of Geophysical Research-Atmospheres 98, 30233029.
Reynolds, R.L., Belnap, J., Reheis, M., Lamothe, P., Luiszer, F., 2001. Aeolian dust in
Colorado Plateau soils: Nutrient inputs and recent change in source. Proceedings of
the National Academy of Science USA 98, 7123-7127.
Reynolds, R., Neff, J., Reheis, M., Lamothe, P., 2006a. Atmospheric dust in modem soil
on aeolian sandstone, Colorado Plateau (USA): Variation with landscape position and
contribution to potential plant nutrients. Geoderma 130, 108-123.
Reynolds, R.L., Reheis, M., Yount, J., Lamothe, P., 2006b. Composition of aeolian dust
in natural traps on isolated surfaces of the central Mojave desert - insights to mixing,
sources, and nutrient inputs. Journal of Arid Environments 66, 42-61.
Reynolds, R.L., Reheis, M.C., Neff, J.C., Goldstein, H., Yount, J., 2006c. Late quaternary
eolian dust in surficial deposits of a Colorado Plateau grassland: Controls on
distribution and ecologic effects. Catena 66, 251-266.
112
Rich P.M., Breshears, D.D., White A.B., 2007. Phenology of mixed woody-herbaceous
ecosystems following extreme events: Net changes from differential responses.
Ecology, in press.
Rosenfeld, D., Rudich, Y., Lahav, R., 2001. Desert dust suppressing precipitation: a
possible desertification feedback loop. Proceedings of the National Academy of
Sciences USA 98: 5975-5980.
Schlesinger, W.H., Reynolds, J.F., Cunningham, G.L., Huenneke, L.F., Jarrell,
W.M.,Virginia, R.A., Whitford, W.G., 1990. Biological feedbacks in global
desertification. Science 247, 1043-1048.
Shao, Y., Dong, C.H., 2006. A review on east Asian dust storm climate, modeling and
monitoring. Global and Planetary Change 52, 1-22.
Shirato Y., Taniyama, I., Zhang, T.H., 2004. Changes in soil properties after afforestation
in Horqin sandy land, North China. Soil Science and Plant Nutrition 50, 537-543.
Toy, T.J., Foster, G.R., Renard, K.G., 2002. Soil erosion: processes, prediction,
measurement and control. John Wiley & Sons, New York.
Trimble, S.W., Crosson, P., 2000a. Land use - US soil erosion rates - Myth and reality.
Science 289, 248-250.
Trimble, S.W., Crosson, P., 2000b. Measurements and models of soil loss rates Response. Science 290, 1301-1301.
Urban, M.A., Daniels, M., 2006. Introduction: Exploring the links between
geomorphology and ecology. Geomorphology 77, 203-206.
113
van Donk, S.J., Huang, X.W., Skidmore, E.L., Anderson, A.B., Gebhart, D.L., Prehoda,
V.E., Kellogg, E.M., 2003. Wind erosion from military training lands in the Mojave
Desert, California, USA. Journal of Arid Environments 54, 687-703.
Vermeire, L.T., Wester, D.B., Mitchell, R.B., Fuhlendorf, S.D., 2005. Fire and grazing
effects on wind erosion, soil water content, and soil temperature. Journal of
Environmental Quality 34, 1559-1565.
Warner, T.T., 2004. Desert Meteorology. Cambridge, New York.
Whicker, J.J., Breshears, D.D., 2004. Assessing wind erosion as a contaminant transport
mechanism at Area G. Los Alamos National Laboratory report LA-UR-04-8553.
Whicker, J.J., Breshears, D.D., Wasiolek, P.T., Kirchner, T.B., Tavani, R.A., Schoep,
D.A., Rodgers, J.C., 2002. Temporal and spatial variation of episodic wind erosion in
unburned and burned semiarid shrubland. Journal of Environmental Quality 31, 599612.
Whicker, J.J., Pinder, J.E., Breshears, D.D., 2006a. Increased wind erosion from forest
wildfire: Implications for contaminant-related risks. Journal of Environmental Quality
35, 468-478.
Whicker, J.J., Pinder III, J.E., Breshears, D.D., Eberhardt, C.F., 2006b. From dust to
dose: effects of forest disturbance on increased inhalation exposure. Science of the
Total Environment 368, 519-530.
Whicker, J.J., Pinder III, J.E., Breshears, D.D., 2007a. Thinning semiarid forests
amplifies wind erosion comparable to wildfire: Implications for restoration and soil
stability. Journal of Arid Environments, in press.
114
Whicker, J.J., Pinder III, J.E., Ibrahim, S.A., Stone, J.M., Breshears, D.D., Baker, K.N.,
2007b. Uranium partition coefficients (Kd) in forest surface soil reveal long
equilibrium times and vary by site and soil size fraction. Health Physics 93:36-46.
Whitford, W.G., 2002. Ecology of Desert Systems. Academic Press: San Diego, CA,
USA.
Wolfe, S.A., Nickling, W.G., 1993. The protective role of sparse vegetation in wind
erosion. Progress in Physical Geography 17, 50-68.
Zhao, C.X., Zheng, D.W., Stigter, C.J., He, W.Q., Tuo, D.B., Zhao, P.Y., 2006. An index
guiding temporal planting policies for wind erosion reduction. Arid Land Research
and Management 20, 233-244.
Zobeck, T.M., Sterk, G., Funk, R., Rajot, J.L., Stout, J.E., Van Pelt, R.S., 2003.
Measurement and data analysis methods for field-scale wind erosion studies and
model validation. Earth Surf Process Landforms 28, 1163-1188.
115
Figures
Figure 1. (a) The grassland-forest continuum as a gradient of increasing cover by woody
plants (and co-occurring increases in woody plant height), spanning shrublands and
woodlands with intermediate levels of woody plant cover. The mean and variance of
ecosystem properties related to energy, water, and biogeochemistry (e.g. carbon) are
hypothesized to exhibit trends along the continuum. Relevant environmental issues
include encroachment of woody plants, thicketization in which savannas become
woodlands, xerification or desertification, deforestation, restoration and thinning, die-off
of woody plants, and fire. Increases in elevation and/or changes to a wetter climate often
result in an increase woody plant canopy cover; conversely, decreases in elevation and/or
a drier climate often result in a decrease in woody canopy cover (except in the case of
xerification). (b) Variance in many ecosystem properties is hypothesized to be greatest at
an intermediate amount of canopy cover; this value is often less than 50% woody canopy
cover because woody plants have strong connectivity with the adjacent intercanopy
patches, a concept that is applicable to near-surface wind flows (modified from
Breshears, 2006).
116
Figure 2. Three major types of flow regimes related to woody plant density (top): (a)
isolated roughness flow, for <16% canopy cover; (b) wake interference flow, for 16-40%
canopy cover; and (c) skimming flow, for >40% canopy cover, as seen in profile (center)
and from above (bottom; modified from Wolfe and Nickling, 1993).
117
Figure 3. Aeolian sediment transport, based on multi-month estimates of sediment
transport (g m-2 d-1), as related to woody canopy cover along the grassland-forest
continuum for relatively undisturbed sites (open symbols, from Table 1) and (b)
including disturbed sites (solid symbols, from Table 2). A forest site that was thinned
(diamonds) is shown at pre- (solid) and post- (open with x inset) thinning values for
canopy cover.
118
Figure 4. (a) Estimates of spatial density C for relatively undisturbed sites (Table 1) as a
function of canopy cover; C as calculated here is based on woody plants as roughness
elements and ignores the finer-scale roughness associated with herbaceous plants. (b)
Values of C near 0.1 have been hypothesized to be associated with larger rates of
sediment transport.
119
Figure 5. Relatively undisturbed sites along the grassland-forest continuum as a function
of canopy cover and height (dashed lines correspond to an approximate envelope around
sites spanning the grassland-forest continuum) and values of woody canopy cover and
height resulting in a spatial density C near 0.1 (solid line), associated with large rates of
erosion , as a function of woody canopy cover and height.
120
Figure 6. Conceptual framework for aeolian sediment transport as sediment transport
along the grassland-forest continuum for undisturbed and disturbed sites. Also depicted
are minimum amount of ground cover (assumed here to equal woody plant cover) and
main types of flow. Relatively undisturbed sites occur near the undisturbed trend line.
Disturbances can increase sediment transport toward an upper bound. Among relatively
undisturbed ecosystems, shrublands inherently have the largest sediment transport. The
range of values between the undisturbed trend line and the disturbed upper bound is
greatest for grasslands. Variation in sediment transport between the undisturbed trend line
and the disturbed upper bound is associated with changes in amount of ground cover in
intercanopy areas, particularly from herbaceous plants but also including biological and
non-biological soil crusts. Variation in soil texture may also influence variation within
this interval. Unusually large amounts of intercanopy ground cover can yield sediment
transport that are less than corresponding values on the undisturbed trend line.
121
Figure 7. Categories of vertical dust flux along the grassland-forest continuum, bounded
between relatively undisturbed and disturbed sites. Estimates are based on assuming
sediment transport = 5% of vertical dust flux (Whicker et al., 2006b).
122
Tables
Table 1. Sediment transport measured in undisturbed ecosystems using BSNE and
Bagnold samplers.
123
Table 2. Sediment transport in undisturbed ecosystem measured using BSNE and
Bagnold samplers.
124
APPENDIX D
BACKGROUND DUST EMISSION FOLLOWING GRASSLAND FIRE: A
SNAPSHOT ACROSS THE PARTICLE-SIZE SPECTRUM HIGHLIGHTS HOW
HIGH-RESOULUTION MEASUREMENTS ENHANCE DETECTION
Luis Merino-Martin1, Jason P. Field2, Juan Camilo Villegas2,3, Jeffrey J. Whicker4,
David D. Breshears2,5, Darin J. Law2, and Anna M. Urgeghe6
1
Departamento de Ecología, Edificio de Ciencias, Universidad de Alcalá, Campus
externo de Alcalá, Ctra. Madrid-Barcelona km 33.600, Alcalá de Henares, E-28871
Madrid, Spain. 2 School of Natural Resources and the Environment, University of
Arizona. Tucson, Arizona, USA. 3Grupo GIGA. Facultad de Ingeniería. Universidad de
Antioquia. Medellín, Colombia. 4Environmental Programs, Los Alamos National
Laboratory, Los Alamos, New Mexico, USA. 5Department of Ecology and Evolutionary
Biology, The University of Arizona. Tucson, Arizona, USA. 6Departamento de Ecología,
Universidad de Alicante, Alicante, Spain.
Document Type: Currently in review as of September 29th, 2009.
Abstract: Dust emission rates vary temporally and with particle size. Many studies of
dust emission focus on a particular temporal scale and the portion of the particle-size
spectrum associated with a single instrument; fewer studies have assessed dust emission
across the particle-size spectrum and associated temporal scales using multiple
instruments. Particularly lacking are measurements following disturbances such as fire
that are high-resolution and focused on finer particles—those with direct implications for
human health and potential for long-distance biogeochemical transport—during less
windy but more commonly occurring background conditions. We measured dust
emissions in unburned and burned semiarid grassland using four different instruments
spanning different combinations of temporal resolution and particle-size spectrum: Big
Springs Number Eight (BSNE) and Sensit instruments for larger saltating particles,
DustTrak instruments for smaller suspended particles, and Total Suspended Particulate
125
(TSP) samplers for measuring the entire range of particle sizes. Unburned and burned
sites differed in vegetation cover and aerodynamic roughness, yet surprisingly differences
in dust emission rates were only detectable for saltation using BSNE and for smaller
aerosols using DustTrak. Our results, surprising in the lack of consistently detected
differences, indicate that high-resolution DustTrak measurements offered the greatest
promise for detecting differences in background emission rates and that BSNE samplers,
which integrate across height, were effective for longer intervals. More generally, our
results suggest that interplay between particle size, temporal resolution, and integration
across time and height can be complex and may need to be considered more explicitly for
effective sampling for background dust emissions.
Introduction
Wind erosion and associated dust emissions are controlled to a large extent by the
type and distribution of vegetation cover, as well as other important site characteristics
such as climatic conditions, soil physical and chemical properties, and topography (Pye,
1987; Visser et al., 2004; Li et al., 2005). Dust emissions can have important ecological
and environmental consequences at scales ranging from individual plants up to the globe
(Goudie and Middleton, 2006; Okin et al., 2006; Field et al., 2010). Wind erosion in
dryland ecosystems can vary substantially spatially and temporally, particularly following
large-scale disturbance (e.g., fire, grazing) that removes much or all of the protective
vegetation cover (Ravi and D’Odorico, 2005; McConnell et al., 2007; Neff et al., 2008;
Field et al., 2009). Event-based dust emissions are strongly dependent not only on the
protective vegetation cover but also on the wind speed and the threshold friction velocity,
126
as well as size of the wind-blown particles being transported (Gillette et al., 1980; Stout
and Zobeck, 1996). Notably, wind erosion and dust emissions are related to the erosive
force of the wind and the overall erodibility of the soil, which is controlled largely by
particle-size distribution, aggregate stability, and particle cohesion due to moisture (Lee
and Tchakerian 1995; Vermeire et al., 2005). Erosive force varies greatly with wind
speed and the structure of the vegetation and the soil surface. Wind-blown sediment of
varying sizes is transported by one of three processes (these categories overlap): surface
creep for soil particle diameters > 500 μm, saltation for diameters ranging from 20 – 500
μm, and suspension for diameters < 20 μm (Bagnold, 1941; Toy et al., 2002). Aeolian
transport at local scales is dominated primarily by surface creep and saltating particles,
whereas at regional to global scales, aeolian transport is dominated primarily by
suspended particles (Stout and Zobeck, 1996; Goudie and Middleton, 2006). Aeolian
sediment transport can occur in both the horizontal and vertical direction as either
horizontal sediment flux, which contributes to localized redistribution, or as vertical dust
flux, which is usually considered available for long-distance transport (Breshears et al.,
2003; Zobeck et al., 2003).
Although the largest dust emission rates typically occur during short, windy
periods, the most frequent conditions affecting potential dust emissions are background
conditions with lower wind speeds. Such background rates of dust emission might be
particularly important following disturbances such as fire that remove ground cover
because even small increases in dust emission can be important for human health and/or
long distance biogeochemical transport. Most studies of dust emission focus on windy
127
conditions, and many of these studies further focus on a particular portion of the particlesize spectrum related to a single instrument and the associated temporal scale for
measurement. Fewer studies have applied multiple instruments to assess background
dust emission across the entire particle-size spectrum and associated different temporal
scales following disturbance.
Dust emissions can increase following different types of disturbance such as
grazing and fire. Fire can be particularly important because it rapidly reduces cover by
consuming it, often resulting in a large increase in bare soil cover. Many soil and
vegetation characteristics, including texture, surface crusting, organic matter content, and
percent canopy cover, can affect the rates of wind erosion following fire (Visser et al.,
2004). The magnitude of wind erosion following fire depends strongly on both climatic
factors, such as wind speed and atmospheric humidity (Ravi and D’Odorico, 2005) and
certain physical characteristics of the soil and associated vegetation. An improve
understanding of the temporal and spatial variability associated with wind erosion prior to
and following a fire is required for accurate long-term assessment of public risk for
nearby residents (Whicker et al., 2002; Whicker et al., 2006).
In summary, many studies of dust emission focus on a particular temporal scale
and on the portion of the particle-size spectrum associated with a single instrument.
However, fewer studies have assessed dust emission using multiple instruments across
the particle-size spectrum and associated different temporal scales. Notably lacking are
high-resolution field measurements of background concentrations focused on finer
particles following disturbances such as fire that increase dust emissions, and which
128
could have direct implications for human health and potential for long distance
biogeochemical transport. To address this issue, we measured dust emission rates in
unburned and burned semiarid grassland using four different instruments that span
different combinations of temporal resolution and particle-size spectrum: Big Springs
Number Eight (BSNE) and Sensit instruments for larger saltating particles, DustTrak
instruments for finer aerosols, and Total Suspended Paticulate (TSP) samplers for all
particle sizes. These samplers collectively span the entire particle spectrum associated
with aeolian transport (Fig. 1). We obtained measurements from semiarid grassland sites
in southern Arizona USA and contrasted an unburned site with a recently (2 months
prior) burned site.
Material and Methods
The study was conducted in semiarid grassland at the Santa Rita Experimental
Range (31°50’’N, 110°50’’W), located 48 km south of Tucson, Arizona (McClaran,
2003). Mean annual precipitation is approximately 350 mm. Soils are predominantly
sandy loam with shallow argillic horizons and moderate infiltration rates (~ 40 mm hr-1).
Grass cover varies from approximately 40-60%, with Lehmann lovegrass (Eragrostis
lehmanniana) constituting the majority (>90%) of the grass cover. Sparse woody shrubs,
primarily Velvet mesquite (Prosopis velutina), constitute less than (5%) of the total
vegetation cover. We established three 50 m by 50 m plots both in an unburned site and
in a site that had recently burned (~2 months prior). The plots were selected to control
for minimal slope (<5%), minimal shrub cover (<10%).
129
Soil samples for estimating texture were obtained from each plot in three different
locations to a depth of 0.05 m and were aggregated and mixed into a single plot sample.
Soil texture was determined using the Bouyoucos' hydrometer method (Bouyoucos
1962). Vegetation cover was measured by visual estimation along 50 m transects by two
methods: Line-point intercept with height and gap intercept (Herrick et al. 2005). A
meteorological station within a few hundred meters of each site provided data on timing
and amount of rainfall at the site (Campbell Scientific, Logan, UT, USA).
We obtained measurements from each of four instruments on each plot. For
measuring saltation, we installed a Sensit (Sensit Company, Portland, ND, USA) wind
erosion mass flux sensor at 0.08 m above the soil surface. The Sensit was programmed
for 1 second measurements. In addition, we also installed a profile of Big Spring
Number Eight (BSNE) suspension samplers to measure horizontal flux mainly due to
saltation. Each profile included samplers at five heights (0.08, 0.17, 0.25, 0.50, and 1.0 m
above the soil) to asses horizontal dust flux using the gradient method and flux equation
derived by Gillette (1977). Sediment was collected from the dust samplers every 7-10
days and oven-dried at 60°C to achieve constant weight. For measuring aerosols, we
measured Total Suspended Particulate (TSP) with a filter-based air sampler (F&J Model
DF-AB-75L, Ocala, FL, USA). TSP concentration measurements were based on mass
collected on five sampling days during the study and had a sampling rate of about 4.5
m3h-1. The TSP measurements were limited in duration because the sampler battery only
allowed about 5 hours of operation. Thus, TSP sampling was made during the day
generally between 12:00 to 17:00, yielding one concentration measurement per 5-hour
130
period. Simultaneously, we also measured finer aerosols using light-scattering laser
photometers (DustTrak model 8533, TSI Inc., Shoreview, MN, USA), which measured
particle concentrations < 10.0 µm and were collected at a height of 25 cm above the soil
surface. The DustTrak measurements were obtained at 1 second intervals and aggregated
to 1 minute intervals.
Supplemental measurements of wind speed were obtained for a single location at
each of the two sites. Three anemometers were installed at 0.25, 0.50, and 1.0 m above
ground to calculate the wind profile, friction velocity, and aerodynamic roughness. In
addition, we measured wind speed for each plot at 25 cm using a data logging weather
sensor (Kestrel 4500, Boothwyn, PA), which were programmed to calculate 10-minute
wind speed averages. The temporal resolution of the Kestrel sensors was less than the
wind anemometers, which could be particularly important during periods dominated by
low wind velocities; however, the Kestrels offered the advantage over the more costly
wind anemometers in that they could be placed within each plot for detailed onsite
measurements of wind speed. These sensors also measured temperature and relative
humidity.
Our analyses included calculating cumulative horizontal dust flux by integrating
from 0 to 1 m using a simple exponential relationship with height (Gillete et al., 1997;
Stout, 2001). To assess friction velocities, we plotted the friction velocities obtained for
each site versus the wind velocity at 1 meter height (at 1 second frequency during 17
sampling days). To test for differences between the unburned and burned site, we used
the Mann-Whitney U test using STATISTICA (Statsoft 2001).
131
Results
Environmental variables and vegetation cover in unburned and burned sites
The sites did not differ significantly in relative humidity between unburned
(44.04% +0.46 s.e.) and burned (44.09% +1.96 s.e.). Soil texture was also consistent
between sites and was a sandy loam. Temperatures were slightly cooler in the unburned
site (26.45 ± 0.08ºC) relative to the burned site (27.15 ± 0.23ºC). The most notable
difference between the sites was the large difference in vegetation cover—the burned site
had much more cover that was bare soil (Fig. 2a).
Aerodynamic roughness lengths and friction velocities between burned and
unburned sites
As expected based on differences in cover, the aerodynamic roughness lengths at
the burned site were significantly lower than at the unburned site (Mann Whitney U test,
Fig. 2b). Consistent with this finding, the friction velocities at the burned site were
significantly less than at the unburned site, and an analysis of the linear relationships of
the friction velocities with wind velocity resulted in significantly different slopes and
intercepts (Fig. 3) (p < 0.0001, Student’s t test).
Dust emission rates across the particle-size spectrum
The degree to which we were able to detect differences in dust emissions between
unburned and burned sites varied with measurement approach. Measurements by BSNE
samplers that generally are dominated by saltation were significantly greater at the
burned site relative to the unburned site (Fig. 4a; p = 0.038). However, total count for
132
saltating particles measured at the soil surface measured by the Sensit were not
significantly different between the unburned and burned sites (Fig. 4b; p = 0.827). For
aerosols, the TSP samplers did not detect significant differences (Fig. 4c; p=0.419). Yet,
for the high-resolution DustTrak measurements, the distribution for maximum 1 minute
aerosol concentration obtained within each hour of sampling was more than twice as
large for the burned site relative to the unburned site (Fig. 4d; p=0.0007).
Aerosol measurements at different temporal scales
Differences in aerosols concentrations between unburned and burned sites as
measured by the DustTrak increased with the degree of temporal resolution considered
(Fig. 5). When concentrations were averaged over the 5-hour measurement period for
each of the 5 days for which they were obtained, there was no significant difference
between unburned and burned sites (Fig. 5a; p = 0.395). However, differences were
significant when the 1-minute maximum concentration within each 5-hour sampling
period was considered (Fig. 5b; p = 0.021), and the site difference was most significant
when 1-minute maximum concentrations within each hour were considered, as reported
previously (Fig. 5c; p = 0.0007). The increase in degree of significance between these
two measurements could be partially related to the fact that the 1-hour data has a larger nvalue, which increases statistical power of detecting treatment differences.
Wind speed and saltation and dust fluxes
Our evaluations of wind speed (10-minute average) relative to dust emission rates
estimated from Sensit or DustTrak instruments yielded counterintuitive results (Fig. 6).
133
Rather than an expected increase in dust emissions with wind speed, our results indicated
a large number of dust emissions associated with wind speeds averaged over 10 minutes
that were low, including many values at a wind speed of 0 m/s.
Discussion
Not surprisingly, our results indicate that fire produced significantly elevated rates
of aeolian sediment transport and associated dust emissions based on most of the
measurements, which spanned a spectrum of particle-sizes and temporal resolution.
More specifically, our results indicate, as expected, that soil at the burned site had lower
friction velocities and aerodynamic roughness and was more erodible by wind than soil at
the unburned site. Our results build on previous related studies of dust emission rates in
burned vs. unburned sites (e.g., Whicker et al., 2002; Visser et al., 2004; Vermeire et al.,
2005), provide data on background emission rates when wind speeds are relatively low
and to provide a multi-instrumental approach that spans the particle-size spectrum
associated with aeolian-induced transport. Notably, we were able to detect significant
differences with some of the instruments (BSNE, DustTrak) but not others (Sensit, TSP).
The instrument-dependent responses that we detected were likely related to the
interplay among a combination of factors that include particle size, temporal resolution,
height-integration, and total sampling time. Overall, the high-temporal resolution
DustTrak measurements, which generally have not been obtained in the field, were the
most effective samplers for detecting statistically significant differences in background
rates of dust emissions between the two sites. The DustTraks provide high-temporal
resolution measurements that proved advantageous over the other instruments that were
134
used in this study, likely because of the dynamic and sporadic nature associated with
event-based aeolian sediment transport. Our results showed that differences in aeolian
sediment transport between the burned and unburned sites increased substantially with
increasing temporal resolution of the DustTrak measurements. In this case, averaging
over longer-time periods apparently marginalized any differences that were observed
between the burned and unburned sites. This highlights the importance of utilizing
various instruments in aeolian research that are capable of spanning the wide range of
particle sizes and temporal dynamics relevant to aeolian sediment transport processes
(Zobeck et al., 2003).
Perhaps most surprisingly, our data showed counterintuitive relationships with
both Sensit and DustTrak measurements. We expected dust emission measured with
either instrument to be minimal for low wind speeds and to increase with wind speed,
especially above some critical threshold. However, this pattern was not apparent (Fig. 6).
Indeed, many dust emission rates corresponded to a mean 10-minute wind speed of 0 m s1
. We anticipated, based on previous research, that 10-minute average wind speed would
be sufficient resolution to develop relationships between wind speed and dust emission
measure by Sensit or DustTrak. However, a plausible explanation for the counterintuitive
results that we observed is that, particularly for background dust emission rates
associated with more common lower wind speeds, high resolution wind data that captures
short gusts is needed (on the order of seconds); that is, the 10-minute average wind
speeds that we obtained may have included a few substantial gusts but over the 10-minute
sampling period were averaged away to a wind speed approaching 0 m s-1.
135
Consequently, future studies focused on background dust emission rates may need to
obtain high-resolution (1 s) wind speed data. Resolving relationships at low wind
velocities might also require having a wind anemometer that has better resolution at the
lower velocities than the Kestrel instruments are capable of.
In conclusion, our results on temporal measurements of dust emissions (Fig. 5)
and related wind speed (Fig. 6) both point to the importance of high-resolution
measurements for assessing background dust emission rates. In the absence of highresolution measurements, the BSNE samplers, which integrate across height, could
provide a sensitive measurement of site differences if sampled over a sufficiently long
period of time. More generally, our results suggest that interplay between particle size,
temporal resolution, height-integration, and total sampling time can be complex, and
although not often considered or measured explicitly (Fig. 1), may need to be considered
more explicitly and in concert to develop effective sampling for background dust
emissions.
Acknowledgements
The authors specifically wish to acknowledge Drs. Enrique de la Montaña, Miguel
Piñeros and Mitch McClaran for their help. This work was supported by Universidad de
Alcalá (LMM), NSF DEB 0816162, and Los Alamos National Laboratory.
References
Bagnold, R.A., 1941. The physics of blown sand and desert dunes. Chapman and Hall
Ltd., London.
136
Bouyoucos, G.J., 1962. Hydrometer method improved for making particle size snalyses
of soils. Agron. J. 54, 464-465.
Breshears, D.D., Whicker, J.J., Johansen, M.P., Pinder III, J.E., 2003. Wind and water
erosion and transport in semi-arid shrubland, grassland and forest ecosystems:
quantifying dominance of horizontal wind-drive transport. Earth Surf. Proc. Land. 28,
1189–1209
Field, J.P., Breshears, D.D., Whicker, J.J., 2009. Toward a more holistic perspective of
soil erosion: Why aeolian research needs to explicitly consider fluvial processes and
interactions. Aeolian Research. doi:10.1016/j.aeolia.2009.04.002
Field, J.P., Belnap, J., Breshears, D.D., Neff, J.C., Okin, G.S., Whicker, J.J., Painter,
T.H., Ravi, S., Reheis, M.C., Reynolds R.L., in press.. The ecology of dust. Front.
Ecol. Environ.
Gillette, D.A., 1977. Fine particulate emission due to wind erosion. Transactions of the
ASAE. 20, 890-897.
Gillette, D.A., Adams, J., Endo A., Smith D., Kihl R., 1980. Threshold velocities for
input of soil particles into the air by desert soils. J. Geophysical Res. 85, 5621-5630.
Gillette, D.A., Fryrear, D.W., Gill, T.E., Ley, T., Cahill, T.A., Gearhart, E.A., 1997.
Relation of vertical flux of particles smaller than 10 µm to total aeolian horizontal
mass flux at Owens Lake. J. Geophys. Res. 102. (D22). 26009-26015.
Goudie, A.S., Middleton, N.J., 2006. Desert dust in the global system. Heidelberg,
Germany, Springer-Verlag.
137
Herrick, J.E., Van Zee, J.W., Havstad, K.M., Burkett, L.M., Whitford, W.G., 2005.
Monitoring Manual for Grassland, Shrubland and Savanna Ecosystems, Volume I,
Quick Start. The University of Arizona Press, Tucson, Arizona.
Lee, J.A., Tchakerian, V.P., 1995. Magnitude and frequency of blowing dust on the
Southern High Plains of the United States, 1947-1989. Annals of the Association of
American Geographers. 85, 684-693.
Li, M., Li, Z., Liu, P., Yao, W., 2005. Using Cesium-137 technique to study the
characteristics of different aspects of soil erosion in the Wind-Water Erosion
Crisscross Region of Loess Plateau of China. Appl. Radiat. Isot. 62, 109–113.
McClaran, M.P., 2003. A Century of vegetation change on the Santa Rita Experimental
Range. In: Santa Rita Experimental Range: 100 Years (1903– 2003) of
accomplishments and contributions. Proceedings RMRS-P-30. U.S. Department of
Agriculture Forest Service, Rocky Mountain Research Station, Fort Collins CO.
McConnell, J.R., Aristarain, A.J., Banta, J.R., Edwards, P.R., Simones J.C., 2007. 20thCentury doubling in dust archived in an Antarctic Peninsula ice core parallels climate
change and desertification in South America. Proc. Nat. Acad. Sci. USA. 104, 57435748.
Neff, J.C., Ballantyne, A.P., Farmer, G.L., Mahowald, N.M., Conroy, J.L., Landry, C.C.,
Overpeck, J.T., Painter, T.H., Lawrence, C.R., Reynolds, R.L., 2008. Increasing
eolian dust deposition in the western United States linked to human activity. Nature
Geoscience. 1, 189 - 195.
138
Okin, G.S., Herrick, J.E., Gillette, D.A., 2006. Multiscale controls on and consequences
of aeolian processes in landscape change in arid and semiarid environments. J. Arid
Environ. 65, 253-75.
Pye, K., 1987. Aeolian dust and dust deposits. Academic Press, Boca Raton, Florida.
Ravi, S., D’Odorico, P.A., 2005. Field-scale analysis of the dependence of wind erosion
threshold velocity on air humidity. Geophys. Res. Lett. 32, L21404.
Statsoft. 2001. STATISTICA. Data Analysis Software System. Tulsa, OK.
Stout, J.E., 2001. Dust and environment in the Southern High Plains of North America.
Journal of Arid Environments. 47, 425-441.
Stout, J.E., Zobeck, T.M., 1996. The Wolfforth field experiment: a wind erosion study.
Soil Science. 161, 616-32.
Toy, T.J., Foster, G.R., Renard, K.G., 2002. Soil erosion: processes, prediction,
measurement and control. John Wiley & Sons, New York, NY.
Vermeire, L.T., Wester D.B., Mitchell, R.B., Fuhlendorf, S.D., 2005. Fire and grazing
effects on wind erosion, soil water content, and soil temperature. Journal of
Environmental Quality. 34, 1559-1565.
Visser, S.M., Sterk G., Ribolzi, O., 2004. Techniques for simultaneous quantification of
wind and water erosion in semi-arid regions. Journal of Arid Environments. 59, 699717.
Whicker, J.J., Breshears, D.D., Wasiolek, P.T, Kirchner, T.B., Tavani, R.A., Schoep,
D.A., Rodgers, J.C., 2002. Temporal and spatial variation of episodic wind erosion in
unburned and burned semiarid shrubland. J. Environ. Qual. 31, 599-612.
139
Whicker, J.J., Pinder, J.E., Breshears, D.D., Eberhart, C.F., 2006. From dust to dose:
effects of forest disturbance on increased inhalation exposure. Science of the Total
Environment. 368, 519–530.
Zobeck, T.M., Sterk, G., Funk, R., Rajot, J.L., Stout, J.E., Van Pelt, R. S., 2003.
Measurement and data analysis methods for field-scale wind erosion studies and
model validation. Earth Surf. Proc. Land. 28, 1163-1188.
140
Figures
Figure 1. Relationship between the sampling resolution based on temporal scale and
particle size for four instruments used to study dust emission rates: Sensit for saltation,
BSNE, or Big Spring Number Eight, for horizontal fluxes that are largely influenced by
slatation, TSP, or total suspended particulate, for aerosols across all sizes, and DustTrak
for measurement of smaller (< 10 μm) aerosols.
141
Figure 2. Site characteristics for unburned and burned sites for for: a) Soil cover and b)
Aerodynamic roughness (p=0.00005). The letters above the bars for panel b represent
significant differences. The error bars represent one standard error.
142
Figure 3. Aerodynamic friction velocities measured at the unburned and burned sites as a
function of the horizontal wind velocity measured at a height of 1m. Both slopes and
intercepts were significantly different (p < 0.0001).
143
Figure 4. Dust emission rates for unburned and burned plots from four instruments that
differ particle size and temporal scale: a) total particle count measured by the Sensit
(p=0.827); b) total horizontal flux (g·m-1·d-1) collected from the BSNE samplers
(p=0.038); c) Total Suspended Particle (TSP) concentration from a 5-h period for each of
5 d (p=0.419); d) maximum 1 min aerosol concentration from a 5-h period based on
evaluation of 1-h maxima, all for 5 different days (p=0.0007). The letters above each bar
represent significant differences. Error bars represent one standard error.
144
Figure 5. Estimates of dust emission measured by DustTrak at different scales of
temporal resolution, based on 1-min measurements: a) mean of five hours (p=0.395), b)
maxima of five hours (p=0.021), c) maxima over 5 h from maxima within each hour
(p=0.0007). The letters above each bar represent significant differences. Error bars
represent one standard error.
145
Figure 6. Scatter plots showing the relationship between wind velocity and dust
movement as measured by a) Sensit or by DustTrak using average 10-min concentrations
(b) or maximum within 10 min intervals (c); relationships between dust emission and
wind speed were not significant in all cases.
146
APPENDIX E
TOWARD A MORE HOLISTIC PERSPECTIVE OF SOIL EROSION: WHY
AEOLIAN RESEARCH NEEDS TO EXPLICITY CONSIDER FLUVIAL PROCESSES
AND INTERACTIONS
Jason P. Field1, David D. Breshears2, and Jeffrey J. Whicker3
1
School of Natural Resources, University of Arizona, Tucson, 85737 AZ USA
School of Natural Resources, Institute for the Study of Planet Earth, and Department of
Ecology and Evolutionary Biology, University of Arizona, Tucson, 85737 AZ USA
3
Environmental Programs, Los Alamos National Laboratory, Los Alamos, NM 87545
USA
2
Document Type: Peer-reviewed publication in Aeolian Research.
Abstract: Soil erosion is driven by not only aeolian but also fluvial transport processes,
yet these two types of processes are usually studied independently, thereby precluding
effective assessment of overall erosion, potential interactions between the two drivers,
and their relative sensitivities to projected changes in climate and land use. Here we
provide a perspective that aeolian and fluvial transport processes need to be considered in
concert relative to total erosion and to potential interactions, that relative dominance and
sensitivity to disturbance vary with mean annual precipitation, and that there are
important scale-dependencies associated with aeolian-fluvial interactions. We build on
previous literature to present relevant conceptual syntheses highlighting these issues. We
then highlight relative investments that have been made in soil erosion and sediment
control by comparing the amount of resources allocated to aeolian and fluvial research
using readily available metrics. Literature searches suggest that aeolian transport may be
somewhat understudied relative to fluvial transport and, most importantly, that only a
relatively small number of studies explicitly consider both aeolian and fluvial transport
147
processes. Numerous environmental issues associated with intensification of land use
and climate change impacts depend on not only overall erosion rates but also on
differences and interactions between aeolian and fluvial processes. Therefore, a more
holistic viewpoint of erosional processes that explicitly considers both aeolian and fluvial
processes and their interactions is needed to optimize management and deployment of
resources to address eminent changes in land use and climate.
Introduction
Aeolian processes in general, and soil transport and erosion in particular, present
widespread and substantial challenges in environmental science and management (Pye,
1987; Toy et al., 2002; Peters et al., 2006; CCSP, 2008). The consequences of aeolian
transport processes have important global implications (Cooke et al., 1993; Goudie,
2008) and are perhaps most evident in major dust storms across regionally degraded
landscapes, as experienced throughout much of North America during the 1930s Dust
Bowl era (Worster, 1979; Peters et al., 2007, 2008) and in China in association with
degraded northern drylands (Chepil, 1949; Shao and Shao, 2001). Although dust
deposition in some regions can have important beneficial effects, such as the transport of
nitrogen, phosphorous, and other essential nutrients to aquatic and terrestrial systems
(Swap et al., 1992; Chadwick et al., 1999; Neff et al., 2008), the detachment and removal
of wind-blown sediment from source areas can significantly lower soil fertility and water
holding capacity (Lal et al., 2003; Li et al., 2007, 2008), alter biogeochemical processes
(Schlesinger et al., 1990; Jickells et al., 2005), and increase land surface inputs of dust to
the atmosphere (Gillette and Passi, 1988; Tegen and Fung, 1994; Reynolds et al., 2001).
148
The catastrophic impacts of the North American Dust Bowl of the 1930s, as well
as the Sirocco dust events of 1901-1903, led to a widespread surge in interest in aeolian
processes and a significant increase in the number of publications in aeolian research
(Stout et al., 2009), many of which focused on basic aeolian transport processes or soil
conservation through improved land management. This interest subsequently benefitted
from novel, quantitative advances developed by Bagnold (1941), which have served as a
basic foundation for much of our current understanding of aeolian transport processes.
The aeolian research community has been growing steadily since Bagnold’s (1941)
classic work on aeolian entrainment and founding studies of the geomorphology of dune
fields (Stout et al., 2009). Aeolian transport is now clearly recognized as critical to land
surface dynamics for the environmental and geosciences research community and by
many within the resource management community (Peters et al., 2006; CCSP, 2008).
Although aeolian transport is generally recognized as important, current
understanding and focus on aeolian processes is often in isolation from the other primary
driver of land surface dynamics: fluvial transport (Heathcote, 1983; Baker et al., 1995;
Breshears et al., 2003; Visser et al., 2004). More specifically, researchers and
practitioners in soil conservation generally segregate into one of two disciplines, those
focusing on wind erosion or those focusing on water erosion. Many geomorphological
studies focus on inferring relative importance of aeolian vs. fluvial processes in soil
profiles, but these studies do not directly quantify concurrent, co-located rates of both
wind and water erosion. Although both wind and water erosion have contributed close to
one billion tons of soil loss per year within the United States (NRCS, 2000a, 2000b), and
149
they operate on many similar fundamentals, there are critical differences between the two
types of processes that drive this separation (Toy et al., 2002; Breshears et al., 2003;
Visser et al., 2004). These include major differences in the density of the transport fluid
(water verses air), directionality of sediment and dust transport, temporal scales of the
erosion events, and spatial scales of the impact (from localized to global). Though
research on aeolian transport has generally proceeded in isolation from fluvial transport,
there are numerous reasons to re-evaluate the interrelationships between wind and water
erosion and associated aeolian and fluvial processes (Heathcote, 1983; Baker et al., 1995;
Breshears et al., 2003; Bullard and McTainsh, 2003; Visser et al., 2004) because such
interrelationships may have important environmental and ecological consequences
(Aguiar and Sala 1999; Peters et al., 2006; Ravi et al., 2007b). The degree and manner in
which aeolian and fluvial transport processes are interrelated could also have important
implications for relative investments in research and soil conservation for controlling
erosion of both types. This issue is particularly pressing given the growing
environmental challenges related to maintaining agricultural productivity, preventing
ecosystem degradation, and adapting to the projected impacts of global climate change
(Lal et al., 2003; Nearing et al., 2005; CCSP, 2008).
The potential for soil erosion and land degradation due to synergistic effects of
aeolian and fluvial transport may well far exceed that of either type of process alone
(Bullard and Livingstone, 2002). Aeolian and fluvial transport processes can degrade
ecosystems and accelerate desertification (Schlesinger et al., 1990; Belnap, 1995; Peters
et al., 2006; Okin et al., 2009), and both processes can contribute substantially to total
150
erosion (Breshers et al., 2003; Bullard and McTainsh, 2003; Visser et al., 2004).
Combined, the effects of aeolian- and fluvial- driven soil loss have resulted in moderate
to severe soil degradation throughout much of the world’s arable land (Oldeman et al.,
1990; Pimentel, 1993). Globally, perhaps as much as one-third of all arable land has
experienced accelerated rates of erosion that undermine long-term productivity (Brown,
1981; USDA, 2006). It is clear that a majority of lands, what ever the use pattern, are
subject to both aeolian and fluvial transport processes and that these processes operate
together to redistribute soil and other critical resources, such as nutrients, organic debris,
seeds, and water (Schlesinger et al., 1990; Aguiar and Sala, 1999; Bullard and McTainsh,
2003). Interactions between aeolian and fluvial processes can have a large influence on
the transport and deposition of fine sediment and sand-sized material in dryland
environments. For example, aeolian entrainment from lake beds, river beds, and flood
plains can transport fluvial sediment long distances and subsequently deposit it as aeolian
material, at which point either fluvial or aeolian processes can further redistribute the
sediment, thus increasing the potential for interactions between aeolian and fluvial
processes (Bullard and Livingstone, 2002). Additional examples of aeolian-fluvial
interactions include glaciogenic outwash in major drainage systems supplying silt for
aeolian entrainment to form loess (Sun, 2002; Muhs et al., 2008), raindrop destruction of
soil aggregates to yield particle sizes suitable for deflation (Cornelis et al., 2004;
Chappell et al., 2005; Erpul et al., 2009), micro-topography formation beneath plant
canopies (Schlesinger et al., 1990; Ravi et al., 2007a), and reworking of hillslope loess to
form pedisediment (Ruhe et al., 1967).
151
Despite the importance of wind and water erosion over vast areas, field studies
comparing the absolute and relative magnitudes of both types of erosion are largely
lacking (Breshears et al., 2003; Visser et al., 2004). Although conceptual models and
limited field measurements suggest that both wind and water erosion can be of similar
magnitude in many environments (Kirkby, 1980; Baker et al., 1995; Valentin, 1996;
Breshears et al., 2003), substantial uncertainty remains about the relative magnitudes of
the two types of erosion and how they interrelate with each other because few studies
explicitly evaluate both processes. In addition, an integrated perspective of how these
processes contribute to total erosion and how they vary with scale and the degree to
which they interact is lacking. Given that recent field measurements and erosion models
indicate that both processes contribute substantially to total erosion (NRCS, 2000a,
2000b; Breshears et al., 2003) and that the ways in which they interact are being
considered more directly (Bullard and Livingstone, 2002; Bullard and McTainsh, 2003;
Visser et al., 2004), a key challenge that lies before the aeolian and fluvial research
communities is to develop a more integrated perspective of aeolian-fluvial dynamics.
The uncertainty about the relative magnitudes of aeolian and fluvial transport processes
needs to be addressed to develop more effective land management and could be useful in
guiding future deployment of resources. Here we address these key issues about aeolian
transport processes in the context of fluvial transport. Specifically, we (1) discuss the
scale-dependent and interactive ways in which aeolian and fluvial transport operate
across humid through arid environments, justifying the need of holistic evaluations; (2)
evaluate relative investments in research as measured through the number of publications
152
globally, and research stations and government funded erosion control based on data
from the United States, as an example, and (3) propose a prospectus for future studies of
aeolian transport in a scale-dependent context that explicitly considers aeolian-fluvial
interactions.
Environmental and Scale-Dependencies
Precipitation has a multifaceted role in soil transport that is particularly relevant in
that the magnitude of aeolian transport relative to fluvial transport likely varies strongly
with precipitation regimes among arid, semiarid/subhumid, and humid environments.
Precipitation amount affects vegetation cover, soil characteristics, and topography, all of
which are critical factors driving the amount of soil transport through both aeolian and
fluvial processes (Pye, 1987; Visser et al., 2004; Li et al., 2005). In general, there is
usually greater potential for soil erosion and transport in arid and semiarid environments
relative to humid environments, especially in areas with large available sediment supplies
(Fig. 1a). One of the most obvious spatial differences between aeolian and fluvial
sediment transport is the direction and dimensions of transport characteristics specific to
each process (Fig. 1a; Bullard and Livingstone, 2002; Breshears et al., 2003; Reiners and
Driese, 2004; Visser et al., 2004). Aeolian transport is two-dimensional, with transport
occurring in both vertical and horizontal directions, and omni-directional, with material
potentially being transported in any wind direction. In contrast, fluvial transport is
primarily one-dimensional, with flow occurring horizontal to slope, and unidirectional,
with the direction of transport being downslope. Fluvial transport is largely irreversible
because material transported in one direction will not be transported back toward the
153
source location in subsequent fluvial events. The spatial scales of aeolian and fluvial
transport are also different. Aeolian transport not only can occur in any changing wind
direction, but wind can transport dust on a much larger spatial scale, including globally
(Prospero et al., 2002; Goudie and Middleton, 2006). In contrast, fluvial transport is not
only limited to downslope direction but is also constrained by topographic barriers
associated with watershed drainage areas (Reiners and Driese, 2004). Although
important temporal differences exist between aeolian and fluvial transport, both processes
can occur over similar time scales in response to unique weather and climatic events. For
example, aeolian and fluvial sediment transport often occur as sporadic, event-based
phenomenon associated with either periods of strong winds (Stout, 2001; Whicker et al.
2002) or intense rain events (Dingman, 1994). However, aeolian sediment transport in
many dryland systems may be expected to occur more frequently than fluvial transport,
perhaps on almost a daily basis, because even short bursts of wind on generally calm days
can result in the detachment and transport of sediment (Stout and Zobeck, 1997).
Both aeolian and fluvial transport are sensitive to disturbances in vegetation cover
and soil surfaces, and these can change the relative magnitudes of the two types of
transport in a precipitation-regime-dependent context (Fig. 1b). Disturbances such as fire
or livestock grazing can alter both aeolian and fluvial transport and may alter their
relative importance (Toy et al., 2002; Whicker et al., 2002; Visser et al., 2004; Breshears
et al., 2009). In arid environments, which are typically characterized by sparse
vegetation cover (Greig-Smith, 1979; Aguiar and Sala, 1999), the effect of disturbance on
total ground cover is often less pronounced compared to changes in total ground cover
154
that would be expected following disturbance in humid environments or other
environments that may have nearly complete vegetation cover prior to disturbance
(Breshears et al., 2009). For example, the loss of protective vegetation cover due to
disturbance in a humid environment can have a dramatic impact on fluvial transport
(Brooks et al., 2003) because vegetation cover can change rapidly from nearly complete
cover to bare following disturbance (White, 1979; Sousa, 1984). A dramatic reduction in
vegetation cover can allow overland flow to become more concentrated and can increase
total runoff and sediment transport potential because less sink areas are available for
water storage (Johansen, et al., 2001; Dunkerley, 2002; Wilcox et al., 2003). Aeolian
transport in humid settings will also likely increase to some degree following disturbance,
but will probably be limited due to the higher soil moisture contents and atmospheric
humidity associated with humid environments (Ravi et al., 2004; Ravi and D’Odorico,
2005). Vegetation in humid environments should generally recover more rapidly than in
dryland environments because xeric soil conditions often cause plant water stress and can
impede vegetation recovery (Burke et al., 1998; Berlow et al., 2003; Chaves et al., 2009),
thereby increasing the total amount of soil erosion over time. Consequently, arid and
semi-arid climates can be viewed as being more vulnerable to long-term increases in
erosion following disturbance (unless, for example, channel erosion in a humid
environment becomes a persistent, reinforcing problem following disturbance).
When aeolian transport is considered in a more holistic context with fluvial
transport, the potential importance of total sediment transport from both processes and
their interactions becomes more readily apparent (Fig. 2a). Total sediment transport
155
could be a simple additive result of aeolian and fluvial transport, a starting assumption we
make here. If so, then the differential dependencies of aeolian and fluvial transport on
precipitation regimes would imply that total sediment transport in relatively undisturbed
systems should be greatest in semiarid environments rather than humid or arid ones.
Fluvial processes are typically thought to dominate sediment transport potential in humid
or mesic environments, whereas aeolian processes are typically thought to dominate
sediment transport in arid or xeric environments (e.g., Schumm 1965; Marshall, 1973;
Kirkby, 1978; Bullard and Livingstone, 2002). However, in semiarid and drylands
systems, which constitute approximately 40% of the earth’s land surface, neither aeolian
nor fluvial processes may dominate, based on the limited relevant studies that have
considered both types of transport directly (Breshears et al., 2003; Visser et al., 2004).
Note that the maximum for potential aeolian sediment transport does not necessarily
occur at the lowest levels of annual precipitation (Fig. 2a) because the lack of moisture in
hyperarid systems can result in the lack of available sediment for transport (Schumm
1965; Bullard and Livingstone, 2002; Gillette and Chen, 2001); exceptions would include
highly weathered or disturbed systems such as dune fields and agricultural land.
Similarly, the maximum potential for fluvial transport does not necessarily occur at the
highest levels of annual precipitation because vegetation cover typically increases as
moisture availability increases, thereby reducing the amount of exposed soil susceptible
to fluvial transport. Given these precipitation-regime dependencies for aeolian and
fluvial transport, the maximum potential for interaction between these two processes is
likely to occur under semiarid climatic conditions where neither process solely dominates
156
(Kirkby, 1978; Heathcote, 1983; Baker et al., 1995; Bullard and Livingstone, 2002;
Breshears et al., 2003). Semiarid systems have the greatest potential for aeolian-fluvial
interactions because these systems are often characterized by sparse vegetation cover,
making them highly erodible under both aeolian and fluvial forces. The greatest potential
for total sediment transport (i.e., combined aeolian and fluvial sediment transport) would
therefore likely be found in semiarid systems (Fig. 2a), where both processes are thought
to contribute substantially to total sediment transport (Breshears et al., 2003; Visser et al.,
2004).
Total sediment transport potential across precipitation regimes should differ
between disturbed and undisturbed conditions (Fig. 2b). We hypothesize that short-term
fluvial transport would increase more dramatically following disturbance relative to
aeolian transport because fluvial transport dominates humid systems, which undergo the
most dramatic change in vegetation cover following disturbance (Heathcote, 1983; Baker
at al., 1995; Johansen et al., 2001; Brooks et al., 2003). When compared to humid
systems, disturbances in arid and semiarid systems likely result in a smaller decrease in
the relative amount of vegetation cover because inherently a large portion of the soil
surface is already void of protective vegetation cover (Aguiar and Sala, 1999). The
maximum potential for aeolian-fluvial interaction, as well as the maximum total sediment
transport potential, is therefore expected to shift toward a more mesic environment
following disturbance and the loss of vegetation cover (Fig. 2a, b).
Additional insights emerge about the spatial and temporal scale-dependencies
associated with both aeolian and fluvial processes, as well as about their precipitation
157
dependencies, when a more holistic perspective is considered (Fig. 3a). As noted
previously, aeolian transport differs fundamentally from fluvial transport in that it can
occur in both the horizontal direction, as horizontal sediment flux that contributes to
localized redistribution, and vertically, with the suspended dust being subject to longdistant redistribution (Breshears et al., 2003; Zobeck et al., 2003). Event-based sediment
transport at small spatial and temporal scales (e.g., 102 m and 10-2 hr, respectively) is
dominated by fluvial processes because larger sediment and rock fragments can
constitute the majority of mass being moved short distances over very short times (e.g.,
during flash floods), whereas the force of wind is not great enough to immobilize these
larger fragments (Brooks et al., 2003). At large spatial and temporal scales, however,
aeolian transport is expected to be dominant because fluvial transport is confined to
channels and rivers within watershed boundaries, whereas aeolian transport is not
confined to watersheds and can therefore transport dust at distances that span the globe
(Fig. 3a). Notably, the greatest potential for aeolian-fluvial interactions occurs at
intermediate spatial and temporal scales (Fig. 3a) because both wind and water have the
potential to transport small- and medium-sized particles (e.g., sand, silt, and clay) over
intermediate distances (e.g., 101 to 106 m). Fluvial processes, in contrast to aeolian
processes, concentrate sediment transport capacity per unit source area with increasing
spatial scale because water flows only downslope thereby cumulatively compounding the
potential to transport sediment while at the same time reducing the relative spatial area
affected by the transported sediment (Fig. 3b; Reiners and Driese, 2004).
158
Because of these fundamental differences, aeolian and fluvial transport processes
are likely to interact to a lesser degree with increasing spatial scale. For example, at the
plot or hillslope scale, both processes can have roughly the same potential transport
capacity (Breshears et al., 2003) and both might be expected to have similar areas
impacted by the deposition of transported sediment (Fig. 3b). However, as spatial scale
increases to the landscape and regional scale, the potential for aeolian-fluvial interactions
likely decreases because aeolian transport capacity becomes weaker as the depositional
area increases, whereas fluvial transport capacity becomes stronger as the depositional
area decreases (e.g., gully and channel erosion; Fig. 3b).
Research Output and Resource Investment
Given that aeolian transport is hypothesized to dominate sediment transport in
many arid environments (Kirkby, 1980; Valentin, 1996), and is expected to be codominant and potentially interactive with fluvial transport in more mesic environments
(Bullard and Livingstone, 2002; Visser et al., 2004), and that these environmental settings
account for a large proportion of the terrestrial biosphere, to what extent does previous
research and investments in soil erosion and sediment control reflect the relative
importance of aeolian transport in these settings? Although there are limitations
associated with simple assessments of trends in peer-reviewed literature, such
assessments can nonetheless provide useful insights on research activity within the
scientific community. We evaluated the literature by conducting literature searches
(using ISI Web of Science) to quantify the number of publications associated with key
words related to aeolian and fluvial processes. We considered some of our results
159
relative to a precipitation gradient distinguishing among arid, semiarid/subhumid, and
humid environments. Our limited evaluation of available scientific papers on aeolian and
fluvial transport suggests that there are more studies on fluvial than aeolian transport,
even in drier environments where aeolian processes may indeed dominate (Fig. 4a).
More importantly, most studies only consider the aeolian or fluvial transport
component—very few explicitly considering both. Similarly, there appears to be more
studies of land surface dynamics and sediment transport processes that focus on fluvial
than aeolian transport processes, particularly those that evaluate erosional processes and
their impacts (Fig. 4b).
The results of our literature search suggest there have been more studies of fluvial
than aeolian transport and are notable given the previously raised point that the limited
relevant studies suggest that aeolian and fluvial transport can be of similar magnitude in
many environments (Kirkby, 1980; Baker et al., 1995; Valentin, 1996; Breshears et al.,
2003). To extend our overview to other metrics, we used the United States as an example
to compare the relative magnitudes of wind and water erosion on agricultural land, the
economic cost of both types of erosion, and the amount of resources devoted to soil
conservation and aeolian and fluvial research. Although these trends likely vary greatly
for other countries and need to be considered in a broader context to assess global trends,
we focused here on wind and water erosion within the United States where relevant data
on both processes were readily available. Wind erosion for United States agricultural
lands, estimated to account for ~8 x 105 tons of soil loss per year, is nearly as large as
water erosion, estimated to be ~1 x 106 tons of soil loss per year (NRCS, 2000a, 2000b).
160
The total area of United States agricultural land (cropland and Conservation Reserve
Program land only; data not readily available for rangeland) that is eroding at a rate
greater than 5 tons per acre per year, which is over twice the national average, is
approximately the same for wind erosion (40 million acres) and water erosion (41 million
acres; NRCS, 2000a, 2000b). Additionally although the western United States is
primarily dominated by wind erosion and the eastern United States is primarily
dominated by water erosion, there are substantial areas in the central, mid-west, and
northwest portions of the United States where neither of the two processes dominates and
both contribute substantially to total erosion rates, based on model assessments (Fig. 5a).
In short, these examples highlight that both aeolian and fluvial transport are
important to consider in many cases when assessing the overall environmental and
economic impact of sediment transport and soil erosion (Pimentel, 2000; Lal et al., 2003;
Visser et al., 2004). The total on-site and off-site costs of wind and water erosion on
United States agricultural land have previously been estimated to be 9.6 and 7.4 billion
United States dollars per year, respectively (Pimentel et al., 1995). Notably, however, the
amount of resources allocated to help combat erosion and its environmental and
ecological impact is not distributed proportionally relative to annual rates and costs
associated with wind and water erosion in the United States. For example, the USDA
Environmental Quality Incentive Program (EQIP) spends up to 60 times the amount per
acre of agricultural land on soil erosion and sediment control practices in the watererosion dominated eastern United States than in the wind-erosion dominated western
United States (Fig. 5b; NRCS, 2008), even though annual rates of wind and water erosion
161
are nearly identical in the United States for agricultural cropland and land in the
Conservation Reserve Program (NRCS, 2000a, 2000b). Resources in the United States
are also not distributed proportionally among money spent on aeolian and fluvial
research. The USDA Agricultural Research Service (ARS) is the primary research
agency in the United States responsible for assessing and mitigating erosional impacts on
agricultural lands. The amount of resources and number of research locations devoted to
aeolian and fluvial research within this agency suggests a disproportionate amount of
resources is directed toward research related to fluvial processes. For example, the
number of USDA-ARS experimental watersheds outnumbers the number of wind erosion
units by a factor of 18 (Fig. 5c). This apparent fluvial bias is somewhat ironic because
much of the current United States soil conservation policy was initiated as a direct
response to the devastating environmental and economic impacts of wind erosion and
dust storms from agricultural lands in the Great Plains during the 1930s Dust Bowl
(Worster, 1979).
These collective points summarizing research and resource investments imply that
there may be a bias toward fluvial processes over aeolian processes such that investments
in research (globally) and erosion control (at least within the United States) have not been
in proportion to the relative importance of aeolian transport processes. If additional
assessment supports the findings of this initial overview, then insights from adopting a
more holistic perspective of aeolian and fluvial processes could aide scientists, land
managers, government agencies, and especially policy-makers in optimizing effective
distribution of resources.
162
Addressing Emerging Challenges
Emerging challenges related to land use intensification and climate change
reinforce the need for a more holistic perspective of soil erosion and associated aeolian
and fluvial processes. Because co-located, simultaneous measurements of wind and
water erosion are lacking, it is difficult to assess how changes in land use and particularly
climate change will alter relative rates of wind and water erosion. In regions such as the
southwestern United States where climate is projected to shift to more arid, “Dust- Bowllike” conditions (Seager et al., 2007) and land use is also projected to intensify (CCSP,
2008), risks to soil surface stability could be substantial. Importantly, assessing such
risks requires evaluating how both wind and water erosion—not simply one or the
other—will likely respond. Little information exists in support of the widespread implicit
assumption that wind erosion is not influenced by water erosion. If aeolian processes
contribute a substantial amount to total erosion, as contended here and elsewhere
(Kirkby, 1980; Baker et al., 1995; Valentin, 1996; Breshears et al., 2003), then we
suggest that aeolian researchers need to explicitly consider how aeolian transport
processes influence and depend on fluvial transport processes in a scale-dependent and
environmental gradient context. We propose some themes that could be the center of a
research agenda addressing the relative importance of aeolian and fluvial transport
processes, their interactions, and implications for land management and economics
(Table 1).
In presenting a holistic perspective that emphasizes considering aeolian transport
relative to fluvial transport, we have highlighted two major points. First, that aeolian and
163
fluvial transport processes need to be considered in concert to appropriately assess and
manage total erosion and the associated scale-dependencies of aeolian-fluvial
interactions. Second, that investments made in aeolian research on dust emission and
management, based on data from the United States—but perhaps relevant elsewhere—
have been smaller than those for fluvial research. These points are based on limited
relevant information but are consistent with available research, previously posed
hypotheses, and available cost and management metrics. The key hypotheses that we
present emerge from previous syntheses and recent research and provide both challenges
and opportunities for aeolian researchers to more directly engage fluvial researchers and
to enhance overall management effectiveness related to erosion. We suggest that
addressing such a research agenda will be important scientifically and could provide a
means for realigning research and management investments with relative magnitudes of
wind and water erosion. Importantly, the implicit assumption that is made in most
studies and assessments that wind and water erosion do not influence each other is one
that needs to be explicitly tested. In conclusion, we suggest that land management that
depends on soil surface stability in the face of changing land use and climate is unlikely
to be effective unless we develop a more holistic understanding of not only aeolian
transport and erosional processes alone but also of their potential interactions with fluvial
transport and erosional processes.
Acknowledgements
This perspective was developed with support of the Arizona Agricultural Experiment
Station (DDB), USDA CSREES (JPF, DDB), and the National Science Foundation
164
(DDB, JPF; NSF-DEB 0816162). We thank D. J. Law for assistance.
References
Aguiar, M.R., Sala, O.E., 1999. Patch structure, dynamics, and implications for the
functioning of arid ecosystems. Trends Ecol. Evol. 14, 273-277.
Bagnold, R.A., 1941. The Physics of Blown Sand and Desert Dunes. Chapman and Hall
Ltd., London.
Baker, M.B. Jr., DeBano, L.F., Ffolliott, P.F., 1995. Soil loss in piñon-juniper ecosystems
and its influence on site productivity and desired future condition, in: Shaw, D.W.,
Aldon, E.F., LoSapio, C. (Eds.), Desired Future Conditions for Piñon-Juniper
Ecosystems, Proceedings of Symposium, 8–12 August 1994, Flagstaff, AZ. General
Technical Report. RM-258. US Department of Agriculture, Forest Service, Rocky
Mountain Forest and Range Experiment Station: Ft Collins, CO, pp. 9–15.
Belnap, J., 1995. Surface disturbances: their role in accelerating desertification. Environ.
Monit. Assess. 37, 39-57.
Berlow, E.L., D’Antonio, C.M., Swartz. H., 2003. Response of herbs to shrub removal
across natural and experimental variation in soil moisture. Ecol. Appl. 13, 1375-1387.
Breshears, D.D., Whicker, J.J., Johansen, M.P., Pinder III, J.E., 2003. Wind and water
erosion and transport in semi-arid shrubland, grassland and forest ecosystems:
quantifying dominance of horizontal wind-drive transport. Earth Surf. Proc. Land. 28,
1189-1209.
Breshears, D.D., Whicker, J.J., Zou, C.B., Field, J.P., Allen, C.D., 2009. A conceptual
framework for dryland aeolian sediment transport along the grassland–forest
165
continuum: effects of woody plant canopy cover and disturbance. Geomorphology
105, 28-38.
Brooks, K.N., Ffolliott, P.F., Gregersen, H.M., Debano, L.F., 2003. Hydrology and the
Management of Watersheds. Iowa State Press, Ames, Iowa.
Brown, L.R., 1981. World population growth, soil erosion, and food security. Science
214, 995-1000.
Bullard. J.E., Livingstone, I., 2002. Interactions between aeolian and fluvial systems in
dryland environments. Area 34, 8-16.
Bullard, J.E., McTainsh, G.H., 2003. Aeolian-fluvial interactions in dryland
environments: examples, concepts and Australia case study. Prog. Phys. Geog. 27,
471-501.
Burke, I.C., Lauenroth, W.K., Vinton, M.A., Hook, P.B., Kelly, R.H., Epstein, H.E.,
Aguiar, M.R., Robles, M.D., Aguilera, M.O., Murphy, K.L., Gill, R.A., 1998. Plantsoil interactions in temperate grasslands. Biogeochemistry 43, 121-143.
CCSP, 2008. The Effects of Climate Change on Agriculture, Land Resources, Water
Resources, and Biodiversity in the United States. United States Climate Change
Science Program. Washington, D.C.
Chadwick, O.A., Derry, L.A., Vitousek, P.M., Huebert, B.J., Hedin, L.O., 1999.
Changing sources of nutrients during four million years of ecosystem development.
Nature 397, 491-497.
166
Chappell, A., Zobeck, T.M., Brunner, G., 2005. Using on-nadir spectral reflectance to
detect soil surface changes induced by simulated rainfall and wind tunnel abrasion.
Earth Surf. Proc. Land. 4, 489-511.
Chaves, M.M., Flexas, J., Pinheiro, C., 2009. Photosynthesis under drought and salt
stress: regulation mechanisms from whole plant to cell. Ann. Bot. 103, 551-560.
Chepil, W.S., 1949. Wind erosion control with shelterbelts in North China. Agron. J. 41,
127-129.
Cooke, R.U., Warren, A., Goudie, A.S., 1993. Desert Geomorphology. UCL Press,
London.
Cornelis, W.M., Oltenfreiter, G., Gabriels, D., Hartmann, R., 2004. Splash-saltation of
sand due to wind-driven rain: vertical deposition flux and sediment transport rate.
Soil Sci. Soc. Am. J. 68, 32-40.
Dingman, S.L., 1994. Physical Hydrology. Prentice-Hall, Upper Saddle River, New
Jersey.
Dunkerley, D., 2002. Systematic variation of soil infiltration rates within and between the
components of the vegetation mosaic in an Australian desert landscape. Hydrol.
Process. 16, 119-131.
Erpul, G., Gabriels, D., Cornelis, W.M., Sammy, H., Guzelordu, T., 2009. Sand transport
under increased lateral jetting of raindrops induced by wind. Geomorphology 104,
191-202.
167
Gillette, D.A., Chen, W., 2001. Particle production and aeolian transport from a ‘supplylimited’ source area in the Chihuahuan Desert, New Mexico, United States. J.
Geophys. Res. 106, 5267-5278.
Gillette, D.A., Passi, R., 1988. Modeling dust emission caused by wind erosion. J.
Geophys. Res.-Atmos. 93, 14233-14242.
Goudie, A.S., 2008. The history and nature of wind erosion in deserts. Annu. Rev. Earth
Planet. Sci. 36, 97-119.
Goudie, A.S., Middleton, N.J., 2006. Desert dust in the global system. Springer-Verlag,
Heidelberg, Germany.
Greig-Smith, P., 1979. Pattern in vegetation. J. Ecol. 67, 755-779.
Heathcote, R.L., 1983. The Arid Lands: Their Use and Abuse. Longman, New York.
Jickells, T.D., An, Z.S., Andersen, K.K., Baker, A.R., Bergametti, G., Brooks, N., Cao,
J.J., Boyd, P.W., Duce, R.A., Hunter, K.A., Kawahata, H., Kubilay, N., laRoche, J.,
Liss, P.S., Mahowald, N., Prospero, J.M., Ridgwell, A.J., Tegen, I., Torres, R., 2005.
Global iron connections between desert dust, ocean biogeochemistry, and climate.
Science 308, 67-71.
Johansen, M.P., Hakonson, T.E., Breshears, D.D., 2001. Post-fire runoff and erosion
from rainfall simulation: contrasting forests with shrublands and grasslands. Hydrol.
Process.15, 2953-2965.
Kirkby, M.J., 1978. The stream head as a significant geomorphic threshold. Department
of Geography, University of Leeds Working Paper 216.
Kirkby, M.J., 1980. The problem, in: Kirkby, M.J., Morgan, R.P.C. (Eds.), Soil Erosion.
168
John Wiley and Sons, New York, pp. 1-16.
Lal, R., Sobecki, T.M., Iivari, T., Kimble, J.M., 2003. Soil degradation in the United
States: extent, severity, and trends. Lewis Publ., Boca Raton, Florida.
Li, J., Okin, G.S., Alvarez, L.J., Epstein, H.E., 2008. Effects of wind erosion on the
spatial heterogeneity of soil nutrients in a desert grassland of southern New Mexico.
Biogeochemistry 88, 73-88.
Li, J., Okin, G.S., Hartman, L.J., Epstein, H.E., 2007. Quantitative assessment of wind
erosion and soil nutrient loss in desert grasslands of southern New Mexico, USA.
Biogeochemistry. 85, 317-332.
Li, M., Li, Z., Liu, P., Yao, W., 2005. Using Cesium-137 technique to study the
characteristics of different aspect of soil erosion in the Wind-Water Erosion
Crisscross Region of Loess Plateau of China. Appl. Radiat. Isot. 62, 109-113.
Marshall, J. K., 1973. Drought, land use and soil erosion, in: Lovett, J.V. (Eds.), The
Environmental, Economic, and Social Significance of Drought. Angus and
Robertson, Sydney, pp. 55-77.
Muhs, D.R., Bettis, E.A., Aleinikoff, J.N., McGeehin, J.P., Beann, J., Skipp, G.,
Marshall. B.D., Roberts, H.M., Johnson, W.C., Benton, R., 2008. Origin and
paleoclimatic significance of late Quaternary loess in Nebraska: evidence from
stratigraphy, chronology, sedimentology, and geochemistry. Geol. Soc. Am. Bull.
120, 1378-1407.
Nearing, M.A., 2005. Soil erosion under climate change: rates, implications and
feedbacks-Introduction. Catena 61, 103-104.
169
Neff, J.C., Ballantyne, A.P., Famer, G.L., Mahowald, N.M., Conroy, J.L., Landry, C.C.,
Overpeck, J.T., Painter, T.H., Lawrence, C.R., Reynolds, R.L., 2008. Increasing
eolian dust deposition in the western United States linked to human activity. Nature Geosciences. doi:10.1038/ngeo133.
NRCS, 2000a. Average Annual Soil Erosion by Water on Cropland and CRP Land, 1997.
Map ID: m5058. NRCS, Resource Assessment Division, Washington, DC.
NRCS, 2000b. Average Annual Soil Erosion by Wind on Cropland and CRP Land, 1997.
Map ID: m5065. NRCS, Resource Assessment Division, Washington, DC.
NRCS, 2008. FY-2007 EQIP Payments for 1997-2007 Soil Erosion and Sediment
Control Practices per Agricultural Acre. Map ID: m10048. NRCS, Resource
Assessment Division, Washington, DC.
Okin, G.S., Parsons, A.J., Wainwright, J., Herrick, J.E., Bestelmeyer, B.T., Peters,
D.P.C., Fredrickson, E.L., 2009. Do changes in connectivity explain desertification?
BioScience 59, 237-244.
Oldeman, L., Hakkeling, R., Sombroek, W., 1990. World Map of the Status of HumanInduced Soil Degradation: An Explanatory Note. International soil reference and
information center, Wageningen, The Netherlands and United Nations Environmental
Program, Nairobi, Kenya.
Peters, D.P.C., Bestelmeyer, B.T., Herrick, J.E., 2006. Disentangling complex
landscapes: new insights into arid and semiarid system dynamics. Bioscience 56, 491501.
170
Peters, D.P.C., Groffman, P.M., Nadelhoffer,K.J., Grimm, N.B., Collins, S.C., Michener,
W.K., Huston, M.A., 2008. Living in an increasingly connected world: a framework
for continental-scale environmental science. Frontiers Ecol. Environ. 6, 229-237.
Peters, D.P.C., Sala, O.E., Allen, C.D., Covich, A., Brunson, M., 2007. Cascading events
in linked ecological and socio-economic systems: predicting change in an uncertain
world. Front. Ecol. Environ. 5, 221-24.
Pimentel, D. (Ed.), 1993. World Soil Erosion and Conservation. Cambridge Univ. Press,
Cambridge, UK.
Pimentel, D., 2000. Soil as an endangered ecosystem. Bioscience 50, 947-947.
Pimentel, D., Harvey, C., Resosudarmo, P., Sinclair, K., Kurz, D., McNair, M., Crist, S.,
Shpritz, L., Fitton, L., Saffouri, R., Blair, R., 1995. Environmental and economic
costs of soil erosion and conservation benefits. Science. 267, 1117-1123.
Prospero, J.M., Ginoux, P., Torres, O., Nicholson, S.E., Gill, T.E., 2002. Environmental
characterization of global sources of atmospheric soil dust identified with the
Nimbus-7 Total Ozone Mapping Spectrometer (TOMS) absorbing aerosol product.
Rev. Geophys. 40, 1002, doi:10.1029/2000RG000095.
Pye, K., 1987. Aeolian dust and dust deposits. Academic Press, Boca Raton, Florida.
Ravi, S., D’Odorico, P.A., 2005. Field-scale analysis of the dependence of wind erosion
threshold velocity on air humidity. Geophys. Res. Lett. 32, L21404.
Ravi, S., D’Odorico, P., Okin, G.S., 2007a. Hydrologic and aeolian controls on
vegetation patterns in arid landscapes. Geophys. Res. Lett. 34, L24S23.
171
Ravi, S., D’Odorico, P., Over, T.M., Zobeck, T.M., 2004. On the effect of air humidity
on soil susceptibility to wind erosion: the case of air-dry soils. Geophys. Res. Lett.
31, L09501.
Ravi, S., D'Odorico, P., Zobeck, T.M., Over, T.M., Collins, S., 2007b. Feedbacks
between fires and wind erosion in heterogeneous arid lands. J. Geophys ResBiogeosciences 112, G04007.
Reiners, W.A., Driese, K.L., 2004. Transport processes in nature: propagation of
ecological influences through environmental space. Cambridge University Press,
United Kingdom.
Reynolds, R., Belnap, J., Reheis, M., Lamothe, P., Luiszer, F., 2001. Aeolian dust in
Colorado Plateau soils: nutrient inputs and recent change in source. Proc. Nat. Acad.
Sci. USA 98, 7123-7127.
Ruhe, R.V., Daniels, R.B., Cady, J.G., 1967. Landscape evolution and soil formation in
southwestern Iowa. USDA Tech. Bull. 1349, Washington, D.C.
Schlesinger, W.H., Reynolds, J.F., Cunningham, G.L., Huenneke, L.F., Jarrell, W.M.,
Virginia, R.A., Whitford, W.G., 1990. Biological feedbacks in global desertification.
Science 247, 1043-1048.
Schumm, S.A., 1965. Quaternary paleohydrology, in: Wright, H.E. Jr., Frey, D.G. (Eds.),
The Quaternary of the United States, A review volume for the VII Congress INQUA.
Princeton, New Jersey, pp. 783-794.
Seager, R., Ting, M.F., Held, I., Kushnir, Y., Lu, J., Vecchi, G., Huang, H.P., Harnik, N.,
Leetmaa, A., Lau, N.C., Li, C.H., Velez, J., Naik, N., 2007. Model projections of an
172
imminent transition to a more arid climate in southwestern North America. Science
316, 1181-1184.
Shao, Y.P., Shao, S.X. 2001. Wind erosion and wind erosion research in China: a
review. Ann. Arid Zone 40, 317-336.
Sousa, W.P., 1984. The role of disturbance in natural communities. Ann. Rev. Ecol. Syst.
15, 353-391.
Stout, J.E., 2001. Dust and environment in the Southern High Plains of North America. J.
Arid Environ. 47, 425-441.
Stout, J.E., Warren, A., Gill, T.E., 2009. Publication trends in aeolian research: an
analysis of the Bibliography of Aeolian Research. Geomorphology 105, 6-17.
Stout, J.E., Zobeck, T.M., 1997. Intermittent saltation. Sedimentology 44, 959-970.
Sun, J.M., 2002. Provenance of loess material and formation of loess deposits on the
Chinese Loess Plateau. Earth Planet Sci. Lett. 203, 845-859.
Swap, R., Garstang, M., Greco, S., Talbot, R., Kallberg, P., 1992. Saharan dust in the
Amazon Basin. Tellus B Chem. Phys. Meteorol. 44, 133-149.
Tegen, I., Fung, I., 1994. Modeling of mineral dust in the atmosphere – sources,
transport, and optical-thickness. J. Geophys. Res.-Atmos. 99, 22897-22941.
Toy, T.J., Foster, G.R., Renard, K.G., 2002. Soil erosion: processes, prediction,
measurement and control. John Wiley & Sons, New York.
USDA, 2006. Conservation Resource brief: Soil Erosion. United States Department of
Agriculture, Natural Resources Conservation Service February 2006, Brief Number
0602.
173
Valentin, C., 1996. Soil erosion under global change, in: Walker, B., Steffen, W. (Eds.),
Global Change and Terrestrial Ecosystems. Cambridge University Press, New York,
pp. 317-338.
Visser, S.M., Sterk, G., Ribolzi, O., 2004. Techniques for simultaneous quantification of
wind and water erosion in semi-arid regions. J. Arid Environ. 59, 699-717.
Whicker, J. J., Breshears, D.D., Wasiolek, P.T, Kirchner, T.B., Tavani, R.A., Schoep,
D.A., Rodgers, J.C., 2002. Temporal and spatial variation of episodic wind erosion in
unburned and burned semiarid shrubland. J. Environ. Qual. 31, 599-612.
White, P.S., 1979. Pattern, process, and natural disturbance in vegetation. Bot. Rev. 45,
229-299.
Wilcox, B.P., Breshears, D.D. Allen, C.D., 2003. Ecohydrology of a resource-conserving
semiarid woodland: effects of scale and disturbance. Ecol. Monogr. 73, 223-239.
Worster, D., 1979. Dust Bowl: the southern plains in the 1930s. Oxford University Press,
New York.
Zobeck, T.M., Sterk, G., Funk, R., Rajot, J.L., Stout, J.E., Van Pelt, R. S., 2003.
Measurement and data analysis methods for field-scale wind erosion studies and
model validation. Earth Surf. Proc. Land. 28, 1163-1188.
174
Figures
Figure 1. Erosion behavior under (a) undisturbed vegetation cover and (b) postdisturbance vegetation cover along a hypothetical moisture gradient spanning humid to
arid environments. Length of arrow approximates transport distance; width of arrow
approximates transport capacity or total sediment flux by aeolian (red) and fluvial (blue)
processes; vertical arrows indicate vertical dust flux and length represents the degree of
connectivity.
175
Figure 2. Hypothesized trends of potential sediment transport capacity as a function of
mean annual precipitation to highlight the potential total sediment transport for (a)
undisturbed, and (b) disturbed sites (modified from Schumm, 1965; Marshall, 1973;
Kirkby, 1978). At the most arid sites, aeolian sediment transport is supply limited
(except for sand dunes and highly disturbed systems), and at the most mesic sites, fluvial
sediment transport is limited due to high vegetation cover. Note that the scales differ,
with curves from (a) provided for reference in (b). Potential for increased sediment
transport following disturbance is greater for fluvial transport than for aeolian transport
because there is a greater potential for a large reduction in vegetation cover in humid
environments relative to arid environments.
176
Figure 3. (a) Transport distances and event-based transport/entrainment times,
highlighting differences between fluvial vs. aeolian dominance and scales at which
aeolian-fluvial interactions are potentially most important. (b) Scale-dependent
interactions between aeolian and fluvial transport, highlighting maximum interactions at
plot scale. Width between red or blue lines indicates the maximum depositional area.
Note that the potential for fluvial sediment transport capacity increases with increasing
scale, but the maximum deposition area simultaneously decreases. Horizontal aeolian
sediment flux can move to hillslope and landscape scales, whereas vertical dust flux can
extend to regional scales and has the maximum deposition area.
177
Figure 4. Literature search results based on ISI Web of Knowledge: (a) total number of
aeolian and fluvial papers as a function of aridity; and (b) total number of aeolian and
fluvial papers by erosional process and erosional impact. Searches were based on the
following criteria Timespan = (1978-2007). Databases = SCI-EXPANDED. Refined by:
Document Type = (ARTICLE OR PROCEEDINGS PAPER OR REVIEW). Keywords
for Aeolian Processes: Topic = (aeolian OR eolian OR loess OR "wind erosion" OR
"wind-driven transport" OR "surface creep" OR saltation OR "dust flux" OR "dust
transport" OR "dust load"). Keywords for Fluvial Processes: Topic = (fluvial OR
alluvium OR alluvial OR "water erosion" OR "sheet erosion" OR "rill erosion" OR "gully
erosion" OR "channel erosion" OR "suspended sediment load" OR bedload). Keywords
for Both Processes: at least one term from both the aeolian search and fluvial search.
178
Figure 5. (a) Agricultural areas (cropland and Conservation Reserve Program lands only)
dominated by wind and water erosion are similar in magnitude and area (from NRCS, 2000a,
2000b) and are similar with respect to total off-site costs of erosion (Pimentel et al., 1995). (b)
The amount of EQIP funds spent on soil erosion and sediment control (from NRCS, 2008) is
much greater in areas dominated by water erosion than in areas dominated by wind erosion. (c)
Locations of the Agricultural Research Service sites focused on fluvial transport (experimental
watersheds) and on aeolian transport (wind erosion units).
179
Tables
Table 1. Key knowledge gaps about aeolian transport relative to fluvial transport.
Relative
magnitudes across
precipitation
gradients
Interactions
Management and
economics
Total sediment transport from aeolian and fluvial
processes is usually greatest in semiarid ecosystems
relative to arid, subhumid, and humid ecosystems.
Figs. 1a
Aeolian transport is usually most sensitive to
disturbance in semiarid ecosystems, whereas fluvial
transport is usually most sensitive to disturbance in
humid ecosystems because vegetation cover can be
reduced from complete to nothing.
Aeolian and fluvial processes can be closely
interrelated at intermediate scales (e.g., aeolian
transport can prime fluvial transport; fluvial transport
can concentrate or expose new sediment, increasing
availability for aeolian transport; rainsplash can
simultaneously affect both processes).
Aeolian and fluvial processes are likely to exhibit
maximum potential for interactions at small spatial
and temporal scales; at large scales the processes are
likely to be more decoupled because fluvial transport
is unidirectional and concentrates sediment with
increasing spatial scale, whereas aeolian transport is
omni-directional and disperses sediment with
increasing spatial scale.
Investments in soil erosion and sediment control
within the United States disproportionally match risks
associated with wind and water erosion.
The amount of resources and number of research
locations within the USDA Agricultural Research
Service suggests that a disproportionate amount of
resources is directed toward research related to fluvial
processes.
Figs. 1b
and 2a
and 2b
Fig. 3a
Fig. 3b
Fig. 5a, b
Fig. 5c
180
APPENDIX F
A DUSTIER FUTURE UNDER GLOBAL-CHANGE-TYPE EROSION?
Jason P. Field1, David D. Breshears1,2, Jeffrey J. Whicker3 & Chris B. Zou4
1
School of Natural Resources and the Environment, University of Arizona, Tucson, AZ
85721 USA; 2Department of Ecology and Evolutionary Biology, Institute of the
Environment, University of Arizona, Tucson, AZ 85721; 3Environmental Programs, Los
Alamos National Laboratory, Los Alamos, NM 87545; 4Department of Natural Resource
Ecology and Management, Oklahoma State University, Stillwater, OK 74078.
Document Type: In review as of Dec 4, 2009.
Abstract: Drought and intense precipitation events are projected to increase in many
dryland regions as consequences of global climate change. These comingled climate
changes are likely to alter soil erosion through changes in wind-driven dust emissions and
water-driven sediment production, yet because wind and water erosion are generally
studied independently, holistic knowledge of soil erosion response under projected
climate change remains elusive. Our co-located erosion measurements in a semiarid
grassland show that wind-driven sediment transport exceeded water-driven sediment
transport by a factor of ~40 during a year with wet-dry extremes—global change-type
erosion—corresponding to a more than an order-of-magnitude increase relative to
baseline conditions. We suggest that climate change that combines more intense
precipitation events and extreme droughts could substantially increase rates of wind
erosion relative to water erosion, such that vast drylands could potentially become dustier
in the future as a result of global climate change.
181
Introduction
Both wind and water erosion are fundamental drivers of land surface dynamics
and both remain a serious and persistent environmental problem worldwide, affecting soil
productivity, air and water quality, and ecosystem health (Trimble and Crosson 2000;
MEA 2005). Notably, roughly two-thirds of the world’s arable land is affected by
moderate to severe soil degradation (Pimentel et al 1995), most of which is attributed to
wind and water erosion (Oldeman et al. 1990). The combined impact of wind and water
erosion on agricultural land translates directly into considerable financial costs (Pimentel
et al 1995), and their effects permeate across all major types of ecosystem goods and
services (MEA 2005). Wind and water erosion can be drivers of desertification and both
can contribute substantially to total erosion in drylands (Breshears et al. 2003; Field et al.
2009), which comprise a large fraction of the terrestrial land surface. Further, synergistic
relationships between wind and water erosion could be particularly important in dryland
ecosystems because aeolian and fluvial processes can interact on the land surface to
redistribute soil and other critical resources, such as nutrients, seeds, and water (Aguiar
and Sala 1999). The degree to which wind and water erosion interact could have
important implications not only for maintaining agricultural productivity but also for
maintaining ecosystem health and for adapting to projected impacts of climate change
(Pimentel et al. 1995; MEA 2005). For example, at regional scales dust emissions from
drylands can suppress surface precipitation via effects on atmospheric radiative forcing
and condensation nuclei (Rosenfeld et al. 2001) and can reduce the duration of mountain
snow cover via effects on albedo (Painter et al. 2007)—both of which can have profound
182
impacts on water resources for many dryland regions. In addition, the transport and
deposition of dust from regional source areas, such as the Sahara in North Africa, can
have important global implications for terrestrial and aquatic productivity through the
addition of essential elements (e.g., nitrogen, phosphorous) to nutrient-poor environments
such as intercoastal waters and highly weathered tropical soils (Goudie and Middleton
2001).
Wind and water erosion can occur at the same location over similar time scales in
response to unique weather and climatic events, and both can contribute substantially to
overall erosion, as inferred indirectly from soil pedology (Sweet 1999), experimental
field measurements (Breshears et al. 2003), and erosion modeling (Trimble and Crosson
2000). Unfortunately, wind and water erosion are almost always studied independently
(Field et al. 2009), and rates of wind and water erosion have rarely, if ever, been
measured directly at the same location and during the same time period. Such direct
comparisons have generally been precluded by challenges associated with differences in
spatial qualities, in that flux footprints of aeolian and fluvial processes are dynamic and
unlikely to be coherent at any given point in time. However, these challenges can be
addressed by measurements of mass transport of wind- and water-driven sediment
passing per unit length perpendicular to the erosional force vector for each (Breshears et
al. 2003); both these measures have been shown to be related to wind and water erosion
(i.e., soil loss), respectively (Moss and Walker 1978; Gillette et al. 1997). Simultaneous
measurements of wind and water erosion using this new approach are notably lacking.
Consequently, considerable uncertainty remains not only for relative rates of wind and
183
water erosion under baseline conditions of the current climate but importantly under
projected changing conditions associated with future climate change. For many drylands,
projected climate changes include warmer temperatures and increasing frequency of both
extreme droughts and intense precipitation events (IPCC 2007), which could
differentially affect wind and water erosion and increase the importance of assessing
sequences of such wet-dry extremes.
Methods
Site Characteristics
We measured wind and water erosion at the same location over two years,
estimated as cumulative wind-driven horizontal dust flux and water-driven sediment
transport, respectively (Breshears et al. 2003). The measurements were made in a
semiarid grassland in southern Arizona, an ecosystem type of interest due not only to its
global extent but more importantly to its vulnerability to degrade to an eroding shrubdominated system, which is a major land management concern globally. The site was
located in a semiarid rangeland on the Santa Rita Experimental Range (31°50’’N,
110°50’’W) approximately 50 km south of Tucson, Arizona (McClaran et al. 2002). The
study plots were located on Sandy Loam Upland (SLU) Ecological Sites that occupy
recently deposited Holocene alluvial fan and fan terrace surfaces with ≤8% slopes, sandy
loam soils to about 15-cm depth, and 5-25% gravel at the surface (Breckenfeld and
Robinett 2003). The site was located approximately 1100 m above mean sea level.
Mean annual air temperature at this location is approximately 16ºC, with daily maximum
air temperatures exceeding 35ºC in summer (McClaran et al. 2002).
184
Long-term (1923-2003) annual precipitation near the study plots is approximately
350 mm and is bimodally distributed with more than half of the total annual precipitation
occurring during the summer monsoon (July-September) with drier fall and spring
months separating the wetter winter (January-March) and summer months. The first year
of this study (July 2005 – June 2006) included the unusual conditions of both a 25-year
precipitation event in August followed directly by the driest 9-month (September – May)
period on the >100-year instrumental record, whereas the second year (July 2006 – June
2007) served as a representative baseline year, with relatively normal conditions (e.g.,
mean annual precipitation within 1% of 1971 – 2000 mean). Meteorological data
corresponding to the erosion measurements was collected onsite (Campbell Scientific,
Logan).
Experimental Design
We established 3 study plots (50 x 50 m) in a relatively undisturbed semiarid
grassland. Herbaceous cover prior to disturbance was approximately 60%, with
Lehmann lovegrass (Eragrostis lehmanniana) constituting the majority (>90%) of the
grass cover. Woody plant cover was approximately 10%, with Velvet mesquite
(Prosopis velutina) constituting the majority of the shrub cover.
Laboratory and Field Measurements
Soil samples were collected from four locations within each of the nine plots at
the end of the study period to avoid disturbing the plot surface. Soil cores (5-cm
diameter, 20-cm deep) were collected near each of the plot corners and aggregated into
185
single composite samples for each of the nine study plots. Soil samples were oven-dried
at 60ºC to achieve constant weight. Particle-size distribution was determined using the
hydrometer method. Vegetation cover was measured within each plot using the linepoint intercept method along two 3-m transects that extend the full length of the plots.
Meteorological data, including wind speed and wind direction, was collected onsite
within 100 m of the study plots (Campbell Scientific, Logan, UT).
Each of the study plots were instrumented with a series of Big Spring Number
Eight (BSNE) samplers at five heights aboveground (0.08, 0.17, 0.25, 0.50, and 1.0 m)
and a pair of bordered (3 m by 10 m) and unbordered sediment check dams to estimate
annual rates of wind and water erosion, respectively. Sediment samples were collected
from the dust samplers and check dams every 7 to 14 days and over-dried at 60°C to
achieve constant weight. Dust samples were collected by rinsing the BSNEs with DI
water and collecting the rinsate in 20-ml glass vials. Cumulative horizontal dust flux was
calculated by integrating flux measurements from 0 to 1 m above the soil surface using a
simple exponential relationship with height (Gillete et al. 1997).
Results
The two one-year intervals in this study included a “baseline” year, for which
annual precipitation was near normal (~350 mm), and a “global-change-type” year, for
which annual precipitation was about 50% below normal and was characterized by wetdry extremes including an intense 25-year precipitation event followed directly by the
driest 9-month period on the >100 year instrumental record (Fig. 1a). Under the globalchange-type conditions, annual rates of water-driven sediment transport decreased by
186
more than 80% relative to baseline conditions (e.g., mean annual precipitation within 1%
of 1971 – 2000 mean) despite the large 25-year precipitation event (Fig. 1b). Surface soil
moisture conditions—reflecting an integrated response to precipitation, temperature,
atmospheric humidity and wind speed and which have direct effects on soil erodibility by
wind (Ravi et al. 2006)—were significantly drier during much of the global-change-type
year (Fig. 2a, b). Under the global-change-type conditions, annual rates of wind-driven
horizontal dust flux increased by 60% relative to baseline conditions (Fig. 2c).
Combining the results for both wind and water erosion revealed that during the
global-change-type year, wind-driven dust flux exceeded water-driven sediment transport
by a factor of ~40, representing more than an order-of-magnitude increase over the
baseline year, for which wind erosion exceeded water erosion by a factor of only ~3 (Fig.
3a). This jump in the relative rates of wind-to-water erosion is a non-intuitive result
given that the study period included the large 25-year precipitation event (Fig. 3b).
Discussion
Although the duration of our study precludes robustly assessing future trends in
erosion by wind and water, nonetheless, the simultaneous measurements of both wind
and water erosion that spanned both types of projected extremes for many drylands—
increased drought and more intense precipitation events—provide a glimpse into changes
in erosion that might occur under projected climate change. Several caveats apply. The
relative rates of wind and water erosion should vary with site conditions (e.g., soil
texture, percent slope), as well as climate. Under the time span of our study the effects of
variation in vegetation cover were minimized, but over longer time frames spanning
187
multiple years with wet-dry extremes, the relative rates of wind and water erosion would
likely be further altered due to climate-induced changes in vegetation. Notably,
vegetation cover during this study, which was dominated by perennial grasses, was
similar in both years and ranged between 60% and 80% in each year. Therefore, drier
soil conditions associated with the extreme drought appear to have driven the substantial
increase in dust emissions because differences in other parameters that affect wind
erosion, including wind speed and vegetation cover, were relatively small over the
duration of this study. In addition, because wind speed and vegetation cover were
relatively constant between years, our results are likely relatively insensitive to the
sequence of years (the year with wet dry extremes proceeded the baseline year). Our
results for baseline conditions, as well as those for global-change-type conditions, are
also notable in that measureable amounts of wind erosion occurred over every one-to
two-week sampling interval throughout the duration of this study, highlighting that even
though wind erosion is affected by periods with large dust events, it is a much more
continuous process than water erosion. Though we have used measurements of soil
transport as a proxy for erosion (Moss and Walker 1978; Gillette et al. 1997; Breshears et
al. 2003) and more information is needed to better quantify how these two types of fluxes
are related, the potential large amplification in dust emissions that we quantified could
have important implications for land surface-atmosphere interactions under global change
conditions.
The more than an order-of-magnitude jump in dominance of wind erosion over
water erosion reveals the potential for wind-driven sediment transport to increase relative
188
to water-driven sediment transport under changes analogous to those projected to occur
for the southwestern USA and many other drylands. These results may be more broadly
applicable to other drylands. If so, such a change could trigger a substantial shift of land
surface-atmosphere interactions for drylands. Further, an increase in atmospheric
aerosols due to amplified dust emissions from dryland regions has the potential to exert
widespread influence on global climate even though uncertainty remains with respect to
the direction and magnitude of change (IPCC 2007). Because wind erosion is generally
less in grasslands than in other dryland ecosystems (Breshears et al. 2003; Breshears et al.
2009), such amplifications in dust emissions relative to water-driven sediment production
may be even greater in other dryland ecosystems, particularly shrublands and highly
degraded or barren sites. Our results are of particular significance given that the
southwestern United States is projected to transition within years to decades to a more
arid climate comparable to that of the 1930s Dust Bowl (Seager et al. 2007) and that
many other dryland regions worldwide are also likely to experience increased aridity in
the near future due to reduced precipitation and increased temperatures (IPCC 2007).
Dust emissions are highly sensitive to changes in soil moisture and atmospheric humidity
(Ravi et al. 2006), as well as the amount and distribution of protective vegetation cover at
the soil surface (Breshears et al. 2009). Projected increases in drought frequency and
severity for many drylands will likely increase soil susceptibility to wind erosion by
drying out surface soil and resulting in a reduction in the amount of protective vegetation
cover (IPCC 2007), both of which could further amplify dust emissions and land surfaceatmospheric interactions over vast areas worldwide. Dust emissions from dryland areas
189
could be further amplified due to projected increases in land-use intensity that are likely
to occur in the future as a result of increasing human demand for ecosystem goods and
services (MEA 2005). Because erosion depends on land management as well as climate,
our results highlight the need for careful land management to avoid further amplification
of dust emissions such as those that resulted from post-1880 cattle grazing in the western
United States (Neff et al. 2008).
More generally, wind and water erosion processes are recognized as collectively
influencing land surface dynamics (Aguiar and Sala 1999) but have most often been
studied in isolation of one another (Field et al. 2009), or relative rates have only been
inferred from soil pedology and biogeochemistry rather than by direct measurement.
Direct measurements are particularly relevant for shorter time scales of months to years
associated with land management. The importance of considering wind and water
erosion as interrelated processes has arguably been overlooked, yet is fundamental and
analogous to the way in which the C:N ratio in soils must be explicitly considered for
effective land management, rather than considering C or N in isolation of one another. In
short, our results from simultaneous measurements of both wind and water erosion
conceptually suggest that under projected climate changes for many drylands, wind
erosion may substantially increase relative to water erosion, resulting in a significant shift
to a dustier future under global-change-type erosion.
Acknowledgements
We thank Chris McDonald and Guy McPherson for setting up the experimental design
and Travis Huxman, Darin Law, and Lisa Graumlich for comments. This study was
190
supported by the Arizona Agricultural Experiment Station (DDB), the U.S. Department
of Agriculture Cooperative State Research, Education, and Extension Service (JPF, DDB;
CSREES 2005-38420-15809), the National Science Foundation (DDB, JPF; NSF-DEB
0816162), and the Department of Energy (JJW; DE-AC52-06NA25396).
References
Aguiar MR and Sala OE. 1999. Patch structure, dynamics, and implications for the
functioning of arid ecosystems. Trends Ecol Evol 14: 273-277.
Breckenfeld DJ and Robinett D. 2003. Soil and ecological sites on the Santa Rita
Experimental Range. In: McClaran MP, Ffolliott PF, Edminster CB (tech. coords.),
Santa Rita Experimental Range: 100 years (1903-2003) of Accomplishments and
Contributions; Conference Proceedings. US Department of Agriculture, Forest
Service, Rocky Mountain Research Station Proceedings, Tucson, AZ, RMRS-P-30,
October 30-November 1, pp. 157-165.
Breshears DD, Whicker JJ, Johansen MP, Pinder III JE. 2003. Wind and water erosion
and transport in semi-arid shrubland, grassland and forest ecosystems: quantifying
dominance of horizontal wind-drive transport. Earth Surf Proc Land 28: 1189-1209.
Breshears DD, Whicker JJ, Zou CB, et al. 2009. A conceptual framework for dryland
aeolian sediment transport along the grassland–forest continuum: effects of woody
plant canopy cover and disturbance. Geomorph 105: 28-38.
Field JP, Breshears DD, and Whicker JJ. 2009. Toward a more holistic perspective of soil
erosion: why aeolian research needs to explicitly consider fluvial processes and
interactions. Aeolian Research 1: 9-17.
191
Gillette DA, Fryrear DW, Gill TE, et al. 1997. Relation of vertical flux of particles
smaller than 10 μm to aeolian horizontal mass flux at Owens Lake. J Geophys Res
102: 26009-26015.
Goudie AS and Middleton NJ. 2006. Desert dust in the global system. Heidelberg,
Germany: Springer-Verlag.
IPCC. Climate Change 2007: The Physical Science Basis, Contribution from Working
Group 1 to the Third Assessment Report, Intergovernmental Panel for Climate
Change (Cambridge University Press, Cambridge UK, 2007).
McClaran MP. 2003. A Century of vegetation change on the Santa Rita Experimental
Range. In: Santa Rita Experimental Range: 100 Years (1903– 2003) of
accomplishments and contributions. Proceedings RMRS-P-30. U.S. Department of
Agriculture Forest Service, Rocky Mountain Research Station, Fort Collins CO.
Millennium Ecosystem Assessment. 2005. Ecosystems and Human Well-being:
Desertification Synthesis. Washington, DC: World Resources Institute.
Moss AJ and Walker PH. 1978. Particle transport by continental water flows in relation
to erosion, deposition, soils, and human activities. Sedimen. Geol 20: 81-139.
Neff JC, Ballantyne AP, Farmer GL, et al. 2008. Increasing eolian dust deposition in the
western United States linked to human activity. Nature - Geosciences
doi:10.1038/ngeo133.
Oldeman L, Hakkeling R, and Sombroeck W. 1990. World Map of the Status of Human-
Induced Soil Degradation: An Explanatory Note (International Soil Reference and
192
Information Center, Wageningen The Netherlands, and United Nations Environment
Programme, Nairobi Kenya).
Painter TH, Barrett AP, Landry CC, et al. 2007. Impact of disturbed desert soils on
duration of mountain snowcover. Geophys Res Lett 34: L12502,
10.1029/2007GL030208.
Pimentel D, Harvey C, Resosudarmo P, et al. 1995. Environmental and economic costs of
soil erosion and conservation benefits. Science 267: 1117-1123.
Ravi S, D'Odorico P, Herbert B, et al. 2006. Enhancement of wind erosion by fireinduced water repellency. Water Resources Res 42: W11422.
Rosenfeld D, Rudich Y, and Lahav R. 2001. Desert dust suppressing precipitation:
desertification feedback loop. Proc Nat Acad Sci USA 98: 5975-5980.
Seager R, Ting MF, Held I, et al. 2007. Model projections of an imminent transition to a
more arid climate in southwestern North America. Science 316: 1181-84.
Sweet ML. 1999. Interaction between aeolian, fluvial and playa environments in the
Permian Upper Rotliegend Group, UK southern North Sea. Sedimentology 46: 171187.
Trimble SW and Crosson P. 2000. U.S. Soil erosion rates – myth and reality. Science
289: 248-250.
193
Figures
Figure 1. Annual cumulative precipitation (A) relative to water-driven sediment transport
(B) for baseline conditions and global-change-type conditions.
194
Figure 2. Mean daily wind speed at 3 m above ground surface (A) and soil moisture at 2
cm below ground (B) relative to wind-driven horizontal dust flux (C) for baseline
conditions and global-change-type conditions.
195
Figure 3. Cumulative wind-driven dust flux and water-driven sediment transport in an
Arizona semiarid grassland for a period including a 25-year precipitation event (August)
followed by the driest 9-month period (September – May) on the >100-year instrumental
record (A); wind-to-water erosion ratio for baseline conditions and global-change-type
conditions associated with year of study (reference totals indicated by arrows in A) (B).
196
APPENDIX G
HOW GRAZING AND BURNING DIFFERENTIALLY ALTER WIND AND WATER
EROSION: IMPLICATIONS FOR RANGELANDS
Jason P. Field1, David D. Breshears1,2, Jeffrey J. Whicker3 & Chris B. Zou4
1
School of Natural Resources and the Environment, University of Arizona, Tucson, AZ
85721 USA; 2Department of Ecology and Evolutionary Biology, Institute of the
Environment, University of Arizona, Tucson, AZ 85721; 3Environmental Programs, Los
Alamos National Laboratory, Los Alamos, NM 87545; 4Department of Natural Resource
Ecology and Management, Oklahoma State University, Stillwater, OK 74078.
Document Type: In review as of Dec 4, 2009.
Abstract. Rangelands are globally extensive, provide fundamental ecosystem services,
and are tightly coupled human-ecological systems. Long-term rangeland sustainability
depends in large part on soil erosion, which can be driven by both wind and water.
Management for rangeland sustainability centers primarily on alternate grazing and
burning practices. Because few studies directly measure both wind and water erosion,
lacking is an assessment of how both may differentially respond to grazing and burning.
Here we report co-located, simultaneous estimates of both wind- and water-driven
erosion, estimated as sediment transport, in a semiarid grassland over three years for four
land management treatments: control, grazed, burned, and burned+grazed. For all
treatments, cumulatively wind erosion exceeded water erosion due to a combination of
non-trivial ongoing background rates and elevated rates preceding the monsoon season.
For erosion by both wind and water, rates differed consistently by treatment:
burned+grazed > burned >> grazed > control, with effects immediately apparent after
burning but delayed after grazing until the following growing season. Notably, the wind-
197
to-water erosion ratio decreased following burning but increased following grazing. Our
results show how rangeland practices differentially alter wind and water erosion—
differences that could potentially help explain divergence between rangeland
sustainability and degradation.
Introduction
Rangelands are the most abundant type of human-managed ecosystems in the
world and account for about 50% of the terrestrial land surface (Holechek et al. 2001).
Notably, rangelands play a major role in supporting human populations and are estimated
to provide more than $900 billion dollars worth of ecosystems services annually, such as
food production, water regulation, erosion control, and recreation (Costanza et al. 1997,
Millennium Ecosystem Assessment 2005, Havstad et al. 2007, Greiner et al. 2009). The
long-term sustainability of these rangelands depends in large part on factors that
adversely impact soil quality and fertility, in particular soil erosion. Soil erosion has long
been recognized as a serious problem on rangelands (Sampson and Weyle 1918, Bennet
and Chapline 1929, Weaver and Noll 1935) and is considered to be the most severe
consequence of rangeland mismanagement or overgrazing (Holechek et al. 2001). High
rates of soil erosion adversely affects all natural and human-managed ecosystems, and
their effects are often pervasive and long lasting because erosion can reduce the
productivity of the land and ultimately lead to a reduction in the diversity of plants,
animals, and microbes (Larson et al. 1983, Pimentel and Kounang 1998, Lal 2003).
Because soil formation on rangelands is a slow process, often taking thousands of years
to form several inches of topsoil, under mismanaged or overgrazed systems can be lost in
198
a matter of years due to accelerated rates of wind and water erosion (Stevens and Walker
1970, Dregne 1983, Trimble and Mendel 1995, Pillans 1997, Chartier et al. 2009).
Due to the slow rate at which soils forms and their underlying importance in
maintaining land productivity, a critical component of any range management plan is to
maintain adequate vegetation cover to protect the soil surface from wind and water
erosion (Thurow and Taylor 1999, Emmerich and Heitschmidt 2002). Management for
rangeland sustainability centers primarily on alternate grazing and burning practices,
which control to a large extent the amount and distribution of vegetation cover.
Vegetation cover can have a large influence on the absolute and relative magnitudes of
wind and water erosion, in some cases even more so than topography and soil texture
(Breshears et al. 2003, Visser et al. 2004, Field et al. 2009). For example, wind is
thought to be the dominant force controlling soil erosion and redistribution in systems
characterized by spotted vegetation patterns, whereas water is thought to be the dominant
force controlling soil erosion and redistribution in systems characterized by banded
vegetation (Aguiar and Sala, 1999). The redistribution of sediment and other materials,
such as nutrients and organic matter, by both wind and water can greatly alter the surface
characteristics of rangeland soils, which in turn can modify certain hydrological
processes, including water infiltration rates, water storage capacity, runoff/runon patterns,
and erosion rates (Rostagno and del Valle, 1988, Parsons et al. 1992, Bochet et al. 2000,
Nash et al. 2004). Both wind and water erosion can redistribute essential resources in
these ecosystems, such as nutrient-rich sediment, organic matter, and seeds but unlike
wind, water-driven erosion may also result in the redistribution of surface water and is
199
therefore particularly important in arid and semiarid rangelands, where soil water
availability is often the most critical factor controlling plant productivity and
reproduction (Noy-Meir 1973, Dunkerley 2002, Wilcox et al. 2003). Interactions
between wind and water erosion and biotic processes can lead to increased resource
heterogeneity on rangelands and can have important consequences on the structure and
composition of vegetation within these human-managed ecosystems (Schlesinger et al.
1990).
Interactions between erosional processes and vegetation dynamics can lead to
rapid shifts in the amount and distribution of vegetation cover and can ultimately result in
the degradation of rangelands and other ecosystems that are susceptible to erosion
(Turnbull et al. 2008, Breshears et al. 2009, Okin et al. 2009). Rangeland management
practices that involve livestock grazing or prescribed burning usually result in at least a
temporary reduction in the amount of protective vegetation cover, and in the case of
grazing, some amount of surface disturbance, both of which can greatly increase soil
susceptibility to the erosive forces of wind and water (Belnap et al 1995, Whicker et al.
2002, Ravi et al. 2007, Neff et al. 2008, Field et al. 2009). Surface disturbances, such as
a reduction in the extent or quality biological soil crust following grazing or fire-induced
soil-water repellency in soils, can result in a decrease in the erosion threshold for both
wind- and water-driven sediment transport (Belnap 1995, Nash et al. 2004, Ravi et al.,
2006). Fire is an important process in rangeland ecosystems and its ecological and
environmental consequences are related to several factors including the timing, severity,
and frequency of fire (DeBano et al. 1998). Fire can affect nutrient loss pathways such as
200
volatilization, ash convection, wind erosion, runoff, and leaching of fire-released
nutrients (Schoch and Binkley 1986). In addition, feedbacks between wind and water
erosion following fire can promote the redistribution of soil resources from vegetative
patches to nutrient-depleted interspaces and result in more homogeneous distribution of
vegetation and soil resources (Ravi et al. 2009). Because of its effectiveness, fire is
frequently used as an effective management tool on rangelands to reduce fuel loads,
control exotic and competitive understory species, facilitate seeding of native plant
species, and increase seedling growth and survival (DeBano et al. 1998).
Although soil erosion is widely recognized as an important process on rangelands
and its adverse ecological effects are widely documented (Thurow and Taylor 1999,
Emmerich and Heitschmidt 2002, Holechek et al. 2001), few studies have directly
measure rates of both wind and water erosion, particularly following disturbance
(Breshears et al. 2003, Visser et al. 2004, Field et al. 2009). Assessments of how both
types of erosion may differentially respond to disturbances such as fire and grazing is
largely lacking, precluding to a large part out understanding of the dynamic nature of
wind and water erosion and their potential interactions and consequences. Accurate
assessment of wind and water erosion is critical to managing the health and sustainability
of rangelands (Weltz et al. 2003) because both types of erosion can constitute a
substantial fraction of total erosion in most arid and semiarid ecosystems (Bullard and
Livingstone 2002, Breshears et al. 2003, Bullard and McTainsh 2003, Visser et al. 2004,
Field et al. 2009). Here we evaluate rates of soil erosion under natural field conditions
for disturbed and relatively undisturbed rangelands using a recently developed approach
201
that provides a fundamental means for comparing wind- vs. water-driven transport
(Breshears et al., 2003). We measured rates of annual erosion and sediment transport in a
semiarid grassland because these systems represent a large fraction of the earth’s land
surface and are inherently vulnerable to both types of erosion due to the patchy
distribution of vegetation cover characteristic of these systems (Belnap 1995, Aguiar and
Sala 1999, Stout 2001). Although the environmental and economic importance of wind
and water erosion has been well documented in human-managed ecosystems, most
studies consider only one erosional process (Field et al. 2009); therefore, critical
knowledge gaps remain about their relative magnitudes and potential interactions that
must be addressed to enable accurate and effective assessment of the consequences of
soil erosion on rangelands.
Materials and Methods
Site Characteristics
The site was located in a semiarid rangeland on the Santa Rita Experimental
Range (31°50’’N, 110°50’’W) approximately 50 km south of Tucson, Arizona
(McClaran et al. 2002). The study plots were located on Sandy Loam Upland (SLU)
Ecological Sites that occupy recently deposited Holocene alluvial fan and fan terrace
surfaces with ≤8% slopes, sandy loam soils to about 15-cm depth, and 5-25% gravel at
the surface (Breckenfeld and Robinett 2003). The site was located approximately 1100 m
above mean sea level, which is near the low elevation limit of the desert grassland and
above the desert shrub (McClaran 2003). Mean annual air temperature at this location is
approximately 16ºC, with daily maximum air temperatures exceeding 35ºC in summer
202
(McClaran et al. 2002). Long-term (1923-2003) annual precipitation near the study plots
is approximately 350 mm and is bimodally distributed with more than half of the total
annual precipitation occurring during the summer monsoon (July-September) with drier
fall and spring months separating the wetter winter (January-March) and summer months
(Sheppard et al. 2002).
Experiment Design and Treatments
We established 12 study plots (50 x 50 m) in three replicated blocks in a
relatively undisturbed semiarid grassland. Herbaceous cover prior to disturbance was
approximately 60%, with Lehmann lovegrass (Eragrostis lehmanniana) constituting the
majority (>90%) of the grass cover. Woody plant cover was approximately 10%, with
Velvet mesquite (Prosopis velutina) constituting the majority of the shrub cover. Four of
the 12 study plots were undisturbed and served as a control with the other were randomly
assigned one of the following treatments: prescribed burn (B), livestock grazing (G), and
prescribed burn followed by livestock grazing (BG). The prescribed burn was conducted
on July 29, 2005 and was characterized as a light- to moderate-severity fire. The
livestock grazing treatment was characterized as short-duration, moderate-intensity and
was conducted from August to September 2005 by rotating approximately half a dozen
cattle through each of the 50 x 50 m study plots until nearly all of the herbaceous cover
was removed, which typically took about 7-10 days per plot.
203
Laboratory and Field Measurements
Soil samples were collected from four locations within each of the nine plots at
the end of the study period to avoid disturbing the plot surface. Soil cores (5-cm
diameter, 20-cm deep) were collected near each of the plot corners and aggregated into
single composite samples for each of the nine study plots. Soil samples were oven-dried
at 60ºC to achieve constant weight. Particle-size distribution was determined using the
Bouyoucos' hydrometer method (Bouyoucos 1962). Vegetation cover was measured
within each plot using the line-point intercept method (Herrick et al. 2005) along two 3-m
transects that extend the full length of the plots. Meteorological data, including wind
speed and wind direction, was collected onsite within 100 m of the study plots (Campbell
Scientific, Logan, UT).
Each of the study plots were instrumented with a series of Big Spring Number
Eight (BSNE) samplers at five heights aboveground (0.08, 0.17, 0.25, 0.50, and 1.0 m)
and a pair of bordered (3 m by 10 m) and unbordered sediment check dams to estimate
annual rates of wind and water erosion, respectively. Both the dust samplers and the
sediment check dams are estimated to have a capture efficiency of approximately 90%
(Fryear 1986, Hastings et al. 2005). Sediment samples were collected from the dust
samplers and check dams every 7 to 14 days and over-dried at 60°C to achieve constant
weight. Dust samples were collected by rinsing the BSNEs with DI water and collecting
the rinsate in 20-ml glass vials. Cumulative horizontal dust flux was calculated by
integrating flux measurements from 0 to 1 m above the soil surface using a simple
exponential relationship with height (Gillete et al. 1997, Stout 2001).
204
Data Analysis
We used a one-way ANOVA to test for significant differences (P < 0.05) in the
mean values of wind- and water-driven sediment transport in burned, grazed, burned and
grazed, and undisturbed plots. Analysis of variance was carried out according to the
general linear model procedure of the statistical analysis system and the type III sums of
squares using JMP® 8.0 (SAS, Cary, NY). Results were considered significant at the α
level of 0.05. When significant differences among mean values were detected, either the
Student’s t-test or Tukey’s HSD (Honesty Significant Difference) test were used to
separate means.
Results
Our simultaneous, co-located measurements of wind and water erosion, estimated
as wind-driven horizontal dust flux and water-driven sediment transport, indicate that
both types of erosion contribute substantially to total erosion rates in semiarid rangelands
and grasslands. Water erosion was characterized by sporadic, infrequent precipitation
events that were driven primarily by convective thunderstorms associated with the
monsoon season (July – September; Fig. 1a). Wind erosion, in contrast, was
characterized by more consistent and frequent events that were driven by a combination
of small but persistent background dust emissions associated with average daily wind
speeds and large but less frequent dust events associated with strong wind gusts induced
by diurnal temperature fluctuations, as well as frontal and convective thunderstorms (Fig.
1b). Vegetation cover was reduced to about 20% following the prescribed fire and
remained low relative to unburned plots for approximately two years following the fire,
205
whereas vegetation cover was reduced to a lesser extent and for a shorter duration
following the livestock grazing (Fig. 1c). The reduction in vegetation cover explains to a
large degree the substantial increase in both wind and water erosion that was observed in
the burned plots but not in the grazed only plots. Although the first year following
treatment was characterized by extreme wet/dry events, the lack of vegetation cover in
the burned plots appears to be the dominant factor driving the large increases in wind and
water erosion observed during year 1 of the study because control plots had more
consistent annual rates of wind- and water-driven sediment transport throughout the
three-year study period (Fig. 1d, 1e). For all treatments (i.e., control, graze, burn,
burn+graze), the cumulative amount of wind-driven sediment transport exceeded that of
water due to a combination of non-trivial ongoing background rates and elevated rates
preceding the monsoon season.
Grazing and burning had differential effects on the absolute and relative rates of
wind and water erosion. Although there were no statistically significant increases in rates
of wind or water erosion for the first year following the short-duration, moderateintensity grazing treatment, mean daily wind-driven sediment transport was elevated by
more than 30% relative to the control plots (Fig. 2a, 2b). The prescribed fire significantly
increased rates of wind- and water-driven sediment transport during the first two years
following treatment; however, the effect of fire increased water-driven sediment transport
to a much greater extent than wind-driven sediment transport following fire (Fig. 2c, 2d).
The combination of burning and grazing had a synergistic effect that resulted in a
significant increase in wind-driven sediment transport, but this effect was only observed
206
during the second year of the study and only for the case of wind-driven sediment
transport. Rates of wind-driven sediment transport during the second year of the study
were elevated by nearly 40% relative to control, but similar to the first year, no
statistically significant differences were detected. Burning and grazing had no effect on
wind-driven sediment transport by the third year of the study; however, rates of waterdriven sediment transport were still slightly elevated three years following the fire
although still substantially less than rates of wind erosion during the third year of the
study (Fig. 2e, 2f).
Our simultaneous measurements of both types of erosion suggest that the wind-towater erosion ratio can be significantly altered by range management practices that
utilized grazing and burning. Notably, the wind-to-water erosion ratio decreased
following burning but increased or remained relatively unchanged compared to control
following grazing (Fig. 3). Although the prescribed fire significantly increased rates of
water erosion relative to that of wind, the wind-to-water erosion ratio was still positive
(i.e., wind erosion > water erosion) during the first year following treatment (Fig. 3a),
likely because of the extreme 9-month drought that occurred, which resulted in
accelerated rates of wind erosion relative to water erosion, despite the 25-year
precipitation event that occurred during this period (Field et al. in review). In the second
year of the study, which was characterized by near normal amounts of annual
precipitation, rates of water erosion actually exceed rates of wind erosion in the burned
plots (Fig. 3b), suggesting that disturbance alone might by sufficient to cause a shift the
dominate form of erosion on some semiarid rangelands. Although the magnitudes of the
207
wind-to-water erosion ratios were similar for all treatments three years following
disturbance, significant differences were still observed between the burned and control
plots, indicating possible residual effects of burning on soil erosion up to three years
following light- to moderate-intensity rangeland fires.
Discussion
Differential responses of wind and water erosion to rangeland treatments
Notably, our field-based measurements of both processes indicate that even in
semiarid rangelands with substantial ground cover and seasonally intense thunderstorm
activity, wind-driven transport can be a substantial component of total erosion through
small, persistent events. Further, our results highlight that wind erosion and associated
sediment transport maybe underappreciated in drylands systems and that more generally,
simultaneous measurements of coupled aeolian and fluvial processes are necessary to
improve model predictions of dust flux and sediment transport and their associated
feedbacks on land use and climate change (Visser et al. 2004, Field et al. 2009).
Although the first year following treatment was characterized by wet/dry extremes, the
lack of vegetation cover in the burned plots appears to be the dominant factor driving the
large increases in wind and water erosion that were observed during the first year of the
study. For all treatments, cumulatively wind erosion exceeded water erosion due to a
combination of non-trivial ongoing background rates and elevated rates that preceded
thunderstorm activity associated with the monsoons.
Although rates of wind- and water-driven sediment transport exhibited differential
responses to grazing and burning, overall trends among treatments were consistent
208
between both types of erosion: burned+grazed > burned >> grazed > control, with effects
immediately apparent after burning but delayed after grazing until the following growing
season. Many studies have reported increases in either wind or water erosion following
disturbance such as fire and grazing (Johansen et al. 2001, Field et al. 2005, Ravi 2007,
Goudie 2008) but essentially no field measurement are available for both types of erosion
at the same location following, particularly following disturbance (Breshears et al. 2003,
Visser et al. 2004, Field et al. 2009). Notably, the wind-to-water erosion ratio decreased
following burning but increased following grazing and in years with extreme drought.
For example, the first year of the study was characterized by a 9-month drought, which
increased the wind-to-water ratio from background levels (around ~3 to 4) to more than
40, representing more than an order of magnitude increase (Field et al. in review). Our
results suggest that wind-induced erosion and transport is a substantial component of total
erosion rates in rangelands and that the influence of climate and land management
practices on the absolute and relative magnitudes of wind and water erosion can have
important implications for sustainable management of rangelands and should be consider
as a critical part of any range management plan (Thurow and Taylor 1999, Holechek et
al. 2001, Emmerich and Heitschmidt 2002).
Implications for management and rangeland vegetation-soil dynamics
The long-term sustainability of rangelands depends in large part on maintaining
adequate vegetation cover to protect the soil surface from wind and water erosion
(Thurow and Taylor 1999, Emmerich and Heitschmidt 2002). High rates of soil erosion
adversely affects all natural and human-managed ecosystems and their effects are often
209
pervasive and long lasting because erosion can reduce the productivity of the land and
ultimately lead to vegetation change and a reduction in the diversity of plants (Belnap
1995, Costanza et al. 1997, Pimentel and Kounang 1998, Aguiar and Sala 1999, Neff et
al. 2005). The differential responses of wind and water erosion that were observed in our
study following disturbance, including drought, grazing and burning, suggest that the
type of disturbance plays an important role in determining the relative and absolute
responses of wind and water erosion. Our simultaneous measurements of wind- and
water-driven sediment transport indicate that burning has a disproportionally greater
potential to increase rates of water erosion relative to wind erosion and that grazing can
disproportionally increase rates of wind erosion relative to water erosion (Fig. 4). These
results are consistent with long-term observations of rates of wind and water erosion on
US cropland and rangelands. For example, the dramatic increase in livestock grazing
throughout much of the western US in the late 1800s and early 1900s is believed to be the
primary cause for the 500% increase in dust emissions relative to background (late
Holocene) deposition (Neff et al. 2008). Although rates of wind erosion appear to be
increasing in many areas, particularly in the western US, rates of water erosion appear to
be declining (U.S. Department of Agriculture 1971; 1989, Lee 1990, Pimentel et al.
1995). This trend could partially be explained by widespread decreases in the extent and
frequency of fire across many rangelands and widespread increases in grazing pressure
on these systems (e.g., Millennium Ecosystem Assessment 2005).
In summary, our results suggest that differential responses of wind- and waterdriven sediment transport following disturbance may be partially responsible for
210
facilitating some of the recent, widespread observed changes in vegetation composition
across many rangelands and other human-managed ecosystems worldwide. Notably, the
wind-to-water erosion ratio decreased following burning but increased following grazing.
Our results show how rangeland practices differentially alter wind and water erosion—
differences that could potentially help explain divergence between rangeland
sustainability and degradation. In conclusion, our results highlight the pressing need for
more simultaneous field measurements of wind and water erosion to better assess
rangeland health and the overall environmental and economic impacts of soil erosion.
Acknowledgements
This study was supported by the Arizona Agricultural Experiment Station (DDB), the
U.S. Department of Agriculture Cooperative State Research, Education, and Extension
Service (JPF, DDB; CSREES 2005-38420-15809), the National Science Foundation
(DDB, JPF; NSF-DEB 0816162), and the Department of Energy (JJW; DE-AC5206NA25396).
References
Aguiar, M. R., and O. E. Sala. 1999. Patch structure, dynamics and implications for the
functioning of arid ecosystems. Trends in Ecology and Evolution 14(7): 273-277.
Bennett, H. H., and W.R. Chapline. 1928. Soil erosion: a national menace. USDA
Circulation 33.
211
Bochet, E., J. Poesen, and J. L. Rubio. 2000. Mound development as an interaction of
individual plants with soil, water erosion and sedimentation processes on slopes.
Earth Surface Processes and Landforms 25: 847-867.
Chartier, M. P., C. M. Rostagno, F. A. Roig. 2009. Soil erosion rates in rangelands of
northeastern Patagonia: A dendrogeomorphological analysis using exposed shrub
roots. Geomorphology 106(3-4): 344-351.
Belnap, J. 1995. Surface disturbances: their role in accelerating desertification.
Environmental Monitoring and Assessment 37: 39-57.
Breckenfeld, D. J., and D. Robinett. 2003. Soil and ecological sites on the Santa Rita
Experimental Range. In: McClaran, M. P., Ffolliott, P. F., Edminster, C. B. (tech.
coords.), Santa Rita Experimental Range: 100 years (1903-2003) of
Accomplishments and Contributions; Conference Proceedings. US Department of
Agriculture, Forest Service, Rocky Mountain Research Station Proceedings, Tucson,
AZ, RMRS-P-30, October 30-November 1, pp. 157-165.
Breshears, D. D., J. J. Whicker, M. P. Johansen, and J. E. Pinder III. 2003. Wind and
water erosion and transport in semi-arid shrubland, grassland and forest ecosystems:
quantifying dominance of horizontal wind-drive transport. Earth Surface Processes
and Landforms 28: 1189-1209.
Breshears, D. D., J. J. Whicker, C. B. Zou, J. P. Field, and C. D. Allen. 2009. A
conceptual framework for dryland aeolian sediment transport along the grassland–
forest continuum: effects of woody plant canopy cover and disturbance.
Geomorphology 105: 28-38.
212
Bouyoucos, G. J. 1962. Hydrometer method improved for making particle size analyses
of soils. Agron. J. 54: 464-465.
Bullard. J. E., and I. Livingstone. 2002. Interactions between aeolian and fluvial systems
in dryland environments. Area 34: 8-16.
Bullard, J. E., and G. H. McTainsh. 2003. Aeolian-fluvial interactions in dryland
environments: examples, concepts and Australia case study. Prog. Phys. Geog. 27:
471-501.
Costanza, R., R. d’Arge, R. De Groot, S. Farber, M. Grasso, B. Hannon, K. Limburg, S.
Naeem, R.V. O’Neill, J. Paruelo, R.G. Raskin, P. Sutton, and M. van den Belt. 1997.
The value of the world’s ecosystem services and natural capital. Nature 387(6630):
253-260.
Debano, L. F, Neary, D. G., and Ffolliot, P. F. 1998. Fire’s Effects of Ecosystems. John
Wiley & Sons, Inc. Hoboken, New Jersey.
Dregne, H. E. 1983. Desertification of arid lands. Harwood Academic Publishers, New
York, pp. 1-15.
Emmerich, W. E., and R. K. Heitschmidt. 2002. Drought and grazing: II. Effects on
runoff and water quality. Journal of Range Management 55(3): 229-234.
Field, J. P., D. D. Breshears, and J. J. Whicker. 2009. Toward a more holistic perspective
of soil erosion: why aeolian research needs to explicitly consider fluvial processes
and interactions. Aeolian Research 1(1-2): 9-17.
213
Field, J. P., K. W. Farrish, B. P. Oswald, M. T. Roming, and E. A. Carter. 2005. Forest
site preparation effects on soil and nutrient losses in east Texas. Transactions of the
ASABE 48(2): 861-869.
Fryrear, D. W. 1986. A field dust sampler. Journal of Soil and Water Conservation 41(2):
117–120.
Gillette, D. A., D. W. Fryrear, T. E. Gill, T. Ley, T .A. Cahill, and E. A. Gearhart. 1997.
Relation of vertical flux of particles smaller than 10 µm to total aeolian horizontal
mass flux at Owens Lake. Journal of Geophysics Research 102 (D22): 26009-26015.
Goudie, A. S., 2008. The history and nature of wind erosion in deserts. Annu. Rev. Earth
Planet. Sci. 36, 97-119.
Greiner, R., I. Gordon, and C. Cocklin. 2009. Ecosystem services from tropical savannas:
economic opportunities through payments for environmental services. Rangeland
Journal 31(1): 51-59.
Hastings, B. K., D. D. Breshears, and F. M. Smith. 2005. Spatial variability in rainfall
erosivity versus rainfall depth: Implications for sediment yield. Vadose Zone Journal
4(3): 500-504.
Havstad, K. M., D. P. C. Peters, R. Skaggs, J. Brown, B. Bestelmeyer, E. Fredrickson, J.
Herrick, and J. Wright. 2007. Ecological services to and from rangelands of the
United States. Ecological Economics 64(2): 261-268.
Herrick, J. E., J. W. Van Zee, K. M. Havstad, L. M. Burkett, and W. G. Whitford. 2005.
Monitoring Manual for Grassland, Shrubland and Savanna Ecosystems, Volume I,
Quick Start. The University of Arizona Press, Tucson, Arizona.
214
Holechek, J. L., R. D. Pieper, and C. H. Herbel. 2001. Range Management Principles and
Practices, Fourth Edition. Prentice Hall, New Jersey.
Johansen, M. P., Hakonson, T. E., Breshears, D. D., 2001. Post-fire runoff and erosion
from rainfall simulation: contrasting forests with shrublands and grasslands.
Hydrological Processes15: 2953-2965.
Lal, R., T. M. Sobecki, T. Iivari, and J. M. Kimble. 2003. Soil degradation in the United
States: extent, severity, and trends. Lewis Publication, Boca Raton, Florida.
Lee, L.K. 1990. The dynamics of declining soil-erosion rates. Journal of Soil and Water
Conservation 45(6): 622-624.
Larson, W. E., F. J. Pierce, and R. H. Dowdy. 1983. The threat of soil-erosion to longterm crop production. Science 219(4584): 458-465.
McClaran, M. P., D. L. Angell, and C. Wissler. 2002. Santa Rita Experimental Range
Digital Database: User’s Guide. US Department of Agriculture, Forest Service,
Rocky Mountain Experiment Station General Technical Report RM GTR-100.
McClaran, M. P. 2003. A Century of vegetation change on the Santa Rita Experimental
Range. In: Santa Rita Experimental Range: 100 Years (1903– 2003) of
accomplishments and contributions. Proceedings RMRS-P-30. U.S. Department of
Agriculture Forest Service, Rocky Mountain Research Station, Fort Collins CO.
Millennium Ecosystem Assessment. 2005. Ecosystems and Human Well-Being:
Desertification Synthesis. World Resources Institute, Washington D.C.
215
Nash, M. S., E. Jackson, and W. G. Whitford. 2004. Effects of intense, short-duration
grazing on microtopography in a Chihuahuan Desert grassland. Journal of Arid
Environments 56: 383-393.
Neff, J. C., A. P. Ballantyne, G. L. Famer, N. M. Mahowald, J. L. Conroy, C. C. Landry,
J. T. Overpeck, T.H. Painter, C. R. Lawrence, and R. L. Reynolds. 2008. Increasing
eolian dust deposition in the western United States linked to human activity. Nature Geosciences. doi:10.1038/ngeo133.
Neff, J. C., R. L. Reynolds, J Belnap, and P. Lamothe. 2004. Multi-decadal impacts of
grazing on soil physical and biogeochemical properties in southeast Utah. Ecological
Applications 15: 87-95.
Noy-Meir I. 1973. Desert ecosystems: environment and producers. Annual Reviews of
Ecological Systems 4: 25-51.
Okin, G. S., A. J. Parsons, J. Wainwright, J. E. Herrick, B. T. Bestelmeyer, D. C. Peters,
and E. L. Fredrickson. 2009. Do changes in connectivity explain desertification?
Bioscience 59(3): 237-244.
Parsons, A. J., A. D. Abrahams, and J. R. Simanton. 1992. Microtopography and soilsurface materials on semi-arid piedmont hillslopes, southern Arizona. Journal of Arid
Environments 22: 107-115.
Pillans, B. 1997. Soil development at snail’s pace: evidence from a g Ma soil
chronosequence on basalt in north Quennsland, Australia. Geoderma 80(1-2): 117128.
216
Pimentel, D., C. Harvey, P, Resosudarmo, K. Sinclair, D. Kurz, M. Mcnair, S.Crist, L.
Shpritz, L. Fitton, R. Saffouri, and R.Blair. 1995. Environmental and economic costs
of soil erosion and conservation benefits. Science 267: 1117-1123.
Pimentel, D., and N. Kounang. 1998. Ecology of soil erosion in ecosystems. Ecosystems
1(5): 416-426.
Ravi, S., P. D'Odorico, B. Herbert, T. M. Zobeck, and T. M. Over. 2006. Enhancement of
wind erosion by fire-induced water repellency, Water Resources Research 42:
W11422.
Ravi, S., P. D’Odorico, L. Wang, C. S. White, G. S. Okin, S. A. Macko, and S. L.
Collins. 2009. Post-fire resource redistribution in desert grasslands: A possible
negative feedback on land degradation. Ecosystems 12(3): 434-444.
Ravi, S., P. D'Odorico, T. M. Zobeck, T. M. Over, and S. Collins. 2007. Feedbacks
between fires and wind erosion in heterogeneous arid lands. J. Geophys ResBiogeosciences 112: G04007.
Rostagno, C. M., H. F. del Valle, and L. Videla. 1988. Mounds associated with shrubs in
aridic soils of North-eastern Patagonia: characteristics and probable genesis. Catena
15: 347-359.
Sampson, A. W., and L. H. Weyl. 1918. Range preservation and its relation to erosion
control on Western grazing lands. USDA Bulletin 675.
Schlesinger, W. H., J. F. Reynolds, G. L. Cunningham, L. F. Huenneke, W. M. Jarrell, R.
A. Virginia, W. G. Whitford. 1990. Biological feedbacks in global desertification.
Science, 247: 1043-1048.
217
Schoch, P., and D. Binkley. 1986. Prescribed burning increased nitrogen availability in a
mature loblolly pine stand. Forest Ecology and Management 14: 13-22.
Sheppard, P. R., A. C. Comrie, G. D. Packin, K. Augersbach, and M. K. Hughes. 2002.
The Climate of the US Southwest. Climate Research 21: 219-238.
Stevens, P. R., and T. W. Walker. 1970. Chronosequence concept and soil formation.
Quarterly Review of Biology 45(4): 333.
Thurow, T. L., and C. A. Taylor. 1999. Viewpoint: the role of drought in range
management. Journal of Range Management 52(5): 413-419.
Trimble, S.W., and A. C. Mendel. 1995. The cow as a geomorphic agent – A critical
review. Geomorphology 13(1-4):233-253.
Turnbull, L., J. Wainwright, and R. E. Brazie. 2008. A conceptual framework for
understanding semi-arid land degradation: ecohydrological interactions across
multiple-space and time scales, Ecohydrology 1(1): 23-34.
U.S. Department of Agriculture. 1971. Agriculture and the environment. U.S.
Department of Agriculture, Economic Research Service, Washington DC.
U.S. Department of Agriculture. 1989. The second RCA appraisal: Soil, water, and
related resources on nonfederal land in the United States: Analysis of conditions and
trends. U.S. Department of Agriculture, Washington DC.
Visser, S. M., G. Sterk, and O. Ribolzi. 2004. Techniques for simultaneous
quantification of wind and water erosion in semi-arid regions. Journal of Arid
Environments 59: 699-717.
218
Weaver, J. E. ,and W. M. Noll. 1935. Measurement of runoff and soil erosion by a single
investor. Ecology 16(1):1-13.
Weltz, A. M., G. Dunn, J. Reeder, and G. Frasier. Ecological sustainability of rangelands.
Arid Land Research and Management 17(4):369-388.
Whicker, J. J., D. D. Breshears, P. T. Wasiolek, T. B. Kirchner, R. A. Tavani, D. A.
Schoep, and J. C. Rodgers. 2002. Temporal and spatial variation of episodic wind
erosion in unburned and burned semiarid shrubland. Journal of Environmental
Quality 31: 599-612.
219
Figures
Figure 1. Time series of (a) precipitation intensity, (b) average daily wind speed at 300
cm, (c) percent vegetation cover, (d) water-driven sediment transport, and (e) winddriven sediment transport in a semiarid rangeland following grazing and burning. Error
bars represent the standard error of the mean.
220
Figure 2. Annual rates of wind- and water-driven sediment transport in a semiarid
rangeland following grazing and burning for (a,b) the first, (c,d) second, and (e,f) third
year of treatment. Note y-axis is plotted on log-scale. Error bars represent the standard
error of the mean; means for a given vector (wind or water) for a given year with the
same letter do not differ significantly (P < 0.05).
221
Figure 3. Wind-to-water erosion ratio in a semiarid grassland following grazing and
burning for (a,b) the first, (c,d) second, and (e,f) third year of treatment. Note y-axis is
plotted on log-scale. Error bars represent the standard error of the mean; means with the
same letter do not differ significantly (P < 0.05).
222
Figure 4. Conceptual model grazing and burning impacts on vegetation structure and
resultant wind and water-driven erosion vectors (note: length and thickness of arrows
represent sediment transport capacity).