Automatic Near‐Real‐Time Flood Mapping from Geostationary Low Earth Orbiting Satellite Observations

Near-real-time flood maps derived from satellite data are invaluable to river forecasters and decision makers for disaster monitoring and relief efforts. Combining utilization of the low earth orbiting (LEO) and geostationary (GEO) satellite imagery shows great advantages in flood mapping. Under clear-sky coverage, the floodwater detected in LEO satellite imagery shows rich inundation detail; while under nonclear-sky conditions, composition from multiple GEO satellite images provides more clear-sky coverage for flood detection and the clear-sky information can be used to fill the gaps of clouds and cloud shadows in LEO imagery. With support from the National Oceanic and Atmospheric Administration (NOAA) and National Aeronautics and Space Administration Joint Polar Satellite System and Geostationary Operational Environmental Satellites (GOES) R programs, the flood mapping software has been developed to derive flood maps from Suomi National Polar-orbiting Partnership and NOAA-20/Visible Infrared Imaging Radiometer Suite imagery, and GOES-16 and 17/Advanced Baseline Imager imagery. These flood maps are distributed via the Unidata Local Data Manager, reviewed by river forecasters in the second generation of the Advanced Weather Interactive Processing System and applied in flood operations. Initial feedback from operational forecasters on the product accuracy and performance has been largely positive. Offline evaluation efforts include visual inspection, an intercomparison with the Moderate-Resolution Imaging Spectroradiometer automatic flood products, and a quantitative validation using Landsat imagery. The steady performance indicates encouraging feasibility of these flood maps to be provided at the product level.