Search

This study was conducted in the region of Tahoua the northeastern part of the Southern Illulemeden sedimentary basin. In this region the transboundary aquifer system of Continental Intercalaire (CI) and the Continental Hamadien (CH) are potentially rich in underground water especially in free and captive aquifers. More, this water is the main source of drinking water for the populations, the animals and for agriculture and extractive industries. However, the population growth, the agricultural development, and the climate change have a real impact on the dynamics of groundwater and the exploitation of these resources. Hence, this study aims at analyzing the behavior of aquifers and assessing the impact of various natural and anthropogenic factors. Especially it matters to analyze the drilling parameters and cuttings, logging data and pumping test of nine new boreholes. The results reveal that the depths equipped with boreholes catching the CI/CH complex vary from 70 to 800 meters from East to West while the average thickness of the reservoir CH is 50 meters and it mainly consists of coarse sandstone. Drilling penetrated more than 100 meters into the CI without reaching the base. The CI tank is composed of several layers of gravel and clay whilst the CI/CH aquifers flow rates range from 0.2 to 22.8 m3/h/m. So this is cause of great variability of the water potential within this aquifer. Lastly, the transmissivity coefficients are more important in the CH than in the CI, and the water flow average rates range respectively from 29 m/year to 7 m/year.

In a semi-arid to arid climate context, dependency on groundwater resources may lead to overexploitation and deterioration of water quality. The Continental Intercalaire (CI) aquifer is one such continental-scale aquifer (more than a million of km2), which is mainly confined, poorly recharged but intensely abstracted. To date, the management of this resource relies on hydrogeological modelling and key parameters such as recharge/discharge rate and groundwater dynamics. We use a combination of residence time indicators (14C, 36Cl, 4He) and stable isotopes of water (2H and 18O) to give greater constraint on the groundwater residence time in the CI. In previous studies, 14C measurements and steady state modelling indicate a residence time of less than 100 ka whereas in others, 36Cl measurements and transient scenarios modelling suggest a longer residence time (>500 ka). In this study, most of the 14C measurements are below the limit of detection, establishing residence times greater than 40 ka and confirming the necessity of strict sampling protocols to exclude all air and AMS measurements when low 14C concentrations are expected. In the Tunisian recharge area, detectable 14C indicate sporadic recharge episodes (3–7 ka and 29–43 ka), whereas 4He and 36Cl concentrations in central areas suggest very old ( We characterise five distinct flowpaths reaching the Tunisian discharge area using their isotopic signatures. According to our mixing model, the average contribution from the main recharge area, the Algerian Atlas Mountains, is around 88%. This value is close to hydrogeological models. Conversely, the contribution from the Dahar Mountains is lower than in the hydrogeological modelling (2% against 10%) whereas the Tinhert shows a greater contribution (10% against 1%). Increase of abstraction from the CI can potentially activate the circulation of old brackish groundwaters and dramatically decrease the water quality in the whole system.

International audience; δ18O and δD of fluid inclusions in carbonates provide insights into temperatures and fluid chemical compositions prevailing during the carbonate precipitation, however, various analytical restrictions limit a wider application of this proxy. This paper presents a new fluid inclusions isotopic analytical line coupled to an online cavity ring‐down spectrometer that increased the analytical productivity up to 10 carbonate samples per working day. This efficiency allowed for the first time to assess the reliability a large set of water samples with size ranging from 0.1 to 1 µL. Good reproducibility (±0.5‰ for δ18O and ± 2‰ δD; 1σ) is obtained for water quantity superior or equal to 0.3 μL and no evidence of memory effect is found. The line is further tested using two types of natural carbonates: (1) modern speleothems samples from caves for which δ18O and δD values of drip water were measured and (2) diagenetic carbonates for which the δ18O of the parent water were independently back‐calculated from carbonate clumped isotope Δ47 measurements. Speleothem fluid inclusion values despite falling close to the Global Meteoritic Water Line are not always representative of the isotopic composition of the parent drip water. Results on diagenetic cements show that the δ18Owater values measured in fluid inclusions agree, within 1%, with the δ18Owater independently derived from Δ47 measurements. Overall, this study confirms the reliability and accuracy of the developed analytical line for carbonate fluid inclusion analyses with a good reproducibility obtained for water quantity above 0.3 μL.

Past variations of the recharge is estimated in the Atlas Mountains, the main recharge area of the Continental Intercalaire (CI) aquifer, one of the major Saharan aquifers, over the last 775 kyr. In this Mediterranean climatic context, continental archives generally record single humid events, grouped in hypothetical time interval and do not offer continuous chronicles of precipitation. We propose to use spatially distributed 36 Cl / Cl data as a temporal constraint, to infer the past recharge in the Atlas area. Based on a simplified but robust climatic scenario, assuming a piston model, we apply a Markov Chain Monte Carlo (MCMC) inversion approach to attribute a specific recharge to the last nine interglacial periods and an undifferentiated recharge to the glacial periods. The interglacial recharge values vary from a few mm yr−1 to more than 60 mm yr−1. Glacial recharge is less than 1 mm yr−1. These values are then analyzed in terms of intensity and allow questioning some initial hypothesis, especially the generally accepted value of the initial 36 Cl / Cl value (around 133 × 10 - 15 at at - 1 ). Our analysis suggests a higher value, around 175 × 10 - 15 at at - 1 . This approach allows us to bring out paleoclimatic information retained in these continental archives. Especially, computed recharges provide reliable evidence that Marine Isotope Stage (MIS) 5 was more humid than the Holocene period in agreement with marine archives that document wet condition in the North Sahara during that period.

This study presents data on chloride and bromide concentrations in porewater, water and anion diffusion coefficients, and their accessible porosities based on radial diffusion experiments on rock samples collected from a 2000 m-deep borehole (EST433) drilled by Andra in the eastern Paris Basin. The distributions of water stable isotope and chloride concentrations in porewater along this column reveal transient flows of water and solutes between the aquitard layers and the surrounding aquifers. These distributions confirm the occurrence of two separate aquitard/aquifer systems: “Oxfordian/Callovo–Oxfordian/Dogger” and “Dogger/Liassic/Rhaetian”. This separation is confirmed by Cl−/Br− ratios, which are low in Liassic and Dogger porewater, suggesting the influence of primary brines, and which are close to the marine ratio in the Dogger groundwater. Based on these results, transport simulations in the two systems were carried out according to different scenarios. The simulation results confirm that transport properties obtained in the laboratory at sample scale may be extrapolated to the formation scale. It is highly probable that diffusion is the main transport process in the Callovo–Oxfordian formation. This may also be the case in the Dogger/Liassic layer, although a limited contribution from advection cannot be totally excluded. By testing different scenarios of boundary conditions in diffusive models, it is proposed that the Dogger aquifer was first activated in the Early Miocene.

International audience; Speleothem-like deposits that develop underground in urban areas are an archive of the environmental impact of anthropic activities that has been little studied so far. In this paper, the sulfate content in shallow groundwater from northern Paris (France) is compared with the sulfur content in two 300-year-old urban carbonate deposits that grew in a historical underground aqueduct. The present-day waters of the aqueduct have very high sulfur and calcium contents, suggesting pollution from gypsum dissolution. However, geological gypsum levels are located below the water table. Sulfur content was measured by micro-X-ray fluorescence in these very S-rich carbonate deposits (0.5 to 1% of S). A twofold S increase during the second half of the 1800s was found in both samples. These dates correspond to two major periods of urbanization above the site. We discus three possible S sources: anthropic sources (industries, fertilizers…), volcanic eruptions and input within the water through gypsum brought for urbanization above the studied site (backfill with quarry waste) since the middle of the 19th century. For the younger second half of the studied section, S input from gypsum brought during urbanization was confirmed by the study of isotopic sulfur composition (δ34S = + 15.2‰ at the top). For the oldest part, several sulfur peaks could be related to early industrial activity in Paris, that caused high local air pollution, as reported in historical archives but also to historical gypsum extraction. This study provides information on the origin and timing of the very high SO42 − levels measured nowadays within the shallow groundwater, thus demonstrating the interest in using carbonate deposits in urban areas as a proxy for the history of urbanization or human activities and their impact on water bodies.

International audience; Following the explosion of reactor 4 at the Chernobyl power plant in northern Ukraine in 1986, contaminated soil and vegetation were buried in shallow trenches dug directly on-site in an Aeolian sand deposit. These trenches are sources of radionuclide (RN) pollution. The objective of the present study is to provide constraints for the Chernobyl flow and RN transport models by characterising groundwater residence time. A radiochronometer 3H/3He method (t1/2 = 12.3 a) and anthropogenic tracers including CFC and SF6 are investigated along with the water mass natural tracers Na, Cl, 18O and 2H. The groundwater is stratified, as evidenced by Na and Cl concentrations and stable isotopes (18O, 2H). In the upper aeolian layer, the Na-Cl relationship corresponds to evapotranspiration of precipitation, while in the underlying alluvial layer, an increase in Na and Cl with depth suggests both water-rock interactions and mixing processes. The 3H/3He and CFC apparent groundwater ages increase with depth, ranging from 'recent' (1-3 a) at a 2 m depth below the groundwater table to much higher apparent ages of 50-60 a at 27 m depth below the groundwater table. Discrepancies in 3H/3He and CFC apparent ages (20-25 a and 3-10 a, respectively) were observed during the 2008 campaign at an intermediate depth immediately below the aeolian/alluvial sand limit, which were attributed to the complex water transfer processes. Extremely high SF6 concentrations, well above equilibrium with the atmosphere and up to 1112 pptv, are attributed to significant contamination of the soils following the nuclear reactor explosion in 1986. The SF6 concentration vs. the apparent groundwater ages agrees with this interpretation, as the high SF6 concentrations are all more recent than 1985. The persistence of the SF6 concentration suggests that SF6 was introduced in the soil atmosphere and slowly integrated in the groundwater moving along the hydraulic gradient. The apparent age distribution in the lumped parameter models suggests an exponential or piston flow model in the upper geological section, followed by more pronounced mixing processes in the lower section.

A geochemical and isotopic (water and dissolved sulphate) study was performed on groundwater in the region of Marrakech, Morocco, with the aim of better understanding the regional hydrogeological system in order to improve water resources management. Significant differences in stable isotope contents and chemical compositions were observed between groundwater collected in the northern part of the region (Jbilets massif), where the basement schists outcrop, and that sampled in the southern part (Haouz basin), where the basement schists are overlaid by Plio-Quaternary deposits. The stable isotope composition of the groundwater showed that in the southern part the aquifer is mainly recharged from high-altitude precipitation over the High-Atlas Mountains, which may reach 600 mm per year or more, whereas, in the northern part, it is only recharged by lower-altitude local precipitation, which does not exceed 240 mm per year. Because of this limited supply, the groundwater flux in the northern compartment is much lower than in the southern compartment. This affects the water-rock interaction and the modalities of groundwater mineralization: the schist alteration is more developed in the southern compartment than in the northern one, leading to different behaviours of the conservative elements dissolved in groundwater. The observed geochemical and isotopic zonation of the studied area corresponds to a hydrogeological compartmentalisation, where the wells located in the north of the area generally produce less water than those located in the south of the area.

International audience; A vertical distribution of helium-4 was obtained in pore-water from the Toarcian/Domerian argillaceous formation at Tournemire (South Aveyron, France). Most of the studied core samples were collected from two vertical air-drilled boreholes across the impervious argillaceous formation and penetrating the bounding aquifers. Accessible porosities for helium were assumed similar to those of water and calculated by weighting before and after heating the core samples. Helium-4 concentrations were obtained after out-gassing the rock samples in tight containers. The comparison of the total amount of 4He produced since the deposition of the formation with that presently measured in the rock reveals that more than 97% of 4He has been lost from the solid phase to pore-water and afterwards to the surrounding aquifers. Helium distribution in pore water shows a complex profile because of the presence of unexpectedly 4He-rich samples in the lower part of the Upper Toarcian level. In order to understand this distribution and to assess the transport of helium at the formation scale, a series of preliminary model calculations were performed using helium diffusion coefficients three times higher than those of water isotopes. The comparison between simulated and experimental data indicates that: (1) the helium profile could be similar to those of deuterium and chloride in pore-water if some unexpectedly 4He-rich samples are not considered; (2) the estimated transport parameters (at small scale) may be transposable at the formation scale with some uncertainties; (3) a steady state has most probably been reached between the in situ helium production and its loss towards the aquifers; (4) only 17-30Ma of diffusion time would be necessary to obtain steady state; this is short compared to the formation age (180Ma) and thus the helium profile cannot give very precise information about diffusion time in the massif, but is in agreement with the hypothesis that transport processes are dominated by diffusion. © 2011 Elsevier Ltd.

International audience; The Taleghan fault (TF) is a major active fault of the Central Alborz mountain range in Iran. Located 50 km northwestwards of Tehran, this 80-km-long fault represents one of the major structures threatening 15 million people living in the capital of Iran and the surrounding cities (e.g. Karaj). The TF could be the source of some of the strongest historical earthquakes recorded in the Tehran region, notably the 958 AD event (estimated magnitude M 7.7). To characterize the kinematics and activity of the fault, we carried out a detailed morphological and palaeoseismological study combining aerial photographs, digital elevation models and fieldwork. We show that, unliked described so far, the TF is not a reverse fault but a left-lateral strike-slip fault with a normal component. Its strike, dip and rake within its eastern part are 105, 60 and -20/-40, respectively. Our palaeoseismological analysis shows that a sequence of 2-3 events with magnitudes M-w >= 7 occurred during the past 5300 years. If we consider a three-event scenario, the average recurrence interval is similar to 2000 years, and the most recent event is younger than 80 AD. If we consider a two-event scenario, the time interval between the second and the first events ranges between 3760 and 830 years, and the elapsed time since the last event ranges between 3529 and 1599 years. Combined with morphotectonics data, our palaeoseismological analysis allows estimating a minimum horizontal slip rate of 0.6-1.6 mm yr(-1) and a minimum vertical slip rate of similar to 0.5 mm yr(-1). Taking the similar to 450-m total vertical displacement observed across the fault, we conclude that the kinematical change along the TF (from reverse to left-lateral + normal) occurred similar to 1 Ma.

In order to characterize the large-scale transport properties of the Toarcian/Domerian formation in the Tournemire Underground Research Laboratory, the isotopic composition ( δ 18 O and δ 2 H) of pore water was determined on samples from a 250 m borehole drilled from the tunnel ground down to the lower aquifer. The isotope data were obtained from core samples by diffusive vapour exchange. The comparison with previous profiles obtained from vacuum distillation shows a clear discrepancy for all the stratigraphic levels. This demonstrates that artefacts, induced by distillation and already highlighted on only one stratigraphic level, occurred all over the shale sequence. Moreover, diffusion parameters were derived from both radial diffusion experiments and petrophysical determinations. To evaluate the large-scale transport properties of the formation, we performed a series of diffusive model calculations and compared them to experimental data. In agreement with the hydrogeological history, the best simulations considered an activation of the embedding aquifers 11–13 Ma ago, corresponding to the beginning of the diffusive process through the argillaceous formations.

To characterize snow isotopic signatures, monitoring of snowmelt was carried out at two sites (Oukaimden and Ifni) in the Moroccan High Atlas Mountains. For the Oukaimden site, samples of snow were taken by two methods to compare sampling techniques: (1) coring with a polyvinyl chloride (PVC) tube and (2) passive capillary sampling (PCS) installed at the snow/soil interface. The analyses show variable isotope contents, ranging from -14.7 to -2.5 parts per thousand for oxygen-18 and from -116 to -28.2 parts per thousand for deuterium. The most depleted values are observed in March 2013 at high elevation (3229 m asl). The majority of snow core samples display fractionation by sublimation, whereas those collected by the PCS sampling method are close to the Global Meteoric Water Line. The isotopic signature is comparable for snow, surface water and groundwater samples, indicating that snowmelt plays an important role in recharging aquifers, lakes, and rivers on the southern and northern sides of the Atlas Mountains. Recharge by snowmelt allows the dilution of salinity in adjacent aquifers. Characterization of the stable isotopic composition of snow obtained from snow cores is limited in comparison with the PCS method, which provides realistic compositions of the melt water contribution to water resources in this semi-arid area.

Summary The “Continental Intercalaire” from Sahara is one of the largest confined aquifers in the world. In the northern part of the basin, the depth of the aquifer ranges between 400 and 1800 m and the main flow direction is from west to east. Chlorine-36 analyses were performed on groundwater samples mainly collected along this flowpath. They show a wide range of 36 Cl contents, from 8 to 99 × 10 −15 at at −1 expressed as 36 Cl/Cl atomic ratio. In order to evaluate the epigene production, measurements were also performed on chloride extracted by leaching from a soil profile, near the recharge zone. This showed that the contribution of epigene production remains limited. A range of 36 Cl/Cl ratios from 116 to 133 × 10 −15 was adopted for groundwater recharging the aquifer, including both meteoric production and epigene production. For most of the samples collected along the main flowpath, a significant decay effect was observed. An attempt of dating was made by using an equation that takes into account radioactive decay of the meteoric–epigene input, deep production and chloride dissolution within the aquifer. The calculated 36 Cl residence time varies from 16 to 500 ka for the minimum ages, and from 25 to 1200 ka for the maximum ages. An increase of residence time is observed up to about 500 km from the recharge zone. This corresponds to mean flow velocities of 0.2–0.5 m a −1 and to an average permeability of 1.1 ± 0.6 × 10 −5 m s −1 , at the scale of several hundreds of kilometres.

Argillites are one of the geological formations studied by IRSN for their confining properties for isolation of radioactive wastes. One of the main objectives is the study of water transport through rocks with very low water content and very low hydraulic conductivity by modeling of natural tracer profiles. This paper presents the protocol developed for and applied to the acquisition of data for chloride content in interstitial water of the Toarcian argillites at the Tournemire site (Southern France). This protocol is based on laboratory diffusion experiments and on modeling. Experimental data obtained during the transient and steady-state parts of diffusion experiments enable, respectively, the assessment of the diffusion coefficient and the determination of Cl concentration in pore water. Using this protocol, profiles with depth for both of these data sets have been acquired along the geological sequence. Taking into account the present knowledge of the geological and hydrogeological history of the Tournemire massif, a conceptual model granting the main role for mass transport to diffusion has been proposed. According to this conceptual model, a one-dimensional numerical model was built for simulating the mass transport of chloride through the sedimentary column, over 53 Ma. The good agreement between experimental data and calculated values for both diffusion coefficients and concentrations of chloride confirms that diffusion is likely the main process for mass transport in the massif. This model was also tested with the deuterium content of interstitial water, applying variable concentrations at the aquifer system boundaries for reflecting the thermal dependency of isotopic composition in precipitation. These simulations also reveal the likely important role of heterogeneities, such as fractures, in the variability of tracer concentrations with regards to a simple diffusion profile. © 2004 Elsevier B.V. All rights reserved.

With the assistance of the ASTER team (M. Arnold, G. Aumaître, D. Bourlès, and K. Keddadouche); International audience; The Bulnay earthquake of 23 July 1905 (Mw 8.3–8.5), in north‐central Mongolia, is one of the world’s largest recorded intracontinental earthquakes and one of four great earthquakes that occurred in the region during the twentieth century. The 375 km long surface rupture of the left‐lateral, strike‐slip, N095°E‐trending Bulnay fault associated with this earthquake is remarkable for its pronounced expression across the landscape and for the size of features produced by previous earthquakes. Our field observations suggest that in many areas the width and geometry of the rupture zone is the result of repeated earthquakes; however, in those areas where it is possible to determine that the geomorphic features are the result of the 1905 surface rupture alone, the size of the features produced by this single earthquake are singular in comparison to most other historical strike‐slip surface ruptures worldwide. Along the 80 km stretch, between 97.18° E and 98.33° E, the fault zone is characterized by several meters width and the mean left‐lateral 1905 offset is 8.9±0.6 m with two measured cumulative offsets that are twice the 1905 slip. These observations suggest that the displacement produced during the penultimate event was similar to the 1905 slip. Morphotectonic analyses carried out at three sites along the eastern part of the Bulnay fault allow us to estimate a mean horizontal slip rate of 3.1±1.7 mm/yr over the Late Pleistocene–Holocene period. In parallel, paleoseismological investigations show evidence for two earthquakes prior to the 1905 event, with recurrence intervals of ∼2700–4000 yrs.

The study of the isotopic composition ( 18 O and 2 H) of groundwater collected in the Sfax basin (Tunisia), helped to understand the behaviour of the different aquifers. It showed that the groundwater in the deep aquifer is old, probably slow moving and recharged under a colder climate than at present. The increasing exploitation of the shallow aquifers probably favoured upward leakage from the deep aquifer. Isotope balance equations allowed us to estimate the contribution of the deep aquifer to the shallow aquifer recharge.

International audience; In the Paris Basin, in France, the Callovo-Oxfordian (COx) is currently studied over a 250 km2 surface area by the French national radioactive waste management agency in order to assess the feasibility of long-term underground nuclear waste repository. The COx is a 140 m thick clay-rich layer, which is part of the 2000 m aquitard/aquifer system constituting the sedimentary cover. In such sedimentary context, the transport processes of potential contaminants can be represented by both vertical diffusion and horizontal advection through the most permeable layers. Chloride is used as a natural conservative tracer, and is monitored in term of concentrations and isotopic composition (δ37Cl) for both pore and groundwater. During this study, the samples were collected from three boreholes located in the center of the studied zone, one of them (EST433) going down to 2000 m depth. The main solute transport process is shown to be vertical diffusion from the massive Keuper halite level to the rest of the sedimentary pile. This global diffusive system can be occasionally disturbed by horizontal circulation of groundwater occurring in the Oxfordian and Dogger limestone formations. Therefore, these circulations cut the global diffusive system in a succession of independent diffusive systems. In this study the data set was implemented in a simplified 2D solute transport model and scenarii reproducing known history in term of paleo circulations inside the system, were applied and allowed to obtained a good fit of the data. Model results showed that paleo circulations, occurring between −145 Ma and −110 Ma, still have an impact on current distribution of chloride in the system, especially for δ37Cl. The model highlights the need of the presence of a circulation spatially limited at the base of the Liassic formation to fit the data. The best fit obtained indicated current residence time of 500 ka in the Dogger and Oxfordian, with respective onset of the circulations at −20 Ma and −5 Ma.

The uranium content and 234 U/ 238 U ratios have been measured in more than thirty groundwater samples over the Tunisian part of the Continental Intercalaire (CI) aquifer. In this large artesian aquifer, uranium behavior is mainly controlled by redox front systems. The uranium concentrations vary by 5 orders of magnitude from 10-4 to 20 ppb and the ( 234 U/ 238 U) activity ratio from 1 to 10. Close to CI outcrops in the Dahar Mountains, the stepwise decrease in U concentrations combined with a significant increase of the ( 234 U/ 238 U) ratio up to 10 (“augmenting regime”) is followed, downflow by a ( 234 U/ 238 U) decrease up to 3 (“decaying regime”), clearly indicating the occurrence of a redox front. We will discuss the significance of 234 U- 238 U disequilibria in terms of water residence times and its potential to trace water-rock interaction mechanisms.

International audience; This study provides evidence on the sources and pathways of trace-metal contamination using isotopic tracers, δ34S of dissolved sulfate and 206Pb/207Pb of suspended particulate matter (SPM), in an urban French river. The trace-metal contamination of the Orge River watershed was correlated with a highly contrasted land-use pattern with mainly forests and agricultural lands upstream and a highly dense urban area (up to 8000 inhabitants/km2) downstream in the suburbs of the Parisian megacity. The increase in the sulfate concentrations (0.25-1 mmol/l) of the dissolved compartment (< 0.45 μm) monitored during a hydrological year (2010/2011) correlated with the dissolved sulfate δ34S and δ18O ratios (+ 2 to + 12 and + 6 to + 13‰, respectively). This result indicated that the runoff and sewage waters are important sources of water in the river, particularly for the highly urbanized sites. The lead enrichment factor in the SPM was observed to increase according to the urbanization gradient (1 to 6), and the associated 206Pb/207Pb ratio indicated that the signature of the lead-bearing particles transformed from a three-endmember system (2001 samples) to a two-endmember one (2010/2011 samples) over the last decade. These results suggest that gasoline-originated lead has disappeared from particulate fluxes since the lead additive prohibition in 2000. The lead-bearing species in the Orge River were also observed to originate mainly from urban activities, and the reactivity of selected urban lead-bearing species is discussed based on microscopic observations. Finally, coupling the δ34S and 206Pb/207Pb ratios provided a powerful indicator of urban influence as a function of both river hydrodynamics and physico-chemical conditions.

International audience; Water stable isotopes and chloride profiles in pore-water through more than 800 m of sediments were obtained from a 2000 m-deep borehole (EST 433) drilled by Andra in the eastern Paris Basin. Vapour exchange method and aqueous leaching were used to obtain the stable isotope and chloride concentrations of pore-water from 24 rock samples. Petrophysical measurements included water contents, grain densities and porosities of the studied formations. Pore-water and some groundwater samples collected during the drilling are mainly of meteoric origin: they plot near the Global Meteoric Water Line, distributed between heavy-isotope depleted Oxfordian groundwater and enriched Triassic groundwater, in good agreement with previous data. The δ2H and δ18O values describe curved profiles in the Callovo-Oxfordian formation, and show an increase with depth below this formation (Dogger and Liassic). Similar trends were observed for the chloride concentrations, except in the Liassic formation where they are more or less constant. The low chloride concentrations in the basal Jurassic layers indicate that the source of salinity to the Dogger aquifer is likely the middle Liassic formation and not the Triassic salt as previously suggested. A preliminary modelling exercise showed that currently available diffusion parameters (diffusion coefficients and accessible porosities) might be used to properly simulate these exchanges for deuterium. This is not the case for chloride, perhaps because the used values for anion accessible porosity were not relevant and/or the applied modelling conditions were unsuitable.

Summary Groundwater in many arid basins, particularly in developing countries, is the only available water resource that sustains local communities. Yet, information on the basic hydrological parameters and the sustainability of the groundwater exploitation are often lacking. This study investigates the origin of groundwater from the Lower Cambrian carbonate aquifer of the Lakhssas Plateau in the Anti-Atlas Mountains of southwestern Morocco. The study aims to reveal the origin of the groundwater, salinity sources, and the residence time of the water. The study is based on a comprehensive geochemical and isotopic (oxygen, hydrogen, carbon, and strontium) investigation of groundwater from different parts of the basin. The hydrochemical and isotopes results indicated three types of groundwater in the Lakhssas Plateau: (1) thermal water in the southern part of the basin with solute composition that reflects dissolution of calcium–sulfate and calcium carbonate minerals; (2) low-temperature groundwater at the southern margin of the basin with low salinity (chloride content up to 100 mg/L) and chemical composition that is expected from equilibrium with limestone–dolomite rocks; and (3) low-temperature groundwater in the northern, western, and eastern margins of the basin with a wide range of salinity (chloride up to 800 mg/L). The different water types had also different stable isotope composition; the thermal water was depleted in 18 O and 2 H (δ 18 O as low as −7.6‰) relative to the southern (−5.9 to −5.3‰) and northern waters (−5.7 to −3.8‰). The differences in δ 18 O and δ 2 H between the southern and northern waters are related to elevation that induced fractionation of oxygen and hydrogen isotopes in recharge water originated from coastal moisture. The data suggest that the high salinity in groundwater from the northern, western and eastern margins of the Lakhssas Plateau is related to the presence of schist rocks in these areas. The distinctive low Na/Cl and Br/Cl ratios, coupled with high silica contents and high 87 Sr/ 86 Sr ratios (up to 0.713) in the saline groundwater provide additional evidences for the link between salinity and the schist rocks. In contrast, the thermal water had relatively low 87 Sr/ 86 Sr ratio (0.7089), which is identical to the secular seawater Sr isotope ratio during the Early Cambrian period and presumably reflects interaction with the Early Cambrian carbonate and sulfate aquifer rocks. In the northern and southern groundwater, the 87 Sr/ 86 Sr ratios were higher and correlated with the Mg/Ca ratios, inferring mixing between the Early Cambrian limestone and other rocks with higher 87 Sr/ 86 Sr, such as the schist rocks. The radiocarbon data showed 14 C activities ranging from 6 pmC in the thermal water to 62 pmC in the southern water. Age-modeling, corrected for dissolution of the carbonate rocks with dead carbon, simulated mean residence time of 10–15 ka BP for the thermal water and a range of 0 to 3 ka BP for the northern and southern waters, depending on the used models. The integration of the data enables us to establish a conceptual model for the origin of groundwater in the Lakhssas Plateau: (1) recharge to the aquifer from relatively heavy-isotope depleted recharge water, presumably during wetter conditions about 10–15 ka BP. The recharge water interacted at high depth with limestone and calcium sulfate minerals and emerged to the surface as thermal water at the southern part of the basin; (2) more recent recharge from coastal moisture originated from the Atlantic Ocean. The stable-isotope composition of groundwater was controlled by the elevation of their recharge areas: recharge at higher elevation, particularly in the southern margin resulted in lower δ 18 O and δ 2 H values; (3) the recharge water interacted with the carbonate aquifer rocks, particularly with calcite and dolomite minerals. In areas of exposure of schist rocks, the water–rock interaction induced salinization of the groundwater. Overall, our data reveal that the limited water resources in this semi-arid zone of Morocco could be in some parts less renewable and also saline. Future exploitation of this basin will have to account the salinity factor and the suggested contribution of water recharged some thousands years ago.

International audience; This study concerns the Souss upstream basin. The objective is to investigate the characteristics of surface water and groundwater, to assess the impact of artificial recharge as reinforcement of the natural replenishment and assess the renewal of groundwater under semi-arid area. Two major water types are observed: (i) surface waters and upstream springs (least mineralized) and (ii) all groundwater samples (prevailing calcium and magnesium bicarbonate water type). Water isotopes show a low evaporation of precipitations during infiltration. Impoverishment in heavy isotopes is the characteristic of mountain rainfalls, or of a climate colder and wetter than present. Carbon-14 activities (34-94 pmc) indicate a long residence time. The artificial recharge is low compared to the reservoir volume, due to which the renewal rate is also low.

In order to assess the feasibility for long lived radioactive wastes storage facilities in deep geological formation, solute transport processes must be investigated in the vicinity of the host formation. In France, the Oxfordian and Dogger limestone layers surrounding the Callovo-Oxfordian (CO x ) argillite in the east of the Paris Basin are investigated for this purpose. More than 60 samples of Oxfordian and Dogger formation groundwater and one sample of Triassic formation groundwater, located at the bottom section of the investigated sedimentary cover, were collected over a 250 km 2 area, and were analysed for major ions, δ 18 O and δ 2 H of water, 87 Sr/ 86 Sr, and δ 34 S and δ 18 O of dissolved sulphate. Oxfordian and Dogger formation water is from meteoric origin, and no direct water flow between Oxfordian, Dogger, nor Triassic formations was evidenced. Mineralization processes of the Dogger limestone groundwater were fully investigated. These processes correspond to a series of geochemical reactions including: calcite dissolution, incongruent dissolution of dolomite, ion addition from upward vertical diffusion through the sedimentary pile inducing cation exchange and further dissolution due to the increased ionic strength of the solutions. A PHREEQC model was developed to simulate inferred chemical processes. The model output is consistent with observed data.

In arid environments, water rises from the saturated level of a shallow aquifer to the drying soil surface where evaporation occurs. This process plays important roles in terms of plant survival, salt balance and aquifer budget. A new field quantification method of this capillary rise flow is proposed using micro-injections (6 μL) of a deuterium-enriched solution (δ value of 63 000‰ vs. V-SMOW) into unsaturated soil at a 1 m depth. Evaluation of peak displacement from profile sampling 35 days later delivered an estimate that was compared with outputs of numerical simulation based on laboratory hydrodynamic measurements assuming a steady state regime. A rate of 3.7 cm y−1 was estimated at a Moroccan site, where the aquifer water depth was 2.44 m. This value was higher than that computed from the relationship between evaporation rates and water level depth based on natural isotopic profile estimates, but it was lower than every estimate established using integration of the van Genuchten closed-form functions for soil hydraulic conductivity and retention curve.

International audience; Groundwaters from the Tithonian/Kimmeridgian, Oxfordian and Upper Dogger aquifers, within the eastern part of the Paris basin (France), were characterised using 3H, 14C and 36Cl, and noble gases tracers, to evaluate their residence times and determine their recharge period. This information is an important prerequisite to evaluating the confinement properties of the Callovo-Oxfordian clay formation sandwiched between the Oxfordian aquifer and the Dogger aquifer, currently being investigated by the French nuclear waste management agency (Andra) for radioactive waste disposal. Data presented in this paper are used to test 4 hypotheses : (1) The Oxfordian limestone is isolated from the overlying Tithonian/Kimmeridgian surface aquifer. This first hypothesis is supported by the presence of measurable 3H activities in groundwaters from the surface aquifer (6–11 TU), and by its absence in groundwaters derived from deeper aquifers (

Summary Groundwater resources in the Tadla basin stem from surface water recharge and different groundwater inflows, forming a multilayered aquifer system, which hosts one of Morocco’s most important groundwater reservoirs. The hydrodynamic infrastructure; i.e. the relationship between all regional aquifers, recharge, and the residence time of waters poses a serious challenge for current water management and future exploitation in aiming for a long-term sustainable utilization. A combined hydrogeologic and isotopic investigation using hydrochemical and isotopic tracers such as 18 O, 2 H, 3 H, 13 C and 14 C was carried out in order to determine the sources of water recharge to the aquifers, the groundwater flow system, and the residence time of these waters. More than one hundred point measurements throughout the study area in varying wells, boreholes, springs and rivers were investigated. Chemical compositions of the groundwater indicate an important influence by the host carbonate rocks from each of the Tadla aquifers. Stable isotope results indicate the existence of two groups of groundwater corresponding to the unconfined aquifer in the north and the confined aquifer in the south. The δ 18 O, δ 2 H, 3 H, and 14 C data indicate that the High Atlas Mountains in the south and east of the basin, which are characterized by high rainfall and low δ 18 O and δ 2 H values, are currently the major source of recharge for the Tadla aquifers. A significant recharge zone lies in the northern part of the basin where all the aquifers outcrop. The confined zones show depleted 18 O values, corresponding to the signature of recharge water from the Atlas Mountains. Moreover, all isotope data demonstrated clearly that the Tassaout springs, which are located in the southwest of the basin and were previously interpreted as representing natural outlet of the deep aquifers, are comprised of young waters with depleted 18 O and 2 H signatures, suggesting a high altitude recharge from the Atlas Mountain. In contrast, the unconfined parts of the aquifers show higher values of δ 18 O, indicating an evaporation phenomenon, which occurs during infiltration or recharge from irrigation. The mixing process of old and recent waters is confirmed by 14 C and 3 H. The isotopic data also indicates probable interaction and flow between the different aquifers. The isotopic and hydrochemical data is therefore essential in confirming possible mixing relationships between aquifers whereas traditional hydraulic data is not capable to provide a quantitative assessment of these relationships. The data generated in this study will certainly encourage the revision and improvement of the current hydrological water resources model for the Tadla basin. The results provide a framework for development of a comprehensive management plan in which water exploitation shifts towards areas with modern recharge and where young and high quality groundwater is found.

The Aegean volcanic arc is the result of a lithosphere subduction process during the Quaternary time. Starting from the Soussaki area, from west to east, the arc proceeds through the islands of Egina, Methana, Milos, Santorini, the Columbus Bank, Kos and Nisyros. Volcano-tectonic activities are still pronounced at Santorini and Nisyros in form of seismic activity, craters of hydrothermal explosions, hot fumaroles and thermal springs. A significant number of cold water springs emerge in the vicinity of hot waters on these islands. Chemical and isotopic analyses were applied on water and fumaroles samples collected in different areas of the volcanic arc in order to attempt the assessment of these fluids. Stable isotopes of water and carbon have been used to evaluate the origin of cold and thermal water and CO 2 . Chemical solute concentrations and isotopic contents of waters show that the fluids emerging in Egina, Soussaki, Methana and Kos areas represent geothermal systems in their waning stage, while the fluids from Milos, Santorini and Nisyros proceed from active geothermal systems. The δ 2 H–δ 18 O–Cl − relationships suggest that the parent hydrothermal liquids of Nisyros and Milos are produced through mixing of seawater and Arc-Type Magmatic Water (ATMW), with negligible to nil contribution of local ground waters and with very high participation of the magmatic component, which is close to 70% in both sites. A very high magmatic contribution to the deep geothermal system could occur at Santorini as well, perhaps with a percentage similar to Nisyros and Milos, but it cannot be calculated because of steam condensation heavily affecting the fumarolic fluids of Nea Kameni before the surface discharge. The parent hydrothermal liquid at Methana originates through mixing of local groundwaters, seawater and ATMW, with a magmatic participation close to 19%. All in all, the contribution of ATMW is higher in the central–eastern part of the Aegean volcanic arc than in the western sector. This difference, which is spotted in the variable isotopic composition of the sampled fluids from west to east along the arc, is probably due to several causes, including the tectonic regime, the depth of the deep reservoir below sea level, the age of volcanic activity and in general the geomorphologic state of each island.

Chemical (Na, K, Mg, Ca, Cl, SO4, HCO3, Si, Al, Fe, Mn, TOC) and isotopic (18O[H2O], 2H[H2O], 3H[H2O], 14C[DIC], 13C[DIC], 34S[SO4], 18O[SO4]) data were obtained on groundwater collected from fractures inside the consolidated argillaceous formation of Tournemire (Aveyron) and from the aquifers surrounding this formation. Because of the very low transmissivity of such fractures (10−10 m2 s−1), specific devices were developed, for limiting out-gassing and air-contamination of groundwater during the sampling. Two modelling approaches are proposed to account for the chemical evolution of fracture groundwater from Na–Cl–HCO3 to Na–SO4 type. The first one is based on mineral equilibria only and the second one also takes into account the ion-exchange reactions. The two modelling approaches are able to reproduce K+ concentrations and they led to an almost constant pCO2 (10−2.6 atm), which only depends on equilibrium with the calcite–dolomite–kaolinite–quartz assemblage. The modelling of Na+, Ca2+ and HCO 3 + behaviours is improved by taking the ion-exchange reactions into account. In some boreholes, the time evolution of the chemical composition of fluids was greatly influenced by SO4 reduction combined with oxidation of dissolved organic C. The isotopic composition of aqueous SO4 in fracture groundwater confirms that SO4 reduction processes have occurred since the drilling of these boreholes, but also previously, during the natural evolution of groundwater. These redox reactions constitute a source of aqueous inorganic C that has to be considered in the computation of groundwater “ages” with 14C. For the studied fracture groundwaters, the maximum age estimates range from 17 to 29 ka. This shows that fractures may induce relatively fast groundwater flow, compared to the very long migration times that are calculated by only considering diffusive transport in the rock microporosity.

Summary Groundwater and surface water in Souss–Massa basin in the west-southern part of Morocco is characterized by a large variation in salinity, up to levels of 37 g L −1 . The high salinity coupled with groundwater level decline pose serious problems for current irrigation and domestic water supplies as well as future exploitation. A combined hydrogeologic and isotopic investigation using several chemical and isotopic tracers such as Br/Cl, δ 18 O, δ 2 H, 3 H, 87 Sr/ 86 Sr, δ 11 B, and 14 C was carried out in order to determine the sources of water recharge to the aquifer, the origin of salinity, and the residence time of water. Stable isotope, 3 H and 14 C data indicate that the high Atlas mountains in the northern margin of the Souss–Massa basin with high rainfall and low δ 18 O and δ 2 H values (−6 to −8‰ and −36 to −50‰) is currently constitute the major source of recharge to the Souss–Massa shallow aquifer, particularly along the eastern part of the basin. Localized stable isotope enrichments offset meteoric isotopic signature and are associated with high nitrate concentrations, which infer water recycling via water agricultural return flows. The 3 H and 14 C data suggest that the residence time of water in the western part of the basin is in the order of several thousands of years; hence old water is mined, particularly in the coastal areas. The multiple isotope analyses and chemical tracing of groundwater from the basin reveal that seawater intrusion is just one of multiple salinity sources that affect the quality of groundwater in the Souss–Massa aquifer. We differentiate between modern seawater intrusion, salinization by remnants of seawater entrapped in the middle Souss plains, recharge of nitrate-rich agricultural return flow, and dissolution of evaporate rocks (gypsum and halite minerals) along the outcrops of the high Atlas mountains. The data generated in this study provide the framework for a comprehensive management plan in which water exploitation should shift toward the eastern part of the basin where current recharge occurs with young and high quality groundwater. In contrast, we argued that the heavily exploited aquifer along the coastal areas is more vulnerable given the relatively longer residence time of the water and salinization processes in this part of the aquifer.

International audience; In Northeast Thailand lowlands with shallow saline watertable, rainfed paddy fields often present high salt concentration in the dry season, forming patches or spots of salt crusts on the soil surface. In this context, the mechanisms implied in salt concentration during dry season were studied by establishing salt budget with evaporation and drainage estimates inside and outside a saline patch. Drainage was estimated by Hydrus-1D modelling constrained by an hydrodynamic characterization and the profile of water contents at the end of dry season. Evaporation rates at the end of the dry season were computed by interpreting natural detailed profiles of deuterium (D) and chloride (Cl) contents. Because of the drastic diminution of hydraulic conductivity at saturation with depth and the decrease of groundwater level at the end of the cropping season, simulated hydrological balance with Hydrus-1D pointed out zero cumulated fluxes for depths of 39.5 cm (outside the saline patch) and 37.5 cm (inside the saline patch). Therefore, ail the chloride accumulated in the very upper layers during dry season comes from the chloride that was present in the 0-39.5 cm layers before the beginning of the drying. Inside the saline patch, the tentative Cl budget is coherent with the hypothesis of saturation of the profile by aquifer saline water during the flooding. Evaporation rates computed from the diffusion of chloride and deuterium at the end of the drying season, when the aquifer level was 1.4 m deep, range between 0.121 and 0.378 mm d(-1). This does not sustain the assumption of a considerable salinity contribution from the aquifer during the dry season. Moreover, evaporation estimates based on Cl and D diffusion equilibrium showed depleted rates (38-63%) inside the saline patch due to salt accumulation in the first 12 cm of the soil. In the vapour transfer Layer, estimated evaporation rate based on the vapour movement of D was in the same order of magnitude than computed rate assuming liquid Cl diffusion. This coincidence is attributed to the liquid fluxes that occurred during the expansion of the vapour transfer Layer during the progression of the evaporation front.

[1] On the basis of chloride concentrations of pore water in the Tournemire massif (part 1), a conceptual model for mass transport in argillites by diffusion is proposed. From this conceptual model and current knowledge of the geological history of the massif, one-dimensional numerical simulations are formulated for chloride transport in Tournemire massif over the past 53 Ma. Good agreement between experimental data and calculated values for both diffusion coefficients and concentrations of chloride confirms that diffusion is the main process for mass transport in the massif. This model is also tested using deuterium contents of pore water, applying variable concentrations to meteoric water (circulating in system boundary layers) based on the thermal dependency of its isotopic composition. These simulations reveal the likely important role of lithologic heterogeneities, such as fractures, in the horizontal distribution of tracer concentrations.

Some granitic deep groundwaters have high chloride contents of which the origin is still unclear. Among the different possibilities, leakage of fluid inclusions trapped in primary minerals was proposed. This hypothesis is examined by studying four granitic sites with various hydrogeological features: drill-cores from Aspo (Sweden); Boettstein and Leuggern (northern Switzerland); outcropping granites and drill-core samples from Cauterets and Luchon in the Pyrenees (France); In Tounine granitic massif (Central Ahaggar, Algeria). With the exception of In Tounine, a part of the granite samples selected in each study site displays salt-rich fluid inclusions with salinities ranging from 22 to 26 eq. wt.% NaCl, calcium contents and minimal trapping temperatures typical of deep sedimentary brines. Most salt-rich fluid inclusions occur in secondary fluid inclusion planes in magmatic quartz, but were also found within fissure-in-filling calcite at Boettstein and Aspo. The wide ranges of Cl/Br ratios in single same rock samples, due to heterogeneities in fluid inclusion distribution in rock-forming minerals, and the variability of Cl/Br ratios in fluid reservoirs render the use of Cl/Br ratios for fluid-source tracing inaccurate. Only salt-rich inclusions in the Pyrenees samples show a relatively narrow range of Cl/Br ratios, similar to those of associated groundwaters. Mass-balance calculations based on salinity estimates from microthermometry, crush-leach analyses and inclusion counting, coupled with the study of possible mechanisms for leakage, demonstrate that the relatively high salinity of deep groundwaters (Boettstein and Aspo) cannot only be accounted for by the leakage of fluid inclusions trapped in quartz. It cannot be totally excluded that fluid inclusions may be responsible for the major part of chloride in groundwaters with relatively low salinity (Leuggern 1647 m, Cauterets, Luchon, Ahaggar). The leakage of secondary fluid inclusions trapped in primary minerals cannot be considered as the main process responsible for the high salinity of present-day deep groundwaters in the studied sites. However, the presence of secondary salt-rich inclusions indicates that sedimentary brines have already percolated the granites, a datum which is important for the evaluation of the hydrogeological history and for the understanding of the modalities of chlorine introduction in the crystalline basement.