The efficiency to improve the water quality of industrial and municipal wastewater in a constructed wetland/filter basin treatment system was investigated. The wetland system was constructed in a closed phosphate mine used for clay settling and sand tailings in Polk County, Florida. During 18-months of monitoring the chemical/microbiological composition of treated wetland water remained relatively constant, despite significant seasonal variations in temperature, rainfall and humidity. The following changes in water quality between input and output were observed: substantial decrease of water temperature (up to 10°C), reduction of As, SO4, F, Cl, NO3, NO2, Br, Na, K, Ca, and Mg, change in pH from 9 to 6.5-7, increase of H2S (up to 1060 micrograms/L), and a change from positive to negative ORP. There were no exceedances of the primary drinking water standards, volatile organic compounds, synthetic organic compounds, and radionuclides, but a number of exceedances for the secondary drinking water standards (Al, F, Fe, Mn, color, odor, total dissolved solids, and foaming agents). The concentration of fecal and total coliform bacteria in the wetland water was high, but subsequently reduced during filtration in the filter basin from 30 - 730 and 1000 - 7000 count/100 mL to < 2 and < 100 count/ 100 mL, respectively. To resolve the complex hydrogeological conditions a combined isotope/chemical mass-balance approach was applied. The results were the following: (1) the composition of water in the wetland varied throughout the period of the study; (2) a change in isotopic composition along the wetland flow path; (3) the wetland contained mainly wastewater (88 - 100 %) during normal pumping operations; however, hurricanes and inconsistent pumping added low conductivity water directly and triggered enhanced groundwater inflow into the wetland of up to 78 %; (4) the composition of water in monitor wells was mostly groundwater dominated; however periodically seepage from a water body to the north was detected; and (5) seepage from adjacent water bodies into the wetland was not identified during operation, which would indicate a potential water loss from the wetland. To test if the wetland system could be a prospective pre-treatment option for water used in aquifer storage and recovery (ASR) scenarios, a set of bench-scale leaching experiments was carried out using rocks from the Avon Park Formation, the Suwannee Limestone and the Ocala Limestone. Since As in the Floridan Aquifer was mainly present as an impurity in the mineral pyrite the elevated iron and sulfide concentrations in the wetland water were thought to prevent pyrite dissolution. The experiments which covered a range of redox conditions showed that the amount of As released from the aquifer matrix was not perfectly correlated with the bulk rock As concentration, nor the redox state of the water. The following important results were obtained: (1) the highest concentration of As was leached from the Avon Park Formation and the lowest - from the Suwannee Limestone, although the Ocala Limestone had the lowest bulk rock As; (2) minor to no As was released using native Floridan groundwater; (3) Tampa tap water, which chemically and physically resembled the ASR injection water, caused the As leaching of up to 27 micrograms/L, which was higher than the As drinking water standard; (4) the wetland and filter basin waters caused the highest release of As (up to 68 micrograms/L), which was unexpected because those water types were less oxygenated than Tampa tap water and thus should be less aggressive; (5) the in-situ filtration of the wetland water through a 0.2 micrometer membrane resulted in a reduction of As from 30 microgram/L to 16 microgram/L; and (5) the UV treatment significantly reduced both fecal and total coliform bacteria, but facilitated the increase of DO in initial waters, a change from negative to positive ORP, and the increase of As concentration in leachates. The experiments confirmed that perturbations of native aquifer conditions caused the release of As from the Floridan aquifer matrix, although the reaction may not be as simple as the dissolution of pyrite by oxygen, but additionally governed by a complex set of factors including the ORP of the system, SO4²?/S², Fe³?/Fe²?, dissolved organic carbon and microbial activity. In addition, the trend of As leaching could be governed by a set of factors, such as the porosity and permeability of the aquifer matrix influencing the rate and degree of free water saturation, amount of pyrite to be exposed to the preferential water flow paths, limited surface reactivity of pyrite with favored reactions on fractured mineral surfaces, the concentration and the selective leaching of As from individual pyrite crystals. To characterize and verify the geochemical processes in the column experiments, the Geochemist's Workbench reactive transport models (React and X1t) were developed. Results from the models correlated well to those from the column experiments and confirmed the following: (1) the water-rock reaction between the aquifer matrix and native groundwater was favorable for pyrite stability preventing the release of As into solution; (2) the injection of oxidizing surface water into reducing native groundwater caused a change in redox potential of the system thus promoting the dissolution of pyrite, and (3) 1D reactive transport model of water-rock reaction between the aquifer matrix and surface water indicated a diverse behavior of As along the column, such as the oxidative dissolution of pyrite, mobilization and simultaneous sorption of As onto neo-formed HFO, followed by the reductive dissolution of HFO and secondary release of adsorbed As, and the potential non-oxidative dissolution of pyrite contributing the additional source of As to the solution.

This dissertation presents the design of an integrated watershed model, WASH123D version 3.0, a first principle, physics-based watershed-scale model of integrated hydrology/hydraulics and water quality transport. This numerical model is comprised of three modules: (1) a one-dimensional (1-D) simulation module that is capable of simulating separated and coupled fluid flow, sediment transport and reaction-based water quality transport in river/stream/canal networks and through control structures; (2) a two-dimensional (2-D) simulation module, capable of simulating separated and coupled fluid flow, sediment transport, and reactive biogeochemical transport and transformation in two-dimensional overland flow systems; and (3) a three-dimensional (3-D) simulation module, capable of simulating separated and coupled fluid flow and reactive geochemical transport and transformation in three-dimensional variably saturated subsurface systems. The Saint Venant equation and its simplified versions, diffusion wave and kinematic wave forms, are employed for surface fluid flow simulations and the modified Richards equation is applied for subsurface flow simulation. The reaction-based advection-dispersion equation is used as the governing equation for water quality transport. Several physically and mathematically based numerical options are provided to solve these governing equations for different application purposes. The surface-subsurface water interactions are considered in the flow module and simulated on the basis of continuity of interface. In the transport simulations, fast/equilibrium reactions are decoupled from slow/kinetic reactions by the decomposition of reaction networks; this enables robust numerical integrations of the governing equation. Kinetic variables are adopted as primary dependent variables rather than biogeochemical species to reduce the number of transport equations and simplify the reaction terms. In each time step, hydrologic/hydraulic variables are solved in the flow module; kinetic variables are then solved in the transport module. This is followed by solving the reactive chemical system node by node to yield concentrations of all species. Application examples are presented to demonstrate the design capability of the model. This model may be of interest to environmental scientists, engineers and decision makers as a comprehensive assessment tool to reliably predict the fluid flow as well as sediment and contaminant transport on watershed scales so as to evaluate the efficacy and impact of alternative watershed management and remediation techniques prior to incurring expense in the field.

The construction of infrastructures such as dams requires for the foundations to lay on stable ground. One way to do so is to grout the rock fractures present in the bedrock using concrete, and this also ensures that lower amounts of groundwater reach the infrastructure itself. However, the continuous flow of groundwater carrying various dissolved chemical compounds, will trigger over time the deterioration of the concrete grout by dissolving the mineral phases resulting from the hydration process. This study aims at determining the variations in porosity and hydraulic conductivity induced by the dissolution of portlandite content present in the grout following the groundwater flow. Eight different cases have been simulated in COMSOL Multiphysics®, based on various combinations of fracture geometry (rough and smooth), hydraulic gradient, and inflowing groundwater composition. Each model has two main components, one for the flow simulation and one for the reactive transport. The modules implemented from COMSOL Multiphysics® are Darcy’s Law, Chemistry, and Transport of Diluted Species in Porous Media. The software PHREEQC has been used to determine the chemical species concentration for both the initial and boundary conditions. The results show that, at the end of a 100-day period, the mineral concentration decreases from 56.27% to 61.27% depending on the simulation considered. This leads to an increase in porosity ranging from 1.94% to 2.23%, while hydraulic conductivity displays a minimum growth of 6.42% and a maximum of 7.43%. The sensitivity analysis results reveal that the most influencing factor on the degradation is the hydraulic gradient, which is followed by the fracture geometry, while the inflowing groundwater composition impact is not as high as the previous ones. / Byggandet av infrastrukturer som dammar kräver att grunden läggs på stabil mark. Ett sätt att uppnå detta är att fylla sprickor i berggrunden med betong, vilket också säkerställer att mindre mängder grundvatten når infrastrukturen. Det kontinuerliga flödet av grundvatten, som bär med sig olika lösta kemikalier, kan dock med tiden leda till att betongen bryts ner genom att mineralfaserna som bildas vid hydratiseringsprocessen löses upp. Denna studie undersöker hur porositeten och den hydrauliska ledningsförmågan förändras när portlandit i betongfogar löses upp av grundvattenflödet. Åtta olika scenarier har simulerats i COMSOL Multiphysics® med olika kombinationer av sprickgeometri (grov och slät), hydraulisk gradient och inkommande grundvattensammansättning. Varje modell består av två huvuddelar: en för flödessimulering och en för reaktiv transport. Modulerna från COMSOL Multiphysics® som används är Darcys lag, kemi och transport av utspädda ämnen i porösa medier. Programvaran PHREEQC har använts för att fastställa koncentrationen av kemiska ämnen för både initiala och gränsvillkor. Resultaten visar att efter 100 dagar minskar mineralkoncentrationen med mellan 56,27% och 61,27%, beroende på simuleringen. Detta leder till en ökning av porositeten med 1,94% till 2,23%, medan den hydrauliska ledningsförmågan ökar med minst 6,42% och högst 7,43%. Känslighetsanalysen visar att den hydrauliska gradienten är den mest påverkande faktorn för nedbrytningen, följd av sprickgeometrin. Sammansättningen av det inkommande grundvattnet har inte lika stor påverkan.

The geological sequestration of carbon dioxide (CO2) is one of the options of controlling the greenhouse gas emissions. However, leakage of CO2 from the storage reservoir is a risk associated with geological sequestration. Over longer times, large-scale groundwater motion may cause leakage of dissolved CO2 (CO2aq). The objectives of this thesis are twofold. First, the modelling study analyzes the leakage of CO2aq along the conducting pathways. Second, a relatively safer mode of geological storage is investigated wherein CO2aq is injected in a carbonate reservoir. A reactive transport model is developed that accounts for the coupled hydrological transport and the geochemical reactions of CO2aq in the porous media. The study provides a quantitative assessment of the impact of advection, dispersion, diffusion, sorption, geochemical reactions, temperature, and heat transport on the fate of leaking CO2aq. The mass exchange between the conducting pathway and the rock matrix plays an important role in retention and reactions of leaking CO2aq. A significant retention of leaking CO2aq is caused by its mass stored in aqueous and adsorbed states and its consumption in reactions in the rock matrix along the leakage pathway. Advection causes a significant leakage of CO2aq directly from the reservoir through the matrix in comparison to the diffusion alone in the rock matrix and advection in a highly conducting, but thin fracture. Heat transport by leaking brine also plays an important role in geochemical interactions of leaking CO2aq.  Injection of CO2aq is simulated for a carbonate reservoir. Injected CO2-saturated brine being reactive causes fast dissolution of carbonate minerals in the reservoir and fast conversion of CO2aq through considered geochemical reactions. Various parameters like dispersion, sorption, temperature, and minerals reaction kinetics are found to play important role in the consumption of CO2aq in reactions. / <p>Research Funders:</p><p>(i) Higher Education Commission (HEC) of Pakistan</p><p>(ii) Lars Erik Lundberg Scholarship Foundation, Sweden</p>

Le stockage géologique du CO2 est l'une des diverses technologies étant explorées afin de réduire les émissions de carbone atmosphérique des processus industriels (i.e. combustion de l'énergie fossile). L'une des spécifiques caractéristiques de l'injection du CO2 en profondeur reste la possibilité de réactions géochimiques (dissolution-précipitation) entre la saumure réactive mobile (e.g. eau de formation enrichie en CO2) et la roche encaissante durant l'évolution spatiale et temporelle du CO2, conduisant à des modifications dans la structure des pores et par conséquent dans les propriétés d'écoulement du réservoir (e.g. la perméabilité k). Donc, ces changements structuraux peuvent largement contrôler l'injectivité, ainsi que le champ de pression dans le réservoir et aussi la propagation du CO2. Il demeure ainsi crucial d'explorer les changement dans les propriétés de réservoirs (e.g. structurales et hydrodynamiques) induits durant une injection de CO2 et explicitement les relations existantes entre eux (e.g. k ou surface réactive-Sr versus porosité- , k versus hétérogénéité de la roche), afin de développer des outils de modélisation prédictive des processus de transport et réactionnels se produisant durant une injection de CO2 et d'évaluer de façon fiable les risques. Dans le cas des réservoirs carbonatés, l'application des modèles prédictifs de transport réactif demeure toujours un enjeu, car contrainte par la forte hétérogénéité en leur sein ainsi que par l'incertitude dans la cinétique de réactions des minéraux carbonatés dans ce contexte. Dans cette optique, nous avons réalisé des expériences de percolation à travers des échantillons de roches carbonatées dans les conditions thermodynamiques de stockage en profondeur (T = 100°C et P =12 MPa). L'évolution de la perméabilité est suivie au cours des expériences ; et la variation de la porosité est calculée à partir des résultats d'analyses chimiques au ICP-AES des fluides de sortie échantillonnés. L'investigation des modifications apportées à la structure des pores est réalisée par le biais de la Micro-Tomographie haute résolution à rayon X, acquise au synchrotron de Grenoble (e.g. ESRF). Dépendant du régime de dissolution, contrôlé par la fabrique de la roche réservoir et la composition chimique de la saumuré chargée en CO2 (e.g. PCO2 engagée), on a observé qu'une modification de la structure de la roche peut soit améliorer soit détériorer (résultat atypique en contexte de dissolution) la valeur de la perméabilité k. Mots clés : Stockage géologique du CO2, transport, réactions géochimiques, structure des pores, propriétés hydrodynamiques, expériences de percolation de CO2, micro-tomographie à rayon X. / Geological storage of CO2 is one of diverse technologies being explored to reduce atmospheric carbon from industrial processes (i.e. fossil fuel combustion). One of the specific features of CO2 injection is the possibility of geochemical reactions (dissolution – precipitation) between mobile reactive brine (e.g. formation water enriched in CO2) and the host rock during the spatial and temporal evolution of CO2. That leads to modifications in the pore structure which in turn change the flow dynamics of the reservoir (e.g. the permeability k). Then, theses structural modifications can largely control the injectivity, so that the pressure field in the reservoir and also the CO2 propagation. Accordingly, it is crucial to explore the changes in the reservoir properties (e.g. structural and hydrodynamic) induced during a CO2 injection and specially the relationships between them (e.g. k or reactive surface-Sr versus porosity- , k versus rock heterogeneity), for developing predictive modelling tools of the transport and reaction processes occurring during a CO2 injection and reliable risk assessment. In the case of carbonate rocks, the application of the predictive models of transport and reaction is still challenging, because of their high heterogeneity so that the incertitude in the reaction kinetics of carbonate minerals. From this perspective, we realized brine-enriched in CO2 percolation experiments through carbonate rock samples in thermodynamic conditions expected during CO2 injection in deep reservoirs (T = 100°C et P =12 MPa). The permeability changes k(t) is monitored during the experiments and the porosity variation is calculated from chemical analyses of the sampled outlet fluids, using ICP-EAS. The pore structure modifications are investigated from high resolution X ray micro tomography images acquired from the synchrotron of Grenoble (ESRF). Depending to the dissolution regime, controlled by the reservoir rock fabric and the chemical composition of the brine (e.g. PCO2), we observed that a modification of pore structure can either improve (atypical result in dissolution context) or impair the value of the permeability k. Keywords: CO2 geological storage, transport, geochemical reactions, pore structure, hydrodynamic properties, brine enriched in CO2 percolation experiments, X ray microtomography.

La dissolution de verres borosilicatés en présence de minéraux magnésiens a été étudiée. Ces minéraux (dolomite, illite, smectite…) appartiennent à la couche géologique (Callovo-Oxfordien) destinée à accueillir le stockage des déchets nucléaires vitrifiés en France. Ils contiennent du magnésium, élément capable d'entretenir l'altération du verre lorsqu'il est disponible en solution. Dans les milieux confinés du stockage, la réactivité des solides contrôle la composition de la solution et peut être la force motrice de l'altération des verres nucléaires. Les expériences montrent que les carbonates magnésiens (hydromagnésite, dolomite) entretiennent l'altération du verre : la précipitation de silicates de magnésium empêche la recondensation du silicium dans la couche passivante en surface du verre. Plus le minéral magnésien est soluble, plus l'altération du verre est importante. Les phases argileuses purifiées (illite, smectite…) du Callovo-Oxfordien (COx) augmentent également l'altération du verre. La moitié du magnésium échangeable de ces phases a été remplacée par du sodium lors du protocole de purification. Dans ces conditions, l'effet des phases argileuses sur l'altération du verre est en partie dû au pH acide qu'elles imposent. Le modèle d'altération des verres GRAAL implémenté dans le code de transport réactif HYTEC a permis de confirmer et de quantifier les mécanismes identifiés à partir des expériences en système fermé. Des expériences en cellule de diffusion, deux compartiments séparés par une barrière diffusive inerte, ont permis de valider une modélisation du transport réactif. Ces expériences, plus représentatives des conditions de stockage, où le bloc de verre sera séparé du COx par les produits de corrosion des aciers, illustrent le ralentissement des cinétiques attendu compte tenu de l'éloignement du verre et des minéraux réactifs. / Borosilicate glasses dissolution has been studied in presence of magnesium minerals. Those minerals (dolomite, illite, smectite…) belong to the Callovo-Oxfordian (COx) claystone layer, studied in France as a potential site for nuclear waste disposal. Such minerals contain magnesium, an element able to sustain glass alteration when it is available in solution. In the confined media of the wastes disposal, thesolids reactivity controls the solution composition and can be the driving force of nuclear glass alteration. Experiments show that magnesium carbonates (hydromagnesite and dolomite) increase in the glass alteration: the precipitation of magnesium silicates consumes silicon which slows down the formation of the glass passivating layer. The lower the magnesium mineral solubility, the lower the glass alteration.The purified clay phases (illite, smectite…) from the COx layer increase the glass alteration. Half the magnesium was remplaced by sodium during the purification process. In such conditions, the effect of clay phases on glass alteration is in part due to the acidic pH-buffering effect of the clay fraction. The GRAAL model implemented in the geochemical transport code HYTEC has confirmed and quantified the mechanisms put in evidence in the experiments. Cells diffusion experiments where the two solids were separated by an inert diffusion barrier allow to valid reactive transport modelling. Such experiments are more representative of the glass package which will be separated from the COx by corrosion products. They show that glass alteration rate is reduced when solids are not close.

Ce travail entre dans le cadre de l’étude des interactions eau-roche dans le cas du stockage du CO2 en milieu géologique. Un intérêt particulier est accordé aux hétérogénéités des paramètres associés aux phénomènes géochimiques. Ces hétérogénéités peuvent s’observer à différentes échelles: celle des grains (les minéraux présentent des défauts de cristallinité et des impuretés), et l’échelle centimétrique/pluri-décamétrique. En particulier, les paramètres thermodynamiques (logK) et de cinétique chimique (dans ce travail nous avons considéré le produit de la constante cinétique k par la surface spécifique S soit kS comme "paramètre de cinétique chimique") sont connus à partir des expériences de laboratoire pour des échantillons de quelques centimètres de dimension, alors que l’on s’intéresse aux réactions minéralogiques à l’échelle des réservoirs.Nous avons évalué les caractéristiques géostatistiques de la variabilité spatiale après réaction à travers des simulations de transport réactif dans lesquelles différents paramètres (logK et kS) sont perturbés avec une première variabilité imposée. Une combinaison de deux approches est ainsi abordée : déterministe et géostatistique. Le code du transport-réactif COORES (IFP-EN et Ecole nationale supérieure des mines de Saint-Etienne) a été utilisé pour les simulations déterministes et le système géochimique étudié concerne la dissolution du diopside avec précipitation de minéraux secondaires comme la calcite et la magnésite.Après analyse par la méthode des plans d’expériences, les résultats montrent qu’une corrélation spatiale élevée combinée avec une grande variance de dispersion des minéraux favorise une réactivité importante des minéraux lorsqu’on perturbe le paramètre de cinétique chimique kS. Par ailleurs une vitesse d’injection élevée accélère le processus de dissolution du minéral étudié. La variabilité spatiale du paramètre thermodynamique n’a cependant pas d’effet significatif sur les résultats, le système se comporte comme dans le cas homogène. Du point de vue de l’homogénéisation du paramètre kS, on retrouve l’influence de l’historique de dissolution. / The present work is based on the study of water-rock interactions in the case of CO2 storage in geological media. Particular attention is devoted to heterogeneities at different observation scales geochemical phenomena. These heterogeneities can be observed at different scales: the grain (mineral crystallinity present defects and impurities), and the centimeter scale / multi- decametric (rocks are heterogeneous at different scales). In particular, the thermodynamic parameters logK and chemical kinetics kS (in this work we considered the product of the rate constant k by the specific surface area S is kS as "chemical kinetics parameter") are known from laboratory experiments to a few centimeters in size, while we are interested in mineralogical reactions across tanks.We propose to evaluate the geostatistical characteristics of the local variability after reaction through simulations of reactive transport on a small scale in which various parameters (logK and kS) are perturbed with a first spatial variability imposed. A combination of both approaches is discussed: deterministic and geostatistical for the study of geochemical problems at different scales. The reactive transport code - COORES (IFP - EN and Ecole nationale supérieure des mines de Saint -Etienne) was used for deterministic simulations and the geochemical system studied concerns the dissolution of diopside with precipitation of secondary minerals such as calcite and magnesite.After analysis by the method of design of experiments, the results show that high spatial correlation variance combined with high dispersion of minerals promotes a high reactivity when minerals chemically disturbing is the kinetic parameter kS. In addition, a high velocity injection accelerates the dissolution of the mineral studied. However, the effect of spatial variability of the thermodynamic parameter, did not significantly affect the results, the system behaves as in the homogeneous case. From the standpoint of homogenizing the parameter kS, include the influence of the history of dissolution.

L’eau des circuits d’une centrale nucléaire est purifiée à l’aide de résines échangeuses d’ions. La prédiction de leurs performances constitue une aide importante pour l’exploitation de ces réacteurs. Les méthodes du transport réactif sont particulièrement adaptées pour cela et constituent la base du code OPTIPUR, dédié à la modélisation de ces résines. Le travail présenté comporte trois axes principaux. Le premier est l’intégration d’une limitation au transfert de masse dans une colonne de résines échangeuses d’ions, avec une mobilité spécifique à chaque espèce chimique, dans le cadre d’un découplage des calculs de chimie et de transport. Ce modèle permet, sans paramètre ajustable, de reproduire assez fidèlement une série d’expériences réalisées précédemment par le CEA. Le second axe concerne les aspects numériques du transport réactif, avec l’utilisation de la méthode d’Anderson pour accélérer la convergence du couplage chimie-transport dans un schéma itératif. En utilisant les informations issues des itérations précédentes et sans modification majeure du code, la robustesse et les temps de calcul ont pu être nettement améliorés. La troisième thématique abordée est celle de l’équilibre d’échange d’ions. Les bases d’un modèle prenant en considération l’évolution de l’humidité de la résine, ainsi que son élasticité sont proposées ; les interactions entre groupes fonctionnels, contre-ions et eau sont considérées comme des équilibres chimiques. Les constantes d’équilibre sont ajustées à partir de mesures de la teneur en eau de la résine à différentes pressions de vapeur d’eau. Finalement, des coefficients de sélectivité apparents peuvent être calculés et comparés aux mesures disponibles. / In nuclear power plants, the water contained in different circuits is purified by passing through ion exchange resins. Prediction of the performance of these resins is an important help to the plant operators. To this end, the method of reactive transport modeling are well suited and is the basis of the OPTIPUR code that was designed to model the resins. The work presented in this manuscript covers three main aspects. The first one is the integration of a limitation to mass transfer in a ion exchange deep bed, taking into account a specific mobility for each chemical species, in the context of separated calculations for chemistry and transport. This model was shown to reproduce experimental data, without adjustable parameters. The second part of this work deals with the numerical aspects of reactive transport modelling. A method developped by Anderson was used to accelerate the convergence of the chemistry transport coupling in an iterative scheme. Using the information from previous iterations, and without major changes in the code, calculation times were largely decreased, as well as the number of calculations failures. The third topic is ion exchange equilibrium. The basis of a model that takes into account the change in the water content of the resin and its elasticity are described. The interactions between the fonctional groups, the counterions and water are considered as chemical reactions. The corresponding equilibrium constants are fitted to measurements of the water content of the resin at different relative humidity. Finaly, the selectivity coefficients can be calculated and compared to litterature values.

Chemical weathering of silicate rock consumes atmospheric CO2 and supplies the oceans with cations, thereby controlling both seawater chemistry and climate. The rate of CO2 consumption is closely linked to the rate of CO2 outgassing from the planetary interior, providing a negative feedback loop essential to maintaining an equable climate on Earth. Reconstruction of past global temperatures indicates that a pronounced episode of global cooling began ~50 million years ago, coincident with the collision of India and Asia, and the subsequent exhumation of the Himalayas and Tibet. This has drawn attention to the possible links between exhumation, erosion, changes in silicate weathering rates, and climate. However, many of the present-day weathering processes operating on the continents remain debated and poorly constrained, hampering our interpretations of marine geochemical archives and past climatic shifts. To constrain the controls on silicate weathering, this thesis investigates the lithium (Li) isotopic composition of river waters, suspended sediments and bed load sediments in the Alaknanda river basin, forming the headwaters of the Ganges. Due to the large fractionation of Li isotopes in the Earth’s surface environment, Li is sensitive to small changes in silicate weathering processes. As a consequence of the pronounced gradients in climate (rainfall and temperature) and erosion across the basin, the river waters show large variations in their Li isotopic composition (δ7Li), ranging from +7.4 to +35.4‰, covering much of the observed global variation. This allows a detailed investigation of the controls on Li isotope fractionation, and by extension silicate weathering. The Li isotopic composition is modelled using a one-dimensional reactive transport model. The model incorporates the continuous input of Li from rock dissolution, removal due to secondary mineral formation, and hydrology along subsurface flow paths. Modelling shows that the Li isotopic variations can be described by two dimensionless variables; (1) the Damköhler number, ND, which relates the silicate dissolution rate to the fluid transit time, and (2) the net partition coefficient of Li during weathering, kp, describing the partitioning of Li between secondary clay minerals and water, which is primarily controlled by the stoichiometry of the weathering reactions. The derived values of the controlling parameters ND and kp, are investigated over a range of climatic conditions and on a seasonal basis, shedding light onto variations in the silicate weathering cycle. In a kinetically limited weathering regime such as the Himalayan Mountains, both climate and erosion exert critical controls the weathering intensity (the fraction of eroded rock which is dissolved) and the weathering progression (which minerals that are being weathered), and consequently the fractionation of Li isotopes and silicate weathering in general. Modelling of the Li isotopic composition provides an independent estimate of the parameters which control silicate weathering. These estimates are then used to constrain variables such as subsurface fluid flux, silicate dissolution rates, fluid transit times and the fraction of rock which is weathered to form secondary clay minerals. The simple one-dimensional reactive transport model therefore provides a powerful tool to investigate the minimum controls on silicate weathering on the continents.

La protection et la remédiation des ressources en eau sont un enjeu sociétal majeur, ainsi il est nécessaire de comprendre l’évolution de solutés, tels les polluants, au sein de la zone saturée et non saturée. Dans ce but, de nombreux travaux ont été consacrés à la modélisation du transport réactif en milieux poreux. Son déroulement à l’échelle de Darcy dépend des hétérogénéités microscopiques du milieu. Les modèles de réseau de pores qui simplifient la géométrie en un ensemble de pores reliés par des liens de sections constantes, permettent de se placer à une échelle mésoscopique, faisant le lien entre l’échelle porale et l’échelle de Darcy. Sur de telles formes géométriques, l’écoulement admet un traitement analytique. En ce qui concerne le transport réactif des solutés, nous proposons une solution analytique dans les liens qui permet de calculer le débit de masse entre pores. Le modèle de transport se formule alors comme un système d’équations de Volterra de secondes espèces dont les noyaux de convolution sont des séries d’exponentielles décroissantes (hormis le premier terme qui est constant). Leurs temps de relaxation sont pilotés essentiellement par le temps de dispersion td. Dans la limite où td tend vers 0 à Péclet constant, les termes transitoires des noyaux se réduisent à un Dirac, débouchant sur un premier modèle simplifié à réponse instantanée c'est-à-dire un modèle de transport quasi-statique. Dans le cas où les volumes des pores sont suffisamment grands, les noyaux se réduisent à leur premier terme. Ces formulations du transport généralisent celles de la littérature. En particulier pour des Péclet petit ou grand on retrouve respectivement les modèles usuels en régime dispersif et convectif. Numériquement, la décroissance exponentielle des noyaux permet d’optimiser le calcul des convolutions avec une précision arbitrairement fixée, réduisant drastiquement le temps de résolution. / Protection and remediation of ground water resources are a major societal challenge. It implies to understand the evolution of solutes as pollutant in the saturated and non-saturated zones. For that purpose numerous studies have been conducted for modeling the reactive transport in a porous media. At Darcy scale, the behavior of solutes depends on microscopic heterogeneity for the media. The Pore Network Models (PNM) simplifies drastically its geometry and considers pores linked by straight throats the section of which is constant. They give a description which is in between the macroscopic and the pore descriptions. With such geometry it is possible to use a Poiseuille flow modeling the flux. With respect to the reactiontransport equation, we seek the analytical solution of the CDE in throats, which in turn allows computing the mass flux in pores. The transport solution consists of a Volterra equation system. Its convolution kernels result in a summation of time function which is decreasing exponentially with time (except the first term which still constant). The time constant is driven by the diffusion time td. As td goes to zero, keeping the Peclet number fixed, each term of the summation reduces to a Dirac. The response of the system is then instantaneous. When the volume of the pore is large enough it is possible to neglect all the term of the kernel except the constant one. In the limit where the Peclet number goes to zero, usual models are recovered. Numerically, the exponential time decreasing of the kernel allow to optimize their computational time up to an arbitrary fixed precision.