The superior influence of Darwin Sea level pressure anomalies over ENSO as a simple drought predictor for Southern Africa

Summary The dominant climatic mode responsible for seasonal rainfall variability across central southern Africa has been well-established as ENSO. Hence, the El Niño signal of the equatorial Pacific has been used extensively to predict droughts in this sub-region. Although this paper acknowledges that El Niño influences rainfall deficits over eastern southern Africa, an earlier signal of extreme positive sea level pressure (SLP) anomalies at Darwin for the averaged March to June period (MAMJ Darwin) has proved to have a superior remote connection to droughts in the sub-region. Simple linear statistical tools including composite techniques and correlation methods have been employed on century long data sets (1901–2000) to identify the emerging paramount connection between MAMJ Darwin SLP anomalies and southern African rainfall. Both MAMJ Darwin SLP anomalies and the Zimbabwe seasonal rainfall time series are significantly correlated (above the 95% significant level) with sea surface temperature anomalies. These represent the Indian Ocean Dipole mode in the tropical Indian Ocean and ENSO in the tropical Pacific for the averaged September to December period. ‘Pure’ MAMJ Darwin (that occur in the absence of El Niño in the Pacific) coincide with droughts more significantly (83% hit rate) than ‘pure’ El Niño events (not preceded by a high MAMJ Darwin) (38% hit rate). Co-occurrences (MAMJ Darwin preceded by El Niño) do not only have the highest hit rate of 93% but subsequent droughts are noticeably more severe. The ‘pure’ El Niños however, are not only poorly related to Zimbabwe seasonal rainfall deficits, but are apparently not connected to extreme droughts of the 20 th century. Thus, MAMJ Darwin is a good simple predictor of droughts associated with or without ENSO in the Pacific. The high prediction skill of these results, especially the inherent longer lead-time than ENSO, makes MAMJ Darwin SLP anomalies an ideal additional input candidate for sub-regional drought monitoring and forecasting schemes. In this way, drought early warning and disaster preparedness activities can be enhanced over the sub-region.