• Global Rapid Flood Mapping System with Spaceborne SAR Data
Program GEO-Global Flood Risk Mod
PI / Institution Sang-Ho Yun / JPL
Start Date April 11, 2018
End Date April 11, 2021
  • Summary

    According to the International Federation of Red Cross and Red Crescent Societies’ (IFRC’s) World Disasters Report 2015, on average, for the past decade, there have been floods somewhere in the world almost every other day. Earth-observing satellite remote sensing is needed to monitor global floods, but only 15% of the world is cloud-free and under daylight at any given time. Radar “see” through clouds and can image Earth day and night, and is hence an ideal tool for rapid flood mapping. Synthetic Aperture Radar (SAR) is revolutionizing flood response. However, the ability to effectively utilize it for
    rapid flood response has been limited due to latencies in the end-to-end event response process.
     
    We propose to streamline an end-to-end automated flood response process and demonstrate by responding to extreme flood events in 10 GEO member countries partnering with their responding agencies, international GEO member organizations, and USGEO member agencies. We have designed our goals and implementation details strictly based on lessons learned from our flood response efforts in coordination with the Federal Emergency Management Agency (FEMA), U.S. Geological Survey (USGS), and National Oceanic and Atmospheric Administration (NOAA), World Bank, and the International Centre for Water Hazard (ICHARM) at UNESCO.
     
    The Advanced Rapid Imaging and Analysis (ARIA) has a comprehensive system automated to respond to USGS NEIC’s major earthquake event notification. The system is automated to discover, ingest, and process the European Sentinel-1 SAR data to produce maps of ground deformation caused by the earthquake. However, the ARIA system has not been optimized and automated for rapid flood response. We will leverage the ARIA system’s current capability and existing SAR processing modules. Combining them together, we will mature the system with optimized performance for flood mapping and seamless end-to-end automation from triggering to delivery.
     
    By the end of our proposed 3-year effort, we will have responded to about a dozen major global flood events rapidly producing SAR-based flood maps for our end users, using Sentinel-1, COSMO-SkyMed, ALOS-2, RADARSAT-2, and TerraSAR-X data. Tasking request of the satellites will be guided by automatic event notification from existing forecast and observation products. All the flood maps will have been calibrated with ground or airborne observations and compared with water fraction maps derived from NASA’s SMAP and NPP-ATMS observations. Once the water fraction maps turn out to be useful, we will scale up the ARIA system and start producing globally consistent flood maps of all historic events imaged by SAR and SMAP/ATMS sensors.  
     
    The flood maps will be shared in standard GIS formats and WMTS format with the response, relief, and recovery communities in the flood-prone developing countries as well as with our international (World Bank, IFRC, and UNESCO-ICHARM) and national (FEMA, USGS, NOAA) partners and end users. The outcome of our proposed work will build an optimal framework for future near real time flood responses and help improving modeling methods and flood risk analysis.