Climate Forecasting Application in Bangladesh: A New Vision to Manage Floods

A project called Climate Forecasting Application in Bangladesh (CFAB) has been implemented to study and provide short- and long-range forecasts of flood probability in the Bangladesh river delta. It is anticipated that this will allow remedial actions that can minimize the adverse impacts of flooding and provide opportunities to increase agricultural output.

CFAB aims at creating and increasing collaborations between international and national partners in flood forecasting, using state-of-the-art technology for forecasting, involving user communities and increasing dialogue between scientific and user communities. It also plans to develop techniques for the application of flood forecasts in sectors such as agriculture, precipitation forecasting and disease, and to build capacities within the Bangladesh Meteorological Department and wider regional community to produce the ensemble forecasts needed to support flood prediction.

The project is funded by NOAA/OGP, NSF and USAID/OFDA and is implemented in partnership with Bangladeshi and international organizations. Some of the partners are Flood Forecasting and Warning Centre (FFWC), Surface Water Modeling Centre (SWMC), Department of Agricultural Extension (DAE), Bangladesh Meteorology Department (BMD), and the Centre for Environment and GIS Support Project for Water Sector Planning (CEGIS), which together form the CFAB Steering Committee in Bangladesh. The Program in Atmospheric and Oceanic Sciences (PAOS) at the University of Colorado Boulder, the European Centre for Medium Range Weather Forecasts, and the Department of Meteorology, Florida State University Tallahassee, are the international partners.

Since the inception of the project, four trips to Bangladesh have been made. A workshop is planned in Bangladesh for December 2002, to share and evaluate the results of the experimental forecasts with the user community. These will be “hindcasts” for the years 1998, 1999 and 2000. With continued progress, it is anticipated that by 2003 “real-time” experimental forecasts will be generated.