Does soil composition have an influence on the sandblasting process ?
Alfaro S.C., J.L. Rajot, S. Lafon, M. Maillé, and A. Gaudichet
LISA, Université de Paris Val de Marne 61, av. du Général de Gaulle, 94010,
Créteil, France
email : alfaro@lisa.univ-paris12.fr
Abstract
It is now well ascertained that fine dust emission by wind erosion is the result of two
processes: saltation and sandblasting. In saltation, loose soil aggregates present at the
soil surface can be set into motion by wind action provided the aerodynamic stress
overcomes the forces that tend to maintain aggregates on the ground, namely, weight
and interparticle forces. Numerous parameters are necessary to describe this process.
Beside wind speed itself, they are mainly parameters describing the soil surface state:
size distribution of loose soil aggregates, soil roughness length (including effects of
non-erodible elements and soil ridging), humidity, and crusting. To a degree, all these
soil characteristics depend on soil texture and composition. For example, after a rain
strong crusts are more liable to form at the surface of fine textured soils than over
coarser textured ones (Rajot et al., in press). Other factors, such as large contents in
carbonates, in iron oxydes, or in organic matter that tend to promote soil cohesion and
favor formation of larger aggregates or even clods, have an effect at the same time on
aggregate mobility and on soil roughness. Thus soil composition has a direct effect on
saltation.
In the sandblasting process, a fraction of the terminal kinetic energy of the saltating
aggregates is used to release fine particles (PM20) either from the aggregates
themselves, or from the surface on which they impact. In the modeling of
sandblasting, binding energies of the PM20 populations are crucial factors that
condition at the same time the magnitude of fine dust emission fluxes and their size
distribution (Alfaro et al., 1997). After combining the saltation and sandblasting
models to form a Dust Production Model (DPM, Alfaro and Gomes, 2001), it has
been showed (Alfaro et al, in press) that dust fluxes measured over a variety of soils
differing in texture and in composition can be retrieved with a single set of binding
energies. This indirect finding that, contrary to saltation, soil composition has no
influence on PM20 binding energies, and hence on sandblasting, would greatly
simplify applicability of the DPM at regional or larger scale. Thus, it is of the utmost
importance to confirm it.
For this purpose, wind erosion simulations have been performed in the LISA wind
tunnel with 3 soils chosen for being quite different in texture as well as in
composition. With each soil, experiments have been carried out at 3 different wind
speeds. In all cases saltation flux was measured in parallel with PM20 concentrations
in 6 size classes (0.3-0.5, 0.5-0.7, 0.7-1, 1-2, 2-5, >5 µm). Results of these
experiments will be presented and analyzed. They confirm that saltation is sensitive
to soils characteristics, but that, relative to saltation intensity, PM20 emissions do not
significantly differ from one soil to the other.
References



Alfaro, S.C., Gaudichet, A., Gomes, L., and M. Maillé, Modeling the size
distribution of a soil aerosol produced by sandblasting, J. Geophys. Res,
102, n° D10, 11,239-11,249, 1997.
Alfaro S.C. and L. Gomes, Modeling mineral aerosol production by wind
erosion : Emission intensities and aerosol distributions in source areas, J.
Geophys. Res, 106, n° D16, 18,075-18,084, 2001.
Rajot J.L., S.C. Alfaro, L. Gomes, and A. Gaudichet, Influence of sandy soil
crusting on horizontal and vertical wind erosion fluxes, Catena, in press

Alfaro S.C., J.L. Rajot, and W. Nickling, , Estimation of PM20 emissions by
wind erosion: Main sources of uncertainties, Geomorphology, in press
Modelling Saharan dust and first results on vertical distribution based on
2.5 y simulations
P. Alpert, P. Kishcha, A. Shtivelman, S.O. Krichak & J.H. Joseph
Department of Geophysics and Planetary Sciences Tel-Aviv University, Tel-Aviv
69978, ISRAEL
Abstract
The model dust prediction and verification will be presented (Alpert et al, 2002,
Tsidulko et al, 2002, Krichak et al, 2002). Next, first results on vertical distribution
will be shown.
The vertical distribution of aerosols is not well known but is required for the better
understanding of their radiative effects. Within the framework of the Israeli-American
MEIDEX project the 2.5-year database from November 2000 up to the present of 48hour dust forecasts at the Tel Aviv University prediction system has been adapted for
the climatological analysis of dust vertical profiles over North Africa and the
Mediterranean. This climatological analysis is the first one based on a relatively large
archive of vertical profiles over the Sahara; the gross amount of profiles is
approximately 107 profiles per year. Our climatological approach, now under way,
yields relevant information about the characteristic features of dust in the region in
question. Month-to-month variations of horizontal distribution of simulated dust
loading are compared with known seasonal cycles of dust activity based on the
TOMS aerosol index.
The distributions of the largest sources of dust in the Sahara, simulated by the model,
are also in agreement with the TOMS measurements. Vertical distributions of dust
reflect differences between the Atlantic and the Mediterranean dust transport. As a
whole, the Mediterranean dust is found to be within a wider range of altitudes,
penetrating rather higher into the troposphere, and is most probably multilayered. On
average, dust over the Atlantic penetrates up to < 5 km while over the Mediterranean
up to < 8 km. These results are consistent with available lidar measurements. The
characteristic feature of dust vertical profiles over the dust source near Lake Chad is
its maximal concentration near the surface. From April to June averaged profiles over
the Chad basin in the Sahara are restricted below the level of ~4.5 km. In the winter
months and in March, dust concentration over the Chad basin is closer to the surface
beneath 1.5 km. Dust also maximizes near the surface over another dust source in
West Africa, which is the major one in summer.
The resulting averaged 3D-distribution of the Saharan dust supports our derived
profiles of tropospheric temperature response to dust based on the ECMWF
reanalysis increments.
References


P. Alpert, S. O. Krichak, M. Tsidulko, H. Shafir and J. H. Joseph, "A dust
prediction system with TOMS initialization", Mon, Wea. Rev., Vol. 130,
No. 9, pp. 2335-2345, 2002.
M. Tsidulko, S. O. Krichak, P. Alpert, O. Kakaliago, G. Kallos and A.
Papadopulous, " Numerical study of a very intensive E. Mediterranean dust
storm; 13-16 March 1998", JGR-Atmospheres, 107(D21), 4581,


doi:10.1029/2001JD001168, 2002.
P. Kishcha, P. Alpert, J. Barkan, I. Kirchner and B. Machenhauer,
"Atmospheric response to Saharan dust deduced from ECMWF reanalysis
(ERA) temperature increments", Tellus, 2002. Tellus 55 B no 4, September
issue.
S. Krichak, M. Tsidulko, P. Alpert, "A study of an INDOEX period with
aerosol transport to the eastern Mediterranean area", J. Geophys. Res.,
107(D21), 4582, doi:10.1029/2001JD001169, 2002.
Variability of Dust Particle Shapes and Aggregation in Asian, North
African, and North American Aerosols
James R. Anderson
Environmental Fluid Dynamics Program, Arizona State University, Tempe, Arizona,
USA janderson@asu.edu
Abstract
Soil dust particles suspended and transported in the troposphere have a considerable
range in shape, the degree and manner in which they are aggregated, and what
materials they are aggregated with. Indeed, shape, size, and aggregation are
commonly linked in that shape increases in complexity with increasing degree of
aggregation, which in turn tends to increase with increasing size. Because particle
shape and the nature of particle aggregation (the "state of mixing") effect light
scattering and absorption, it is important to assess typical states from a variety of
source areas and types.
A simple two-dimensional shape metric for individual particles is circularity:
Circularity = (Perimeter2)/(4pi x Area)
For the work discussed here, this is determined using binary images derived by
segmenting gray-level backscattered electron images of particles acquired in an
automated scanning electron microscope. As particle size decreases and becomes
vanishingly small (for practical purposes in this case, 100 nm diameter), circularity is
constrained to the minimum value of 1.0, the circularity of a sphere. A convenient
measure of complexity is the circularity of particles with diameters of 10 µm on a fit
of plotted circularity vs. average diameter (the best fit is not necessarily linear).
Common values of circularity for 10 µm particles are in the range 2.5-6.0, but a small
fraction of the total can be exceptional particles with circularities of 10.0 or greater.
It would of course be more desirable to describe shapes in three dimensions and if
indeed each particle was a simple object with no aggregation this might be feasible
for at least the larger particles using image analysis techniques on stereo image pairs.
However, because a significant fraction of the particles are aggregates with all their
inherent difficulties (e.g., interior spaces and complex grain junctions), it is not clear
that any attempt to measure three-dimensional shapes would be worth the effort given
current technology.
The most complex shapes observed for soil dust in the work described here have been
in cases of dust mixed with industrial pollution, including soot, flyash, and other nonsoil materials. In such cases, aggregates only a few microns in diameter can have
circularities as high as 6-8. When dust is not mixed with pollution, there seem to be
some common properties among samples from many locations, provided they are
from similar types of sources. For ACE-Asia samples of dust with little intermixed
pollution collected from outflow from China, a typical circularity for 10 µm particles
is about 4, but varies from case to case. In a study of soil dust from unpaved roads
along the US/Mexico border at Douglas, Arizona this value tends to be closer to 5. In
a study of re-entrained dust from freeways in the Phoenix, Arizona metropolitan area,
the action of vehicles rolling over dust particles on concrete roadway surfaces seems
to have partially disaggregated many with the result that a typical 10 µm particle has
a circularity of about 3. However, a case with some of the lowest circularity values
observed is from the PELTI project in the Caribbean, where the intercept at 10 µm is
about 2.5. These particles are compact aggregates of submicron clay and other
silicates and may have been from ephemeral lakes in North Africa; the drying surface
of an ephemeral lake after a storm is of course a very different source type than
unpaved roads, agricultural fields, and disturbed areas undergoing desertification.
However, the similarity in shape distributions of dust aerosols from places as diverse
as the Loess Plateau in China and the Sonoran Desert in North America suggest that
for unpolluted cases, at least, some reasonable assumptions about shape distributions
can often be made for efforts to model dust optical properties.
Raman-lidar profiling of dust optical properties at UV and visible
wavelengths
Albert Ansmann, Detlef Müller, Ina Mattis, Ulla Wandinger
Institute for TroposphericResearch, Permoserstr. 15, 04318 Leipzig, Germany
albert@tropos.de
Abstract
Introduction
Raman lidars operating at 355 and 532-nm wavelengths are very useful for the
vertical profiling of dust optical properties. In contrast to standard lidars they allow
an unambiguous determination of the volume extinction coefficient of dust particles
(Ansmann et al., 1990). In contrast to insitu observations (aircraft) they allow a
profiling under ambient conditions and over long time periods. In contrast to satellite
and ground-based remote sensing vertically resolved information on dust layering is
obtained so that a clear separation between the optical effects caused by particles in
the boundary layer (mixture of dust, marine and anthropogenic particles) and in the
free troposphere (mainly mineral particles) is possible. It is essential to know the
amount of dust in the free troposphere. This fraction can be transported over
thousands of kilometers and thus can affect climate on a hemispheric scale. Profiling
of the extinction coefficient at 355 and 532 nm is useful because Saharan dust seems
to considerably absorb radiation at 355 nm. At 532 nm, light absorption does not play
an important role.
Furthermore, simultaneous profiling of the extinction coefficient and the backscatter
coefficient (180 degree scattering) with Raman lidar provides us with vertically
resolved values of the extinction-to-backscatter ratio (lidar ratio, Ansmann et al.,
1992). This quantity is very interesting for dust studies because it is sensitive to
particle shape. Backscattering by nonspherical dust particles is believed to be a factor
of 2-3 (Mishchenko et al., 1997) or even 3-10 (Kalashnikova et al., 2002) lower than
backscattering by surface-equivalent spheres. Our observations support this
assumption (Mattis et al., 2002; Müller et al., 2003). The lidar ratio is also an input
parameter in the estimation of dust extinction profiles with standard backscatter
lidars. Thus, observed (not simulated) dust lidar ratios are needed for an improved
dust monitoring with standard lidars In near future satellite-based standard lidars may
map the global aerosol distribution in terms of backscattering. A global aerosol-typedependent lidar ratio climatology is then needed to obtain reliable 3-D particle
extinction information. It should be mentioned that the number of aerosol Raman
lidars is growing. About 50 % of the stations of the European Aerosol Research Lidar
Network (EARLINET, Bösenberg et al., 2001) is equipped with Raman channels.
Many Saharan dust outbreaks were observed with EARLINET between 2000 and
2002.
Observations
Here, we report on two major outbreaks that took place in August and October 2001.
We will present results in terms of measured spectrally resolved backscatter and
extinction profiles, lidar-ratio and depolarization-ratio profiles, and dust optical
depth. The observations were made at Leipzig, Germany, which is a site of the
Aerosol Robotic Network (AERONET) as well as of EARLINET (Müller et al.,
2003). During the two outbreaks, about 80% of the dust was found in the free
troposphere. The dust layers were mainly below 6 km height, with traces of dust up to
10 km height. The particle depolarization ratio ranged from 10%-25%. The African
air masses travelled 3000-5000 km before arriving at Leipzig, Germany. Optical
depths were as high as 0.3 to 0.6 at 532 nm. Extinction coefficients ranged from 100
to 300 Mm-1 at both wavelengths. The Angstrom exponent (355/532 nm) varied from
0-0.5. Single scattering values were close to 0.95 (derived from AERONET Sun
photometer observations). The effective radius of the dust particles, estimated from
the combined lidar/photometer data, was about 0.5 µm. This value is much smaller
than typical literature values. Unexpectedly large lidar ratios, mainly between 50 and
80 sr at 532 nm, were observed in the free-tropospheric dust layers. The values for
355 nm were, on average, 10%-30% larger. Based on Mie scattering calculations dust
lidar ratios were expected to be close to 40 sr (355 nm) and 20 sr (532 nm). The
higher lidar ratio for 355 nm (for spherical particles) results from larger absorption in
the UV. However, a clear interpretation of the wavelength dependence of the
observed dust lidar ratios is not possible because of the dependence of the lidar ratio
on the size distribution, chemical composition, and especially on particle shape. In
contrast to the extinction coefficient (and its spectral slope), particle backscattering
and its wavelength dependence is believed to be strongly influenced by shape effects
(Mishchenko et al., 1997).
Current problems
Multiwavelength Raman lidars (backscatter at 355, 532, 1064nm, extinction at 355,
532nm) can in principle be used to retrieve microphysical properties (surface and
volume concentrations of the particles, refractive index) from the optical data (Müller
et al., 2001). Because of the sensitive impact of particle shape on the backscatter
coefficients such inversions (exclusively based on lidar data) are not possible in the
case of desert dust. Combined photometer/Raman lidar observations may improve the
situation. The latest status in this field is given in the workshop poster presented by
Müller et al.. More work (model calculations, closure studies with lidar, radiometers,
aircraft) is needed to provide the lidar community with a clear picture of the
dependence of the backscatter coefficient (at the widely used wavelengths of 355,
532, and 1064 nm) on realistic dust particle shapes and particle distributions for a
mixture of different shapes. Raman lidar observations in southern Italy (much closer
to the source region, eastern part of the Sahara; dust over Leipzig originated mainly
from the western part of the Sahara) yield much lower lidar ratios (50-60 sr) at 355
nm (Leipzig: 60-90 sr), much closer to the values found from Mie calculations (De
Tomasi et al., 2003). After long-range transport of Asian dust lidar ratios over Japan
ranged mostly from 40-55 sr at 532 nm (Liu et al., 2002). These values are lower than
the lidar ratios observed over Leipzig (mostly 60-75 sr), but clearly larger than the
results from Mie calculations. So, presently there is large room for speculations about
the impact of the shape of the dust particles on the measurable optical properties. It
should also be examined in which way the depolarization ratio measured with lidar
depends on the size distribution and the shape of the particles. How useful are
depolarization ratio observations at two wavelengths (e.g., 532 and 1064 nm)? Could
such observations be used to retrieve the mean size of the particles or to identify the
most probable shape of the particles?
References










Ansmann, A., M. Riebesell, and C. Weitkamp, 1990: Measurements of
atmospheric aerosol extinction profiles with Raman lidar, Opt. Lett., 15,
746-748.
Ansmann, A., U. Wandinger, M. Riebesell, C. Weitkamp, and W. Michaelis,
1992: Independent measurement of extinction and backscatter profiles in
cirrus clouds by using a combined Raman elastic-backscatter lidar, Appl.
Opt., 31, 7113-7131.
Bösenberg, J., et al., 2001: EARLINET: A European Aerosol Research Lidar
Network, in Laser remote sensing of the atmosphere. Selected papers of the
20th International Laser Radar Conference, Vichy, France, A. Dabas, C.
Loth, and J. Pelon, Eds., Ecole Polytechnique, Paris, France, 155-158.
De Tomasi, F., A. Blanco, and M. R. Perrone, 2003: Raman lidar monitoring
of extinction and backscattering of African dust layers and dust
characterization, Appl. Opt., 42, 1699-1709.
Kalashnikova, O. V., and I. N. Sokolik, 2002: Importance of shapes and
compositions of wind-blown dust particles for remote sensing at solar
wavelengths, Geophys. Res. Lett., 29, 10.1029/2002GL014947.
Liu, Z., N. Sugimoto, and T. Murayama, 2002: Extinction-to-backscatter
ratio of Asian dust observed with high-spectral-resolution-lidar and Raman
lidar, Appl. Opt., 41, 2760-2767.
Mattis I., A. Ansmann, D. Müller, U. Wandinger, and D. Althausen, 2002:
Dual-wavelength Raman lidar observations of the extinction-to-backscatter
ratio of Saharan dust, Geophys. Res. Lett., 29, 9, 10.1029/2002GL014721.
Mishchenko, M. I., L. D. Travis, R. A. Kahn, and R. A. West, 1997:
Modeling phase functions for dustlike tropospheric aerosols using a shape
mixture of randomly oriented polydisperse spheroids, J. Geophys. Res., 102,
16831-16847.
Müller, D., U. Wandinger, D. Althausen, M. Fiebig, 2001: Comprehensive
particle characterization from three-wavelength Raman-lidar observations:
case study, Appl. Opt., 40, 4863-4869.
Müller D., I. Mattis, U. Wandiger, D. Althausen, A. Ansmann, O. Dubovik,
S. Eckhardt, and A. Stohl, 2003: Saharan dust over a central European
EARLINET-AERONET site: combined observations with Raman lidar and
Sun photometer, J. Geophys. Res., 108, 10.1029/2002JD002918.
The Dirt on Dust
Richard Arimoto
Carlsbad Environmental Monitoring & Research Center, New Mexico State
University, Carlsbad, NM, USA arimoto@cemrc.org
Abstract
The composition and concentrations of atmospheric dust have been characterized in
various ways, perhaps most commonly by determining the concentrations of dustassociated elements. Typically, Al or Fe or Si or some other element is used as an
indicator of dust, often with the tacit assumption that all other sources for that
indicator are negligible; in some cases water soluble Ca has been used for this
purpose. Elemental ratios are then calculated and compared with a compositional
reference to determine which elements are in crustal proportions and which are not.
There are several problems with the reference element approach for evaluating dust
concentrations and composition, the first being weathering reactions which can cause
coatings, such as rock varnishes, to form on the parent materials from which the dust
is formed. Second, alluvial fans and playas are important sources of dust in some
places, and as a result pedogenic salts can be mixed with crustal material when the
eolian dust forms. These two effects cause differences between the composition of the
eolian dust relative to the crustal reference, but they would likely be minor compared
with the next problem, which is that non-dust sources can emit aerosols that in some
ways resemble mineral dust; that is, the "sole source" assumption for the indicator
element is violated. For instance, the concentrations of some major elements such as
Al, Fe and Ca in coal fly ash are similar to those of typical soils, but various other
elements are strongly enriched in the coal combustion residues.
Recent studies conducted as part of ACE-Asia showed that at Zhenbeitai, People's
Republic of China, which is close to the Asian dust source region, the molar ratio of
sulfate to soluble calcium was ~0.1, but the ratio increased to ~1 at Gosan, South
Korea, which is much farther from the main Asian dust sources. One explanation for
these results is that gas-to-particle conversion caused the observed increase in sulfate
relative to Ca downwind, that is, SO2 gas or H2SO4 vapor became associated with
the dust particles during transit. However, both the percentage of Ca that was soluble
and the ratio of soluble calcium to aluminum were lower at Gosan than Zhenbeitai,
indicating that aerosol mixing also affected the Ca concentrations during transport.
Yet another process that could have contributed to the observed differences between
sites is the size-selective fractionation of dust during transport; this process can
compromise the indicator element approach by preferentially removing minerals that
do not have the indicator and other elements in crustal proportions.
Even natural substances such as atmospheric sea salt can cause complications with
the indicator element approach if the amount of the interfering substance is
sufficiently high. At Bermuda, a small island in the North Atlantic, sea-salt Al can
amount to as much as ~30% of the total monthly aerosol Al, even though such high
percentages occur for only a few months each year when the dust concentrations and
deposition rates were low. While the presence of non-dust materials is probably
inconsequential in many cases, interferences of this nature can call into question
inferences about dust concentrations based on indicator elements. The essence of the
problem in using indicator elements is that no elements are unique to dust.
Studies recently conducted in Carlsbad, NM, USA suggest that mineral dust is
contaminated with transuranic radionuclides from atmospheric nuclear weapons teststhe question raised by these studies is whether the relationship between dust and
bomb-derived nuclides is restricted geographically or a widespread phenomenon,
perhaps even involving global contamination. Related studies by other investigators
have shown that dust concentrations can be correlated with those of organic nitrogen,
and the possibility of associations between microbes, pesticides and persistent
organochlorine compounds with dust also has been raised. An important implication
of the widespread contamination of dust--beyond artifacts in determining dust
concentrations--is that the optical and radiative properties of contaminated dust likely
differ significantly from those of pure mineral aerosol. Without information on the
extent and effects of contamination, the characterization of pure dusts' composition
and properties may prove to be more of academic exercise than a key to improving
assessments of aerosols' impact on climate.
HETEROGENEOUS CHEMISTRY ON MINERAL AEROSOL:
INFLUENCE ON TROPOSPHERIC OZONE
Y. Balkanski, M. Schulz, S. Textor, D. Hauglustaine, S.E. Bauer, R. van Dingenen 2,
P. Bonasoni 3, H. Fischer 4 F. Dentener 2, J. P. Putaud 2
IPSL-LSCE, l'Orme des Merisiers, Bât 709, 91191 Gif/Yvette Cedex France
Joint Research Center, I-21020, Ispra, Italy
CNR ISAC, via Gobetti, 40129, Bologna, Italy
Max Planck Institute for Chemistry, POB 3060, 55020 Mainz, Germany
Abstract
The progresses made in understanding the role of mineral aerosol in chemistry and
climate have been very important in the last few years. Satellite, sunphotometers and
Lidars have allowed to document the large regions prone to the influence of these
aerosols. Very pointed techniques used in the laboratory and in the field have
evidenced that the surface of the aerosol allow for heterogeneous reactions to take
place in particular for HNO3, HO2, H2O2, SO2 which in turn affects NOx, NOy and
O3.
The effect on ozone of heterogeneous reactions at the surface of mineral aerosol was
investigated using a General Circulation Model coupled to a module INCA that treats
INteractions between Chemistry and Aerosols. The accomodation coefficients of
H2SO4, NO3, O3 and N2O5 were chosen according to laboratory experiments that
were conducted within the framework of the European project MINATROC
(MINeral Aerosol and Tropospheric Chemistry). Two field experiments one that took
place in Mt Cimone, Italy in June 2000 and the second one in Izana in July-August
2002 provide a characterisation of the gas and aerosols phase at the sites that is
compared to model results.
The results of a nudged global simulation for the years 2000 and 2002 permit to
estimate the effect of the heterogeneous reactions of mineral dust. Tropospheric
ozone is reduced by 10 to 30% in the tropics and the relative roles of H2SO4, NO3,
O3 and N2O5 in this reduction have been sorted out. We will also point to significant
reduction in H2SO4 and O3 predicted over Southern Europe and Asia.
References



Balkanski, Y., S. E. Bauer, R. van Dingenen, P. Bonasoni, M. Schulz, H.
Fischer, G. P. Gobbi, M. Hanke, D. Hauglustaine, J. P. Putaud, A. Stohl, and
F. Raes , The Mt Cimone, Italy, free tropospheric campaign: principal
characteristics of the gaseous and aerosol composition from European
pollution, Mediterranean influences and during African dust events, Atmos.
Chem. Phys. Discuss., 3, 1753-1776, 2003.
Bauer S. E., Y. Balkanski, M. Schulz, D. Hauglustaine and F. Dentener,
Heterogeneous chemistry on mineral dust aerosol surfaces: Influence on the
global tropospheric ozone chemistry, to be submitted to J. Geophys. Res.,
2003.
Guelle W., Y. Balkanski, M. Schulz, B. Marticorena, G. Bergametti, C.
Moulin, R. Arimoto, And K. D. Perry, Modelling the atmospheric
distribution of mineral aerosol: Comparison with ground measurements and
satellite observations for yearly and synoptic time scales over the North
Atlantic, J. Geophys. Res, 105, 1997-2005, 2000.
Satellite views of spatial, seasonal and interannual variability of AFRICAN
dust and relationships with meteorology and climate: combination of
TOMS and METEOSAT observations
I. Chiapello (1), C. Moulin (2), S. Benaissa (1) and M. Legrand (1)
(1) Laboratoire d'Optique Atmosphérique, Université des Sciences et Technologies
de Lille, France, chiapello@loa.univ-lille1.fr, fax :+33-3-20-43-43-42
(2) Laboratoire des Sciences du Climat et de l'Environnement, CE Saclay, Gif-surYvette, France
Abstract
In the recent years, several studies have investigated the mineral dust variability at
interannual time scales on the basis of individual satellite records, either over Africa
using METEOSAT/infrared observations [Brooks and Legrand, 2000], or over
surrounding oceanic regions using METEOSAT/VIS and TOMS (Total Ozone
mapping Spectrometer) observations [Moulin et al., 1997; Chiapello and Moulin,
2002]. These studies have highlighted the potential impact of several climate
parameters on the dust export, especially the North Atlantic Oscillation (NAO) and
the Sahel drought, but are generally limited in time (~10 years) and regionally. To
better understand these impacts, it is necessary to enlarge at maximum the period for
which satellite observations are available and to combine the different satellite
records available.
The daily TOMS/Nimbus-7 (1979-1992) and TOMS/Earth Probe (1997-2000)
Aerosol Index (AI) have been combined with the METEOSAT/VIS (1984-1997) dust
optical thickness (DOT) over the Atlantic ocean to derive an estimated TOMS DOT
since 1979 over both ocean and land. These estimated TOMS DOT are validated over
Africa by comparison to Sun-Photometer measurements performed during field
campaigns in the 1980's and derived from the AERONET network. The temporal and
spatial variability of African dust derived from these estimates are analyzed and
compared to those derived by METEOSAT/VIS over the Atlantic and by
METEOSAT/Infrared over Sahara and Sahel. For these three data sets, inter-annual
variations of dust loads are investigated in conjunction with annual variations of
rainfall in Sahel and North Atlantic Oscillation.
References



Brooks, N. & Legrand, M. Dust variability over northern Africa and rainfall
in the Sahel, S.J. McLaren and D.R. Kniveton (eds.), Linking Climate
Change to Land Surface Change, Kluwer Academic Publishers, 1-25 (2000).
Chiapello, I. & Moulin, C. TOMS and METEOSAT satellite records of the
variability of Saharan dust transport over the Atlantic during the last two
decades (1979-1997). Geophys. Res. Lett. 29, 17-20 (2002).
Moulin, C. et al. Control of atmospheric export of dust from North Africa by
the North Atlantic Oscillation. Nature 387, 691-694 (1997).
Retrieval of Optical Properties of Desert Dust Aerosol from AERONET
Observations
Oleg Dubovik1,2, Brent Holben1, Tom F. Eck1,2, Alexander Smirnov1,2, Tatyana
Lapyonok1,3, Alexander Sinyuk1,2 , Didier Tanre4 Philippe Goloub4 and Ilya
Slutsker1,3
1 NASA Goddard Space Flight Center, Greenbelt, MD, USA
2 Goddard Earth Science and Technology Center, University of Maryland Baltimore
County, USA
3 Science Systems and Applications Inc., Lanham, USA
4 Universite de Science et Techniques de Lille, Lille, France
dubovik@aeronet.gsfc.nasa.gov, brent@aeronet.gsfc.nasa.gov,
teck@aeronet.gsfc.nasa.gov, asmirnov@aeronet.gsfc.nasa.gov,
sinyuk@aeronet.gsfc.nasa.gov, lapyonok@aeronet.gsfc.nasa.gov,
Didier.Tanre@univ-lille1.fr, goloub@loaser.univ-lille1.fr ,
ilya@aeronet.gsfc.nasa.gov,
Abstract
This presentation outlines the main features of desert dust optical properties observed
in the AERONET retrievals.
INTRODUCTION
AERONET - AErosol RObotic NETwork (Holben et al. 1998) of ~ 150 identical
globally distributed sun and sky scanning ground-based automated radiometers
provides measurements of desert dust optical properties in many locations. The
spectral sky-radiance is measured in a wide angular range from the sun and is
minimally affected by surface reflectance. The standardized network procedures
(Holben et al. 1998, Smirnov et al. 2000) of instrument maintenance, calibration,
cloud screening and data processing allow for quantitative comparison of the aerosol
data obtained at different times and geographical sites.
The inversion algorithm (Dubovik and King, 2000) employed by AERONET
provides aerosol retrievals by fitting the entire measured field of radiances - optical
depths and the angular distribution of sky radiances - at four wavelengths (0.44, 0.67,
0.87 and 1.02 µm) to a radiative transfer model. The radiation field is driven by the
(wavelength dependent) aerosol complex index of refraction and the particle size
distribution (22 size bins in the range: 0.05 ≤ r ≤ 15 µm) in the total atmospheric
column. Only spectral and size smoothness constraints are used, preventing
unrealistic oscillations in either parameter.
RESULTS
Size distributions of desert dust are typically bimodal and dominated by coarse mode
(super micron radius) particles. Correspondingly, the Angstrom parameter is low
(ranges from ~ 0.75 down to - 0.1) and the phase function asymmetry is relatively
high at all wavelengths considered. Some differences for dust of different geographic
origin are also observed. For example, the desert dust from the western part of Africa
and the Saudi Arabian Peninsula (Saudi Arabia and Cape Verde) are strongly
dominated by large particles (Cvc/Cvf ~ 50) and seem to have optical properties more
representative of so-called pure desert dust. The aerosol in Bahrain/Persian Gulf has a
larger fine mode (Cvc/Cvf ~ 10) than observed in Saudi Arabia and in Cape Verde.
This difference relates to the frequent presence in the Persian Gulf of small particles
produced by industrial activity. The median sizes range from ~0.12 to ~0.15 µm for
fine mode and from ~1.9 to ~2.6 µm for coarse modes and, in contrast with biomass
burning and urban/industrial aerosols do not show any pronounced dynamics with
aerosol loading.
Single Scattering albedo ω0 values retrieved by AERONET for Saharan dust (0.96 0.99 for wavelengths greater than 550 nm) are significantly higher than many aerosol
models suggest. Similarly, the retrieved imaginary part of the refractive index k(λ)
ranging from 0.0006 to 0.003 are smaller than the 0.008 value given for the visible
spectrum by several models. Another feature of the retrievals is the pronounced
absorption of desert dust the blue spectral range (ω0(440) ~ 0.92 - 0.93 and k(λ) is 3-4
times higher at 440 than at the longer wavelengths). Such spectral dependence is not
surprising for desert dust and has been reported previously in many studies (however
for higher absolute values of absorption). Asian dust in spring, measured over
urbanized China, exhibited greater absorption (ω 0 ~ 0.92 - 0.94 for wavelengths > 550
nm) due to mixing with absorbing fine mode particles and probable attachment of BC
to the dust surfaces. The typical values of real part of the refractive index retrieved by
AERONET ranging from 1.48 to 1.56 for various dust observation are in general
agreement with most available dust measurements.
Particle nonsphericity was consistently observed in desert dust retrievals as the
appearance of retrieval artifacts (high concentration of very small particles with r <
0.1 µm and strong spectral dependence of n(λ)) associated with the presence of
nonspherical particles (as shown in sensitivity studies by Dubovik et al, 2000). The
similarity of these retrieval features with ones observed in numerical tests suggested
that a nonspherical scattering model of randomly oriented spheroids is rather
adequate for desert dust aerosol. Therefore, we have developed an approach allowing
(Dubovik et al, 2002) the use of the model of dustlike aerosol particles as
polydisperse, randomly oriented spheroids in retrieving aerosol optical properties
from remote measurements of atmospheric radiances. The application to the entire
AERONET database has shown significant improvements in the retrieved size
distribution, refractive index, and phase function for aerosols in desert dustdominated or influenced locations.
References





Dubovik, O., A. Smirnov, B. N. Holben, M. D. King, Y. J. Kaufman, T. F.
Eck and I. Slutsker, 2000: Accuracy assessment of aerosol optical properties
retrieval from AERONET sun and sky radiance measurements. J. Geophys.
Res., 105, 9791-9806.
Dubovik, O., and M. D. King, 2000: A flexible inversion algorithm for
retrieval of aerosol optical properties from Sun and sky radiance
measurements. J. Geophys. Res., 105, 20673-20696.
Dubovik, O., B. N. Holben, T. Lapyonok, A. Sinyuk, M. I. Mishchenko, P.
Yang and I. Slutsker, Non-spherical aerosol retrieval method employing
light scattering by spheroids, Geophys. Res. Lett., 10.1029/2001GL014506,
2002.
Holben, B. N., T. F. Eck, I. Slutsker, D. Tanré, J. P. Buis, A. Setzer, E.
Vermote, J. A. Reagan, Y. J. Kaufman, T. Nakajima, F. Lavenu, I.
Jankowiak and A. Smirnov, 1998: AERONET-A federated instrument
network and data archive for aerosol characterization. Remote Sens.
Environ., 66, 1-16.
Smirnov A., B. N. Holben, T. F. Eck, O. Dubovik and I. Slutsker, 2000:
Cloud screening and quality control algorithms for the AERONET data base.
Remote Sens. Environ., 73, 73,337-73349.
About Mineral Dust Deposition
François Dulac
Laboratoire des Sciences du Climat et de l'Environnement (LSCE), Unité Mixte de
Recherche CNRS-CEA No. 1572, CEA Saclay 709, F-91191 Gif-Sur-Yvette Cedex,
France
e-mail fdulac@cea.fr
Abstract
This presentation aims at pointing out the need for experimental studies of mineral
dust deposition and at raising some relevant points and questions.
The study of mineral dust deposition is of interest because Aeolian erosion in arid and
semi-arid areas and subsequent long-range transport of mineral dust particles are
responsible for huge transfers of matter from continent to continent and from
continent to oceans, which affect soil and sediment formation and biogeochemical
cycles.
First, it is of interest to keep in mind that dust concentration in ice or sediment cores
is primarily a record of deposition processes. Present observations show that
atmospheric dust load and deposition at a given place may be uncorrelated and
exhibit different seasonal patterns.
A number of dust deposition studies indicate that the annual fallout is often
dominated by a few events. Although wet deposition generally dominates yearly
budgets, dry deposition may be dominant during half-year long dry seasons so that
both should be assessed separately. There are also large interannual variations in
mineral dust deposition. Strategies for measuring mineral dust deposition should
therefore rely on a continuous monitoring basis over multi-year periods, with a high
temporal resolution of the order of one or few days. The use of automated rain
collectors makes it possible, and should be recommended for monitoring wet
deposition on an event basis. Immediate filtration to apportion chemicals between the
soluble and particulate phases should also be encouraged.
Dust deposition is, indeed, known to be a source of limiting nutrients (e.g. Si, Fe, P,
...) for surface waters of the open ocean. Model studies suggest that high dust
deposition rates to the world ocean at the Last Glacial Maximum may have impacted
atmospheric CO2 concentrations by 30-50 ppm through Fe fertilization. However, we
have a very limited understanding of the extent to which dust deposition provides bioavailable nutrients. First laboratory studies suggest that there are probably differences
in the bio-available fraction from deposited dust between wet and dry deposition, and
this may be of interest to study biogeochemical impacts. It can also be questioned
whether monthly averaged deposition fluxes are relevant to study biogeochemical
impact on surface ocean, due to the sporadic nature of deposition events.
Deposition data have proved much helpful for validating models of aerosol transport.
There has been, however, few attempts to validate mineral dust deposition fields, and
only integrated deposition budgets are generally considered. In order to promote
detailed model validation, a tentative inventory of existing data on mineral dust
deposition will be made, and technical aspects of measurements discussed.
Finally, some data on particle size distribution from deposition samples will be
considered. Deposition sampling appears like a good tool to study the largest particle
size fraction of mineral dust which is likely underestimated by classical
measurements made on aerosol particles or on sampled air.
Passive Visible and Infrared Observations of Dust Aerosol Properties
Philip A. Durkee
Professor of Meteorology Naval Postgraduate School Monterey, California USA
Abstract
The effects of dust have been observed in satellite imagery since the early visible
wavelength radiometers. Today, on the order of two dozen satellites orbit the earth
with the ability to sense dust and its properties. Detecting the presence of dust is
relatively easy compared to many other aerosol types. However, quantifying the
radiative properties of dust and its impact on radiative heating, visibility, and general
human activity is quite difficult. This talk will describe some of those difficulties,
illustrate the consequences of various assumptions and approximations, and describe
some techniques for limiting the ambiguity inherent in the problem through the use of
multiple satellite/sensor analysis.
The workhorse for climatological aerosol observations, beginning in the 1970's, has
been the Advanced Very High Resolution Radiometer (AVHRR) on board the NOAA
series of satellites. AVHRR currently measures radiance in 5 wavelength bands
including three reflected solar bands (0.63, 0.86, and 1.6 ?m band center
wavelengths). With three bands, analysis of AVHRR radiance measurements can
provide aerosol optical depth estimates by constraining the possible combinations of
particle size and absorption characteristics that are consistent with the radiance
measurements. The problem of course has many more than three degrees of freedom,
so the results are dependent on the a priori assumptions about particle size and
absorption properties of the aerosol.
Recent multispectral radiometers such as the Sea-viewing Wide Field-of-view Sensor
(SeaWiFS) or the MOderate Resolution Imaging Spectroradiometer (MODIS) on the
Terra and Aqua satellites display beautiful renditions of dust events by exploiting the
subtle differences in the reflectance of airborne dust compared to clouds or land
surfaces. These sensors, along with the new MEdium Resolution Imaging
Specrometer (MERIS) on ENVISAT, and the GLobal Imager (GLI) on the Midori-II
satellite, provide an unprecedented multispectral view of the reflected solar radiance
from the Earth's aerosol systems. A summary of the sensor characteristics is provided
below.
Solar spectral radiance measurements do not tell the whole story. Radiance
measurements as a function of scattering angle and polarization also provide
important information to constrain estimates of aerosol properties. The Multiangle
Imaging SpecrtroRadiometer (MISR) on the Terra satellite measures radiance at 4
wavelengths from 9 view angles and the Advanced Along Track Scanning
Radiometer (AATSR) on ENVISAT measures radiance at 7 wavelengths from two
directions. The POLarization and Directionality of the Earth's Reflectances
(POLDER) instrument on Midori-II measures polarization parameters.
In addition, since dust aerosol is of sufficient size and concentration, infrared
radiance also carries information about aerosol properties. MODIS, AVHRR, GOES,
and the new Meteosat Second Generation (MSG) sensors all measure radiance in
multiple infrared channels that provide information about dust aerosol.
This talk will illustrate the various dust observation techniques using combinations of
passive visible and infrared sensors. The focus will be on a summary of what is
possible with these techniques and not a comprehensive review due to time
constraints and some techniques will be more adequately described by other speakers.
The limitations of these techniques will also be discussed. Finally, future
developments and improvements in dust aerosol characterization from passive visible
and infrared measurements will be described.
Solar Bands IR Bands
Spatial
(visible/near- (absorption Resolution
IR)
window)s
(km)
AVHRR
3
2
1
Sensor
SeaWiFS
8
0
1
MODIS
19
3
0.25, 0.5,
MERIS 15 0 0.3
and 1
GLI
23
13
0.25 and 1
MISR
4
0
0.275 and
1.1
AATSR
4
3
1
POLDER
8
0
6x7
MSG
3
4
1 (vis), 3
(IR)
Vertical profiling of mineral dust properties
Paola Formenti 1, James M. Haywood 2, Simon Osborne 2, and Meinrat O. Andreae
3
1 LISA, Créteil, France, email: formenti@lisa.univ-paris12.fr
2 UK MetOffice, Farnborough, UK, email: jmhaywood@metoffice.com and
simon.osborne@metoffice.com
3 MPIC, Mainz, Germany, email: andreae@mpch-mainz.mpg.de
Abstract
The physical and optical properties of Saharan dust aerosol measured by the Met
Office C-130 during the SaHAran Dust Experiment (SHADE) are presented. SHADE
was conducted in September 2000 over the Atlantic Ocean between Sal Island and
Senegal.
Dust was encountered in the altitude range from 0.5-1 to 4-4.5 km. Sub-layers with
different particle size distributions and different source areas could be distinguished
within the main dust plume. Dust loadings containing up to 54 µg m-3 Al (submicron
plus supermicron fraction) were measured during particularly intense dust events. The
geochemical signature of mineral dust was consistent with previous results in the
area. Si, Fe, and Ti were not enriched with respect to the soil composition, while
other elements, such as Ca and S, were. Mixing of dust with anthropogenic aerosols,
mainly NH4HSO4, was observed in the fine fraction.
Our results suggest that the absorption by Saharan dust is significantly overestimated
in the solar spectrum if standard refractive indices are used. Our measurements
suggest an imaginary part of the refractive index of 0.0015i is appropriate at a
wavelength, λ, of 0.55µm. Additional radiation measurements enable the
determination of the aerosol optical depth, δa, and the direct radiative effect, DRE, of
the mineral dust. At 0.55 µm, δa approached 1.5 during the period of heaviest dust
loading, which equates to an instantaneous top of the atmosphere DRE over ocean of
approximately -130 W m-2, or a surface DRE of -210 W m-2.
Finally, an attempt will be made to discuss the results of SHADE in the light of those
obtained during the various dust airborne experiments - dedicated and of opportunity
- that have been undertaken in the recent years.
Optical Modelling of Asian Dust Aerosol based on ACE-Asia Ground
Observation, Particle Transport Model, and SeaWiFS Observation
Hajime Fukushima,1) Yuichiro Hagihara, 1) Hiroshi Kobayashi,2) Toshiyuki
Murayama,3) Sachio Ohta4) and Itsushi Uno 5)
1) School of High-Technology for Human Welfare, Tokai University, Numazu, 4100395 Japan E-mail: hajime@fksh.fc.u-tokai.ac.jp
2) Faculty of Engineering, Yamanashi University
3) Tokyo University of Mercantile Marine, Tokyo, Japan
4) Faculty of Engineering, Hokkaido University
5) Research Institute for Applied Mechanics, Kyushu University, Fukuoka, Japan
Abstract
Modeling optical properties of Asian dust is critical in terms of proper and unbiased
evaluation of radiation budget and atmospheric correction of satellite data. The
objectives of this study are 1) to define an optical model of the Asian dust aerosol
based on sky-radiometer and LIDAR observations as well as laboratory-measured
refractive index of Chinese soil particles, and 2) to evaluate the performance of the
model by comparing model-predicted and satellite-derived top-of-atmosphere (TOA)
reflectances. Predicted results of a particle transport model calculation are also used
to model the vertical distributions and the mixture ratio of each aerosol species.
In this study, the particle size distribution (mono-modal, log-normal) and the vertical
profile of dust aerosol model were determined from sky-radiometer and LIDAR
observations conducted in Tokyo urban area on April 10, 2001, when SeaWiFS made
a contemporaneous observation. Laboratory-measured complex refractive indices of
sampled Chinese dust particles were adopted to conduct a radiative transfer (RT)
simulation to synthesize TOA spectral reflectance, which compared well with the
SeaWiFS observation (within few per cent error in reflectance), when the aerosol
optical thickness was adjusted to meet with the SeaWiFS 865 nm band observation.
The same model, with vertical distribution modeled after Chemical Weather Forecast
System (CFORS) prediction, was applied to four data points in Japan Sea over the
SeaWiFS imagery on April 10, when and where a major dust airmass was observed.
Although the RT simulated TOA reflectance reproduced the SeaWiFS reflectance
with 3-4 % error under moderately loaded dust cases, significant differences in
spectral dependency (RT-predicted TOA reflectance in shorter wavelength region
was almost 10% lower than the satellite-observed) was observed under dense dust
condition. We will discuss the reasons of this discrepancy, including the variability in
size distribution and the effect of non-sphericity of the dust particles.
In the presentation, we also plan to discuss the validity of several other cases,
including "anthropogenically contaminated dust aerosol" case such as studied in Li et
al.(accepted). Variability of the refractive indices of dust aerosol obtained from
different sources including skyradiometer observations (AERONET and SKYNET)
will be also discussed.
References

Li, L.-P., H. Fukushima, R. Frouin, B. G. Mitchell, M.-X. He, I. Uno, T.
Takamura, and S. Ohta (accepted): Influence of sub-micron absorptive
aerosol on SeaWiFS-derived marine reflectance during ACE-Asia, J.
Geophys. Res.(AAC).
Dust characterization and its implications for global biogeochemical cycles
Yuan Gao
Princeton University, USA
Abstract
In addition to its direct and indirect climate effects, dust is a source of iron (Fe), an
essential micronutrient for phytoplankton growth in many areas of the surface ocean.
Recent Fe fertilization experiments conducted in the equatorial Pacific, Southern
Ocean and sub-Arctic Pacific confirmed that Fe supplies strongly regulate
phytoplankton growth in these oceanic regions. The Fe input from dust may, in turn,
regulate the global carbon cycle and thus affect climate. However, the distributions of
dust (or aeolian Fe) delivery to the ocean vary strongly with season and from one
ocean region to another. Dust particles also provide reaction sites for many
heterogeneous reactions involving SO2, NOy, HOx, O3, etc. during the long-range
transport. Those processes may alter dust properties, including aeolian Fe solubility in
the air and then its bio-availability in the surface ocean. Thus characterization of dust
properties is crucially important to understand its roles not only in the Earth's
radiation budget but also in global biogeochemical cycles. This presentation intends
to share recent findings on dust properties, including preliminary results obtained
from shipboard observations in the North Pacific during the Asian-Pacific Regional
Aerosol Characterization Experiment.
Empirical TOMS Index for Dust Aerosol: Applications to Model
Validation and Source Characterization
Paul Ginoux
NOAA GFDL, Princeton University, Forrestal Campus, Route 1, Princeton NJ
08542-0308
Abstract
An empirical relation is developed to express the TOMS aerosol index (AI) for the
case of dust plumes, as an explicit function of four physical quantities: the single
scattering albedo, optical thickness, altitude of the plume and surface pressure. This
relation allows sensitivity analysis of the TOMS AI with physical properties,
quantitative comparison with dust model results and physical analysis of dust sources,
without the necessity of cumbersome radiative calculation.
Two applications are presented: 1) the case study of a dust storm over the North
Atlantic in March 1988, and 2) the characterization of 13 major dust sources. The
first application shows that simulated dust distribution can be quantitatively
compared to TOMS AI on a daily basis and over regions where dust is the dominant
aerosol.
The second application necessitates to further parameterize the relation by replacing
the optical thickness and the altitude of the plume by meteorological variables. The
advantage is that surface meteorological fields are easily available globally and for
decades but the formulation only applies to dust sources. The daily, seasonal and
inter-annual variability of the parameterized Index over major dust sources reproduce
correctly the variability of the observed TOMS AI. The correlation between these two
indices is used to determine the surface characteristics and physical properties of dust
aerosol, over the sources.
Modelling size-segregated soil dust aerosol during Ace-Asia, 2001:
Implications for Trans-Pacific Transport
S.L. Gong 1, 2 and X.Y. Zhang 2,
1Air Quality Research Branch, Meteorological Service of Canada 4905 Dufferin
Street, Toronto, Ontario M3H 5T4, CANADA
2State Key Laboratory of Loess & Quaternary Geology Institute of Earth
Environment, Chinese Academy of Sciences 10 Fenghui S. Rd., PO Box 17, XiAn
710075, CHINA
Abstract
A size segregated soil dust emission and transport model NARCM was used to
simulate the production and transport of Asian soil dust during the ACE-Asia period
from March to May 2001 [Gong et al., 2003]. The model is driven by the NCEP reanalyzed meteorology and has all the atmospheric aerosol physical processes of soil
dust: production, transport, growth, coagulation, and dry and wet deposition. A
Chinese soil texture map that infers the soil grain size distribution with 12 categories
was generated to drive the size distributed soil dust emission scheme [Alfaro et al.,
1997; Marticorena and Bergametti, 1995]. The size distribution of vertical dust flux
was derived from the observed surface dust size distribution in the desert regions.
Parameters applicable to the Asian deserts for the dust emission scheme are assessed.
Comparisons of model simulations were carried out with ground base measurements
in East Asia and North America, and satellite measurements for the same period of
time. The model captured most of the dust mobilization episodes during this period in
China and reasonably simulated the concentrations in source regions and downwind
areas from east China to western North America. About 252.8 Mt of soil dust below d
< 40 µm was estimated to be emitted in the East Asian deserts between March 1 to
May 31, 2001 with ~60% attributed to four major dust storms. The vertical dust
loadings above 700 hPa correlate reasonably well with TOMS aerosol index (AI)
observations. The sensitivity analysis of model performance to soil size distribution,
water moisture and meteorology was carried out with the observational data to
establish the most appropriate parameters and conditions for the Chinese soil dust
production and transport.
Size-segregated budgets of soil dust aerosols in Asia during ACE-Asia were also
investigated [Zhao et al., 2003]. Simulated mass size distributions of dust deposition
showed a similar size distribution to the dust emission fluxes over the source regions
and a decreased peak corresponding to a 1-3 µm diameter range over down-wind
regions. The simulations suggest that dry deposition was a dominant dust removal
process near the source areas and the removal of dust particles by precipitation was
the major process over the trans-Pacific transport pathway, where wet deposition
exceeded dry deposition by up to a factor of ten. The Asian dust deposition from the
atmosphere to the North Pacific Ocean was correlated not only with precipitation
over the North Pacific, but also with the dust transport patterns. Variations of monthly
Asian dust outflow were identified with the latitudinal centre of transport at 38oN in
March; 42oN in April and 47oN in May. The monthly trans-Pacific transport patterns
of Asian Dust in spring were characterized. The transport-axis extended around 300N
and 400N from the East Asian subcontinent to the North Pacific in March. A zonal
transport pathway around 40 0N was well developed in April over the North Pacific
and reached North America. However, the transport in May was separated into two
pathways: an eastward zonal path over the North Pacific and a meridional path from
the source regions to the Northeast Asian continent. Based on the averaged dust
budgets during spring 2001, it was found that the major sources of Asian dust were
located in the desert regions in China and Mongolia with an estimated dust emission
of 21.5 tons.km-2, and the regions from the Loess Plateau to the North Pacific were
sinks of soil dust aerosols with the Loess Plateau as the main sink for Asian dust.
References



Alfaro, S.C., A.Gaudichet, L.Gomes, and M. Maillé, Modeling the size
distribution of a soil aerosol produced by sandblasting, J. Geophys. Res,
102, 11,239-11,249, 1997.
Gong, S.L., X.Y. Zhang, T.L. Zhao, I.G. McKendry, D.A. Jaffe, and N.M.
Lu, Characterization Of Soil Dust Distributions In China And Its Transport
During ACE-ASIA 2. Model Simulation and Validation, Journal of
Geophysical Research, 108 (D9), 2003.
Marticorena, B., and G. Bergametti, Modeling the atmospheric dust cycle.
Part 1: Design of a soil-derived dust emission scheme, J. Geophys. Res.,
100, 16415-16430, 1995.

Zhao, T.L., S.L. Gong, X.Y. Zhang, and I.G. McKendry3, Modelled sizesegregated wet and dry deposition budgets of soil dust aerosol during ACEAsia, 2001: Implications for Trans-Pacific Transport, Journal of Geophysical
Research, in repss, 2003.
Local and Global Dust over North America
Rudolf B. Husar
Center for Air Pollution Impact and Trend Analysis, Washington University, St.
Louis, MO, USA rhusar@me.wustl.edu
Abstract
There is considerable research literature on the dust aerosol pattern and characteristics over North America. The extensive
previous work by Prospero, Gillette, Cahill, Malm and others has clearly indicated that Sahara dust is an important
component of fine and coarse particles in the south eastern US. However, both the recent satellite and previous research
on North American dust is fragmented and uneven in spatial, temporal and compositional coverage. An integrated
assessment of the North American dust using the rich literature and the most recent are therefore desirable.
The objectives of this work are to (1) establish the spatio-temporal and chemical pattern of the airborne dust over North
America (2) characterize the features of dust from the different sources and to (3) attribute the dust over NAM to the
major source regions. The approach is to integrate data from surface and satellite observations and to combine spatial,
temporal and compositional analysis.
The dust over North America originates from local sources as well as from the Sahara and Gobi Deserts. Each dust source
region has distinct chemical signature in the crustal elements. The pattern of different dust contributions varies in space as
well as by season, episodicity and vertical distribution. New satellite sensors allow monitoring of the spatial and temporal
pattern of dust events on a daily basis. Satellite observations of Sahara dust also provide compelling evidence for
intercontinental dust transport. The Sahara dust transport accross the Atlantic has also been frequently photographed by
the astronouts.
Virtually all dust mass is over 1 µm in size. The mass mean diameter (MMD) of dust near the source is over 5-10 µm.
However, long-range transported dust (3-10 days old) has MMD of 2-5 µm. Hence, local dust is virtually all in the coarse
mode (>2.5 µm) while long-range dust has 30-50% of the mass in the PM2.5 range. Comparison of dust elemental
composition at Denali NP, AK (Asian dust) and at Virgin Islands NP. (Sahara dust) show major differences. Al/Si ratio
(Sahara- 0.66; Asian - 0.4) and in K/Si ratio (Sahara- 0.15; Asian - 0.08).
Based on chemical composition measurements (Sisler and Malm) the highest average fine particle dust concentration
occurs in Texas, where it accounts for 10-25% of the fine particle mass. In the northeastern US dust accounts for <5% of
the fine particle mass.
Each dust source has a unique seasonal, synoptic scale (5-10 days) and diurnal pattern. This pattern can be used for source
identification. However, the resulting concentrations at distant receptor sites are modulated by transport and removal
processes. In most regions the seasonality has a single peak, but in some regions the fine particle dust concentrations peak
in two seasons. For example, the dust concentration at Big Bend, TX peaks in July and April. The combination of dust
composition data and seasonal transport pattern reveal that the July peak at Big Bend, TX is due to Sahara dust in the 3-5
µm size range, while the April dust peak is due to local wind blown dust of 5-10 µm dust.
Figure 1. Fine particle dust chemical composition pattern at Big Bend, TX.
In July the Sahara dust contribution to fine particle mass is 4-8 µg/m3 throughout the Southeast. During this month the
Sahara dust contributions exceed the local source contribution by factor of 2-4.
The extensive previous work by Prospero, Cahill, Malm and others has clearly indicated that Sahara dust is an important
component of fine and coarse particles in the southeastern US. Our analysis of the EPA AIRS PM10 data base combined
with the transport analysis has revealed several major dust incursions over the Gulf Coast. In three such episodes July 5,
1992, June 30, 1993 and June 21, 1997, the PM10 concentration exceeded 80 µg/m3 and over multistate regions of the
Southeast.
Asian dust is generated over the Gobi desert most frequently in the spring season. The Gobi dust clouds frequently
traverse the Pacific and the fraction reaches North America. Multi-year satellite data from the AVHRR sensor show the
prominent spring-time aerosol plume (AOT>0.3). Dust is a major contributor to the spring-time Asian plume along with
biomass smoke and industrial sulfate-organic haze. Trans-Pacific dust transport events have been documented
extensively, including the April 1998 and the April 2001 events. Aerosol chemical climatology for Alaska and West Coast
of North America confirms the spring-time Asian dust incursions. However, unlike the Sahara dust the long-term average
contribution of Asian dust to North America is not yet established.
Measurement of free iron content in desert dust : effect on light absorption,
size dependence and soil influence
S. Lafon, J.-L. Rajot, S. C. Alfaro, A. Gaudichet
LISA , UMR CNRS 7583, Universités Paris 7 et 12, Créteil, France
Abstract
In order to predict dust optical properties from their physical and chemical
characterization, it is necessary to better quantify free iron content. Iron is a key
element for solar light absorption and the quantification of its different forms in
aerosol is still a great source of uncertainty. The light absorbent agent is the free part
of iron. Free iron is defined as iron oxides or hydroxides under the form of discrete
particles or of coating particles bound to silicates surface. The other part of iron in
mineral dust is the structural iron, mainly present under the form of substitution
cation in aluminosilicate particles.
A method to apportion the free iron content in aerosol samples collected in small
quantities on polycarbonate filter was developed. Measurements performed on bulk
aerosols samples collected on the field showed that free iron content varies in natural
mineral aerosols. Firstly, we want to establish the quantitative relation between light
absorption capacity of the aerosol and its free iron content. The second point is to
provide information on size resolved free iron content in aerosol. Finally, the
relationships between free iron content of aerosol and their parent soils are studied
with special emphasis on size distribution of both aerosol and soil particles.
To answer these questions, we use aerosol samples generated from three natural soils
collected in different sources regions : Sahel (Niger), North Sahara (Tunisia) and
Ulan Buh desert (China). Dusts are generated in wind tunnel and collected with a
special isokinetic device allowing to collect simultaneously samples by different
ways : Filtration to obtain bulk samples for free iron quantification, and for white
light absorption measurements with aethalometer, Cascade impactor collection to
quantify size resolved free iron content.
The three kinds of aerosol samples present free iron content relative to the total oxide
mass of sample that are significantly different : 6.6 % (sd 0.1), 3.7 % (sd 0.2) and 3.2
% (sd 0.2) for Nigerien, Tunisian and Chinese samples respectively. Absorption in
the visible spectrum is linearly correlated to the mass of free iron contained in studied
samples and this whatever the sample origin. Such an unique correlation doesn't exist
with sample mass or total iron mass. This confirms that free iron is the only
compound responsible for white light absorption by mineral dust.
Chemical analysis of impactor samples are used to determine size distribution in
mass. All the three aerosol samples contain the same two major particles populations
with geometric mean diameter of 3 µm and 12 µm. Free iron contents are analyzed
for impactor stages that better correspond to these particle sizes. Relative to the mass
of aerosol on each stage, free iron proportion is more important in the largest mode.
This means that each fractionation process during dust cycle may generate evolution
of bulk aerosol free iron content and hence of light absorption properties.
The soil /aerosol relationship was studied by quantifying free iron in the parent soils
fractions smaller than 20 µm. Differences in free iron contents of bulk aerosol are
found to reflect differences between parent soils. This influence of the parent soil
composition is also observed in each of the aerosol fraction. Differences between the
various aerosols size fractions and those due to difference in soil compositions are in
the same order of magnitude. Thus to assess the free iron content all along the aerosol
cycle, it is necessary to take into account 1, the parent soil free iron content and 2, the
size distribution changes during transport.
Thermal Infrared Radiometry and Microphysical Properties of Mineral
Dust
Michel Legrand and Ovidiu Pancrati
LOA/USTL, Batiment P5, 59655 Villeneuve d'Ascq, France
legrand@loa.univ-lille1.fr
Abstract
Introduction
The microphysical and optical properties of mineral dust are intricate and variable,
including the particle size distribution, shape and mineralogical composition. In part
for this reason, and also due to the large and fast spatial and temporal changes of its
atmospheric concentration, dust is insufficiently well described, particularly in view
of its radiative forcing determination (Sokolik et al., 2001). In the visible and NIR
parts of the spectrum, the ground-based photometric method and the satellite
reflectance method over ocean and low-albedo surfaces, can be efficiently used
through photometric networks such as AERONET (Holben et al., 1998). This
approach is still to be improved with the enhanced capabilities of the new airborne
sensors (King et al., 1999). Over land, especially for high albedo surfaces of arid
regions where mineral dust originates, satellite remote sensing of dust is efficient only
in the UV (Herman et al., 1997) and in the TIR (Legrand et al., 2001) parts of the
spectrum. The dust indices derived at these wavelengths are considered as semi
quantitative because they can be related to visible optical thickness - the key
parameter for all aerosols atmospheric content - empirically only (Hsu et al., 1999;
Legrand et al., 2001).
The TIR multiwavelength radiometer CLIMAT
Obviously, for ground truth operations to be carried out along with satellite remote
sensing of mineral dust in the TIR, it is advisable to use a ground-based radiometer
measuring the downwelling radiance at the satellite TIR wavelengths, instead of (or
in addition to) a photometer working at visible and NIR wavelengths. It is one of the
objects of the development in the 90s of the TIR multiwavelength radiometer
CLIMAT, designed to ground-based measurements (Sicard et al., 1999; Legrand et
al., 2000; Brogniez et al., 2003). This is a self-sufficient portable field instrument
designed to measure radiances or brightness temperatures from the ground, either
manually or automatically. Its FOV is 10° wide. It can hold up to six channels in the
TIR spectrum. The quality of its long term stability has been controlled. In the field, it
is operated with a robot and a black cavity. The optical head of the radiometer is
successively directed by the robot for measurements to the sky (target) and to the
black cavity. The latter is dedicated to the control of in situ measurement.
The Sahelian campaign NIGER 98 and its results
A joint LISA/IRD/LOA/AGRHYMET/ACMAD campaign was carried out during the
period February-May, 1998, on the site of Banizoumbou (50 kilometers east of
Niamey), Niger, with the project of characterizing mineral dust, its emission and its
transport. The LOA participated to this experiment with the radiometer CLIMAT and
a hand-held photometer Cimel. The following results have been derived from the
data:
° after corrections using the blackbody, the measured TIR brightness temperatures are
derived with an accuracy of a few hundredths of kelvin;
° the measured radiances in the various channels of the instrument are very sensitive
to the presence of mineral dust;
° after elimination of cloudy measurements and correction of water vapor variations,
a radiance varying quasi linearly with the photometric dust optical depth is obtained;
° the instrument channels radiance is evenly enhanced as a function of the particles
size;
° the channels provide a spectral signature depending strongly on dust mineralogical
composition through its complex index.
So, we observe that a dust model using a refractive index from the compilation by
d'Almeida et al. (1991), or a pure quartz dust model, cannot fit the measured spectral
signature. On the other hand, if this dust model is dominated by a clay component,
such as described by Caquineau et al. (2002), its spectral signature calculated using
the complex index given by Sokolik and Toon (1999) turns to an outstanding
agreement with the measurements. In addition, the performances of the radiometer
could be easily improved for future campaigns, by increasing the number of filters
(up to six) and by fitting their wavelengths and widths according to the absorption
peaks of the various mineralogical species of concern. Then, for example illite and
kaolinite could be separated.
Conclusion and prospects
The radiometer CLIMAT used in a field experiment dedicated to mineral dust, is able
through the measured TIR radiances to supply information about (large) particle size
distribution and about mineral composition. For this, it could be used with benefit
with a photometer such as that used in AERONET network, in order to complete the
microphysical description of mineral dust. The addition of a lidar to these instruments
would add the aerosol vertical profile and a sensitive cloud detection. As shown by
comparing the ground-based measured TIR radiances from NIGER 98 with the
coincident TIR dust index IDDI from Meteosat IR, the radiometer CLIMAT is
adequate for control, validation, correction and analysis of the TIR satellite data. It
will be suitable for the implementation of the future multispectral IDDI from MSG
(four channels expected) taking advantage of the spectral signature of dust to retrieve
its atmospheric content and composition over land.
References












d'Almeida, G.A., P. Koepke and E. Shettle, 1991: Atmospheric aerosols,
global climatology and radiative characteristics, A. Deepak Publishing,
Hampton, VA, 561 pp.
Brogniez, G., C. Pietras, M. Legrand, P. Dubuisson and M. Haeffelin, 2003:
A high-accuracy multiwavelength radiometer for in situ measurements in the
thermal infrared. Part II: Qualification in field experiments, J. Atmos.
Oceanic Technol., 20, 1023-1033.
Caquineau, S., A. Gaudichet, L. Gomes and M. Legrand, 2002: Mineralogy
of Saharan dust transported over northwestern tropical Atlantic Ocean in
relation to source regions, J. Geophys. Res., 107, doi 10.1029/
2000JD000247.
Herman, J.R., P.K. Bhartia, O. Torres, C. Hsu, C. Seftor and E. Celarier,
1997: Global distribution og global absorbing aerosols from Nimbus-7
TOMS data, J. Geophys. Res., 102, 16,911-16,922.
Holben, B.N., et al., 1998: AERONET - A federated instrument network and
data archive for aerosol characterization, Remote Sens. Environ., 66, 1-16.
Hsu, N.C., J.R. Herman, O. Torres, B.N. Holben, D. Tanré, T.F. Eck, A.
Smirnov, B. Chatenet and F. Lavenu, 1999: Comparisons of TOMS aerosol
index with sun-photometer aerosol optical thickness: Results and
applications, J. Geophys. Res., 104, 6269-6279.
King, M.D., Y.J. Kaufman, D. Tanré and T. Nakajima, 1999: Remote
sensing of tropospheric aerosols from space: Past, present and future, Bull.
Am. Meteorol. Soc.s 80, 2229-2259.
Legrand, M., A. Plana-Fattori and C. N'doumé, 2001: Satellite detection of
dust using the IR imagery of Meteosat 1. Infrared difference dust index, J.
Geophys. Res., 106, 18,251-18,274.
Legrand, M., C. Pietras, G. Brogniez, M. Haeffelin, N.K. Abuhassan and M.
Sicard, 2000: A high-accuracy multiwavelength radiometer for in situ
measurements in the thermal infrared. Part I: Characterization of the
instrument, J. Atmos. Oceanic Technol., 17, 1203-1214.
Sicard, M., P.R. Spyak, G. Brogniez, M. Legrand, N.K. Abuhassan, C.
Pietras, J.-P. Buis, 1999: Thermal-infrared field radiometer for vicarious
cross-calibration: Characterization and comparison with other field
instruments, Opt. Eng., 38, 345-356.
Sokolik, I.N., D.M. Winker, G. Bergametti, D.A. Gilette, G. Carmichael,
Y.F. Kaufman, L. Gomes, L. Schuetz and J.E. Penner, 2001: Introduction to
special section: Outstanding problems in quantifying the radiative impacts of
mineral dust, J. Geophys. Res. 106, 18,015-18,027.
Sokolik, I.N. and O.B. Toon, 1999: Incorporation of mineralogical
composition into models of the radiative properties of mineral aerosol from
UV to IR wavelengths, J. Geophys. Res., 104, 9423-9444.
On the interactions of mineral dust particles and clouds
Zev Levin
Tel Aviv University, Ramat Aviv 69978, Israel zev@hail.tau.ac.il
Abstract
The direct effects of mineral dust particles on climate are not yet understood. Their
effects depend on the size distribution of the particles and their chemical composition
which vary depending on the sources and the distance traveled. The indirect effect of
these particles on climate is even less understood.
Mineral dust particles are normally relatively poor cloud condensation nuclei, CCN,
but are good ice nuclei IN. However, as these aerosols are carried over long distances,
they become modified through interactions with pollution, originating from other
sources. Published work shows that many of these modified particles are coated with
soluble material, such as sulfate and nitrate, making them efficient giant CCN. The
way in which these particles get coated with soluble material is not yet clearly
resolved. Simple calculations show that Brownian scavenging cannot account for the
aggregation of pollution particles on dust, because the particles are too small and their
aggregation efficiency is too small. Calculations show that one possible route is
through clouds. Namely, cloud drops grow on sulfate or nitrate particles and only
then collect interstitial dust particles. Since most clouds evaporate, the released
residue consists of particles with dust cores coated with sulfate or nitrate.
Measurements conducted during the MEIDEX (Mediterranean Israeli Dust
Experiment) revealed the presence of large dust particles coated with sea salt.
Calculations show that when such large particles enter clouds, precipitation will be
enhanced significantly, leading to shorter cloud lifetime and increased effective cloud
drop size. Unfortunately, this increase in cloud drop effective radius cannot be seen
remotely from space observations because the center of the precipitation is well
below cloud top.
On the other hand, calculations also show that incorporation of too many either small
or large CCN into clouds can lead to reduced rain amounts due to the increased
competition among the drops for the available water vapor.
In the lecture some of the recent calculations and results from measurements in
MEIDEX will be described.
Retrieval of the surface characteristics controlling mineral dust emissions: a
focus on the aeolian roughness length
B. Marticorena
LISA, UMR CNRS 7583, Universités Paris 7 et 12 61, av. du Général de Gaulle,
94010, Créteil, France
email : marticorena@lisa.univ-paris12.fr
Abstract
Dust emissions models describing explicitely the involved processes have
demonstrate their capability to reproduce both the frequency and the intensity of dust
emissions. Moreover, they allow the simulation of the size-distribution associated
with the dust emissions fluxes and its variations with the dynamical conditions
prevailing during the emission events. Since they are physically explicit, they can be
used to simulate the dust emissions for different climatic conditions and makes
possible the simulations of interactions between the mineral dust cycle and the
climatic conditions.
One of the main limitations of such models is the availability of the required input
data. Among them, the surface roughness and the soil size-distribution are known as
the key factors controlling the dust event frequency and the intensity and the sizedistribution of the dust fluxes.
In arid regions, the surface exhibits very different features ranging from very flat and
smooth areas to rocky mountain. The surface roughness is controlled by inert nonerodible elements (boulders, pebbles, gravel, sand grains, etc.) and should not vary
significantly at the time scale mineral dust is generally studied. As a result, a
roughness data base can be established and would not require frequent updating. The
possibilities of estimating the aeolian surface roughness for such surfaces from
remote sensing have been investigated, at different spatial scales and resolutions.
Greeley et al. [1991] proposed an empirical relation between the aerodynamic
roughness lengths z0 measured over various arid surfaces (Death Valley and Lunar
Lake, USA; Gobabeb, Namibia) and the radar backscatter cross-section σ0 from the
Airborne Synthetic Aperture Radar system (AIRSAR). This method, applied to radar
space-borne measurements (onboard the US "Shuttle": SRL-1 in April 1994, and
SRL-2 in October 1994) with a horizontal resolution of (50×50) m2, allows the
retrieval of z0 within half an order of magnitude [Greeley et al., 1997]. More recently,
another approach has been developped to retrieve the roughness length from the socalled protrusion coefficient PC, computed from POLDER surface products derived
from bidirectional reflectance measurements. The POLDER surface products are
available on a global scale with a spatial resolution of about 6×6 km for eight months
(November 1996 to June 1997). An empirical relationship has been established
between the PC's and the logarithm of the aerodynamic roughness length estimated
over the Sahara based on a geomorphological approach [Callot et al., 2000]. This
relation is confirmed by the few experimentally determined values of roughness
length available in the literature [Marticorena et al. in press].
To further investigate the relationship between these two remotely sensed parameters
and the surface roughness, ground measurements of geometric and aerodynamic
roughness have been performed during one month experiment in the South-Tunisia.
A direct comparison between the experimental data and the backscatter coefficients
derived from high resolution ERS/SAR measurements provides an empirical
relationship comparable to the one proposed by Greeley et al. [1991]. Due to the
natural variability of the surface roughness over the selected sites, a direct
comparison between field measurements and the protrusion coefficient derived from
POLDER measurements was not possible. However, the radar backscatter
coefficients averaged over homogeneous surfaces having the dimensions of a
POLDER pixel have been found linearly correlated to the protrusion coefficient.
These results suggest that both methods allow the retrieval of z0. Depending on the
required dimensions and spatial resolution, one or the other method can be used to
established maps of the aerodynamic roughness length over arid regions with a spatial
resolution ranging from a 1/4 of degree square (Polder PC) to 12.5 m x 12.5 m
(Radar).
A similar approach, based on vegetation index can be developed for semi-arid areas
where vegetation is the main roughness element. However, modelling the vegetation
remains the only approach to account for retroactions between the climatic
conditions, vegetation and the mineral dust emissions. Such an approach has been
investigated using a model specifically designed to simulate the growth of the
Sahelian steppic vegetation during the wet season [STEP model, Mougin et al.,
1995]. This model described the soil water budget as a function of the soil texture
profiles, based on daily climatologic parameters (mean temperature, precipitation, etc.
). The model main ouputs are the green and the dry biomass, from which the mean
characteristics of the annual herbaceous vegetation are derived : mean height and
cover rate of the green and dry vegetation, on a daily time scale. Annual simulations
of the vegetation over the Sahelian belt have been performed with a resolution of
1°×1° for two years exhibiting different annual precipitation rates and spatial pattern.
The aim of these simulations is to provide an estimation of the mineral dust emissions
in a "natural" semi-arid areas, without any human disturbance. The simulated biomass
of green vegetation was generally found in good agreement with vegetation indexes
derived from AVHRR observations, the level of agreement depending on the
precipitation regimes and the soil types. Indeed, the vegetation simulations are
sensitive to the precipitation fields (occurence of the first precipitation, and
precipitation amount cumulated over the growing period) and soil texture profile ( %
of clay as a function of depth) used as input data. From these simulations, the
influence of this seasonal vegetation of the frequency of dust emissions on the
Sahelian region has been estimated (assuming that the Sahel is not anthropogenically
perturbed). The results suggest that a "natural" Sahel does not represent a significant
source of mineral dust, compared to the Sahara.
SULFATE AND NITRATE COATINGS ON MINERAL DUSTS:
CRYSTALLINE OR AQUEOUS?
Scot T. Martin, Hui-Ming Hung, and Jeong-Ho Han
Division of Engineering and Applied Sciences Harvard University, Cambridge, MA
02138, USA
Abstract
Keywords: Phase transition; Atmospheric Aerosols; Ammonium sulfate; Ammonium
nitrate
Observational evidence shows that mineral dusts in Asian outflows become coated by
sulfates and nitrates. Layer thickness can range to hundreds of nanometers. At high
relative humidity, the layers uptake water and are in aqueous form. For example, the
deliquescence relative humidity of ammonium sulfate is 80% at 298 K while that of
ammonium nitrate is 60%. The water content has several important effects. The
particle volume increases with concurrent increases in the mass extinction coefficient,
the single scatter albedo (in the infrared), and the asymmetry parameter. The aqueous
coatings also provide milieu for aqueous chemical reactions, such as sulfate oxidation
or N2O5 hydrolysis. At lower relative humidity, the aqueous coating crystallizes.
There is a concomitant release of the water content to the vapor phase and associated
decrease in particle volume. The mass extinction coefficient, single scatter albedo,
and asymmetry parameter decrease. Important chemical transformation pathways are
shut off.
At what critical relative humidity does the phase change from an aqueous to a
crystalline coating occur?
The answer to this question shows that dependent factors are the size and the
chemistry of the mineral dust core. Through detailed laboratory experiments, the
governing polynomial are found to be :
for one hour observation times. The symbols are CRH = crystallization relative
humidity; D = diameter (nm) of mineral core particle; AS = ammonium sulfate; AN =
ammonium nitrate; c = corundum; and h = hematite.
The implication of the above equations is that there is a synergistic interaction
between hygroscopic components of the aerosol such as sulfates and nitrates and
insoluble components such as mineral dusts on the chemical and radiative properties
of atmospheric aerosols.
The CRH equations are derived from laboratory experiments with submicron aqueous
ammonium sulfate particles containing hematite (á-Fe2O3) and corundum (á-Al2O3)
inclusions in an aerosol flow tube at 298 K. Ammonium sulfate coatings of different
layer thicknesses are deposited on metal oxide particles generated by spray pyrolysis
methods. The heterogeneous nuclei (i.e., the mineral dust cores) regulate the RH of
the phase transition from 35% up to 60% RH in the case of ammonium sulfate and 0
to 10% in the case of ammonium nitrate as the inclusion size varies from 50 to 500
nm. The strong size dependence is inconsistent with the application of classical
nucleation theory on defect free surfaces. However, an active site model successfully
interprets the data. Model optimization yields 1010 sites cm-2 and m < 0 for a-Al2O3
and 109 sites cm-2 and m < 0 for a-Fe2O3 particles.
Figure 1. Plot of crystallization relative humidities of aqueous ammonium sulfate and
ammonium nitrate outer layers on hematite and corundum cores.
References







Buseck, P.R., D.J. Jacob, M. Posfai, J. Li, and J.R. Anderson, Minerals in
the air: An environmental perspective, Int. Geol. Rev., 42, 577-593, 2000.
Fletcher, N.H., Actives Sites and Ice Crystal Nucleation, J. Atmos. Sci., 26,
1266-1271, 1969.
Gorbunov, B.Z., and N.A. Kakutkina, Ice crystal formation on aerosol
particles with a non-uniform surface, J. Aerosol Sci., 13, 21-28, 1982.
Martin, S.T., Phase Transitions of Aqueous Atmospheric Particles, Chem.
Rev., 100, 3403-3453, 2000.
Martin, S.T., and J.H. Han, An Aerosol Chemical Reactor for Coating Metal
Oxide Particles with (NH4)2SO4-H2SO4-H2O. 3. Manipulation of the
Sulfate Coating, J. Crystal Growth, 219, 290-299, 2000.
Martin, S.T., J.H. Han, and H.M. Hung, The Size Effect of Hematite and
Alumina Inclusions on the Efflorescence Relative Humidities of Aqueous
Ammonium Sulfate Particles, Geophys. Res. Lett., 28, 2601-2604, 2001.
Posfai, M., J.R. Anderson, P.B. Buseck, T.W. Shattuck, and N.W. Tindale,
Constituents of a Remote Pacific Marine Aerosol: A TEM Study, Atmos.
Environ., 28, 1747-1756, 1994.
How well can we constrain the single scatter albedo of dust?
Sarah Masonis*, Tad Anderson*, William Conant@, Antony Clarke#, Steven
Howell#, Patricia Quinn$, Anne Jefferson%, and Barry Huebert#
*University of Washington, Seattle, WA, USA;
@California Institute of Technology, Pasadena, CA,USA;
#University of Hawaii, Honolulu, HI, USA;
$NOAA-PMEL, Seattle, WA, USA;
%NOAA-CMDL, Boulder, CO, USA
sarahd@atmos.washington.edu, tadand@atmos.washington.edu,
wconant@caltech.edu, tclarke@soest.hawaii.edu, showell@soest.hawaii.edu,
Patricia.K.Quinn@noaa.gov, Anne.Jefferson@noaa.gov, huebert@hawaii.edu
Abstract
Global-average estimates of dust radiative forcing are currently highly uncertain, both
in magnitude and sign (IPCC, 2001). The uncertainty in sign results in large part from
uncertainty in dust's single scatter albedo, ω. Model estimates of dust single scatter
albedo, which are based on an assumed size distribution and mineral composition,
often yield values of ω<0.9 at visible wavelengths (cf. Myhre and Stordal, 2001).
However, in-situ measurements by Haywood et al. of Saharan dust (2001 & 2003)
and in-situ measurements of Asian dust (ACE-Asia; Anderson et al., 2003) both yield
values of ω~0.96-0.97 at 550nm. Satellite measurements also indicate dust ω>0.9 at
visible wavelengths (Kaufmann et al., 2001).
Unfortunately, the longwave impact of dust is still highly uncertain. While most
atmospheric aerosols only interact with visible-wavelength radiation, dust also has an
impact on the earth's longwave radiative balance. Because dust is often lofted high
into the atmosphere during the passage of frontal systems, if its single scatter albedo
at infrared wavelengths is low enough it could act as a warming agent, even if the
shortwave TOA forcing is negative. In-situ measurements of dust single scatter
albedo to date have only been made in the visible because the most common
instrument used to measure light absorption only operates at one wavelength
(~550nm). Recently developed instrumentation will at least allow for measurement of
the wavelength-dependence of ω within the visible range (460, 530 and 660nm), but
we still must rely on models to determine the infrared single scatter albedo for dust.
The index of refraction (in particular the imaginary component) used for dust in
models is highly variable, so the infrared effects of dust are still very uncertain.
Nonetheless, the very high visible ω measured in-situ for the Saharan and Asian dust
do suggest that the top-of-the-atmosphere radiative forcing of dust to be negative in
sign for two of the most significant sources of atmospheric desert dust (Conant et al.,
2003). Here the in-situ measurements of the dust optical properties from ACE-Asia
will be presented, and the accuracy of these measurements discussed. Data from the
NCAR C-130 aircraft are of particular interest because they are vertically-resolved,
they were made using an inlet that efficiently samples coarse mode aerosol, and
because light scattering and absorption for the fine (D<1µm) and coarse modes were
separately determined, allowing for independent determination of the optical
properties of the dust and pollution aerosol. These data will be presented in detail and
compared with similar measurements from land- and ship-based platforms.
References






Anderson, T. L., S. J. Masonis, D. S. Covert, N. C. Ahlquist, S. G. Howell,
A. D. Clarke and C. S. McNaughton (2003): Variability of aerosol optical
properties derived from in situ aircraft measurements during ACE-Asia, J.
Geophys. Res., in press.
Conant, W. C., J. H. Seinfeld, J. Wang, G. R. Carmichael, Y. Tang, I. Uno,
P. Flatau, K. M. Markowicz, and P. K. Quinn, (2003): A model for the
radiative forcing during ACE-Asia derived from CIRPAS Twin Otter and
R/V Ronald H. Brown data and comparison with observations, J. Geophys.
Res., in press.
Kaufman, Y. J., D. Tanre, O. Dubovik. A. Karnieli, L. A. Remer, (2001):
Absorption of sunlight by dust as inferred from satellite and ground-based
remote sensing, Geophys. Res. Lett., 28, 1479-1482.
Haywood, J. M., P. N. Francis, M. D. Glew and J. P. Taylor, (2001): Optical
properties and direct radiative effect of Saharan dust: a case study of two
Saharan outbreaks using data from the U.K. Met. Office C-130, J. Geophys.
Res., 106, 18,417-18,430.
Haywood, J., P. Francis, S. Osborne, M. Glew, N. Loeb, E. Highwood, D.
Tanre, G. Myhre, and P. Formenti, (2003): Radiative properties and direct
radiative effect of Saharan dust measured by the C-130 aircraft during
SHADE: 1. Solar spectrum, J. Geophys. Res., in press.
Myhre, G. and F. Stordal, (2001): Global sensitivity experiments of the
radiative forcing due to mineral aerosols, J. Geophys. Res., 106 (D16),
18,193-18,204.
Incorporating the Effect of Subgrid Scale Circulations Upon Dust Emission
in an AGCM
Reha Cakmur 1,2 and Ron Miller 1,2
1 Department of Applied Physics and Applied Math, Columbia University
2 NASA Goddard Institute for Space Studies, New York, New York
Abstract
Realistic simulation of dust emission in an atmospheric general circulation model (AGCM) is inhibited by the model's coarse resolution
compared to the scale of the circulations observed to mobilize dust. We construct a probability distribution of wind speed within each AGCM
grid box that depends upon the resolved grid box speed and the magnitude of fluctuations. This magnitude is calculated by incorporating
information from the AGCM's parameterizations of the planetary boundary layer along with dry and moist convection. Emission depends on
the fraction of the wind speed distribution above the threshold value. As a consequence of the distribution, emission can occur even if the
resolved wind speed is less than the threshold value as long as the subgrid scale variability is large enough. Subgrid wind fluctuations are
dominated by dry convection. This favors dust emission over deserts, where there is continuous mixing within the boundary layer due to
intense solar heating of the surface. Given our representation of subgrid variability, the AGCM's dust aerosol burden improves significantly,
compared to the TOMS AOT retrieval, over the Sahara/Sahel and the Taklimakan, regions considered to be major sources of dust emission.
This mechanistic representation of subgrid variability allows us to calculate the atmospheric burden of dust under different climates, where
emission can change due to altered boundary layer variability in addition to changes in the mean wind speed.
Capabilities of the Photoacoustic Technique for the Measurement of Light
Absorption by Mineral Dust
Hans Moosmüller and W. Patrick Arnott
Desert Research Institute, University of Nevada System, 2215 Raggio Parkway,
Reno, NV 89512, USA hansm@dri.edu
Abstract
Introduction
Aerosol light absorption makes an important contribution to atmospheric radiative
transfer (Chylek and Wong, 1995) and visibility (Watson, 2002). It causes heating
within an aerosol layer, reduces the amount of light reaching the atmosphere below
and the earth's surface, and perhaps contributes to a lack of cloudiness as clouds are
burned-off by the heat provided in the light absorption of sunlight (Ackerman et al.,
2000; Lohmann and Feichter, 2001). Aerosol light absorption in the atmosphere is
generally dominated by black carbon (BC) particles, which have an absorption
efficiency of about 10 m2/g in the visible, significantly higher than that of other
particles suspended in the atmosphere.
Mineral dust aerosols can also cause significant light absorption, depending on the
particle composition, size distribution, and the wavelength of interest. In contrast to
BC, mineral dust light absorption has a strong wavelength dependence. The
characterization of mineral dust light absorption has similar requirements as that of
BC with the added need for wavelength dependent measurements and for the
sampling of coarse particles. Besides the need of characterizing mineral dust light
absorption for radiative transfer and visibility applications, in depth knowledge of
light absorption at specific wavelengths is also needed for remote sensing
applications such as the use of the TOMS aerosol index (e.g., Alpert et al., 1999; Hsu
et al., 2000).
Measurement of Light Absorption
Aerosol light absorption can be measured with filter techniques or with in situ
techniques. Filter based methods concentrate aerosol particles on a filter substrate and
measure light extinction through the loaded filter, either long after exposure in the
laboratory (e.g., Lin et al., 1973) or in real time (Hansen et al., 1984; Bond et al.,
1999). However, it is well known that the interaction of scattering from the
concentrated aerosol and the filter medium can lead to errors on the order of a factor
of two or three, as the filter instruments lack absolute calibration (Horvath, 1997;
Bond et al., 1999). This situation can be improved by using correction methods
utilizing the simultaneous measurement of aerosol scattering (Horvath, 1997;
Moosmüller et al., 1998). However, an absolute calibration is still needed for filter
methods. The photoacoustic technique (Petzold and Niessner, 1996; Arnott et al.,
1999) is an in situ technique, which measures aerosol light absorption with the
particles in their natural suspended state with an absolute calibration based on first
principles.
Photoacoustic Technique
An alternative to filter-based methods for light absorption measurement is to use a
photoacoustic instrument (Petzold and Niessner, 1996; Arnott et al., 1999) No filters
are used in these instruments, but instead, the sample air is continuously drawn
through an acoustical resonator. A periodically modulated laser beam also passes
through the resonator. Concomitant with light absorption by either gas or particles is
heat transfer to the surrounding air. The resonator can be designed with an acoustic
resonance frequency such that all of the heat from light absorption is transferred
during the acoustic period. Upon receiving this heat, from light absorption, the
surrounding air expands and this expansion contributes to the acoustic standing wave
in the resonator. Measurement is made with a microphone. The microphone signal is
linearly proportional to the aerosol light absorption coefficient. In practice, if one is
seeking to measure particulate light absorption, the choice of laser wavelength is
made to minimize influence of standard atmospheric gases. The typical wavelength
that the Desert Research Institute (DRI) uses for the visible region is 532 nm, where
high efficiency, frequency-doubled, compact ND-YAG lasers are available. In the
near infrared (IR), 1047 nm lasers are used.
The photoacoustic measurement is a zero-based measurement in the sense that no
light absorption corresponds to no microphone signal. By comparison, typical
extinction measurements require careful monitoring of a reference level. The art to
making the photoacoustic method viable in practice is to design the system so that the
influence of ambient and pump acoustic noise are minimized (Arnott et al., 1999).
The theoretical calibration of the DRI photoacoustic instrument has been thoroughly
tested against light absorption measurements on absorbing gases - something that
simply can not be done with a filter based method (Arnott et al., 2000). The
photoacoustic method has been touted as the desired way to obtain aerosol light
absorption (Andreae, 2001).
Recent improvements have yielded a sensitivity corresponding to a light absorption
lower limit of Babs = 0.15 Mm-1 (corresponding to an equivalent BC mass
concentration of 30 ng m-3) for the 1047 nm laser, laser power of 200 mW, and 2
minutes averaging time. This is a factor of 3 better than our previous efforts reported
earlier (Arnott et al., 1999). The same instrument has been used to measure the real
time BC content of diesel vehicle exhaust for heavy-duty diesel trucks on
dynamometer test stands at loadings in excess of 30 mg m-3, indicating an excellent
dynamic range of 106, far exceeding that of filter-based instruments. We have also
succeeded in making the photoacoustic instrument very insensitive to ambient noise
as indicated by airborne operation in a small plane and ground-based operation only
meters away from a fighter jet operating at 80% of maximum power.
Photoacoustic Technique for Mineral Dust Light Absorption
So far the photoacoustic technique has been mainly used for the measurement of BC
light absorption and the light absorption of ambient aerosols dominated by BC. On an
exploratory basis, the photoacoustic technique has been used during a dust
entrainment study near El Paso, TX. Photoacoustic measurements indicated that the
light absorption measured here was several orders of magnitude smaller than the
measured scattering, indicating an aerosol albedo extremely close to one. This
preliminary result is a consequence of the white carbonate particles encountered at
this location. Currently, the DRI photoacoustic instrument is being modified for
seven-wavelength operation with wavelengths ranging from 355 nm in the ultraviolet
(UV) to 1047 nm in the IR. This modification will allow for the determination of the
wavelength dependence of aerosol light absorption, making it possible to distinguish
between BC and several kinds of mineral dust. In addition, we are working on
incorporating a reciprocal nephelometer in the photoacoustic resonator to obtain a
measurement of aerosol albedo with a single instrument.
While BC particles are typically quite small with a typical mass mean diameter of
about 100 nm, mineral dust can contain very large particles. This will make it
necessary to characterize and possibly modify the sampling arrangement of the
photoacoustic instrument for low-loss sampling of large particles.
References

Ackerman, A. S., O. B. Toon, D. E. Stevens, A. J. Heymsfield, V.
Ramanathan, and E. J. Welton (2000). "Reduction of Tropical Cloudiness by
Soot." Science 288, 1042-1047.














Alpert, P., J. Herman, Y. J. Kaufman, and I. Carmona (1999). "Response of
the Climatic Temperature to Dust Forcing, Inferred from Total Ozone
Mapping Spectrometer (TOMS) Aerosol Index and the NASA Assimilation
Model." Atmos. Res. 53, 3-14.
Andreae, M. O. (2001). "The Dark Side of Aerosols." Nature 409, 671-672.
Arnott, W. P., H. Moosmüller, C. F. Rogers, T. Jin, and R. Bruch (1999).
"Photoacoustic Spectrometer for Measuring Light Absorption by Aerosol:
Instrument Description." Atmos. Environ. 33, 2845-2852.
Arnott, W. P., H. Moosmüller, and J. W. Walker (2000). "Nitrogen Dioxide
and Kerosene-Flame Soot Calibration of Photoacoustic Instruments for
Measurement of Light Absorption by Aerosols." Rev. Sci. Instrum. 71,
4545-4552.
Bond, T. C., T. L. Anderson, and D. Campbell (1999). "Calibration and
Intercomparison of Filter-Based Measurements of Visible Light Absorption
by Aerosols." Aerosol Sci. Tech. 30, 582-600.
Chylek, P. and J. Wong (1995). "Effect of Absorbing Aerosol on Global
Radiation Budget." Geophys. Res. Lett. 22, 929-931.
Hansen, A. D. A., H. Rosen, and T. Novakov (1984). "The Aethalometer An Instrument for the Real-Time Measurement of Optical Absorption by
Aerosol Particles." Sci. Total Environ. 36, 191-196.
Horvath, H. (1997). "Experimental Calibration for Aerosol Light Absorption
Measurements Using the Integrating Plate Method - Summary of the Data."
J. Aerosol Sci. 28, 1149-1161.
Hsu, N. C., J. R. Herman, and C. Weaver (2000). "Determination of
Radiative Forcing of Saharan Dust Using Combined TOMS and ERBE
Data." J. Geophys. Res. 105, 20649-20661.
Lin, C. I., M. B. Baker, and R. J. Charlson (1973). "Absorption Coefficient
of Atmospheric Aerosols: A Method for Measurement." Appl. Opt. 12,
1356-1363.
Lohmann, U. and J. Feichter (2001). "Can the Direct and Semi-Direct
Aerosol Effect Compete with the Indirect Effect on a Global Scale."
Geophys. Res. Lett. 28, 159-161.
Moosmüller, H., W. P. Arnott, C. F. Rogers, J. C. Chow, C. A. Frazier, L. E.
Sherman, and D. L. Dietrich (1998). "Photoacoustic and Filter
Measurements Related to Aerosol Light Absorption during the Northern
Front Range Air Quality Study (Colorado 1996/1997)." J. Geophys. Res.
103, 28149-28157.
Petzold, A. and R. Niessner (1996). "Photoacoustic Soot Sensor for in-Situ
Black Carbon Monitoring." Appl. Phys. B 63, 191-197.
Watson, J. G. (2002). "2002 Critical Review -- Visibility: Science and
Regulation." J. Air & Waste Manage. Assoc. 52, 626-713.
Estimation of the contribution of mineral dust to the total aerosol optical
depth: Particular focus on Atlantic Ocean
G. Myhre 1,2, A. Grini 1, T.K. Berntsen 1,3, I.S.A. Isaksen 1
1 Department of Geophysics, University of Oslo, Norway. 2 Norwegian Institute for
Air Research (NILU), Kjeller, Norway. 3 Center for International Climate and
Environmental Research - Oslo (CICERO), Oslo, Norway.
Abstract
Satellite retrievals show that the highest aerosol optical depth (AOD) over ocean is
west of Africa. The magnitude and distribution of the AOD varies significantly
during the year. Further, satellite retrievals have markedly differences in AOD. A
quantification of aerosol sources and aerosol properties is important in this region as
the radiative effect of aerosols in this region can have a magnitude of several tens of
Wm-2. Emission of mineral dust and biomass burning aerosol are particularly large
over the African continent and the aerosols are transported over ocean. In addition sea
salt particles are generated over ocean and contribute to the AOD. The aerosol
properties of these species are quite different, with regard to size and absorption.
With a global aerosol transport model we will estimate the contribution of mineral
dust particles to the total AOD and compare this to an estimate based on satellite
retrievals. The latter estimate will be based on retrievals of Ångstrøm exponent and
sunphotometer measurements. The aim of this study is to understand if there are
particular regions or time periods where one aerosol type dominates the others, but
also to try to understand why there are discrepancies between the model and the
satellites as well as between the different satellite retrievals.
The Role of Shear Stress Partitioning on Dust Emissions
William G. Nickling 1 and John A. Gillies 2
1 Wind Erosion Laboratory, Department of Geography, University of Guelph,
Guelph, Canada
nickling@uoguelph.ca
2 Division of Atmospheric Sciences, Desert Research Institute, Reno, USA
jackg@dri.edu
Abstract
Most currently available aeolian sediment transport models assume a homogeneous
surface with some coverage of relatively small-scale non-erodible roughness
elements. Many natural surfaces from which dust is emitted, have large non-erodible
roughness elements that affect the emission process. These elements can range in size
from centimeters to meters and can be solid or in the case of vegetation, porous.
These roughness elements offer a measure of protection from erosive winds by
several mechanisms including, sheltering, direct covering of the surface, and by
decreasing the available shear force through processes of momentum extraction.
Characterizing and quantifying the effect of momentum partitioning on surface
shearing stress and sand transport, which in large part controls dust emissions by
wind, remains an important research challenge. Several approaches have been
undertaken to account for surface roughness effects in aeolian sediment transport
including the models of Raupach et al. (1993) and Marticorena and Bergametti
(1995).
A major focus of our recent research has been the refinement and modification of the
Raupach et al. (1993) shear stress partitioning model to account for the effect of
larger-scale surface roughness on aeolian sediment transport. The approach we have
taken has combined both laboratory and field experimentation. In these experiments,
we have utilized specially-designed instrumentation that independently and
simultaneously measures the total drag force of the wind as well as the shearing stress
on the individual elements and the intervening surface. The total shear stress is
determined through the measurement of the vertical wind speed profile in both the
wind tunnel and atmospheric boundary layers. In both the laboratory and the field,
newly-designed drag plates and force balances were used to measure the element and
surface borne shear stresses. In addition, Irwin sensors were used to quantify and
characterize the magnitude and spatial distribution of shear stresses on the intervening
surface area.
Laboratory wind tunnel studies were carried out to evaluate roughness element form
and distribution effects on shear stress partitioning for solid elements (e.g., Crawley
and Nickling, 2002), sand drift through porous element arrays, and the aerodynamics
of different plant morphologies to assess porosity effects on shear stress partitioning
(Gillies et al., 2002). These wind tunnel experiments undertaken to evaluate and
constrain the parameters of the Raupach et al. (1993) model, indicated that it provides
a useful approach to predict the role of surface roughness on sediment entrainment
and transport. Our research however, has indicated that components of the model
require further consideration. In particular, the parameter that characterizes the spatial
heterogeneity of shear stress (m) on the intervening surface is not independent and
appears to be a function of the aspect ratio of the roughness elements (Crawley and
Nickling, 2002). Our more recent investigations using cubes of differing sizes and
spacings have indicated that roughness concentration (i.e., λ) does not adequately
characterize the shear stress partitioning effect. We have observed that for 0.1 ≤ λ ≤
0.5, common to many natural surfaces covered with roughness elements (e.g., rocks,
sparse vegetation), shear stress reduction on the intervening surface is not
independent of the height and width of the elements. In general we found that the
height of the element has a greater effect on shear stress reduction than element width
most likely due to the greater momentum absorption by elements protruding further
into the logarithmic boundary layer. Outside the stated range of λ the shear stress
partitioning effect appears to be independent of element shape. It was also found that
configurations with the same λ, but with a greater number of elements, were more
effective in decreasing shear stress on the intervening surface. Our results indicate
that the single parameter λ, which describes size, shape, and distribution of elements
on the surface does not fully account for the observed partitioning of shear stress.
This dimensionless parameter needs to be modified or replaced with a parameter or
parameters that account for specifically the effects of height and number of elements.
Few studies have field evaluated the Raupach et al. (1993) model at full-scale. Fullscale shear stress partitioning measurements in sparsely vegetated desert
environments and on surfaces with different distributions of solid elements have been
undertaken to corroborate the laboratory studies. In one of our field studies large
elements (0.30 m width × 0.36 m high cylinders) were placed in similar
configurations of λ to those in our laboratory experiments using 500 to 2000
roughness elements distributed over an area 2000 m2 on a flat, bare, crusted field. In
this study we also used Irwin sensors to measure the intervening surface shear stress
and force balances to determine the drag on individual elements. In addition, sand
transport through the roughness array was measured with self-orienting traps that
measured the saltation flux at 1 Hz. The results of the field study clearly indicated the
efficacy of the Raupach model for these arrays. We observed that the shear stress on
the intervening decreased proportionally with increasing λ for this single element
shape. It was also found that saltation flux decreased rapidly from the leading edge of
the array reaching equilibrium within a relatively few roughness element rows and
was proportional to the shear stress on the intervening surface. Equilibrium sediment
flux rate measured at the downwind edge of the element array was found to be
proportional to λ.
In arid and semi-arid environments were dust emission occurs, large-scale roughness
elements are frequently grasses, shrubs, and bushes. Our research has indicated that
these flexible and porous elements have different aerodynamic properties and
behaviors in airflow than solid elements of similar size and shape. Gillies et al. (2000,
2002) observed that vegetation typically has a higher drag coefficient than a solid
element of similar size. Higher drag coefficients indicate that vegetation absorbs
greater amounts of momentum than similar sized solid elements resulting in greater
protection of bare intervening surfaces. Gillies et al. (2002) also observed non-linear
drag coefficient behavior with increasing Reynolds number in shrubs and grasses,
which suggests that the ratio of total stress to the stress on the intervening surface is
also dependent on Reynolds number. This was especially true for grass, which
continually lowers its drag coefficient in response to increasing wind speeds. In the
future we plan to incorporate the effects of porosity into the Raupach et al. (1993)
shear stress partitioning model.
Shear stress partitioning is a common physical process in most natural environments
and must be accounted for in any dust emission model. From recent research it is
clear that shear stress partitioning is a complex, multivariate, non-linear process,
which is further complicated for sparsely vegetated surface where the non-linear
aerodynamic behavior of vegetation may have to be considered. Currently available
dust emission models do not adequately account for these effects on the aeolian
sediment transport system. It is hoped that our recent investigations into the
characterization of the physical controls on this process will provide a basis for
improving models of the dust emission process.
References





Crawley, D. and Nickling, W.G., 2002. Drag partition for regularly arrayed
rough surfaces. Boundary-Layer Meteorology, Vol. 10, 445-468.
Gillies, J.A., N. Lancaster, W.G. Nickling, and D. Crawley (2000). Field
determination of drag force and shear stress partitioning effects for a desert
shrub (Greasewood, Sarcobatus vermiculatus), J. Geophys. Res.,, 105, 24,
871-24,880.
Gillies, J.A., Nickling, W.G. and King, J., 2002. Drag coefficients and plant
form response in three plant species: Burning Bush (Euonymus alatus),
Colorado Spruce (Picea pungens glauca.) and Fountain Grass (Pennisetum
setaceum), J. Geophys. Res., 107, D 24, 10-1 - 15-1.
Marticorena, B. and G. Bergametti (1995). Modeling the atmospheric dust
cycle: 1 Design of a soil derived dust emission scheme, J. Geophys. Res.,
100, 16415-16430.
Raupach, M.R., D.A. Gillette, and J.F. Leys (1993). The effect of roughness
elements on wind erosion threshold, J. Geophys. Res., 98, 3023-3029.
DREAM Dust Model: Ongoing and Future Developments
S. Nickovic
Euro-Mediterranean Centre on Insular Coastal Dynamics (ICoD,) Malta
Abstract
The desert dust model DREAM (Nickovic at al., 2001) is a component of the ICoD
environmental modelling system that integrates different environmental modules.
This integrated system couples the atmosphere, ocean, soil and aerosol processes and
provides different linkages between them. Whenever possible, mutual and
simultaneous interactions of various natural environments is established. Currently,
the following two-way couplings are already part of the integrated model: atmosphere
<–> dust, dust <–> soil, atmosphere <–> ocean and atmosphere <–> soil. In the
future, dust <–> ocean component and possibly some others are to be incorporated.
Technically, the integrated system is designed so that a) each environmental module
of the system is developed as a callable routine called by the major driver - the
atmospheric model, and b) any of the mentioned interactions is performed on-line
within the course of the integrated model execution.
The ongoing and future developments of DREAM are to a large extent motivated to
further improve simulation of different linkages between components of the ICoD
integrated modelling system. In the recent study (Nickovic 2002; Nickovic, Özsoy,
Pejanovic, unpublished), the dust aerosol is linked to the atmospheric radiation as online module. Both short- and long-wave radiation spectra interconnect with the
multiple-particle dust concentration simulated by DREAM. Preliminary results
indicate that during major dust storms over the Sahara/Mediterranean region dust may
reduce the lower atmosphere temperatures by several degrees. A slight increase of the
temperature of less than a degree is simulated in the upper atmosphere. To improve
the atmosphere <–> dust linkages in the model, an increase from 4 to 8 particle sizes
is planned in the near future.
Further improvements of the dust <–> soil interactions are based on replacement of
the previous global 1/6-degree topography (US Navy) and vegetation data with the
corresponding new global USGS 1-km data sets. The next phase of developments will
also include seasonal changes of the vegetation cover and tests with refined dust
production schemes (e.g. Shao and Li, 2002).
Next DREAM developments anticipate extensive use of lidar dust profile
observations. The EARLINET (European Lidar Network) containing more than 20
stations over the Euro-Mediterranean region is close to become operational in the
recent future. The dust modelling community for the first time has an opportunity to
better validate 3D structure of dust concentration by combining lidar data with other
available data (e. g. satellites, sun-photometers) (Ansman et al., 2003, submitted to
JGR). In addition to validation, lidar data could be also used for data assimilation in
dust models; preparations to incorporate lidar profiles in DREAM are ongoing in
order to further improve quality of dust forecasting operations. Finally, use of
DREAM results in regional climate scenario experiments (especially its dustradiation interaction component) within the EU ProMed Initiative could further refine
understanding of the role of aerosols in climate variability. Incorporation of dust
cloud physics and dust atmospheric chemistry linkages as planned in the Initiative
could substantially contribute to better understanding dust impacts to different
environmental systems.
SUMMARY OF SIZE-RESOLVED MINERAL DUST
MEASUREMENTS FROM THE ACE-ASIA (2001) AND ITCT (2002)
EXPERIMENTS
Kevin D. Perry
Meteorology Department, University of Utah 135 S 1460 E, Rm 819, Salt Lake City,
UT 84112-0110, USA
Abstract
Measurements of the size-resolved aerosol elemental composition were made at sites
in China, Taiwan, Japan, Korea, Alaska, Hawaii, and the west coast of the United
States in support of the ACE-Asia and ITCT experiments. The purpose of these
measurements was to characterize Asian aerosol near the source regions, downwind
of the source regions, and at distant receptor sites. By using identical instrumentation
and analytical techniques, both the spatial patterns of the size-resolved aerosol
composition and the net effect of physical and chemical transformations that occurred
during transport have been documented. During these experiments, the aerosols were
collected on greased-Mylar substrates using a combination of 8-stage and 3-stage
rotating drum impactors. The impaction substrates were slowly rotated under the
slotted orifices for a 6-week period to preserve a continuous record of the sizeresolved aerosol. At the conclusion of the experiments, the substrates were analyzed
by synchrotron X-ray fluorescence (S-XRF) using a microprobe at beamline 10.3.1 of
the Advanced Light Source (Lawrence Berkeley National Laboratory). The S-XRF
analysis yielded quantitative measurements of the size-resolved aerosol elemental
composition with 3-hour time resolution for elements ranging from sodium through
uranium.
Simultaneous measurements were made during ACE-Asia at two sites in China. The
Beijing (116.4°E, 39.9°N, 55 MSL) and Hefei (117.2°E, 31.9°N, 35 MSL) sites were
situated along an approximate north-south line but were sufficiently far apart to show
the regional nature of the dust events. During the ACE-Asia sampling period
(March/April 2001) the frequency and severity of the dust episodes at Beijing were
about twice those observed at the more southerly Hefei site. The mineral dust mass in
Beijing was dominated by particles with aerodynamic diameters > 5 µm while the
dust mass in Hefei peaked in the 2.5 - 5 µm size range. The higher concentrations of
coarse dust in Beijing resulted from its closer proximity to a mineral dust source
region. Although regional dust episodes were observed, Positive Matrix Factorization
(PMF) was able to identify multiple dust sources based upon temporal variations and
differences in chemical composition as a function of particle size. In particular, the
PMF analysis revealed a calcium-rich source, an iron-rich source, and a source that
closely mirrors the average crustal. The calcium- and iron-rich sources were observed
over all size ranges, but were most apparent in the coarse and fine fractions,
respectively. The 3-hour resolution of the data also revealed a distinct diurnal cycle of
mineral dust concentrations within the boundary layer. Maximum dust concentrations
at the surface were observed just before sunrise while minimum dust concentrations
were observed in the late afternoon. This pattern most likely resulted from changes in
the boundary layer (BL) depth. Because the BL grows by entrainment of air from the
free atmosphere, the decrease in surface concentrations during the day indicates that
on average, mineral dust concentrations immediately above the BL are less than those
within the BL. This result is consistent with LIDAR measurements in China which
tend to show large dust concentrations within the BL with several distinct dust layers
at higher elevations. Measurements made at Gosan (126.2°E, 33.3°N, 78 MSL) and
Mt. Hallasan (126.5°E, 33.4°N, 1100 MSL), Korea during ACE-Asia showed that
this pattern of high concentrations within the BL and lower concentrations in the free
atmosphere existed downwind from the Asian continent as well. On average, the
mineral dust concentrations at Mt. Hallasan were <40% of those at Gosan. These
measurements indicate that the low- and high-altitude dust layers are generally
decoupled from each other.
Measurements made along the west coast of the United States during ACE-Asia and
ITCT documented several examples of long-range transport (LRT) of mineral dust
from Asia. These episodes were most apparent at high elevation sites in the Cascade
and Sierra Nevada Mountain ranges, but were also identifiable at the Trinidad Head
supersite located along the California coast (124.2°W, 41.1°N, 107 MSL). Dust
concentrations in the BL were <25% of those observed at the higher elevation sites.
The mineral dust concentrations at the Trinidad Head site exhibited a diurnal pattern
with the exact opposite phase of those observed immediately downwind of Asia. In
this case, dust concentrations in the BL were highest in the afternoon when
entrainment of air from the free atmosphere was maximized. The maximum
concentrations of mineral dust during the LRT episodes were typically observed in
the 1.15 - 2.5 µm diameter size range. However, significant quantities of mineral dust
were still observed in the 2.5 - 5 µm and 5 - 10 µm size ranges during several of the
episodes. As expected, the coarse dust was maximized when the meteorological
conditions favored rapid transport (i.e., < 1 week).
Simultaneous measurements of the size-resolved aerosol elemental composition at
ambient and dry relative humidity (RH) conditions (i.e., <55% RH) were also made at
the Trinidad Head site to determine if the size of the mineral dust was affected by
changes in RH after undergoing LRT from Asia. The results of this study indicate that
the Asian dust is somewhat hygroscopic upon arrival at the west coast of the United
States. In all LRT episodes, the mineral dust mass distribution shifted to smaller sizes
as the RH was reduced. Since fresh mineral dust near the source is unlikely to exhibit
this behavior, it is likely that the mineral dust was coated with a more hygroscopic
species during transport across the Asian continent or transport across the Pacific
Ocean. While the exact nature of this hygroscopic coating cannot be determined from
the S-XRF analysis, measurements made other scientists near the Asian continent
during ACE-Asia showed that nitrate was primarily confined to particles with
diameters > 1 µm while sulfate was primarily confined to submicron particles. Thus,
particulate nitrate is the most likely candidate species responsible for increasing the
hygroscopicity of the aged Asian dust.
MODIS Retrieval of Dust Aerosol
Lorraine A. Remer 1, Yoram J. Kaufman 1, Didier Tanré 2
1 Laboratory for Atmospheres, NASA/Goddard Space Flight Center, Greenbelt MD,
USA
2 Laboratoire d'Optique Atmospherique, Universite de Lille 1, Lille, France
Abstract
The MODerate resolution Imaging Spectroradiometer (MODIS) currently aboard
both the Terra and Aqua satellites produces a suite of products designed to
characterize global aerosol distribution, optical thickness and particle size. Never
before has a space-borne instrument been able to provide such detailed information,
operationally, on a nearly global basis every day. The three years of Terra-MODIS
data have been validated by comparing with co-located AERONET observations of
aerosol optical thickness and derivations of aerosol size parameters. Some 8000
comparison points located at 133 AERONET sites around the globe show that the
MODIS aerosol optical thickness retrievals are accurate to within the pre-launch
expectations. However, the validation in regions dominated by desert dust is less
accurate than in regions dominated by fine mode aerosol or background marine sea
salt. The discrepancy is most apparent in retrievals of aerosol size parameters over
ocean. In dust situations, the MODIS algorithm tends to under predict particle size
because the reflectances at top of atmosphere measured by MODIS exhibit the
stronger spectral signature expected by smaller particles. This pattern is consistent
with the angular and spectral signature of non-spherical particles. All possible aerosol
models in the MODIS Look-Up Tables were constructed from Mie theory, assuming
a spherical shape. Using a combination of MODIS and AERONET observations, in
regimes dominated by desert dust, we construct phase functions, empirically, with no
assumption of particle shape. These new phase functions are introduced into the
MODIS algorithm, in lieu of the original options for large dust-like particles. The
results will be analyzed and examined.
Adsorption of Semivolatile Organic Compounds Pertinent to Urban
Environments onto Mineral Dust Particles
Y. Rudich 1, A. H. Falkovich 1, G. Schkolnik 1, E. Ganor 2
1 Department of Environmental Sciences, Weizmann Institute, Rehovot 76100 Israel
2 Department of Geophysics and Planetary Sciences, Tel Aviv University, Tel Aviv,
Israel
Abstract
INTRODUCTION
The interaction of mineral dust particles from the Sahara with semi-volatile organic
compounds (SVOC) over an urban region in the Israeli costal plain was studied by
collecting dust samples in more than 15 dust storms during spring 2001, under
varying meteorological conditions. Organic compounds adsorbed on collected
mineral dust particles were analyzed in an integrated, multi-technique study which
employed scanning electron microscope equipped with energy dispersion system
(SEM-EDS) and bulk aerosol analysis consisting of gas chromatography / mass
spectrometry (GC/MS) and ion chromatography (IC). The SEM-EDS analysis
exemplifies the co-existence of inorganic and organic species on individual mineral
dust particles. Using the GC/MS and IC analysis, specific tracers for urban air
pollution and photodegradation products of agriculture emissions have been identified
and their size distributions were obtained. Redistribution of semi-volatile organics
such as polycyclic aromatic hydrocarbons (PAH) and pesticides from the sub-micron
to larger particle size fractions, governed by the mineral dust transport trajectory and
size distributions, are observed. Non-volatile species, such as anhydrous sugars and
large PAH, do not redistribute between the phases due to their low vapor pressure.
The concentrations of short chain carboxylic acids increased with higher ambient
relative humidity, suggesting water-assisted uptake onto the mineral particles.
RESULTS
In this study we characterized organic species adsorbed onto mineral dust particles
during a series of dust storms using several analytical methods. The analyzed
compounds represent several classes of compounds pertinent to urban and polluted
environments. Therefore this study can be regarded as a survey for the ability of dust
to transport such species in the atmosphere. The multi-technique approach employed
here reduces sampling artifacts and help generalize the observed phenomena.
Individual particle analysis with SEM-EDS using uncoated conductive supports (Si or
Be plates) enables the direct detection of organic carbon and demonstrated the copresence of inorganic and organic species on mineral dust particles. The analysis
shows that most of the particles have a complex nature. They are aggregates of
various minerals and with a wide variety of different organic compounds adsorbed on
their surface. The organic carbon may be distinguished from soot and carbonate. The
SEM analysis allows estimating the size range of the particles containing organics. It
shows that in many case the organics do not cover the whole particle surface and are
spread in a heterogeneous manner. Such information cannot be obtained by the bulk
analytical methods. Using GC/MS and ion chromatography, the concentrations of
specific organic tracers was determined in size-segregated samples.
The following characteristic features of the interaction between mineral dust and
organic compounds were determined by the chromatographic techniques:
1.Mineral dust particles adsorb polar and non-polar organics (e.g. PAH, oxy-PAH,
pesticides and carboxylic acids) leading to higher aerosol-phase concentrations of
these species compared to the concentrations found in ambient aerosol collected in
the same sites on dust-free days.
2.The gas-particle redistribution of organic compounds during the dust storm is
influenced by their vapor pressure. More volatile organics are more affected by the
presence of dust and react more quickly to the changing aerosols loading. Many
organics that are usually found in fine aerosol mode are found in the larger aerosol
sizes typical of dust.
3.The interaction between water-soluble species such as carboxylic acids and mineral
particles is affected by relative humidity as well as by the dust concentration. This is
in contrast to hydrophobic compounds such as PAHs that are more affected by the
dust surface area distribution alone.
4.Weakly polar and non-polar organics adsorbed on mineral particles exhibit several
trends. If there is permanent emission of a species' (e.g. pesticides), then the higher
the dust concentration, the higher is the species consecration. If time is required to
build a steady state gas phase concentration of organics, the first "wave" of the dust
storm absorbs the ambient organics, and their particle content will subsequently
decrease. The phenomenon suggests that a dust storm "cleans" the atmosphere from
these organic species.
Particle-associated organics concentration increase in dust events and long-range
transport of pollutants via the particles can be very efficient. This occurs both in the
small and large size fractions. Therefore, this study suggests that dust can be an
efficient medium for transport of organic material (pollutants, natural emissions) and
deposition in the troposphere.
Dust Emission, Dust Emission Scheme and Dust Storm Prediction
Yaping Shao
Department of Physics and Materials Science City University of Hong Kong, Hong
Kong SAR, PRC email: apyshao@cityu.edu.hk
Abstract
The mechanisms for dust emission will be examined by considering the aerodynamic, gravitational and inter-particle cohesive forces acting on soil
particles. The balance of these forces determines the entrainment of particles into the atmosphere. The entrainment mechanisms for sand and dust differ
because of the relative importance of these forces. Threshold friction velocity for dust particles is a stochastic variable that satisfies certain probabilistic
distributions. While it is meaningful to define the threshold friction velocity as a single value for sand, it is not so for dust.
Three mechanisms for dust emission can be identified: (1) aerodynamic entrainment: dust particles can be lifted from the surface directly by
aerodynamic forces. However, dust emission arising from direct aerodynamic lift is in general small; (2) saltation bombardment: as sand-sized particles
strike the surface, they cause localized impacts which overcome the binding forces on dust particles, leading to dust emission; and (3) aggregates
disintegration: dust particles often exist as dustcoats attached to sand grains in sandy soils or as aggregates in soils with high clay content. During weak
wind erosion, sand particles coated with dusts and clay aggregates behave as individuals and dust particles may not be released, while during strong
wind erosion, dustcoats and soil aggregates disintegrate resulting in dust emission.
A dust emission scheme is described which is designed to estimate (1) friction velocity; (2) threshold friction velocity; (3) sand drift intensity; and (4)
dust emission rate for various particle size groups. The dust scheme has been implemented in the integrated wind erosion modeling system (IWEMS).
IWEMS is used for 24, 48 and 72hr forecasts of Northeast Asian dust events for March and April, 2002. The predictions are validated with synoptic
records from the meteorological network and dust concentration measurements at 12 stations in China, Japan and Korea. The validation confirms the
capacity of the modeling system in quantitative forecasting of dust events in real time. Dust emission, deposition and load in Northeast Asia are
determined. During an individual dust episode, dust sources and intensities vary in space and time, but on average the Gobi Desert, the Hexi Corridor,
the Chaidam Basin, the Tulufan Basin and the fringes of the Talimu and Zhunge'er Basins are the main source regions. The Gobi is the strongest source
of dust emission, where the maximum dust emission reaches 5000 µg m-2 s-1 and the net dust emission reaches 16 tn km-2 day-1 in March and April,
2002.
Recent advances and remaining challenges in predicting radiative
properties of mineral dust
Irina N. Sokolik
Program in Atmospheric and Oceanic Sciences (PAOS) University of Colorado at
Boulder
E-mail: irina.sokolik@colorado.edu
Abstract
Radiation/climate modeling and remote sensing both heavily rely on the ability to
accurately model the optical properties of dust in a wide spectral range from the UV
to the IR. In my talk I will briefly review recent advances in modeling main optical
characteristics (such as spectral extinction and absorption coefficients, single
scattering albedo, and scattering phase function) of mineral aerosols, as well as
highlight emerging challenges.
Recent field measurements (e.g., PRIDE and ACE-Asia) and laboratory studies
confirm that dust particles have complex shapes and composition that vary with
particle size. This complexity renders dust optical properties difficult to model.
Kalashnikova and Sokolik (2002, 2003) proposed a method to compute the optical
properties of nonspherical particles by applying the discrete dipole approximation
method to representative composition-shape-size (CSS) distributions reconstructed
from the electron microscopy data of several atmospheric dust samples. Modeling
results revealed various differences between optics of CSS distributions and those of
volume-equivalent spheres and spheroids. These differences are sufficiently large as
to affect the radiative properties of dust important to both radiation/climate modeling
and remote sensing applications. However, the above approach relies on several
critical assumptions largely due to the inability of electron microscopy (SEM and
TEM) to provide a complete set of input parameters required for optical calculations.
Sokolik et al. (2002) investigated whether the combination of electron microscopy
data and aerosol time-of-flight mass spectroscopy could provide additional constrains
for optics modeling. An integrated analysis of the aerosol time-of-flight mass
spectroscopy and electronic microscopy data collected from the Ron Brown ship
during ACE-Asia was performed to identify the composition and morphology of dust
particle as a function of size in the clean and polluted marine conditions under
varying dust loading. The analysis revealed that mineral dust particles in the MBL
were often internally mixed with other chemical species, forming complex
multicomponent aerosols containing dust (MCA-D). This further complicates
prediction of dust radiative properties. I will highlight the most critical assumptions
used in both methods and suggest the measurements needed to constrain the models.
It is important to point out that new instrumentation to measure optical properties
(especially, spectral single scattering albedo and scattering phase function) of dust
and complex internally mixed aerosols is urgently needed.
One of the problems is that the experimental data covering the microphysical
properties of atmospheric dust particles is limited. Ultimately, as new data will
become available, a statistical analysis of a large number of dust samples would be
required to determine and classify the morphology and composition as a function of
size. From the other hand, analysis of a large number of particles would require a new
theoretical framework. I will discuss several feasible approaches for classification of
dust particles (based on particle concavity, fractal dimension and a Fourier analysis of
the shape-wave).
The second part of my talk will be focused on the spectral radiative signature of
atmospheric dust and MCA-D and implications to radiation/climate modeling and
interpretation of data acquired with satellite narrowband and high spectral resolution
sensors. Since both climate studies (e.g., IPCC) and remote sensing retrievals treat
atmospheric aerosols as an external mixture of distinct aerosol types (such as dust,
sulfates, black carbon), it is important to determine the magnitude of radiative
impacts of MCA-D.
Sokolik (2002) found that mineral dust has a unique radiative signature in the IR
(termed a "negative slope") which separates the effect of dust from that of clouds and
gases. Our analysis revealed that narrowband sensors (e.g., MODIS, AVHRR,
GOES) have different sensitivity to dust composition depending on a particular
channel. We conclude that narrowband satellite sensors are capable of detecting dust
but the quantitative characterization of dust properties requires the high spectral
resolution observations. I will demonstrate that the high spectral resolution remote
sensing provides the unique capability for constraining the composition of windblown atmospheric dust. I will also present some results of our ongoing work towards
developing the new generation of atmospheric aerosol models for high spectral
resolution remote sensing.
Summarizing, I would like to stress several outstanding problems which need to be
addressed to develop improved dust models for remote sensing applications and
climate predictions: the need for a better understanding of key physical and chemical
properties of mineral dust and processes that control spatial and temporal variability
of dust in the atmosphere; and the need for new innovative techniques for integrating
various data (satellite, in situ, ground-based, lab) and modeling at various spatial and
temporal scales.
References




Kalashnikova, O., and I.N. Sokolik, Importance of shapes and composition
of wind-blown dust particles for remote sensing at solar wavelengths.
Geophys. Res. Let., 29, No.10, 10.1029/2002GL014947, 2002.
Kalashnikova, O., and I.N. Sokolik, Modeling optical properties of
nonspherical soil-derived dust aggregates. J. Quant. Spectrosc. Radiative
Transfer, 2003, (in review).
Sokolik I.N., The spectral radiative signature of wind-blown mineral dust:
Implications for remote sensing in the thermal IR region. Geophys. Res.
Let., 29, 10.1029/2002GL015910, 2002.
Sokolik, I.N., J. Anderson, , S. A. Guazzotti, D. A. Sodeman, and K. A.
Prather, The radiative impacts of multicomponent aerosols containing dust
(MCA-D) over the ACE-Asia study domain. EOS Trans. AGU, Fall
Meeting, Suppl. 2002.
Measurement and modeling of the Saharan dust radiative impact: Overview
of the SaHAran Dust Experiment (SHADE).
D. Tanré 1, J. Haywood 2, J. Pelon 3, J.F. Léon 1, B. Chatenet 4, P. Formenti 5, P.
Francis 2, P. Goloub 1, E.J. Highwood 6, G. Myhre 7, F. Lavenu et A. Diallo
1- Laboratoire d'Optique Atmosphérique, Centre National de la Recherche
Scientifique et Université des Sciences et Technologies de Lille, Villeneuve d'Ascq,
France
2- Met Office, Bracknell, UK.
3- Service d'Aéronomie, Centre National de le Recherche Scientifique, Université
Pierre et Marie Curie, France
4- Laboratoire Inter-universitaire des Systèmes Atmosphériques, Centre National de
le Recherche Scientifique, Université Paris 12, France
5- Max Planck Institute for Chemistry, Mainz, Germany.
6- Department of Meteorology, University of Reading, Reading, UK.
7- Department of Geophysics, University of Oslo, Norway
Abstract
Recent studies have shown that a significant proportion of mineral dust in the
atmosphere may be of anthropogenic origin and therefore they may have an important
role in climate change by exerting a significant radiative forcing. However, the
optical and radiative properties of dust are not very well determined and SHADE was
designed to better determine the parameters that are relevant for computing the direct
radiative effect.
The experiment took place in September 2000 off the coast of west Africa near the
Cape Verde Islands. During the experiment, the UK C-130 aircraft flew below,
within, and above dust layers for measuring the in-situ physical, chemical, and
radiative properties and remotely sensed the aerosols using radiation equipment. The
French Mystere20 aircraft flew above the aerosol layer and made remote sensing
measurements during MODIS daytime overpasses with a combined radiation equ
ipment, the airborne POLDER simulator and the backscatter Lidar LEANDRE.
Ground-based sunphotometers were deployed in Sal Island of the Cape Verde
archipelago and in M'Bour, 80 km south of Dakar, Sénégal.
The dust physical and optical properties were derived with concurrent measurements
of the radiances and irradiances within the solar and the terrestrial spectrum. A good
consistency between the various measurements and the results of the modelling is
observed. It validates both the modelling and the dust properties on which the
computations of the Chemical Tranport Model are based. Results from SHADE
strongly suggest that the mineral dust has a cooling effect and the model estimates a
global net DRE of the Saharan dust of -0.4Wm-2. Concerning the use of satellite for
aerosol remote sensing, the new sensors explicitly designed for that objective are
efficient. The combination of passive and active observations is very powerful. The
location of the aerosol layer within the atmospheric column can be determined and
properties like the effective radius can be derived as a function of the altitude, which
is very important to understand the indirect aerosol effect.
Modelling Global Dust Emission Trends
Ina Tegen and Martin Werner
Max Planck Institute for Biogeochemistry, Jena, Germany itegen@bgc-jena.mpg.de
Abstract
Soil dust aerosol is an integral part of the vegetation-climate system. It is suspected to
have large impacts on the atmospheric radiation balance, on marine, and possibly
terrestrial biological productivity. The magnitude and distribution of atmospheric soil
dust is strongly controlled by dust emissions, which depend on the extent and type of
terrestrial vegetation and land use, as well as on soil properties and meteorological
variables. Assessment of the role of dust in a climate change scenario requires
understanding of the controls on dust emission, including the role of changes in
anthropogenic land use.
We estimated the influence of land-use change on dust emissions with an offline
model (Tegen et al., 2002), comparing model results for natural and agricultural areas
from two independent historical datasets with a compilation of dust storm data
(Engelstädter et al., 2002). The model takes into account the dependence of dust
emissions on wind speed and soil parameters, as well as vegetation type and cover,
and the presence of preferential "hot spot" sources. Major uncertainties in the
computation of dust emissions are caused by the absence of global data on soil
properties like the particle size distribution and surface crusting, in particular in
regions that are predicted to be preferential dust sources (topographic depressions).
We find that for modern conditions, less than 10% of global dust emissions originate
from cropland or rangeland areas, which is considerably lower than the previous
estimates of up to 50% contribution of anthropogenic soil sources to modern dust
emissions (IPCC, 2001). The offline model was used to estimate future changes in
dust emissions for IPCC scenarios with increased greenhouse gas concentrations,
using meteorological fields extracted from future scenario results of the ECHAM4
and HADCM3 models, and taking into account expected changes in vegetation and
land use to compute dust emissions. We find that expected future changes in
meteorological parameters and changes in natural vegetation cover as consequence of
increased greenhouse gases have a stronger influence on dust emissions than changes
in cultivation. The estimates of future changes in dust emissions are strongly model
dependent, and range from a 9% decrease to a 19% increase in global annual
emissions for the different model scenarios. Apart from these discrepancies, which
are the result of cancellation of positive and negative changes in dust emissions in
different parts of the world, we find common features in the different model
scenarios. For example, in both models increasing dust emissions in the northern
Sahara are caused by stronger winds, and decrease of dust emissions in North
America occurs because of an increase in vegetation cover, and therefore a decrease
in source areas. The reliability of such future estimates depends to a large degree on
the performance of the climate models that are used to produce the meteorological
fields that drive dust emissions.
Magnitudes and patterns of dust emission, transport and deposition are also computed
online in the ECHAM5 model. The vegetation parameters controlling the extent of
dust sources are currently prescribed by satellite-derived vegetation parameters, but
will be replaced by vegetation parameters that are computed on-line in the near
future. Tests of the sensitivity of the dust parameters on model resolution revealed
that the dust emission model performed well with T63 resolution, when compared to
the higher T106 resolution. The simulated dust concentrations and deposition fluxes
of several model years are checked against in-situ observations. This model will
ultimately be used to estimate recent changes in dust radiative forcing, and climate
response, caused by changes in climate conditions, changes in natural vegetation and
land use changes.
References



Engelstaedter, S., K.E. Kohfeld, I. Tegen, S.P. Harrison., 2003. Controls of
dust emissions by vegetation and topographic depressions: an evaluation
using dust storm frequency data. Geophysical Research Letters, 30(6), 1294,
doi:10.1029/2002GL016471.
Tegen, I., S.P. Harrison, K. Kohfeld, I.C. Prentice, M. Coe and M. Heimann,
2002. The impact of vegetation and preferential source areas on global dust
aerosol: Results from a model study. Journal of Geophysical Research 107
(D21), 4576 DOI: 10.1029/2001JD000963.
IPCC, 2001, Climate Change 2001: The Scientific Basis. Contribution of
Working Group I to the Third Assessment Report of the Intergovernmental
Panel on Climate Change, J. T. Houghton, et al. (Eds), Cambridge Univ.
Press, Cambridge, USA.
The detection and characterization of mineral dust from space-based nearUV measurements: recent developments
Omar Torres
University of Maryland, Baltimore County, Baltimore, Md, 21250 NASA Goddard
Space Flight Center, Greenbelt, Md, 20771
Abstract
Since the first workshop on mineral dust, a great deal of progress has been made on
the use of TOMS near UV measurements for the detection and characterization of
mineral dust. The following topics will be briefly discussed.
Near UV aerosol products
In addition to the well-known Aerosol Index (AI) product, TOMS near-UV
observations (330-380 nm) have been used to derive the optical depth (OD) and the
single scattering albedo (SSA) of the atmospheric aerosol load [Torres et al., 2002].
Because of the large sensitivity to UV-absorbing aerosols, and the capability to
observe aerosols over arid and semi-arid environments, the TOMS aerosol products is
a particularly useful tool in studies of windblown mineral dust.
Recently, a new version of the OD and SSA algorithm has been applied to the TOMS
observations. The most significant improvement of the new algorithm is the use of a
realistic representation of the particle size distribution (PSD) of mineral dust,
carbonaceous and sulfate aerosols, in the calculation of the inversion algorithm lookup tables. The new PSD's are based on multi-year statistics from AERONET
observations. Other important changes include improved cloud mask, and radiometric
calibration update based on the version 8 of the TOMS Total Ozone Algorithm.
The long-term record on near-UV aerosol properties will continue with the launch of
the Ozone Monitoring Instrument (OMI) on the AQUA satellite (January 2004). The
OMI sensor is a hyper-spectral (270-500 nm) instrument, with a footprint of about
13x24 km2. An aerosol inversion scheme similar to the one in the TOMS algorithm
will be applied to OMI observations. Measurements in the visible part of the
spectrum will be used to differentiate between absorbing aerosols types (i.e., dust
from smoke), which is one of the TOMS aerosol algorithm limitations.
Imaginary Component of the Refractive Index of Saharan Dust in the near-UV
During the First International Dust Workshop in 1999, the issue of the optical
properties of dust was identified as a high priority research need. Results of analysis
using remote sensing observations [Kaufman et al, 2001] show that Saharan dust is
not as absorbing in the visible as previously assumed based on the refractive index
data of Patterson el at [1977]. The first version of the TOMS aerosol algorithm used
the data of Patterson et al [1977] to characterize dust absorption in the near UV.
Because of the lower (than Patterson's) values found in the visible spectral region, it
became necessary to examine the validity of the assumption on the near UV values,
which are also taken from Patterson's work.
Two separate studies, both involving TOMS observations, were carried out to infer
the imaginary refractive index of Saharan dust in the near UV. The first one [Colarco
et al, 2002], used the TOMS Aerosol Index in conjunction with a transport model that
calculates dust vertical and particle size distributions to obtain the Aerosol Optical
Depth. Comparing the modeled AOD/AI ratio to the observed relationship, the best
refractive index was determined at each of several locations in the analysis, during
the ACE-2 field experiment. The obtained refractive index at 360 nm was about 70%
lower than the Patterson et al [1977] value. The second analysis [Sinyuk et al, 2003]
used TOMS measured radiances collocated in time and space with several
AERONET sites. The obtained results are similar to the ones by the Colarco et al
[2002] analysis.
Sensitivity of TOMS Aerosol Index to altitude of mineral dust aerosols
The TOMS Aerosol Index is a measure of the difference in spectral contrast between
observed and calculated radiances in the near UV. The main source of spectral
contrast variability in the Aerosol Index is the aerosol absorption of Rayleigh
scattering. Because of the strong dependence of molecular scattering on atmospheric
pressure, the sensitivity of the AI to absorbing aerosols goes down rapidly as the
aerosol layer altitude above the surface decreases. Thus, for gray aerosols (i.e.,
spectrally independent imaginary part of refractive index), the AI sensitivity to
aerosol absorption is lost, when the aerosol layer is at an altitude lower than about 2
km. The spectral dependence of the imaginary part of refractive index associated with
colored aerosols, such as mineral dust, constitutes an additional source of spectral
contrast change that contributes to the Aerosol Index. Since this source of spectral
contrast variability does not depend on the height of the aerosol layer, mineral dust
aerosols are detected by the AI regardless of the aerosol layer height.
References





Colarco P.R., O. B. Toon, O. Torres and P. J. Rasch, Determining the UV
imaginary index of refraction of Saharan dust particles from TOMS data
using a three dimensional model of dust transport, J.Geophys. Res, 107,
10.129/2001JD000903, 2002
Kaufman, Y.D., et al, Absorption of sunlight by dust as inferred from
satellite and ground based measurements, Geophys. Res. Lett., 28, 14791482, 2001
Patterson, E.M., D.A. Gillete, and B.H. Stockton, Complex index of
refraction between 300 and 700 nm for Sharan aerosol, J. Geophys. Res, 82,
3153-3160, 1977
Sinyuk A., O. Torres, and O. Dubovik, Imaginary refractive index of desert
dust using satellite and surface observations, Geophys. Res. Lett., 30 (2),
1081, doi: 10.1029/2002GL016189, 2003.
Torres, O., P.K. Bhartia, J.R. Herman, A. Syniuk, P. Ginoux, and B. Holben,
A long term record of aerosol optical depth from TOMS observations and
comparison to AERONET measurements, J. Atm. Sci., 59, 398-413, 2002
Numerical Analysis of Inter-Annual Variation of Dust Emission and
Transport in East Asia
Itsushi UNO 1, Yukari HARA 2, Sinsuke SATAKE 2 and Zifa WANG 3
1 Research Institute for Applied Mechanics, Kyushu University, Fukuoka Japan
2 Graduate School of Engineering and Science, Kyushu University, Fukuoka Japan
3 Institute of Atmospheric Physics, CAS, Beijing, China
Abstract
Asian dust (Yellow Sand) is a significant spring phenomena in East Asia. It is
estimated that several tens of millions of tons of mineral dust are transported every
year from desert areas in China and Mongolia to western Pacific regions. Mineral
dust has various effects on the atmospheric environment, including chemical and
radiative effects, and also on the oceanic environment. Effects of heavy dust events
on human health have also become a concern recently. Several statistics indicate that
the number of dust events has been increasing in recent years. Especially a maximum
number of dust observation days/year was renewed every year between 2000 and
2002 by the Japan Meteorological Agency(JMA), and similar dust observation trend
were also reported in Korea. A heavy dust event, the largest in the last ten years, was
also observed in Beijing on 20 March, 2002. Therefore detailed analysis of dust
phenomena why the number of yellow sand days were significantly increasing and
what meteorological parameters control the dust event frequencies become very
urgent scientific and social issues. However, both observational and numerical studies
for quantitatively estimating the inter-annual variation of dust emission and transport
processes are quite limited.
In this study, a Chemical Weather Forecast System(CFORS) was applied to simulate
the recent dust emission and transport in the every spring time between the year 1993
and 2003 (from Feb. 20 to April 30) over East Asian domain. Here, CFORS was
developed based on a 3D on-line regional scale chemical transport model fully
coupled with the Regional Atmospheric Modeling System (RAMS) [Pielke et al.,
1992]. CFORS treats several chemical tracers including the mineral dust. Mineral
dust emissions (12 bins, ranging from 0.1 to 20 µm in radius) are calculated on-line
using a vertical dust deflatation scheme as a power law function of surface friction
velocity u* [Gillette and Passi, 1988]. Dust emission areas are defined as desert and
semi-desert areas in the US Geological Survey vegetation data base (based on
NOAA/AVHRR data obtained in 1992/93). Monthly snow cover data is used to mask
emission areas. ECMWF Global data set was used for the RAMS boundary
conditions. The simulation domain adopted is centered at 25°N, 115°E. The
horizontal grid consists of 100 by 90 grid points, with a resolution of 80 km. More
detailed information of CFORS dust calculation was described by Uno et al. [2003;
JGR in press].
Dust simulation results were examined by using the TOMS Aerosol Index, LIDAR
observation data, WMO GTE SYNOP reports, dust observation statistics by JMA and
hourly PM10 observation from Ministry of Environment, Japan. It was found that the
10 years dust simulation results explained the observed dust episodes, the annual dust
variation and the recent increase of dust episode in Japan and Korea. Annual dust
emission intensities were strongly controlled by the large scale wind field and snow
cover information, and they are not so sensitive to the soil water content information.
We also found that a large scale anomaly of meteorological field (such as 500 hPa
geopotenital height) is a one of the important factor which controls the dust storm
frequency and dust transport paths. Severe dust episodes in years 2001 and 2002 were
considered as the typical example for discussing the importance of such an anomaly.
We found that in the dust season of 2002, dust was transported to Japan more
frequently than in 2001, and several severe dust phenomena were also reported in
Korea. The total dust emission simulated by the CFORS was similar in each year but
the CFORS model studies revealed that the transport pattern was different. This was
probably related to the fact that the perturbations in the westerly jet was smaller in
2002. The present study indicated that a change in transport pattern due to a slight
change in climate can also cause a large difference in dust phenomena in the
northwestern Pacific region.
In the current calculation with CFORS, the same surface land use categories were
used for every years. To understand these dust phenomena more quantitatively, we
must further investigate surface conditions in the next step of model studies.
Online Scattering Matrix Database for Mineral Particles
Hester Volten 1,2, Olga Muñoz 3, Joop Hovenier 1, Johan de Haan 4, Wim Vassen
5, Wim van der Zande 2, and Rens Waters 1
1 Astronomical Institute "Anton Pannekoek", Kruislaan 403, 1098 SJ Amsterdam,
Netherlands
2 FOM-Institute AMOLF, Kruislaan 407, 1098 SJ Amsterdam, Netherlands
3 Instituto de Astrofísica de Andalucía, CSIC, Granada, Spain
4 Meteorological Institute, KNMI, De Bilt, Netherlands
5 Department of Physics and Astronomy, Free University, Amsterdam, Netherlands
volten@amolf.nl
Abstract
Contents of the database
In recent years a considerable number of experimental single scattering matrices as
functions of the scattering angle obtained with the light scattering facility in
Amsterdam (Hovenier, 2000; Hovenier et al. 2003) have become available for
samples of randomly oriented micron-sized mineral particles in air with broad ranges
of sizes and shapes (Volten et al. 2001, Muñoz et al. 2000, Muñoz et al. 2001, Muñoz
et al 2002). >From these data it has become clear that particle shape is highly
important in determining the overall light scattering behavior of these samples. This
has important implications. First, it confirms that the use of Mie calculations to
interpret data involving light scattering by irregular mineral particles e.g. the Earth
atmosphere, is unlikely to give accurate results (see e.g. Veihelmann et al. 2003).
Second, it implies that light scattering approaches pretending to describe scattering by
irregular particles should take the nonspherical shapes of these particles adequately
into account.
To provide an incentive for further research and applications we have decided to
make our experimental data more easily available by storing our data in digital form
in a database freely accessible through the Internet. All data in this database have
been previously published in scientific journals in graphical form. The database will
contain the following data for several samples of mineral aerosols in random
orientation:
a)Tables of all scattering matrix elements as functions of the scattering angle from 5173 degrees at two wavelengths, 441.6 nm and 632.8 nm.
b)Tables of size distributions measured with a laser diffraction method.
c)Scanning Electron Microscope (SEM) images of the particles that are indicative of
their shape characteristics.
d)Information about the origin, composition and/or the complex refractive index of
the samples, when available.
Where possible information on the accuracy of the data is provided. We intend to
update this database regularly with new measured scattering matrix results.
Applications
There are several ways in which the data from the database can be useful. The data
can be used in a direct manner, e.g. in comparisons with observations of light that has
been scattered once or to assess results of numerical light scattering methods for
nonspherical particles (e.g. Volten et al. 2001, Nousiainen et al. 2002). Also, the data
may be used in an indirect manner. For example, if a method is applied to extrapolate
the measured curves to the full scattering angle range, including forward and
backward scattering, the extrapolated functions may serve as input for model
computations (e.g. Liu et al. 2003, Mishchenko et al. 2003, Veihelmann et al. 2003).
Another way to employ the data in an indirect way, is to find a fit to the experimental
results applying theoretical techniques using parameterized size and shape
distributions. These distributions and techniques can then be used to obtain the
scattering and absorption properties at other scattering angles, wavelengths and/or
sizes where experiments are impossible or not practical, e.g. in the middle and far
infrared.
We like to note that a strong point of the data in the database is that it provides the
complete scattering matrix as a function of the scattering angle and not one or two
elements. This not only facilitates checks of systematic errors in the data, by e.g.
applying "eye ball" tests or the Cloude test (e.g. Hovenier and Van der Mee, 1996),
but also makes it possible to perform multiple scattering calculations including
polarization.
References











Hovenier JW, Measuring scattering matrices of small particles at optical
wavelengths, 2000: in Light Scattering by Non-Spherical Particles, edited by
MI Mishchenko, JW Hovenier, and LD Travis, pp. 355-365, Academic, San
Diego.
Hovenier JW, CVM van der Mee, 1996: Testing scattering matrices, a
compendium of recipes, J. Quant. Spectrosc. Radiat. Transfer, 55, 649-661.
Hovenier JW, H Volten, O Muñoz, WJ van der Zande and LBFM Waters,
2003: Laboratory studies of scattering matrices for randomly oriented
particles. Potentials, problems, and perspectives, J. Quant. Spectrosc. Radiat.
Transfer, 79-80, 741-755.
Liu L, Mishchenko MI, Hovenier JW, Volten H, Muñoz O, 2003: Scattering
matrix of quartz aerosols: comparison and combination of laboratory and
Lorenz-Mie results, J. Quant. Spectrosc. Radiat. Transfer, 79-80, 911-920.
Mishchenko MI, Geogdzhaev I, Liu L, Orgen A, Lacis A, Rossow W,
Hovenier JW, Volten H, Muñoz O, 2003: Aerosol retrievals from AVHRR
radiances: effects of particle nonsphericity and absorption and an updated
long-term global climatology of aerosol properties, J. Quant. Spectrosc.
Radiat. Transfer, 75-80, 953-972.
Muñoz O, H Volten, JF de Haan, W Vassen, JW Hovenier, 2000:
Experimental determination of scattering matrices of olivine and Allende
meteorite particles, Astron. Astrophys., 360, 777-788.
Muñoz O, H Volten, JF de Haan, W Vassen, JW Hovenier, 2001:
Experimental determination of scattering matrices of randomly oriented fly
ash and clay particles at 442 and 633 nm, J. Geophys. Res., 106, 22,83322,844.
Muñoz O, H Volten, JF de Haan, W Vassen, JW Hovenier, 2002:
Experimental determination of the phase function and degree of linear
polarization of El Chichon and Pinatubo volcanic ashes, J. Geophys. Res.,
107, 10.1029/2001JD000983, 4174.
Nousiainen T, Muinonen K, Räisänen P, 2003: Scattering of light by large
Saharan dust particles in a modified ray-optics approximation. J. Geophys.
Res., 108, 10.1029/2001JD001277.
Veihelmann B, H Volten, WJ van der Zande, 2003: Simulations of light
reflected by an atmosphere containing irregularly shaped mineral aerosol
over the ocean, 7th Conference on Electromagnetic and Light Scattering by
Nonspherical Particles: Theory, Measurements, and Applications, Bremen.
Volten H, O Muñoz, E Rol, JF de Haan, W Vassen, JW Hovenier, K
Muinonen, T Nousiainen, 2001: Scattering matrices of mineral particles at
441.6 nm and 632.8 nm, J. Geophys. Res., 106, 17375-17401.
Dust Sources and Transport
Richard Washington, Andrew Goudie and Martin Todd
Climatology Research Group, University of Oxford, Mansfield Rd, Oxford, OX1
3TB, UK Richard.Washington@geog.ox.ac.uk
Abstract
We investigate the location of the key sources of global dust from a variety of data
sets and attempt to explain the origin of these sources and the interannual variability
of dust emissions from them by means of three dimensional wind data and a fourth
order trajectory model.
Surface data from operational meteorological observations (including visibility and
dust storm frequency) are analysed along with daily data from the well known Total
Ozone Monitoring Spectrometer's (TOMS) Aerosol Index (AI). Atmospheric
circulation data is derived from the NCEP-NCAR reanalysis project which provides 6
hourly three dimensional winds on a 2.5 by 2.5 degree grid from 1948 onwards.
We start by presenting a dust climatology calculated from the mean fields of the
surface based and satellite data. Agreement and points of departure in the mean fields
and seasonal cycle are noted, with an emphasis on North Africa. Objectively defined
source regions for the Sahara are determined from eigenvector analyses applied to the
TOMS data. Regions of coherent variability, defined by eigenvectors of the
covariance matrix of TOMS AI monthly anomalies, are presented. These include the
Bodele Depression in Chad, the Djouf region of Mali/Mauritania, the Chotts of
Tunisia and Algeria, a large region of central Libya as well as the dust plumes over
the Atlantic. The remarkable degree of co-location between some of these source
regions, the potential sand flux, q, (q = 2.61 U3* pg -1 (1-U*/U*)(1+U*/U*)2 as
computed from 6 hourly wind fields from 1948) and the surface elevation is
remarkable. On the other hand there are some important regions that surface
observations indicate as being important that do not appear consistently in the TOMS
data. Possible reasons for these discrepancies are explored.
Interannual variability of dust emissions from some of the primary dust sources is
examined. Anomalously high and anomalously low emissions from the Bodele
Depression are sampled from an area index of TOMS AI values and visibility/sand
storm data from Bilma in Niger and Faya in Chad. Atmospheric circulation anomalies
in the form of 3 dimensional winds from the reanalysis data sets are examined to
determine the degree to which the variability in dust output relates to the atmosphere
alone. Transport of the dust from the Bodele Depression and other key Saharan
sources is examined in a fourth order trajectory model.
Observations of Dust using the NASA Geoscience Laser Altimeter System
(GLAS): new measurements of aerosol vertical distribution from space
Ellsworth J. Welton 1, James D. Spinhirne 1, Steven P. Palm 2, Dennis Hlavka 2,
and William Hart 2
1. NASA Goddard Space Flight Center, Greenbelt, MD, USA
2. Science Systems and Applications Inc., Greenbelt, MD, USA Corresponding
author's email address: Ellsworth.J.Welton@nasa.gov
Abstract
On January 12, 2003 NASA launched the first satellite-based lidar, the Geoscience
Laser Altimeter System (GLAS), onboard the ICESat spacecraft. The GLAS
atmospheric measurements introduce a fundamentally new and important tool for
understanding the atmosphere and climate. In the past, aerosols have only been
studied from space using images gathered by passive sensors. Analysis of this passive
data has lead to an improved understanding of aerosol properties, spatial distribution,
and their effect on the earth's climate. However, these images do not show the
aerosol's vertical distribution. As a result, a key piece of information has been
missing. The measurements now obtained by GLAS will provide information on the
vertical distribution of aerosols and clouds, and improve our ability to study their
transport processes and aerosol-cloud interactions. Here we show an overview of
GLAS, provide an update of its current status, and present initial observations of dust
profiles. In particular, a strategy of characterizing the height profile of dust plumes
over source regions will be presented.
1. Introduction Satellite observations of aerosols have proven to be an invaluable
asset in the array of measurement capabilities available to researchers (Kaufman et
al., 2002). Satellite imagery has enabled studies of regional and global aerosol
distributions and their climatic impact. However, these satellite measurements have
been passive in nature and do not provide information on the aerosol's height
distribution. Active remote sensing, such as lidar, is required to provide aerosol
profiles. In the past, technology and cost constraints prohibited a widespread use of
active remote sensing from space. The first attempt at such measurements was from
the NASA Lidar In-space Technology Experiment (LITE) in 1994 (Winker et al.,
1996). However, LITE was flown on the space shuttle and only acquired a limited
data set before the end of the mission. In order to obtain continuous long-term
atmospheric profiles, a satellite-based lidar is required. NASA launched the first such
satellite-based lidar, the Geoscience Laser Altimeter System (GLAS), onboard the
ICESat spacecraft on January 12, 2003.
2. The Geoscience Laser Altimeter System (GLAS) An overview of the ICESat
mission and the GLAS sensor is discussed by Zwally et al. (2002). A more detailed
presentation of the GLAS atmospheric measurements is provided by Palm et al.
(2002), a short summary is given here. The orbital altitude of ICESat is 600 km at an
inclination of 94º with an 8-day repeat track. After the completion of the calibrationvalidation phase (later this year) the orbit is scheduled to switch to a 183-day repeat
track. GLAS acquires atmospheric profiles of molecules, aerosols, and gases from 40
km to the surface using a two-wavelength backscatter lidar (1064 and 532 nm). Upon
striking the surface, the GLAS laser footprints are 70 m wide and spaced about 175 m
apart. The atmospheric profiles are recorded with a vertical resolution of 76.8 m. The
laser pulse repetition frequency is 40 Hz, but temporal averaging of the raw signals
occurs according to height. The data are stored at 1 Hz from 40 to 20 km, 5 Hz from
20 to 10 km, and the full 40 Hz from 10 km to the surface. Further temporal and
vertical averaging may be necessary to detect some aerosol layers. For instance, subvisual cirrus and tenuous aerosols may require a ~7 second average (or approximately
50 km horizontal resolution) to be detected accurately.
3. History and Current Status GLAS data acquisition began at the end of February
2003 and continued until the end of March when a malfunction occurred in the first of
three lasers onboard GLAS. The instrument has remained off while the instrument
team investigates the cause of the laser failure. GLAS is expected to begin normal
operations again this summer using the next laser (each laser transmits both
wavelengths). The initial GLAS data set contains continuous global atmospheric
profiles at 1064 nm during a 1-month time-span. The 532 nm detectors were not
scheduled to turn on until after March because of their long out-gassing period. Both
1064 nm and 532 nm data will be available after GLAS restart this summer.
4. Observations of Dust Plumes and Sources Using GLAS During February and
March 2003 dust plumes were studied using nearly coincident observations from
GLAS and MODIS on TERRA and AQUA. These studies demonstrate the usefulness
of combined passive and active analysis. In particular, the heights of the dust plumes
and clouds in the MODIS images were obtained from GLAS, and the results show
areas were aerosol-cloud interactions can occur. The height profile of dust plumes
emanating from their source has also been studied with GLAS. The repeat orbit tracks
cross over or close to almost all major sources in the "dust belt" (Prospero et al.,
2002). The aim of this study is to characterize the plume height directly over and
immediately downwind of dust sources. Such information will improve the initiation
of dust transport models (Ginoux and Torres, 2003).
References






Ginoux, P., and O. Torres, 2003: Empirical TOMS Aerosol Index:
Applications to Model Validation and Dust Source Characterization, J.
Geophys. Res., in press.
Kaufman, Y.J., D. Tanre, and O. Boucher, 2002: A satellite view of aerosols
in the climate system, Nature, 419, 215-223.
Palm, S., W. Hart, D. Hlavka, E.J. Welton, A. Mahesh, and J. Spinhirne,
2002: Geoscience Laser Altimeter System (GLAS) Algorithm Theoretical
Basis Document Version 4.2: GLAS Atmospheric Data Products,
http://glo.gsfc.nasa.gov/publications/, 1-137.
Prospero, J.M., P. Ginoux, O. Torres, S.E. Nicholson, and T.E. Gill, 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,
doi:10.1029/2000RG000095.
Winker, D.M., M.P. McCormick, and R. Couch, 1996: An Overview of
LITE: NASA's Lidar In-space Technology Experiment, Proc. IEEE, 84, 164180.
Zwally, H.J., B. Schutz, W. Abdalati, J. Abshire, C. Bentley, A. Brenner, J.
Bufton, J. Dezio, D. Hancock, D. Harding, T. Herring, B. Minster, K. Quinn,
S. Palm, J. Spinhirne, and R. Thomas, 2002: ICESat's laser measurements of
polar ice, atmosphere, ocean, and land, J. Geodyn., 34, 405-445.
High-Resolution Modeling of Dust Sources and Dust Storms in East and
Southwest Asia
Douglas L. Westphal*, Ming Liu, Annette L. Walker, Kim Richardson, Teddy R.
Holt and Steven D. Miller
Naval Research Laboratory, Marine Meteorology Division, Monterey, CA 93943
*Tel. (831) 656-4742, westphal@nrlmry.navy.mil
Abstract
Dust storms generated by wind erosion over arid or semiarid land surfaces and
transported long distances downwind can cause adverse environmental effects,
disrupt transportation, and may alter the Earth's weather and climate. Satellite
imagery reveals that many of these events originate from small-scale individual
sources. Downwind, the plumes merge together to form the synoptic-scale
plumes that are modeled by GCM and synoptic-scale models. To date, few
studies have been made regarding both the fine-scale distribution of these
sources or the dynamical forcing and subsequent merging of the plumes. In this
paper we discuss a mesoscale (9-km horizontal resolution) dust aerosol model
that has been developed based on the Navy's mesoscale weather model
(COAMPSTM.) The predictive dust module is part of COAMPS and uses the
exact meteorological fields at each time step and at each grid point of all nests
resulting in dust storm simulations with more fidelity, realism and accuracy than
off-line or large-scale dust models. Examples of COAMPSTM dust forecasts for
China, Mongolia, Afghanistan, and Iraq will be presented including validation
against satellite, sun photometer, and surface observations.
The potential benefits of this high-resolution approach can only be achieved if
the model is provided with a compatible high-resolution dust source inventory.
All existing databases have resolutions of one degree or coarser. This is
inadequate for mesoscale simulations with resolutions as high as 3-km. We will
describe our approach to solving this problem, present our results to date, make
recommendations, and solicit feedback from the workshop participants.
CALIPSO: New Capabilities for Global Observations of Dust
David M. Winker
NASA Langley Research Center, Hampton, VA, USA
david.m.winker@nasa.gov
Abstract
It is widely recognized that mineral dust impacts the radiation balance of the
atmosphere on regional and global scales. The lifetime of dust in the
atmosphere, and thus its total radiative impact, depends strongly on its altitude.
If dust is lofted into the free troposphere it can be transported long distances,
affecting even the global scale. The climate effects of dust are poorly understood
because of limited abilities to observe the geographical distribution and vertical
extent of dust. It is difficult for conventional passive satellites to observe
atmospheric dust near its sources, which tend to be areas of high surface albedo,
and passive observations provide no information on the vertical distribution of
dust. Chemical transport models are now beginning to incorporate sophisticated
source models for mineral dust, but the global transport of the dust is highly
dependent on the vertical mixing predicted by the models. The vertical
distributions predicted by these models are uncertain because the observational
datasets to adequately test the vertical transport schemes do not currently exist.
Space lidar can help address these issues. Lidar provides vertical resolution on
the order of 100 meters - over bright deserts as well as the dark ocean - and
allows the observation and retrieval of aerosol properties even in the presence of
overcast thin cirrus or underlying low clouds. The Lidar In-space Technology
Experiment (LITE) in 1994 demonstrated the potential of space lidar for
profiling clouds and aerosols (Winker, et al., 1996; Karyampudi, et al., 1999).
CALIPSO is a satellite mission which builds on the experience of LITE with a
payload consisting of a two-wavelength polarization-sensitive lidar, and passive
imagers operating in the visible and infrared spectral regions (Winker, et al.,
2003). CALIPSO is being developed within the framework of a collaboration
between NASA and the French space agency, CNES, for launch in late 2004.
The CALIPSO lidar can identify situations where multiple aerosol layers are
present and can partition the column aerosol optical depth between an elevated
dust layer and aerosol in the boundary layer. Quijano et al. (2000) have shown
that to understand the radiative effects of aerosols it is important to be able to
observe the vertical layering of different aerosol types. CALIPSO will be able to
discriminate aerosol dominated by large or small particles by analysis of the
backscatter signals at the two wavelengths, 532 nm and 1064 nm. The lidar
depolarization measurement will discriminate between dry and hydrated
aerosols. Thus, CALIPSO will be able to distinguish elevated layers of dust and
smoke, and can distinguish dust from hydrated sea salt aerosol or fine mode
pollution aerosol. The aerosol profile observations can be combined with backtrajectory analysis to provide further insight into aerosol composition and
sources.
Retrieval of aerosol optical depth from a simple backscatter lidar requires a
priori knowledge of the aerosol extinction-to-backscatter ratio, Sa, and this is the
dominant factor determining the accuracy of the retrievals except at very low
loadings where signal-to-noise ratio becomes important. The value of Sa
depends on aerosol size distribution, composition, and shape. Fortunately, dust
is often observed in elevated layers with relatively clean air underneath. This
allows the optical depth of the upper layer to be determined from the layer
transmittance obtained by comparing the lidar return from clear air on either
side of the layer. Combined with the integrated backscatter signal through the
layer, an estimate of Sa can be made directly from the data. This estimate can
then be used in extinction retrievals in locations where the layer is not elevated.
CALIPSO will fly in formation with the Aqua, Aura, PARASOL, and CloudSat
satellites as part of the "Aqua constellation" or "A-train". The CALIPSO orbit
will be maintained relative to Aqua so that observations are acquired within
about 1.5 minutes. This time-space coincidence allows additional observational
synergies by combining CALIPSO profile measurements with passive
measurements from other instruments in the A-train. This talk will discuss the
capabilities of CALIPSO to observe mineral dust and how these new
observations will contribute to our understanding of the role of dust in the
climate.
References




Karyampudi, V. M., et al., 1999: Validation of the Saharan Dust Plume
Conceptual Model Using Lidar, Meteosat, and ECMWF Data, Bull.
Amer. Meteor. Soc., 80, 1045-1075.
Quijano, A. L., I. N. Sokolik, and O. B. Toon, 2000: Influence of the
aerosol vertical distribution on the retrievals of aerosol optical depth
from satellite radiance measurements. Geophys. Res. Lett., 27, 34573460.
Winker, D. M., M. P. McCormick, and R. Couch, 1996: An Overview
of LITE: NASA's Lidar In-space Technology Experiment, Proc. IEEE
84, 164-180.
Winker, D. M., J. Pelon, and M. P. McCormick, 2003: The CALIPSO
Mission, Proc. SPIE, vol 4893, 1-11.
Recent Results of Individual Asian Dust Particle Analysis
Daizhou Zhang (1), Yasunobu Iwasaka (2) and Guangyu Shi(3)
1Prefectural University of Kumamoto, Kumamoto 862-8502, Japan
(email:zdz@pu-kumamoto.ac.jp)
2Graduate School of Environmental Sciences, Nagoya University, Nagoya 4648601, Japan
3Institute of Atmospheric Physics, Chinese Academy of Science, Beijing
100029, China
Abstract
Our group are measuring Asian dust particles at source areas (Dunhuang: the
east edge of Taklamaka desert), the coastal zone of China (Qingdao), and
Japanese islands. Besides surface measurements, much effort has been done to
get particles at different altitudes by balloon-borne or aircraft-borne samplers
and particle counters. Collected particles are investigated using electron
microscopes and an energy dispersive X-ray spectrometer. Measured items
include dust particle size and shape, their mixture state with sea salt, sulfate and
nitrate, and vertical distributions. In this report, we will report some major new
results of the measurements at different regions and identify some questions
which need to be addressed for evaluating Asian dust.
Measurements at source areas: Particles in free troposphere over the desert areas
in Northwestern China were collected using balloon-borne samplers. The
analysis results of collected particles indicate there is a dust layer at about
3~5km over the desert areas in summer 2002 when no dust events were reported
or recorded. Si-rich and Ca-rich particles occupied more than 70% of particles
of diameter>1µm. This layer was also confirmed by balloon- particle counters
and surface Lidar observations. The emission sources of these particles were not
identified although they seemed to be from the Taklamakan desert. There is no
further data to explain the presence of the layer, to discuss the transport of the
particles in the layer, and the ranges these particles can reach to.
Measurements at the coastal zone: The analysis of particles collected at a coastal
site of Qingdao during a dust storm event in spring 2002 reveals that the dust
plume did not mix with polluted air masses and they were separated by the cold
front of the event. The post-frontal mineral particles are the so-called Asian dust
particles and they were rarely polluted. The pre-frontal ones are from
anthropogenic emissions and contained significant anthropogenic pollutants.
Similar weather patterns were also confirmed from other dust storm events.
These results suggest that the timing of particle collection is an important factor
for Asian dust observations at the continent. If the collection is started before the
front arrival, the obtained samples may contain significant pollutants irrelevant
to dust particles.
The results from three dust storm events at the coastal site indicate that,
although a number of dust particles contained Na, S, and Cl, the elements in
most of the particles were more likely from crustal origins. Sulfate was rarely
detected in the particles. Since sulfur compounds in surface soils at dust source
areas are almost completely water soluble, the crustal contributions of Na, S and
Cl need to be evaluated in the investigation of anthropogenic pollutants in bulk
dust samples collected in the continental atmosphere.
Measurements at Japan: More than 60% dust particles collected at Japan are
internally mixed with sea salt. It seems dust particles can efficiently combine
with sea salt even with no cloud or fog processing. So the coagulation of dust
particles and sea salt in marine atmosphere cannot be ignorant in accessing the
influence of Asian dust in the marine atmosphere. Although the relative contents
of mineral components and sea salt are different in different dust particles, the
size distributions of dust particles segregated by the mixture amounts from high
mineral content with low sea salt to low mineral content with high sea salt are
similar to each other and have maxima around 3 µm diameter with the range of
1~8 µm. Out of this range, few dust particles were detected. It is confirmed that
the coagulation of dust particles with sea salt can cause large increase of the dust
particle size. So the decrease of particle concentrations in the range of larger
than 3 µm may have an important meaning. That is the critical diameter for dust
particle dispersion is around 3 µm and a particle will be removed rapidly when
its diameter is or becomes larger than this scale. This hypothesis provides a
plausible interpretation for why the distribution modes of dust storm particles
observed at Korea, Japan and even North America during Asian dust are
approximately consistent. In addition, it is found that dust particle can modify
the chlorine chemistry in marine atmosphere through surface uptake of gaseous
HCl and restraint of HCl release from sea salt, and the Cl deposition onto dust
particle seems not ignorant compared with S deposition.
Aircraft-borne measurements held in the free troposphere over central Japan
revealed a steady transport of dust in a small scale between 2~6km during spring
including days with no evident dust outbreak. Such dust was also observed in
some summer observations in levels higher than 4 km under the influence of
remaining westerly winds. Concerning the summer dust layer in free
troposphere at source areas, Asian dust may not be a phenomenon confined in
spring but may continue to summer under persisting westerlies in the middle
latitude of the northern hemisphere.
Surface of observation of an Intense Dust Storms in Beijing on 20
March 2002
Renjian Zhang
State Key Laboratory of Atmospheric Boundary Layer Physics and
Atmospheric Chemistry, Institute of Atmospheric Physics, Chinese Academy of
Sciences, Beijing, 100029, China
Abstract
On March 20, 2002, the strongest dust storm over the last decade attacked
Beijing. The change of meteorological parameters in the boundary and physical
properties were studied. Back trajectory method is applied to investigated the
origin of this dust storm The results show that dust storm lead to the distinct and
sharp increase of wind speed, decrease of relative humidity, development of
turbulence and mixing capability in boundary layer. In dust period, TSP mass
concentrations reached 12 mg/m3(Figure 1). The coarse particle(>2.1µm)
account for 61% in no dust period and 95% in dust period. The mass
concentration of coarse particle and fine particle are 9.4 and 1.2 times as that in
no dust period, respectively. Back trajectory analysis shows that this heavy dust
storm on March 20, 2002 was originally from Xinjiang in China, Mongolia, and
Inner Mongolia in China by long-range transportation.
Figure 1. Mass concentrations of dust in Beijing in spring 2002
References


1.Zhang Renjian et al., 2003, Inorganic chemical composition and
source signature of PM2.5 in Beijing during ACE-Asia period, Chinese
Science Bulletin, 48(10),1002-1005.
2.Zhang Renjian et al., 2000, Research on elemental concentrations and









distributions of aerosols in Winter/spring in Beijing, Climate and
Environmental Research,5(1), 6~12(In Chinese, English abstract).
3.Zhang Renjian et al., 2000, Analysis on the chemical and physical
properties of "2000.4.6" super dust storm in Beijing, Climate and
Environmental Research, 5(3),259-266 (In Chinese, English abstract).
4.Zhang Renjian et al., 2001, Preliminary research on the size
distribution of aerosols in Beijing, Adv. Atmos.Sci., 18(2), 225~230.
5.Zhang Renjian et al., 2001, Observation and analysis on elemental
composition of atmospheric aerosols over Mount Qomolongma
Region, Plateau Meteorology, 20(3),234-238 (In Chinese, English
abstract),2001.
6.Zhang Renjian et al., Focus on Dust Research in China, Plenary
lecture in 2nd Asian Aerosol Conference, July 1-4,2001, Korea,
Pushan.
7.Zhang Renjian et al., 2001, Chemical composition and size
distribution of aerosols in winter/spring in Beijing, Proceeding of
International Conference on Engineering and Technological Sciences
2000, Section 7:Environmental Protection & High Technology, Song
Jian et al., Eds., 235-240, China Ocean Press, Beijing.
8.Zhang Renjian et al., 2002, Dust storm weather in China: New
characteristics and origins, Quaternary Sciences, 22(4),374-380(In
Chinese, English abstract).
9.Zhang Renjian et al., 2002, Chemical composition of dust and PM2.5
at Beijing in ACE-Asia, Proceeding of Sixth International Aerosol
Science, International Aerosol Research Assembly, 127-128, Taibei,
September 2002 (ISBN 986-80544-1-9).
10.Zhang Renjian et al., 2002, Elemental concentration and distribution
of atmospheric particles before and in heating period in Beijing,
Journal of Graduate School of the Chinese Academy of Sciences,
19(1), 75-81, (In Chinese, English abstract).
11.Zhang Renjian et al., 2002, Chemical composition of aerosols in
winter/spring in Beijing, Journal of Environmental Sciences, 14(1), 711.
Linkage of Saltation and Dust Emissions in Bare Agricultural Fields
Ted M. Zobeck(1), R. Scott Van Pelt(1), Jim Kjelgaard(2), and Brenton
Sharratt(3)
(1)1Wind Erosion and Water Conservation Research Unit, USDA, Agricultural
Research Service, 3810 4th Street Lubbock, TX 79415,
tzobeck@Lbk.ars.usda.gov
(2)Biological Systems Engineering, Washington State University, Pullman,
WA 99164-6120, jkjelgaard@wsu.edu
(3)Land Management and Water Conservation Group, USDA-ARS, Pullman,
WA 99164-6120, sharratt@wsu.edu
Abstract
Vertical dust flux is often estimated using the gradient method and flux equation
derived by Gillette (1977) (eg. Nickling and Gillies, 1993; Stetler and Saxton,
1996; Lopez et al, 1998; Rajot et al., 2003). In addition, wind tunnel studies and
theoretical considerations show that the dust (<20 µm) emission rate is
proportional to the horizontal saltation flux, and hence proportional to u*3, if u*
is much larger than the threshold friction velocity for saltation (Shao et al.,
1993; Shao, 2000). These studies were generally based on time-integrated
measurements of dust flux. Recent advances in sensor technology have allowed
for the measurement of wind velocity measurements and saltation and dust
fluxes at high frequency, enabling more detailed analyses of the interaction of
wind, saltation, and dust. This presentation will report the results from two field
dust sampling campaigns, conducted on agricultural fields in the US, that
employed fast-response wind, saltation, and dust sensors.
One site (WA) was located on a silt loam soil in the northwest US, in the
southeast quadrant of the state of Washington, a region of extensive loess
deposits known as the Columbia Plateau. The second site (TX) was located on a
fine sandy loam soil in the Southern High Plains of west Texas, a region of
extensive sandy aeolian deposits. Sampling equipment deployed on the 9 ha WA
site included BSNE saltation samplers, Sensit saltation monitors, sonic
anemometers at two heights, tapered element oscillating microbalances
(TEOMs) at two heights, high volume dust samplers, and meteorological
equipment. The meteorological equipment included three-cup anemometers at
six heights, air temperature sensors at five heights, wind vane, relative humidity
sensor, infrared thermometer to monitor soil surface temperature, net
radiometer, pyranometer, rain gauge, soil heat flux plates at a depth of 10 cm,
and thermocouples at two depths in the soil. Sampling equipment deployed on
the 3 ha TX site included near-surface creep/saltation samplers, BSNE saltation
samplers, Sensit saltation monitors, sonic anemometers at two heights, three
towers with Dusttrak aerosol monitors installed at three heights, and
meteorological equipment. The meteorological equipment included three-cup
anemometers at two heights, air temperature, wind vane, relative humidity
sensor, rain gauge, pyranometer, evaporation pan, barometric pressure sensor,
and thermocouples at 2 depths in the soil.
Temporal resolution of saltation and dust sampling equipment varied with the
type of samplers. The near-surface/creep and BSNE saltation samplers collected
samples for the duration (time-integrated) of each high wind event (HWE). The
Sensit detected impacts of saltating sand grains at a frequency of 1 Hz
(dynamic). The high volume dust samplers, TEOMs and Dusttraks were
designed to measure particulate matter with a 10 µm diameter cut-point ( PM10)
at different rates. The high volume dust samplers collected samples on a filter
for the duration of the HWE. The TEOMs and Dusttraks measured dust
concentrations at frequencies of 0.1 Hz and 1 Hz, respectively. The sonic
anemometers measure wind speed in three directions at a frequency of 10-20
Hz.
Horizontal saltation flux measured with BSNE showed good correlation with
PM10 measured over the same time period. Dust concentration with height
varied with storm intensity. Dust concentrations were similar at different heights
during intense storms but seemed to vary more with height during less intense
HWE. Evidence for the saltation effects on dust flux varied with soil. Strong
evidence for a saltation effect on dust flux was found in the fine sandy loam TX
site. Conversely, few saltation particle impacts were observed during HWEs on
the silt loam WA site and an aerodynamic analysis based on wind velocity
profile regression also suggested little saltation occurred during HWEs.
References

Gillette DA. 1977. Fine particulate emissions due to wind erosion.






Trans. ASAE 20: 890-897.
Nickling WG, Gillies JA. 1993. Dust emission and transport in Mali,
West Africa. Sedimentology 40:859-868.
Stetler LD, Saxton KE. 1996. Wind erosion and PM10 emission from
agricultural fields on the Columbia Plateau. Earth Surface Processes
and Landforms 21:673-685.
Lopez MV, Sabre M, Garcia R, Arrue JL, Gomes L. 1998. Tillage
effects on soil surface conditions and dust emission by wind erosion in
semiarid Aragon (NE Spain). Soil Tillage Res. 45:91-105.
Rajot JL, Alfaro SC, Gomes L, Gaudichet A. 2003. Influence of sandy
soil crusting on its susceptibility to wind erosion. Catena (In press).
Shao Y, Raupach MR, Findlater PA. 1993. The effect of saltation
bombardment on the entrainment of dust by wind. Journal of
Geophysical Research 98:12719-12726.
Shao Y. 2000. Physics and Modelling of Wind Erosion. Kluwer
Academic Publishers. Dordrecht, The Netherlands.