Evaluation of extreme precipitation events as simulated by CMIP6 models over Central Africa: spatial variability

Abstract

Extreme precipitation events have substantial implications for water resources, ecosystems, and human populations in Central Africa (CA). Consequently, understanding the spatial variability of these events is crucial for effective climate change adaptation and water management strategies. In this study, we assess the performance of the state-of-the-art global climate models from Phase 6 of the Coupled Model Intercomparison Project (CMIP6) in simulating extreme precipitation events over CA. By considering three observational datasets, we evaluated the ability of sixteen CMIP6 models as well their multi-model ensemble-mean (MME), to capture the patterns of extreme precipitation events. We then focus on key metrics such as duration and intensity, based on a total of ten indices of extreme precipitation events, defined by the Expert Team on Climate Change Detection and Indices (ETCCDI). The results showed that the individual models as well as the MME exhibited acceptable performance in reproducing the patterns of extreme precipitation events, especially when dealing with wet-day amount, frequency, dry spells duration and persistence of extreme precipitation. Moreover, the analyses revealed that CMIP6 models generally have great difficulty in simulating not only the wet spells duration, but also heavy precipitation indices, highlighting the remaining challenges in reproducing local-scale processes governing precipitation variability and extremes precipitation events over the region. The results of this study provide an understanding of both the strengths and limitations of the CMIP6 models over CA, which would help to improve regional climate projections and strengthen the ability of decision makers to assess the future risks associated with extreme precipitation events in the region.