A Scalable Earth Observations‐Based Decision Support System for Hydropower Planning in Africa

Abstract Hydropower is a key part of the increasing shift in power production from nonrenewables to renewable energy. In regions such as Africa, hydropower reservoirs are vital for achieving several sustainable development goals, including clean water, energy, and poverty elimination. However, the operations of hydropower reservoirs are often suboptimal due to the lack of hydrologic data for generating reliable inflow forecasts. Here, we present a decision support system (DSS) framework for hydropower planning at daily to seasonal time scales by combining data from earth observation satellites (EOS) with ensemble climate forecasts from dynamical models and hydrologic modeling. The large uncertainty inherent in satellite‐based datasets is overcome by using a data validation framework which does not require ground‐based measurements. In addition, an EOS evapotranspiration product is used as a proxy for streamflow in calibrating hydrologic models. Compared to a DSS forced with a climatological forecast (zero‐skill), the hydropower production with the new DSS increased by 20%. The study highlights the advantage of using data from EOS in overcoming the issue of data scarcity in water resources applications, particularly in developing regions of the world such as Africa.