Modeling of Saharan dust outbreaks over the Mediterranean by RegCM3: case studies

Abstract. The regional climate model RegCM3 coupled with a radiatively active aerosol model with online feedback is used to investigate direct and semi-direct radiative aerosol effects over the Sahara and Europe in a test case of July 2003. The aerosol model includes dust particles in addition to sulfates, hydrophobic and hydrophilic black carbon and organic carbon. The role of the aerosol online feedback on the radiation budget and the direct radiative forcing (short-wave and long-wave) by dust particles are investigated by intercomparing results from three experiments: REF, including all interactive aerosol components, Exp1, not accounting for the aerosol radiative feedback, and Exp2 not accounting for desert dust particles. The comparison of results in the REF experiment with satellite observations, sun/sky radiometer measurements, and lidar profiles at selected Central Mediterranean sites reveals that the spatio-temporal evolution of the aerosol optical depth is reasonably well reproduced by the model during the entire month of July. Results for the dust outbreaks of 17 and 24 July, averaged over the simulation domain, show that the daily-mean SW direct radiative forcing by all particles is −24 Wm⁻² and −3.4 Wm⁻² on 17 July and −25 Wm⁻² and −3.5 Wm⁻² on 24 July at the surface and top of the atmosphere, respectively. This is partially offset by the LW direct radiative forcing, which is 7.6 Wm⁻² and 1.9 Wm⁻² on 17 July and 8.4 Wm⁻² and 1.9 Wm⁻² on 24 July at the surface and top of the atmosphere, respectively. Hence, the daily-mean SW forcing is offset by the LW forcing of ~30% at the surface and of ~50% at the ToA. It is also shown that atmospheric dynamics and hence dust production and advection processes are dependent on the simulation assumptions and may significantly change within few tens of kilometers. The comparison of REF and Exp1 shows that the aerosol online feedback on the radiation budget decreases the domain-average daily-mean value of the 2 m-temperature, aerosol column burden (CB), and short-wave (SW) atmospheric forcing by −0.52 °C, 14%, and 0.9%, respectively on 17 July and by −0.39 °C, 12% and 12%, respectively on 24 July. The comparison of REF and Exp2 reveals that on 17 July, radiatively-active dust particles decrease the daily-mean 2 m-temperature averaged over the whole simulation domain by 0.4% even if are responsible for 99.8% and 97% of the daily-mean aerosol column burden and SW atmospheric forcing, respectively.