Résumé
Southern France is frequently hit with severe flash floods. These events have caused numerouscasualties as well as considerable economic losses. In France, crisis inducing floods are managed ata local scale by municipalities. Current forecasting models are rarely used by these crisis managers.This is due to the fact that these models often fail to convey relevant information in an appropriateform. This work aims to determine the adequate output variable in a crisis management contextwith a special focus on the needs and vision of local crisis managers.The study area is the 545 sq. km. sized Gardon in Anduze in the Cévennes range, France, which isoften subject to important flash floods. A feed-forward multilayer perceptron, a type of neuralnetworks where one of the inputs is the measured output, is used to forecast water discharge.Neural networks are good candidates to model nonlinear phenomenon. Both the properties ofparsimony - they provide relevant results while requiring a limited number of parameters – and ofuniversal approximation are very helpful in hydrology. Various rainfall and discharge measurementsare considered as inputs. The output variable for this model is water flow. Data ranges from 2002 to2019 at a 30 minutes’ time step. Events are extracted from this database to compose a training set,a test set and an early stopping set. Complexity and variable selection is made using cross-validation, allowing to select the most robust model. To make the model easier to understand byend users, discharges are replaced by water levels. Both these results are compared, especiallyregarding their ability to reach crisis management plans levels without delay. Secondly, ensemblemodels based on different initializations of the parameters of the neural model during the trainingstep are developed. They give an uncertainty margin, often desired by end users.The results are enriched through the collection of background information (end users’ opinions andhabits) in order to enhance the assessment of performance. Further analysis is carried to determinethe pros and cons of each approach (discharge or levels to crisis management plans levels anduncertainties representation and communication).