Use of parameter optimization to estimate a flood wave: Potential applications to remote sensing of rivers

In this paper, the potential for identifying discharge and/or flood hydrograph from remotely sensed data is explored. The parameter identification process is based on the minimization of the difference between the solution of the model equations and the observed system response which consists in maximum inundation extent. The river geometry is supposed to be known, effect of the accuracy of these data on the estimation has been tested. Sensitivity of the model to individual parameters is then assessed using an extension of the generalized sensitivity analysis. Synthetic data have been used to test the methodology. Results show that the Nash criterion of the estimated flood hydrograph is higher than 0.9 for all the tested cases.