Physics and modelling of wind erosion

Physics and Modelling of Wind Erosion


ATMOSPHERIC AND OCEANOGRAPHIC SCIENCES LIBRARY
VOLUME 37

Editors
Lawrence A. Mysak, Department of Atmospheric and Oceanographic Sciences,
McGill University, Montreal, Canada
Kevin Hamilton, International Pacific Research Center, University of Hawaii,
Honolulu, HI, U.S.A.
Editorial Advisory Board
L. Bengtsson
A. Berger
J.R. Garratt
G. Geernaert
J. Hansen
M. Hantel
H. Kelder
T.N. Krishnamurti
P. Lemke
P. Malanotte-Rizzoli
D. Randall
J.-L. Redelsperger
A. Robock
S.H. Schneider
G.E. Swaters
J.C. Wyngaard

Max-Planck-Institut für Meteorologie, Hamburg, Germany

Université Catholique, Louvain, Belgium
CSIRO, Aspendale, Victoria, Australia
DMU-FOLU, Roskilde, Denmark
MIT, Cambridge, MA, U.S.A.
Universität Wien, Austria
KNMI (Royal Netherlands Meteorological Institute),
De Bilt, The Netherlands
The Florida State University, Tallahassee, FL, U.S.A.
Alfred-Wegener-Institute for Polar and Marine Research,
Bremerhaven, Germany
MIT, Cambridge, MA, U.S.A.
Colorado State University, Fort Collins, CO, U.S.A.
METEO-FRANCE, Centre National de Recherches
Météorologiques, Toulouse, France
Rutgers University, New Brunswick, NJ, U.S.A.
Stanford University, CA, U.S.A.
University of Alberta, Edmonton, Canada
Pennsylvania State University, University Park, PA, U.S.A.

For other titles published in this seires, go to
w ww.springer.com/series/5669


Physics and Modelling
of Wind Erosion
by

Yaping Shao
University of Cologne, Germany

ABC


Dr. Yaping Shao
University of Cologne
Germany


ISBN 978-1-4020-8894-0

e-ISBN 978-1-4020-8895-7

Library of Congress Control Number: 2008932207
All Rights Reserved
c 2008 Springer Science + Business Media B.V.
No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by
any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written
permission from the Publisher, with the exception of any material supplied specifically for the purpose
of being entered and executed on a computer system, for exclusive use by the purchaser of the work.
Printed on acid-free paper
9 8 7 6 5 4 3 2 1
springer.com


Preface 1

Wind erosion occurs in many arid, semiarid and agricultural areas of the
world. It is an environmental process influenced by geological and climatic
variations as well as human activities. In general, wind erosion leads to land
degradation in agricultural areas and has a negative impact on air quality.

Dust emission generated by wind erosion is the largest source of aerosols which
directly or indirectly influence the atmospheric radiation balance and hence
global climatic variations. Strong wind-erosion events, such as severe dust
storms, may threaten human lives and cause substantial economic damage.
The physics of wind erosion is complex, as it involves atmospheric, soil
and land-surface processes. The research on wind erosion is multidisciplinary,
covering meteorology, fluid dynamics, soil physics, colloidal science, surface
soil hydrology, ecology, etc. Several excellent books have already been written
about the topic, for instance, by Bagnold (1941, The Physics of Blown Sand
and Desert Dunes), Greeley and Iversen (1985, Wind as a Geological Process on Earth, Mars, Venus and Titan), Pye (1987, Aeolian Dust and Dust
Deposits), Pye and Tsoar (1990, Aeolian Sand and Sand Dunes). However,
considerable progress has been made in wind-erosion research in recent years
and there is a need to systematically document this progress in a new book.
There are three other reasons which motivated me to write this book. Firstly,
in most existing books, there is a general lack of rigor in the description of
wind-erosion dynamics; secondly, the emphasis of the existing books appears
to be placed primarily on sand-particle motion, while topics related to the
modelling of dust entrainment, transport and deposition have not been presented in great detail and thirdly, the results presented in the existing books
appear to be mainly experimental and lacking in documentation of the computational modelling effort involved.
My intention is to provide a summary of the existing knowledge of wind
erosion and recent progress in that research field. The basic contents of the
book include the physics of particle entrainment, transport and deposition
and the environmental processes that control wind erosion. It is intended to
treat the physics of wind erosion as rigorously as possible, from the viewpoint
v


vi

Preface 1

of fluid dynamics and soil physics. A considerable proportion of the book
is devoted to the computational modelling of wind erosion. I hope that this
book can be used as a reference point for both wind-erosion researchers and
postgraduate students. My basic consideration is that wind erosion can only
be understood from a multidisciplinary viewpoint and the computational
modelling of wind erosion should focus on the development of integrated simulation systems. Such a system should tightly couple dynamic models, such
as atmospheric prediction models and wind-erosion schemes, with real data
that characterises soil and surface conditions. In the introductory chapter of
the book, this basic concept is reiterated, while in Chapter 9 examples of the
advocated modelling approach are given. Chapter 2 provides a summary of
wind-erosion climatology in the world and selected regions. Chapters 3 and
4 are devoted to the description of atmospheric modelling and land-surface
modelling, as these are the prerequisite for the modelling of wind erosion.
Chapter 5 is a description of the basic aspects of wind-erosion theory, while
Chapters 6, 7 and 8 are dedicated to the entrainment, transport and deposition of sand and dust particles. In Chapter 9, the integrated wind-erosion
modelling system and the data requirement are described. The concluding
remarks are given in Chapter 12.
Cologne, Germany

Yaping Shao
November 1999


Preface 2

Since the publication of the first edition of this book in 1999, much progress
has been made in the field of wind-erosion research, especially on dust. This
is mainly due to the strong interests in understanding the impacts of mineral
aerosol on climate change and the role of dust in bio-geochemistry. In this

edition, I have updated the contents of the book to reflect the new developments and corrected the mistakes known to me in the first edition. I have also
improved the text and the illustrations.
Many colleagues have helped with the preparation of this edition. In
particular, I wish to thank Drs Masao Mikami, Irina Sokolik, Karl-Heinz
Wyrwoll, Qingcun Zeng, Gongbing Peng, Chaohua Dong, Zhaohui Lin,
Masaru Chiba, Naoko Seino, Taichu Y. Tanaka, Masahide Ishizuka, Eunjoo
Jung and Youngsin Chun for their support. I also wish to thank Ms. Dagmar
Jansen for her careful proofreading of the manuscript and Ms. Martina Klose
for helping with the manuscript preparation using LaTeX.
Cologne, Germany

Yaping Shao
March 2008

vii


Acknowledgements

About 10 years ago, Dr. M. R. Raupach introduced me to the research of
wind erosion. I have ever since maintained a strong interest in this field.
During these years, I came to know many colleagues, including Professor
L. M. Leslie, Dr. J. F. Leys, Dr. G. H. McTainsh, Mr. P. A. Findlater,
Professor W. G. Nickling, Dr. D. A. Gillette, Professor H. Nagashima,
Dr. B. Marticorena, Dr. G. Bergametti and Dr. I. Tegen among many
others, who helped me to develop a understanding of the topics presented
in this book. I am grateful to them for the valuable discussions and arguments during the years and to many of them for providing me with their
research results for inclusion in this book. In the wind-erosion research community, there prevails truly a collaborative spirit. The development of the
integrated wind-erosion modelling system described in Chapter 9 has been
a team effort, and I acknowledge explicitly the significant contributions to

the project made by my colleagues and friends, especially, Dr. H. Lu, Dr.
P. Irannejad, Dr. R. K. Munro, Dr. C. Werner and Mr. R. Morison. The assistance of Dr. P. Irannejad and Mr. H. X. Zhuang in preparing the graphs
and the manuscript has been very helpful. The painstaking final corrections
by Dr. R. A. Byron-Scott have resulted in improvements to a text which
has been written uncomfortably in my second language. Several chapters of
the book were drafted during my stay at the Institute for Geophysics and
Meteorology, University of Cologne, in 1999 when I was an Alexander von
Humboldt Research Fellow. My stay in Germany has been a happy one, and I
thank Professor Dr. M. Kerschgens and the Humboldt Foundation for making
that possible. My thanks also go to Dr. M. de Jong from Kluwer Academic
Publishers for her enthusiastic and patient approach toward publishing this
book. Finally, I would like to take this opportunity to express my gratitude
to Professor P. Schwerdtfeger, Dr. J. M. Hacker and Dr. T. H. Chen for their
continuous encouragements throughout my scientific career.

ix


Contents

Preface 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v

Preface 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
1

Wind Erosion and Wind-Erosion Research . . . . . . . . . . . . . . . . .
1.1 Wind-Erosion Phenomenon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1.2 Wind-Erosion Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3 Integrated Wind-Erosion Modelling . . . . . . . . . . . . . . . . . . . . . . . .

1
1
7
9

2

Wind-Erosion Climatology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1 Climatic Background for Wind Erosion . . . . . . . . . . . . . . . . . . . . .
2.2 Geographic Background for Wind Erosion . . . . . . . . . . . . . . . . . .
2.3 Atmospheric Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.1 Monsoon Winds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.2 Cyclones and Frontal Systems . . . . . . . . . . . . . . . . . . . . . . .
2.3.3 Squall Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4 Global Wind-Erosion Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5 Major Wind-Erosion Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.1 Dust Weather Records and Satellite Remote
Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.2 North Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.3 The Middle East . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.4 Central Asia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.5 Southwest Asia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.6 Northeast Asia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.7 The United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.8 Australia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

13

13
18
20
21
23
24
26
29
29
30
34
36
37
39
44
45

xi


xii

3

Contents

Atmospheric Boundary Layer and Atmospheric
Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1 Atmospheric Boundary Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2 Governing Equations for Atmospheric Boundary-Layer Flows .

3.3 Reynolds Averaging and Turbulent Flux . . . . . . . . . . . . . . . . . . . .
3.4 Equations for Mean Flows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.5 Equations for Turbulent Fluxes and Variances . . . . . . . . . . . . . . .
3.5.1 Turbulent Dust Flux and Dust Concentration
Variance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.5.2 Turbulent Kinetic Energy . . . . . . . . . . . . . . . . . . . . . . . . . .
3.5.3 Features of Different Atmospheric Boundary Layers . . . .
3.6 Surface Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.6.1 Flux-Gradient Relationship . . . . . . . . . . . . . . . . . . . . . . . . .
3.6.2 Friction Velocity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.6.3 Logarithmic Wind Profile and Roughness Length . . . . . .
3.6.4 Stability Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.7 Similarity Theories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.7.1 Monin–Obukhov Similarity Theory . . . . . . . . . . . . . . . . . .
3.7.2 Mixed–Layer Similarity Theory . . . . . . . . . . . . . . . . . . . . .
3.8 Turbulent Flow Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.9 Meso-scale, Regional and Global Atmospheric Models . . . . . . . .

49
49
52
56
59
60
60
61
63
67
67
68

71
72
74
75
78
79
85

4

Land-Surface Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
4.1 General Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
4.2 Surface Energy Balance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
4.3 Soil Moisture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
4.4 Soil Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
4.5 Calculation of Surface Fluxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
4.6 Land-Surface Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
4.7 Examples of Land-Surface Simulation . . . . . . . . . . . . . . . . . . . . . . 110
4.8 Treatment of Heterogeneous Surfaces . . . . . . . . . . . . . . . . . . . . . . 112

5

Basic Aspects of Wind Erosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
5.1 Soil-Particle Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
5.2 Forces on an Airborne Particle . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
5.3 Particle Terminal Velocity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
5.4 Modes of Particle Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
5.5 Threshold Friction Velocity for Sand Particles . . . . . . . . . . . . . . . 134
5.5.1 The Bagnold Scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
5.5.2 The Greeley-Iversen Scheme . . . . . . . . . . . . . . . . . . . . . . . . 138

5.5.3 The Shao–Lu Scheme and the McKenna Neuman
Scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
5.6 Threshold Friction Velocity for Dust Particles . . . . . . . . . . . . . . . 142
5.6.1 Relative Importance of Forces . . . . . . . . . . . . . . . . . . . . . . . 142
5.6.2 Stochastic Nature of Threshold Friction Velocity . . . . . . 145


Contents

xiii

6

The Dynamics and Modelling of Saltation . . . . . . . . . . . . . . . . . 149
6.1 Equations of Particle Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
6.2 Uniform Saltation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
6.3 Non-Uniform Saltation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
6.4 Streamwise Saltation Flux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
6.5 The Bagnold-Owen Saltation Equation . . . . . . . . . . . . . . . . . . . . . 157
6.5.1 The Bagnold Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
6.5.2 The Owen Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
6.6 Other Saltation Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
6.7 The Owen Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
6.7.1 The Formulation of Owen . . . . . . . . . . . . . . . . . . . . . . . . . . 165
6.7.2 The Formulation of Raupach . . . . . . . . . . . . . . . . . . . . . . . . 166
6.7.3 Other Formulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
6.7.4 Profile of Saltation Flux . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
6.8 Independent Saltation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
6.9 Supply-Limited Saltation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
6.10 Evolution of Streamwise Sand Transport with Distance . . . . . . . 176

6.11 Splash Entrainment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
6.11.1 Wind-Tunnel Observations . . . . . . . . . . . . . . . . . . . . . . . . . 178
6.11.2 Numerical Simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
6.12 Numerical Modelling of Saltation . . . . . . . . . . . . . . . . . . . . . . . . . . 186
6.12.1 Simple Flow Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
6.12.2 Large-Eddy Simulation Model . . . . . . . . . . . . . . . . . . . . . . . 187
6.12.3 Particle Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
6.12.4 Aerodynamic Entrainment . . . . . . . . . . . . . . . . . . . . . . . . . . 190
6.12.5 Splash Scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
6.13 Understanding of Saltation from Numerical
Simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
6.13.1 Importance of Splash Entrainment . . . . . . . . . . . . . . . . . . . 194
6.13.2 Particle-Momentum Flux, Saltation Flux
and Roughness Length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
6.14 Saltation in Turbulence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
6.14.1 Intermittency of Saltation . . . . . . . . . . . . . . . . . . . . . . . . . . 201
6.14.2 Aeolian Streamers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
6.14.3 Dynamical Similarity of Saltation . . . . . . . . . . . . . . . . . . . 206

7

Dust Emission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
7.1 Dust Flux and Friction Velocity . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
7.2 Mechanisms for Dust Emission . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
7.3 Aerodynamic Dust Entrainment . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
7.4 Energy-Based Dust-Emission Scheme . . . . . . . . . . . . . . . . . . . . . . 222
7.5 Volume-Removal-Based Dust-Emission Scheme . . . . . . . . . . . . . . 226
7.5.1 Motion of Ploughing Particle and Volume
Removal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
7.5.2 Vertical Dust Flux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232

xiv

Contents

7.6 Comparison of Dust Schemes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
7.7 Spectral Dust-Emission Scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
7.8 Discussions on Dust Schemes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
8

Dust Transport and Deposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
8.1 Evidence of Dust Transport and Deposition . . . . . . . . . . . . . . . . . 247
8.2 Lagrangian Dust-Transport Model . . . . . . . . . . . . . . . . . . . . . . . . . 252
8.3 Eulerian Dust-Transport Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
8.4 Vertical Dust Transport by Diffusion . . . . . . . . . . . . . . . . . . . . . . . 261
8.5 Vertical Dust Transport by Convection . . . . . . . . . . . . . . . . . . . . . 273
8.5.1 Convective Adjustment . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
8.5.2 Cumulus Parameterisation . . . . . . . . . . . . . . . . . . . . . . . . . . 275
8.6 Dry Deposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
8.6.1 Two-Layer Dry-Deposition Model: Smooth
Surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
8.6.2 Two-Layer Dry-Deposition Model:
Vegetation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
8.6.3 Single-Layer Dry-Deposition Model . . . . . . . . . . . . . . . . . . 286
8.7 Wet Deposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
8.7.1 The Theory of Slinn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
8.7.2 Scavenging Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
8.7.3 Scavenging Ratio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299

9

Integrated Wind-Erosion Modelling . . . . . . . . . . . . . . . . . . . . . . . 303
9.1 System Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
9.2 Wind-Erosion Parameterisation Scheme . . . . . . . . . . . . . . . . . . . . 307
9.3 Threshold Friction Velocity for Natural Surfaces . . . . . . . . . . . . . 308
9.3.1 Drag Partition: Approach I . . . . . . . . . . . . . . . . . . . . . . . . . 310
9.3.2 Drag Partition: Approach II . . . . . . . . . . . . . . . . . . . . . . . . 316
9.3.3 Relationship of λ and z0 . . . . . . . . . . . . . . . . . . . . . . . . . . . 317
9.3.4 Double Drag Partition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321
9.3.5 Soil Moisture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
9.3.6 Chemical Binding and Crust . . . . . . . . . . . . . . . . . . . . . . . . 327
9.4 Sand Drift and Dust Emission of Soils with Multiple
Particle Sizes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328
9.5 Climatic Constraints on Dust Emission . . . . . . . . . . . . . . . . . . . . . 333
9.5.1 Erodibility Derived from Synoptic Data . . . . . . . . . . . . . . 333
9.5.2 Erodibility Derived from Satellite Data . . . . . . . . . . . . . . . 336
9.5.3 Wind-Erosion Hot Spots . . . . . . . . . . . . . . . . . . . . . . . . . . . 336
9.6 Land-Surface Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336
9.6.1 Soil Particle-Size Distribution . . . . . . . . . . . . . . . . . . . . . . . 337
9.6.2 Soil-Binding Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
9.6.3 Frontal-Area Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344
9.6.4 Soil Moisture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
9.7 Manipulation of GIS Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347


Contents

xv

9.8 Examples of Integrated Wind-Erosion Modelling . . . . . . . . . . . . . 350
9.8.1 Wind-Erosion Time Series . . . . . . . . . . . . . . . . . . . . . . . . . . 350
9.8.2 Wind-Erosion Predictions on Global, Regional
and Local Scales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351
9.9 Data Assimilation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359
10 Sand Dunes, Dynamics and Modelling . . . . . . . . . . . . . . . . . . . . . 361
10.1 Classification of Sand Dunes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363
10.2 Migration Speed of Transverse Dunes . . . . . . . . . . . . . . . . . . . . . . 370
10.3 Basic Features of Flow over a Sand Dune . . . . . . . . . . . . . . . . . . . 373
10.4 Sand Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377
10.5 Computational Fluid Dynamic Simulation . . . . . . . . . . . . . . . . . . 381
10.5.1 Flow-Model Implementation: Non-hydrostatic Model . . . 382
10.5.2 Flow-Model Implementation: Large-Eddy Model . . . . . . . 384
10.5.3 Computation of Erosion and Deposition Rates . . . . . . . . 385
10.6 Discrete Lattice Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 388
11 Techniques for Wind-Erosion Measurements . . . . . . . . . . . . . . . 391
11.1 Wind-Tunnel Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391
11.2 Sand Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393
11.2.1 Passive Samplers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393
11.2.2 Active Samplers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 396
11.2.3 Impact Sensors: Sensit, Saltiphone and Safire . . . . . . . . . 397
11.2.4 Sand Particle Counter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399
11.3 Dust Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400
11.3.1 High- and Low-Volume Air Samplers . . . . . . . . . . . . . . . . . 400
11.3.2 Optical Particle Counter . . . . . . . . . . . . . . . . . . . . . . . . . . . 402
11.4 Deposition Collectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 403
11.5 Field Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404
11.6 Particle-Size Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407
11.6.1 Dry Sieving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407
11.6.2 Settling Tube and Elutriator . . . . . . . . . . . . . . . . . . . . . . . . 408

11.6.3 Electro-Sensing Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . 410
11.6.4 Laser Granulometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412
11.7 Abrasion Emitter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412
12 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415
12.1 Current Research Topics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415
12.2 Dust Cycle in the Earth System . . . . . . . . . . . . . . . . . . . . . . . . . . . 419
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447


1
Wind Erosion and Wind-Erosion Research

1.1 Wind-Erosion Phenomenon
Wind erosion is a process of wind-forced movement of soil particles. This process has the distinct phases of particle entrainment, transport and deposition
(Fig. 1.1). It is a complex process because it is affected by many factors which
include atmospheric conditions (e.g. wind, precipitation and temperature), soil
properties (e.g. soil texture, composition and aggregation), land-surface characteristics (e.g. topography, moisture, aerodynamic roughness length, vegetation and non-erodible elements) and land-use practice (e.g. farming, grazing
and mining). During a wind-erosion event, these factors interact with each
other and, as erosion progresses, the properties of the eroded surface can be
significantly modified.
In the first instance, wind erosion is a geological and climatic phenomenon
which takes place over long periods of time in deserts and arid regions. Most of
the time, wind-erosion events proceed unnoticed but sometimes they are most
spectacular. Figure 1.2 shows the satellite image of a massive dust storm over
the Atlantic on 26 February 2000. During this event, dust from the Sahara
Desert was lifted to up to 5,000 m above ground and blown off the African continent by an easterly wind. Dust storms of this magnitude have been observed
elsewhere in the world, for instance in the Middle East, China and Australia
(Figs. 1.3, 2.4). Between 15 and 19 April 1998, severe dust storms developed
over the Gobi Desert in Mongolia and China. In the following days, the duststorm front moved across China and, by April 20, the elongated dusty belt

covered a 2,000-km stretch of the east coast of China. The dust clouds were
moving across the Pacific on 23 and 24 April and arrived in North America
by 27 April (Husar et al. 2001).
Wind erosion is the main mechanism for the formation and evolution of
sand seas in the world and the long-range transport of sediments from continent to ocean. Recent studies suggest that the global dust emission amounts to
3,000 Mt yr−1 (estimates vary between 1,000 and 10,000 Mt yr−1 ), and a considerable proportion of this dust is deposited in the ocean (Duce et al. 1991).
Y. Shao, Physics and Modelling of Wind Erosion,
c Springer Science+Business Media B.V. 2008

1


2

1 Wind Erosion and Wind-Erosion Research

Impact on radiation
(Optical thickness, backscatter)

Transport by
wind & clouds

Wet deposition

Condensation nuclei

Convection

Dry deposition

Wind

Turbulent diffusion
Roughness elements
Trapped particles

Soil texture
& surface crust

Dust emission
Saltation

Soil moisture

Fig. 1.1. An illustration of the three phases of wind erosion: entrainment, transport
and deposition. Atmospheric conditions, soil properties, land-surface characteristics
and land-use practice control the erosion process (Modified from Lu, 1999)

Large quantities of minerals and organic matter are carried with the dust
particles and redistributed around the world. The Loess Plateau in China
has a soil depth ranging from 30 to 120 m and its formation is believed to
be largely due to the deposition of wind-transported particles from the Gobi
Desert over many millions of years. On geological time scales, wind erosion
contributes greatly to the global mineral and nutrient circulation and to the
evolution of surface topography.
Particles suspended in the atmosphere, the aerosols, play an important
role in the climate system, as they influence the atmospheric radiation balance directly, through scattering and absorbing various radiation components,
and indirectly, through modifying the optical properties and lifetime of clouds.
Aerosols generated by wind erosion (mineral aerosol) are the most important
aerosols in the atmosphere. The global dust emission of 3,000 Mt yr−1 is comparable with the global sea-salt emission, which is estimated to be around

3,300 Mt yr−1 (Penner et al. 2001). Both estimates for the global dust and
sea-salt emissions have large uncertainties, probably a factor of two. The radiative forcing of tropospheric aerosols on the atmosphere is currently an active
research topic. For climate studies, the key research topics related to wind erosion are the global dust cycle, namely, the emission, transport and deposition
of dust, and the atmospheric processes which involve mineral aerosols, such
as radiation, cloud formation and precipitation.


1.1 Wind-Erosion Phenomenon

3

Fig. 1.2. Satellite image of a dust storm over the Atlantic. Dust from the Sahara was
blown off the African continent by an easterly wind on 26 February 2000 (NOAA,
acknowledgement)

Wind erosion also causes air-quality hazards in populated areas adjacent
to major dust sources. In Beijing, for example, the measured near-surface dust
concentration during severe dust storms has been reported to be as high as
5–10 mg m−3 . Near dust sources, dust concentration can exceed 20 mg m−3
(Yabuki et al. 2002). The northeast Asian dust storm that occurred between
18 and 24 March 2002 caused severe disruptions of social activities in the
northern part of China and Korea (e.g. closure of airports and schools). During the 21–23 October 2002 Australian dust storm, the PM10 concentrations
measured in some coastal cities of Australia (e.g. Brisbane) were close to
1 mg m−3 (Chan et al. 2005). Many contaminants that pose risks to human
health and the environment are found or associated with dust, including metal,
pesticides, dioxins and radionuclide. Thus, to quantify dust sources and to estimate airborne dust concentrations are also important to air-quality studies.
In the recent history, human activities have created profound disturbances
to the natural environment. Excessive clearance of native vegetation, over
grazing and inadequate agricultural practices have resulted in increased frequency and intensity of wind erosion in some parts of the world. Tegen
and Fung (1995) estimated that 20–50% of the global dust load is derived

from human-disturbed soils. This estimate has been recently repudiated by
Prospero et al. (2002), but there is evidence that over human-disturbed surfaces, the rate of wind erosion can be many times that over undisturbed


4

1 Wind Erosion and Wind-Erosion Research

Fig. 1.3. Image of dust storms in the Sahara captured by the space shuttle (NASA,
acknowledgement)

natural surfaces. During the 1930s, for example, decreased precipitation coupled with intensive agricultural activities lead to severe wind erosion in the
Great Plains of the United States, which became known as the dust bowl
of the USA. In the Sahel, drought conditions combined with overpopulation
also resulted in a considerable increase of wind-erosion events. In China, cultivation on the Loess Plateau may have contributed much to the severe dust
storms in northeast Asia. In Australia, some of the recent severe dust storms
have originated in the agricultural areas, where the native vegetation has been
cleared over the past 200 years. Figure 1.4 shows a dust storm over the Murray River near Mildura (Australia), a farming area claimed from the forests
of native Mallee trees.
Wind erosion in agricultural areas leads to land degradation. During an
erosion event, fine soil particles rich in nutrient and organic matters, are carried away by wind over large distances and this results in the loss of soil nutrients. According to Raupach et al. (1994), the February 1982 Melbourne dust
storm generated a loss of 2 million tonnes of topsoil, including 3,400 t of nitrogen and 10 t of phosphorus. The May 1994 dust storms in Australia caused
a soil loss between 10 to 20 million tonnes. The preferential removal of fine
particles by wind erosion leaves coarser and less fertile material behind. Consequently, eroded soils become less productive and have a smaller water-holding
capacity. For land-care purposes, the major tasks of wind-erosion research are


1.1 Wind-Erosion Phenomenon

5

Fig. 1.4. Dust clouds over the Murray River near Mildura (142◦ E, 32◦ S, Australia).
The origin of the dust were the nearby farming areas (J. F. Leys, acknowledgement)

to quantify the risks of wind erosion on different temporal and spatial scales,
to identify the responsible factors and to develop wind-erosion prevention
measures.
Wind erosion involves complex physics that is not yet fully understood.
Its study requires the knowledge of a wide range of disciplines including atmospheric sciences (climatology, synoptics, remote sensing, cloud physics and
atmospheric boundary layers), fluid dynamics, soil physics, surface hydrology, colloidal sciences, and ecology as well as agricultural sciences and land
management.
Almost all physical processes related to wind erosion are particle-size dependent. We often use the term ‘sand’ to describe particles in the size range
between 60 and 2,000 µm and the term ‘dust’ to describe particles smaller
than 60 µm. More precise definitions are given in Chapter 5.
Wind erosion is the consequence of two types of forces at work: the
aerodynamic forces that tend to remove particles from the surface, and
the forces, such as gravity and inter-particle cohesion that resist the removal.


6

1 Wind Erosion and Wind-Erosion Research

The former can be quantified by the friction velocity, u∗ , a measure of wind
shear at the surface, and the latter by the threshold friction velocity, u∗t ,
which defines the minimum friction velocity required for wind erosion to occur. While u∗ is related to atmospheric flow conditions and surface aerodynamic properties, u∗t is related to a range of surface properties, such as soil
texture, soil moisture and vegetation cover. For dry and bare sandy soil surfaces, u∗t is small, and therefore it is not surprising that wind erosion occurs
mainly under such conditions. The balance between u∗ and u∗t is governed
by, and is sensitive to, a number of environmental factors, namely, (1) weather
(wind, temperature, rainfall, etc.); (2) soil type (soil texture, hydraulic properties, etc.); (3) soil state (wetness, compactness, aggregation, etc.); (4) surface

microscopic conditions (aerodynamic roughness length, vegetation coverage,
etc.) and (5) surface macroscopic conditions (landforms, windbreaks, etc.). As
a consequence, wind erosion is strongly variable in space and intermittent in
time. The sporadic nature of wind-erosion events makes the modelling and
prediction of wind erosion, even in a qualitative sense, a formidable task.
For soil particles to become airborne, lift forces associated with wind shear
near the surface or caused by particle impacts must overcome the gravitational
and cohesive forces acting upon them. We call the entrainment of particles
aerodynamic entrainment if it is dominated by aerodynamic forces, and refer to it as splash or bombardment entrainment if it is mainly caused by
the impact of other moving particles. In either situation, the forces involved
in the entrainment process vary strongly from case to case, depending on a
range of factors, but in particular on particle size. Consequently, the dominant mechanism for particle entrainment also depends on particle size. For
sand particles, the entrainment is essentially aerodynamic, while for dust particles the entrainment is primarily due to the impact of saltating sand grains,
a phenomenon known as saltation bombardment (Gillette et al. 1982; Shao
et al. 1993b).
The motion of airborne particles in the atmosphere has two modes, known
as saltation and suspension. Saltation refers to the small hopping motion of
sand-sized particles in the direction of the wind, while suspension refers to
the floating motion of dust-sized particles in the atmosphere.
Through saltation, soil particles are transported in the direction of wind
during an erosion event. Saltation is the mechanism for the evolution of sand
seas on regional scales and the development of sand dunes and fence-line drifts
on local scales. It is an interesting dynamic problem which involves the interactions between the fluid phase, the particulate phase and the surface. As
particles saltate through the atmospheric surface layer with a strong wind
shear, they absorb momentum from the airflow and generate a momentum
transport in the vertical direction. During the impact on the surface, the
saltating particles may splash more particles into the atmosphere. The deposition of saltating particles is of great significance to the evolution of the
land surface. Particles in saltation may be deposited as wind speed reduces
due to changes in atmospheric conditions or changes in surface roughness

1.2 Wind-Erosion Research

7

(e.g. shrubs) or topography (e.g. hills). Saltation is also associated with a
large degree of randomness originating from the lift-off velocities, lift-off angles and turbulent fluctuations in the atmosphere. It is of particular interest in
wind-erosion studies to estimate the intensity of horizontal sand drift related
to saltation and to statistically describe the stochastic features of this mode
of particle motion.
In contrast to sand particles, dust particles, once airborne, can remain
suspended in the atmosphere for some time and be dispersed beyond the atmospheric surface layer by turbulence and transported over large distances.
This process leads to net soil losses from areas of wind erosion. The dispersion
of dust particles is a difficult fluid-dynamic problem, known as heavy-particle
diffusion. Because dust particles have a density more than 2,000 times larger
than the air density, dispersion of dust particles differs from that of neutrallybuoyant fluid parcels. In addition, unless the atmospheric patterns and turbulence properties are adequately pre-specified, it is not possible to predict
the transport of dust particles in the atmosphere with a reasonable accuracy.
Dust particles suspended in the atmosphere are eventually delivered back
to the surface through dry and wet deposition. Dry deposition is the transfer of
airborne dust particles to the surface through turbulent and molecular diffusion and gravitational settling, while wet deposition is the transfer of airborne
dust particles through precipitation. Both processes are of similar importance.
In the first instance, dry deposition is a fluid-dynamic problem dealing with
the diffusion of particles through a thin layer immediately adjacent to the
deposition surface. It also involves many physical processes that are difficult
to quantify, such as the static electrical charge. The difficulty in studying dry
deposition also lies in the lack of detailed knowledge of the flow structure in
the very thin layer immediately adjacent to the deposition surface. On the
other hand, wet deposition involves the process of rain droplets capturing
dust particles suspended in air. The study of wet deposition requires detailed
understanding of raindrop size distribution, particle concentration and the

capture mechanisms.

1.2 Wind-Erosion Research
Wind-erosion research has been progressing along several different streams.
Wind-Tunnel Experiments
Wind-tunnel experiments have been carried out to investigate the physics of
wind erosion, both in laboratory and in field. These studies have focused on the
estimates of threshold friction velocity for different particle sizes, sand-drift
intensity under various wind and surface conditions, dust-emission mechanisms, sand-dune evolution and the impacts of surface roughness elements
and vegetation on wind erosion. These studies have contributed greatly to the
core knowledge on wind erosion.


8

1 Wind Erosion and Wind-Erosion Research

Field Measurements
Field measurements of wind erosion have been carried out on different scales in
many parts of the world. Sand drift has been measured using saltation traps,
and the impacts of land-surface parameters on wind erosion have been studied. Measurements of wind and dust concentration profiles have been made
using anemometers and dust samplers mounted on towers. In addition, networks of air samplers and deposition traps have been setup in recent years for
measuring dust concentration and deposition over large areas. Such networks
are providing valuable data for studying dust movement in the entrainment
phase (tower measurements), the transport phase (tower measurements and
air samplers) and at the deposition phase (deposition traps).
Wind-Erosion Assessment
Assessment of wind erosion on continental scales has been performed by considering wind erosivity and wind erodibility. Wind erosivity describes the
potential of wind to generate erosion, while erodibility describes the potential
of the surface to be eroded. Chepil and Woodruff (1963) proposed to use a

wind-erosion index and developed a model for calculating such indices with
the data of wind speed, precipitation and evaporation. McTainsh et al. (1990)
applied the model of Chepil and Woodruff to determining wind-erosion indices for Australia. With the development of Geographic Information Systems
(GIS), more attention has been paid to soil and land-surface factors. Studies
of wind-erosion climate based on dust-storm records have been carried out by,
for instance, Middleton (1984), Littman (1991), Goudie and Middleton (2001),
Qian et al. (2002) and Kurosaki and Mikami (2005) among many others.
Satellite Remote Sensing
Satellite remote sensing is advantageous in dust-storm monitoring. Sensors
on board of satellites detect the radiances of various surfaces of the Earth
through different spectral channels. These channels are set in correspondence
to the atmospheric radiation windows and water vapour absorption bands.
Various satellite-sensed signals are combined (1) to identify and monitor dust
storms in real time (Carlson, 1979; Ackerman, 1989); (2) to derive land-surface
and atmospheric parameters for dust modelling; (3) to retrieve dust quantities, such as dust load, optical thickness, particle size, etc. (Ackerman, 1997;
Zhang et al. 2006); and (4) to derive long-term dust climatology. For example,
Prospero et al. (2002) have used the NIMBUS 7 Total Ozone Mapping Spectrometer (TOMS) aerosol index over a 13-year period (1980–1992) to examine
the distribution of dust sources on the globe.
Empirical Wind-Erosion Modelling
Empirical wind-erosion models have been under development for some time.
The most widely used is the Wind-Erosion Equation (WEQ) (Woodruff and


1.3 Integrated Wind-Erosion Modelling

9

Siddoway, 1965), an empirical model in which the driving parameters are descriptors of soil type, vegetation, roughness, climate and field length. The
original WEQ used annual averages of these descriptors to estimate annual
average soil loss. For estimates over shorter periods, the WEQ was modified

by Bondy et al. (1980) and Cole et al. (1983). More recent revisions have
led to the Revised Wind-Erosion Equation (RWEQ) which includes input parameters such as planting date, tillage method and amount of residue from
the previous crop; a weather generator is then used to predict future erosion (Comis and Gerrietts, 1994). The empirical nature of the WEQ limits
its transferability from the central Great Plains of the USA, for which it was
originally developed, to other areas of the world. Also, the complex interactions between the variables controlling wind erosion are not fully accounted
for in the WEQ. For this reason, a more process-oriented model called the
Wind-Erosion Prediction System (WEPS) has been developed. The WEPS
includes submodels for weather generation, crop growth, decomposition, soil,
hydrology, tillage and erosion (Hagen, 1991).
Large-Scale Field Experiments
Several large-scale field experiments have been recently carried out, dedicated
to wind erosion, dust storms and dust cycle. The Aeolian Dust Experiment on
Climate Impact (ADEC, Mikami et al. 2006) and the Asian Pacific Regional
Aerosol Characterization Experiment (ACE-Asia, Huebert et al. 2003; Arimoto et al. 2006) are two examples. Networks have also been constructed to
obtain dust observations over large areas. For example, the Aerosol Robotic
Network (AERONET) is a federation of ground-based remote sensing aerosol
networks. AERONET assesses aerosol optical properties and validates satellite
retrievals of these properties. The data include globally distributed observations of spectral aerosol optical depths and precipitable water. The network
has been operating since 1993 and has been carrying out routine measurements at around 150 stations distributed all over the world.

1.3 Integrated Wind-Erosion Modelling
The approach advocated in this book is integrated wind-erosion modelling.
An integrated wind-erosion modelling system enables the simulation and prediction of all aspects of wind erosion, from particle entrainment, transport to
deposition. The aim of such a system is to provide quantitative assessment and
prediction of wind erosion on scales from local to global. To this end, the integrated system needs to be constructed with six basic components: an atmospheric model, a wind-erosion model, a land-surface scheme, a dust-transport
scheme, a data-assimilation scheme and a geographic-information data base.
The atmospheric model provides the data required to drive the winderosion scheme, such as friction velocity, u∗ , wind field for dust advection,


10

1 Wind Erosion and Wind-Erosion Research

turbulence intensity for dust diffusion and deposition, and precipitation for
wet deposition. In addition, the atmospheric model provides the data, such
as radiation, required by the land-surface scheme for modelling the environmental variables, such as soil moisture and vegetation cover, which strongly
influence wind erosion. Most atmospheric models are coupled with radiation
schemes to deal with the impact of mineral aerosols on radiation transfer in
the atmosphere.
The wind-erosion model enables the quantification of the entrainment,
transport and deposition of soil particles of all sizes. For instance, as far as
particle entrainment is concerned, it enables the prediction of (1) the threshold
friction velocity for wind erosion, (2) the rate of sand transport and (3) the
rate of dust emission.
Data assimilation is a technique which combines model and data to achieve
an optimal simulation or prediction of a problem of concern. This technique
has been very successfully applied to atmospheric and oceanic predictions.
Because of the lack of dust measurement data, very little has been done so far
in applying data assimilation to dust modelling. However, dust measurements
are becoming increasingly available. For example, satellites can now provide
continuous dust monitoring over large areas and the developments of inverse
methods are producing quantitative estimates of dust load in an atmospheric
column. Networks of lidar are being established, which provide dust-profile
estimates at a number of locations. Further, stations equipped with dust samplers and radiometers are being set up. We expect that in the near future,
data assimilation will be an important component of integrated wind-erosion
modelling.
Reliable land-surface data is important to wind-erosion modelling. They
are required, because the properties of the land surface control the erosion
threshold friction velocity, the capacity of the soil to release dust and the
partitioning of wind-shear stress acting on non-erodible roughness elements

and the erodible surface. Three categories of parameters can be distinguished.
The first consists of parameters related to soil properties, e.g., soil particlesize distribution and soil-binding strength. The second consists of aerodynamic
parameters related to surface roughness and drag partitioning. The third category consists of parameters which specify the soil thermal and hydraulic
properties. For the purpose of modelling wind erosion on regional to continental scales, these soil and land-surface parameters can be stored as layers
in a geographic information system (GIS).
The first attempt of developing an integrated wind-erosion modelling system was probably made by Gillette and Hanson (1989), who used extensive
atmospheric and land-surface data to determine the spatial and temporal
variations of dust production in the United States. Gillette and Hanson did
not use an atmospheric model and did not consider dust transport and deposition. Earlier versions of dust models, more from the atmospheric perspective, have been developed by Westphal et al. (1988) and Tegen and Fung
(1994, 1995). In these early dust models, rather crude wind-erosion schemes


1.3 Integrated Wind-Erosion Modelling

11

and land-surface data were used. Marticorena and Bergametti (1995), Shao
et al. (1996) and Marticorena et al. (1997) developed physics-based winderosion models and applied them to improve the simulations of wind erosion.
Shao and Leslie (1997) and Lu (1999) developed an almost fully integrated
wind-erosion modelling system which couples a physics-based wind-erosion
scheme with an atmospheric model, a land-surface scheme and a geographicinformation database. They have implemented the system for the prediction of
dust storms in Australia. Since the late 1990s, a number of dust storm models
for global, regional and local dust problems have been developed. Examples of
global dust models include those of Zender et al. (2003), Ginoux et al. (2004)
and Tanaka and Chiba (2006). Examples of regional dust models include the
studies of Nickovic et al. (2001), Liu et al. (2001), Shao et al. (2003) and Uno
et al. (2005). Seino et al. (2005) simulated dust storms in the Tarim Basin
using a meso-scale dust model.
Integrated modelling is a new approach to studying wind erosion. It takes
the advantage of the recent rapid expansion in computing power, developments in atmospheric and land-surface modelling, and the increasing availability of land-surface and remote-sensing data. This approach is a major

step forward in the quantitative prediction of wind erosion, the comprehensive analysis of wind-erosion processes and the identification of the natural
and human factors that affect wind erosion. However, integrated systems are
complex. As will become evident in this book, nearly all wind erosion processes
are sensitive to parameters which cannot be derived with great certainty. For
example, threshold friction velocity is sensitive to soil moisture and vegetation cover. As a consequence, it is difficult to predict wind erosion with
great accuracy. Nevertheless, recent studies have demonstrated that integrated
wind-erosion modelling systems can produce results (sand-drift intensity, dust
emission, dust concentration, etc.) which are comparable in magnitude with
observed data, and the uncertainties embedded in the modelling systems are
comparable with the uncertainties of observations.