Search

This study provides the first geographic and temporal large-scale analysis of Cu stable isotope compositions in indigenous oysters to biomonitor anthropogenic Cu inputs in aquatic systems. It includes oyster samples from French marine environments with distinct Cu-pollution levels and histories and extends over several decades of biomonitoring. Sample series composed of oysters from the same season and similar size/age intended to avoid biological biases. Oysters in macrotidal estuaries (Loire and Gironde, Western France) display fluctuating Cu concentrations without clear temporal trends, challenging to infer anthropogenic influence. Conversely, the time series of their Cu isotope ratios indicate a constant proportion of natural and anthropogenic Cu sources since the 1980s. In contrast, Cu isotope compositions in oysters from the progressively developing locales (urbanization and recreational boating) of Arcachon and Vilaine bays shift positively over time, along with increasing Cu concentrations. This finding suggests a gradual augmentation in the bioaccumulation of anthropogenic Cu over time. Thus, the time series of Cu isotope compositions in oysters is an unambiguous and robust approach to pinpoint anthropogenic Cu contamination. Furthermore, Cu concentrations combined with isotope compositions yield site-specific fingerprints allowing geographic discrimination among oysters. This two-dimensional Cu signature is promising as a tool for seafood traceability.

Copper (Cu), mercury (Hg) and lead (Pb) are elements and contaminants of historical and emerging concerns in coastal environments. Although Cu is considered essential for living organisms, it is bioactive in a narrow range of optimal concentrations. Hg and Pb have no known biological role and are considered toxic towards organisms. Marine consumers incorporate and bioaccumulate these metals mainly through trophic pathways, which are then important to delineate. Since recently, metal stable isotopes are promising to trace origin and processes of the contamination of marine matrices, including biota. As part of the present study conducted within the SCOTTTI and Pollusols programs, we merged integrated trophic markers (carbon and nitrogen stable isotopes, lipid profiles) and metal stable isotopes to investigate i) the trophic transfer of Cu, Hg and Pb within a model food web of a French human-impacted coastal area (Toulon Harbor, NW Mediterranean); and ii) the potential sources of food web contamination. Whenever possible, samples included sediment, suspended organic matter, size fractions of plankton, filter-feeding bivalves (mussels, oysters) and planktivorous fishes, collected at different seasons. Metal- and species-specific patterns throughout the food web were observed, highlighting the role of biological/physiological processes on the observed metal levels and isotope distributions.

The application of zinc (Zn) isotope compositions in bivalve organisms to quantify anthropogenic Zn bioaccumulation in marine biota is of great interest to environmental marine management programs such as the “Mussel Watch Program”. Field studies, however, are urgently needed to test its practical value. To this end, we investigated Zn isotope variations in the oyster Crassostrea gigas collected over four decades near the Loire estuary (France), where previous geochemical studies provided evidence for a regionally uniform but temporally variable metal contamination. We show that the Zn temporal isotope profile of oysters matches that of the sedimentary records with an isotope offset of approximately +0.5–0.7‰, tentatively attributed to compromised estuarine processes and trophic transfer. A Zn isotope model for quantifying anthropogenic Zn bioaccumulation suggests an overall decrease in anthropogenic estuarine Zn levels over the past 40 years. This first successful application of Zn isotope ratios in a bivalve species to quantify anthropogenic Zn bioaccumulation confirms their utility for supporting environmental management strategies in marine biomonitoring programs.

In order to move forward in the acceptance of a novel contaminant monitoring technique (Diffusive Gradients in Thin-films: DGT) for assessment of marine water bodies, sensu the WFD, an Inter-Laboratories Comparison (ILC) exercise (nine Europeans laboratories) was organized in the framework of the Interreg Atlantic Area MONITOOL project, which focused on the use of the DGT technique for the measurement of WFD priority metals (Cd, Ni and Pb). Reproducible results were obtained for each metal by several laboratories, supporting the assertion that DGT analysis can be performed satisfactorily by laboratories experienced in measuring metals at trace levels in marine environments, even if they have limited practice in DGT analysis. According to the Z-score analysis, among the 9 participating laboratories, 3 had 100 % of satisfactory results for Cd, Ni, and Pb, 3 had80 % satisfactory results and 2 had about 60 % satisfactory results. This work highlights the need to clearly describe the DGT method in order to control sources of contamination during analytical steps, in particular the resin gel retrieval and the elution steps. Such international intercomparison exercise is an important step to develop the laboratory network involved in DGT analysis and contributes to the improvement of data quality.

SF and the Si isotope measurements at STAiG were supported by NERC grant NE/R002134/1 to PS; PS would also like to cite the support of a Carnegie Trust Research Incentive Grant, which helped the setup of various isotope techniques in the St Andrews Isotope Geochemistry (STAiG) laboratories. C.D. Coath is thanked for laboratory support and the European Research Council is acknowledged for funding (ICY-LAB, grant agreement 678371). The use of silicon (Si) isotopes has led to major advances in our understanding of Si cycling in modern and past environments. This inter‐laboratory comparison exercise provides the community with the first set of soil and plant reference materials with an analytically challenging matrix containing organic material that is known to induce isotopic bias, for use as secondary reference materials in Si isotope measurement. Seven laboratories analysed four soil reference materials (GBW‐07401, GBW‐07404, GBW‐07407, TILL‐1) and one plant reference material (ERM‐CD281). Participating laboratories employed a range of chemical preparation methods and analytical setups but all analyses were performed by MC‐ICP‐MS. Irrespective of the chemical preparation method or analytical conditions, the results show excellent agreement among laboratories within 2s for at least three replicates. Data were combined together to calculate δ29Si and δ30Si mean values (relative to NBS 28) and their expanded uncertainties (U, coverage factor k = 2). The δ30Si values are as follow: GBW‐07401: ‐0.27 ± 0.06 ‰, GBW‐07404: ‐0.76 ± 0.12 ‰, GBW‐07407: ‐1.82 ± 0.17 ‰, TILL‐1: ‐0.16 ± 0.06 ‰ and ERM‐CD281: ‐0.28 ± 0.11 ‰. Also, a compilation of published data provides an up‐to‐date mean δ30Si for BHVO‐2 of ‐0.28 ± 0.08 ‰. Postprint

International audience; Colloids have been recognized as key vectors of pollutants in aqueous environment. Amongst them, those formed by iron (Fe) and organic matter (OM) are of major importance due to their ubiquity in the surface environment and strong affinity for metals. In the recent years, Fe stable isotopes have been increasingly used to elucidate the sources and biogeochemical cycling of Fe in Earth's surface environments. In this study, we aim to elucidate (i) the possible Fe isotopic signature resulting from the Fe/OM colloid formation and (ii) the mechanisms involved in the development of such isotopic signature. For this purpose, Fe-OM associations were synthesized through binding and titration experiments. Various pH levels were used in order to study the isotope behavior of Fe occurring as free species at pH 1, as Fe-OM complexes at pH 2 and as mixed Fe-oxyhydroxide/OM nanoaggregates or particles at pH 6.5. Organic matter-free, Fe-free and OM membrane-deposition experiments were also performed. These suspensions were (ultra)filtered at 0.2 µm, 30 kDa and 5 kDa to evidence the possible Fe isotope fractionation between fractions. This protocol allowed also testing the potential of (ultra)filtration techniques to generate isotope fractionation. The results provided evidence that abiotic Fe precipitation, (ultra)filtration techniques and OM deposition were not able to produce significant Fe isotope fractionation under the experimental conditions. However, at circum-neutral pH, the Fe-OM binding and titration experiments displayed a significant enrichment of heavy Fe isotopes in the < 30 kDa fractions relative to the total Fe pool δ56Fe = 0.35 ± 0.05‰ and 0.26 ± 0.05‰ (95% confidence interval, 2σ and relative to international standard IRMM-14), respectively. Mass balance and error propagation calculation showed Fe isotope fractionation in binding and titration experiments between the > 30 kDa and < 30 kDa fractions for -0.35 ± 0.05‰ and -0.27 ± 0.05‰, respectively. This Fe isotope fractionation could be due to the complexation of Fe by OM in the < 30 kDa fractions. At pH 2, the OM-free experiment, the < 30 kDa fraction showed Fe isotope ratio δ56Fe = 0.75 ± 0.03‰ with an enrichment in heavy Fe isotopes of δ56Fe’ = 0.14 ± 0.04‰ relative the total Fe pool (δ56Fe’ is δ56Fe value which was corrected by δ56Fe of total fraction). This enrichment in heavy Fe isotopes induced an isotopic fractionation factor of -0.87 ± 0.26‰ between the > 30 kDa and < 30 kDa fractions produced by the complexation between the heavy Fe isotopes and OH- ligands in the < 30 kDa fraction. Natural Fe-OM associations were further investigated through oxidation experiments of a reduced wetland soil solution. The oxidized soil solution was (ultra)filtered at 5 µm, 3 µm, 0.2 µm, 30 kDa and 5 kDa. The highest δ56Fe was obtained in the smallest size fraction, i.e. < 5 kDa fraction, yielding a negative isotopic fractionation Δ56Fe >5kDa

Metal release into the environment from anthropogenic activities may endanger ecosystems and human health. However, identifying and quantifying anthropogenic metal bioaccumulation in organisms remain a challenging task. In this work, we assess Cu isotopes in Pacific oysters (C. gigas) as a new tool for monitoring anthropogenic Cu bioaccumulation into marine environments. Arcachon Bay was taken as a natural laboratory due to its increasing contamination by Cu, and its relevance as a prominent shellfish production area. Here, we transplanted 18-month old oysters reared in an oceanic neighbor area into two Arcachon Bay mariculture sites under different exposure levels to continental Cu inputs. At the end of their 12-month long transplantation period, the oysters’ Cu body burdens had increased, and was shifted toward more positive δ65Cu values. The gradient of Cu isotope compositions observed for oysters sampling stations was consistent with relative geographic distance and exposure intensities to unknown continental Cu sources. A binary isotope mixing model based on experimental data allowed to estimate the Cu continental fraction bioaccumulated in the transplanted oysters. The positive δ65Cu values and high bioaccumulated levels of Cu in transplanted oysters support that continental emissions are dominantly anthropogenic. However, identifying specific pollutant coastal source remained unelucidated mostly due to their broader and overlapping isotope signatures and potential post-depositional Cu isotope fractionation processes. Further investigations on isotope fractionation of Cu-based compounds in an aqueous medium may improve Cu source discrimination. Thus, using Cu as an example, this work combines for the first time a well-known caged bivalve approach with metal stable isotope techniques for monitoring and quantifying the bioaccumulation of anthropogenic metal into marine environments. Also, it states the main challenges to pinpoint specific coastal anthropogenic sources utilizing this approach and provides the perspectives for further studies to overcome them.

Copper (Cu) isotope compositions in bivalve mollusks used in marine-monitoring networks is a promising tool to monitor anthropogenic Cu contamination in coastal and marine ecosystems. To test this new biomonitoring tool, we investigated Cu isotope variations of two bivalves-the oyster Crassostrea gigas and the mussel Mytilus edulis-over 10 years (2009-2018) in a French coastal site contaminated by diffuse Cu anthropogenic sources. Each species displayed temporal concentration profiles consistent with their bioaccumulation mechanisms, that is, the Cu-regulating mussels with almost constant Cu concentrations and the Cu-hyperaccumulating oysters with variable concentrations that track Cu bioavailability trends at the sampling site. The temporal isotope profiles were analogous for both bivalve species, and an overall shift toward positive δ65Cu values with the increase of Cu bioavailabilities was associated with anthropogenic Cu inputs. Interestingly, mussels showed wider amplitudes in the isotope variations than oysters, suggesting that each species incorporates Cu isotopes in their tissues at different rates, depending on their bioaccumulation mechanisms and physiological features. This study is the first to demonstrate the potential of Cu isotopes in bivalves to infer Cu bioavailability changes related to anthropogenic inputs of this metal into the marine environment.

We have examined the distribution and spectiation mercury and trace metals under contrasted season, river flow conditions and tidal amplitudes with a series of cruises taking place between 2012 and 2015. Our laboratory had alredy undertaken a similar study in the early 1990's. The latest data acquisition efforts are an opportunity to examine the changes that occurred within that time span for mercury and its species, and to discuss the causes of the observed changes. In particular, we will replace the observations in the perspectives of environmental regulation, and of field and laboratory practices observed for environmental chemical characterization. Early 1980's estuarine transects show total dissolved Hg levels near 5nM in the Loire river (Frenet, 1981). In the early 1990's, a careful and detailed study shows estuarine levels between 1 and 6pM (Coquery, 1994) with a strong dependency on the hydrological regime. New results confirm the riverine clevels in the low pM range. Concentrations decrease progressively during mixing with seawater and the oceanic end-member concentrations are near 0.5pM. Monitoring of mercury levels and speciation in the Loire river and the upper reaches of the estuary show that these Hg levels and species fluctuate with water flow and areal flooding upstream, AND on the history of that hydrological regime.

The advent of Multicollector ICP-MS advent inaugurated the analysis of new metal isotope systems, the so-called “non-traditional” isotopes. They are now available tools to study geochemical and ecotoxicological aspects of marine metal contamination and hence, to push the frontiers of our knowledge. However, such applications are still in their infancy, and an accessible state-of-the-art describing main applications, obstacles, gaps, and directions for further development was missing from the literature. This paper fills this gap and aims to encourage the marine scientific community to explore the contributions of this newly available information for the fields of chemical risk assessment, biomonitoring, and trophic transfer of metal contaminants. In the current “Anthropocene” epoch, metal contamination will continue to threaten marine aquatic ecosystems, and “non-traditional” isotopes can be a valuable tool to detect human-induced changes across time-space involving metal contaminants, and their interaction with marine biota.

During the MARSITECruise expedition in November 2014 on board the RV Pourquoi Pas?, multidisciplinary sampling was carried out with the ROV Victor 6000 in order to investigate biogeochemical processes taking place at cold seep environments in the Sea of Marmara. Pore water, bottom water, sediment and authigenic carbonate samples were collected from two short push cores (MRS-DV5-PC04 − 8 cm, MRS-DV5-PC01 − 12.5 cm) at an active methane bubbling site in the southeastern part of the Cinarcik Basin. Sulfate sulfur and oxygen isotope data as well as sulfide isotope data indicate that sulfate-dependent anaerobic oxidation of methane is the dominant process in the sediments. This is confirmed by archaeal lipids diagnostic for anaerobic methane oxidizers detected with strong 13C-depletions. The available data even allows to distinguish the dominant AOM assemblages. Specific lipid patterns are consistent with a dominance of ANME-2 archaea within the microbial community. Abundant authigenic carbonates (mostly aragonite), found at all depths, show a narrow range in δ13C values between −27.69‰ and −33.40‰. The carbon isotopic composition of the dissolved inorganic carbon as well as strontium and calcium isotopes confirm that the current reaction zone (sulfate-methane transition zone) starts at the bottom of the core. All shallower carbonates are witnesses of paleo seepage activity. U-Th dating of four pure aragonite samples show the short time span that is preserved in core MRS-DV5-PC01 (235 ± 60 yr B.P.). Two major earthquakes of 1766 CE and 1754 CE in the Cinarcik Basin might potentially have triggered the increased seepage of methane at this location.

The δ30Si stable isotopic composition of silicon in soils and fine-grained sediments can provide insights into weathering processes on continents, with important implications on the Si budget of modern and past oceans. To further constrain the factors controlling the distribution of Si isotopes in sediments, we have analysed a large number (n = 50) of separate size-fractions of sediments and suspended particulate materials collected near the mouth of rivers worldwide. This includes some of the world’s largest rivers (e.g. Amazon, Congo, Mackenzie, Mississippi, Murray-Darling, Nile, Yangtze) and rivers from the case study areas of the Congo River Basin and Northern Ireland. Silt-size fractions exhibit a mean Si isotopic composition (δ30Si = −0.21 ± 0.19‰; 2 s.d.) similar to that previously inferred for the upper continental crust. In contrast, clay-size fractions display a much larger range of δ30Si values from −0.11‰ to −2.16‰, which yield a global δ30Siclay of −0.57 ± 0.60‰ (2 s.d.) representative of the mean composition of the average weathered continental crust. Overall, these new data show that the Si isotopic signature transported by river clays is controlled by the degree of chemical weathering, as inferred from strong relationships with Al/Si ratios. At a global scale, the clay-bound Si isotopic composition of the world’s largest river systems demonstrates a link with climate, defining a general correlation with mean annual temperature (MAT) in corresponding drainage basins. While the distribution of Si isotopes in river sediments also appears to be influenced by the tectonic setting, lithological effects and sediment recycling from former sedimentary cycles, our results pave the way for their use as paleo-weathering and paleo-climate proxies in the sedimentary record.

Estuaries can be considered as biogeochemical reactors, acting both as sinks or sources for trace metals depending on environmental factors. Distribution of dissolved and particulate Cd, Co, Cu, Mn, Ni, and Zn were determined in samples collected from the Loire river estuary (France) during seasonal contrasting flow conditions (summer, spring and winter). Mean concentrations of dissolved Cd (0.13 ± 0.07 nM) were below concentrations previously observed in the estuary, while Cu (23.00 ± 8.07 nM) concentrations were similar to those found ten years ago. The results also demonstrated a wide fluctuation of the TMZ (Turbidity Maximum Zone) following the water discharge conditions and various reactivity behaviours depending on the type of metal even if all metals presented stable log KD values along the estuary. Rapid adsorption of suspended matter occurs at the salinity front, and only Ni and Cu are conservative through the salinity gradient. Organic matter appeared to be a major factor in metals fractioning and its role though the estuary is discussed. Riverine particulate and dissolved metal inputs, and estuary dissolved efflux have been estimated, and show a large export of Cu and Zn to the sea. However, the Loire river outputs only play a minor contribution to the ocean at the global scale. These results provide unique information concerning the contemporary distribution of dissolved and particulate trace metals in the Loire estuary, and provide a geochemical baseline dataset to assess the evolution of historical contaminants in the estuary. Ongoing studies on emerging metallic elements must be carried out to understand the impact of urban development in this part of the river.

Ultramafic soils constitute an extreme environment for plants because of specific physico-chemical properties and the presence of Ni, Cr, and Co. We hypothesized that type of ultramafic parent rock depending on their origin affects the composition of soils and plants. Therefore, phytoavailability of metals would be higher in soil derived from serpentinized peridotite compared to serpentinite because of differences in susceptibility of minerals to weathering. Based on DTPA-CaCl2 extractions, we noted that soil derived from the serpentinized peridotite is characterized by a higher phytoavailability of Ni compared to soil derived from the serpentinite. On the contrary, plant species growing on soil derived from the serpentinite contain higher concentrations of metals. Our study suggests that the metal uptake by plants is controlled by the mineral composition of parent rocks, which results from both their original magmatic composition and later metamorphic processes. Chemical extractions show that the phytoavailability of Ni and Co is higher in soil derived from the serpentinized peridotite than the serpentinite. Surprisingly, plants growing on the soil derived from the serpentinite contain higher levels of metals compared to these from the serpentinized peridotite derived soil. This contrasting behavior is due to higher abundances of Ca and Mg, not only Ni and Co, in soil derived from the serpentinized peridotite as compared to those in the soil derived from the serpentinite. Calcium and Mg are favored by plants and preferably fill the available sites, resulting in low Ni and Co intake despite their higher abundances.

Epithermal mineralization was recently described in synrift volcanic rocks from the SE Afar Rift in Djibouti. To infer fluid sources in this system, coupled sulfur and strontium isotopic analyses were performed on mineralized samples from hydrothermal veins and stockwork zones. Mineralization includes gypsum and chalcedony, and/or quartz ± carbonate containing gold and sulfides. Sulfur isotopic compositions of sulfides and sulfates were determined using Multicollector-Inductively Coupled Plasma Mass Spectrometer (MC ICPMS); Laser Ablation MC ICPMS; and Continuous Flow Elemental Analyzer-Isotope Ratio Mass Spectrometry while strontium isotopes were determined by Thermal Ionisation Mass Spectrometry (TIMS) methods. Sulfur isotopic composition of sulfides (mainly pyrite) varies from − 9.2 to + 6.6‰. These values are classically reported for hydrothermal sulfides resulting from the mixing of variable S sources such as seawater sulfate and magmatic source. In addition, individual pyrite grains within the same sample show both positive and negative δ34S values suggesting multi-phase mineralization processes. Coupled sulfur and strontium isotopic compositions of seven sulfate samples hosted in (i) gypsum mound, (ii) evaporites, (iii) stockwork veins in sediments or in volcanic rocks were also investigated to characterize the sources of Sr and S. The δ34S values of the sulfate ranged from − 0.8 to + 14.3‰ while the 87Sr/86Sr ranged from 0.70389 to 0.70639 consistent with variable mixing ratios of Sr and S derived from volcanic sources and seawater-derived sulfate. This study shows that, although precious-metal deposition is related to low-sulfidation epithermal environment, both acidic fluid of magmatic and saline origins contributed to the hydrothermal system in the SE Afar Rift.

International audience; Amazonian goliath catfishes are widespread in the Amazon Basin. Recently, otolith 87Sr:86Sr analyses using laser ablation–multi‐collector–inductively coupled plasma mass spectrometry (LA‐MC‐ICPMS) revealed a >8,000 km trans‐Amazonian natal homing in Brachyplatystoma rousseauxii among fish caught and hatched in the largest Amazon River tributary, the upper Madeira basin. Although also suspected for fish in the upper Amazon, homing could not be demonstrated owing to less distinct environmental 87Sr:86Sr gradients along the Amazon mainstem. Using scanning X‐ray fluorescence microscopy (SXFM), a separate study provided evidence that Se:Ca and Sr:Ca are useful markers for identifying migration into Andean headwaters and the estuarine environment. We analysed otoliths of known 87Sr:86Sr profiles using SXFM mapping to test if Sr:Ca and Se:Ca patterns could demonstrate natal homing for three fish caught in the upper Amazon, using as reference two individuals that were natal homers and two forced residents (hatched after the construction of hydroelectric dams on the Madeira River) from the upper Madeira River. As hypothesised, although the Sr isotope profiles of the upper Amazon individuals were uninformative, two of them presented similar alternating mirror patterns of Sr:Ca and Se:Ca to those of the upper Madeira natal homers, indicating migrations out of the Andean region and into the estuary area. Both were therefore natal homers from the upper Amazon. The third individual from the upper Amazon presented similar Sr:Ca and Se:Ca patterns to those of the upper Madeira residents, suggesting it was a natural resident from the upper Amazon. By combining the results of 87Sr:86Sr analyses (LA‐MC‐ICPMS) and Sr:Ca and Se:Ca mappings (SXFM) that are completely independent of one another, we demonstrated that B. rousseauxii also performs natal homing in the upper Amazon. Our results indicate that the life cycle of B. rousseauxii is more complex than previous literature hypothesised, with the existence of partial migration, even in absence of physical barriers. Quantifying the relative importance of these different life‐history strategies will have important implications for fisheries management. Our results also lay the groundwork for conservation efforts in the context of hydropower development in the Amazon Basin and set testable hypotheses of the potential impacts of the Madeira River dams.

Toulon bay is severely impacted by metal contamination induced by past and recent naval activities. In this work, Cu, Zn and Pb isotope compositions and elemental concentrations of fifty-five surface sediments were determined in order to map the spatial distribution of anthropogenic and natural sources along this land-sea continuum. Two sub-systems of Toulon Bay, the Small and Large bays, showed well-marked patterns on metal contamination levels and isotope signatures for Cu and Pb. The Small bay had the highest metal concentrations, and displayed average Pb and Cu isotope compositions of 1.1664 ± 0.0043 (1s, expressed as 206Pb/207Pb ratios) and −0.17 ± 0.19‰ (1s, expressed as δ65CuNIST values), respectively. It contrasted with the Large bay, with moderate to pristine metal concentrations and average Pb and Cu isotope compositions of 1.1763 ± 0.0079 (1s) and +0.08 ± 0.23‰ (1s), respectively. Lead isotope systematics indicated a binary source mixing process involving industrial and natural sources, while Cu isotope systematics showed a ternary mixing process involving two distinct anthropogenic signatures, interpreted as associated to new diffuse anthropogenic sources and old warfare material. In contrast, Zn isotope compositions in the Small and Large bays were practically the same: +0.06 ± 0.05‰ and +0.06 ± 0.11‰ (1s, expressed δ66ZnJMC values), respectively, denoting an overlap between isotope signatures of natural and anthropogenic sources. This study presents the first detailed spatial distributions of Cu and Zn isotope compositions for an aquatic system, and demonstrates the feasibility to use Cu isotopes as tracers of anthropogenic sources in coastal environments.

In this work, a multi-elemental approach combining Cu and Zn stable isotopes is used to assess the metal contamination evolution in the Loire estuary bulk sediments. Elemental geochemical data indicate an increase of metal concentrations from the beginning of the industrial period peaking in the 1990s, followed by an attenuation of metal contamination inputs to the estuary. Zinc isotope compositions suggest a binary mixing process between Zn derived from terrigenous material and multi-urban anthropogenic sources. Copper isotope systematics indicate a single natural dominant source represented by weathered silicate particles from soils and rocks. This work demonstrates the applicability of Zn isotopes to identify anthropogenic Zn sources in coastal systems, even under a low to moderate degree of contamination. Further studies are required to constrain Cu sources and to elucidate possible effects of grain-size and mineralogy in the Cu isotope composition of sediment in the Loire estuary.

International audience; In the Amazon, migratory catfishes of the genus Brachyplatystoma are apex predators that are important for fisheries and conservation. The life cycle of Brachyplatystoma platynemum Boulenger, 1898 is poorly known, although it has been hypothesized to be very similar to that of Brachyplatystoma rousseauxii Castelnau, 1855, which uses the entire length of the Amazon basin to complete its life cycle (from the Andes to the estuary). This study provides the first data on the migratory patterns of B. platynemum at the individual level using otolith microchemistry. In total, 94 individuals were sampled close to major breeding areas in the Amazon basin (78 fish from the middle and upper Madeira River and 14 fish from the upper Amazon), and their lifetime movements were assessed by measuring variations in 87 Sr/ 86 Sr along transverse sections of their otoliths (ear stones), using laser ablation multi-collector mass spectrometry (LA-MC-ICP-MS). The migrations of B. platynemum are not as extensive as those of B. rousseauxii, and do not involve natal homing. Furthermore, the estuary is not a nursery area, at least for fish hatched in the Madeira. Nevertheless, B. platynemum migrates several thousand kilometres within the Amazon basin, with transboundary displacements between at least Bolivia, Brazil, and Peru. Current and planned hydroelectric development in the Amazon basin will severely disrupt both migration and access to breeding grounds, ultimately affecting the recruitment and population dynamics of these apex predators. The conservation of B. platynemum is crucial for the stability of the Amazonian aquatic food webs. This requires building effective fish passage on the two existing Madeira dams and considering alternative options to the large-scale hydropower development in the Amazon basin

There is increasing evidence that abrupt vegetation shifts and large-scale erosive phases occurred in Central Africa during the third millennium before present. Debate exists as to whether these events were caused by climate change and/or intensifying human activities related to the Bantu expansion. In this study, we report on a multi-proxy investigation of a sediment core (KZR-23) recovered from the Congo submarine canyon. Our aim was to reconstruct climate, erosion and vegetation patterns in the Congo Basin for the last 10,000 yrs, with a particular emphasis on the late Holocene period. Samples of modern riverine suspended particulates were also analyzed to characterize sediment source geochemical signatures from across the Congo watershed. We find that a sudden increase of bulk sediment aluminium-to-potassium (Al/K) ratios and initial radiocarbon ages of bulk organic matter occurred after 2,200 yrs ago, coincident with a pollen-inferred vegetation change suggesting forest retreat and development of savannas. Although hydrogen isotope compositions of plant waxes (δDwax) do not reveal a substantial hydroclimate shift during this period, neodymium isotopes and rare earth elements in detrital fractions indicate provenance changes for the sediment exported from the Congo Basin at that time, hence suggesting a reorganization of spatial rainfall patterns across Central Africa during this event. Taken together, these findings provide evidence for changing landscapes in Central Africa from about 2,200 yrs ago, associated with synchronous events of vegetation changes and enhanced erosion of pre-aged and highly weathered soils. These events coincided remarkably well with the arrival of Iron Age communities into the rainforest, as inferred from comparison to regional archaeological syntheses. While the human impact on the environment remains difficult to quantify at the scale of the vast Congo Basin, we tentatively propose that strengthening of El Nino-Southern Oscillation (ENSO) variability at that time played a key role in triggering the observed environmental changes, and possibly acted as a driver for the eastward migration of Bantu-speaking peoples across Central Africa.

International audience; Ultramafic derived soils are characterized by low nutrient soils, a low Ca:Mg ratio, and high metal contents such as Ni, Co and Cr. The vegetation growing on these soils is highly adapted and includes both Ni hyperaccumulator and accumulator species. Today, approximately 530 Ni hyperaccumulator species are listed worldwide and the Ni concentration can be extremely high, e.g. up to 25% in latex from Pycnandra acuminata (Sapotaceae), a tree found in New Caledonia. The aim of this study is to identify the potential role of Ni hyperaccumulator plants in the Ni biogeochemical cycle at the soil surface by using Ni isotopes. A set of Ni hyperaccumulator and Ni accumulator plants as well as topsoils were sampled on the Barro Alto and Niquelândia ultramafic complexes (Goiás State, Brazil). Three Ni hyperaccumulator plants were collected: Justicia lanstyakii, Heliotropium aff. salicoides, Cnidoscolus aff. urens, as well as one Ni accumulator plant, Manihot sp. The isotopic compositions of the whole plants were determined and compared to those of the bulk topsoils and DTPA-extractable Ni. The topsoils exhibited δ60Ni values ranging from −0.30 ± 0.06‰ to 0.16 ± 0.05‰. The DTPA-extractable Ni in the topsoils ranged from 94 to 623 mg kg−1, i.e. 0.9–4.9% of the total soil Ni and was found to be isotopically heavier than the corresponding topsoil (from −0.30 ± 0.05‰ to 0.34 ± 0.08‰). The δ60Ni values for the Ni accumulator plants showed an enrichment in heavy Ni isotopes in the aerial parts of the plant compared to the roots, whereas similar δ60Ni values for the roots, stems and aerial parts suggested that no significant fractionation results from Ni uptake and translocation in Ni hyperaccumulator plants. Moreover, the aerial parts (i.e. leaves and flowers) from all of the plants analyzed showed the highest Ni concentrations and the heaviest δ60Ni values up to 1.21 ± 0.05‰. The enrichment in heavy Ni isotopes in the leaves (0.09 ± 0.06‰

Ultramafic (UM) rocks are known to be nickel (Ni) rich and to weather quickly, which makes them a good candidate to look at the Ni isotope systematics during weathering processes at the Earth’s surface. The present study aims at identifying the Ni solid speciation and discussing the weathering processes that produce Ni isotope fractionation in two deep laterite profiles under tropical conditions (Barro Alto, Goias State, Brazil). While phyllosilicates and to a lower extent goethite are the main Ni-bearing phases in the saprolitic part of the profile, iron (Fe) oxides dominate the Ni budget in the lateritic unit. Nickel isotopic composition (δ60Ni values) has been measured in each unit of the regolith, i.e., rock, saprock, saprolite and laterite (n = 52). δ60Ni varies widely within the two laterite profiles, from −0.10 ± 0.05‰ to 1.43 ± 0.05‰, showing that significant Ni isotope fractionation occurs during the weathering of UM rocks. Overall, our results show that during weathering, the solid phase is depleted in heavy Ni isotopes due to the preferential sorption and incorporation of light Ni isotopes into Fe oxides; the same mechanisms likely apply to the incorporation of Ni into phyllosilicates (type 2:1). However, an isotopically heavy Ni pool is observed in the solid phase at the bottom of the saprolitic unit. This feature can be explained by two hypotheses that are not mutually exclusive: (i) a depletion in light Ni isotopes during the first stage of weathering due to the preferential dissolution of light Ni-containing minerals, and (ii) the sorption or incorporation of isotopically heavy Ni carried by percolating waters (groundwater samples have δ60Ni of 2.20 and 2.27‰), that were enriched in heavy Ni isotopes due to successive weathering processes in the overlying soil and laterite units.

Vertical distributions of iron (Fe) concentrations and isotopes were determined in the total dissolvable and dissolved pools in the water column at three coastal stations located along the Peruvian margin, in the core of the Oxygen Minimum Zone (OMZ). The shallowest station 121 (161 m total water depth) was characterized by lithogenic input from the continental plateau, yielding concentrations as high as 456 nM in the total dissolvable pool. At the 2 other stations (stations 122 and 123), Fe concentrations of dissolved and total dissolvable pools exhibited maxima in both surface and deep layers. Fe isotopic composition (δ56Fe) showed a fractionation toward lighter values for both physical pools throughout the water column for all stations with minimum values observed for the surface layer (between -0.64 and -0.97 ‰ at 10 - 20 m depth) and deep layer (between -0.03 to -1.25 ‰ at 160 - 300 m depth). An Fe isotope budget was established to determine the isotopic composition of the particulate pool. We observed a range of δ56Fe values for particulate Fe from + 0.02 to -0.87 ‰, with lightest values obtained at water depth above 50 m. Such light values in the both particulate and dissolved pools suggest sources other than atmospheric dust deposition in the surface ocean, including lateral transport of isotopically light Fe. Samples collected at station 122 closest to the sediment show the lightest isotope composition in the dissolved and the particulate pools (-1.25 and -0.53 ‰ respectively) and high Fe(II) concentrations (14.2 ± 2.1 nM) consistent with a major reductive benthic Fe sources that is transferred to the ocean water column. A simple isotopic model is proposed to link the extent of Fe(II) oxidation and the Fe isotope composition of both particulate and dissolved Fe pools. This study demonstrates that Fe isotopic composition in OMZ regions is not only affected by the relative contribution of reductive and non-reductive shelf sediment input but also by seawater-column processes during the transport and oxidation of Fe from the source region to open seawater.

Within the Sea of Marmara, the highly active North Anatolian Fault (NAF) is responsible for major earthquakes (Mw 7), and acts as a pathway for fluid migration from deep sources to the seafloor. This work reports on pore water geochemistry from three sediment cores collected in the Gulfs of Izmit and Gemlik, along the Northern and the Middle strands of the NAF, respectively. The resulting data set shows that anaerobic oxidation of methane (AOM) is the major process responsible for sulfate depletion in the shallow sediment. In the Gulf of Gemlik, depth concentration profiles of both sulfate and alkalinity exhibit a kink-type profile. The Sulfate Methane Transition Zone (SMTZ) is located at moderate depth in the area. In the Gulf of Izmit, the low concentrations observed near the seawater-sediment interface for sulfate, calcium, strontium, and magnesium result from rapid geochemical processes, AOM, and carbonate precipitation, occurring in the uppermost part of the sedimentary column and sustained by free methane accumulation. Barite dissolution and carbonate recrystallization have also been identified at deeper depth at the easternmost basin of the Gulf of Izmit. This is supported by the profile of the strontium isotope ratios (87 Sr/ 86 Sr) as a function of depth which exhibits negative anomalies compared to the modern seawater value. The strontium isotopic signature also shows that these carbonates had precipitated during the reconnection of the Sea of Marmara with the Mediterranean Sea. Finally, a first attempt to interpret the sulfate profiles observed in the light of the seismic activity at both sites is presented. We propose the hypothesis that seismic activity in the areas is responsible for the transient sulfate profile, and that the very shallow SMTZ depths observed in the Gulf of Izmit is likely due to episodic release of significant amount of methane.

International audience; In this preliminary study, we investigate living (stained) foraminifera from the Sea of Marmara. We focus on the faunal composition and geochemical signatures (trace elements, carbon and oxygen stable isotopes) in foraminiferal tests at two deep-sea sites (329 and ~ 1240 m depth respectively). Documented by ROV observations and sampling, both study areas are heterogeneous (including bacterial mats and carbonate concretions), proximal to cold seeps and consist of dysoxic bottom water (O2 < 20 µmol/L). The prevailing dysoxia at both study areas restricts foraminiferal diversity to very low values (S < 9, H’ < 0.97). Stress-tolerant species Bolivina vadescens and Globobulimina affinis dominate living faunas at both sites. The highest foraminiferal standing stock is recorded at the shallowest site underneath a spreading bacterial mat. No benthic foraminifera from either site possess geochemical signatures of methane seepage. Our biogeochemical results show that use of foraminiferal Mn/Ca ratios as a proxy for bottom water oxygenation depends strongly on regional physiography, sedimentary processes and water column structure.

International audience; MarsiteCruise was undertaken in October/November 2014 in the Sea of Marmara to gain detailed insight into the fate of fluids migrating within the sedimentary column and partially released into the water column. The overall objective of the project was to achieve a more global understanding of cold-seep dynamics in the context of a major active strike-slip fault. Five remotely operated vehicle (ROV) dives were performed at selected areas along the North Anatolian Fault and inherited faults.To efficiently detect, select and sample the gas seeps, we applied an original procedure. It combines sequentially (1) the acquisition of ship-borne multibeam acoustic data from the water column prior to each dive to detect gas emission sites and to design the tracks of the ROV dives, (2) in situ and real-time Raman spectroscopy analysis of the gas stream, and (3) onboard determination of molecular and isotopic compositions of the collected gas bubbles. The in situ Raman spectroscopy was used as a decision-making tool to evaluate the need for continuing with the sampling of gases from the discovered seep, or to move to another one. Push cores were gathered to study buried carbonates and pore waters at the surficial sediment, while CTD-Rosette allowed collecting samples to measure dissolved-methane concentration within the water column followed by a comparison with measurements from samples collected with the submersible Nautile during the Marnaut cruise in 2007.Overall, the visited sites were characterized by a wide diversity of seeps. CO2- and oil-rich seeps were found at the westernmost part of the sea in the Tekirdag Basin, while amphipods, anemones and coral populated the sites visited at the easternmost part in the Cinarcik Basin. Methane-derived authigenic carbonates and bacterial mats were widespread on the seafloor at all sites with variable size and distributions. The measured methane concentrations in the water column were up to 377 μmol, and the dissolved pore-water profiles indicated the occurrence of sulfate depleting processes accompanied with carbonate precipitation. The pore-water profiles display evidence of biogeochemical transformations leading to the fast depletion of seawater sulfate within the first 25-cm depth of the sediment. These results show that the North Anatolian Fault and inherited faults are important migration paths for fluids for which a significant part is discharged into the water column, contributing to the increase of methane concentration at the bottom seawater and favoring the development of specific ecosystems.

The isotope composition of boron (B) in marine biogenic carbonates has been predominantly studied as a proxy for monitoring past changes in seawater pH and carbonate chemistry. In order to derive seawater pH from boron isotope ratio data, a number of assumptions related to chemical kinetics and themodynamic isotope exchange reactions are necessary. Furthermore, the boron isotope composition (δ 11 B) of biogenic carbonates (δ 11 B CaCO3 ) is assumed to reflect the δ 11 B of dissolved borate (B(OH) 4 − ) in seawater. Here we report the development of methodology for measuring the δ 11 B in biogenic carbonate samples at the multi-collector inductively coupled mass spectrometry facility at Ifremer (Plouzane, France) and the evaluation of δ 11 B CaCO3 in a diverse range of marine calcifying organisms. We evaluated the δ 11 B CaCO3 of 6 species of marine calcifiers (a temperate coral, Oculina arbuscula ; a coralline red alga, Neogoniolithion sp.; a tropical urchin, Eucidaris tribuloides ; a temperate urchin, Arbacia punctulata ; a serpulid worm, Hydroides crucigera ; and an American oyster, Crassostrea virginica ) that were reared for 60 days in isothermal seawater (25 °C) equilibrated with an atmospheric p CO 2 of ca. 409 μatm. We observe large inter-species variability in δ 11 B CaCO3 (ca. 20 ‰) and significant discrepancies between measured δ 11 B CaCO3 and δ 11 B CaCO3 expected from established relationships between δ 11 B CaCO3 and seawater pH. We discuss these results in the context of various proposed mechanisms of biocalcification, including the potential dominant role that internal calcifying site pH plays in regulating CaCO 3 saturation state and borate δ 11 B at the site of calcification and, thus, the δ 11 B composition of calcifers’ shells and skeletons.

International audience; Marine sediments at ocean margins vent substantial amounts of methane. Microbial oxidation of the methane released can trigger the precipitation of carbonate within sediments and support a broad diversity of seafloor ecosystems. The factors controlling microbial activity and carbonate precipitation associated with the seepage of submarine fluid over geological time remain poorly constrained. Here, we characterize the petrology and geochemistry of rocks sampled from metre-size build-ups of methane-derived carbonate chimneys located at the Amon mud volcano on the Nile deep-sea fan. We find that these carbonates comprise porous structures composed of aggregated spherules of aragonite, and closely resemble microbial carbonate reefs forming at present in the anoxic bottom waters of the Black Sea. Using U-series dating, we show that the Amon carbonate build-ups formed between 12 and 7 thousand years ago, contemporaneous with the deposition of organic-rich sediments in the eastern Mediterranean, the so-called sapropel layer S1. We propose that the onset of deep-water suboxic or anoxic conditions associated with sapropel formation resulted in the development of intense anaerobic microbial activity at the sea floor, and thus the formation of carbonate chimneys.

Ferromanganese (Fe-Mn) crusts are potential archive of the Ni isotope composition of seawater through time. In this study we aim at (1) understanding Ni isotope fractionation mechanisms and metal enrichment processes in Fe-Mn deposits, (2) addressing global vs. local control of Ni isotope composition of these deposits. Two Fe-Mn crusts from the North Pacific Ocean (Apuupuu Seamount, Hawaii) and two Fe-Mn crusts from the South Pacific Ocean (near Rurutu Island, Austral archipelago of French Polynesia) were characterized for their elemental geochemistry and Ni isotope composition. Geochemical analyses were performed at millimeter intervals in order to provide time-resolved record of Ni isotopes. Chronology and growth rates were determined using cosmogenic 10Be isotope abundances. The results show that, despite different growth rates, textures and geochemical patterns, Fe-Mn crusts from both North and South Pacific Oceans have fairly homogenous Ni isotope compositions over the last ∼17 Ma, yielding average δ60/58Ni values of 1.79 ± 0.21‰ (2sd, n = 31) and 1.73 ± 0.21‰ (2sd, n = 21) respectively. In one crust sample, however, layers directly in contact with the altered substrate show anomalously light δ60/58Ni values down to 0.25 ± 0.05‰ (2se) together with rejuvenated 10Be/9Be ratios correlating with elevated Ni/Mn ratios. Such patterns are best explained by protracted fluid–rock interactions leading to alteration of Mn-phases after crust formation. Isotopically light Ni would be the result of Ni isotope fractionation during adsorption rather than the contribution of external Ni sources (e.g. hydrothermal sources) having light Ni isotope compositions. The combination of our results with previously published data on Fe-Mn crusts indicates that the average Ni isotope composition in deep waters has not changed through the Cenozoic (∼70 Ma). We propose that Ni isotope variations in Fe-Mn crusts may not only record variations of Ni sources to the oceans, but also post-depositional processes depending on the growth history and geological settings of Fe-Mn crusts.

The hafnium and neodymium radiogenic isotope systems behave differently during Earth surface processes, causing a wide dispersion of Hf and Nd isotopic compositions in sediments and other sedimentary rocks. The decoupling between Hf and Nd isotopes in sediments is generally attributed to a combination of preferential sorting of zircon during sediment transport and incongruent weathering processes on continents. In this study, we analysed size-fractions of sediment samples collected near the mouth of 53 rivers worldwide to better understand the factors controlling the distribution of Hf and Nd isotopes in sediments. Our results for rivers draining old cratonic areas and volcanic provinces demonstrate that both granite and basalt weathering can lead to significant grain-size dependent Hf isotopic variability. While silt-size fractions mainly plot along the Terrestrial Array, World river clays are systematically shifted towards more radiogenic Hf isotopic compositions, defining together with published data a new Clay Array ( e Hf = 0.78 × e Nd + 5.23 ). The Hf–Nd isotope decoupling observed in volcanogenic sediments is best explained by selective alteration of Lu-rich mineral phases (e.g. olivine) and preferential enrichment of resistant unradiogenic minerals, such as spinel and ilmenite, in silt fractions. We also show that the extent to which World river clays deviate from the Clay Array ( Δ e Hf clay ) is not linked to the presence of zircons. Instead, it correlates positively with weathering indices and climatic parameters (temperature, rainfall) of the corresponding drainage basins. Overall, these findings demonstrate that the distribution of Hf–Nd isotopes in clay-size sediments is related to a large extent to weathering conditions on continents, although the precise mechanisms controlling this relationship remain unclear. We finally propose that the Hf–Nd isotope pair proxy could be used in palaeoenvironmental studies to provide semi-quantitative information on past climates.

Environmental transport of Tl is affected by redox reaction between Tl(I) and Tl(III) and ligand exchange reactions of them. In order to deepen the knowledge of Tl chemistry, we investigated fractionation of Tl stable isotopes (Tl-203 and Tl-205) in a chemical exchange system. Tl isotopes were fractionated in a liquid-liquid extraction system, in which aqueous and organic phases are hydrochloric acid solution and dichloroethane including a crown ether, respectively. After purification by ion-exchange chemistry, the isotope ratio of Tl-205/Tl-203 in equilibrated aqueous phase was measured precisely by multiple-collector-inductively-coupled-plasma-mass-spectrometry. A large isotope fractionation > 1 aEuro degrees was found. Electronic structures of possible Tl species (hydrated Tl+, Tl3+, and Tl chlorides) were calculated by ab initio methods, and the isotope fractionation factor was theoretically obtained. The isotope fractionation via intramolecular vibrations was calculated to be much smaller than the experimental result. The isotope fractionation via isotopic change in nuclear volume, named the nuclear field shift effect, was calculated to be > 1 aEuro degrees in Tl(I)-Tl(III) redox systems and/or ligand exchange systems of Tl(III). The nuclear field shift effect was found to be the major origin of Tl isotope fractionation.

The metrological compatibility of n ( 234 U)/ n ( 238 U), n ( 235 U)/ n ( 238 U) and n ( 236 U)/ n ( 238 U) isotope amount ratio measurement results obtained by gas source mass spectrometry (GSMS), thermal ionisation mass spectrometry (TIMS) and multi-collector inductively coupled plasma mass spectrometry (MC-ICPMS) techniques in a set of low-enriched uranium samples is demonstrated in this study. The impact of the correlation between certified isotope reference materials on the metrological compatibility of measured isotope amount ratios is also discussed. The metrological concepts defined in the VIM 3rd edition and the measurement uncertainty evaluation according to the BIPM-GUM guide was thoroughly applied in this work.

This paper presents the investigation of strontium isotope ratios of about 650 different European natural mineral waters as part of the food traceability project “TRACE” funded by the EU. Analysed 87Sr/86Sr values in the natural mineral waters range from 0.7035 to 0.7777, which indicates an influence by a great diversity of rocks from young mantle derived basaltic rocks to very old silicic continental crust. The results of the large scale investigation are used to elaborate a novel spatial prediction for strontium isotope ratios by combining the measured data with a GIS based geological map of Europe. The resulting map can be used to predict the strontium isotopic composition of groundwater and thus the composition of bio-available strontium, which is available for uptake by plants and subsequently transferred into the food chain. We also show, as an example, that the strontium isotopic composition of honey and wheat from specific sample regions within the TRACE project correlates well with that of the natural mineral water as predicted by our map. The proof of principle shown in our paper is highly relevant for geographical food authentication as it allows an assessment of the origin of food products without the immediate need for geographically authenticated materials which may not always be available. Our approach provides a cost effective first instance screening tool.

A robust method allowing the routine determination of n(87Sr)/n(86Sr) with at least five significant decimal digits for large sets of mineral water samples is described. It is based on 2 consecutive chromatographic separations of Sr associated to multi-collector inductively coupled plasma mass spectrometry (MC-ICPMS) measurements. Separations are performed using commercial pre-packed columns filled with “Sr resin” to overcome isobaric interferences affecting the determination of strontium isotope ratios. The careful method validation scheme applied is described. It included investigations on all parameters influencing both chromatographic separations and MC-ICPMS measurements, and also the test on a synthetic sample made of an aliquot of the NIST SRM 987 certified reference material dispersed in a saline matrix to mimic complex samples. Correction for mass discrimination was done internally using the n(88Sr)/n(86Sr) ratio. For comparing mineral waters originating from different geological backgrounds or identifying counterfeits, calculations involved the well known consensus value (1/0.1194) ± 0 as reference. The typical uncertainty budget estimated for these results was 40 ‘ppm’ relative (k = 2). It increased to 150 ‘ppm’ (k = 2) for the establishment of stand alone results, taking into account a relative difference of about 126 ‘ppm’ systematically observed between measured and certified values of the NIST SRM 987. In case there was suspicion of a deviation of the n(88Sr)/n(86Sr) ratio (worst case scenario) our proposal was to use the NIST SRM 987 value 8.37861 ± 0.00325 (k = 2) as reference, and assign a typical relative uncertainty budget of 300 ‘ppm’ (k = 2). This method is thus fit for purpose and was applied to eleven French samples.

Nickel isotope ratios were measured in ores, fly ash, slags and FeNi samples from two metallurgical plants located in the Goias State, Brazil (Barro Alto, Niquelândia). This allowed investigating the mass-dependent fractionation of Ni isotopes during the Ni-laterite ore smelting and refining. Feeding material exhibits a large range of δ 60 Ni values (from 0.02 ± 0.10‰ to 0.20 ± 0.05‰, n = 7), explained by the diversity of Ni-bearing phases, and the average of δ 60 Ni feeding materials was found equal to 0.08 ± 0.08‰ (2SD, n = 7). Both δ 60 Ni values of fly ash (δ 60 Ni = 0.07 ± 0.07‰, n = 10) and final FeNi produced (0.05 ± 0.02‰, n = 2) were not significantly different from the feeding materials ones. These values are consistent with the very high production yield of the factories. However, smelting slags present the heaviest δ 60 Ni values of all the smelter samples, with δ 60 Ni ranging from 0.11 ± 0.05‰ to 0.27 ± 0.05‰ (n = 8). Soils were also collected near and far from the Niquelândia metallurgical plant, to evaluate the potential of Ni isotopes for tracing the natural vs anthropogenic Ni in soils. The Ni isotopic composition of the non-impacted topsoils developed on ultramafic rocks ranges from −0.26 ± 0.09‰ to −0.04 ± 0.05‰ (n = 20). On the contrary, the Ni isotopic composition of the non-ultramafic topsoils, collected close to the plant, exhibit a large variation of δ 60 Ni, ranging from −0.19 ± 0.13‰ up to 0.10 ± 0.05‰ (n = 4). This slight but significant enrichment in heavy isotopes highlights the potential impact of smelting activity in the surrounding area, as well as the potential of Ni isotopes for discerning anthropogenic samples (heavier δ 60 Ni values) from natural ones (lighter δ 60 Ni values). However, given the global range of published δ 60 Ni values (from −1.03 to 2.5‰) and more particularly those associated to natural weathering of ultramafic rocks (from −0.61 to 0.32‰), the use of Ni isotopes for tracing environmental contamination from smelters will remain challenging.

International audience; Over the past decades, rare earth elements (REE) and their radioactive isotopes have received tremendous attention in sedimentary geochemistry, as tracers for the geological history of the continental crust and provenance studies. In this study, we report on elemental concentrations and neodymium (Nd) isotopic compositions for a large number of sediments collected near the mouth of rivers worldwide, including some of the world's major rivers. Sediments were leached for removal of non-detrital components, and both clay and silt fractions were retained for separate geochemical analyses. Our aim was to re-examine, at the scale of a large systematic survey, whether or not REE and Nd isotopes could be fractionated during Earth surface processes. Our results confirmed earlier assumptions that river sediments do not generally exhibit any significant grain-size dependent Nd isotopic variability. Most sediments from rivers draining old cratonic areas, sedimentary systems and volcanic provinces displayed similar Nd isotopic signatures in both clay and silt fractions, with Delta epsilon Nd(clay-silt) < vertical bar 1 vertical bar. A subtle decoupling of Nd isotopes between clays and silts was identified however in a few major river systems (e.g. Nile, Mississippi, Fraser), with clays being systematically shifted towards more radiogenic values. This observation suggests that preferential weathering of volcanic and/or sedimentary rocks relative to more resistant lithologies may occur in river basins, possibly leading locally to Nd isotopic decoupling between different size fractions. Except for volcanogenic sediments, silt fractions generally displayed homogeneous REE concentrations, exhibiting relatively flat shale-normalized patterns. However, clay fractions were almost systematically characterized by a progressive enrichment from the heavy to the light REE and a positive europium (Eu) anomaly. In agreement with results from previous soil investigations, the observed REE fractionation between clays and silts is probably best explained by preferential alteration of feldspars and/or accessory mineral phases. Importantly, this finding clearly indicates that silicate weathering can lead to decoupling of REE between different grain-size fractions, with implications for sediment provenance studies. Finally, we propose a set of values for a World River Average Clay (WRAC) and Average Silt (WRAS), which provide new estimates for the average composition of the weathered and eroded upper continental crust, respectively, and could be used for future comparison purposes. (C) 2015 Elsevier Ltd. All rights reserved.

Boron and Li are light, incompatible elements that preferentially partition into the liquid phase, whether melt or aqueous fluid, and thus are useful for tracking fluid-related processes in rocks. Most of the Li isotopic data presently available on subduction-related rocks are from whole-rock analyses; and the B isotopic analyses of subduction material have been carried out either on whole-rocks or in-situ on an accessory phase, such as tourmaline. The new method presented here couples an ESI New Wave UP-193-FX ArF* (193 nm) excimer laser-ablation microscope with a Neptune Plus (Thermo Scientific) MC-ICP-MS aiming to measure both Li and B isotopes in situ with good spatial resolution (metamorphic minerals are commonly chemically zoned, and whole-rock analyses lose this detail). The data thus obtained are compared with SIMS analyses on the same mineral samples for B, and with MC-ICP-MS analyses on whole-rock or mineral separates from the same sample for Li. Additionally, data acquired on tourmaline standards were compared to SIMS values. The results show that for B concentrations above 5 μg/g, the data obtained by LA-MC-ICP-MS and by SIMS are identical within error, for mica (phengitic muscovite), pyroxene (jadeite), serpentine (antigorite), and tourmaline. For Li concentrations above 10 μg/g, the data obtained by LA-MC-ICP-MS and by MC-ICP-MS are also identical, within error, for mica (phengitic muscovite), and pyroxene (jadeite). However, analyses of tourmaline standards have shown significant differences with reference values, so LA-MC-ICP-MS does not yet appear to be an appropriate method to analyze Li isotopes in tourmalines. Thus, LA-MC-ICP-MS is a suitable method to measure Li and B isotopes with good spatial resolution in major rock-forming silicates from subduction-related rocks where concentrations exceed 10 μg/g and 5 μg/g, respectively, with an error on individual measurements equal to or less than previously used methods, but obtainable in a significantly shorter amount of time. The external reproducibility is ± 2.88 to 3.31‰ for B and ± 1.50 to 1.75‰ for Li, which is lower than or equal to the variations encountered within a given chemically zoned sample (up to 10‰ of variation within a given natural sample).

We report rare earth element (REE) and neodymium (Nd) isotope data for the organic fraction of sediments collected from various depositional environments, i.e. rivers (n = 25), estuaries (n = 18), open-ocean settings (n = 15), and cold seeps (n = 12). Sedimentary organic matter (SOM) was extracted using a mixed hydrogen peroxide/nitric acid solution (20%-H2O2–0.02 M-HNO3), after removal of carbonate and oxy-hydroxide phases with dilute hydrochloric acid (0.25 M-HCl). A series of experimental tests indicate that extraction of sedimentary organic compounds using H2O2 may be complicated occasionally by partial dissolution of sulphide minerals and residual carbonates. However, this contamination is expected to be minor for REE because measured concentrations in H2O2 leachates are about two-orders of magnitude higher than in the above mentioned phases. The mean REE concentrations determined in the H2O2 leachates for samples from rivers, estuaries, coastal seas and open-ocean settings yield relatively similar levels, with ΣREE = 109 ± 86 ppm (mean ± s; n = 58). The organic fractions leached from cold seep sediments display even higher concentration levels (285 ± 150 ppm; mean ± s; n = 12). The H2O2 leachates for most sediments exhibit remarkably similar shale-normalized REE patterns, all characterized by a mid-REE enrichment compared to the other REE. This suggests that the distribution of REE in leached sedimentary organic phases is controlled primarily by biogeochemical processes, rather than by the composition of the source from which they derive (e.g. pore, river or sea-water). The Nd isotopic compositions for organic phases leached from river sediments are very similar to those for the corresponding detrital fractions. In contrast, the SOM extracted from marine sediments display eNd values that typically range between the eNd signatures for terrestrial organic matter (inferred from the analysis of the sedimentary detrital fractions) and marine organic matter (inferred from the analysis of local surface seawater). A notable exception is the case of organic matter (OM) fractions leached from cold seep sediment samples, which sometimes exhibit eNd values markedly different from both terrigenous and surface seawater signatures. This suggests that a significant fraction of organic compounds in these sediments may be derived from chemosynthetic processes, recycling pore water REE characterized by a distinct isotopic composition. Overall, our results confirm that organic matter probably plays an important role in the oceanic REE budget, through direct scavenging and remineralization within the water column. Both the high REE abundances and the shape of shale-normalized patterns for leached SOM also suggest that OM degradation in sub-surface marine sediments during early diagenesis could control, to a large extent, the distribution of REE in pore waters. Benthic fluxes of organic-bound REE could hence substantially contribute to the exchange processes between particulates and seawater that take place at ocean margins. Neodymium isotopes could provide useful information for tracing the origin (terrestrial versus marine) and geographical provenance of organic matter, with potential applications in paleoceanography. In particular, future studies should further investigate the potential of Nd isotopes in organic compounds preserved in sedimentary records for reconstructing past variations of surface ocean circulation.

International audience; The Paleoproterozoic Era was a time of remarkable importance in the redox evolution of Earth's atmosphere and oceans. Here, we present a multi-proxy study of Mo, U and Fe isotopes together with Fe speciation of black shales and siltstones from the upper Zaonega Formation of the Onega Basin in Karelia. We attempt to better understand oceanic redox conditions during the 2.05 Ga Shunga Event as the next step following the Great Oxidation Event (GOE) and the Lomagundi carbon isotope excursion Event. A cautious examination of the Fe speciation data shows that the studied section was deposited under dominantly euxinic conditions (anoxic and sulfidic) and that the lower part of the section experienced metamorphism through which pyrite was altered to pyrrhotite. During this episode, the system was closed with respect to Fe but not sulfur. The Mo and U isotopic compositions (corrected for detrital input) were not affected by the metamorphism and loss of S and are fairly uniform throughout the entire section. The Fe isotope compositions are exceptionally heavy in the lower part of the section (up to δ56FeIRMM-14 = 0.83‰) and become lighter towards the upper intervals, which also show significant [Mo] and [U] enrichments. We suggest that this pattern reflects changes in the position of the deposition site relative to the redox structure of the water column. The upper part was deposited within a locally euxinic portion of the basin where H2S availability was highest, removal of Mo and U was more efficient and precipitated pyrite captured relatively non-fractionated dissolved Fe. In other words, quantitative uptake of Fe was favored. In contrast, the lower interval was deposited on the lower margin of a euxinic wedge where H2S availability was lower, and removal of Mo and U was less efficient. Pyrite precipitation in this part of the water column reflected a more fractionated dissolved Fe reservoir due to more protracted, non-quantitative Fe uptake because of less efficient pyrite formation under lower sulfide conditions and greater access to the large oceanic pool of Fe. The U isotopic signal was corrected for detrital contribution giving compositions similar to the riverine input and suggesting that co-precipitation into carbonates was the main process of U removal at this time. We estimate the Mo isotope composition of the contemporaneous ocean to be δ98MoSW = 0.85 ± 0.21‰. This is the lowest value yet reported for the Proterozoic ocean, suggesting that the oceanic Mo cycle was dominated by euxinic and anoxic sinks with negligible Mo removal into oxic environments. Recent studies have proposed a sharp increase in ocean-atmosphere oxygen levels during the Lomagundi Event followed by a dramatic crash. Our results from black shales of the 2.05 Ga Shunga Event are consistent with a post-Lomagundi decrease in biospheric oxygen levels.

The factors controlling fluid emission dynamics at ocean margins are poorly understood. In particular, there are significant uncertainties on how fluid seepage at cold seeps may have responded to abrupt environmental changes in the geological past. This study reports on a detailed geochemical investigation of seafloor carbonate crusts sampled at cold seeps along the submerged part of the North Anatolian Fault system in the Sea of Marmara – an inland sea, which has experienced major paleo-environmental changes over the last deglaciation period. We also analyzed a series of authigenic carbonate concretions recovered from two sediment cores at the Western-High ridge, an active fluid venting area. The ages of seafloor carbonate crusts derived from isochron U–Th dating cover the last 7 kyr, suggesting that fluid activity along the fault system remained continuous over that time interval. In the sediment cores, carbonate concretions are concentrated at the lacustrine-to-marine transition, which corresponds to the period when Mediterranean waters flowed into the Marmara Basin about 12–14 kyr ago. U–Th isotopic data indicate that most of these concretions formed later during the Holocene, around 9–10 kyr ago, a period coinciding with an important anoxic event that led to the deposition of a sapropel layer in the Sea of Marmara. Based upon these results, we suggest that the absence of carbonate concretions in the lacustrine sediment unit indicates that dissolved sulfate concentrations in the Marmara lake pore waters during glacial time were too low to promote significant anaerobic methane oxidation, thereby preventing sedimentary carbonate authigenesis. In contrast, the progressive inflow of Mediterranean waters into the glacial Marmara lake after 15 ka provided a source of dissolved sulfate that allowed anaerobic oxidation of methane to proceed within the anoxic sediment. Importantly, the synchronism between the main phase of authigenic carbonate precipitation at the studied sites (average 9.4 ± 1.8 ka , n = 16 ) and the regional anoxic sapropel event support the idea that the drop in bottom water dissolved oxygen content was probably a key factor to enhance microbial activity and associated carbonate precipitation at that time. Overall, these results provide straightforward evidence that fluid emission dynamics and hydrocarbon oxidation at cold seeps can be directly related to changing environmental conditions through time.

Although initial studies have demonstrated the applicability of Ni isotopes for cosmochemistry and as a potential biosignature, the Ni isotope composition of terrestrial igneous and sedimentary rocks, and ore deposits remains poorly known. Our contribution is fourfold: (a) to detail an analytical procedure for Ni isotope determination, (b) to determine the Ni isotope composition of various geological reference materials, (c) to assess the isotope composition of the Bulk Silicate Earth relative to the Ni isotope reference material NIST SRM 986 and (d) to report the range of mass-dependent Ni isotope fractionations in magmatic rocks and ore deposits. After purification through a two-stage chromatography procedure, Ni isotope ratios were measured by MC-ICP-MS and were corrected for instrumental mass bias using a double-spike correction method. Measurement precision (two standard error of the mean) was between 0.02 and 0.04 parts per thousand, and intermediate measurement precision for NIST SRM 986 was 0.05 parts per thousand (2s). Igneous- and mantle-derived rocks displayed a restricted range of Ni-60/58 values between -0.13 and +0.16 parts per thousand, suggesting an average BSE composition of +0.05 parts per thousand. Manganese nodules (Nod A1; P1), shale (SDO-1), coal (CLB-1) and a metal-contaminated soil (NIST SRM 2711) showed positive values ranging between +0.14 and +1.06 parts per thousand, whereas komatiite-hosted Ni-rich sulfides varied from -0.10 to -1.03 parts per thousand., Bien que les premières études démontrent la pertinence des isotopes du nickel en cosmochimie et en tant que signature biologique, la composition isotopique du nickel des roches ignées et sédimentaires terrestres ainsi que celle des dépôts de minerais est encore très peu connue. Notre contribution s'organise en quatre axes, (a) détailler la procédure analytique pour la détermination des compositions isotopiques en nickel, (b) déterminer la composition isotopique de matériaux géologiques de référence variés, (c) estimer la composition isotopique de la Terre Silicatée Globale (BSE) par rapport au standard isotopique de référence de nickel NIST SRM 986, et (d) reporter la gamme des fractionnements isotopiques du nickel dépendants de la masse dans les roches magmatiques et les dépôts de minerais. Après purification suivant une procédure de chromatographie en deux étapes, les rapports isotopiques du nickel ont été mesurés par MC-ICP-MS puis corrigés du biais de masse instrumental par la méthode du double-spike. La fidélité de nos mesures (erreur standard de la moyenne) est comprise entre 0.02 et 0.04‰, et la précision de mesure intermédiaire pour le NIST SRM 986 est de 0.05‰ (2s). Les roches ignées et mantelliques montrent une gamme restreinte de valeurs isotopiques δ60/58Ni entre −0.13 et +0.16‰, ce qui suggère une composition moyenne de la BSE à +0.05‰. Les nodules de manganèse (Nod A1; P1), le shale (SDO-1), le charbon (CLB-1) et le sol contaminé en métaux (NIST SRM 271) donnent des valeurs δ60/58Ni positives comprises entre +0.14 et +1.06‰, tandis que les sulfures riches en nickel présents dans les komatiites ont des valeurs négatives allant de −0.10 à −1.03‰.

Continental ice sheets are a key component of the Earth’s climate system, but their internal dynamics need to be further studied. Since the last deglaciation, the northern Eurasian Fennoscandian Ice Sheet (FIS) has been connected to the Black Sea (BS) watershed, making this basin a suitable location to investigate former ice-sheet dynamics. Here, from a core retrieved in the BS, we combine the use of neodymium isotopes, high-resolution elemental analysis, and biomarkers to trace changes in sediment provenance and river runoff. We reveal cyclic releases of meltwater originating from Lake Disna, a proglacial lake linked to the FIS during Heinrich Stadial 1. Regional interactions within the climate–lake–FIS system, linked to changes in the availability of subglacial water, led to abrupt drainage cycles of the FIS into the BS watershed. This phenomenon raised the BS water level by ∼100 m until the sill of the Bosphorus Strait was reached, flooding the vast northwestern BS shelf and deeply affecting the hydrology and circulation of the BS and, probably, of the Marmara and Aegean Seas.

The Amsterdam-St Paul (ASP) oceanic plateau results from the interaction between the ASP hot spot and the Southeast Indian ridge. A volcanic chain, named the Chain of the Dead Poets (CDP), lies to its northward tip and is related to the hot spot intraplate activity. The ASP plateau and CDP study reveals that ASP plume composition is inherited from oceanic crust and pelagic sediments recycled in the mantle through a 1.5 Ga subduction process. The ASP plateau lavas have a composition (major and trace elements and Sr-Nd-Pb-Hf isotopes) reflecting the interaction between ASP plume and the Indian MORB mantle, including some clear DUPAL input. The Indian upper mantle below ASP plateau is heterogeneous and made of a depleted mantle with lower continental crust (LCC) fragments probably delaminated during the Gondwana break-up. The lower continental crust is one of the possible reservoirs for the DUPAL anomaly origin that our data support. The range of magnitude of each end-member required in ASP plateau samples is (1) 45% to 75% of ASP plume and (2) 25% to 55% of Indian DM within 0% to a maximum of 6% of LCC layers included within. The three end-members involved (plume, upper mantle and lower continental crust) and their mixing in different proportions enhances the geochemical variability in the plateau lavas. Consequently, the apparent composition homogeneity of Amsterdam Island, an aerial summit of the plateau, may result from the presence of intermediate magmatic chambers into the plateau structure.

Neumann et al . argue that terrestrial evidence does not support our interpretation of large-scale human land use in Central Africa about 2500 years ago and that climate was the main driver of the rainforest crisis at that time, and Maley et al . raise a number of concerns about our interpretation of data from chemical weathering proxies. Taking into account existing palaeoclimatic data and clarifying some misconceptions, we reassert that humans must also have contributed fundamentally to this large vegetation change.