Radiative Effects of Dust - Mineral Dust: A Key Player in the Earth System - page 278

11.6 Implications for Including the Radiative Impact

of Dust in Models

It is important that we have confidence in our ability to model the interaction of
dust with radiation, since dust responds to climate change induced by other forcings
and is therefore an important feedback mechanism on a wide range of spatial and
temporal scales (Chaps. 13 , 16 , 17 and 18 ) . Changes in soil moisture, vegetation
cover, source areas and circulation patterns could result in significant changes in
dust concentrations. Recent assessments of the anthropogenic contribution to dust
lie around 25 % globally (Ginoux et al. 2012 ). 8 % of this comes from the Sahel,
and 13 % from Australia (although 75 % of Australia's total dust is anthropogenic).
Dust sources created by cultivation may increase in the future as the need for food
rises with the growing population. Thus, dust forcing may increase due to higher
dust burden in the atmosphere. Additionally, anthropogenic aerosol emissions are
projected to decrease in the future (Lamarque et al. 2011 ), although the extent to
which these projections will be realised is a matter of some debate.
The current approach taken to include the radiative impact of dust in models
depends on the application and the level of complexity of the dust scheme in the
radiation, climate or NWP model concerned (see Chaps. 9 and 10 for the latter
two). In general, all models require some specification of scattering and absorption
efficiency and phase function or asymmetry parameter. Since these properties are a
function of size, many climate models with interactive dust schemes use size-based
pre-computed look-up tables of optical properties. Thus, the optical properties of
dust can to some extent evolve within the model in so far as the size distribution
changes. However, most models use only a limited number (or even one) of
assumption of refractive index, which implies a uniform dust composition. Studies
such as Su and Toon ( 2011 ) and Moosmüller et al. ( 2012 ) suggest that this is likely
to be inappropriate globally or even regionally, but to do otherwise is likely to be
computationally expensive as models will need to track the transport of individual
minerals or at least the hematite fraction. Current research can more realistically
assess the sensitivity of radiative effects, temperature changes, heating rate changes
and other climate variables to the assumptions made concerning optical properties.
Once this has been examined systematically in both weather and climate models
and the results evaluated against observations, we will have a clearer idea of the
properties and processes that need further observation or understanding.
The increasing numbers of in situ observations of dust radiative properties
such as highlighted in this chapter and the size distribution and composition
understanding discussed elsewhere in this volume clearly go some way to providing
data for model evaluation (e.g. Greed et al. 2008 ). However, for many areas of
the globe, the best prospect for enhanced observations is via satellites, not least
because many of the regions where dust is the primary aerosol are logistically and
physically challenging places to work. This approach will have its own challenges,
not least the inconvenient fact that dust sources tend to be located within unvegetated
desert regions where the surface is bright and satellite retrievals are difficult.