Data assimilation of remotely sensed information in operational flood forecast in Southwest Germany

Floods are the most affecting natural disasters in Europe. Flood damage can be reduced if the runoff is predicted with sufficient accuracy and well enough in time so that technical and precaution measures can be applied. Hydrological models are the tools to simulate runoff and allow flood forecast. Since models are only as good as their input data, one possibility to improve flood forecast is to provide a better spatial characterisation of the watershed and to observe its change with time. Remote sensing methods allow the observation of some of these hydrological relevant watershed properties.