| River modeling is an important component of flood forecasting system that can simulate the water flow dynamics of a stream network and forecast river levels in flood prone regions. Around the world, especially in the developing regions, many large river basins are mostly ungauged. For these basins river model setup is very challenging due to lack of necessary in-situ and routine measurement of river bathymetry, flood plain and river boundary data. Moreover, lack of data sharing among the countries occupying the trans-boundary rivers, also a hurdle to river model development. For such basins, proxy approaches depending on the satellite based remotely sensed data could be an alternative solution. In this study, one dimensional hydrodynamic model has been developed for the Ganges, Brahmaputra and Meghna basin region using the Hydrologic Engineering Center River Analysis System (HEC-RAS). Only 7% of the total basin area has good quality in-situ measurement of river hydraulics. For the remaining part, remotely sensed data have been utilized for river model development. This study utilized: a) LANDSAT/MODIS for identifying flow path of river network, b) Shuttle Radar Topographic Mission (SRTM) for extracting river profile c) Radar altimeter for establishing depth-width relationship and d) Precipitation data to generate sub basin wise flow data. Simulated model results have been tested at two downstream low lying locations. The outcome of the study showed significant improvement of root mean square error (RMSE) for river level simulation from 3.0m to 1.0m. A step by step `rule book' has been documented to facilitate the setting up river models for similar type basins around the world for operational water agencies. |
