Assessing flood risks in Burkina Faso’s urban centers using bias-corrected projected extreme precipitation data from CMIP6 models
- 2025 IEEE Multi-conference on Natural and Engineering Sciences for Sahel's Sustainable Development (MNE3SD) : 1-10
Résumé
In the context of escalating climate change impacts, particularly intensified extreme precipitation and flood risks in West Africa, this study addresses Burkina Faso’s vulnerability to hydroclimatic extremes. Focusing on four representative cities Ouagadougou, Bobo-Dioulasso, Fada-N’Gourma and Ouahigouya across three climatic zones, the research utilizes daily precipitation data from the Agence Nationale de la Météorologie (ANAM; 1981–2020) and simulations from 17 CMIP6 models for historical (1981–2014) and future SSP2-4.5/SSP5-8.5 scenarios. Objectives include comparing CMIP6 simulations with observations, selecting top-performing models via a weighted composite score, and projecting extreme indices (ETCCDI: Rx1day, Rx5day, P95, P99, CDD, CWD) and return levels (RL2, RL5, RL10) using Peaks-Over-Threshold with Generalized Pareto Distribution. Bias correction was applied using trend-preserving quantile mapping (CDF-t) with occurrence adjustment, calibrated over 1981–2000 and validated over 2001–2014. Based on this score, the three best-performing CMIP6 models for each city are MRI-ESM2-0, BCC-CSM2-MR, and NorESM2-MM for Bobo-Dioulasso; ACCESS-ESM1-5, CMCC-CM2-SR5, and CMCC-ESM2 for Ouagadougou; NorESM2-MM, CMCC-ESM2, and CMCC-CM2-SR5 for Ouahigouya; and CMCC-ESM2, BCC-CSM2-MR, and ACCESS-ESM1-5 for Fada-Ngourma. Results further show robust post-correction alignment of distributions and precipitations, with heterogeneous projections: moderate near-term (2021–2040) increases in extremes under SSP5-8.5, escalating mid- (2041–2060) and far-term (2081–2100), amplified by CWD persistence (e.g., +100% in Ouagadougou under SSP2-4.5), heightening urban flood risks. Findings inform adaptive policies, emphasizing runoff management. Future perspectives include high-resolution downscaling, spatial studies integrating Antecedent Precipitation Index for soil moisture, and soil texture variations to refine localized flood dynamics.
Mots-clés
Extreme precipitation, CMIP6 models, Flood risk, Burkina Faso, Climate adaptation