River discharge estimation under uncertainty from synthetic SWOT-type observations using variational data assimilation

Abstract

Variational data assimilation is used for discharge estimation for the Garonne River downstream reach, using the 1D full Saint-Venant hydraulic model SIC². Based on water surface elevation observations with the level of noise expected for the forthcoming Surface Water and Ocean Topography (SWOT) mission, discharge estimation problem under uncertainty in the bathymetry and the friction has been considered. Two observation sampling periods has been tested, with the purpose of investigating the influence of the temporal frequency of observations. We emphasize that the temporal sampling frequency should be consistent with the characteristic (travel) time of the dynamical system, in which case the discharge is successfully estimated (rRMSE = 4.6%). Larger temporal sampling intervals require complementary information given by a-priori information on discharge or combined in-situ and remote sensing data assimilation. Simultaneous estimation of discharge and the spatially distributed functions may lead to the equifinality issue. In this case, the estimates of the bathymetry and the friction only help to achieve a better estimation quality of discharge, but may not be usable for subsequent experiments.