Improving Rainfall Estimates for Flood Forecasting in Cambodia

Period of Performance

May 2017 - September 2018

Through the Improving Rainfall Estimates for Flood Forecasting in Cambodia Service, SERVIR Mekong supported the Department of Hydrology and River Work (DHRW) of the Ministry of Water Resources and Meteorology (MOWRAM) in Cambodia to improve flood forecasting efforts. This service worked with DHRW to incorporate the Virtual Rain and Stream Gauge Information System (VRSGIS) data stream into their hydrological models. The VRSGIS data stream contained high-resolution satellite precipitation datasets, critical for regions where ground observations are sparse and precipitation information is needed to generate flood forecasts. These included: 

  • Climate Hazards Group InfraRed Precipitation with Station (CHIRPS)
  • Global Precipitation Measurement (GPM) mission
  • Tropical Rainfall Measurement Mission (TRMM)
  • NOAA Climate Prediction Center Morphing Technique (CMORPH)
  • Integrated Multi-Satellite Retrievals for GPM (IMERG)

These data streams fed into existing hydrological models and a new modeling system developed by MOWRAM through an Asian Development Bank project. Addressing data sparsity through the addition of near real-time precipitation data improved the quality of satellite rainfall estimates that fed into existing hydrological models in Cambodia to generate better early warnings for floods.


Flash flooding caused by monsoon rains and tropical storms, along with frequent seasonal flooding are recurrent issues in lower Mekong countries. To monitor, prepare for, and respond to flood risk, national and regional level organizations and agencies require a range of information, forecasts, and decision-support tools. By working with the DHRW to calibrate their system through incorporating reference data and ground validation, the staff will be able to use, maintain, and adapt the system and understand the benefits of this new method of estimating rainfall and river levels for their modeling systems. Inserting additional data provided from the VRSGIS into Cambodia’s hydrological models will help generate improved flood forecasts for use by MOWRAM and the National Committee for Disaster Management to distribute for disaster mitigation, appropriate reservoir management, and early response action at the community level.