Aerosol radiative forcing during African desert dust events (2005–2010) over South-Eastern Spain
1Departamento de Física Aplicada, Universidad de Granada, Granada, Spain
2Centro Andaluz de Medio Ambiente (CEAMA), Granada, Spain
Abstract. The instantaneous values of the aerosol radiative forcing (ARF) at the surface and the top of the atmosphere (TOA) were calculated during desert dust events occurred at Granada (Southeastern Spain) from 2005 to 2010. For that, the SBDART radiative transfer model was utilized to simulate the global irradiance values (0.3–2.8 μm) at the surface and TOA using as input the aerosol properties derived from a CIMEL sun-photometer measurements and an inversion methodology that uses the sky radiance measurements in principal plane configuration and non-spherical particle shapes approximation. The SBDART modeled global irradiances at surface have been successfully validated against experimental measurements obtained by CM-11 pyranometer, indicating the reliability of the radiative transfer model used in this work for the ARF calculations.
The monthly ARF values at surface ranged from −32 W m−2 to −46 W m−2, being larger in April and July than in the rest of months. The seasonal ARF evolution was inconsistent with seasonal aerosol optical depth (AOD) variation due to the effects induced by other aerosol parameter such as the single scattering albedo. The ARF at TOA changed from −9 W m−2 to −29 W m−2. Thus, the atmospheric ARF values (ARF at TOA minus ARF at surface) ranged from +15 to +35 W m−2. These results suggest that the African dust caused local atmospheric heating over the study location.
The instantaneous aerosol radiative forcing efficiency (ARFE), aerosol radiative forcing per unit of AOD (440 nm), at surface and TOA during African desert dust events was evaluated according to the desert dust source origins. The ARFE values at surface were relatively high (in absolute term) and were −157 ± 20 (Sector A), −154 ± 23 (Sector B), and −147 ± 23 (Sector C) W m−2. These values were larger than many of the values found in literature which could be due to the presence of more absorbing atmospheric particles during African desert dust intrusions over our study area. Finally, our ARF computations showed good agreement with the corresponding ARF calculated by AERONET network.