Assimilation of SAR-derived flood extent observations for improving fluvial flood forecast – A proof-of-concept

Abstract As the severity and occurrence of flood events tend to intensify with climate change, need for forecasting capability increases. In this regard, Flood Detection, Alert rapid Mapping (FloodDAM) project, funded by Space Climate Observatory initiatives, was set out develop pre-operational tools dedicated enabling quick responses in flood-prone areas, improve reactivity decision support systems. This work focuses on assimilation 2D extent data (expressed terms wet surface ratios) in-situ water level representation plain dynamics a Telemac-2D model an Ensemble Kalman Filter (EnKF). The EnKF control vector composed friction coefficients corrective parameter input forcing. It is then augmented state averaged over several floodplain zones. conducted context Observing System Simulation Experiments (OSSE) based real event occurred January-February 2021 Garonne Marmandaise catchment. allows validate observation operator associated ratio observations as well dual state-parameter sequential correction implemented work. merits assimilating SAR-derived complementary are shown spaces 1D assessment metrics. also that hydraulic significantly improved dynamics, especially during recession. proof-of-concept study paves way towards near-real-time forecast, making most remote sensing-derived observations.