Author(s): Maurizio Mazzoleni; Biswa Bhattacharya; Miguel Angel Laverde Barajas; Dimitri Solomatine

Linked Author(s): Dimitri Solomatine

Keywords: Brahmaputra basin; Data fusion; Distributed hydrological modelling; Flood prediction; Rainfall remote sensing

Abstract: An important aspect in hydrological modelling is the accurate quantification and prediction of rainfall. In ungauged or poorly gauged basins ground data is sparse and often is complemented by rainfall satellite products, which brings additional uncertainties. The main objective of this research is to assess performance of distributed hydrological models using the remotely sensed rainfall estimates as forcings for the model. The model, based is based on the conceptual HBV-96 model and the PCRaster framework, is implemented for the Brahmaputra basin. Three different remote sensed datasets of precipitation (MSWEP, TMPA and PERSIANN-CDR) are used. Simple fusion methods are used to combine models results generate by the dataset of precipitation. The preliminary results of this study show that better model results are achieved merging the output results. Using MSWEP and TMPA as the forcing data provides satisfactory model results. On the other hand, use of PERSIANN-CDR leads to better prediction of flow peaks but overestimations of the hydrographs’falling limbs.

DOI:

Year: 2018