Environmental Modelling & Software 19 (2004) 141–151
www.elsevier.com/locate/envsoft
A numerical simulation of dust storms in China
Zhenxin Song ∗
The Atmospheric Science Department of Lanzhou University, Lanzhou, 730000, People’s Republic of China
Received 30 September 2002; received in revised form 20 January 2003; accepted 18 February 2003
Abstract
Wind erosion occurs in many arid, semiarid and agricultural areas of the world. The desert areas of China, which occupy
approximately 13% of China’s total surface area, are major sources of Asian dust. The major wind-erosion areas are the sandy
lands in western and northwestern China together with the extensive regions of the Gobi desert in northern and northeastern China,
especially along the basin of the Yellow River. In this paper, dust storms which occurred in China in the spring of 2002 were
simulated using an integrated numerical modeling system.
The purpose of the simulation is to produce quantitative predictions of wind erosion on regional scales. The integrated wind
erosion modeling system used in this study coupled the following three major components: (1) An atmospheric prediction model,
together with a land-surface model; (2) a wind-erosion model and (3) a geographic information database. The atmospheric model
provides the necessary input data for the wind erosion scheme, including wind speed and precipitation. It also provides input data
for the land-surface model that produces predictions for soil moisture. Dust transport and deposition are also considered in the
atmospheric model. The wind-erosion model predicts streamwise saltation and dust emission rate for given atmospheric, soil and
land surface conditions. The geographic information database provides spatially distributed parameters, such as soil type and vegetation coverage, for the atmospheric, land surface and wind erosion models.
Dust storms in China occur mainly in spring and winter, but most frequently in April. In spring, surface soils frozen in the
previous winter become especially loose, creating a favorable condition for wind erosion. As an example, the severe dust storms
of 15–20 March were simulated. The results show the integrated modeling system can simulate the main characteristics of the dust
storms. The system produced estimates of wind erosion intensity and patterns that are in agreement with observations. Such a
system offers the possibility of determining wind erosion patterns on broad scales with high spatial resolution, as well as dust
transport and deposition.
 2003 Elsevier Ltd. All rights reserved.
Keywords: Wind erosion; Dust storm; Numerical simulation; Integrated modeling system
1. Introduction
Wind erosion is a serious environmental problem in
arid and semi-arid regions of China and in many other
parts of the world. Strong wind erosion events, such as
severe dust storms, may threaten human lives and cause
substantial economic damage. The northwestern China
region is one part of the central Asia dust storm area.
The desert areas of China, which occupy approximately
13% of China’s total surface areas, are major sources of
Asian dust. These areas include the temperate arid land
∗
Present address: National Meteorological Centre, Zhong Guancun
South Street 46, Beijing 100081, China. Tel.: +86-10-6840-7469; fax:
+86-10-6840-8584.
E-mail address: [email protected] (Z. Song).
1364-8152/$ - see front matter  2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S1364-8152(03)00116-6
from 75°E to 125°E and from 35°N to 50°N (Liu, 1985).
The major wind erosion areas are sandy lands in western
and northwestern China together with the extensive
regions of the Gobi desert in northern and northeastern
China, especially along the basin of the Yellow River
(Liu, 1985; Walker, 1982). The dust storms occurring in
the north part of China and Mongolia are called East
Asian dust weather. Recently, dust storms occurred frequently in the spring in China, which caused the wide
attention of the public and the government. Wind erosion
is an environmental process influenced by geological and
climatic variations as well as human activities. It occurs
when a soil surface is unprotected by vegetation cover
and sufficiently dry, under such conditions, wind is able
to pick up sand sized particles, which bounce along the
surface and eject more particles, including dust particles.
142
Z. Song / Environmental Modelling & Software 19 (2004) 141–151
Those dust particles, which usually contain most of the
organic matter and nutrients, may be carried a long distance by the wind, notably as dust storm or dust hazes.
It reduces soil productivity and leads to land degradation. Wind erosion causes loss to public utilities. For
instance, dust suspension reduces visibility, sandblasting
destroys young crops, and dust related air pollution
causes a health hazard, etc. Hence, the simulation and
forecast of a dust storm is not only important for long
term sustainable agriculture but also has significant
economic benefits.
Considerable insight has been gained into the physics
of wind erosion since Bagnold published his pioneer
work The Physics of Blown Sand and Desert Dunes in
1941 (Bagnold, 1941). Suspension, saltation, and creep
are the three distinct modes which occur during wind
erosion (Bagnold, 1941). Shao (2000) treats the physics
of wind erosion rigorously from the viewpoint of fluid
dynamics and soil physics. The purpose of developing a
wind erosion modeling system is to produce a quantitative prediction of wind erosion on scales from paddock
to global. The system should have the capacity of modeling the complete wind erosion process, from particle
entrainment through transport to deposition. It is a formidable task because wind erosion is governed by a wide
range of factors involving atmospheric conditions, soil
states and surface properities. A lot of progress on the
simulation of dust weather has made. The first attempt to
combine the information of atmospheric data with landsurface data for wind erosion assessment was made by
Gillette and Hanson (1989) in their investigation of the
spatial and temporal variations of dust production in the
United States. In atmospheric studies, dust emission and
transport have been under research since the late eighties
(e.g. Westphal et al., 1988; Tegen and Fung, 1994,
1995). However, in most of these studies, crude wind
erosion schemes and coarse land surface data were used,
which limited the reliability of the modeling results.
Marticorena and Bergametti (1995); Shao et al. (1996)
and Marticorena et al. (1997) have developed better
wind erosion schemes which account for the impact of
surface properties on sand drift and dust emission. Shao
and Leslie (1997) and Lu and Shao (2001) have
developed and implemented an almost fully integrated
wind erosion modeling and prediction system.
In the spring of 2002, a research group was established
in
CMA
(Chinese
Meteorological
Administration). Members of the group come from NMC
(National Meteorological Centre), NSMC (National Satellite Meteorological Centre), IAP CAS (Institute of
Atmospheric Physics, Chinese Academy of Sciences)
and IGE CAS (Institute of Geography, Chinese Academy of Sciences). The Group used an integrated wind
erosion modeling system developed by Shao and Li
(1999 and Shao and Lu, 2000), land surface data and
GIS data to make real time forecast of dust storms that
occurred in China from March to May in 2002. During
these periods, NMC provided numerical forecasts products on dust weather every day. It is the first real time
forecast of dust weather in China. In this paper, we
report the basic facts on dust simulation and prediction
in China in the spring of 2002. At the same time the
model results are compared with observation images.
2. An integrated wind erosion prediction system
2.1. System structure
The framework of an integrated wind erosion modeling system is as illustrated in Fig. 1. It is composed of
an atmospheric model, a land surface scheme, a wind
erosion scheme, a transport and deposition scheme and
a geographic information database. The atmospheric
model provides input data for other three model components. The land surface scheme simulates energy,
momentum and mass exchanges between the atmosphere, soil and vegetation, but more important in the
context of wind erosion modeling, it produces the soil
moisture as an output. The wind erosion scheme obtains
friction velocity from the atmospheric model, soil moisture from the land surface scheme and other spatially distributed parameters from the GIS database. The wind
erosion scheme predicts streamwise saltation flux and
dust emission rate for different particle-size groups. The
transport and deposition model obtains flow velocity,
turbulence data and precipitation from the atmospheric
model and dust emission rate and particle-size information from the wind erosion scheme. Fig. 1 also illustrates a possible computational procedure, the atmospheric model is first run after initialization for
Fig. 1. The structure of integrated wind erosion modeling system
consisting of an atmospheric prediction model, a land surface model,
wind erosion model, a transport and deposition scheme and a GIS database.
Z. Song / Environmental Modelling & Software 19 (2004) 141–151
atmospheric dynamics and atmospheric physics. This is
followed by running the land surface scheme and wind
erosion scheme. Finally, the calculation of dust transport
and deposition is carried out.
2.2. Weather prediction model
The atmospheric model of the integrated system is a
high resolution limited area weather prediction model
developed at The University of New South Wales by
Leslie and his colleagues (Leslie and Purser, 1991),
referred to as HIRES (High Resolution Limited Area
Model). It is a primitive equation model on a LambertConformal projection and utilizes the s coordinate with
the Arakawa C grid. The equation system used for
numerical weather prediction consists of seven basic
equations for velocity components, the continuity equation, the thermodynamic equation, the moisture equation and the equation of state. As dust transport is also
of concern, the dust concentration equation has been
added to the equation system. The simulation area is
30°E, 5°N to 180°E, 65°N with spatial resolution of 50
km. The area of data analysis is 72°E, 5°N to 148°E,
53°N. The atmospheric data required for HIRES initialisation and boundary conditions are derived from the
T213-GCM of the China Meteorological Administration.
In the vertical, the atmosphere is divided into 16 layers.
An advanced soil moisture parameterization scheme has
been linked (Irannejad and Shao, 1998).
ctotal ⫽
冘
143
N
ci
i=1
. The transport model predicts atmospheric dust concentration by solving a continuity equation of dust written
in the form of Eqs. (1) and (2). Writing the equations to
a s coordinate, produces
앫
∂
∂psc(d) ∂psuc(d) ∂psvc(d)
⫹
⫹
⫹ c(d)(pss
∂t
∂x
∂y
∂s
∂c(d) / r
∂c(d) / r
∂
∂
⫹ ps Kphr
(1)
⫹ grwt) ⫽ ps Kphr
∂x
∂x
∂y
∂y
⫹
g2 ∂
∂c(d) / r
K r3
ps ∂s ph
∂s
with boundary conditions
앫
∂c(d)
g2
c(d)(pss ⫹ grwt)⫺ Kpzr3
⫽ grF(d) at the surface
ps
∂s
∂c(d) / r
⫽0
∂z
(2)
at the top
where c(d) is the concentration of dust particles of diameter d, wt is the settling velocity of particles (which is a
function of d), and F is the vertical dust flux; u, v, s앫
and ps are wind velocity and surface pressure, respectively, and r is air density. The horizontal dust particle
diffusivity Kph is assumed to be equal in the x and y
directions. The vertical dust particle diffusivity Kpz is
assumed to be a function of the particle diameter d.
2.4. Dust emission model
2.3. Wind erosion model
The wind erosion model comprises three key parameterizations representing: (i) the erosion threshold friction velocity u∗t , (ii) the streamwise sand flux Q, (iii)
the dust emission flux F(i) for N size classes of dust
particles. The modeling of these processes is based,
respectively, on a model of the wind erosion attenuation
by roughness elements, the saltation model of Owen
(1964). The main outputs from the wind erosion model
are threshold velocity u∗t (m/s), horizontal sand flux Q
(of dimensions M L⫺2 T⫺1), and vertical dust flux F
(g/m2s). The vertical dust flux F then become an input
as the dust source term in the dust transport model. In
our simulation and forecast six particle bins are used in
the model. A division of dust particles into different size
groups has been proposed to be dⱕ2 µm, 2 ⬍ dⱕ11
µm, 11 ⬍ dⱕ22 µm, 22 ⬍ dⱕ40 µm, 40 ⬍ dⱕ80
µm and d ⬎ 80 µm. It is assumed that particles suspended in the atmosphere are composed of N particle
size , each with a size di (i=1,…N). The discussion is
limited to a single particle size, where the multi-particle
case can be reproduced by superimposing the single particle situations. If the concentration of the ith particle is
ci = c(di), then the total concentration is
Lu and Shao (1999) have proposed a dust-emission
model which, in contrast to energy-based models, estimates dust emission on the basis of the volume removed
by impacting sand grains as they plough into the soil
surface. Also, in this model, saltation bombardment is
considered to be the main mechanism for dust emission.
In our simulation, dust emission model developed by
Shao (2001) was used. Three mechanism responsible for
dust emission can be identified: (1) direct liftoff of dust
particles by aerodynamic forces; (2) release of dust particles as saltating particles strike the surface causing
abrasion; and (3) disintegration of dust coats on sand
grains and clay aggregates during saltation. The dust
emission rate related to these three mechanisms can be
formally expressed as
F ⫽ Fa ⫹ Fb ⫹ Fc
(3)
where Fa is aerodynamic lift, which is insignificant in
general, because particles lifted by fa (aerodynamic
forces) are weak in normal wind erosion conditions. Fb is
saltation bombardment, which refers to striking particles
overcome fi (inter-particle binding forces) and result in
strong emission. Fc is aggregates disintegration, which
means fine particles exist as aggregates. In weak events,
144
Z. Song / Environmental Modelling & Software 19 (2004) 141–151
they behave as grains. While in strong events, they disintegrate.
2.5. Dry and wet deposition
Dust particles are delivered back to the surface by
both dry and wet deposition. Dry deposition is the dust
flux from the atmosphere to the surface through molecular and turbulent diffusion and gravitational settling,
while wet deposition is the dust transfer to the surface
through precipitation. Dry deposition dust flux, Fd , can
be expressed as
Fd ⫽ ⫺rwd[c(z)⫺c(0)]
(4)
where c(0) and c(z) are, respectively, dust concentration
at the surface and at the reference level z and wd is the
dry-deposition velocity. Raupach (1991) have proposed
a single-layer dry-deposition model which is less
demanding on data and parameterizations. In this model,
the dry-deposition velocity is treated as a bulk singlelayer conductance made up of three components acting
in parallel
wd ⫽ ⫺wt ⫹ gbb ⫹ gbm
(5)
where gbb is molecular conductance and gbm is impaction
conductance. Wet deposition is not considered in this
work.
2.6. Input GIS (Geographic Information System) data
and model output
The stationary land surface parameters required for
the model are: the soil type index, vegetation type index,
vegetation height, leaf area index (LAI) and land use
status index. We needed to handle the GIS data before
these data were used in the model system. There are
three steps in the pre-processor. The first step is to analyze original GIS data. Most of the GIS parameters are
regrouped into three categories: water, erodible and nonerodible. For a water surface, the parameter values will
be set as 0; if the soil index is equal to those non-erodible
soil indices, the soil index will be re-set to 999. In other
cases, the parameter value is kept as the original. The
second step is to calculate mapping enlargement factor
R. Without considering the effect of soil moisture, threshold friction velocity can be treated as a stationary parameter and pre-calculated if monthly or even short time
prediction is of interest. For each GIS grid, the enlargement factor R is calculated. Finally the soil type index,
enlargement factor, and the fraction of uncovered surface
area need to be input into HIRES. Each Hires grid is
divided into several fractions according to the soil type
index. The sub-areas (GIS grids) with same soil index
are added together regardless of their location within the
HIRES grid.
GIS data is important in wind erosion simulation.
Wind erosion modelling requires spatial-distributed data
for soil and vegetation. The resolution of the GIS data
is 5 km. Soils are normally divided into a number of
soil classes. In our simulation, soils are divided into 30
primary classes and many secondary classes. Although
the classification may not be directly useful for wind
erosion modelling, it provides the basic for further
manipulation. Among the 30 soil classes, 10 are nonerodible soil (7 stabilized soils plus rocks, peats or saline
lakes). The rest of the soil classes can be regrouped into
11 USDA soil-texture classes, according to the descriptive information or to the particle-size analysis for each
primary class. A particle-size distribution, both minimally-dispersed and fully-dispersed can be assigned for
each USDA soil texture class. Vegetation data provide a
range of parameters such as vegetation height, fractional
vegetation cover and leaf-area index. From the vegetation database, a reasonable estimate can be made of
quantities such as vegetation height and vegetation-cover
fraction. The estimate of leaf-area index can draw on the
remotely-sensed NDVI (Normalized Difference Vegetation Index) data. We used the remotely-sensed NDVI
data from March to May in 2002 in our simulation
experiments.
All the variables including 3D variables (such as
wind, pressure, and dust concentration fields) and 2D
surface variables (such as vertical dust flux F, soil moisture w) are stored as binary access format. The basic data
unit is a 2D horizontal plane which is ‘sliced’ for every
vertical level. For a 2D variable, only one unit is used.
For a 3D variable, sixteen (corresponding to the number
of vertical levels used in this study) units are necessary.
For short time prediction, output is written in an hourly
interval; for long time simulation, output is written in 6hourly interval. In this study, we make a 72 h prediction
and 3-hourly interval is used.
3. Results of the simulation
There are obvious difficulties in quantitative winderosion modeling, as both dust-emission rate and streamwise saltation flux are sensitive to input data, such as
soil moisture and frontal-area index, which are difficult
to determine accurately. Nevertheless, wind erosion
models developed recently have produced estimates of
wind erosion intensity and patterns which are reasonable
agreement with observations (Marticorena and Bergametti, 1995; Shao et al., 1996; Shao and Leslie, 1997
and Lu and Shao, 2001). The integrated wind erosion
modeling system is nested with T213 global model in
NMC/CMA, which provides the initial and boundary
data for the integrated wind erosion system. In the
experiment in spring of 2002, we developed the pre-process program to read T213 forecast database and transform them from isobar level to the atmospheric vertical
Z. Song / Environmental Modelling & Software 19 (2004) 141–151
level. In dry climatic conditions, wind erosion risk in
spring in China is high. As a consequence, a considerable proportion of land surface had little protective
annual vegetation and the land surface was susceptible
to wind erosion.
In this paper, the integrated wind erosion modeling
system described above was applied to simulate the dust
storms during March, April and May, 2002 over China.
The integrated wind erosion prediction system simulated
the total process of the dust storm and forecasted many
variables of the atmosphere, soil and wind erosion,
which can be used to describe the synoptic system, the
distribution and strength of the dust sources, the concentration, transport and deposition of dust storms.
Throughout the simulation, Beijing local time is used. It
is 8 hours before Coordinated Universal Time (UTC)
(Greenwich Mean Time).
3.1. Dust sources
145
3.2. The content of dust in the atmosphere
The physical variable used to describe the content of
dust-sized particles in the atmosphere is written as C,
which is the dust mass content per volume. The dimension of C is [ML⫺3]. The variation of C is determined
by mass conservation equation, and is influenced by
advection, diffusion, wind erosion and deposition processes. All the physical process are associated with the
size of the particles, which were also divided into several
groups, such as d ⬍ 2 µm, 2ⱕd ⬍ 11 µm and 11ⱕd
⬍ 22 µm etc. For each particle type, the content of dust
in the atmosphere is calculated. Fig. 3 shows the content
of four different particle types in the near surface layer.
As a example of the strong dust weather of 15 March
2002, it is found that particles with a diameter smaller
than 11 µm have a similar distribution pattern. The
maximum content of particles with 11ⱕd ⬍ 22 µm in
the atmosphere exceeds 300 µg/m3.
3.3. The deposition of dust
The physical variable describing the sources of dust
weather was written as F, which is the dust vertical flux
in the land surface, and the dimension of F is
[ML⫺2T⫺1]. The variable F denotes the mass flux of dust
per time and per area. For example, F also can be used
to illustrate how much dust was emitted during one day
and per square kilometer area. The temporal and spatial
variation of F also represents the temporal and spatial
change of dust sources. Fig. 2 gives such an example of
the distribution of dust sources which arouse the dust
storm weather of 15 March 2002. Fig. 2 shows the
locations of dust sources in Mongolia and the deserts of
the north part of China. The integrated prediction system
forecast successfully the dust sources, which mainly corresponded with the distribution of deserts. The Badain
Juran desert and the Loess Plateau region to the west of
Beijing also are major dust sources.
Fig. 2. Dust sources distribution of the strong dust storm weather on
15 March, 2002.
The physical variable of describing the dust deposition
is written as D, which is the vertical flux of dust to the
land surface. D has the same dimension as F, that is,
[ML⫺2 T⫺1]. At the same time D represents the mass
flux of dust particle pre time and area. Fig. 4 shows the
dry deposition from the strong dust events of 15 March
2002. The location and pattern of dry deposition is
focused on the NNW part of China and Mongolia. From
the predicted dry deposition, we find the dust swept
Fig. 3. The dust concentration distribution of strong dust weather on
15 March 2002, C1 represents concentration of particles with diameter
d ⬍ 2 µm; but C2 and C3 represent particles of 2ⱕd ⬍ 11 µm and
11ⱕd ⬍ 22 µm.
146
Z. Song / Environmental Modelling & Software 19 (2004) 141–151
Fig. 4. Dry deposition distribution of the strong dust weather of 15
March 2002.
dust storm events of 14–18 March 2002 over China.
From the figure of s = 0.6 (about 600 hpa), on 14 March
we found that the dust began to occur in the boundary
regions between inner Mongolia and Gansu province and
moved towards the southeast with time. After 11 BST
of 15 March, most parts of the dust storm moved out of
China and influenced Korea. In the afternoon of 15
March, the dust storm strengthened in the original source
area. From the figure of s = 0.998 (about 1000 hpa), the
same situation can also be found. In the near surface,
the dust followed the same process of developing, moving and finally reaching Korea. Fig. 5 clearly shows the
whole pathway of the dust storm. We also believe the
integrated prediction system has the ability to forecast
the path of motion of the dust storm.
3.5. The spatial distribution of dust
through the desert regions of the northwestern and the
northern China and were carried to the region of Korea.
The modeling system successfully forecast the dry deposition of the dust.
3.4. The movement and the path of dust
An integrated wind erosion modeling system offers
the possibility of determining wind erosion patterns on
broad scales with high spatial resolution, as well as dust
transport. At the same time, the movement and the path
of dust under all kinds of weather conditions were also
calculated and analyzed by using the integrated prediction system. In Fig. 5, we present an example of applying
a wind erosion modeling system to the prediction of the
Fig. 5.
The integrated prediction system can be used to analyze the spatial structure of dust concentration. Fig. 6
shows the distribution of dust concentration at different
vertical levels, but at the same time period. Fig. 6(a) is
the distribution of dust concentration at 400 hpa, on 14
BST of 14 March. We can find at 400 hpa, the location
and structure of the concentration of dust storm matched
that of 500 hpa. The values of the dust concentration at
higher levels are lower than the concentrations at lower
levels. Most of the dust lies in the near surface level.
The concentrations of dust decrease as the heights
increase. As heights increase, the location of highest dust
concentration moves towards the east. We also image the
three dimensional distribution of the dust concentration.
The development and movement of the strong windy weather.
Z. Song / Environmental Modelling & Software 19 (2004) 141–151
Fig. 6.
147
The dust concentration distribution at different heights on 14 March 2002.
3.6. Columnar mass of dust
The columnar dust mass in the atmosphere can be
expressed as Mt = 兰Ctdz, and the dimension is [ML⫺2].
Fig. 7 gives us an example of dust load on 15 March
2002 in the vertical direction. The main parts of the dust
load are located in the north and northwest of China.
Furthermore, the location of dust concentration moves
out of China and extends to South Korea, Korea and
Japan. The dust load is important to determine the content of dust in the atmosphere.
4. Comparison with observation
4.1. The improvement of numerical simulation and
prediction results
Fig. 7.
The columnar dust mass in the atmosphere.
Northeast Asian dust storms were active between
March and May 2002 and a severe event occurred on 19
and 20 March 2002. We carried out intensive numerical
experiments using the integrated wind erosion modelling
system and were able to successfully predict all major
dust storm events during the period between March and
May. In our paper, we give an example of prediction of
a dust storm. These predictions are in excellent agreement with the surface station observations, demonstrating the capacity of the integrated modelling system. It
can also show that the spatial and temporal evolution of
entire dust storm episodes are well predicted.
148
Z. Song / Environmental Modelling & Software 19 (2004) 141–151
In the spring of 2002, we made a 72-hour forecast on
East Asian weather every day. During these periods, we
found and resolved, step by step, some problems including the modeling system itself and GIS data. The forecast
results of the integrated modeling system has been
improved dramatically. The top part of Fig. 8 gives us
the results of the dust concentration simulated at early
stage. The bottom part of Fig. 8 shows us the results of
the dust concentration simulated after the GIS data and
model system were revised. The results before revision
have some errors, for example, in the southwest part of
China occurred virtual dust regions. Furthermore, the
area of dust region is larger and displaced towards the
south compared with the revised results. At the early
stage of the experiment, forecasters and the authors
found these problems and we tried to resolve them.
We corrected and revised GIS data because we found
the results of friction velocity and threshold friction
velocity are not correct in some regions. A key variable
to determine in a wind erosion scheme is the threshold
friction velocity, u∗t. Several surface and soil-related factors strongly affect the magnitude of u∗t, including soil
texture, soil moisture, salt concentration, surface crusting
and presence of surface roughness elements, such as
vegetation and pebbles. Some of them may be modified
during a wind erosion event. For example, the particlesize distribution of the topsoil may become coarser as
small particles are transported away from the source and
aerodynamic roughness length may increase as large soil
Fig. 8. The forecast results before revision compared with simulated
results after revision for the strong dust weather of 20 March 2002.
aggregates emerge from the surface. Consequently, u∗t
may also change during the wind erosion process. In old
simulation cases, GIS data is not accurate in some areas
and the simulation results are also affected by GIS data.
4.2. The simulation of strong dust weather—20 March
2002
We forecast all the dust weather occurring from
March to April in 2002 by using the integrated prediction modeling system. The revised system predicted the
strong dust weather processes. In the following is an
example of strong dust weather. From 19–20 March
2002, severe dust storms occurred in Beijing, which
affected people’s lives, traffic and caused substantial
economic damage. Fig. 9(a) shows us the observation
results. The meteorological station observed the dust
weather and recorded the observations. We find from
Fig. 9(a) that at 08 BST on 20 March, the dust weather
had occurred in the south part of Mongolia, inner Mongolia, Xinjiang, Gansu provinces and Beijing. From Fig.
Fig. 9. The observation of ground station (a) and the simulated dust
concentration (b) at 08 BST on 20 March 2002.
Z. Song / Environmental Modelling & Software 19 (2004) 141–151
149
10(a) we also found the dust weather system moving
towards the east and the areas of dust concentration
became larger than the area of dust concentration at the
time of 20 BST.
Given the uncertainties involved in the model, it is
important to verify the simulated results with observations. The predictions can be compared with observation image. Compare Fig. 9(a) with Fig. 9(b), also at
the same time, comparing Fig. 10(a) with Fig. 10(b), we
find the forecast regions of dust concentration are in
good agreement with the observations. It is the first real
time forecast of dust storms in China.
Fig. 11 shows the simulated surface wind field. It can
be seen from Fig. 11(a) that strong wind regions
occurred in Mongolia and in inner Mongolia, furthermore, strong cyclonic circulation was developed at that
time. The center locations of the strong wind areas are
in agreement with the observation. At the same time, a
good agreement is also found between the location and
regions of strong wind and the simulated dust concentration. On 21 March, in inner Mongolar and Korea, peninsula strong winds occurred and also formed a cyclone
system which is in good agreement with the simulated
dust concentration in these two regions. The experiments
from March to April have proven that the integrated system is a successful dust weather prediction system.
Many issues in dust modelling are unsolved, but the
Fig. 11. Simulated wind field in the near surface, (a) 08 BST 20
March 2002 (b) 08 BST 21 March 2002.
Fig. 10. The observation of ground station (a) and the simulated dust
concentration (b) at 20 BST on 20 March 2002.
accurate estimation of dust emission seems to be the key.
Even with the best model currently available, the uncertainties in the modelling of dust events are very large.
Observation data are often unavailable or insufficient for
model validation. Under such circumstances, a synoptic
recorder of dust activities is of particular importance
because it provides a general guidance to numerical
simulation. It is a pity that we did not get any data of
dust concentration and I think we will compare our forecasts of airborne dust concentrations with measured data
in the future. Yet, there exist few observations that are
150
Z. Song / Environmental Modelling & Software 19 (2004) 141–151
dedicated to the quantification of dust events, although
efforts are being made by the China Meteorological
Administration for specialized measurements. Hence,
traditional weather records are to be the best data.
Fig. 12 shows the simulated temporal evolution of
dust concentration section across Beijing occurring on
20 March. It can be found that on 20 March Beijing
suffered a strong dust storm. The value of the dust concentration reached 1 mg/m3 . On 21 March, Beijing also
experienced dust weather, but compared with the dust
storm on 20 March, the intensity was weak. The forecast
results are in agreement with observations.
5. Summary and conclusions
An integrated wind erosion prediction system has
been briefly described with emphasis on the physically
based wind erosion model and its linkage to a detailed
GIS database. A systematic approach has been taken into
modeling wind erosion by using a wind erosion model
and a dust transport model and the coupling these
components with a high-resolution weather prediction
model. The integrated prediction system not only forecasts the emission sources, temporal and spatial structure
distribution of dust, but can also be used for operational
dust weather forecasting every day. The simulated evolution of dust storms is in qualitative agreement with the
observations. Quantitative agreement is not verified for
lack of observational data. The total simulation of the
15–20 March 2002 dust storm events in China compared
reasonably well with observations. In CMA the first real
time forecast of dust weather has been carried out and
also a great success.
In the simulation of dust storm events, the reliability
of the atmospheric forcing data and the availability of
land-surface parameters are two additional constrains
imposed upon wind erosion modeling. Wind erosion
events are often associated with the development of certain synoptic and sub-synoptic severe weather events,
and these types of weather events are often the most
difficult to describe and predict using atmospheric models. We require high-resolution land–surface parameters
for soil texture, soil hydraulic properties, vegetation
characteristics and surface aerodynamic properties. The
resolution of the data used in the modeling system is
still too coarse. Although we have some problems to
overcome, our experiments in the spring of 2002 have
demonstrated that it is not impossible to predict the
occurrence of individual wind erosion events with
reasonable confidence and estimate their intensity to the
correct order of magnitude. It is therefore hopeful that
as various aspects of the modeling system improve, the
simulation and prediction of dust weather will become
satisfactory.
Acknowledgements
Professor Wang Jianjie of the Numerical Meteorological Centre in CMA has been very helpful in providing
computers and work environments. The author acknowledges that the Computational Modeling System
(CEMSYS4) used in this study is provided by Dr. Yaping Shao through a collaborative project between City
university of Hong Kong and the China Meteorological
Administration, and the atmospheric prediction model,
as part of CEMSYS4, was originally developed by Prof.
Lance M. Leslie. Mr. Zhang Shihuang assisted in the
preparation of GIS data.
References
Fig. 12. Simulated temporal evolution of dust concentration section
across Beijing.
Bagnold, R.A., 1941. The Physics of Blown Sand and Desert Dunes.
Methuen, London.
Gillette, D.A., Hanson, K.J., 1989. Spatial and temporal variability of
dust production caused by wind erosion in the United States. J.
Geophys. Res. 94D, 2197–2206.
Irannejad, P., Shao, Y., 1998. Description and validation of the atmosphere land surface interaction scheme (ALSIS) with HAPEX and
Cabauw data. Global Planet. Change 19, 87–114.
Leslie, L., Purser, R., 1991. High-order numerics in a three-dimensional time-split semi-Lagrangian forecast model. Mon. Wea. Rev.
119, 1612–1632.
Liu, T.S., 1985. Loess and Environment. China Ocean Press, Beijing
pp 215.
Z. Song / Environmental Modelling & Software 19 (2004) 141–151
Lu, H., Shao, Y., 1999. A new model for dust emission by saltation
bombardment. J. Geophys. Res. 104, 16827–16842.
Lu, H., Shao, Y., 2001. Toward quantitative prediction of dust storms:
An integrated wind erosion modelling system and its application.
Env. Modelling & Software 16, 233–249.
Marticorena, B., Bergametti, G., 1995. Modeling the atmospheric dust
cycle: 1. Design of a soil-derived dust emission scheme. J. Geophys. Res. 100, 16415–16430.
Marticorena, B., Bergametti, G., Aumont, B., Callot, Y., N’Doume,
C., Legrand, M., 1997. Modeling the atmospheric dust cycle: 2.
Simulation of Saharan dust sources. J.Geophys. Res. 102D (40),
4387–4404.
Owen, R., 1964. Saltation of uniform grains in air. J. Fluid Mech. 20,
225–242.
Raupach, M.R., 1991. Saltation layer vegetation canopies and roughness lengths. Acta Mechanica 1 (Suppl.), 135–144.
Shao, Y., 2001. A model for mineral dust emission. J. Geophys. Res.
106, 20239–20254.
Shao, Y., Lu, H., 2000. A simple expression for wind erosion threshold
friction velocity. J. Geophys. Res. 105, 22437–22443.
151
Shao, Y., Li, A., 1999. Numerical modeling of saltation in atmospheric
surface layer. Boundary-Layer-Meteorol. 91, 199–225.
Shao, Y., Raupach, M.R., Leys, J.F., 1996. A model for predicting
Aeolian sand drift and dust entrainment on scales from paddock to
region. Aust. J. Soil Res. 34, 309–342.
Shao, Y., Leslie, L.M., 1997. Wind erosion prediction over Australian
continent. J. Geophys. Res. 102, 30091–30105.
Shao, Y., 2000. Physics and Modeling of Wind Erosion. Kluwer Academic Publishers, USA.
Tegen, I., Fung, I., 1994. Modeling of mineral dust in the atmosphere:
sources, transport, and optical thickness. J. Geophys. Res. 99,
22897–22914.
Tegen, I., Fung, I., 1995. Contribution to the atmospheric mineral aerosol load from land surface modification. J. Geophys. Res. 100,
18707–18726.
Walker, A.S., 1982. Deserts of China. Am. Sci. 70, 366–376.
Westphal, D.L., Toon, O.B., Carson, T.N., 1988. A case study of mobilization and transport of Saharan dust. J. Atmos. Sci. 45, 2145–
2175.