Le devenir des polluants dans les sols constitue un enjeu environnemental majeur. Dans ce travail, nous apportons une contribution à quelques méthodes numériques pour la simulation de l’écoulement et du transfert de polluants en milieu poreux variablement saturés. La propagation d’un contaminant dans les milieux souterrains dépend en premier lieu des caractéristiques de l’écoulement qui le transporte. Dans la première partie de ce travail, nous présentons la méthode des éléments finis mixtes hybrides pour la résolution de l’équation de Richards. Une procédure de condensation de la masse est proposée pour éviter l’apparition d’oscillations non physiques, notamment lors de la simulation de problèmes d’infiltration dans un milieu initialement sec.Dans la deuxième partie de ce travail, la méthode ELLAM (Eulerian-Lagrangian Localized Adjoint Method) est utilisée pour la modélisation du transport réactif en milieux fortement hétérogènes. En effet, les résultats obtenus pour le transport linéaire, décrit par l’équation d’advection-dispersion, avec les ELLAM sont très encourageants. La méthode ELLAM permet (i) de s’affranchir des contraintes de discrétisations spatiale ettemporelle imposées avec les méthodes eulériennes classiques, (ii) de conserver la masse et (iii) de traiter toutes les conditions aux limites. Par ailleurs, nous proposons une nouvelle formulation des ELLAM (C_ELLAM) permettant d’éviter les oscillations numériques et de limiter la diffusion numérique générées parla formulation standard.Dans la dernière partie, le code de calcul élaboré avec la formulation C_ELLAM est utilisé pour la caractérisation de la macrodispersion dans les milieux hétérogènes. Pour ce faire, il est indispensable de disposer d’outils de simulation précis et efficaces car cette étude est basée sur une méthode Monte Carlo nécessitant la réalisation d’un très grand nombre de simulations sur des grilles de calcul de l’ordre du million de mailles. Les résultats obtenus sont comparés avec une étude antérieure basée sur le Random WalkParticle Method. / The fate of contaminants in soils is a major environmental challenge. In this work, we develop efficient and reliable numerical tools for simulation of water flow and distribution prediction of pollutants in variably saturated porous media. In the first part of this document, the mixed hybrid finite element method is presented for solving Richard’s equation. A mass lumping technique is proposed to avoid unphysical oscillations when sharp infiltration fronts are simulated. In the second part of this work, the Eulerian Lagrangian Localized Adjoint Method (ELLAM) is used for modeling reactive transport in highly heterogeneous domains. Solute transport is described mathematically by the advection-dispersion and results obtained with ELLAM are very encouraging. ELLAM allows (i)overcoming spatial and time discretizations constraints imposed by classical Eulerian method, (ii)conserving mass and (iii) treating general boundary conditions naturally in the formulation. Moreover, we introduce a new ELLAM scheme (C_ELLAM) which avoid unphysical oscillations and reduce the numerical dispersion generated by the standard formulation.In the last part of this document, the C_ELLAM scheme is used to characterize the macrodispersion of a nonreactive solute in heterogeneous domains. This study is based on Monte Carlo simulations andtherefore requires highly efficient simulators. Our results are compared with previous work using Random Walk Particle Method to solve the advection-dispersion equation.

The heart is one the most important organs in the human body and many other live creatures. The electrical activity in the heart controls the heart function, and many heart diseases are linked to the abnormalities in the electrical activity in the heart. Mathematical equations and computer simulation can be used to model the electrical activity in the heart. The heart models are challenging to solve because of the complexity of the models and the huge size of the problems. Several cell models have been proposed to model the electrical activity in a single heart cell. These models must be coupled with a heart model to model the electrical activity in the entire heart. The bidomain model is a popular model to simulate the propagation of electricity in myocardial tissue. It is a continuum-based model consisting of non-linear ordinary differential equations (ODEs) describing the electrical activity at the cellular scale and a system of partial differential equations (PDEs) describing propagation of electricity at the tissue scale. Because of this multi-scale, ODE/PDE structure of the model, splitting methods that treat the ODEs and PDEs in separate steps are natural candidates as numerical methods. First, we need to solve the problem at the cellular scale using ODE solvers. One of the most popular methods to solve the ODEs is known as the Rush-Larsen (RL) method. Its popularity stems from its improved stability over integrators such as the forward Euler (FE) method along with its easy implementation. The RL method partitions the ODEs into two sets: one for the gating variables, which are treated by an exponential integrator, and another for the remaining equations, which are treated by the FE method. The success of the RL method can be understood in terms of its relatively good stability when treating the gating variables. However, this feature would not be expected to be of benefit on cell models for which the stiffness is not captured by the gating equations. We demonstrate that this is indeed the case on a number of stiff cell models. We further propose a new partitioned method based on the combination of a first-order generalization of the RL method with the FE method. This new method leads to simulations of stiff cell models that are often one or two orders of magnitude faster than the original RL method. After solving the ODEs, we need to use bidomain solvers to solve the bidomain model. Two well-known, first-order time-integration methods for solving the bidomain model are the semi-implicit method and the Godunov operator-splitting method. Both methods decouple the numerical procedure at the cellular scale from that at the tissue scale but in slightly different ways. The methods are analyzed in terms of their accuracy, and their relative performance is compared on one-, two-, and three-dimensional test cases. As suggested by the analysis, the test cases show that the Godunov method is significantly faster than the semi-implicit method for the same level of accuracy, specifically, between 5 and 15 times in the cases presented. Second-order bidomain solvers can generally be expected to be more effective than first-order bidomain solvers under normal accuracy requirements. However, the simplest and the most commonly applied second-order method for the PDE step, the Crank-Nicolson (CN) method, may generate unphysical oscillations. We investigate the performance of a two-stage, L-stable singly diagonally implicit Runge-Kutta method for solving the PDEs of the bidomain model and present a stability analysis. Numerical experiments show that the enhanced stability property of this method leads to more physically realistic numerical simulations compared to both the CN and Backward Euler (BE) methods.