Search

International audience; The real-time vision of a large portion of French rivers is not a coloured one, not even a black and white, but a blind one, du to the large number of basins which are not monitored. The event of June 2010 in Southern France, where 25 people died during the flood of the Nartuby and Argens rivers gives a dramatic example of this blind vision. In such a context a simple insight, aiming at providing flood alerts rather than pretending to represent exactly the physical process at stake in the catchment, seems welcomed. The aim of this study is to assess the ability of a simple conceptual and parsimonious rainfall-runoff model to deliver flood alerts in ungauged catchments. We use a downward approach: the model is calibrated on gauged catchments and its performances are checked on nested catchments which are not used for the calibration. The model is calibrated over 20 catchments of southern France with areas ranging from 150 km² to 2170 km². The rainfall information is provided by a radar network at an hourly time-step for 15 events that have occurred between 2005 and 2008. The "ungauged" catchments used for the model evaluation are 47 nested basins located within the 20 calibration catchments. We derive the model parameters for these ungauged catchments, from the gauged ones. The model used belongs to the GR models family developed at the Cemagref institute. It works on en event basis thank to a production parameter S and two routing parameters: a shape parameter B and a lag parameter C. Our knowledge of the production parameter benefits from a previous regionalisation work based on several characteristics of the catchments such as the description of their drainage network, the ground-water typology and a climatic variable. The model calibration rests upon the production parameter only. We implement two versions of the model: a lumped one and a distributed one. The results of this downward approach show that the model tested here is able to deliver successful flood alerts in ungauged areas. The distributed version performs slightly better than the lumped version. This study also show that fully distributing the routing parameters results in better alerts than using uniform routing parameters in the distributed version. To fully assess the performance of the model, further evaluation should be carried out in an operational mode, i.e. using an initialisation procedure to determine the value of the S parameter at the beginning of each event instead of using a calibrated value. We anticipate this study by checking how well the calibrated S correlate with the parameter of a continuous soil moisture accounting model run at a daily time-step. The correlation being better in the case of the distributed version of the model, this version seems hence quite promising for an operational use.

Does allowing spatial variability into a hydrological model improve flood forecasting? This study tries to contribute modestly to this important question, still matter of debate in the hydrological community. To do so, different modelling tests are carried out. For all of them, the model structure is kept unchanged, while the model spatial discretisation is modified. The first test consists to represent each catchment by a simple lumped model, receiving an average basin rainfall. The other tests attempts to take advantage of the spatial variability within the catchment, in particular of rainfall. To do so, the catchment is divided into 1km 2 cells, each containing a replica of the lumped model tested at the first stage. Then, different approaches are compared, for model parameters estimation, from 'uniform' parameters (ie a unique set for all catchment cells), to 'distributed' parameters (ie with inter-cells variability). The rainfall-runoff model common to all these tests is very simplified and depends on only 3 parameters, which enables automatic calibration to be done. These tests are applied on two catchments located in South of France. For each of them, parameters are calibrated using outlet discharge, and results are validated on inner available gauged points. Results suggest that introducing spatial variability into the model can improve simulations, but not in any case. These differences are discussed, and weaknesses in our approach mentioned.

/ Les systèmes technologiques doivent être conçus puis maîtrisés de manière à produire ou fournir un service dans les meilleures conditions économiques, assurer une qualité constante et conforme aux spécifications, garantir la sécurité des employés, des usagers ou consommateurs, de l'environnement et de l'outil de travail, innover pour satisfaire les consommateurs. Dans ce mémoire, nous nous intéressons au contrôle de systèmes technologiques axé vers la maîtrise de deux types de caractéristiques : la qualité des produits et la sécurité des systèmes technologiques, éléments importants pour de nombreuses entreprises. La maîtrise de ces caractéristiques passe par la maîtrise des processus physiques (transformation de matières premières en un produit fini, vieillissement d'un ouvrage de génie civil), informationnels (collecte et traitement de données) et décisionnels (agrégation de données pour prendre une décision), dans un champ de contraintes notamment économiques et réglementaires. L'amélioration des processus informationnels et décisionnels afin de maîtriser les processus physiques repose sur le développement de modèles de mesure, d'évaluation, de contrôle, de diagnostic et de pronostic. Or, le développement de modèles peut poser des difficultés dans le cas de certains systèmes technologiques. En effet, pour des systèmes tels que les procédés alimentaires ou les barrages, les approches classiques de modélisation ne peuvent pas toujours être employées du fait de la compréhension incomplète des réactions se déroulant dans le système, de l'occurrence de phénomènes aux origines variées (chimiques, physiques, biologiques) et souvent en interaction, du nombre de variables très important, du caractère unique du système (cas des barrages), de cinétiques mal connues et non linéaires, de données d'entrée non continues. Par ailleurs, les mesures disponibles pour caractériser le système ont des caractéristiques et des formats très différents : mesures globales et mesures ponctuelles, données symboliques et numériques, mesures directes et indirectes, mesures temps réel et mesures différées, mesures brutes ou mesures traitées, valeurs absolues et tendances, mesures plus ou moins disponibles, précises et certaines. Enfin, aucune donnée, prise séparément, n'est suffisante pour évaluer le système. Dans ce cadre, il est tout à fait pertinent de prendre en compte des connaissances de différentes natures et notamment celles détenues par des experts, pour établir des méthodes et des outils d'aide à la mesure, à l'évaluation et au contrôle des systèmes technologiques considérés. Nous avons ainsi proposé des méthodologies d'acquisition, de formalisation et de modélisation adaptées au système étudié et à l'objectif de la recherche : mesure, évaluation, contrôle, diagnostic et pronostic.

This thesis presents the research activity done to obtain the French "Habilitation à Diriger des Recherches". To define quickly this activity, one could speak of "Computational Mechanics". We present only two research areas: topological shape optimization (by Finite Element Method) and simulation of two-fluid flows at low Mach Number (by finite volume method). / Ce mémoire présente les travaux effectués en vue de l'obtention de l'Habilitation à Diriger des Recherches. Si l'on devait qualifier rapidement ces activités, on pourrait parler de "Mécanique numérique". Dans le cadre de ce mémoire, on présente exclusivement deux domaines de recherche : l'optimisation topologique de forme (par éléments finis) et la modélisation d'écoulements bi-fluide à faible Mach (par volumes finis).

An original approach is proposed to estimate the impacts of climate change on extreme events using an hourly rainfall stochastic generator. The considered generator relies on three parameters. These parameters are estimated by average, not by extreme, values of daily climatic characteristics. Since climate changes should result in parameters instability in time, the poster focuses on testing the presence of linear trends in the generator parameters. A general regionalization procedure is also proposed which offers the possibility to work on both local and regional scales. From the daily information of 139 rain gauge stations between 1960-2003, changes in heavy precipitations in France and their impacts on quantile predictions are investigated. It appears that significant changes occur mainly between December and May for the rainfall occurrence which increased during the four last decades, except in the Mediterranean area. Using the trend estimates, one can deduced that these changes, up to now, do not affect quantile estimations.