• Real-time Global Flood Analyses and Forecasts Using Satellite Rainfall and Hydrological Models
Program Disasters Program
PI / Institution Robert Adler / U. of Maryland
Start Date October 1, 2012
End Date September 30, 2013
  • Summary

    Global flood model products will be developed, tested and implemented to produce a real-time system running routinely at the NOAA/NESDIS Cooperative Institute for Climate and Satellites (CICS) at the University of Maryland. This will allow for continued use and evaluation by a wide array of users and also serve as a key step into full integration into the NOAA NESDIS operational system. Development and evaluation will be accomplished in close coordination with the user community in the U.S. and internationally, with the DoD-sponsored University of Hawaiis Pacific Disaster Center (PDC) providing a key link to the international disaster management community. The first (feasibility) year of the proposal will be spent examining various elements and designs to be included in the development and implementing a preliminary, incomplete version of this new Global Flood Monitoring System (GFMS) system in the NOAA/CICS computer system. The implementation and testing of the full system will be made in proposal years 2-4. The current UMD/Goddard real-time global flood model products (at 1/8th degree) already provide information useful to national and international agencies in understanding the intensity, timeline and impact on populations of these significant flood events. The quality of such applied hydrological estimations should improve under this proposal due to continuation and improvement of multi-satellite precipitation observations through NASAs Global Precipitation Measurement (GPM) mission and the further development of the hydrological models. Improvements will include 1) better precipitation analyses utilizing space-time interpolations and improvements for shallow, orographic rainfall systems, 2) higher resolution flood models with accurate routing and regional calibration, 3) improved use of forecast precipitation to augment the satellite-observed estimates and extrapolate the flood and estimates for 1-5 day forecasts, 4) identification of basins affected by presence of major dams and modification of flood calculations in these locations based on adjusted calculations tuned through use of retrospective model runs, 5) implementation of snowmelt and soil frost processes in the hydrological model to help in high latitude and high altitude regimes, and 6) use of satellite soil moisture information from NASAs Soil Moisture Active-Passive (SMAP) for more accurate estimation of pre-event conditions. A two-level approach will be used where the global, relatively coarse resolution products (~ 10 km) will be available to serve as background (e.g., identifying emerging flood hazards) and provide routine information across the globe. For those areas identified as having floods (or forecast floods) from the global 10 km resolution run, we will provide high-resolution flood products (~ 1 km). This nested approach will provide high-resolution products for all identified flood areas (roughly 10 at any one time) for use in pinpointing the hazard locations and evolution.