Résumé
Casamance, Gambia and Senegal rivers basins in West Africa
The challenges in accessing reliable rainfall and hydrometric gauge data in Africa, gridded precipitation products (satellite, reanalysis, and in situ) provide exciting opportunities for hydrological studies. However, these products are indirect measurements and this requires hydrological validation of these products in relation to ground observations in order to guarantee their relevance and reliability. This study assesses the ability of twenty-three gridded precipitation products to develop monthly (GR2M) and daily (GR4J) rainfall-runoff models to simulate streamflows in the Senegal, Gambia, and Casamance river basins. The Kling-Gupta Efficiency (KGE) and Percentage Bias (PBIAS) metrics were used to evaluate catchment rainfall and discharge derived from the 23 products against in situ observations of rainfall and streamflow.
Multi-source datasets, integrating satellite, reanalysis, and in situ observations, provide better performance for streamflow simulation at both monthly and daily time steps. Among the evaluated products, IMERGDF emerges as the most reliable, followed by MSWEP, GPCP, and TAMSAT. Conversely, CPC, which relies on interpolated ground data, exhibited unexpectedly poor performance. These findings highlight the remarkable performance of certain datasets including at a daily time step, establishing their ability to simulate runoff in poorly gauged catchments. In the event of inaccessibility or unavailability of observed data, the best products of this study can be used for hydrological applications in Senegal's main hydrosystems.
[Display omitted]
•Twenty-three gridded precipitation products are used as inputs to the GR2M and GR4J models to simulate flows.•IMERGDF, MSWEP, TAMSAT and GPCP are the best performing products for flow simulation.•These are composite products combining several data sources (observed, satellite and reanalysis).