Le diagnostic du SAGE Loire au niveau de la plaine du Forez (42) relate une dégradation du potentiel hydrogéologique et une pollution des nappes superficielles par les nitrates et les pesticides. Ces problèmes nécessitent une gestion durable et équilibrée de la ressource en eau. Une partie de la demande en eau est assurée par la ressource souterraine. Les eaux souterraines exploitées sont : les eaux des formations alluviales superficielles pour l’irrigation et l’AEP ; et les eaux minérales des niveaux aquifères profonds du bassin sédimentaire sous-jacents aux alluvions pour le thermalisme et de mise en bouteille.La synthèse des connaissances disponibles a révélé des insuffisances dans la description des aquifères superficiels. En effet, elles ne permettent pas d’appréhender comment fluctue la ressource dans le temps et ne prennent pas en compte l’origine de la recharge des aquifères superficiels. Ainsi, le travail de thèse se focalise sur ces deux points particuliers :- l’identification du fonctionnement transitoire des aquifères est abordée par caractérisation physique des écoulements s’appuyant sur l’analyse de 4 points de suivi piézométrique permanent et la modélisation numérique des écoulements permettant de reproduire les fluctuations piézométriques observées.- l’origine de la recharge des aquifères alluviaux est abordée par caractérisation géochimique des écoulements (analyses en ions majeurs et isotopes δ180, δ2H, δ13C et A14C). Cette approche a mis en évidence des apports d’eaux profondes alimentant localement les aquifères superficiels.La méthodologie proposée, couplée physique et géochimie, permet de mieux préciser la vulnérabilité des aquifères investigués, en vue d’une meilleure gestion future. / The Water Management Scheme in the Forez plain (France 42) has concluded to a degradation of the water resources and a serious pollution of the shallow aquifers by nitrogen and pesticides. These issues require a sustainable and balanced management of the water resources. A part of the water demand is met by groundwater. The exploited groundwater come from: shallow alluvial formations exploited for irrigation and drinking water and deep aquifers in sedimentary basin under the alluvium. Mineral waters from the deep aquifers are exploited for thermo-mineral activities and bottling.The analysis of the data available in the studied area revealed a lack of knowledge in the understanding of the groundwater flow in the shallow aquifers. Is not sufficient to explain how the water resource varies with time and the origin of the recharge of the shallow aquifers. As consequence, this thesis work will focus on these two main topics: - Identification of the transient functioning of the aquifers based on a physical characterization of groundwater flow. This method is based on piezometric head analysis carry out on 4 stations with continuous measurements and numerical modeling to compare with the observed piezometric fluctuations.- Identification of the origin of the recharge of the alluvial aquifers through a geochemical characterization of groundwater flow (major ions and isotopes δ180, δ2H, δ13C and A14C). This approach established the partial contribution of the deep aquifers to the groundwater flow in the shallow alluvium. This methodology based on a coupled geochemical and physical approach allows to assess the vulnerability of the studied aquifers and contributes to improve the future water resources management in this area.

Le bassin crétacé de Tsagaan Els est situé dans le désert de Gobi en Mongolie. L’objectif de cette thèse est de mieux comprendre le fonctionnement hydrogéologique et hydrochimique de ce bassin endoréique afin de faciliter l’implantation de futurs projets miniers. Dans cette optique, un modèle hydrogéologique régional 3D a tout d’abord été développé sous MODFLOW. La calibration à partir des niveaux piézométriques mesurés sur les ouvrages présents au sein des licences et quelques puits nomades, aboutit à plusieurs solutions. Elles présentent chacune des conductivités hydrauliques en accord avec les tests de pompages, et des taux de recharge entre 0.6 et 3.1 mm/an, cohérents avec le climat actuel. Ce modèle a ainsi permis d’étendre nos connaissances piézométriques au-delà des licences et d’effectuer un bilan hydrique régional. Ensuite, l’âge des eaux a été estimé numériquement via ce modèle puis par des datations 14C. Une comparaison des résultats de ces méthodes a indiqué que les solutions présentant des taux de recharges élevés étaient plus probables que celles en présentant de plus faibles à conditions que les écoulements soient restés les mêmes durant les trente derniers milliers d’années. En parallèle, une étude hydrochimique portant sur les ions majeurs, le brome et les isotopes stable de l’eau (18O et 2H) a montré que l’origine de la salinité des eaux était principalement liée à la dissolution d’évaporites superficielles. En dernier lieu, un modèle 3D local, centré sur la playa terminale du bassin a permis de mieux comprendre le fonctionnement actuel et passé du bassin en s’appuyant sur l’évolution spatiotemporelle de la salinité. / The Tsagaan Els basin is located in the Gobi desert, in Mongolia. This thesis aims to better understand the hydrogeology and hydro-chemistry of the basin in order to optimize the future mining projects. To this end, a 3D groundwater flow model of the basin has first been developed under MODFLOW. The latter was calibrated using piezometric network data and levels in a few nomadic wells. Several solutions were found with hydraulic conductivity values in agreement with pumping tests and recharge rates between 0.6-3.1 mm/yr in agreement with the current climate. The model helped us to extend our piezometric knowledge beyond AREVA’s mining licenses and to establish a regional water balance model. Groundwater age was then estimated with this model and with 14C dating. A comparison of the results of these two methods showed that calibrated solutions with higher recharge are more likely than those with lower recharge under the hypothesis that groundwater flow has not changed significantly during the last 30 kyrs. Meanwhile, a hydrochemical study using major ions, bromines and the stable isotopes of water (18O and 2H) showed that groundwater salinity mainly comes from dissolution of surficial evaporites: likely gypsum and halite and possibly thenardite. Even though evaporation is the main driving force of the groundwater flow, this process, when taking place in subsurface, seems to have no real influence on δ18O and δ2H values of the groundwater. Last a 3D local groundwater flow model centered on the terminal playa and based on the spatiotemporal evolution of the salinity, led us to a better understanding of the past and current hydrogeological functioning of the basin.

Groundwater studies face computational limitations when providing local detail within regional models. The researchers are concentrated on applying the numerical models to minimize the difference between the physical reality and the implemented numerical model by considering the minimum computational cost. This work consists of the study of line-elements (such as line-doublets, circles, polygons, fractures) using the Analytic Element Method (AEM) for groundwater flow. In this work, we consider the study of two-dimensional groundwater flow in fractured porous media by the Analytic Element Method. We develop a numerical solution based on a series expansion for a problem with more than one fracture. Each fracture has an influence that can be expanded in a series that satisfies Laplaces equation exactly. In the series expansion, the unknown coefficients are obtained from the discharge potentials of all other elements that are related to the expansion coefficients. Sizes, locations and conductivities for all inhomogeneities are selected arbitrarily. This work also discusses a matrix method obtained by imposing the intern boundary conditions for the Analytic Element Method. The convergence analysis of a Gauss-Seidel type iterative method is also discussed. / Estudos de águas subterrâneas enfrentam limitações computacionais ao fornecer detalhes locais em modelos regionais. Os pesquisadores estão concentrados na aplicação dos modelos numéricos para minimizar a diferença entre a realidade física e o modelo numérico implementado considerando o custo computacional mínimo. Este trabalho consiste no estudo de elementos de linha (como line-doublets, círculos, polígonos, fraturas) usando o Método de Elemento Analítico (AEM) para o fluxo de águas subterrâneas. Neste trabalho, consideramos o estudo do fluxo bidimensional de águas subterrâneas em meios porosos fraturados pelo Método dos Elementos Analíticos. Desenvolvemos uma solução numérica baseada em uma expansão em série para um problema com mais de uma fratura. Cada fratura tem uma influência que pode ser expandida em uma série que satisfaça exatamente a equação de Laplace. Na expansão da série, os coeficientes desconhecidos são obtidos a partir dos potenciais de descarga de todos os outros elementos que estão relacionados aos coeficientes de expansão. Tamanhos, locais e condutividades para todas as não-homogeneidades são arbitrariamente selecionados. Este trabalho também discute o método da matriz obtido impondo as condições de contorno do interno para o Método do Elemento Analítico. A análise de convergência de um método iterativo tipo Gauss-Seidel também é discutida.

La caractérisation de l'agencement spatial des propriétés hydrauliques est essentielle pour prédire les écoulements et le transport des solutés dans les milieux hétérogènes. Les méthodes de tomographie hydraulique, principalement développées pour estimer les propriétés des milieux poreux, n'ont qu'une faible résolution spatiale qui ne reflète pas la vraie hétérogénéité des distributions de fractures des milieux fracturés. Le principal objectif de cette thèse est de développer une nouvelle méthode d'inversion spécifique pour imager les propriétés hydrauliques et de transport des milieux fracturés à l'échelle du site. Pour atteindre ces objectifs, des expériences in situ ainsi qu'une nouvelle approche de modélisation inverse sont proposées, notamment en utilisant la température comme marqueur des écoulements. Nous proposons tout d'abord la tomographie d'écoulement basée sur des tests séquentiels de débimétrie entre puits, comme une nouvelle approche pour caractériser la connectivité des fractures ainsi que leur transmissivité. À partir de simulations numériques reproduisant des cas d'études synthétiques, nous montrons que l'approche par tomographie réduit significativement l'incertitude sur les paramètres estimés, et fournit une caractérisation détaillée du réseau de fracture sans requérir à l'utilisation d'obturateurs hydrauliques. Nous montrons ensuite comment les mesures de température peuvent être utilisées pour quantifier les écoulements dans les milieux fracturés. Le grand intérêt d'utiliser la température est d'obtenir facilement et de façon continue en puits des profils de température. En utilisant un modèle numérique d'écoulement et de transfert de chaleur à l'échelle du puits, une méthode d'inversion pour estimer les vitesses d'écoulement dans le puits à partir des données de température est proposée. Nous couplons ensuite les deux approches présentées précédemment dans une nouvelle approche expérimentale consistant en des enregistrements séquentiels de température dans un puits dans des conditions de pompage entre puits. L'application de cette approche de tomographie en température sur le site de Stang er Brune montre des résultats encourageants pour l'identification du réseau global de connectivité et des zones d'écoulement principales. Enfin, nous discutons de l'intérêt d'utiliser la chaleur comme traceur par rapport à l'utilisation de traceurs classiques. Nous montrons que réaliser des tests de traçage thermiques en milieu fracturé fournit des contraintes supplémentaires importantes sur les propriétés de transport du milieu. / The accurate characterization of distribution of hydraulic properties and connectivity distribution is essential to predict flow and transport in fractured media. Classical approaches were developed for homogeneous aquifers and result in smooth tomograms that often do not match true heterogeneity distribution of fractured media. The main goal of this thesis is to develop new inverse approaches specifically for imaging hydraulic and transport properties in fractured media at the field-scale. To attain this objective new in situ measurement methods as well as new inverse modelling frameworks are proposed.We first propose flow tomography (i.e., sequential cross-borehole flowmeter tests) as a new approach for characterizing fracture connectivity and transmissivities. Based on a discrete fracture network approach, we present a general method to invert flow tomography data. From synthetic case studies, we show that the tomographic approach reduces significantly the uncertainty on the parameter estimation. Flow tomography approach provides detailed characterization on fracture networks without the necessity of using packers. We then study the contribution of temperature measurements for quantifying flow in fractured media. The advantage of using temperature data is that temperature profiles can be obtained more easily and continuously in space, compared to flowmeter profiles. Using a numerical model of flow and heat transfer at the borehole scale, a method to invert temperature measurements to derive borehole flow velocities was proposed. We then couple the two previously proposed approaches in a new experimental approach which we call temperature tomography. This experiment consists of sequential borehole temperature logging under cross-borehole flow conditions. The full inverse framework is then presented to interpret temperature tomography experiments. Application of the temperature tomography approach to Stang er Brune field site showed encouraging results for the identification of general connectivity patterns and transmissivities of the main flowpaths. Finally, we explore the interest of using push-pull thermal tracer tests. Through field experiments and numerical modelling, we demonstrate that conducting push-pull heat tracer tests provide important constraints on the effective transport behavior.

Os eventos de escassez hídrica observados nos últimos anos e o aumento populacional no Estado de São Paulo têm conduzido à maior explotação de água subterrânea do Sistema Aquífero Guarani (SAG) para abastecimento público. Somente no município de Bauru/SP, está previsto um aumento da vazão de água extraída deste Aquífero de 3.699 m3/h (2014) para 4.465 m3/h (2034). Entretanto, a superexplotação em longo prazo do Aquífero pode comprometer a quantidade de água disponível. O Método de Elementos Analíticos foi utilizado para a modelagem do escoamento em regime permanente, considerando o escoamento regional, a extração de água por poços e os principais condicionantes geológicos locais. Observou-se que o gradiente hidráulico do SAG na cidade de Bauru é de aproximadamente 0,82 m/km. Os poços públicos que operam em regime de escoamento confinado estão localizados a nordeste da cidade. O rebaixamento esperado devido ao cenário previsto pelo DAE, em 2034, é de até 15 m com relação ao cenário atual (2014). Este rebaixamento pode ser controlado em até 5 m, reduzindo em 20% a vazão média anual de todos os poços, e em até 10 m, reduzindo em 10% a vazão média anual de todos os poços localizados na área central. A extração de água do SAG em Bauru/SP pode causar o rebaixamento de até 15 m nas cidades mais próximas (Piratininga e Agudos) e pouco interfere no escoamento de água do SAG das cidades mais distantes. Finalmente, os mapas potenciométricos e de rebaixamento gerados pelo modelo analítico são adequados para a análise do escoamento de água subterrânea e podem auxiliar no gerenciamento de recursos hídricos. / The recent drought events and the population growth in São Paulo State (Brazil) have caused many municipalities to increase the groundwater exploitation of the Guarani Aquifer System (GAS). In the city of Bauru/SP, the extraction of water from the Aquifer is expected to increase from 3,699 m3/h, in 2014, to 4,465 m3/h, by 2034. However, the long-term overexploitation of the Aquifer may compromise the amount of available water. The Analytical Element Method was used for the groundwater flow modeling in steady state. The regional flow, the water extraction by wells and the main local geological conditions were considered. The hydraulic gradient in the GAS in Bauru is approximately 0.82 m/km. The public wells in operation in confined flow regime are located in the eastern region of the city. The expected drawdown caused by the scenario predicted by the Department of Water and Wastewater of Bauru, in 2034, can reach 15 m, in comparison with the current situation (2014). Such lowering can be controlled up to 5 m, reducing 20% of the average annual groundwater flow rate of all wells, and up to 10 m, reducing 10% of the annual flow rate of all wells located in the critical area. The GAS groundwater extraction in Bauru/SP may cause the lowering of up to 15 m in nearby cities (Piratininga and Agudos) and has negligible interference in the groundwater flow in the GAS of distant cities. Finally, the potentiometric and drawdown maps generated by the analytical model are suitable for the analysis of groundwater flow and can be useful for the management of water resources.

Os eventos de escassez hídrica observados nos últimos anos e o aumento populacional no Estado de São Paulo têm conduzido à maior explotação de água subterrânea do Sistema Aquífero Guarani (SAG) para abastecimento público. Somente no município de Bauru/SP, está previsto um aumento da vazão de água extraída deste Aquífero de 3.699 m3/h (2014) para 4.465 m3/h (2034). Entretanto, a superexplotação em longo prazo do Aquífero pode comprometer a quantidade de água disponível. O Método de Elementos Analíticos foi utilizado para a modelagem do escoamento em regime permanente, considerando o escoamento regional, a extração de água por poços e os principais condicionantes geológicos locais. Observou-se que o gradiente hidráulico do SAG na cidade de Bauru é de aproximadamente 0,82 m/km. Os poços públicos que operam em regime de escoamento confinado estão localizados a nordeste da cidade. O rebaixamento esperado devido ao cenário previsto pelo DAE, em 2034, é de até 15 m com relação ao cenário atual (2014). Este rebaixamento pode ser controlado em até 5 m, reduzindo em 20% a vazão média anual de todos os poços, e em até 10 m, reduzindo em 10% a vazão média anual de todos os poços localizados na área central. A extração de água do SAG em Bauru/SP pode causar o rebaixamento de até 15 m nas cidades mais próximas (Piratininga e Agudos) e pouco interfere no escoamento de água do SAG das cidades mais distantes. Finalmente, os mapas potenciométricos e de rebaixamento gerados pelo modelo analítico são adequados para a análise do escoamento de água subterrânea e podem auxiliar no gerenciamento de recursos hídricos. / The recent drought events and the population growth in São Paulo State (Brazil) have caused many municipalities to increase the groundwater exploitation of the Guarani Aquifer System (GAS). In the city of Bauru/SP, the extraction of water from the Aquifer is expected to increase from 3,699 m3/h, in 2014, to 4,465 m3/h, by 2034. However, the long-term overexploitation of the Aquifer may compromise the amount of available water. The Analytical Element Method was used for the groundwater flow modeling in steady state. The regional flow, the water extraction by wells and the main local geological conditions were considered. The hydraulic gradient in the GAS in Bauru is approximately 0.82 m/km. The public wells in operation in confined flow regime are located in the eastern region of the city. The expected drawdown caused by the scenario predicted by the Department of Water and Wastewater of Bauru, in 2034, can reach 15 m, in comparison with the current situation (2014). Such lowering can be controlled up to 5 m, reducing 20% of the average annual groundwater flow rate of all wells, and up to 10 m, reducing 10% of the annual flow rate of all wells located in the critical area. The GAS groundwater extraction in Bauru/SP may cause the lowering of up to 15 m in nearby cities (Piratininga and Agudos) and has negligible interference in the groundwater flow in the GAS of distant cities. Finally, the potentiometric and drawdown maps generated by the analytical model are suitable for the analysis of groundwater flow and can be useful for the management of water resources.

As part of the new navigation lock for Bonneville Dam a new water source for the Bonneville Fish Hatchery must be supplied. The hatchery is located on the Oregon side of the Columbia River downstream of the dam. It requires large quantities of water free from chemical and biological contamination. In addition, the water has to be in a narrow temperature range. Currently the fish hatchery receives its water from a well field that is located on the alluvial terrace downstream of Bonneville Dam. The well field lies in the proposed approach channel for the new lock and has to be abandoned during construction of the lock. For the continued water supply of the hatchery, a new well field will be developed north of the approach channel. Early in the planning phase for the new lock, concerns were raised about the potential impact of the relocation of the well field and the excavation of the new approach channel on the hatchery. To assess these concerns and to assure a continuous water supply during and after construction, a hydrogeologic investigation was initiated. Within the framework of the investigation this study focuses on the analysis of pumping test data and the development of a three-dimensional ground water flow model for the site. In the first phase of the study, data from eight pumping tests were analyzed. Hydrogeologic properties of the sedimentary units that make up the downstream terrace were determined. The focus was the pre-slide alluvium (PSA) aquifer, the water source for the existing and the future well field. In addition, the nature and location of hydrogeologic boundaries for the ground water system were determined. The results, in conjunction with information from subsurface exploration and laboratory tests, were used to develop a conceptual understanding of the ground water system at the site. The PSA aquifer receives its recharge primarily from leakage through the overlying confining layers over a large area. A direct connection between the Columbia River and the PSA aquifer could not be detected. They appear to be separated by a continuous aquitard layer or by a layer of fine-grained sediments on the river bottom. Based on these findings, in the second phase of the study, the ground water modeling program HST3D (Kipp, 1987) was used to develop a three-dimensional ground water model for the site. The model was calibrated with data from one of the pumping tests. The calibration was then verified with a second set of conditions including pumping from shallow and deep wells. Water levels in the deep PSA aquifer and the shallow unconfined aquifer were successfully matched. A satisfactory match of observed conditions was possible with only slight modifications of the hydrogeologic parameters determined by pumping test analysis and based on the conceptual model developed in the first phase of the study. It appears that a continuous aquitard layer separating the Columbia River and the PSA aquifer, with the aquifer receiving recharge through vertical leakage over a large area, is a valid representation of the aquifer system.

Fault zones are potential paths for release of radioactive nuclides from radioactive-waste repositories in granitic rock. This research considers detailed maps of en echelon fault zones at two sites in southern Sweden, as a basis for analyses of how their internal geometry can influence groundwater flow and transport of radioactive nuclides. Fracture intensity within these zones is anisotropic and correlated over scales of several meters along strike, corresponding to the length and spacing of the en echelon steps. Flow modeling indicates these properties lead to correlation of zone transmissivity over similar scales. Intensity of fractures in the damage zone adjoining en echelon segments decreases exponentially with distance. These fractures are linked to en echelon segments as a hierarchical pattern of branches. Echelon steps also show a hierarchical internal structure. These traits suggest a fractal increase in the amount of pore volume that solute can access by diffusive mass transfer, with increasing distance from en echelon segments. Consequences may include tailing of solute breakthrough curves, similar to that observed in underground tracer experiments at one of the mapping sites. The implications of echelon-zone architecture are evaluated by numerical simulation of flow and solute transport in 2-D network models, including deterministic models based directly on mapping data, and a statistical model. The simulations account for advection, diffusion-controlled mixing across streamlines within fractures and at intersections, and diffusion into both stagnant branch fractures and macroscopically unfractured matrix. The simulations show that secondary fractures contribute to retardation of solute, although their net effect is sensitive to assumptions regarding heterogeneity of transmissivity and transport aperture. Detailed results provide insight into the function of secondary fractures as an immobile domain affecting mass transfer on time scales relevant to field characterization and repository safety assessment. In practical terms, secondary fractures in these en echelon zones are not indicated to limit release of radiation to the surface environment, to a degree that is significant for improving repository safety. Thus en echelon zones are to be regarded as detrimental geologic features, with potentially complex transport behavior which should be considered in the interpretation of in-situ experiments. / Graduation date: 2005

The decision to bury high level nuclear wastes in deep geological formations led to the study of the Hanford Nuclear Reservation as one of three possible sites for the first nuclear waste repository in the United States. To adequately evaluate the environmental impact of siting nuclear waste repositories in basalt aquicludes, it is essential to know the effects on parameter identification algorithms of thermal gradients that exist in these basaltic aquicludes. Temperatures of approximately 60° C and pressures of approximately 150 atms can be expected at potential repository sites located at depths of approximately 1000m. The phenomenon of over-recovery has been observed in some pumping tests conducted at the Hanford Nuclear Reservation. This over-recovery phenomenon may possibly be due to variations in the fluid density caused by thermal gradients. To asses the potential effects of these thermal gradients on indirect parameter identification algorithms, a systematic scaling of the governing field equations is required in order to obtain dimensionless equations based on the principle of similarity. The constitutive relationships for the specific weight of the fluid and for the porosity of the aquiclude are assumed to be exponentially dependent on the pressure gradient. The dynamic pressure is converted to the piezometric head and the flow equation for the piezometric head is then scaled in radial coordinates. Order-ofmagnitude estimates are made for all variables in unsteady flow for a typical well test in a basaltic aquiclude. Retaining all nonlinear terms, the parametric dependency of the flow equation on the classical dimensionless thermal and hydraulic parameters is demonstrated. These classical parameters include the Batchelor, Fourier, Froude , Grashof, and Reynolds Numbers associated with thermal flows. The flow equation is linearized from order-of-magnitude estimates based on these classical parameters for application in the parameter identification algorithm. Two numerical solutions are presented which predict hydraulic head given a continuous set of flow parameters. The first solution uses a totally numerical finite difference scheme while the second combines an analytical solution with a numerical solution. A radial coordinate system is utilized for describing an anisotropic confined aquifer. The classical inverse parameter identification problem is solved using an indirect method. This method is based on the minimization of a objective function or error criterion consisting of three parts: 1) least-squares error of head residuals; 2) prior information of flow parameters; and 3) regularization. An adjoint equation is incorporated into the method to eliminate the need to differentiate the heads with respect to the parameters being identified, increasing the stability of the algorithm. Verification of the parameter identification algorithm utilizes both "synthetic", computed generated input data and field data from a well test for a confined aquifer within the Columbia Plateau near Stanfield, Oregon. The method used is found to give parameter estimates which are both stable and unique. / Graduation date: 1992

We quantified U removal and investigated the efficacy of uranium as a quantitative tracer of groundwater discharge in a headwater salt marsh of the Okatee River, Bluffton, SC. Determining the magnitude of U removal is important for advancing U as a tracer of paleo-oceanic conditions. Since salt marsh groundwater is typically enriched in nutrients and other biologically and chemically reactive species, quantifying groundwater discharge from marshes is critical for understanding the ability of salt marshes to modify the chemistry of important species in surface waters. We hypothesized that water-column U(VI) was removed by tidally-induced advection of surface water into permeable, anoxic salt marsh sediments, a process resulting in bacterially-mediated precipitation of insoluble U(IV)O2 and/or sorption of uranium to iron-oxides at the oxic/anoxic sediment interface. Furthermore, we suggested that hydraulic pressure gradients established by marsh-surface tidal inundation and seasonally-variable rainfall promote the discharge of salt-marsh-processed, uranium-depleted groundwater to tidal creeks, producing the surface-water U-removal signal. Groundwater and surface water data revealed non-conservative uranium behavior. We documented extensive uranium removal from shallow marsh groundwater and seasonally variable uranium removal from surface waters. These observations allowed for the calculation of seasonally-dependent salt marsh uranium removal rates. On a yearly basis, our removal rate (58 to 104 mol m-2 year-1) reemphasized the importance of anoxic coastal environments for U removal. High uranium removal, high barium concentration water observed seeping from creek banks at low tide supported our hypothesis that groundwater discharge must contribute to uranium removal documented in tidal surface waters. Average site groundwater provided an analytically reasonable endmember for explaining uranium depletion in surface water. Therefore, we used three endmember mixing models for estimating the fraction of surface water with presumed a groundwater signature. Our discharge estimates of 8 to 37 L m-2 day-1 agreed closely with previously published salt marsh values. Seasonality in discharge rates can be rationalized with appeal to seasonal patterns in observed rainfall, tidal forcing, and marsh surface bioturbation. Although more work is needed, the results of this portion of the study suggest that U may be an effective quantitative tracer of groundwater discharge from salt marshes.