The influence of distributed input data on the hydrological modelling of monthly river flow regimes in West Africa

An extensive data collection exercise was undertaken for an area mainly covering Burkina Faso, Cote d'Ivoire and Mali. This exercise led to the development of three gridded monthly rainfall surfaces and three monthly potential evapotranspiration (PET) surfaces. Two semi-distributed hydrological models were used for the analysis. Both models utilize soil data which can be defined as being close to a Soil Water Holding Capacity (SWHC). For this, two gridded data sets were used. When applied to different catchments covering geographically close hydrographical areas, but sometimes with distinct characteristics, the models produced results that varied according to the input data sets, catchment characteristics and model structure. These preliminary results highlight the important influence of the nature and origin of input data for hydrological modelling in Africa.