Potential of Satellite Rainfall Estimates as Inputs for Flood Forecasting: Case Study Gash River, Sudan

Abstract

This study is an attempt to test the potential of using real time Satellite Rainfall Estimates (SRE) data for hydrological modeling. Tropical Rainfall Measurement Mission (TRMM-3B42RT V7) SRE was evaluated against observed rain gauge data in Gash river catchment. The TRMM was evaluated against intensity as well as elevation dependency. The Hydrologic Engineering Center’s Hydrologic Modeling System (HEC-HMS) software of the Army Corps of Engineering of the USA was used to simulate the rainfall - runoff process. The performance of TRMM was found to underestimate the rainfall for most of the events, the underestimation increases with the increase in elevation. TRMM data set was biased corrected and used as input to derive the hydrological model. Observed hydrographs at the catchment outlet were compared to the simulated flow hydrographs using events and continuous modeling. The results of hydrological modeling showed that events based modeling performed better, the coefficient of determination (R2) vary between 0.87 to 0.96 while Nash-Sutcliffe Efficiency (NSE), vary from 0.84 to 0.96 and Root Mean Square Error (RMSE) vary from 45 to 118.3 m3/s. While the same statistics for continuous modeling showed, (NSE = 0.65) and (RMSE 44.5 m3/s). These results reflect the high potential of TRMM data set as inputs for hydrological modeling and flood forecasting in the Gash river and other basins with similar characteristics.