Dust Deposition - Mineral Dust: A Key Player in the Earth System - page 177

three steps of the dust cycle: dust emissions, mainly from the arid and semiarid
regions of the Earth; transport through the atmosphere depending strongly on the
meteorological situations; and finally deposition of dust along their atmospheric
path by wet or dry processes. The global atmospheric dust burden at any one
moment is the result of the balance between the sources and sinks. Thus, a complete
closure of the dust budget at the global scale needs to quantify precisely the amount
of emitted dust, the atmospheric dust load, and the deposited dust mass (dry and
wet). However, this budget is sufficiently constrained if, at least, two of these three
terms are quantified with a good confidence level.
During the 1990s, dust emissions have been considered to be the main source of
errors in dust models (Joussaume 1990 ; Genthon 1992 ; Marticorena and Bergametti
1995 ) (see Chap. 9 ) . Large uncertainties were attached to (1) the representation of
the involved processes and/or their parameterization, (2) the quality of the input data
required by these dust emission models to characterize the main surface properties
of the desert areas, and (3) the localization of the dust source regions (see Chap. 3 ) .
During the last twenty years, significant progress has been made on dust emission
modelling (Shao et al. 1993 ; Marticorena and Bergametti 1995 ;AlfaroandGomes
2001 ;Shao 2004 ; Marticorena 2014 ) and on the dust source monitoring, especially
by using satellite observations (Prospero et al. 2002 ; Washington et al. 2003 ).
However, quantitative estimates of dust emissions in 3D models are still affected
by large uncertainties (Textor et al. 2006 ; Huneeus et al. 2011 ), mainly because a
quantitative validation of dust emissions at a large scale is presently not possible.
The description of the atmospheric dust content in terms of both spatial distri-
bution and temporal variability has also been significantly improved through the
development of aerosol products from remote-sensing instruments. Despite some
limitations of these techniques (see Chap. 7 ) , the atmospheric dust content is
clearly the term of the dust mass budget for which the quantitative estimates are
most precise today (Textor et al. 2007 ). As a consequence, most of the dust cycle
simulations are mainly compared to large datasets of aerosol optical depth (AOD)
or dust concentrations measured during specific intensive campaigns.
Dry and wet depositions are the processes by which atmospheric dust particles
are removed from the atmosphere (see, e.g., Duce et al. ( 1991 ); Schulz et al. ( 2012 )).
In the vicinity of source regions, dry deposition is generally the dominant process,
due both to the presence of large dust particles and to the dry climate prevailing
in deserts. Far from source regions, since the larger particles have already been
deposited, the dust size spectrum is narrower with a typical mass median diameter
close to 2 m (e.g., Schütz ( 1980 )). Thus, wet deposition is usually the dominant
removal process for dust particles far from the source regions (e.g., Bergametti
et al. ( 1989 )). These two sinks, which counterbalance dust emissions on the global
scale, control the atmospheric lifetime of dust particles and their biogeochemical
impact by defining how much, when, and where dust is deposited onto continental
or oceanic surfaces.
However, rather little attention has been paid to dust deposition. It was generally
assumed that the existing schemes were sufficiently efficient for representing dust
deposition in models. As a consequence, few experiments were dedicated to test