Search in KarstBase
Ce travail de these s’inscrit dans la problematique generale de la caracterisation de l’alea inondation. Plus precisement, cette etude cherche a caracteriser l’influence des eaux souterraines sur la genese et la propagation des crues en surface dans les cas d’un bassin versant a forte composante karstique. Le site experimental du bassin versant du Coulazou, riviere temporaire qui traverse le massif karstique du Causse d’Aumelas a l’Ouest de Montpellier a ete retenu pour etudier de maniere approfondie les interactions entre les ecoulements de surface et les ecoulements souterrains en situation de crue.
L’etude hydrodynamique de ce systeme karst/riviere s’appuie sur un dispositif experimental adapte a l’observation des phenomenes hydrologiques (pluie, ruissellement) et hydrogeologiques (piezometrie en forage et dans les drains karstiques, suivi hydrodynamique des exutoires du systeme) tres rapides et tres intenses. La dynamique de ces ecoulements est liee au contexte climatique Mediterraneen mais aussi aux structures de drainage en surface et en souterrain qui permettent un transfert et un transit tres rapide des eaux au sein du systeme karst/riviere.
Une description hydrodynamique classique est completee par une approche fonctionnelle des echanges karst/riviere dans le but de mieux comprendre le fonctionnement hydrodynamique d’un tel systeme et de mettre en avant des indicateurs utilisables dans une demarche de modelisation des echanges surface/souterrain. Un premier modele est presente dans la derniere partie de ce document.
**********
This work aims at assessing the flooding hazard. More precisely, the study focuses on the influence of groundwater on the genesis and propagation of surface flows in the case of a highly karstified watershed. The experimental site of the Coulazou River, a temporary River which crosses the karstified formation of the Causse d’Aumelas (western Montpellier) has been selected to study hydrodynamic interactions between surface flows and groundwater flows during flood.
The hydrodynamic study of this karst/River system is based on a suitable experimental monitoring of both fast and intense hydrological (rain, runoff) and hydrogeological (water level in wells and karst drains, discharge measurements at the main outlets of the system) phenomenon. The specific hydrological response of this watershed is due to the Mediterranean climate but also to surface and underground drainage structures which allow very fast water flows within the karst/river system.
A common hydrodynamic description is followed by a functional approach of karst/river exchanges in order (i) to better understand the hydrodynamic behaviour of such a system and (ii) to highlight some indicators that can be used in a modelling approach. A first conceptual model of surface water/groundwater exchanges in karst terranes is presented in the latter part of the manuscript.
The historical approach of the karst always gave preferential treatment to the study of the superficial phenomena or underground cavities explored by human. However as demonstrated by hydrogeologists, the main karst development keep out of reach because of the too small sizing of drains or due to its filling. Consequently appears the question about the inception drain, the way used by the water from the sink hole to reach its resurgence. Some authors consider this primary link as obvious when the practice demonstrates clearly that the hydrodynamic continuity results of a very long, complex and selective evolution, essentially geochemical. This is the field of the drain inception stages, that we can spell “prekarst” or “primokarst”. Those stages include the successive fields of isalterite and alloterite. This last one opens by compaction a free space and allows a concentrated hydrodynamic flow. These processes, at the origin of the endokarst initiation, can develop on the side of synchronous mechanical dynamics if in the same drain or under a regolith coverage, something divides the bedrock and the quick flow. Without any doubt, this is the purview of the cryptokarst and of the cave walls under filling. We can observe it in the progradation front of the introduction karst or in the retrogradation front of the restitution karst.
Improving knowledge of karst hydrodynamics represents a global challenge for water resources because karst aquifers provide approximately 25% of the world population in fresh water. Nevertheless, complexity, anisotropy, heterogeneity, non-linearity and possible non-stationarity of these aquifers make them underexploited objects due to the difficulty to characterize their morphology and hydrodynamics. In this context, the systemic paradigm proposes others methods by studying these hydrosystems through input-output (rainfall-runoff) relations.
The approach proposed in this thesis is to use information from field measurement and from systemic analyses to constrain neural network models. The goal is to make these models interpretable in terms of hydrodynamic processes by making model functioning to be similar to natural system in order to obtain a good representation and extract knowledge from model parameters.
This work covers the association of information available on the hydrosystem with correlation and spectral analyses to develop a temporal multiresolution decomposition of variables and to constrain neural network models. A new method for variable selection, adapted to represent long term hydrodynamics of the system, has been proposed. These constrained models show very good results and allow, through their parameters, to study the temporal contribution of inputs variables to the output.
Modeling nonlinear and non-stationary hydrosystems with neural network has been improved by a novel implementation of data assimilation. More precisely, when non-stationarity is attributed to the catchment, data assimilation is used to modify the model parameters. When the inputs are non-stationary, data assimilation can be used to modify the inputs.
The modification of inputs opens considerable scope to: i) fill gaps or homogenizing time series, ii) estimate effective rainfall.
Finally, these various analyses and modeling methods, mainly developed on the karst hydrosystem Lez, can improve the knowledge of the rainfall-runoff relationship at different time scales. These methodological tools thus offer perspectives of better management of the aquifer in terms of floods and resources. The advantage of these analyses and modeling tools is that they can be applicable to other systems.