An Integrated Approach for Realtime Floodmap Forecasting on the Belgian Meuse River

The last important floods of the Meuse river have shown the need to design powerful and real-time forecasting tools. With the support of CESAME and the department of Civil and Environmental Engineering at UCL, the Service of Hydrologic Studies (SETHY) of the Walloon Ministry of Equipment and Transport developed two models. Hydromax and Hydroaxe. These two complementary and user-friendly applications work with the data provided by the measurement network of SETHY (raingauges, water levels, discharge measurements, weir-gate positions). Hydromax produces local river flow forecasting for the main natural tributaries of the Meuse. These predictions are used by Hydroaxe to compute discharge propagation and water levels all along the Meuse. In Hydromax, the predictions are produced by a grey box model which involves two main parts. A nonlinear production function computes the effective rainfall from the mean areal rainfall. This part is based on a conceptual approach, the river basin being modelled as a reservoir. In the second part, a linear ARX (AutoRegressive model with eXtra input) transfer function (black box), describes the superficial runoff of the effective rainfall towards the watershed outlet. This transfer function is used to compute short term river flow predictions. Hydroaxe uses a Preissmann finite difference scheme to solve the Saint-Venant equations of shallow-water, completed with the Exchange Discharge Model describing the momentum exchanges between the main channel and the floodplains. The optimisation of the computation time requires a one-dimensional approach, based on a dense (1 point/m2) and accurate (15 cm in x, y, z) topography provided by SETHY and carried out through an original combination of technologies: swath bathymetry and airborne laser (Lidar). With the help of a GIS (Geographic Information System) and the DTM (Digital Terrain Model), the water levels calculated by Hydroaxe are transformed in flooded areas, fitted for an easy and fast overview of the extent of the flood event.