Influence of South Atlantic Sea Surface Temperatures on Rainfall Variability and Extremes over Southern Africa

Abstract It is generally agreed that changing climate variability, and the associated change in climate extremes, may have a greater impact on environmentally vulnerable regions than a changing mean. This research investigates rainfall variability, rainfall extremes, and their associations with atmospheric and oceanic circulations over southern Africa, a region that is considered particularly vulnerable to extreme events because of numerous environmental, social, and economic pressures. Because rainfall variability is a function of scale, high-resolution data are needed to identify extreme events. Thus, this research uses remotely sensed rainfall data and climate model experiments at high spatial and temporal resolution, with the overall aim being to investigate the ways in which sea surface temperature (SST) anomalies influence rainfall extremes over southern Africa. Extreme rainfall identification is achieved by the high-resolution microwave/infrared rainfall algorithm dataset. This comprises satellite-derived daily rainfall from 1993 to 2002 and covers southern Africa at a spatial resolution of 0.1° latitude–longitude. Extremes are extracted and used with reanalysis data to study possible circulation anomalies associated with extreme rainfall. Anomalously cold SSTs in the central South Atlantic and warm SSTs off the coast of southwestern Africa seem to be statistically related to rainfall extremes. Further, through a number of idealized climate model experiments, it would appear that both decreasing SSTs in the central South Atlantic and increasing SSTs off the coast of southwestern Africa lead to a demonstrable increase in daily rainfall and rainfall extremes over southern Africa, via local effects such as increased convection and remote effects such as an adjustment of the Walker-type circulation.