Numerical Dust Models - Mineral Dust: A Key Player in the Earth System - page 200

10-m wind product to compute dust emissions. Subsequently developed global
dust emission and transport models (e.g. Mahowald et al. 1999 ; Ginoux et al.
2001 ; Zender et al. 2003 ; Tanaka and Chiba 2005 ) improved the simulation of
dust emissions by specifying the way in which land-surface conditions influence
dust sources (see also Chap. 3 ) . These specifications include prescribed vegetation
cover from satellite remote sensing or vegetation models to determine potential dust
source regions. In addition, the locations of preferential sources of dust aerosol are
prescribed in several recent models.
The simulated dust concentrations and optical thickness distributions, resulting
from such models, are validated with in situ observations and/or remotely sensed
aerosol products (e.g. Mahowald et al. 2005 ; Huneeus et al. 2011 ). State-of-the-
art global models reproduce the large-scale regional and seasonal patterns of the
atmospheric dust loads (e.g. the Saharan dust plume over the North Atlantic, the
transport of Asian dust to the Pacific and the Arabian dust plume in the northern
hemisphere summer) (see also Chap. 7 ) . However, uncertainties in description of
source areas and processes controlling dust emission and deposition in global-scale
models result in a wide range of simulated dust fluxes. The complex refractive
indices of dust particles depend on the mineral composition of soils in the source
areas. Since the refractive indices influence the optical properties of dust (Chap. 11 ) ,
uncertainties in the location of sources lead also to uncertainties in simulated
dust radiative effects in models. The radiative heating by mineral dust particles is
parameterized in the radiative transfer scheme of the global models according to
prescribed dust optical properties, which can depend on particle size, but normally
constant complex refractive indices are prescribed globally when computing dust
radiative effects. That implies that the varying mineral composition of dust from
different source areas is usually not taken into account (e.g. Balkanski et al. 2007 ).
The performance of many global aerosol models was evaluated in the global
aerosol model intercomparison study AeroCom (Textor et al. 2006 ). Within Aero-
Com, the simulated distribution and radiative effects of natural and anthropogenic
aerosol components, including dust and sea salt particles, sulphate, organic carbon
and black carbon, were compared. Textor et al. ( 2006 ) describe the results of
the individual components of the aerosol cycles for 16 global models. While the
AeroCom study shows that the averages and seasonal variability of vertically
integrated mineral dust parameters like optical thicknesses and Ångstrom exponents
agree within a factor of two with observations, less agreement is found for surface
concentration and deposition fields of mineral dust particles.
Global models are of limited use for the investigation of regional-scale processes,
but they are useful to study large-scale dust transport and to provide the global dust
distribution. Some global models are designed to provide forecasts and boundary
conditions for regional models. The latter are of more use in investigating processes
close to dust source regions, since the typical horizontal grid resolution of several
degrees and simplified emission and deposition parameterizations are too crude to
accurately reproduce smaller-scale dust processes. Regional models are therefore
used for simulations of individual dust storm events as well as for providing