2a: Sources and cycling of mercury in coastal ecosystems

Communities in east Hudson Bay and James Bay (Canada) are concerned about ecosystem changes observed in recent decades, particularly related to sea-ice and oceanographic conditions, and also about potential impacts of contaminants from long-range atmospheric transport and regional human activities. The Arctic Eider Societys Community-Driven Research Network (CDRN) has been established to measure and better understand large-scale cumulative environmental impacts in that marine region. Building on CDRN collaborations and activities in five communities (Sanikiluaq, Kuujjuaraapik, Inukjuak, Umiujaq, Chisasibi), this project funded by the Northern Contaminants Program is generating new information on metal bioaccumulation that will provide a regionally-integrated perspective on metal exposure in the marine environment. The five communities are sampling coastal bioindicator species (blue mussel, common eider) annually for three years (2015 to 2017). Offshore bioindicators (ringed seal, herring gull, plankton, fish) are additionally being collected from two of the communtiies. These locally-important bioindicators of metal accumulation will be used to characterize geographic and habitat-specific variation (coastal and offshore zones) in east Hudson Bay and James Bay.

This poster will present preliminary results from collections in 2015 and 2016 of herring gull eggs, ringed seals, blue mussels, and common eiders. Animal tissues were analyzed for total mercury, methylmercury (on a subset of samples), 24 elements by ICP-MS, and carbon and nitrogen stable isotopes. Spatial variation of tissue mercury concentrations in east Hudson Bay and James Bay will be examined. Food web structure has been characterized using carbon and nitrogen stable isotope ratios, and trophic position and carbon source were found to affect mercury bioaccumulation. Higher mercury levels in blue mussels were associated with bioaccumulation of terrigenous and redox-sensitive elements (e.g., barium, uranium, vanadium, iron, manganese), suggesting that large river inputs may affect mercury bioaccumulation near the base of the food chain in coastal environments. This three-year project will generate important baseline information on mercury levels in the food web to allow for future tracking of impacts from environmental change, long-range atmospheric transport, and regional human activities on this arctic marine ecosystem.

The purpose of this study was to investigate the input and distribution of total mercury (Hg) in the leaf litter fraction and its importance to the sediment of mangrove using the elemental and isotopic composition of carbon and nitrogen to characterize the organic matter in the estuarine area of the Paraíba do Sul river, Southern Brazil. The average concentrations of Hg in sediment were 81 ± 32 ng.g-1 Site 1 (Laguncularia racemosa), 62 ± 15 ng.g-1 Site 2 (Rhizophora mangle) and 110 ± 16 ng.g-1 Site 3 (Avicennia germinans). The carbon isotopic composition (δ13C) in the sediment ranged from -29.40 to -26.49 ‰ and nitrogen (δ15N) from 2.38 to 5.83 ‰. The Hg concentration in annual litter fall were: 21 ± 2 ng.g-1;16 ± 4 μg/m2/ano; 18 ± 1 ng/g-1; 17 ± 3 μg/m2/ano-1; 53 ± 4 ng.g-1; 33 ± 4 μg/m2/ano for Laguncularia racemosa, Rizophora mangle and Avicennia germinans, respectively. The isotopic composition in the litter varied from -28.59 to -26.91 ‰ for δ13C and 4.48 to 7.21 ‰ for δ15N. The elemental and isotopic data in the sediment reflected the values ​​of the dominant plant species at each sampling site. The results show that the dominant species (A. germinans) stands for higher Hg values ​​in the sediments as well as the input for this element via litter. Since Hg in mangrove sediment has a low mobility, our results are suggesting that the atmosphere deposition has been playing an important role in Hg biogeochemical cycle in mangrove ecosystems through atmospheric deposition under the canopy. The Hg input data obtained from the litter, although lower compared to the closest point sources of contamination areas, fall into the mangrove ecosystem as an important integrator for Hg in the air, vegetation and sediment.

Phytoplankton are the typical entry point for mercury in marine food webs, and subsequent mercury bioaccumulation in fish is the principal route of human exposure. Although primary producers form this critical link, the biological and physical factors that influence mercury uptake by phytoplankton are not well understood, particularly in coastal lagoons where our previous work suggests elevated monomethylmercury (MeHg) production relative to coastal seawater. We are combining laboratory and in situ field incubation experiments to determine how cell size and salinity gradients affect mercury bioaccumulation in cultured diatoms (T. weissflogii and T. oceanica), and in natural phytoplankton populations. T. weissflogii (~17 µm length; ~11 µm width) and T. oceanica (~6.5 µm length; ~5 µm width) are centric diatoms with a consistent disk shape that allows comparison of their surface area to volume ratios. Lab cultures of these diatoms were acclimated to a salinity range of 10 to 30 to simulate fresh water and seawater mixing at the land-sea margin. These cultures were grown in ~10 and ~100 pM MeHg, in addition to non-spike controls, to assess the effects of cell size and salinity on MeHg uptake. The companion field incubation experiment was conducted at San Elijo Lagoon in Southern California. Lagoon water was filtered through a 200 µm mesh screen to eliminate zooplankton grazers, then poured into clear polycarbonate carboys and spiked to MeHg concentrations of ~10 and ~100 pM, in addition to no-spike controls. The carboys were incubated in the lagoon for 12 to 36 hours prior to filtration. Chlorophyll-a concentrations indicate that at the time of our study, the phytoplankton community was comprised of 15% picoplankton (0.2-2 µm), 55% nanoplankton (2-20 µm) and 30 % microplankton (20-200 µm). Over the first 24 hours of the incubation, chlorophyll-a in the carboys increased from approximately 6 µg/L to 10 µg/L, while nutrient concentrations decreased, suggesting the absence of grazers enabled a small phytoplankton bloom. This comparison of total mercury and MeHg among various phytoplankton communities will help us understand the drivers of mercury uptake in these understudied coastal lagoon ecosystems.

California’s coastal lagoons provide essential habitat for threatened and endangered species, yet eutrophication causes many lagoons to experience multiple algal blooms each year that could affect mercury (Hg) bioaccumulation in marine food webs. Specifically, anoxia associated with eutrophication favors monomethylmercury (CH3Hg+ or MeHg) production, and phytoplankton blooms provide a mechanism for MeHg entry into the food web. Because coastal lagoons are highly dynamic, we are evaluating both short-term (e.g., tidal) and long-term (e.g., seasonal) trends in mercury cycling. We conducted bimonthly sampling to measure chlorophyll-a, particulate organic carbon, and phytoplankton cell size (flow cytometry), along with the concentration of total Hg and MeHg in filtered (< 0.2 µm) and unfiltered water at San Elijo Lagoon in Southern California. Potential sources and transport mechanisms of MeHg were also evaluated by sampling shallow (< 3 m) groundwater and nearshore seawater at the lagoon mouth. Monitoring data show that dissolved oxygen in San Elijo Lagoon often fluctuates between > 100% (high tide) and < 30% (low tide), possibly due to the discharge of anoxic groundwater, which may be an important source of MeHg. This system also experiences extreme seasonal changes which are, in part, driven by the opening and closing of a sand berm at the lagoon mouth. When the berm is open, the lagoon is in direct exchange with the ocean (i.e., estuarine conditions). During closed conditions, drainage from the watershed remains trapped behind the sand berm and the lagoon becomes eutrophic (multiple blooms with chlorophyll-a > 10 µg/L). Additionally, long periods of drought can induce reverse estuarine conditions, in which seawater entering the lagoon during high tide becomes warmer and more saline than coastal seawater as a result of evaporation and a lack of freshwater inputs. The semi-enclosed geomorphology of coastal lagoons, combined with the occurrence of frequent algal blooms, provide an ideal setting to study the biological and physical drivers of mercury uptake by phytoplankton.

The toxicity of methyl mercury was first discovered after the 1950ties Minamata incident in Japan. Since then, it was found that methyl mercury is a non-threshold pollutant that impairs the development of the human brain. One of the main sources for mercury intoxication is the prenatal uptake of methyl mercury from sea food and the resulting loss of intelligence causes diminished economic productivity that persists over the entire lifetime of affected children.

To better understand the pathways of mercury from anthropogenic emissions to bio-accumulation in marine biota we developed an Eulerian three dimensional multi-media chemistry transport model (MECOSMO) that includes atmosphere, ocean, and ecosystem. The model includes a complete representation of the marine ecosystem from phytoplankton up to higher trophic levels, including fish. We use MECOSMO to simulate the spatial and temporal variability of mercury accumulation in fish in the North- and Baltic Sea. Based on a 60 year hind cast, we determine the natural variability of the system. Moreover, the model is used to project the impact of future perturbations in the system (i.e.: emission reductions, climate change, nutrient control) on the mercury accumulation in sea food. Thereby, supporting the implementation of the Minamata Convention on Mercury on a regional scale.

This work is still ongoing and the results of our model study will be presented for the first time at the ICMGP2017.

Mercury (Hg) pollution remains a concern for human and wildlife populations worldwide. Anthropogenic input of Hg to coastal ecosystems via direct atmospheric deposition and land runoff is expected to decline if negotiated mitigation efforts are implemented. To what extent and within what timeframe the levels of Hg in estuarine fish will be reduced to levels where it no longer poses a threat to human and wildlife health however remains uncertain. One major factor contributing to this uncertainty is the poor understanding of how Hg in recent inputs (‘new’ Hg) differs in availability for methylation and bioaccumulation in comparison to legacy Hg (‘old’ Hg) stored in the sediments. We have conducted mesocosm studies in which the bioavailability of different geochemical pools of Hg was studied under different nutrient and terrestrial matter loading regimes. The systems consisted of intact sediment cores (diameter of 0.63 m) that were placed in 5 m high mesocosm columns filled with brackish water. We then quantified the methylation, demethylation and bioaccumulation of two sets of inorganic Hg (HgII) and methylmercury (MeHg) isotope tracers. One set was added to the sediment (as MeHg and HgII bound to organic matter and HgII added as black cinnabar, β-HgS(s)) representing ‘old’ inputs of Hg, and one set was added as aqueous HgII and MeHg to the water column representing ‘new’ inputs of Hg. In this presentation, we will summarize the key findings from these studies (Jonsson et al., 2014, Nature Communications, Nguyen et al., 2016, Environmental Science and Technology, Jonsson et al. 2017, Science Advances) and discuss how this knowledge can be implemented in predicting future trends in Hg pollution and in the management of contaminated sites. We show for example that the chemical speciation of HgII solid/adsorbed phases controls the sediment Hg pool’s contribution to MeHg, but that input of MeHg from terrestrial and atmospheric sources bioaccumulates to a substantially greater extent than MeHg formed in situ in sediment. Based on the results we also predicted 3-6 times enhanced concentration of MeHg in zooplankton in coastal areas following scenarios with 15–30% increased terrestrial runoff (as suggested for parts of the world due to climate change).

Mercury differs from other heavy metals in that it continuously goes through the deposition and re-emission cycle in the environment. Mercury mainly exist as reactive Hg (Hg2+) in either dissolved or particulate phase in aquatic system. Reactive mercury can be converted to methyl mercury (MeHg) which readily bio-accumulates in aquatic food chains resulting in adverse health effects in human and wildlife, or re-emitted after reduction to elemental mercury. Emission from ocean surface is a great contribution to natural Hg emissions inventory and is an important pathway for removing Hg mass from aquatic system; therefore, quantifying the exchange flux between air and ocean surface is critical for understanding Hg fate and transport in environment. This study was initiated to estimate air-water exchange flux from ocean in Northeast Asian region. Air-water exchange of Hg is strongly dependent on DGM (Dissolved Gaseous Mercury) concentration in water. However, DGM has very low concentration and is supposed to have significant temporal and hourly variation because reduction and oxidation rate constants vary with UV radiation intensity. In this study, we measured total mercury (TM), reducible Hg (RM), and DGM concentrations during three cruises campaigns in open ocean in Northeast Asian region, and obtained the relationships among three Hg species. RM concentration ranged 20-55% of TM concentration, and a statistical correlation between TM and RM/TM ratio was observed. Spatial map of TM concentration in Northeast Asian Ocean was first obtained based on measurements from this study and from literature review, and RM concentration was then calculated based on empirical equation acquired between TM and RM/TM ratio. At the same time, laboratory experiments were carried out to identify the rate coefficients of Hg reduction and oxidation under various UV radiation intensity. DGM concentration will be calculated using reduction and oxidation reaction constants and calculated RM concentration, and air-water exchange flux will be estimated using two-film theory. Detailed results will be showed at the conference.

Dissolved organic matter (DOM) exerts a strong influence on the rate of methylmercury (MeHg) photodemethylation in ocean water. This MeHg photodemethylation can occur via two potential pathways in seawater: photodemethylation of MeHg by photochemically produced reactive intermediates (PPRIs; 1O2, ³DOM*, e¯aq, ·OH) and direct photodemethylation of MeHg-DOM complexes via intramolecular charge transfer. We investigated the potential mechanisms of DOM mediation of MeHg photodemethylation by measuring the demethylation rate constants in artificial seawater under UV-A (365 nm), in the absence and presence of Suwannee River Humic Acid (SRHA) and Pony Lake Fulvic Acid (PLFA). The concentration of MeHg decreased with increasing UV exposure time, and the photodemethylation process followed pseudo-first order kinetics. A predominance of MeHg-SRHA and MeHg-PLFA species over MeHgCl increased the photodemethylation rate constant by 2-5 times that of MeHgCl. The presence of SRHA decreased the concentration of MeHg by 54% in absence of a quencher (0.010 h⁻¹), whereas it decreased by 38% and 20% in the presence of the ∙OH quencher (0.0057 h-1) and 1O2 quencher (0.026 h⁻¹), respectively. Amendment with PLFA decreased MeHg by 70% in absence of a quencher (0.016 h⁻¹), and by 64% and 50% in presence of the ∙OH quencher (0.0125 h-1) and 1O2 quencher (0.0085 h⁻¹), respectively. These results imply that microbial fulvic acid (FA) has higher photosensitization effects than terrestrial humic acid (HA), and 1O2 mediation is more dominant than ∙OH mediation. Additional experiments with other model compounds, aimed at identifying the role of functional groups in HA and FA, showed that 1) photodemethylation was faster in the presence of aromatic thiols (thiosalicylic acid, 0.37 h-1) than aromatic carboxyls (salicylic acid, 0.010 h-1) or nonaromatic thiols (glutathione, 0.019 h-1), and 2) the photodemethylation rates of aromatic carboxylates and nonaromatic thiols were similar to those of SRHA and PLFA. The rate constants were lower in the presence of the ∙OH quencher (0.0066 h-1 for glutathione and 0.0055 h-1 for salicylic acid) than with the 1O2 quencher (0.012 h-1 for glutathione and 0.0074 h-1 for salicylic acid). The rate constants for thiosalicylic acid (aromatic thiols) were similar for both quenchers (0.45 h-1 for 1O2 quencher and 0.40 h-1 for ∙OH quencher) and these values were comparable to those determined without quenchers, implying that the role of PPRIs is not significant. The overall rate constants were compared to those for natural seawater to obtain insights into the photodemethylation mechanisms occurring in marine water.

Mercury (Hg) is a natural element toxic to all living organisms. Its ocean biogeochemical cycle is dominated by atmospheric deposition, which human activities contribute to disrupt significantly, and to a lesser extent by riverine discharge. This element is bioamplified and bioaccumulated in marine food webs. Since mercury concentrations in some coastal animal species of high trophic level approach sanitary thresholds, understanding biogeochemical processes and mechanisms leading to these elevated Hg concentrations becomes important. The stable isotopic geochemistry of metal is indeed a very promising way to trace processes and transfers from one mercury biogeochemical reservoir to another (Sonke and Blum, 2013). The stable isotopes of Hg undergo fractionation that can be either dependent (MDF) or independent (MIF) of their mass, thereby potentially enabling to track biological (MDF) and transfer processes between geochemical reservoirs (MIF). Their coupling with stable isotopes of carbon (C) and nitrogen (N) in biota, clarifies the functioning of trophic networks, and the bioaccumulation processes of contaminants such as Hg.

Since the 1970’s, the ROCCH, a Mussel Watch-like program, tracks contaminants on the French metropolitan coastline using of bivalves as quantitative indicators of coastal chemical contamination. The aim of the project was to map for the first time and on a national scale the temporal and spatial isotopic variations of Hg, C and N. This work also attempts to better discriminate the sources and trace the origin of Hg in bivalves (e.g., is Hg from the global ocean, associated with fluvial OM, or atmospheric local deposition…).

Along with its companion paper dealing with Hg speciation, we will present here mercury isotopic composition data for our entire dataset. The initial results of mercury speciation suggest that total Hg is not a limiting factor for Hg methylation (Briant et al., 2016). Thus, with global change and the predicted temperature rise, the associated increase in bacterial activity and eutrophication of coastal waters could increase MeHg production as proposed by Soerensen et al. (2016). Specifically, preliminary Hg isotope data show a regional signature (Atlantic Ocean vs. Mediterranean Sea), which may come from different trophic regimes as shown by N and C stable isotopes. In contrast to marine-influenced sites without temporal variations, river-influenced sites show temporal δ202Hg increase (e.g., -0.5 to +0.25‰), suggesting a decadal change in the origin of Hg reaching the ocean.

This study is part of the Trococo project, funded by Ifremer, Région Pays de la Loire (Pollusols), and INSU/EC2CO/Dril.

Mercury can be introduced into the marine environment in many different ways. In the case of the Baltic Sea, rivers and atmospheric deposition are predominant ones. However, in the face of ongoing climate change new potential source, coastal erosion, starts to become more significant and currently is considered as third most important source of mercury in the Gulf of Gdansk region. It is especially crucial given coastlines where due to increased occurrence of extreme natural phenomena such as storms, heavy rains, floods, the erosion processes along the coast begin to increase. Particularly vulnerable are cliff coasts which cover about 30% of the length of Polish coastline.

In recent studies considering coastal erosion as a source of mercury to the Gulf of Gdansk, annual loads of Hg were estimated. The aim of this study was to evaluate the amount of labile mercury in cliff sediments. Samples were collected in years 2015-2017 from four cliffs situated in the region of the Gulf of Gdansk, southern Baltic Sea. From each cliff 3 cores (0-65cm) were collected: from the slope of the cliff, and its top. Additionally sediments from the beach and coastline waters (1 m depth) were collected. Samples were kept in -20C until analysis. Later sediments were lyophilized and homogenized before mercury analysis. Mercury speciation was performed on DMA-80 mercury analyzer using thermodesorption method. It enabled to distinguish between labile and stable forms of Hg in sediment cores.

Cliffs along southern Baltic coast comprise mostly of Pleistocene glacial tills (boulder clay) thus mercury concentration in this kind of sediments was the most important one to focus on. Studies indicate that in boulder clay about 60% of mercury is present in bioavailable, labile form. Median concentrations of mercury in collected sediments samples did not exceed natural background values. Although taking into consideration the total mass of sediment which is annually introduced into the Baltic sea, and large masses introduced each time during natural extreme phenomena, it can be regarded as a significant route through which Hg enters the marine environment. High content of labile form suggests that mercury form coastal erosion processes can influence the marine trophic chain.

Experimental plots in Great Sippewissett Marsh (Falmouth, MA USA) have been undergoing long-term (>42 years) fertilization through the application of commercially available, sewage sludge-based fertilizer. This material, while delivering organic carbon and nitrogen to the marsh, supplies elevated amounts of mercury (Hg) and other metals as well. This experiment provides a unique opportunity to test hypotheses regarding the Hg-related response of coastal marine ecosystems to eutrophication as well as assess the efficacy of salt marshes to act as sinks for increased loadings of Hg to the coastal zone. We found Hg inventories in the salt marsh sediments that were usually equivalent to, or greater than, the inadvertent loadings from fertilizer in the treatment plots and from the atmosphere in the control plots, and in both high and low sections of the marsh. We also found that the control plots appeared to retain all atmospheric Hg loadings, though their faithfulness as historical archives maybe questionable. Likewise, the distribution of Hg in the fertilized plots appeared to be shifted relative to our reconstructed history of loadings, implying either some level of Hg mobility or inaccuracies in our reconstructions. The abundance of monomethylmercury (CH3Hg+) within the plots varied dramatically with the amount of fertilizer applied as well as sediment total sulfur, with higher percentages of Hg as CH3Hg+ and amounts of S in the control plots, and lowest percent CH3Hg+ and S in the most fertilized plots. Artifact CH3Hg+ formation was observed through the use of Hg stable isotope tracing during extraction and analysis and found to dominate uncorrected measured CH3Hg+ in the plots undergoing the highest level of fertilization.

These results showed no systematic change of Hg mobility relative to control plots other than the increased load of Hg. Thus, it would appear that New England salt marshes possess a strong ability to retain Hg loadings and that this ability is resistant to degradation by low- to moderate-level eutrophication. Finally, the prediction made by Driscoll and colleagues that eutrophication leads indirectly to less CH3Hg+ production appears borne out in this particular ecosystem.

Mercury cycling in shallow coastal bays and sounds remains poorly understood even though these are important in terms of local and subsistence fisheries. Long Island Sound (LIS), a temperate shallow estuary with an average depth of ~30 m, receives riverine discharge exclusively on its north shore and from the East River flowing through New York City at its western end. Approximately 90% of riverine influx to LIS is derived from the 653.59km-long Connecticut River, originating in Canada and entering LIS towards its eastern boundary. Water of LIS exchanges with the Atlantic Ocean primarily via a narrow, deeper (~50 m) passage, the race, situated on LIS eastern end, where there are strong tidal currents. A previously constructed budget for LIS assumed values for tidal exchange and in situ production, which have not been supported by rigorous measurements. Our recent study measured total mercury (THg) and methylmercury (MeHg) during 3 cruises that took place in May, August and November of 2016 along the transect connecting the north shore of LIS at the mouth of Connecticut River and the south shore of LIS at Orient Point (Long Island, NY). Water collected with Teflon-lined Niskin bottles from 3 depths (e.g. surface, mid and bottom) at 5 stations along the transect at low and high tide, was transferred using trace metal clean approaches into 2L Teflon bottles. During the cruise in November, water was also sampled during incoming and outgoing tides. Collected water was filtered via combusted quartz fiber filters. Both the filtrate as well as the collected particulate matter were analyzed for THg and MeHg. While analyses are ongoing, our initial results show that in spring and in the fall, the dissolved as well as the particulate MeHg concentrations were elevated near the river mouth as well as 2-5 m off the bottom. Spatial and temporal Hg distribution patterns for the filtrate and particles are here used to update the previous budget, while better constraining the importance of seasons, tides, and exchange across the race. Mass balance modeling was applied to simulate the effect of tide on the THg and MeHg cycling in the LIS.

Sediment is the main site of mercury methylation. Accurate quantification for methylmercury (MeHg) flux at the sediment-water interface is vital to better understand the biogeochemical cycling of mercury, especially the toxic MeHg species. Methylmercury is non-conservative in aquatic systems and many factors affect its concentration. Traditional approaches, such as core incubation, are hard to maintain at the in-situ conditions, leading to over/underestimation of benthic fluxes. The 224Ra/228Th disequilibrium method for tracing the transfer of dissolved substances just across the sediment-water interface, has proven to be a reliable proxy to quantify the benthic flux. In this study, both 224Ra/228Th disequilibrium and core incubation methods were applied to examine the benthic fluxes of both 224Ra and MeHg in Barn Island, Connecticut, USA (a wetland) from May to August, 2016. The results showed that the two methods were comparable for 224Ra but contradictory for MeHg. The radiotracer method indicated sediment was always a dominant source of both THg and MeHg. Core incubation produced similar results in May and August but an opposite pattern that sediment was a sink of MeHg in June and July, which was contrary to the evidence of significant MeHg gradients at the sediment-water interface. Overall, we conclude that the 224Ra/228Th disequilibrium approach is preferred for estimating the benthic flux of MeHg and that sediment is indeed an important MeHg source in this wetland.

From 2010 we have been researching the bioaccumulation of mercury in several fishes in Minamata Bay and surrounding areas in Yatsushiro-Kai, Kumamoto, Japan. We selected several sampling points that showed different environmental conditions, species composition and food web patterns. For the determination of feeding types of 60 species fishes (600 samples) sampled by gill net, we measured mercury levels of each sample and distinguished food habit, such as carnivore, omnivore, herbivore and detritivore from direct observation and DNA checking of fish gut contents. At this time, we introduced a stable isotope analysis for checking the food history and feeding habits of dominant fish. In about 300 individuals of 30 species of dominant fish selected from the 600 samples, we measured the stable nitrogen and carbon isotope ratios (δ15N, δ13C) of each sample. We also measured δ15N and δ13C of phytoplankton, zooplankton and benthic microalgae. Checking the food items in the gut of fishes, more than 80% were carnivorous, and showed different selectivity of food items, such as fish and crustacean. From the results of stable isotope ratios, benthic fish tended to show a higher ratio of δ13C. Usually benthic microalgae evidenced a higher ratio of δ13C than planktonic microalgae, and the ratio conservative through the food chain. In general, δ15N increases through the food chain with +3 to +4 ‰ enrichment per trophic step. In these data, carnivorous fishes of benthic and pelagic type showed medium and high ratios of δ15N. From comparing the stable isotope ratio to the mercury concentration of fishes, all of the high-mercury fishes belonged to benthic and carnivorous types. We consider the joint method of food web analysis and stable isotope analysis to be useful for understanding the mechanism of mercury bioaccumulation through the food web.

Sediments are the largest repository for mercury (Hg) in estuarine systems, but their role as a source of methylmercury (MeHg) to the coastal water column is under critique. We used a unique sampling approach to examine particulate MeHg exchange between estuarine sediments and the water column via resuspension over naturally relevant tidal and storm shear stress conditions. The resuspended fraction was collected using a Gust Microcosm Erosion System which induces step-wise increasing shear stresses to undisturbed sediment cores. The cores were collected from an upstream (8.1 ppt) and a downstream location (28.4 ppt) in the Delaware River. Within each location two sites were selected, one organic carbon (OC) poor and one OC rich. Particles were collected from the erosion effluent and analyzed for MeHg, total Hg, chlorophyll a, and C, N, S concentration and isotopic signature. We also collected a traditional bulk sediment sample (0-4 cm homogenized) to compare these results to. The results suggest that easily eroded particles have a different Hg composition than the bulk sediment for both OC conditions. At high OC sites, the resuspended fraction had less MeHg than the bulk sediment while the opposite was true for the low OC sites. The resuspended particles were similar to the suspended particulate MeHg burden at the more terrestrial site, but differed from both the sediments and the water column at the more marine site. This could be due to a higher abundance of living plankton at the more marine site, as suggested by high concentrations of chlorophyll a in the 0.2-20 μm size fraction, that would likely not be interacting with the surface sediments. The C isotopic signature for particles at the marine site was -21‰ indicative of marine phytoplankton, whereas the terrestrial site δ13C was -24.0‰ representing a mixture of marine and terrestrial carbon. This suggests that the MeHg dynamics at the sediment-water interface are distinct from the bulk sediment, and sampling of the bulk sediment does not provide an accurate reflection of the potential input of MeHg to the water column from resuspension. Therefore, to fully understand benthic-pelagic coupling in terms of MeHg in coastal ecosystems, the near-surface particulate MeHg should be sampled and considered when relating sediment-MeHg contributions to the overlying estuarine water columns.

Monomethylmercury (MeHg) is a potent neurotoxin that has a unique ability to biomagnify in aquatic food webs to harmful levels. Previous studies have shown that MeHg preferentially binds to organic compounds containing reduced sulfur groups (thiols). Thus, the concentration and composition of dissolved organic matter (DOM) play an important role in chemical speciation and bioavailability of MeHg. Terrestrial input of DOM furthermore promotes the growth of heterotrophic bacteria and therefore increases their share of the biomass production in the bottom of the pelagic food-web.

The Baltic Sea is an ideal study area for processes responsible for mercury transformation and its pathway in the environment due to large differences in the concentration of terrestrially derived DOM between the north and the south. Hence, different food web structures are present with bacteria-based (microbial) food web in the northern and phytoplankton-based (herbivorous) in the southern areas. As a result of elevated input of DOM in the northern basins and the differences in the food web structures, we hypothesize that MeHg concentrations in the higher trophic level of the pelagic food web (zooplankton) significantly differ between the two regions.

In this study we determine concentrations of MeHg in water and plankton samples obtained from different subbasins within the Baltic Sea using gas chromatography inductively coupled plasma mass spectrometry and calculate bioaccumulation factors through the food web up to zooplankton. We furthermore determine the speciation of sulfur by S-XANES (sulfur X-ray absorption near the edge structure) and identify dominant individual low molecular mass (LMM) thiols using liquid chromatography electrospray ionization mass spectrometry. These data are used to assess MeHg bioavailability and biomagnification from the dissolved phase and up to zooplankton.

Initial results from 2015 suggest that dissolved MeHg concentrations are slightly higher in the southern Baltic offshore surface waters (24±16 fM) than in the north (11±11 fM). However, we find that MeHg concentrations in plankton and the bioaccumulation factors decrease along the north-south gradient. This could be tightly connected with the microbial food web structure. With additional speciation data from 2016 cruise campaigns we will further address MeHg bioavailability and elucidate on how it affects the bioaccumulation process.

Improving our understanding of mercury cycling within estuaries is an important step in modeling mercury uptake by biota and quantifying mercury fluxes to the coastal ocean. We sampled 11 U.S. rivers (Suwannee, Apalachicola, Escambia, Mobile, Pascagoula, Pearl, Mississippi, Lower Atchafalaya, Sabine, Trinity, and Brazos rivers) discharging into the Gulf during Spring and Fall 2012. We measured total and methyl mercury (MeHg) concentrations in particulate and filtered phases as well as other parameters including total suspended solids, dissolved oxygen, and dissolved organic carbon. Mercury concentrations varied along salinity gradients with most systems characterized by mid-estuary maxima in MeHg concentrations. The mean fraction of total mercury as MeHg in each estuary ranged between 0.8 and 3.4%. We found significant correlations between particulate total mercury and total suspended solids. Seasonal differences in the partition coefficients of total mercury appear to be driven by changes in the concentration of filtered mercury.

Wetlands are known to be active sites of mercury (Hg) methylation and can be important sources of methylmercury to downstream systems and the food web. Salt marshes play a multifaceted role in estuarine ecosystems, such as cycling nutrients, providing habitat and mitigating floodwaters. Invasions of plants such as Phragmites australis have raised concerns over how invasive plants impact wetland ecosystem function. The role of P. austalis is complex. In some cases its presence may expand habitat, but in others habitat quality is decreased resulting in an altered food web.

Apart from its role in nitrogen cycling, our understanding of the impact of changes following a P. australis infestation on biogeochemical cycles is not well known. In P. australis patches of the Rhoade River Salt Marsh in the Chesapeake Bay Md, we have observed dramatic differences in sulfur and carbon cycles between P. austalis communities and other plant types. We believe this is in large part because of the active pumping of oxygen to the root zones. Sulfide is all but eliminated in peat pore water surrouning the P. australis.

Concentrations of mercury in the peat pore water around the plant roots are much higher than found in other plant communities in the marsh. Concentrations of methylmercury in the pore water are as high or higher than found in pore water of the other plant communities in the marsh. Given the differences in redox condition, Hg cycling is unlikely to be the same, driven by differences in Hg supply and microbial activity. In this presentation we will first examine factors influencing Hg cycling in the P. australis infested Rhode River Marsh. Salt marshes are import sources of carbon estuaries. Export of dissolved organic matter is substantial, and being located on the fringes of the marsh, P. australis dominated patches perhaps have a greater impact on the type and amount of carbon exported, and hence the Hg and meHg exported, than other plant communities. We will discuss C and Hg export from the Marsh and examine any linkages back to plant community source.