This study aims to analyze and compare the performance of two regional climate models (RegCM4.5 and WRF3.5) in simulating extreme rainfall over West Africa. We performed two simulations respectively at a spatial resolution of 50 km with the RegCM4.5 model at a spatial resolution of 12 km with the WRF3.5 model. These runs cover the period 1981-2010 and the driving fields (lateral boundary conditions) are from the Era-Interim reanalysis. The RegCM4.5 model simulates dry (wet) biases over the Sahel (Guinea Coast) while the WRF3.5 model simulates an opposite bias. This could be explained partly by the fact that the RegCM4.5 (WRF3.5) model underestimates (overestimates) the relative humidity and the monsoon flow over the Guinea Coast compared to the ERA-Interim reanalysis. Results also show that the spatial distribution and the annual cycle of rainfall over West Africa are well simulated by the two regional climate models despite the presence of some biases. The number of rainy days decreases from the southern to the northern Sahel for CHIRPS data and both regional climate models. All datasets show the highest rainfall intensities and the strongest values of the intense rainfall events over the Fouta Jallon highlands, Jos Plateau and Cameroun Mountains. The maxima of the mean 95th percentile of daily rainfall is located over the Guinea zone for CHIRPS datasets and both regional climate models. All datasets show a spatial distribution of the consecutive wet days similar to the number of rainy days with strong values over the orographic regions. When considering the consecutive dry days, all datasets exhibit strong values of this parameter north of 17.5°N (northern Sahel). The shorter consecutive dry days are observed over the area of the maximum precipitation (over the southern Sahel and the orographic regions). In terms of model biases, this study shows substantial differences between the two regional climate models used in this study suggesting the necessity to perform models intercomparison during the present-day before any choice for future projections.

The objective of this study is to estimate the potential changes in extremes rainfall over three orographic regions of Africa (Guinea, Nigeria and Cameroon) under the 1.5°C and 2°C warming scenarios. The results project an increase of the number of rainy days over Guinea under both scenarios by 2059. Over Nigeria, this index will decrease in the north of the country under both scenarios. Over Cameroon, this index will decrease over the whole country under the 1.5°C warming and will increase in the southeast and northeast parts of the country under the 2°C warming. Rainfall intensity will increase over the northern and southwestern parts of Guinea under both scenarios. The maximum one-day rainfall accumulation will increase in almost all three considered countries in the near future. However, compared to Cameroon, the maximum one-day rainfall accumulation will be more important in the northern and central parts of Guinea and Nigeria. There will be a decrease of the number of consecutive wet days in the southwest and central parts of Guinea under the 1.5°C scenario and over most parts of Guinea under the 2°C scenario. Over Nigeria, this index will decrease in the south and center of the country. The number of consecutive wet days will decrease in the southwest and northern Cameroon and increase slightly in the southeast of this country under both warming scenarios. The number of consecutive dry days will decrease over the whole Guinea, over the northern and central parts of Nigeria under both scenarios. The increase noted in extreme events could cause flooding. This work can be considered as a support for the Guinean, Nigerian and Cameroon policymakers for implementing sustainable mitigation and good adaptation measures to limit and combat the adverse effects of climate change (floods, heat waves and drought). Finally, this study shows that the three considered regions will need additional irrigation as an adaptation strategy for the protection of the agricultural sector by 2059 under both scenarios due to the decrease of the number of consecutive wet days.