2i: Regional and local scale stable isotope studies of mercury biogeochemical cycling and bioaccumulation

Atmospheric Hg deposition to Polar Regions has been extensively studied in coastal zones with the goal to better understand the nature and relevance of AMDE's. Little attention has been given so far to the interior, continental Arctic. We will present a comprehensive Hg stable isotope mass balance of the arctic tundra made at Toolik Field station, AK and compare our findings with extensive terrestrial - atmosphere Hg0 flux measurements conducted at the same site. We measured stable Hg isotope signatures of all important atmospheric (Hg0, HgII in precipitation, HgII from AMDE's) and geogenic (bedrock) sources and major ecosystem reservoirs (vegetation, organic and mineral soils).

The results suggest that vegetation uptake of gaseous elemental mercury from the atmosphere is the dominant deposition pathway to the terrestrial ecosystem, and contributes about 70% to Hg in organic soils. We have observed large mass-independent Hg isotope anomalies in snow during spring, with minimal D199Hg values of -1.4‰, which are in agreement with previously observed Hg isotope signatures during the AMDE season at the Arctic coast. However, the transfer of HgII from AMDE's to the terrestrial ecosystem does not represent a significant source (0-5 %), suggesting that most HgII deposited during AMDE's was re-emitted prior to snow-melt. The Hg isotope signatures in the soils and the atmosphere show no indication for substantial re-emission of gaseous elemental mercury from the soils.

The Hg stable isotope results agree well with micro-metrological flux measurements conducted at the same location. Both independent approaches suggest that vegetation uptake of gaseous elemental Hg represents the dominant Hg flux between the atmosphere and Arctic tundra soils and they thus represent a net sink for atmospheric Hg.

The Huron-Erie Corridor (HEC) within the Great Lakes of North America is a large riverine connecting channel that flows 157 km through the St. Clair River (SCR), Lake St Clair (LSC), and the Detroit River (DR). The HEC is one of the most industrialized and environmentally impacted areas of the Great Lakes, with highly Hg-contaminated sediment located predominantly on the Canadian side of the SCR and on the US side of the DR. The distribution of Hg in sediment of the HEC is generally thought to be a function of localized anthropogenic inputs; however, tracking the transport and fate of contaminated sediments in large riverine ecosystems is challenging. We used stable Hg isotopes to fingerprint and assess the redistribution of sediment Hg hotspots. Sediment samples from the HEC had a large range in Hg concentration (30-6,760 ng/g) and isotopic composition (δ202Hg = -2.60 to -0.35‰, n=43). Mass independent fractionation was typically small (mean ∆199Hg = 0.06 ± 0.05‰, 1SD). Local background sediment (30 ± 1 ng/g Hg, 1SD, n=3) had a mean δ202Hg value of -1.59 ± 0.15‰ (1SD). Mercury-contaminated sediment from Canadian waters of the SCR (2,560-3,100 ng/g) displayed the most negative δ202Hg values (mean δ202Hg: -2.55 ± 0.07‰, 1SD, n=2), resulting from historic chlor-alkali operations. Within the upstream reach of the DR, sediment had moderately elevated Hg concentrations (150-250 ng/g, n=4) and was isotopically similar in Canadian and US waters (δ202Hg values from -2.03 to -1.49‰ and -1.89 to -1.65‰, ± 0.11‰, 2SD; respectively), with the exception of one sample from along the US shoreline (80 ng/g Hg, δ202Hg = -1.19 ± 0.11‰, 2SD), generally suggesting the influence of upstream inputs from SCR and LSC. Within the midstream and downstream DR, in contrast, the isotopic composition of Hg in sediment from US waters (δ202Hg = -1.53 to -0.43 ± 11‰, 2SD, n=15) was distinct from Canadian waters (δ202Hg = -1.94 to -1.60 ± 0.11‰, 2SD, n=11), excluding one outlier. Our results suggest a complex mosaic of localized mercury sources with limited cross-channel transport and demonstrates the utility of Hg isotopes to improve our understanding of contaminated sediment redistribution in large riverine ecosystems. Moreover, the large variation in isotopic composition among high Hg concentration sediment hotspots demonstrates that industrial processes can fractionate Hg prior to release, and that caution should be used when choosing mixing model end members to represent industrial releases on regional scales.

Lakes are important ecosystems where mercury (Hg) may be methylated into methylmercury (MMHg) which can pose serious threat on aquatic ecosystem through food web bioaccumulation. Therefore, it’s crucial to understand the sources and biogeochemical processes of Hg in lake systems. Recent studies demonstrated that mercury isotope composition is a powerful tool for tracing its behaviors in the environment (1). Nevertheless, limited by pre-treatment method of dilute Hg in natural water, few studies were carried out on Hg isotope compositions in lake systems (2), and weather the Hg isotopic fractionation could be triggered by in-lake processes still remain unclear. The forested lakes characterized by high TOC concentration in high latitude region are such important concern, because the stratification in water column caused by temperature gradient in summer and the decrease of illumination intensity downward (due to the high TOC content) may trigger different Hg isotopic fractionation.

In this study, samples from water columns of seven lakes in Finland and Sweden were collected to investigate the fractionation mechanisms of Hg isotopes. Preliminary results showed that the stratification phenomenon is obvious in these lakes in summer. A concomitant decrease of δ 202Hg and Δ199Hg values towards deeper water was observed, likely caused by the various isotopic fractionations induced by in-lake processes at different layers. In fact, the Δ199Hg/ Δ201Hg ratio in surface water is about 1.03, very close to that found in the photoreduction of Hg2+ (3), indicating that the Hg isotopic signature in surface water is potentially derived from photoreduction of Hg2+, with a possible contribution of rainwater. The Δ199Hg/ Δ201Hg slope in middle water is 1.19, between 1.00 of Hg2+ photoreduction and 1.36 of photo-degradation of MMHg, possibly suggesting a mixture of these two processes. Finally, the Δ199Hg/ Δ201Hg ratio in bottom water is around 1.64, similar to the Δ199Hg/ Δ201Hg (~1.6) introduced by dark reduction of Hg (4), suggesting a possible control of dark reduction. Thus, the application of mercury isotopes may be helpful for better understanding the biogeochemical cycle of Hg in complicated lake systems.

1) Blum J. D., Sherman L. S. and Johnson M. W., AREPS. 2014, 42, 249-269. 2) Chen J. B. et al., CG. 2016, 426, 33-44. 3) Bergquist, B. A. and Blum, J. D., Sci. 2007, 318, 417-420. 4) Zheng W. et al, JPCA 2010, 114, 4238-4245.

Historic point source mercury (Hg) inputs from industrial processes on the South River (Waynesboro, Virginia) ended many decades ago, however concentrations of Hg in the physical reservoirs of the South River remain elevated relative to regional background concentrations. To better understand Hg sources, mobility, and bioavailability in the South River, we analyzed THg concentrations and Hg stable isotope compositions of streambed sediments, bank soils, suspended particles, and filtered surface and bank pore waters. An end-member mixing model is proposed which identifies three isotopically distinct pools of Hg in the South River channel environment near the historic point source. In addition to isotopically distinct regional background (low THg conc., δ202Hg≅-1.22‰) and contamination-derived Hg sources (high THg conc., δ202Hg≅-0.59‰), a third end-member was identified with high THg concentrations and relatively low δ202Hg values (δ202Hg≅-1.00‰). To further elucidate the potential source of this previously unidentified Hg end-member and to constrain the end-member isotopic compositions, higher spatial resolution sampling was conducted of previously collected abiotic Hg reservoirs and of channel-margin hyporheic waters near the historic Hg point source. These analyses will better constrain the Hg isotope composition in the river reaches in which the unknown Hg end-member is present and may allow for determination of the processes from which the unknown Hg end-member originates. Additionally, high volume sampling of uncontaminated waters upstream of the former industrial facility was conducted to more accurately determine the Hg isotopic composition of the regional background end-members. Analysis of the Hg isotope composition of dated sediment cores from a Hg contamination impacted floodplain will further constrain the temporal variability of the Hg isotopic composition of legacy Hg released from the former industrial facility. By improving understanding about the spatial and temporal variations in Hg isotopic composition, this study aims to provide insights into the processes that control Hg mobility in the South River.

Even though Hg has now been phased out in most industrial applications, a large number of legacy sites exist that have been affected by historical Hg releases. Elevated Hg levels in soils and waters at these sites represent a serious threat for the environment at local and regional scales. The long-term fate, mobility, and bioavailability of Hg strongly depend on its speciation, which is determined by the initial Hg compound from the industrial contamination source as well as biogeochemical transformation processes after release into the environment. Understanding the governing processes and controls on Hg speciation in contaminated subsurface environments is essential for risk assessment and site management.

Transformations of Hg species cause mass-dependent and mass-independent fractionation of stable Hg isotopes. Thus, variations in Hg isotope signatures may help identifying and quantifying transformation processes of Hg species. Previous studies at contaminated sites have largely focused on Hg isotopes as source tracer, whereas the application as process tracer has not been sufficiently explored yet.

Here, we present Hg speciation and isotope data from soil and groundwater collected at a former industrial facility in SW-Germany where wood was preserved by HgCl₂ treatment (kyanization). Up to 25 tons of HgCl₂ have been released to soil and aquifer during operation, resulting in a groundwater contamination plume of 1.3 km length. However, fast flowing groundwater and high Hg concentrations contrast sharply with extremely slow plume movement. It is unclear whether sorption/desorption or species transformation processes control the strong Hg retardation in the subsurface.

Thermodesorption analyses indicated the presence of different Hg species and partial transformation of HgCl₂ to Hg⁰ in the contaminated soil layers (up to 635 µg g^-1 Hg). Isotope analyses by CV-MC-ICP-MS revealed significant d²⁰²Hg variations between sections of a soil core taken close to the contamination source and groundwater collected nearby. Most groundwater samples (up to 192 µg L^-1 Hg) were enriched in heavy Hg isotopes (up to +0.75‰) relative to NIST-3133. The most contaminated soil layers exhibited d²⁰²Hg values of -0.4‰, consistent with signatures of industrial Hg sources, whereas layers with lower Hg contents were isotopically heavier. Sequential extractions revealed significant variations between different Hg soil pools, e.g., isotopically heavy water-extractable Hg relative to bulk soil in the most contaminated layers. Our findings provide evidence for Hg isotope fractionation in the subsurface of contaminated sites and indicate the potential of Hg isotope signatures as tracer for species transformation processes controlling Hg mobility.

Researchers have frequently sought to use environmental archives of sediment, peat and glacial ice to try and assess historical trends in atmospheric mercury (Hg) deposition to aquatic ecosystems. While this information is valuable in the context of identifying temporal source trends, these types of assessments cannot account for likely changes in bioavailability of Hg sources that are tied to the formation of methylmercury (MeHg) and accumulation in fish tissues. For this study we propose the use of long-term fish archives and Hg stable isotope determination as an improved means to relate temporal changes in fish Hg levels to varying Hg sources in the Great Lakes. For this study we acquired 180 archived fish composites from Lake Michigan over a 40-year time period (1975 to 2014) from the Great Lakes Fish Monitoring and Surveillance Program, which were analyzed for their total Hg content and Hg isotope abundances. The results reveal that Hg sources to Lake Michigan trout (Salvelinus namaycush) have encountered considerable changes as well as a large shift in the food web trophic position as a result of the introduction of several invasive species, especially the recent invasion of dreissenid mussels. Total Hg concentrations span a large range (1,600 to 150 ng g-1) and exhibit large variations from 1975 to 1985. Δ199Hg signatures similarly exhibit large variation (3.2 to 6.9‰) until 1985, followed by less variation through the end of the data record in 2014. δ202Hg showed a relatively consistent value of about 0.4 – 0.8‰ from 1975 to 1988, followed by a significant (P<0.0001) shift of about 0.7‰ between 1989 and 1996, with relative stability through the end of the record (1.2 - 1.6‰). This δ202Hg shift may be linked to a combination of the contemporaneous implementation of Clear Air Act rules limiting Hg emissions from waste incineration and to changes in carbon cycling resulting from invasive mussels. Utilizing annual Hg stable isotope variability, we propose a dietary convergence in trout isotope values is observable following the onset of dreissenid invasion. We aim to independently confirm this observation using more traditional stable isotopes (C and N) as well as statistical modeling of isotopic shifts. We propose that with the combined use of Hg, C and N stable isotopes, we can begin to identify how Hg in fish have been affected by shifting sources portfolios of Hg and restructuring of food webs over the past 40 years in Lake Michigan.

Idrija mercury mine is world second largest with 500 years history and total production of 144.800 tons of mercury, which represents over 13% of the entire world production. During that time, 107.700 tons of commercial mercury have been sold and 37.100 tons of mercury were lost into the environment which is after 20 years of mine closure still persistent in local environment. Mercury is present in two main forms as cinnabar ore (70%) and as native Hg(0) (30%). Nowadays, the main sources of Hg in Idrija are: still active mine ventilation shafts, evaporation of Hg from the heavily polluted surroundings of the former smelting plant, mineralized rock dumps of primary or partially exploited ore, outcrops of the ore deposit, and ore residues treated in various ways. Novel methodologies and approaches are developed and used, based on identification of mercury isotope fractionation caused by different processes and sources of Hg during formation of Idrija mine in Triassic. This will enable identification of mercury isotope ratios (IR) characteristic for different environmental Hg sources and reconstruction of Hg mass balance in wider Idrija environment, and will significantly contribute to assess dynamics and transformations of mercury between former mining and smelting operations and local environment. The objective of this research was to study Hg isotopic compositions and evaluate Hg isotopic fractionation of different ores from the mercury mine, processed (retorted) ore, native elemental Hg present in the mine and appeared in near-by Idrijca river, and elemental Hg, produced in Idrija smelting plant. Analysis has been done on Cold Vapour Generator system (CVG) coupled with MC-ICP-MS (Nu plasma II, Nu instruments Ltd, UK). Initial measurements showed huge difference in δ202Hg values were obtained for elemental Hg, from -4.06 ± 0.19 ‰ in produced mercury to -0.74 ± 0.12 ‰ in the native mercury from the mine, whereas the native mercury from the Idrijca river presented δ202Hg value of -0.35 ± 0.15‰. Soil core samples from different locations in the mining district were analysed and δ202Hg values from -1.11 to 0.26 ‰ (2SD = 0.20 ‰) are presented. Hg isotope fingerprints in core samples close to the smelting plant show a big difference in comparison with other cores. The results, which give new information about Hg sources in Idrija mining region, will be presented.

French Guiana is subjected to intense gold-mining since 1850. Amalgamation of gold with liquid Hg(0) is the main exploitation process and, once in the environment, Hg is subjected to various biogeochemical transformations leading to its methylation in monomethylmercury (MMHg). This study aims to determine the influence of gold-mining activities on fish Hg contamination. Fish constitutes the main source of proteins of indigenous populations and their high MMHg concentrations could impact human health. We studied total concentration, speciation and stable isotopes fractionation of Hg in fish of both mining (Camopi) and pristine (Trois Sauts) areas along the Oyapock River and tributaries, in French Guiana.

In six fish piscivorous and periphytophageous species studied, a bioaccumulation of Hg along the trophic chain was clearly shown, with average THg concentrations varying from 20 ng.g-1 d.w. in herbivorous species to up to 1750 ng.g-1 d.w. in piscivorous species for the whole Oyapock R. basin. In the mining area, average THg concentrations in piscivorous species (2450 ng.g-1 d.w.) are significantly higher than in the pristine area (1570 ng.g-1 d.w.) while no significant difference is noticed for periphytophageous species. Speciation analysis showed that Hg in flesh was mainly in the organic form even in periphytophageous fish, representing 75% to 90% of the THg. Analyses of N and C stable isotopes also revealed differences between the two locations: two species (one herbivorous and one periphytophageous) presented different trophic level between Camopi and Trois Sauts, d13C being slightly higher in the gold-mining area than in the pristine one. Consequently, diet of low trophic levels fish species seemed different between both locations. Hg sources in fish could be discriminated with Hg isotope signatures only in piscivorous species. Anomalies D199Hg vs D201Hg linear regression showed a slope of 0.89 at Camopi and 1.30 at Trois Sauts meaning that at Camopi, inorganic Hg undergone photoreduction before methylation and ingestion by piscivorous fish. d13C is correlated with D201Hg in piscivorous fish showing that Hg in Camopi river has different sources than in the Oyapock R. upstream. The positive shift observed in D201Hg in piscivorous (+0.12 ‰) fish species between the pristine and the gold mining area suggests another Hg source in the hydrosystem possibly linked to the gold-mining activities. Periphytophageous species have D201Hg and d202Hg higher than the ones of piscivorous species reflecting the difference of Hg sources inside the trophic chain. The Hg speciation and stable isotopes fractionation study in the aquatic food chain in French Guiana is relevant to limit the Hg exposure of native population.