An Extreme Value Analysis of Drought in Africa

Droughts and their devastating consequences are most common in Africa. In this paper, we provide an extreme value analysis of drought indices for all counties in Africa covering the period 1993–2022. Extreme values were defined as annual minimums and annual maximums of drought indices. The generalized extreme value distribution and its reversed version were fitted to them with one of its three parameters allowed to vary linearly, quadratically, or cubicly with respect to year number. Ten of the countries with respect to annual maximums and 23 of the countries with respect to annual minimums were found to exhibit significant trends in drought indices. The trends were linear, quadratic, or cubic with respect to year. The adequacy of fits was assessed by probability plots, quantile plots, and the Kolmogorov-Smirnov test. The best-fitted models were used to give predictions for 2023–2030.