Verification of inflow into hydropower reservoirs using ensemble forecasts of the TIGGE database for large scale basins in Brazil

• Ensemble forecasts assessment for three large scale basin reservoirs in Brazil.
• First use of TIGGE archive for hydrological ensemble forecasting in South America.
• Evaluation of different meteorological models for multiple large scale basins.
• Assessment on the use of combined meteorological models in a Super Ensemble.
• Discussions on the forecasts potential benefits for hydropower and flood control.

Study regionThis paper describes a major ensemble-forecasts verification effort for inflows of three large-scale river basins of Brazil: Upper São Francisco, Doce, and Tocantins Rivers.Study focusIn experimental scenarios, inflow forecasts were generated forcing one hydrological model with quantitative precipitation forecasts (QPF) from three selected models of the TIGGE database. This study provides information on the regional ensemble performance and also evaluates how different QPF models respond for the different basins and what happens with the use of combined QPF in a greater ensemble.New hydrological insights for the regionThis work presents one of the first extensive efforts to evaluate ensemble forecasts for large-scale basins in South America using TIGGE archive data. Results from these scenarios provide validation criteria and confirm that ensemble forecasts depend on the particular EPS used to run the hydrological model and on the basin studied. Furthermore, the use of the Super Ensemble seems to be a good strategy in terms of performance and robustness. The importance of the TIGGE database is also highlighted.