Soil Dust Modeling-Feedbacks of Dust and Climate

.Soil dust aerosol is emitted from desert surfaces as well as from soil surfaces that have been disturbed by human impact, like overgrazing and agriculture in semi-arid regions, or by deforestation. Since the atmospheric loading of soil dust can be very high, it is expected to impact climate at least regionally. And since the magnitude of dust deflation depends on climate variables like precipitation and wind speed, feedback processes between dust production and climate need to be taken into account to understand the impact of dust on the climate. To study the sensitivity of the climate to soil dust forcing and feedback of dust forcing upon dust production and transport, dust needs to be included as dynamic and radiatively interactive tracer in the GISS GCM. To assess the sensitivity of the climate response to dust radiative properties (which is not very well known on the global scale), the GCM was integrated using different dust single scattering albedos, and in a control case dust was included as dynamic tracer without interaction with radiation. These integrations were carried out with fixed sea surface temperature as well as mixed layer ocean as boundary conditions.

The results show that soil dust radiative forcing generally causes a negative feedback on atmospheric dust loads, i.e. a reduction in the dust load corresponding to a decrease in dust source flux. This reduction varies both with the season and with the single scattering albedo of the dust, depending on the region where dust deflation takes place. The climate response and dust feedback to dust forcing turns out to be highly sensitive to the dust single scatter albedo. Also, the results indicate areas, where changes in dust deflation is controlled by changes in surface winds, and those areas, where dust deflation is controlled by changes in the hydrological cycle. Discrepancies of the model results with observed interannual variations in dust loads can indicate regions, where dust deflation was additionally influenced by changes in soil surface conditions