Rapid Detection Of Debris Flows Using Synthetic Aperture Radar: A Coherence-Based Methodology

The ability to detect and map landslides triggered by intense rainfall in quasi-real time is essential to mitigate their impact and for effective crisis management. The manipulation of space-borne synthetic aperture radar (SAR) images has proven to be one of the most valuable and inalienable asset for this type of investigation. In mountainous areas, morphological variations, related to surface processes and activated by forceful meteorological events, can be usually detected by applying three type of approaches: amplitude-based methods, phase-coherence-based methods and polarimetric techniques. This study present a rapid, effective and straightforward coherencebased methodology which, using just three SAR images, can detect the activation of debris flows with a latency solely related to the repeat cycle of the SAR mission. The technique has been tested in the Italian Dolomites using the dataset of European Space Agency's mission Sentinel-1, showing promising results. This research is carried out in the framework of Project VAILAND, a joint research agreement funded by the Veneto Region (Italy).