NH036-05
Rapid Flood Severity Classification and Alerting for the Spring 2020 Africa Floods: A Case Study
Abstract:
We seek to rapidly classify flood severity and alerts based on the potential for impacts in a way similar to the USGS PAGER impact analysis for earthquakes. Our approach is to use a model of models leveraging existing capabilities and products, as well as incorporating remotely sensed products for ground-truthing model results and delineation of flood areas. Ground-truthing provides continued adjustment of parameters in the model of models approach. Administrative area resilience indicators for flood affected areas will support the incorporation of vulnerability and lack of capacity into the rapid assessment of potential impacts.
The NASA Disasters Program activated in May of 2020 to investigate the flooding in Central and East Africa. We examined this event as a case study for testing the capabilities of our system. The system combines model outputs from the Global Flood Monitoring System (GFMS) and the Global Flood Awareness System (GloFAS) with data on watershed risk using a novel risk function algorithm. Results are then validated using Synthetic Aperture Radar (SAR) or optical data for flood inundation and depth, if available. This data will be integrated into the Pacific Disaster Center (PDC) DisasterAWARE platform, which provides a source of global information on floods that is supported by a common, normalized data model.
DisasterAWARE is operated by PDC and currently provides global multi-hazard alerting and Situational Awareness information to the public. DisasterAWARE supports hazard monitoring and early warning needs of Disaster Managers across the globe, as well as providing the Common Operating Picture in support of US DOD’s humanitarian assistance and disaster response mission. However, the current systems lack global flood alerting or incorporation of a remote sensing component that will allow near real-time validation of simulated flood modeling results.