Developed through a collaborative effort by the SERVIR-HKH Initiative, the Streamflow Prediction Tool for Bangladesh was used by the country’s Flood Forecasting and Warning Centre (FFWC) for inputs into its own in-house hydraulic model and flood mapping applications during the 2018 and 2019 floods, resulting in reliable baseline inputs and accurate forecasts. FFWC has also been using observational data from select locations to validate the generated forecasts.

This is an important tool given that Bangladesh experiences recurring floods during monsoon which heavily impact lives and livelihoods along the floodplains of the Ganges, Brahmaputra, and Meghna rivers. As these river systems originate outside Bangladesh’s national border and further upstream, FFWC does not have access to reliable river flow data that could help mobilize and prepare for the sudden onset floods. The web-based tool provides 10 to 15-day probabilistic forecasts for 21 locations along the Bangladesh border. This will aid the accuracy of FFWC’s flood forecasting models.

To build the FFWC staff’s capacity, we organized multiple training events on the development and use of the tool, data processing requirements, and validation of generated forecasts. We also helped FFWC organize a pre-monsoon stakeholder consultation workshop in 2019 (expected to be held annually) to improve flood forecasting. Further, an android-based mobile application – BWDB Flood Apps – has been developed to make flood warning information more accessible to the public.

The Streamflow Prediction Tool – developed in collaboration with Brigham Young University, the National Aeronautics and Space Administration’s (NASA) Jet Propulsion Laboratory, and FFWC – has also been customized for use in Nepal and Bhutan. We are now working with national partners on joint validation for the tool’s adoption and use.