An automatic system for near-real time flood extent and duration mapping based on multi-sensor satellite data

Abstract

This chapter presents an automatic and operational system for near real-time mapping of flood extent and duration using multi-sensor satellite data. The system is based on four processing chains for the derivation of the flood extent from Sentinel-1 and TerraSAR-X radar as well as from optical Sentinel-2 and Landsat satellite data. While the systematic acquisition of Sentinel-1 and optical data allows a continuous monitoring of inundated areas at an interval of a few days, the TerraSAR-X flood service can be activated on-demand in case of emergency situations. All processing chains contain the following generic steps: data ingestion, preprocessing of satellite data, computation and adaption of global auxiliary data, classification, and dissemination of the crisis information via web-services. Flood duration layers are generated to indicate the temporal stability and evolution of inundation events. The flood mapping system is demonstrated on a flood situation in Mozambique caused by cyclone Idai in 2019.