Résumé
Accurate flood inundation forecasts have the potential to minimize socioeconomic losses, but uncertainties in inflows propagated from the precipitation forecasts result in large prediction errors. Recent studies suggest that by assimilating independent flood observations, inherent uncertainty in hydraulic flood inundation modeling can be mitigated. Satellite observations from Synthetic Aperture Radar (SAR) sensors, with demonstrated flood monitoring capability, can thus be used to reduce flood forecast uncertainties through assimilation. However, researchers have struggled to develop an appropriate cost function to determine the innovation to be applied at each assimilation time step. Thus, a novel likelihood function based on mutual information (MI) is proposed here, for use with a particle filter‐based (PF) flood extent assimilation framework. Using identical twin experiments, synthetic SAR‐based probabilistic flood extents were assimilated into the hydraulic model LISFLOOD‐FP using the proposed PF‐MI algorithm. The 2011 flood event in the Clarence Catchment, Australia was used for this study. The impact of assimilating flood extents was evaluated in terms of subsequent flood extent evolution, floodplain water depths, flow velocities and channel water levels (WLs). Water depth and flow velocity simulations improved by ∼60% over the open loop on an average and persisted for up to 7 days, following the sequential assimilation of two post‐peak flood extent observations. Flood extents and channel WLs also showed mean improvements of ∼10% and ∼80% in accuracy, respectively, indicating that the proposed MI likelihood function can improve flood extent assimilation.
Accurate forecasts of flood inundation can minimize socioeconomic losses from the frequent and often disastrous flood events occurring world‐wide. However, numerical models used to generate flood predictions are strongly dependent on the quality of inputs such as inflows and topography, which typically fail to meet the required accuracy standards. Recent studies suggest that integrating independent flood observations into these numerical models can mitigate some of the errors and increase the reliability of the resulting flood forecasts. Remotely sensed radar data, which has all‐weather/all‐day imaging capabilities, can thus accurately observe flooded areas. These flood extent observations can then be used to improve flood forecasts through model‐data integration, but studies have struggled to develop an effective approach to combine these with numerical models, as the area under water only varies slightly with time. Consequently, this study proposed a novel model‐data integration method, sensitive to slight variations in the flooded area, and verified its performance through synthetic experiments. At each time step shared information between the model predicted and the observed flooded area is quantified and used to combine their information content. This led to persistent improvements in predictions of flood extent, depth, and velocity, demonstrating the potential of the proposed integration method.
A novel mutual information‐based metric is proposed as the likelihood function for particle filter‐based flood extent assimilation Distributed impacts of the assimilation on simulated flood depth and flow velocities are illustrated for different lead times Improvements in simulated water levels of ∼80% over the open loop are shown, persistent for up to one week after the assimilation