The transport of contaminants in groundwater system is strongly influenced by various aquifer heterogeneity factors such as spatial aquifer heterogeneity of hydraulic conductivity and reactive substances distribution. The contaminants transport can be simulated by using numerical reactive transport models, and their fate can be possibly even predicted. Furthermore, reactive transport modeling is an essential tool to get a profound understanding of hydrological-geochemical complex processes and to make plausible predictions of assessment. The goal of this work is to improve our understanding of the groundwater contaminants fate and transport processes in heterogeneous aquifer systems, with a focus on nitrate problems. A large body of knowledge of the fate and transport of nitrogen species has been achieved by previous works, however, most previous models typically neglect the interrelation of physical and chemical aquifer heterogeneities on the contaminant fate and redox transformation, which is required for predicting the movement and behavior of nitrate and quantifying the impact of uncertainty of numerical groundwater simulation, and which motivates this study. The main research questions which are answered in this work are how aquifer heterogeneity influences on the nitrate fate and transport and then, what is the most influential aquifer heterogeneity factor must be considered. Among the various type of aquifer heterogeneity, physical and chemical aquifer heterogeneities are considered. The first part of the work describes groundwater flow system and hydrochemical characteristics of the study area (Hessian Ried, Germany). Especially, data analyses are performed with the hydrochemical data to identify the major driving force for nitrate reduction in the study area. The second part of the work introduces a kinetic model describing nitrate removal by using numerical simulation. The resulting model reproduces nitrate reduction processes and captures the sequence of redox reactions. The third and fourth parts show the influence of physical and chemical aquifer heterogeneity with varying variance, correlation length scale, and anisotropy ratio. Heterogeneous aquifer systems are realized by using stochastic approach. Results, in short, show that the most influential aquifer heterogeneity factors could change over time. With abundant requisite electron donors, physical aquifer heterogeneity significantly influences the nitrate reduction while chemical aquifer heterogeneity plays a minor role. Increasing the spatial variability of the hydraulic conductivity increases the nitrate removal efficiency of the system in addition. If these conditions are reversed, nitrate removal efficiency varies by the spatial heterogeneity of the available initial electron donor. The results indicate that an appropriate characterization of the physical and chemical properties can be of significant importance to predict redox contamination transport and design long-term remediation strategies and risk assessment.

Contaminate transport in vadose is a huge concern since the vadose zone is the main passage way for ground water recharge. Understanding this process is crucial in order to prevent contamination, protect and rehabilitate ground water resources. Reactive transport models are instrumental for such purposes and there are numerous solute transport simulation programs for both ground water and vadose zone but most of this models are limited to simple Linear, Langmuir and Freundlich sorption models and first order decay and fail to simulate more complex geochemical reactions that are common in the vadose zone such as cation exchange, surface complexation, redox reaction and biodegradation. So it is necessary to enhance capabilities of solute transport models by incorporating well tested hydrogeochemical models like PHREEQC in to them to be able closely approximate the geochemical transport process in the subsurface. In this PhD research a new reactive transport model called VS2DRT was created by coupling existing public domain solute and heat transport models VS2DT, VS2DH with hydro-chemical model PHREEQC using non-iterative operator splitting technique. VS2DRT was compiled using MinGW compiler using tools like autotools and automake. A graphical user interface was also created using QT creator and Argus ONE numerical development tools. The new model was tested for one dimensional conservative Cl transport, surface complexation, cation exchange, dissolution of calcite and gypsum, heat and solute transport as well as for two dimensional cation exchange cases. Their results were compared with VS2DT, VS2DH, HP1 and HP2 models and the results are in good agreement.

Reactive transport modelling has wide applications in geosciences. In the field of hydrogeology, it has been utilised to simulate the biogeochemical processes that disperse and degrade contaminants in the aquifer. For geotechnical applications, such as geological CO2 sequestration, the reaction of CO2 with the ambient saline aquifer determines the final success of storage. In a radioactive waste repository, scientists rely on reactive transport models to predict the mobilisation of hazardous radionuclides within space and time. In this work, the multi-component mass transport code OpenGeoSys, was coupled with two geochemical solvers, the Gibbs Energy Minimization Selektor (GEM) and the Biogeochemical Reaction Network Simulator (BRNS). Both coupled codes were verified against analytical solutions and simulation results from other numerical models. Moreover, the coupling interface was developed for parallel simulation. Test runs showed that the speed-up of reaction part had a very good linearity with number of nodes in the mesh. However, for three dimensional problems with complex geochemical reactions, the model performance was dominated by solving transport equations of mobile chemical components. OpenGeoSys-BRNS was applied to a two dimensional groundwater remediation problem. Its calculated concentration profiles fitted very well with analytical solutions and numerical results from TBC. The model revealed that natural attenuation of groundwater contaminants is mainly controlled by the mixing of carbon source and electron donor. OpenGeoSys-GEM was employed to investigate the retardation mechanism of radionuclides in the near field of a nuclear waste repository. Radium profiles in an idealised bentonite column was modelled with varying clay/water ratios. When clay content is limited, Ba-Sr-Ra sulfate solid solutions have a very strong retardation effect on the aqueous radium. Nevertheless, when clay mineral is abundant, cation exchange sites also attract Sr and Ba, thus dominates the transport of Ra.

L’industrie du nucléaire produit de grandes quantités d’effluents radioactifs de sources diverses nécessitant des traitements spécifiques en fonction de leur composition chimique. Le césium 137 fait partie, avec le strontium 90, des radioéléments majoritairement présents dans ces effluents qui doivent être extraits le plus efficacement possible en produisant un minimum de déchets secondaires. Le traitement en colonne est parmi les procédés les plus adaptés pour ce type d’extraction sur support solide. Son principe repose sur la capacité de sorption du radioélément par des hexacyanoferrates (HCF) de cuivre, des échangeurs ioniques minéraux très sélectifs du césium. Des investigations sur les HCF ont permis de mettre en avant les modifications structurales intervenant lors de l’échange avec le césium, à l’origine de leur forte affinité pour cet ion. La fonctionnalisation des HCF sur des monolithes de silice à porosité hiérarchique a mis à profit les propriétés remarquables de ces supports pour une utilisation en colonne. Les cinétiques de sorption évaluées jusqu’aux concentrations traces ont montré une capture rapide du césium qui justifie l’intérêt de ce matériau pour un emploi en colonne. Par ailleurs, la compétitivité des monolithes fonctionnalisés par rapport à des lits particulaires a été démontrée. Ces matériaux ont enfin été mis en œuvre pour la décontamination simultanée du césium et du strontium par des mécanismes couplés d’échange d’ions et de coprécipitation du sulfate de baryum, rendue possible par la grande perméabilité des monolithes. Un modèle simplifié du transport réactif basé sur la morphologie du monolithe a été développé avec le code HYTEC en supposant un écoulement dispersif dans les canaux du squelette et la diffusion dans les parois du squelette et les agrégats de HCF. / The nuclear industry produces high amounts of contaminated water from various sources that require specific treatments depending on their chemical composition. Cesium-137 and strontium-90 are among the most abundant radionuclides in those effluents, which have to be removed as efficiently as possible in order to generate the lowest amount of waste. The column process is one of the most suitable processes to achieve this solid-phase extraction. Its principle is based on the sorption capacity of the radionuclide by copper hexacyanoferrates (HCF), highly cesium-selective mineral ion-exchangers. Investigations on HCF pointed out the structural effects of the cesium insertion within the crystal, which were linked to the high affinity of HCF for this ion. The functionalization of HCF on silica monolith with hierarchical pore structure was carried out in order to benefit the remarkable properties of these supports used as a column. Sorption kinetics evaluated down to trace concentrations have shown a fast capture of the cesium, which proves the interest of this material for a column process purpose. In addition, the performances of functionalized silica monolith have been highlighted in comparison with those made of particulate fixed beds. Finally, those materials were implemented for a simultaneous decontamination of cesium and strontium by a double extraction mechanism of ion exchange and coprecipitation of barium sulfate, allowed by the high permeability of the monolith. A simplified model of reactive transport was built with the HYTEC code, based on the actual morphology of the monolith. To do so, a dispersive flow in the macroporous intraskeletal channels and a diffusive flow inside the walls of the structure and the HCF aggregates were assumed.

Impacts from mining waste deposits on groundwater resources have been recognized invarious parts of the world; though varied in scale depending on the composition of mineralsbeing mined, the level of technology employed and environmental commitment of thedevelopers. Mining activities usually involve milling, concentrating, and processing of oreswhich will result in a huge amount of waste, called tailings, usually deposited inimpoundments as a slurry, composed of fine grained geological material (uneconomicalminerals), chemicals utilized in the processs, and water. Oxidation of these deposits, usuallycontaining sulphide minerals, may result in generation of an acidic, metal laden leachate,callled Acid Mine Drainage (AMD), which may have a devastating impact on thesurrounding groundwater resources. In this study, the stochastic LaSAR-PHREEQC reactive transport modeling approach is usedin order to evaluate the coupled effect of geochemical reactions and physical heterogeneity ofthe subsurface in the breakthrough of acidity and metal downstream of the source while theAMD transported in the water saturated zone of an impoundment. The tailings depositcalled Impoundment 1 at the Kristineberg mining site at the Skellefteå field, in northernSweden, is used as a case study to simulate pH buffering processes and attenuation of Zn.The objectives of the study are 1) to evaluate the relevance of different possible geochemicalprocesses in pH buffering and Zn attenuation; 2) to evaluate the effect of spatial variability ofthe physical processes of the groundwater system on the breakthrough of contaminants; and3) to evaluate the effect of the location and distribution of the source zone in terms of thedistance from the impoundment boundary. Simulation results of the presented model revealed that pH buffering from calcite andchlorite are important processes capable of counteracting the acidification from AMD.Dissolution of secondary Al(OH)3(s) is another important process capable of buffering pH.Precipitation of smithsonite, ZnCO3, is an important process for attenuation of Zn2+.Moreover, sorption of Zn2+ on ferric iron surfaces is found to be an important process forattenuation of the metal, depending on the available sorption surface sites. Flow variabilityhighly affects the breakthrough of the contaminants such that with increasing subsurfaceheterogeneity, earlier breakthrough of contaminants occurs. Moreover, increased variabilityresults in decreased peak loads, but longer duration of the load. / www.ima.kth.se

Soil and groundwater pollution has become a global issue since the advent of industrialization and mechanized agriculture. Some contaminants such as PAHs may persist in the subsurface for decades and centuries. In a bid to address these issues, protection of groundwater must be based on the quantification of potential threats to pollution at the subsurface which is often inaccessible. Risk assessment of groundwater pollution may however be strongly supported by applying process-based simulation models, which turn out to be particularly helpful with regard to long-term predictions, which cannot be undertaken by experiments. Such reliable predictions, however, can only be achieved if the used modeling tool is known to be applicable. The aim of this work was threefold. First, a source strength function was developed to describe the leaching behavior of point source organic contaminants and thereby acting as a time-dependent upper boundary condition for transport models. For general application of these functions dimensionless numbers known as Damköhler numbers were used to characterize the reaction of the pollutants with the solid matrix. Two functions were derived and have been incorporated into an Excel worksheet to act as a practical aid in the quantification of leaching behavior of organic contaminant in seepage water prognoses. Second, the process based model tool SMART, which is well validated for laboratory scale data, was applied to lysimeter scale data from two research centres, FZJ (Jülich) and GSF (München) for long term predictions. Results from pure forward model runs show a fairly good correlation with the measured data. Finally, the derived source term functions in combination with the SMART model were used to assess groundwater vulnerability beneath a typical landfill at Kwabenya in Ghana. The predicted breakthrough time after leaking from the landfill was more than 200 years considering the operational time of the facility (30 years). Considering contaminant degradation, the landfill would therefore not cause groundwater pollution under the simulated scenarios and the SMART model can be used to establish waste acceptance criteria for organic contaminants in the landfill at Kwabenya / Seit dem Beginn der Industrialisierung und der mechanisierten Landwirtschaft wurde die Boden- und Grundwasserverschmutzung zu einem weltweiten Problem. Einige Schadstoffe wie z. B. PAK können für Jahrzehnte oder Jahrhunderte im Untergrund bestehen. Um diese Probleme behandeln zu können, muss der Schutz des Grundwassers basierend auf der Quantifizierung potentieller Gefährdungen des zumeist unzugänglichen Untergrundes erfolgen. Risikoabschätzungen von Grundwasserverschmutzungen können jedoch durch die Anwendung prozess-basierter Simulationsmodelle erheblich unterstützt werden, die sich besonders im Hinblick auf Langzeitvorhersagen als hilfreich erweisen und nicht experimentell ermittelbar sind. Derart zuverlässige Vorhersagen können jedoch nur erhalten werden, wenn das verwendete Modellierwerkzeug als anwendbar bekannt ist. Das Ziel dieser Arbeit bestand aus drei Teilen. Erstens wurde eine Quellstärke-funktion entwickelt, die das Ausbreitungsverhalten organischer Schadstoffe aus einer Punktquelle beschreibt und dadurch als zeitabhängige obere Randbedingung bei Transportmodellen dienen kann. Im Hinblick auf die allgemeine Anwendbarkeit dieser Funktion werden als Damköhler-Zahlen bekannte, dimensionslose Zahlen verwendet, um die Reaktion von Schadstoffen mit Feststoffen zu charakterisieren. Zwei Funktionen wurden abgeleitet und in ein Excel-Arbeitsblatt eingefügt, das ein praktisches Hilfsmittel bei der Quantifizierung des Freisetzungsverhaltens organischer Schadstoffe im Rahmen der Sickerwasserprognose darstellt. Der zweite Teil dieser Arbeit beinhaltet die Anwendung des prozessbasierten und mittels Laborexperimenten validierten Modellwerkzeugs SMART für Langzeitprognosen auf der Lysimeterskala anhand von Daten zweier Forschungszentren, FZJ (Jülich) und GSF (München). Ergebnisse reiner Vorwärtsmodellierungsläufe zeigten gute Übereinstimmungen mit den gemessenen Daten. Im dritten Teil wurden die erhaltenen Quellstärkefunktionen in Kombination mit dem SMART-Modell eingesetzt, um das Grundwassergefährdungspotential unter einer typischen Deponie in Kwabenya, Ghana, einzuschätzen. Die vorhergesagten Durchbruchszeiten nach einer Leckage in der Deponie betragen über 200 Jahre bei einer Betriebszeit von 30 Jahren. Unter Berücksichtigung des Schadstoffabbaus verursacht die Deponie somit keine Grundwasserverunreinigung im Rahmen der simulierten Szenarien und das SMART-Modell kann verwendet werden, um Schadstoffgrenzwerte für organische Schadstoffe in der Deponie in Kwabenya festzulegen.

The intrusion of chlorides and carbon dioxide into a reinforced concrete (RC) structure can initiate corrosion of the reinforcing steel, which, due to its expansive nature, can damage the structure and adversely affects its serviceability and safety. Corrosion will initiate if at the steel surface the concrete free chloride concentration exceeds a defined limit, or its pH falls below a critical level. Hence, determination of the time to reaching these critical limits is key to the assessment of RC structures durability and service life. Due to the ionic nature of the chlorides and the bicarbonate anion (HCO3-) formed by the CO2 in the multi-ionic pore solution, the transport of both species is driven by Fickian diffusion combined with electromigration and ionic activity, which can be mathematically expressed by the Nernst-Planck-Poisson (NPP) equations. For a complete representation of the phenomenon, however, the NPP equations must be supplemented by the relevant chemical equilibrium equations to ensure chemical balance among the various species within the concrete pore solution. The combination of NPP with the chemical equilibrium equations is often termed the NPP reactive transport model. In this study, such a model is developed, coded into the MATLAB platform, validated by available experimental data, and applied to analyze the time-dependent concrete carbonation and the movement of chlorides in carbonated and non-carbonated concrete. The results of these analyses can be used to predict the time to corrosion initiation. The transient one-dimensional governing equations of NPP are numerically solved using the Galerkin’s finite element formulation in space and the backward (implicit) Euler scheme in the time domain. The associated system of chemical equilibrium equations accounts for the key homogeneous and heterogeneous chemical reactions that take place in the concrete during carbonation and chlorides transport. At each stage of the analysis, the effects of these reactions on the changes in the pore solution chemical composition, pH, cement chloride binding capacity, concrete porosity, and the hydrated cement solids volumetric ratio are determined. The study demonstrates that given accurate input data, the presently developed NPP reactive transport model can accurately simulate the complex transport processes of chlorides and CO2 in concrete as a reactive porous medium, and the ensuing physical and chemical changes that occur due to the reaction of these species with the pore solution and the other cement hydration products. This conclusion is supported by the good agreement between results of the current analyses with the corresponding available experimental data from physical tests involving carbonation, and chloride diffusion in non-carbonated and carbonated concrete. / Thesis / Doctor of Philosophy (PhD)

Aerated treatment wetlands are an increasingly recognized nature–based technology for thetreatment of domestic and industrial wastewater. As biodegradation is the most importanttreatment mechanism in aerated wetlands, these systems heavily rely on mechanical aerationmediated oxygen transfer to supply the dissolved oxygen demand of the associated microbialcommunity. In the last decade, research on aerated wetlands has evolved, however, majorquestions on aeration, the associated oxygen transfer and the quantitative link to treatmentperformance still remain unknown. Answering these questions can further improve aeratedwetland design to optimize treatment efficacy and economical efficiency. This dissertation investigated the link of oxygen transfer to the air flow rate of aerationand elucidated the associated impact on treatment performance for organic carbon and nitrogenin horizontal flow aerated wetlands. Therefore, a numerical process model includingone dimensional reactive transport was developed. This model describes the main processesinvolved in horizontal flow aerated wetlands: water flow, heat transport, transport of solubleand particulate wastewater pollutants, biodegradation by a network of bacterial communitiesand oxygen transfer through mechanical aeration. For model calibration and validation, pilot–scale experiments in horizontal flow aerated wetlands treating real wastewater were conducted.These included conservative tracer experiments as well as monitoring steady–state operationat variable air flow rates and aeration interruption. In general, the model was able to simulate conservative tracer transport as well as treatmentperformance for organic carbon and nitrogen at steady–state operation and aeration interruptionwith sufficient accuracy. A local sensitivity analysis of the calibrated parameters revealedporosity, hydraulic permeability and dispersion length as well as the oxygen transfer coefficientkLa as most important. When operating the wetland systems at steady–state, aeration provideda mostly aerobe environment, except at the influent zone. However, when aeration wasinterrupted, anaerobe process started to take over and treatment performance declined within3–4 days. The modeling elucidated that methanogenic and sulphate reducing bacteria can playa significant role for organic carbon removal during aeration interruption. Moreover, the modelrevealed a non–linear declining relationship of the air flow rate with oxygen transfer coefficientkLa and of kLa with treatment performance. The alteration of oxygen transfer by wastewaterpollutant concentration was then investigated in a laboratory–scale column experiment. Basedon this experiment, an empirical equation describing the inhibitory effect of soluble chemicaloxygen demand (CODs) on the oxygen transfer coefficient kLa was derived and incorporatedinto the process model. With the extended model several simulation scenarios were analyzedto quantify the impact of the inhibited oxygen transfer on treatment performance. It turnedout that the reduction of oxygen transfer by CODs will, most likely, be relevant only at highinfluent wastewater strength (CODs 300 mg L-1), low aeration (air flow rate 50 L m-2h-1) or when the aerated wetland design includes zoned aeration. With respect to secondarytreatment of domestic effluents at similar strength using a spatially uniform aeration, an airflow rate of approximately 150–200 L m-2 h-1 can be recommended as a reasonable compromisebetween treatment efficiency and robustness. If zoned aeration is intended (e.g. to create a redox zonation), however, the air flow rate should be increased to approximately 400 L m-2 h-1 to supress the inhibition of oxygen transfer by CODs concentration. Furthermore, the air flow rate at steady–state operation (50–500 L m-2 h-1) did not substantially affect the response in effluent concentrations for organic carbon and nitrogen. This means that at steady–state air flow rates of 50–500 L m-2 h-1 operation, treatment efficacy during aeration interruption will deteriorate and recover in a similar time. In conclusion, this dissertation provides quantitative insights into the mechanisms of aeration and treatment performance for organic carbon and nitrogen in horizontal flow aerated treatment wetlands. The findings obtained can support aerated treatment wetland design for research experiments and engineering applications. Therefore, this dissertation represents a significant advancement in the field of aerated treatment wetland research.

Ce travail vise à étudier le couplage entre écoulement et interactions physico-chimiques dans les sols, dans différentes conditions de saturation en eau, afin d’améliorer la prédiction du devenir des polluants. Il s’agit de comprendre en quoi le taux de saturation du milieu affecte la réactivité du sol vis-à-vis des polluants, et d’évaluer le pouvoir prédictif du transport de solutés réactifs étudié en milieu saturé sur la réactivité en conditions non saturées. Différents processus sont considérés : l’échange de cations calcium-zinc sur un milieu poreux modèle (sable-kaolinite), la sorption et désorption d’un composé organique sur une terre non contaminée, le transport de polluants prioritaires tels que les HAP sur une terre de friche industrielle. Dans chaque cas, des expériences en colonne de laboratoire ont été conduites en conditions d’écoulement saturé et non saturé permanent, permettant tout d’abord la caractérisation de l’hydrodynamique, puis l’étude du couplage avec la réactivité. Les courbes de percée obtenues ont été ensuite modélisées avec des codes tels que CXTFIT. On a montré l’influence de la teneur en eau du milieu sur le transport réactif, variable suivant le type de réaction considéré, la structure des milieux jouant également un rôle important. L’échange d’ions sur le milieu modèle n’est globalement pas affecté par la teneur en eau, dans une gamme proche de la saturation. En revanche, une plus forte sorption et une plus faible mobilisation des polluants organiques ont été observées en conditions non saturées. Le transport réactif de ces composés ne peut donc pas être prédit en conditions non saturées à partir de mesures en milieu saturé, qui peuvent surestimer le transport / The aim of this work was to study the link between water flow and physical and chemical interactions in soils under variably water flow conditions, in order to improve the prediction of contaminants fate. It deals with understanding how the porous media water content can modify soil reactivity towards contaminants, and assessing the possibility to predict reactivity under unsaturated conditions with reactive solute transport studied in saturated porous media. Various processes were considered: cations exchange calcium-zinc on a model porous media (sand-kaolinite), sorption and desorption of an organic compound on a non polluted soil, transport of priority contaminants such as PAHs on an industrial contaminated soil. In each case, experiments were carried out with soil columns at the laboratory scale under saturated and unsaturated steady-state flow conditions, in order to characterize at first hydrodynamics and then to study the link with reactivity. Modeling of the breakthrough curves was then performed with codes such as CXTFIT. We showed an influence of porous media water content on reactive transport which was different as a function of the interaction. Porous media structure must also be taken into account. Ions exchange on a model porous media was not globally modified by the water content varying in a range close to saturation. On the contrary, higher sorption and lower migration of organic contaminants were observed under unsaturated conditions. Reactive transport of these compounds cannot therefore be predicted under unsaturated conditions with tests performed on saturated porous media which may overestimate transport

L'atténuation du changement climatique est l'un des défis majeurs de notre époque. Les émissions anthropiques de gaz à effet de serre ont augmenté de façon continue depuis la révolution industrielle, provoquant le réchauffement climatique. Un ensemble de technologies très diverses doivent être mises en œuvre pour respecter les accords internationaux relatifs aux émissions de gaz à effet de serre. Certaines d'entre elles ont recours au sous-sol pour le stockage de diverses substances. Cette thèse traite plus particulièrement de la dynamique du stockage souterrain du dioxyde de carbone (CO2) et de l'hydrogène (H2). Des modèles numériques de transport réactif et multiphasiques ont été élaborés pour mieux comprendre la migration et les interactions des fluides dans des milieux poreux de stockage souterrain. Ils fournissent des recommandations pour améliorer l'efficacité, la surveillance et la sécurité du stockage. Trois modèles sont présentés dans ce document, dont deux dans le domaine du captage et du stockage du CO2 (CCS pour Carbon Capture and Storage), et le troisième s'appliquant au stockage souterrain de l'hydrogène (UHS pour Underground Hydrogen Storage). Chacun d'entre eux traite plus spécifiquement un aspect de la recherche : Modèle multiphasique appliqué au CCS L'efficacité et la sécurité à long terme du stockage du CO2 dépend de la migration et du piégeage du panache de CO2 flottant. Les grandes différences d'échelles temporelles et spatiales concernées posent de gros problèmes pour évaluer les mécanismes de piégeage et leurs interactions. Dans cet article, un modèle numérique dynamique diphasique a été appliqué à une structure aquifère synclinale-anticlinale. Ce modèle est capable de rendre compte des effets de capillarité, de dissolution et de mélange convectif sur la migration du panache. Dans les aquifères anticlinaux, la pente de l'aquifère et la distance de l'injection à la crête de l'anticlinal déterminent la migration du courant gravitaire et, donc, les mécanismes de piégeage affectant le CO2. La structure anticlinale arrête le courant gravitaire et facilite l'accumulation du CO2 en phase libre, en dessous de la crête de l'anticlinal, ce qui stimule la mise en place d'une convection et accélère donc la dissolution du CO2. Les variations de vitesse du courant gravitaire en raison de la pente de l'anticlinal peuvent provoquer la division du panache et une durée différente de résorption du panache en phase libre, qui dépend de l'endroit de l'injection. / Climate change mitigation is one of the major challenges of our time. The anthropogenic greenhouse gases emissions have continuously increased since industrial revolution leading to global warming. A broad portfolio of mitigation technologies has to be implemented to fulfill international greenhouse gas emissions agreements. Some of them comprises the use of the underground as a storage of various substances. In particular, this thesis addresses the dynamics of carbon dioxide (CO2) and hydrogen (H2) underground storage. Numerical models are a very useful tool to estimate the processes taking place at the subsurface. During this thesis, a solute transport in porous media module and various multiphase flow formulations have been implemented in COMSOL Multiphysics (Comsol, 2016). These numerical tools help to progress in the understanding of the migration and interaction of fluids in porous underground storages. Three models that provide recommendations to improve the efficiency, monitoring and safety of the storages are presented in this manuscript: two in the context of carbon capture and storage (CCS) and one applied to underground hydrogen storage (UHS). Each model focus on a specific research question: Multiphase model on CCS. The efficiency and long-term safety of underground CO2 storage depend on the migration and trapping of the buoyant CO2 plume. The wide range of temporal and spatial scales involved poses challenges in the assessment of the trapping mechanisms and the interaction between them. In this chapter a two-phase dynamic numerical model able to capture the effects of capillarity, dissolution and convective mixing on the plume migration is applied to a syncline-anticline aquifer structure. In anticline aquifers, the slope of the aquifer and the distance of injection to anticline crest determine the gravity current migration and, thus, the trapping mechanisms affecting the CO2. The anticline structure halts the gravity current and promotes free-phase CO2 accumulation beneath the anticline crest, stimulating the onset of convection and, thus, accelerating CO2 dissolution. Variations on the gravity current velocity due to the anticline slope can lead to plume splitting and different free-phase plume depletion time is observed depending on the injection location. Injection at short distances from the anticline crest minimizes the plume extent but retards CO2 immobilization. On the contrary, injection at large distances from anticline crest leads to large plume footprints and the splitting of the free-phase plume. The larger extension yields higher leakage risk than injection close to aquifer tip; however, capillary trapping is greatly enhanced, leading to faster free-phase CO2 immobilization. Reactive transport model on convective mixing in CCS. Dissolution of carbon-dioxide into formation fluids during carbon capture and storage (CCS) can generate an instability with a denser CO2-rich fluid located above the less dense native aquifer fluid. This instability promotes convective mixing, enhancing CO2 dissolution and favouring the storage safety.