Automated mapping of surface water and flood extent from TerraSAR-X

Abstract

The activities presented here are targeted towards the development of dedicated processing chains for a) an automatic and near real-time mapping of flood situations and b) the mapping of surface water on a global scale based on TerraSAR-X data. On the one hand, the mapping of flood situations demands for automatic and computationally inexpensive algorithms which remain accurate even in complex scenarios such as floods in urban areas or underneath vegetation layers. On the other hand, a global mapping of surface water requires methods which are as transferable as possible while remaining robust enough to cope with the large data volumes necessary for this task. In particular, the algorithms need to account for the different backscatter profiles related to environmental factors (e.g. wave effects) as well as acquisition parameters (e.g. incidence angle).