Experimental determination of short- and long-wave dust radiative effects in the Central Mediterranean and comparison with model results

• Desert dust impact on short- and long-wave irradiance measurements.
• Experimental determination of desert dust radiative forcing.
• Surface albedo impact on the dust radiative forcing daily cycle.
• Land surface temperature and long-wave radiative forcing relationships.
• Sensitivity of modeled dust radiative forcing to refractive index values.

Downward and upward irradiance measurements, in the short-wave (SW) and long-wave (LW) spectral range, have been used in combination with simultaneous aerosol optical depths (AODs) to experimentally determine the instantaneous and clear-sky aerosol Direct Radiative Forcing (DRF) at the surface, during a desert dust outbreak which affected the Central Mediterranean from 9 to 13 July 2012. AODs were retrieved from AERONET (AErosol RObotic NETwork) sun/sky photometer measurements collocated in space and time. The importance of downward and upward radiative flux measurements to properly account for both the surface albedo dependence on the solar zenith angle, and the land surface temperature (TLS) has been highlighted. Measured radiative fluxes were in reasonable agreement with the CERES (Clouds and the Earth's Radiant Energy System) and AERONET corresponding ones collocated in space and time. SW and LW downward fluxes at the surface decreased up to 9% and increased up to 13%, respectively, as a consequence of a factor 5 increase of the AOD at 675 nm (AOD675). This is due to the cooling and warming effect of desert dust in the SW and LW spectral range, respectively. In fact, we have also found that the TLS increased at a rate of about 250 K per unit increase of the AOD675. The aerosol DRF at the surface varied from − 8 to − 74 W m− 2 and from + 1.2 to + 9.6 W m− 2 in the SW and LW spectral domains, respectively. In particular, we have found that the LW-DRF on average offsets 14% of the related SW component. It is shown that a two-stream radiative transfer model can reproduce the experimental findings at the surface by replacing the refractive indices typical of dust particles with the ones obtained for a mixture made of dust and soot particles. The dust contamination by anthropogenic particles during its transport to the monitoring site located several hundred kilometers away from the source region was responsible for this last result. We have also found by model simulations that the LW-DRF increased linearly with TLS both at the surface and at the top of the atmosphere.