SAR images and interferometric coherence for flood monitoring

Abstract

SAR backscatter intensity has already been successfully exploited for the detection of different types of changes in a scene, including seasonal changes, ice floes, landslides, earthquake damage and flooding. A drawback is that the backscatter intensity is affected by the presence of wind fields, water. Flood detection performed by only involving SAR intensity values could be not reliable, unless the weather information is integrated. On the contrary, interferometric coherence, usually low in presence of water, is not sensitive to weather conditions. The authors demonstrate that water can be easily identified by a data fusion which involves multitemporal backscatter intensities and multitemporal interferometric coherence: the additional information provided by the absence of coherence over water allows for a more accurate identification of the flooded areas with respect to simply exploiting the backscatter intensity.