Abstract
In Mediterranean regions, flash floods are critical issues for human life and properties over the past few decades with the increment in rainfall intensity and population. To prevent causalities, it is necessary to create flash flood forecast models and early warning system for urbanized and rural areas. However, it is not always a downhill mission especially for complex hydrosystems having heterogeneous and nonlinear precipitation trend. In this contexte, the French Flood Forecasting Service (called SCHAPI for Service Central d'Hydrométéorologie et d'Appui à la Prévision des Inondations) initiated the BVNE (Digital Experimental Basin, for Bassin Versant Numérique Expérimental) project in order to enhance flash flood forecasts. For this purpose, my research goal was designed to provide:i) solutions for the problems in hydrological modeling of flash floods in the complex and heterogeneous hydro systems subjected to heavy rainfalls.ii) an artificial neural network model dedicated to predicting flash floods in advance, strengthening the system by combining multilayer perceptron models in a data assimilation framework and/or coupling different neural network models (e.g., feed forward and recurrent models etc.)Proposed location for the thesis work is the Gardon de Sainte-Croix basin, one of the basins in Cevennes border in France due to complexity of hydrological processes, variability in the spatial distribution of the rainfalls and very short response time. Moreover, the region and rivers are nested with urban and agricultural lands which will increase the risks of life and property loss.