Operational Dust Prediction - Mineral Dust: A Key Player in the Earth System - page 244

2012a ). Another limitation is the low aerosol detectability over bright surfaces,
which affects instruments operating in the visible part of the spectrum. The new
generation of high-resolution IR spectrometers and interferometers on polar orbiting
satellite platforms (e.g. AIRS, IASI) has the potential to provide good quality dust
information (Hilton et al. 2012 ). Algorithms are currently being developed and
validated (Peyridieu et al. 2010 ; Klüser et al. 2011 ), and it is likely that these
products will become prominent both for evaluation and assimilation.
Regions with air quality monitoring networks are the main surface data source
for point evaluation of dust concentrations predicted by dust models. As with the
satellites, air quality measurements integrate the contribution of the different types
of atmospheric aerosol. Furthermore, observational values are usually limited to the
concentration of particulate matter with an aerodynamic diameter less than 10 m
(PM10), which does not always encompass the full size range of dust particles
suspended in the atmosphere. Finally, it is important to consider the selection of
stations, since many of them are located in cities, industrial parks or roads, where
local human activity is the main source of particles, obscuring the contribution of
dust to measured quantities.
Visibility data included in meteorological observations have sometimes been
used as an alternative source of information (Shao et al. 2003 ). Visibility is mainly
affected by the presence of aerosol and humidity in the atmosphere. If visibility
is recorded manually, illumination is also important for intermediate visibilities.
Therefore, the use of visibility data must be complemented with information on
present weather to discard those cases where visibility is reduced by the presence
of hydrometeors (fog, rain, etc.). Several empirical relationships between visibility
and dust surface concentration can be found in the literature (d'Almeida 1986 ;Ben
Mohamed et al. 1992 ;Shaoetal. 2003 ). However, the validity of these relationships
is very limited because the visibility reduction depends not only on the dust mass
concentration but also on the size spectrum of particles, as well as their density,
chemical and mineralogical composition and atmospheric humidity.
Direct-sun photometric measurements are a powerful remote sensing tool that
provides retrieval of column-integrated aerosol microphysical and optical proper-
ties. In particular, AERONET is a comprehensive set of continental and coastal
sites complemented with several sparsely distributed oceanic stations that provides
large and refined data sets in near real-time (Holben et al. 1998 ; Dubovik and
King 2000 ). AERONET measurements are by far the most commonly used in dust
model evaluation. Integral parameters such as AOD are complemented with spectral
information, which permit retrieval of aerosol microphysical and composition
properties (Dubovik et al. 2002 ). A major shortcoming of these measurements is
their unavailability under cloudy skies and during night-time.
Finally, lidar and the most recent generation of ceilometers permit routine
measurement of aerosol vertical profiles. However, continuous measurements in
ground-based stations are only performed in a few stations that are, in general, far
from the main dust sources. On the other hand, space-borne lidars (e.g. CALIOP)
provide global spatial coverage, but their temporal coverage is limited.