An integrated error parameter estimation and lag-aware data assimilation scheme for real-time flood forecasting

Abstract

For operational flood forecasting, discharge observations may be assimilated into a hydrologic model to improve forecasts. However, the performance of conventional filtering schemes can be degraded by ignoring the time lag between soil moisture and discharge responses. This has led to ongoing development of more appropriate ways to implement sequential data assimilation. In this paper, an ensemble Kalman smoother (EnKS) with fixed time window is implemented for the GR4H hydrologic model (modèle du Ge´nie Rural à 4 paramètres Horaire) to update current and antecedent model states. Model and observation error parameters are estimated through the maximum a posteriori method constrained by prior..