Geoscience Reference
In-Depth Information
significant impact on the formation of stone pavements
(McFadden, Wells and Jercinovich, 1987) and surface
crusts and duricrusts (Eckardt
et al.
, 2001). Far-travelled
and fine-textured dust can become deposited in oceans
and polar regions and the analysis of desert dust in ice
cores and ocean sediments has proved an important source
of evidence for both global palaeoclimatic reconstruction
(De Deckker
et al.
, 2010) and contemporary environmen-
tal change (McConnell
et al.
, 2007).
lites such as the Infrared Difference Dust Index (IDDI)
(Legrand, Plana-Fattori and N'doume, 2001) and observa-
tions from the Meteosat Second Generation (MSG) Spin-
ning Enhanced Visible and Infrared Imager (SEVIRI)
instrument (Schepanski
et al.
, 2007, 2009), which pro-
vide temporal resolution of around 15 minutes day and
night, although their spatial resolution is much lower
(1-4 km). The impact of satellite data on desert dust re-
search has been profound and has been widely used in
planning ground campaigns, evaluating numerical mod-
els and understanding source regions and transport. Instru-
ments on satellites measure radiation and use algorithms
to convert that radiation into measures of dust loadings, al-
though many of these measures are nondimensional (unit-
less), relating only qualitatively to dust concentrations.
Additionally, the algorithms rely on assumptions such as
dust particle sphericity and atmospheric water vapour con-
tent for which there are no reliable data.
20.1.2
Measuring dust
The major advances over the last decade in our under-
standing of desert dust have come from improved tech-
niques in measuring and numerically simulating dust.
Prior to the availability of remote sensing techniques suit-
able for the detection of dust, much of the data on global
dust distribution were founded on synoptic weather re-
ports and visibility estimates from weather stations (Mid-
dleton, 1986). Numerous problems beset these data, such
as the inconsistency of reporting dust storms, lack of direct
quantification of dust loadings and the subjective nature of
visibility measurement. Their most important limitation is
still that weather stations tend to be located at desert mar-
gins and almost all are now known, from remote sensing of
dust, to be distant from key source regions. Nevertheless,
they offer a useful resource, particularly in comparison
with top-down satellite approaches.
20.1.2.2 Aircraft campaigns
Most satellite-derived dust data sets provide a top-down,
two-dimensional view of dust. Aircraft provide a means to
sample the characteristics of suspended dust and its influ-
ence on the Earth System in situ. This advantage has led to
numerous aircraft campaigns to measure dust, e.g. the Sa-
haran Dust Experiment (SHADE) (Haywood
et al.
, 2003),
the Mediterranean Dust Experiment (MEIDEX) (Alpert
et al.
, 2004), the Cooperative LBA Airborne Experiment
(CLAIRE) (Formenti
et al.
, 2001) and the Aerosol Char-
acterization Experiment (ACE-Asia) (Anderson
et al.
,
2003). The Dust and Biomass Experiment (DABEX) and
the Dust Outflow and Deposition to the Ocean (DODO)
Experiment were based out of Niamey, Niger, and Dakar,
Senegal, respectively. Collectively, these campaigns have
constrained and quantified the influence of dust on the
radiation budget and yielded comprehensive data sets
on dust characteristics, such as transport pathways, dust
plume heights and dust size, shape and colour. A notable
feature of these campaigns is that almost all have been
distant from dust source regions.
20.1.2.1
Remote sensing of dust
Satellites make no discrimination against measurements
from remote regions of the planet. In fact, the lack of cloud
in deserts makes these regions particularly well suited to
remote sensing using visible channels. Multiple satellite-
derived dust data sets have been produced from many
different sensors since the early 2000s (Table 20.1). Some
of the most widely used are: Total Ozone Mapping Spec-
trometer Aersol Index (TOMS AI) (Herman
et al.
, 1997),
the TOMS aerosol optical thickness (AOT) (Torres
et al.
,
2002), multiangled imaging spectroradiometer (MISR)
(Meloni
et al.
, 2004) and algorithms from the Moderate
Resolution Imaging Spectroradiometer (MODIS) (Kauf-
man
et al.
, 2005), such as the 'Deep Blue' algorithm (Hsu
et al.
, 2004). Others, such as Synergetic Aerosol Retrieval
(SYNAER) from ENVISAT observations (Holzer-Popp
et al.
, 2008) are also coming on stream. All these data
sources derive from polar orbiting satellites and are mostly
observations in shortwave (blue to ultraviolet). In addition,
20.1.2.3
Ground-based measurements and campaigns
In comparison with the coverage from satellite-derived
data, ground measurements and campaigns remain very
limited in their spatial and temporal coverage. One of
the most widely used and longest running data sets on
dust concentrations is the Miami Aerosol Group measure-
ments, although measurements from the Aerosol Robotic
