Applied Geochemistry (v.40, #C)

Selenium (Se) is an important co-existing elemental component of the mineral matrix of mercury (Hg) ore deposits. The hazards associated with Se contamination of the aquatic ecosystems in Hg mining areas; however, are often overlooked by environmental researchers due to a preoccupation with Hg. Selenium may also pose a long-term risk to the local ecosystem, and further complicate the situation as Se may also play an important antagonistic role against Hg. Furthermore, most studies on Se pollution have focused only on total Se, whereas the toxicity, bioavailability, and bioaccumulation of Se in aquatic ecosystems is primarily determined by its site-specific individual species. In this study, the concentrations of total Se, inorganic Se (tetravalent and hexavalent), and organic Se were determined in water samples collected from 41 typical sites selected in rivers, tributaries, and springs in Wanshan, China, where Hg and Se co-occur due to historic Hg mining and retorting activities. Se concentrations were observed to decrease with distance from mine-waste calcines, which indicated that mine-waste calcines may be significant sources of the elevated Se in the rivers, especially in downstream areas within 8 km from the mine-waste calcines. The concentration of total aqueous Se throughout the study area was highly variable (3.8 ± 6.0 μg L−1) and on average was one order of magnitude greater than that in natural river systems worldwide (0.1–0.3 μg L−1). The majority of the Se was hexavalent (3.1 ± 4.9 μg L−1; 65%), followed by tetravalent (0.53 ± 0.86 μg L−1; 15%) and organic forms (0.85 ± 1.5 μg L−1; 20%), possibly due to the generally alkaline conditions. Se concentrations in some sampling sites exceeded certain recommended limit of values. However, the existing criteria for Se in aquatic system are mainly based on total Se and the recommended limit of values in different countries or organizations are inconsistent with one another. Therefore, the need to consider Se speciation rather than only total Se is highlighted for future studies.

With an aim to increase the understanding about the isotopic and chemical heterogeneity of calcites in water-conducting fracture zones with different crystalline wall rock compositions at different depths, we present trace element chemistry, isotopic composition (δ18O, δ13C, 87Sr/86Sr) and biomarkers of euhedral low-temperature fracture-coating calcite. Paleohydrogeological fluctuations and wall rock influence on the hydrochemistry in the deep groundwater are explored. Samples are from several fracture zone sub-fractures (at −360 to −740 m), retrieved during an extensive core drilling campaign in Sweden.Calcite generally showed fracture zone specific values of δ13C, δ18O and 87Sr/86Sr, which indicates precipitation from relatively homogeneous fluid (similar to the modern groundwater at the site) at the same event in each fracture zone. δ18O and δ13C in the different fracture zones were consistent with precipitation from waters of different salinity and decreasing organic input with depth, respectively. The latter is also supported by biomarkers showing clear indications of SRB-related organic compounds (e.g. iso- and anteiso-C17:0-branched fatty acids), except in the deepest zone. In contrast to the isotopes, variation in trace elements within the fracture zones was generally up to several orders of magnitude. Manganese and REE, as oppose to the other metals, were higher in the shallow fracture zones (112–1130 and 44–97 ppm, respectively) than in the deeper (28–272 and 5–11 ppm, respectively), in agreement with the groundwater composition. Although the rock types varied between and within the different fracture zones, this had insignificant influence on the trace element chemistry of the calcites. Co-variation was generally relatively large for many trace elements, with isometric logratio correlation generally better than 0.75, which indicates that their variation in the calcites is due to variation of Ca in the fracture water, but other local factors, especially uptake in co-precipitating minerals (clay minerals, barite, pyrite and zeolites), but also microbial activity and metal speciation may have influenced the metal incorporation into calcite. These detailed studies of fracture calcite are of importance for the understanding of variation in fluid chemistry and trace metal uptake in fracture zones, adding together with hydrochemical studies detailed information optimal for site characterisation.

Competitive effect of the metallic canister and clay barrier on the sorption of Eu3+ under subcritical conditions by Said El Mrabet; Miguel A. Castro; Santiago Hurtado; M. Mar Orta; M. Carolina Pazos; María Villa-Alfageme; María D. Alba (25-31).
An in depth knowledge and understanding of high activity radionuclide (HLRW) immobilization processes on the materials composing the engineered barrier (clay and metallic canister) is required to ensure the safety and the long-term performance of radioactive waste disposal procedures. Therefore, the aim of this study was to understand the mechanisms involved in the retention of Eu3+ by two components of the multibarrier system, the bentonite barrier and the canister. As such, a comparative study of the interaction of trivalent Eu3+, used to simulate trivalent actinides, with both bentonite and a metallic canister has been undertaken in this work. To this end, we designed a minireactor into which the bentonite was introduced and compacted. The minireactor-bentonite system was then submitted to a hydrothermal reaction with a 7.9 × 10−2  M solution of Eu3+ at 300 °C for 4.5 days. SEM and XRD results revealed that both bentonite and the container were involved in the immobilization of europium by the formation of insoluble europium silicate phases. The presence of europium silicate adsorbed on the surface of the metallic canister indicates the competitive effect of both components of the engineered barrier (bentonite and metallic canister) in HLRW immobilization. These results suggested that the canister could play a role in the HLRW immobilization even during its corrosion process.

Ionic strength and pH dependent multi-site sorption of Cs onto a micaceous aquifer sediment by Adam J. Fuller; Samuel Shaw; Caroline L. Peacock; Divyesh Trivedi; Joe S. Small; Liam G. Abrahamsen; Ian T. Burke (32-42).
Caesium-137 (t 1/2  = 30 years) is a common contaminant at nuclear legacy sites. Often the mobility of 137Cs in the environment is governed by its sorption to charged sites within the sediment. To this end it is important to understand the sorption behaviour of caesium across a wide range of environmental conditions. This work investigates the effect of varying solution composition (pH and competing ions) on the sorption of caesium to micaceous aquifer sediment across a large concentration range (1.0 × 10−11 – 1.0 × 10−1  mol L−1 Cs+). Experimental results show that Cs+ exhibits three distinct sorption behaviours at three different concentration ranges. At very low concentrations < 1.0 × 10−6  mol L−1 Cs+ sorption was unaffected by competition with Na+ or H+ but significantly reduced in high ionic strength K+ solution. Secondly between 1 × 10−6 and 1.0 × 10−3  mol L−1 Cs+ is strongly sorbed in a neutral pH, low ionic strength background but sorption is significantly reduced in solutions with either a high concentration of Na+ or K+ ions or low pH. At high concentrations > 1.0 × 10−3  mol L−1 Cs+ sorption is reduced in all systems due to saturation of the sediment’s sorption capacity. A multi-site cation exchange model was used to interpret the sorption behaviour. From this it was determined that at low concentrations Cs+ sorbs to the illite frayed edge sites only in competition with K+ ions. However, once the frayed edge sites are saturated the Cs+ sorbs to the Type II and Planar sites in competition with K+, Na+ and H+ ions. Therefore sorption of Cs+ at concentrations > 1.0 × 10−6  mol L−1 is significantly reduced in both high ionic strength and low pH solutions. This is a significant result with regard to predicting the migration of 137Cs+ in acidic or high ionic strength groundwaters.

Display OmittedThe migration of metals as gallium (Ga) in the environment is highly influenced by their sorption on clay minerals, as montmorillonite and illite. Given the increased usage of gallium in the industry and the medicine, the Ga-associated waste may result in environmental problems. Ga sorption experiments were carried out on montmorillonite and illite colloids in a wide range of pH, ionic strength and Ga concentration. A Ga sorption model was developed combining ionic exchange and surface complexation on the edge sites (silanol and aluminol-like) of the clay sheets. The complexation constants were estimated as far as possible from the Ga hydrolysis constants applying the linear free energy relationship (LFER), which allowed to reduce the number of free parameters in the model.The Ga sorption behaviour was very similar on illite and montmorillonite: decreasing tendency with pH and dependency on ionic strength at very acidic conditions.The experimental data modelling suggests that the Ga sorption reactions avoid the Ga precipitation, which is predicted in absence of clay colloids between pH 3.5 and 5.5. Assuming this hypothesis, clay colloids would affect Ga aqueous speciation, preventing precipitation in favour of sorption. Ga sorption on montmorillonite and illite can be explained on the basis of three main reactions: Ga3+ exchange at very acidic conditions (pH < ∼3.8); Ga ( OH ) 4 - complexation on protonated weak sites in acidic-neutral conditions (between pH ∼5.2 and pH ∼7.9); and Ga(OH)3 complexation on strong sites at basic conditions (pH > ∼7.9).

Lead concentrations and stable lead isotopes (204Pb, 206Pb, 207Pb, 208Pb) were measured in forest moss samples (Pleurozium schreberi or Scleropodium purum) collected at 273 sites across the Czech Republic during 2010. Continuously decreasing median Pb concentrations in moss were documented over the last two decades: 1995: 11 mg/kg, 2000: 5.66 mg/kg, 2005: 4.94 mg/kg and 2010: 2.85 mg/kg. Several local anomalies have decreased in scale, the overall regional distribution patterns remained, however, the same. The regional Pb isotope ratio distributions show that the ratios show little variation for a large central part of the country and provide the large-scale background isotope ratios for the Czech Republic of about 204Pb/206Pb = 0.0550, 206Pb/207Pb = 1.167, 206Pb/208Pb = 0.478 and 207Pb/208Pb = 0.409 for 2010. This background Pb isotope ratio signal in moss has been locally (900–7500 km2) modified by specific Pb isotopic ratio signals caused by deposition of Pb emissions from known local anthropogenic Pb emission sources, such as industrial combustion of local coal, and a variety of industrial enterprises (metallurgical, engineering and glass works). At some sites where mining of uranium and polymetallic ores took place the moss samples show also a locally specific Pb isotope signal. The in terms of area affected largest deviations in the Pb-isotope ratios, e.g., in the Bohemian Massif, may be due to the input of geogenic dust.

Dissolution rates of pressure solution (PS) for quartz aggregates in 0.002 M NaHCO3 solution were experimentally determined under low effective stress conditions of 0.42–0.61 MPa, and low temperatures of 25–45 °C. At temperatures of 25 °C, 35 °C, and 45 °C, the resultant silicon dissolution rates are 4.2 ± 1.2 × 10−15, 6.0 ± 1.0 × 10−15 and 7.8 ± 1.9 × 10−15  mol/cm2/s, respectively. Ratios between these dissolution rates and those of quartz sand at zero effective stress are 4.1 ± 1.2 at 25 °C, 3.0 ± 0.5 at 35 °C, and 2.4 ± 0.6 at 45 °C. As the uniaxial pressure was increased, the dissolution rate of PS also increased, though gradually decreased when the effective stress was kept constant. After the removal of stress, the dissolution rate was observed to increase once again. The activation energy of our PS experiments was determined to be approximately 24 kJ/mol, lower than the amount required for quartz sand dissolution to commence at zero effective stress. Our results clearly show that, even at such low temperature and effective stress, Si released into solution as a result of PS can be detected. This implies that experimental compaction of quartz aggregates can be measured even under such condition.

Slags from base metal smelting are often deposited for over 100 years with no barriers between the heap and the surrounding environment. Observations at the external surfaces of the slag heaps may not reflect the state of the slag material inside the slag heap. A slag heap from Zn–Pb ore smelting (Świętochłowice, Upper Silesia, Poland) was disturbed during recent slag removal and the freshly uncovered surfaces are examined in this study. The material forming the interior of the slag heap is fine-grained (up to 5 cm) and strongly weathered (called the weathered slag zone) in contrast to the large slag boulders on the surface of the slag heap (up to 2 m), which are only slightly weathered. The weathered slag zone is composed of gypsum and hematite plus a mixture of primary and other secondary phases. The weathered material as a whole is chemically more homogenous than unweathered slags and has lower Si and higher Fe, Pb (up to 3 wt.%) and Cd (up to 560 mg/kg) concentrations. SEM images show that primary slag phases are porous and disintegrated. The examined surfaces are 3–4 m high and 10–30 m wide suggesting that such slag weathering may have occurred in more extensive parts of the slag heap. Weathering affects slag types with different phase compositions and is therefore controlled by specific conditions occurring within the heap, not by the type of slag. These conditions are so far not well defined, but may include prolonged slag deposition, slow water transfer within the heap and interaction of slags with acid rains common in the Upper Silesia region.The weathered slag zone, when confined within the slag heap, acts as a sink for many potentially toxic elements, decreasing their release to the environment. However, this may change dramatically when the zone is uncovered and its deposit conditions change. Potential risks encompass release of potentially toxic elements from the heap, transport of fine grained particles to the surrounding environment and release of elements when the material is deposited in nearby soils or waters, as well as when it is reused as a construction material.

Osmium geochemistry of modern estuarine sediments from the Tama and Yasaka rivers in Japan by Guodong Zheng; Katsuhiko Suzuki; Akihito Kuno; Motoyuki Mastuo; Bokuichiro Takano; Hiroshi Shimizu (82-88).
Concentrations of Re and Os and 187Os/188Os ratios were obtained on the modern sediment samples from two estuarine areas: the Tama River in Tokyo and the Yasaka River in Oita, Japan. The shallow (<depth 25 cm) intervals of the Tama sediments contain Os of 29.0–36.7 ppt that are quite similar to that in the upper continental crust (UCC) of 30 ppt whereas the deep (>depth 25 cm) intervals are 99.2–105.5 ppt Os that are more than 3 times higher than UCC. However, all samples have 187Os/188Os ratios within a relatively narrow range of 0.3029–0.4035, much lower than that of UCC (1.0–1.3), indicating a similar source of Os in both parts. The lower 187Os/188Os ratios and high Os concentrations of the sediments in the Tama River suggest the impact from human activities. The sudden enrichment of Os at the deep interval may be related to the redox condition change. In addition, the Os concentration is vertically constant in either the upper and lower intervals of the sediment profile, suggesting constant burial fluxes of Os in the sediment.The Os in the Yasaka River estuary sediments (9.22–29.4 ppt) are generally lower than that of the Tama River, either similar or much lower (∼1/3) than the average UCC Os concentrations. The samples with high Os concentrations have lower 187Os/188Os ratios whereas the samples with lower Os concentrations have relatively higher 187Os/188Os ratios, probably indicating different Os sources. The relatively higher 187Os/188Os ratios and lower Os concentrations in some Yasaka River estuarine sediments than the Tama River may also suggest less pollution from human activity to the Yasaka River compared to the Tama River.There is a large decrease of Re concentrations with depth in the Yasaka River estuarine sediment whereas less variations of Re in the Tama River estuarine sediments. The ratio of Re/Os is also sharply different between the two estuarine sediments, also indicating different sedimentary environments. The Tama River sediments may receive much more anthropogenic influence while the Yasaka River sediments get strong impact from modern volcanic activity due to their different geographical and geological surroundings: the Tama River flows through a heavily industrial metropolitan area whereas the Yasaka River flows through a volcanic active but less industrial area.

Uptake of dissolved lead by anhydrite surfaces by Juan Morales; José Manuel Astilleros; Amalia Jiménez; Jörg Göttlicher; Ralph Steininger; Lurdes Fernández-Díaz (89-96).
Display OmittedThe fate of harmful metals in the Earth crust is importantly affected by sorption processes on mineral surfaces. Here, a study of the ability of anhydrite surfaces to uptake dissolved Pb is presented. Experiments were conducted at room temperature using initial Pb concentration ([Pbaq]0) ranging between 10 and 1000 mg/L and a batch type set-up. Inductively coupled plasma optical emission spectrometry analyses showed that [Pbaq] progressively decreased as the time of interaction increased, to reach a final steady state value of ∼3.0 mg/L, irrespectively of [Pbaq]0. However, the time elapsed before the steady state value was reached strongly depended on [Pbaq]0, with the drop to this final value occurring in less than 1 day interaction when [Pbaq]0  ⩾ 50 mg/L and after 20 days when [Pbaq]0  < 50 mg/L. Scanning Electron Microscopy and X-ray diffraction analyses confirmed the epitactic growth of anglesite (PbSO4) crystals on anhydrite surfaces when [Pbaq]0  ⩾ 50 mg/L. X-ray Absorption Near Edge Structure spectroscopy points to a different sorption mechanisms when [Pbaq]0  < 50 mg/L. The results show that the epitactic growth of anglesite on anhydrite has no significant impact on the ability of anhydrite surfaces to remove Pbaq, which show equal effectiveness as that of gypsum surfaces. The high reactivity of anhydrite surfaces renders this phase potentially important in the control of the fate of dissolved metals in nature.

Display OmittedLead slags from two different metallurgical technologies (primary slag from Pb-ore processing and secondary slag from car battery recycling) were submitted to 12-year batch leaching at a L/S ratio of 10 and the leaching data were compared with the previously obtained results from a ⩽1-year leaching test. In the long-term, the Pb release is efficiently controlled by the precipitation of cerussite (PbCO3), Pb-phosphates and anglesite (PbSO4) and by sorption onto newly formed hydrous ferric oxides. In contrast, Zn is the most important contaminant that is continuously released from the primary slag and, after leaching for 12 years, its concentration exceeds more than 40× the regulatory limit value for hazardous wastes (200 mg/kg). For this reason, the recycling of primary slag for civil engineering purposes (concrete technology, gravel-like material) cannot be recommended, because of the long-term Zn leaching. In terms of the leaching of metals (Pb, Zn, Cu) and metalloids (As, Sb), the secondary slag mostly fulfils the limit criteria for inert waste according to the EU legislation even for long-term leaching. However, high release of Ba exceeding more than 10× the limit values for hazardous waste was observed, despite the massive formation of barite (BaSO4) on the slag surface. These results indicate that less common contaminants (e.g. Ba), which have been overlooked so far, should also be considered during assessment of the leaching behaviour of smelting slags with a view to their safe recycling in civil engineering.

Using ion and isotope characterization to design a frame of protection of a wetland system (Massif Central, France) by Agnès Brenot; Philippe Négrel; Romain Millot; Clotilde Bertin (104-118).
The bio-diversity (vegetation and fauna) of peatlands, like all wetland ecosystems, is very fragile as it requires specific wet conditions. Over the past 20 years, increasing efforts have been made to restore degraded wetlands, to re-create new wetlands where they were lost, and to sustainably manage for multiple benefits. However, actions to restore and preserve wetlands require an in-depth knowledge of the water cycle in the system. We used chemical and multi-isotopic approaches, combined with hydrological tools (measuring potentiometric levels and spring discharge), for tracing the water and dissolved-element fluxes in the Narces de la Sauvetat peatland (Central France) and for better understanding of water budget components involved in this ecosystem. This multi-pronged approach clearly demonstrated its effectiveness for improving our understanding of the hydrological functioning of this wetland ecosystem. The two main results are that: (1) The water volume flowing out of the peatland through the Fouragettes stream is often negligible; and (2) At least three strong groundwater fluxes with distinct chemical and isotopic signatures supply water to the peatland. This new understanding will help decision makers maintain the water balance of the peatland, which is essential for the preservation of this fragile ecosystem.

Two international inter-comparison exercises devoted to dissolved gases and isotope analyses in groundwater, used as tools for groundwater dating were organized in 2012 in France (IDES – Université Paris Sud – CNRS and OSUR – Université Rennes 1 – CNRS). The goal was to compare sampling and analytical protocols through results obtained by the community of groundwater dating laboratories. The two exercises were: GDAT1 on three supply boreholes in a homogeneous sand-aquifer of Fontainebleau (Paris Basin, France) and GDAT2 on two supply boreholes (shallow and deep) in a fractured rock aquifer in French Brittany. This two-step exercise is the first exercise which included a large number of gases and isotopes usually used in groundwater as dating tools and also permit to discuss the uncertainties related to sampling protocols issuing from each laboratory methods. The two tests allowed 31 laboratories from 14 countries to compare their protocols for both sampling and analyses. This paper presents the participants and parameters measured, and focuses on the validation of the sampling strategy. Two laboratories analyzed CFC and SF6 samples collected at regular intervals during the sampling operations in order to verify water homogeneity. The results obtained by the two “reference” laboratories along with monitoring of field parameters showed no clear trend of gas concentration or physic-chemical properties. It can be concluded that the pumped groundwater composition remained constant during sampling. This study also shows the potential for relatively constant pumped groundwater composition from a specific well despite the complexity and/or mixing processes that may occur at a larger scale in the aquifer.

Extraction and analysis of organic pollutants from matrices such as sediment constitute an essential step in environmental research. However, the extraction for quantitative analysis can turn out to be difficult because these compounds are present in trace levels and can be strongly bound to the sorbent matrix. Consequently, accuracy of environmental analyses mainly depends on the efficiency and the robustness of the extraction step. In this work, a sequential ASE extraction procedure was applied to the extraction of polycyclic aromatic and aliphatic hydrocarbons (PAHs, Me-PAHs and n-alkanes) in sediment samples. The extraction protocol was developed for 26 PAHs, including the 16 PAHs of the United-States Environmental Protection Agency (EPA) priority list, for 17 alkylated PAHs homologues and for 29 n-alkanes (from n-C12 to n-C40). A set of 30 experiments was carried out for the determination of the optimal extraction conditions. The four parameters studied were pressure, temperature, extraction time and nature of the solvent. Extracts were analyzed by gas chromatography (GC-MS and GC-FID) after clean-up and concentration. The optimal extraction conditions selected for pressure, temperature, extraction time and nature of solvent were respectively 14 MPa, 160 °C, 24 min and hexane/acetone (1/1 v/v). The analytical procedure was validated by comparing predicted and experimental values of sediment samples and by analyzing standard reference material. The validated method was then applied to establish a depth profile contamination in the sediment of the Deûle River in Northern France.

The interactions of strontium and technetium with Fe(II) bearing biominerals: Implications for bioremediation of radioactively contaminated land by Clare L. Thorpe; Christopher Boothman; Jonathan R. Lloyd; Gareth T.W. Law; Nicholas D. Bryan; Nick Atherton; Francis R. Livens; Katherine Morris (135-143).
At nuclear contaminated sites, microbially-mediated Fe(III) reduction under alkaline conditions opens up the potential for co-treatment of the groundwater contaminants 99Tc, though reduction to less mobile Tc(IV) phases, and 90Sr, through increased sorption and/or precipitation promoted at higher pH. In the experiments described here, microbial enrichment cultures derived from representative Sellafield sediments were used to probe the effect of microbially-mediated Fe(III) reduction on the mobility of 99Tc and Sr (as stable Sr2+ at elevated concentrations and 90Sr2+ at ultra-trace concentrations) under both neutral and alkaline conditions. The reduction of Fe(III) in enrichment culture experiments at an initial pH of 7 or 9 resulted in the precipitation of an Fe(II) bearing biomineral comprised of siderite and vivianite. Results showed that TcO 4 - added at 1.6 × 10−6  M was removed (>80%) from solution concurrent with Fe(III) reduction at both pH 7 and pH 9. Furthermore, X-ray absorption spectroscopy of the reduced biominerals confirmed reduction of Tc(VII) to Tc(IV). To understand Sr behaviour in these systems, Sr2+ was added to enrichment cultures at ultra-trace concentrations (2.2 × 10−10  M (as 90Sr2+)) and at higher concentrations (1.15 × 10−3  M (as stable Sr2+)). In ultra-trace experiments at pH 7, microbially active systems showed enhanced removal of 90Sr compared to the sterile control. This was likely due to sorption of 90Sr2+ to the Fe(II)-bearing biominerals that formed in situ. By contrast, at pH 9, the sterile control showed comparable removal of 90Sr to the microbially active experiment even though the Fe-minerals formed were of very different character in the active (vivianite, siderite) versus sterile (an amorphous Fe(III)-phase) systems. Overall, 90Sr bioreduction experiments showed 60–70% removal of the added 90Sr across the different systems: this suggests that treatment strategies involving bioreduction and the promotion of Fe(III)-reducing conditions to scavenge Tc(IV) are not incompatible with treatment of groundwater 90Sr contamination. In systems with elevated Sr2+ concentrations and an initial pH of 7, microbially active systems showed <20% removal of added Sr2+ following Fe(III) reduction with little or no removal in sterile controls. At pH 9, significant Sr2+ was removed from solution in both sterile and microbially active experiments and was attributed to Sr-sorption to mineral phases both chemically precipitated in sterile controls, and biologically precipitated in the microbially active systems. These results confirm that in systems with an elevated natural or anthropogenic Sr2+ loading, bioreduction at modestly alkaline pH is compatible with co-treatment of both TcO 4 - and 90Sr2+. These data are discussed in terms of aqueous geochemistry trends, X-ray diffraction and morphological data, and thermodynamic modelling. The results demonstrate the potential for removal of trace levels of 99Tc and 90Sr2+ from groundwaters during stimulated bioreduction and highlight that in the presence of stable Sr2+, optimal removal for technetium and strontium is likely to occur under mildly alkaline, reducing conditions.

Data for Mo in British surface water and groundwater collated from a number of databases show that concentrations are in most cases low, of the order of 2 μg/L or less. However, variability is large and sporadic high values are found in both streamwaters and groundwaters. Data for some 11,600 British streamwater samples indicate a 10–90th percentile range of 0.08–2.44 μg/L with a median of 0.57 μg/L and maximum observation of 230 μg/L. High values tend to be from streams on clay-rich formations and sulphide-mineralised bedrock and streams affected by localised urban and industrial contamination. Monitored lowland river waters also typically have median concentrations <1 μg/L although higher values are present in a number of urban/industrialised English rivers. Highest observed concentrations (median 20 μg/L), occur under low-flow conditions in a river system from an urban, industrial area of NE England and are likely the products of industrial contamination and mine drainage. Concentrations in 96 upland lakewater samples from NW England are universally 0.1 μg/L or less. Data for 1735 groundwater samples from across Britain have a 10–90th percentile range for Mo of 0.035–1.80 μg/L with a median of 0.20 μg/L and maximum observation of 89 μg/L. Relatively high values derive from some Lower Cretaceous greensand, Carboniferous limestone and mudstone (Coal Measures) aquifers, particularly under anaerobic conditions. Release from Fe oxides and possibly Mn oxides by reductive dissolution is a likely mechanism for the concentrations. Under more rarely observed sulphate-reducing conditions in British aquifers, concentrations of Mo diminish, likely due to sequestration by precipitating sulphide minerals. The observed ranges in aqueous samples indicate that most water sources in Britain have Mo concentrations more than an order of magnitude below the WHO health-based value for Mo in drinking water (70 μg/L). Such sources are unlikely to pose a risk to drinking water from the current WHO guidance, provided those impacted locally by mining or other industrial contamination are avoided.

Noble gas, CFC and other geochemical evidence for the age and origin of the Bath thermal waters, UK by W. Mike Edmunds; W. George Darling; Roland Purtschert; José A. Corcho Alvarado (155-163).
The city of Bath is a World Heritage site and its thermal waters, the Roman Baths and new spa development rely on undisturbed flow of the springs (45 °C). The current investigations provide an improved understanding of the residence times and flow regime as basis for the source protection. Trace gas indicators including the noble gases (helium, neon, argon, krypton and xenon) and chlorofluorocarbons (CFCs), together with a more comprehensive examination of chemical and stable isotope tracers are used to characterise the sources of the thermal water and any modern components. It is shown conclusively by the use of 39Ar that the bulk of the thermal water has been in circulation within the Carboniferous Limestone for at least 1000 years. Other stable isotope and noble gas measurements confirm previous findings and strongly suggest recharge within the Holocene time period (i.e. the last 12 kyr). Measurements of dissolved 85Kr and chlorofluorocarbons constrain previous indications from tritium that a small proportion (<5%) of the thermal water originates from modern leakage into the spring pipe passing through Mesozoic valley fill underlying Bath. This introduces small amounts of O2 into the system, resulting in the Fe precipitation seen in the King’s Spring. Silica geothermometry indicates that the water is likely to have reached a maximum temperature of between 69–99 °C, indicating a most probable maximum circulation depth of ∼3 km, which is in line with recent geological models. The rise to the surface of the water is sufficiently indirect that a temperature loss of >20 °C is incurred. There is overwhelming evidence that the water has evolved within the Carboniferous Limestone formation, although the chemistry alone cannot pinpoint the geometry of the recharge area or circulation route. For a likely residence time of 1–12 kyr, volumetric calculations imply a large storage volume and circulation pathway if typical porosities of the limestone at depth are used, indicating that much of the Bath-Bristol basin must be involved in the water storage.

234U/238U and δ87Sr in peat as tracers of paleosalinity in the Sacramento-San Joaquin Delta of California, USA by J.Z. Drexler; J.B. Paces; C.N. Alpers; L. Windham-Myers; L.A. Neymark; T.D. Bullen; H.E. Taylor (164-179).
The purpose of this study was to determine the history of paleosalinity over the past 6000+ years in the Sacramento-San Joaquin Delta (the Delta), which is the innermost part of the San Francisco Estuary. We used a combination of Sr and U concentrations, δ87Sr values, and 234U/238U activity ratios (AR) in peat as proxies for tracking paleosalinity. Peat cores were collected in marshes on Browns Island, Franks Wetland, and Bacon Channel Island in the Delta. Cores were dated using 137Cs, the onset of Pb and Hg contamination from hydraulic gold mining, and 14C. A proof of concept study showed that the dominant emergent macrophyte and major component of peat in the Delta, Schoenoplectus spp., incorporates Sr and U and that the isotopic composition of these elements tracks the ambient water salinity across the Estuary. Concentrations and isotopic compositions of Sr and U in the three main water sources contributing to the Delta (seawater, Sacramento River water, and San Joaquin River water) were used to construct a three-end-member mixing model. Delta paleosalinity was determined by examining variations in the distribution of peat samples through time within the area delineated by the mixing model.The Delta has long been considered a tidal freshwater marsh region, but only peat samples from Franks Wetland and Bacon Channel Island have shown a consistently fresh signal (<0.5 ppt) through time. Therefore, the eastern Delta, which occurs upstream from Bacon Channel Island along the San Joaquin River and its tributaries, has also been fresh for this time period. Over the past 6000+ years, the salinity regime at the western boundary of the Delta (Browns Island) has alternated between fresh and oligohaline (0.5–5 ppt).