Les doublets géothermiques sur aquifères superficiels jouent un rôle important en France pour le chauffage, le rafraichissement et la production d’eau chaude sanitaire des bâtiments. La pérennité de ces installations est cependant conditionnée par la possibilité de pouvoir garantir dans le temps la production puis la réinjection de l’eau souterraine. Le colmatage de la boucle géothermale est un problème majeur qui affecte de nombreuses installations sur l’ensemble du territoire français et remet en cause leur viabilité technico-économique. La compréhension et la prédiction de ce phénomène nécessite de nouveaux moyens de caractérisation et de modélisation des processus biogéochimiques couplés au fonctionnement des boucles géothermales dans des environnements de subsurface hétérogènes. Cette thèse a ainsi fait l’objet d’une nouvelle collaboration entre le laboratoire Géosciences Rennes et l’entreprise Antea group pour identifier les mécanismes à l’origine de ces phénomènes de colmatage et d’en caractériser les paramètres influents dans le but de construire un outil d’analyse des risques permettant l’anticipation de ces processus de colmatage. Le premier volet de la thèse expose la synthèse des typologies de colmatage (biogéochimique, chimique, biologique et physique). Cette synthèse est issue de l’étude de la littérature sur la problématique de colmatage des forages d’eau et des retours d’expérience concernant les doublets géothermiques recensés et étudiés durant cette thèse. Le second volet présente les inventaires régionaux des problématiques de colmatage recensées en France dans différents contextes hydrogéologiques. Ces inventaires ont permis de délivrer un état des lieux à grande échelle de la problématique de colmatage et d’étudier les contextes associés à chaque type de colmatage. Dans le troisième volet, de nouvelles méthodologies de caractérisation in-situ des phénomènes de colmatage biogéochimique liés à l’oxydation du manganèse et du fer sont présentées. Elles ont été développées dans le cadre de ces travaux de thèse à partir des investigations menées sur différents sites confrontés à des problèmes d’exploitation. Ces méthodologies d’investigations pluridisciplinaires couplent la mesure des propriétés hydrauliques, des concentrations en éléments chimiques et de la diversité microbienne afin d’identifier la problématique impactant le fonctionnement de l’installation et d’en définir in fine les causes. Le quatrième volet expose les résultats d’une campagne de terrain effectuée sur un doublet géothermique impacté par un processus de colmatage biogéochimique. Cette campagne a permis d’identifier les composantes clés de la réactivité biochimique impliquée dans le colmatage : distribution du flux, hétérogénéité chimique et diversité microbiologique. A l’aide du code de calcul PHREEQC, un modèle géochimique simulant les cinétiques de précipitation observées a été développé. Il permet l’étude quantitative des mécanismes biogéochimiques favorisant l’apparition rapide du colmatage. Enfin, les retours d’expériences de l’exploitation des doublets sur nappes superficielles ont démontré le besoin d’une méthodologie fiable d’analyse des risques, permettant d’anticiper l’apparition des processus de colmatage à chaque étape de la vie du projet. Ainsi, à partir de la synthèse de la littérature scientifique et technique et des conclusions des études menées durant cette thèse, les facteurs de risques d’apparition des phénomènes de colmatage ont été déterminés. L’analyse des risques de colmatage intégrant ces différents facteurs a été implémentée à travers l’élaboration de méthodes développées sous Python 3. L’outil ARCADE (Analyse des Risques de Colmatage et Aide à la Décision) a été conçu d’une part pour évaluer le risque et d’autre part pour informer l’utilisateur averti des bonnes pratiques et moyens préventifs. Ces bonnes pratiques d’analyse et de gestion préventive sont présentées dans le dernier volet de cette thèse. / The sustainability of geothermal systems using shallow aquifers for heating, cooling and hot water production depends on the possibility to ensure, over long time-scales, the production and the reinjection of groundwater in the aquifer. Clogging of the geothermal loop is a major issue affecting the technical and economic viabilities of numerous operations in France. The understanding and prediction of this phenomenon requires new methods of characterization and modelling of biogeochemical processes coupled to the operation of geothermal loops in heterogeneous subsurface environments. This thesis is thus the result of a new collaboration between the Géosciences Rennes lab and Antea group to identify the mechanisms at the origin of clogging phenomena and characterize their controlling parameters, in order to establish a risk assessment tool allowing the anticipation of clogging processes. The first part of the thesis describes the main clogging processes (biogeochemical, chemical, biological and physical). This synthesis is the result of (1) the study of the literature dealing with the clogging of water wells and (2) our feedbacks on the geothermal doublets identified and studied during this thesis. The second part presents the regional inventories of shallow geothermal systems impacted by clogging problems identified in different hydrogeological contexts in France. These inventories provide a large-scale perspective of clogging phenomena and allow to study the contexts associated with each type of clogging processes. In the third part, new methodologies for the in-situ characterization of biogeochemical clogging phenomena linked to manganese and iron oxidation are presented through case studies of sites affected by clogging issues. These interdisciplinary studies couple the measurement of hydraulic properties, chemical element concentrations and bacterial diversity, to identify the specific issue impacting the operation and to define its causes. The fourth part presents the results of an interdisciplinary field campaign carried out on a geothermal doublet impacted by a biogeochemical clogging process. This campaign documented the key components involved in mixing induced biogeochemical reactivity: flow distribution, chemical heterogeneity and microbiological diversity. Using PHREEQC, a geochemical model simulating observed kinetics of precipitation was developed in order to quantitatively explore the biogeochemical mechanisms favoring rapid clogging. Feedback from shallow geothermal systems operation has demonstrated the need for a reliable risk analysis methodology that allowed to anticipate the apparition of clogging processes at each stage of the project life (part 5). From the synthesis of the scientific & technical literature and the conclusions of the studies carried out during this thesis, the risk factors for the appearance of clogging phenomena were determined. A clogging risk analysis integrating these factors was implemented through the development of methods developed under Python 3. The methodology of the ARCADE tool (Analyse des Risques de Colmatage et Aide à la Décision) is designed to assess the risk and to inform users of good practices and preventive methods. These good practices for analysis and preventive methods are presented in the last part of this thesis.

From the Proceedings of the 1971 Meetings of the Arizona Section - American Water Resources Assn. and the Hydrology Section - Arizona Academy of Science - April 22-23, 1971, Tempe, Arizona / High nitrates in drinking water are significant in relation to an infant disease, methemoglobinemia, and the U.S. public health service has set a limit of 45 ppm for human consumption. This paper illustrates how chemical hydrographs were used in a study of nitrates in the groundwater of the Fresno-Clovis metropolitan area (F.C.M.A.) of semiarid central California. The area comprises about 145 square miles, with a population of 310,000. Urban water use is entirely derived from wells, whereas the surrounding agriculture relies on surface and ground water. In 1965, the California department of water resources noted nitrate concentrations in the F.C.M.A. were exceeding the safe limit. A number of sources of error in chemical analyses of water quality are noted. A measure of the accuracies of analyses and a method of double-checking anomalous results is furnished by plotting chemical hydrographs of individual wells. Seasonal changes in nitrate were consistent for many parts of the area, and were related to hydrogeologic factors and parameters directly affecting nitrification. Nitrate hydrographs were monitored by chloride hydrographs. The highest nitrate concentrations were in the shallower parts of the aquifer, and well deepening and changes in water level, pumping patterns and recharge rates complicated interpretations. However, the hydrographs helped to pinpoint the source of nitrate in areas where several possible sources were present.

From the Proceedings of the 1971 Meetings of the Arizona Section - American Water Resources Assn. and the Hydrology Section - Arizona Academy of Science - April 22-23, 1971, Tempe, Arizona / Recharge wells may be used in various problems relating to chemical water quality because of the phenomenon of in-aquifer mixing. This paper reviews specific recharge well-mixing techniques of possible utility in underground mixing operations for nitrate control. Illustrative data from field studies at a recharge site near Tucson, Arizona are presented. Both single- and 2-well types of mixing were investigated. In single-well operations, effluent recharge and pumping of the subsequent mixture occur at the same well. Differences in chlorine ion levels were used to distinguish between recharge effluent and native groundwater. Undiluted effluent was discharged in single-well operations until a pumped volume ratio of about 0.4 was attained. Dilution increased steadily with increased pumping and the relative concentration versus pumped volume curve was s-shaped. Seven-day pauses after effluent recharge resulted in immediate pumping of almost completely diluted water, probably because groundwater movement swept the effluent beyond the pumping unit during the pause. With 2-well pumping, the chlorine breakthrough curve reached a constant level at about 13 days and was close to that of the pause-type, single-well regime.

From the Proceedings of the 1972 Meetings of the Arizona Section - American Water Resources Assn. and the Hydrology Section - Arizona Academy of Science - May 5-6, 1972, Prescott, Arizona / High concentrations of nitrate have been found in water samples from irrigation wells north of the Tucson Arizona sewage treatment plant. The plant, which had primary treatment prior to 1951, produced 2,800 acre-feet of effluent in 1940, 4,600 acre-feet in 1950, 16,300 acre-feet in 1960, and 33,000 acre-feet in 1970. Large amounts of treated effluent recharge the groundwater system north of the plant. Sources of nitrate contamination beside sewage effluent may be sewage lagoons, sanitary landfills, meat packing and dairy effluent, septic tanks, and agricultural runoff. Sewage effluent is considered to be the primary source of nitrate contamination in the area. Geologic and flow net analysis indicate that aquifer conditions minimize the effects of sewage effluent contamination. Chloride and nitrate migration appears to be similar in the aquifer. Large-capacity wells were sampled to reflect regional conditions, and chemical hydrographs of chloride and nitrate were analyzed. The seasonal nature of these hydrographs patterns depend on total nitrogen in sewage effluent. Management alternatives are suggested to decrease nitrate pollution by sewage effluent.

From the Proceedings of the 1972 Meetings of the Arizona Section - American Water Resources Assn. and the Hydrology Section - Arizona Academy of Science - May 5-6, 1972, Prescott, Arizona / The distribution of transmissivity within the Tucson basin aquifer, as determined by pumping tests and reviewed in the construction of a digital model of the aquifer, was not totally random in space. Data tended to be distributed normally or log-normally for biased samples of developed wells. A frequency distribution of transmissivity derived from a calibrated digital model is more nearly representative of the real world because the aquifer sample is without bias as the sample constitutes the entire aquifer. Geohydrologic setting, electric analog, and digital models of the basin are discussed. The theory of transmissivity distribution in an arid land alluvial aquifer is developed from Horton's laws of exponential relationship between stream order and drainage network parameters. It is hypothesized that there is an exponential relationship between transmissivity of an alluvial aquifer. A statistical study was made of values derived from the digital model to test the probability density function hypothesized for transmissivity. The mean value is a function of climate and drainage area. These hypotheses require further validation.

From the Proceedings of the 1973 Meetings of the Arizona Section - American Water Resources Assn. and the Hydrology Section - Arizona Academy of Science - May 4-5, 1973, Tucson, Arizona / All groundwater in fort valley is presently found in perched aquifers. The regional water table in the area is estimated to lie at a depth of approximately 1750 feet. Groundwater reservoirs are perched on impermeable clay zones located at the base of alluvial units. Groundwater is also found in highly fractured volcanic zones overlaying impermeable clay zones. Perched aquifers also occur in interflow zones above either impermeable clays or unfractured volcanics. Groundwater in fort valley is the result of infiltration or runoff and from precipitation. This recharge water infiltrates the alluvium or fractured volcanic rocks until an impermeable zone is reached where it becomes perched groundwater. Greatest well yields come from these recharge aquifers; their reliability is largely dependent on precipitation and runoff. Most wells in the fort valley area supply adequate amounts of water for domestic use.

From the Proceedings of the 1974 Meetings of the Arizona Section - American Water Resources Assn. and the Hydrology Section - Arizona Academy of Science - April 19-20, 1974, Flagstaff, Arizona / A plan for the reuse of sewage effluent is proposed for the city of Tucson, Arizona. Several kinds of use would be possible, but utilization for irrigation of existing farmland in the Avra-Marana area seems particularly attractive for several reasons: (1) conveyance can be accomplished by gravity flow, (2) no tertiary treatment is required for the presently grown crops, (3) the nutrients in the effluent would be better used, and (4) effluent use would reduce the pumpage of high quality groundwater, conserving it for municipal or other uses. An exchange of wastewater for groundwater for use in the city system is seen as a good alternative to the present practice of the city purchasing farmland in Avra valley in order to acquire the groundwater for conveyance to the Tucson basin. Objectives to maximize the quantity and efficiency of wastewater use may not appear compatible with the profit maximization motive of the individual farmer, and suitable provisions will have to be written into wastewater sales agreements to assure coordination between user and supplier.

From the Proceedings of the 1974 Meetings of the Arizona Section - American Water Resources Assn. and the Hydrology Section - Arizona Academy of Science - April 19-20, 1974, Flagstaff, Arizona / Four sources of water supply for desert cities are rainfall, runoff, groundwater, and imported water, and the potential use for each varies. The government can institute various policy changes to eliminate or reduce the imbalance between water supply and demand. Restrictions should be placed on water-use luxuries such as swimming pools, subdivision lakes, fountains, etc. Water pricing should be progressive; each unit of increased use above a reasonable minimum should be charged for at an increasing rate. Runoff from individual properties, homes, storage, and supermarkets should be minimized through the use of onsite recharge wells, and various collection methods should be initiated. A campaign to acquaint the general public with a new water policy must be inaugurated.

From the Proceedings of the 1974 Meetings of the Arizona Section - American Water Resources Assn. and the Hydrology Section - Arizona Academy of Science - April 19-20, 1974, Flagstaff, Arizona / The inner basin is a collapse and erosional feature in San Francisco Mountain, an extinct volcano of late Cenozoic age, which lies approximately eight miles north of flagstaff, Arizona. The main aquifer's coefficient of transmissibility is approximately 14,000 gallons per day per foot and the storage coefficient was 0.08. Aquifer boundaries increased rates of drawdown of water levels in the inner basin well field. Inner basin springs which issue from perched reservoirs are not affected by pumpage of inner basin wells. Recharge is greater than the average yield from springs and wells in the basin which has an average of 8,000 acre-feet of water in storage in the principal aquifer. A large amount of water is lost from the inner basin aquifer system via leakage into underlying fractured volcanic rocks. It is believed that a part of this water could be intercepted by pumpage from a well constructed in the interior valley.

From the Proceedings of the 1974 Meetings of the Arizona Section - American Water Resources Assn. and the Hydrology Section - Arizona Academy of Science - April 19-20, 1974, Flagstaff, Arizona / The Coconino Sandstone of Permian age is the principal aquifer for the Woody Mountain well field, a source of municipal water for the City of Flagstaff. Wells of highest yield are located where the frequency of occurrence of faults is greatest and where the principal aquifer is down-faulted. The locations and displacements of all but the most prominent faults cannot be determined using conventional geologic mapping techniques because relatively undeformed Late Cenozoic basaltic lavas cover the faulted Paleozoic rock terrain. Approximately 3,500 feet of Paleozoic sedimentary rocks, which have little magnetic effect and which have a density of approximately 2.4, comprise most of the stratigraphic section in the well field. The basalt cover is strongly reversely magnetized and has a density of approximately 2.7. Changes in thickness of the basalt cover cause changes in the geomagnetic and gravitational field strength. Analysis of data from geomagnetic and gravity surveys was used to delineate boundaries and thicknesses of blocks of basalt which fill down -faulted areas. The correlation coefficient (r² = 0.96) for plots of known thicknesses of basalt versus complete Bouguer anomaly supports use of gravity data to estimate displacement of down -faulted blocks.