Long-term dynamics of reservoirs across the globe

In many places of the world, reservoirs play an important role in relation to water security, flood risk, agriculture production, hydropower, hydropower potential, and environmental flows. By limiting the amount of water flowing out of the reservoir, reservoirs control flooding downstream, but they can also increase downstream runoff during drought.  

Detailed information about reservoir management (e.g. inflow, volume and outflow operations) is generally unknown or only available to the local control authority. As a result, large-scale information on reservoir dynamics is currently unknown. Recently, reservoir volume dynamics have been estimated from satellite observations based on reservoir surface area estimates. While Earth observation (EO) has the potential to monitor water from space and fill this gap, temporal resolution of these datasets generally varies between 3-7 days without direct information on reservoir inflow and outflow. Hydrological model reanalysis provides a complementary data source. Using cloud computing infrastructure and a high resolution distributed hydrological model wflow_sbm, we present a novel dataset of historical daily reservoir variations for 3236 headwater reservoirs across the globe for the period 1970-2020. Results derived with wflow_sbm model forced with various forcing sources based on observations and reanalysis (ERA5, EOBS, CHIRPS, NLDAS, BOM, MSWEP) are compared with: 1) measured discharge observations, 2) in situ reservoir elevation and volume measurement, and 3) volume estimates derived using satellite observations. Overall good comparisons between the hydrological model and the different measurement sources are observed, although considerable variations are observed. During the presentations we will zoom in on some of the large-scale changes in reservoir dynamics as observed in South America and Africa and how these potentially impact society.