Flood frequency analysis in West Africa in a climate change context

Floods are a recurring and devastating hazard in West Africa, with significant socio-economic and environmental impacts. A better understanding of their frequency and magnitude is crucial for effective flood risk mitigation, infrastructure design, and water resource management. The lack of reliable hydrometric datasets has hitherto been a major limitation in flood frequency analysis at the scale of West Africa. We combine insights from historical flood frequency analysis and future climate-driven flood projections to provide a more complete description of flood hazards in West Africa. Using a newly developed African hydrological database, annual maximum flow (AMF) time series from 246 river basins (1975–2018) were analyzed with the Generalized Extreme Value (GEV) and Gumbel distributions. The GEV distribution, paired with the Generalized Maximum Likelihood Estimation (GMLE) method, yielded the best results for quantile estimation, enabling the generation of regional envelope curves for the first time in West Africa. Future flood trends have been assessed from the OS LISFLOOD and the HMF-WA large-scale distributed hydrological models, driven by five bias-corrected CMIP6 climate projections under the SSP2-4.5 and SSP5-8.5 scenarios. Both hydrological models consistently projected increases in flood frequency and magnitude across West Africa, despite their differences in hydrological processes representation and calibration schemes. Flood magnitudes are projected to increase in 94% of stations, with some areas experiencing increases exceeding 45%. Significant trends are already observable in many basins as early as the 1980s, emphasizing the robust climate change signal in this region. This combined approach, integrating historical flood frequency analysis with future climate-driven projections, offers critical regional-scale insights into the evolving flood hazards in West Africa.