Accuracy of forecasted rainfall is crucial for forecasting of flood and inundated area in a river basin. A network of 49 automated rain recording stations exist in the Indian part of Rapti basin and provide reliable spatially distributed rainfall over the basin. In the current monsoon season, Rapti basin experienced one day maximum rainfall over the basin on 13th July 2019, followed by second maximum rainfall on 9th July 2019. Rainfall forecast using Global Forecast System (GFS) and Weather Research and Forecasting (WRF) models are used extensively for research and real-time forecasting throughout the world. These models are numerical weather prediction system designed to serve both atmospheric research and operational forecasting needs. For study area, WRF data was available at a resolution of 3km × 3km and GFS data was available at 25 km × 25 km resolution. Global Precipitation Measurement (GPM) (10kmx10km) data, which is satellite based precipitation information, and represent real time rainfall, has also been extracted for the same dates. A comparison of observed rainfall, forecasted rainfall generated through WRF and GFS, and real time satellite based GPM rainfall has been made. Results show that GPM rainfall reasonably matches with observed rainfall for both events. In case of forecast rainfall, WRF rainfall was relatively closer to observed rainfall for 13th July 2019, but it did not match with rainfall occurred on 9th July 2019. GFS data performed very poor for both rainfall events. This study highlights that accuracy of forecasted rainfall is still a serious concern as input in the hydrological models to generate reliable flood forecast. There is need to generate more accurate rainfall forecast for Indian basins.
GFS data, WRF Data, GPM Data, Rainfall forecasting, Flood forecasting
