Development of a methodology for flood hazard detection in urban areas from RADARSAT-2 imagery

Due to their capacity do acquire data day-and-night, whatever the meteorological conditions may be, and to the improved spatial resolution of the new generation of SAR sensors such as RADARSAT-2, COSMO-SkyMed or TerraSAR-X (~ 1 to 3 m with fine acquisition modes), SAR imagery is the most suitable tool for the detection of flooded areas in a context of crisis management. However, due to the complexity of the backscattered signal in urban environments, the development of precise and automated methods for flood detection in urban areas with SAR imagery is still challenging. A new approach for flood delineation and mapping in urban areas, combining RADARSAT-2 imagery and probability of flooding data simulated by a flood dynamic risk mapping tool is presented in this paper.