Evaporation Estimation over the Upper Blue Nile Basin By Combining Satellite Observations and River Flow Gauges

Abstract:

Remote sensing and satellite data have enormously progressed in the last decade and became a potential data source to better model land and water resources spatially and temporally. When several global satellite evaporation datasets are compared over the upper Blue Nile (UBN) basin in Ethiopia, it was found that they do not agree temporally nor spatially. All the datasets underestimate evaporation and thus closure of the water budget is not achievable with errors in the order of 30% of the mean annual precipitation. This study presents a method that combines readily available global satellite and river flow measurements to estimate actual and potential evapotranspiration in the UBN basin. The method used is a model that assimilates satellite measurements of TRMM precipitation, WM potential evapotranspiration, GRACE terrestrial water content and flow gauge stations’ measurements into a one-layer soil water balance model. The model is constrained with mass conservation law, topographic setting and its impact on flow routing and soil water holding capacity. The constrained model produces physically consistent long-term spatial and temporal distributions of the hydrological cycle components over the UBN basin. The basin is then divided into smaller sub-basins to investigate evapotranspiration trends. The method described here is applicable to other data-sparse basins facing future challenges of water scarcity and climate change.