The sensitivity of simulated temperatures in climate models to aerosols over southern Africa

Abstract

The purpose of this study is to investigate the sensitivity of Southern African surface temperatures to aerosols using two general circulation models (GCMs), namely the Goddard Institute for Space Studies (GISS) – modelE and the Conformal-Cubic Atmospheric Model (CCAM). Southern Africa’s seasonal surface temperatures have been simulated with and without the aerosol schemes for the period 2000 – 2014. The simulated surface temperatures obtained without the aerosol scheme were evaluated against gridded surface temperature observations from the Climate Research Unit (CRU). A temporal and spatial distribution of simulated aerosol optical depth at 550 nm (AOD550) were compared against AEROsol Robotic NETwork (AERONET) and satellite observations. In general, the GCMs adequately represent the seasonal surface temperature over Southern Africa. However, there is a general pattern in both GCMs of localized positive bias that ranges between 1.0 – 2.5˚C, over the western part of the domain, and a slight negative bias at the eastern part. It was found that both GCMs were able to capture the temporal and spatial distribution of AOD550 but underestimate the magnitude of AOD550 over Southern Africa. The results indicate that AOD550 over Southern Africa varies with respect to space and time, during September – November (SON) AOD550 reaches a maximum value of 0.8 while relatively low AOD550 occurs in March - May (MAM) season. Sensitivity results have shown that aerosols induce a surface temperature change that is spatially dependent over Southern Africa. Generally, negative changes in surface temperature that range between 0.25 – 1.5 °C and 1.0 – 3.5°C are induced for GISS modelE and CCAM respectively. However, there are localized positive changes in surface temperature of about 0.5 – 1.0°C that were simulated by GISS modelE in some parts of Southern Africa especially during the MAM season.