Flood Forecasting on the Tocantins River Using Ensemble Rainfall Forecasts and Real-Time Satellite Rainfall Estimates (PAP014692)

Código

PAP014692

Autores

Fernando Mainardi Fan, Walter Collischonn, KARENA QUIROZ JIMENEZ, Mino Viana Sorribas, Diogo Costa Buarque, VINÍCIUS ALENCAR SIQUEIRA

Tema

Flood forecasting and early warning systems

Resumo

The Tocantins River, located at the northern region of Brazil with over 300.000km² of drainage area, is a very important water body in terms of hydropower production, having six major hydropower plants on its main channel. The occurrence of floods along the Tocantins River is not only important for hydropower plant operation, but also is a relatively frequent event that affects several cities and their inhabitants. One recent example happened in 2012, when a flood with a relatively low recurrence time (approximately 2.5 years) caused damages to cities located close to the river. Motivated by this flooding susceptibility, a hydrological forecasting system was developed in order to assist the decision making of dam operation along main river axis, taking advantage of the existence of telemetric information from the hydropower plants. The forecasting system is being used operationally since mid-2012 and is based on the MGB-IPH model, a large scale distributed hydrological model. The model uses rainfall information from ground based telemetric gauges and real-time TRMM satellite rainfall estimates, which are merged together in order to reduce the impact of the lack of observed information in the basin. Streamflow forecasts are obtained based on quantitative precipitation forecasts from two different sources: (i) the Brazilian CPTEC Eta 15km regional deterministic model, and (ii) NOAAs Global Ensemble Forecasting System, maintained by the National Center for Environmental Prediction (NCEP-NOAA). All the forecasting system data management and operation is conducted through a specifically-designed computer interface, coupled with an open-access GIS platform. We present the forecasting system and show hindcasting analysis of how the 2011/2012 rainy season flood could have been predicted with the use of ensemble forecasts. We also compare results of deterministic and ensemble forecasts during the rainy season of 2012/2013.