2g: Legacy site assessment and management

Yolo Bypass is the largest flood bypass in the Sacramento Valley, California. During high flow flood events, water is diverted into the Yolo Bypass from the Sacramento River to control river stage and protect the city of Sacramento from flooding. Climate change projections for the region indicate the risk of flooding will increase. An increase in flooding would result in increased connectivity of the flood plain with downstream habitats as well as provide conditions favorable for the in situ production of methylmercury (MeHg) in an area burdened by mercury contamination resulting from historic mining activities upstream. Conversion of inorganic mercury (Hg) to the more toxic organic form, MeHg, in freshwater systems is generally accepted to be mediated by bacterial activity. There are a number of environmental variables (organic carbon, sulfate, oxygen) and conditions (temperature, porosity, soil type) that could influence the net production of MeHg and its ultimate release into the water column. This study investigated sediment-water exchange of both total Hg and MeHg from different habitat types in the Yolo Bypass including wild rice, white rice, seasonal wetlands, irrigated pasture, non-irrigated pasture, fallow land, farm land, freshwater tidal wetland, and agricultural drain. Fluxes were determined by a direct assessment using incubated cores and measuring change in overlying water concentrations over short time scales (<25 hrs). Fluxes of total Hg and MeHg ranged from -621 to 787 ng m-2 d-1 and -11 to 29 ng m-2 d-1 respectively with negative values indicating flux into sediment. Agricultural, pasture, fallow, and seasonal wetlands were sources of MeHg with fallow and seasonal wetlands having the highest fluxes. In contrast, drainage canals and freshwater tidal wetlands were sinks and had fluxes into the sediment. These results will be used in the Dynamic Mercury Cycling Model (D-MCM) and integrated with hydrodynamics of the Yolo Bypass to improve our understanding of factors controlling production and transport of Hg and MeHg in the Yolo Bypass.

Historical mercury (Hg) releases occurred at a textile manufacturing facility on the South River, VA. These releases resulted in increased Hg concentrations in biotic and abiotic media, which have not declined over the past thirty years in some media (e.g., fish tissue), as originally expected. Introduction of legacy Hg impacted soils to the South River through bank erosion is the most significant source of Hg loading to the system. Interim remedial measures targeted at reducing this source within the first few river miles were initiated in 2016.

As part of the remedy selection process, a set of robust statistical models were developed to predict responses in physical and biological media associated with completion of the interim measures (i.e. bank stabilization). The models integrated site-specific data collected over more than a decade and built upon the detailed conceptual model. Model parameters included total mercury (THg) and methylmercury (MeHg) concentrations in a number of physical media (e.g. sediment, soil, and surface water); biological (e.g. fish tissue) as well as meteorological, adjacent land use and stream discharge data. Different remedial scenarios (i.e. 50% and 100% reduction in Hg loading from eroding banks) were evaluated in order to inform remedial decision making. Model runs predicted significant THg reductions in surface water and sediments adjacent to and downstream of the proposed interim measures.

A short-term monitoring (STM) program comprised of multiple lines of evidence is being instituted that aims to document the predicted reduction in Hg concentrations in a range of biotic and abiotic endpoints. The STM program is intended to evaluate changes in Hg concentrations at specific bank segments following remediation over a relatively short time period (i.e. 2-10 years). Key learnings from the Pilot Bank Stabilization project monitoring program on the South River were built into the STM program in order to maximize the ability to not only detect reductions in Hg loading attributable to the interim measures, but also to be able to differentiate those from changes that may be due to other sources.

This presentation will expand upon the conceptual site model, outline the input/output of the statistical model, review baseline STM data and discuss their importance in an adaptive management remedial framework.

This study provides a result of mercury recovery characteristics for waste using thermal technology. Mercury waste was categorized as three types such as waste consisting, waste containing, waste contaminated of mercury or mercury compounds. Those mercury wastes should be properly collected separately and recovered mercury or stabilized for environmentally sound management. Thermal treatment was conducted to recover elemental mercury from mercury containing products (barometer and UV lamp) and Hg contaminated soil. 150 L volume stainless steel drum was introduced inside of the pilot-scale furnace for thermal treatment. An agitator was kept operating during thermal treatment if mixing is needed. Experiments were carried out in temperature range between 550-750℃ with reduced pressure condition up to 1,013.25 Pa. Off-gas generated form thermal furnace was passed through a ceramic filter unit. That is used to filter out impurities in flue gas with higher temperature than boiling point of mercury in case of necessary cleaning selectively. 100 kg of waste per batch could be treated and elemental mercury was recovered with a condensation unit by maintain the temperature around 10℃ during thermal treatment. Activated carbon trap was applied to control mercury in emitting gas. Also mass balance was made to observe mercury distribution in pilot scale process. Treated waste could be disposed as general waste and recovered mercury could be recycled and reused as a resource.

Berrys Creek (New Jersey, USA) is an urbanized tidal tributary to the Hackensack River which is in many ways characteristic of urbanized tidal creek-marsh systems dominated by Phragmites. The detrital food web of the Berrys Creek Study Area (BCSA) supports mummichogs and white perch, and the bird community includes sandpipers, red-winged blackbirds, and herons. In the mid-2000s, in association with an Remedial Investigation/Feasibility Studyconducted under Superfund, we initiated a series of studies (1) to better understand the factors controlling mercury geochemistry, bioavailability, and bioaccumulation in BCSA marshes, and (2) to evaluate the potential efficacy of using activated carbon (AC) -based amendments in situ to reduce environmental risks from sediment-associated mercury and PCBs in these marshes.

Building on data from related studies, a substantial field pilot study was conducted in Nevertouch Marsh within the BCSA, in order to assess the benefits of applying thin-layer treatments with SediMiteTM, Granular Activated Carbon (GAC), and GAC + sand. The purpose was to evaluate in-situ treatment effects on Hg, MeHg and PCB bioavailability (porewater) and biouptake (amphipods). The study involved pre-treatment sampling, and spring and fall sampling and analysis of all media over 2 years. Additionally, in situ and laboratory bioaccumulations studies were done to assess AC effects on Hg/MeHg and PCB biouptake.

Results from the Nevertouch Marsh Field Pilot Study and associated bioaccumulation tests provide important information for evaluating the efficacy of in situ treatment of BCSA marshes with AC-based amendments. Specifically, compared to controls, in situ treatment with AC-based amendments: (1) Was highly effective for reducing the bioavailability and biouptake of PCBs, (2) Reduced Hg concentrations in porewater and detritus, (3) Reduced Hg biouptake in caged amphipods (Leptocheirus), (4) Reduced MeHg levels in porewater in the 28-day bioassay with Leptocheirus, and (5) Reduced MeHg biouptake in the 28-day bioassay with Leptocheirus (on days 7 and 14). SediMiteTM treatment consistently performed better than GAC and GAC + sand. Combined with data from related studies, results from the Nevertouch Marsh Field Pilot Study are very helpful for informing the BCSAfeasibility studyand improving our understanding of Hg dynamics in tidal marsh ecosystems.

Recently, there was a report on high mercury concentration in bivalves (Corbicula leana) and sediments near the confluence of the Hyeongsan River and Chilseong Stream located in the southeast of the Republic of Korea. Hyeongsan River supplies drinking water to citizens of Pohang City and currently, the pollution of Hyeongsan River is a big issue.

Given that both Chilseong and Gumu Streams run through the Pohang Industrial Complex and ultimately flow to Hyeongsan River, it is imperative to study if there are any impacts of industrial effluents on mercury contamination in the two streams and the Hyeongsan River. Thus, the objective of this preliminary study was to examine total and methyl mercury concentrations in sediments collected from Gumu and Chilseong streams and from Hyeongsan River. Sediments and water were collected from over 30 stations along with Hyeongsan River, Gumu Stream, and Chilseong Stream and analyzed for total mercury (THg) and methylmercury (MeHg) using Cold-Vapor Atomic Fluorescence Spectrometry. Average concentration of THg was highest in Gumu Stream (78.8±84.3 mg/kg dw), followed by in Hyeongsan River (4.39±9.70 mg/kg dw) and Chilseong Stream (0.30±0.42 mg/kg dw). Similarly, average MeHg in Gumu Stream, Hyeongsan River, and Chilseong Stream were 151±237, 3.19±6.77, and 0.67±1.12 μg/kg dw, respectively. Further study will be necessary to investigate major sources of Hg in Gumu Stream and to evaluate the impacts of Hg transport to the adjacent Hyeongsan River.

Mercury in fish is a widespread problem and source control alone will not solve this problem in most California reservoirs in a reasonable amount of time. Therefore, California's Environmental Protection Agency (Cal/EPA) is developing an innovative statewide mercury control program for reservoirs. This poster provides an update on California's program since ICMGP 2015 (S04).

Fish methylmercury levels are elevated in about half of all California lakes and reservoirs sampled. In addition, the California Office of Environmental Health Hazard Assessment has issued many advisories for limited or no consumption of many popular sport fish in California lakes and reservoirs. The inability to safely consume fish from many California lakes and reservoirs devalues California fisheries as a food source for humans and wildlife.

Mercury impairment is due to several inter-related factors: inorganic mercury sources; conditions in reservoirs that cause the conversion of inorganic mercury to methylmercury and its subsequent bioaccumulation in the food web; fish species present; and in some cases depressed primary and secondary production. Reservoir creation and management may exacerbate the mercury impairment by increasing methylmercury production and bioaccumulation.

The project involves: identifying mercury sources to reservoirs; evaluating reservoir, watershed, and fisheries conditions; determining the linkage between reservoir fish methylmercury levels, reservoir and watershed conditions, and mercury sources; and identifying controllable factors that determine reservoir fish methylmercury levels.

The potential solutions to reduce fish methylmercury concentrations are three-pronged: (1) mercury source reductions, (2) reservoir water chemistry management, and (3) fisheries management. This proposed statewide mercury control program for reservoirs encourages innovation in reservoir water chemistry (e.g., oxidants to hypolimnion) and fisheries management (e.g., intensive fishing; new or changes to fish stocking practices; and nutrient additions to slightly increase chlorophyll-a concentrations in oligotrophic reservoirs).

More information is available on the project website.

Surface applications of black carbons and other treatments were tested as potential tools for methylmercury (MeHg) risk mitigation in a Hg-contaminated tidal salt marsh in the Penobscot River, ME. Four surface amendments were tested, activated carbon, a pine dust biochar, ferrous chloride (FeCl2), and lime. Field studies across the Penobscot system pointed to salt marshes as sites of particularly high methyl Hg production and accumulation, and therefore a key target for remediation. The study design was a fully-crossed small plot study, with five treatments at each of two sites on a large marsh platform at mid-salinity. The sites were chosen to represent two different major habitats in the marsh.

Plots were initially treated in Sept. 2010, and were sampled four times post-amendment, through Sept. 2012. The main study endpoints were pore water MeHg concentrations and sediment water partition coefficients, presumptive indicators of bioavailability to benthic biota. The treatments were generally well-retained in the plots, and it was visually obvious that the carbons penetrated deeper into surface soils over time.

Both activated carbon and biochar amendments were effective in reducing Hg and especially MeHg concentrations in pore waters at both study sites in Mendall Marsh. AC amendments reduced pore water MeHg concentrations at the both sites by >90% at the one month time point and by 60% to 70% on average across all the four time points through two years.

Biochar was only a little less effective than AC. Lime and FeCl2 additions had no significant impacts on either total Hg of MeHg pore water concentrations. In general, the black carbon amendments did not have any significant impact on surficial pore water chemistry.

AC and biochar reduced MeHg risk primarily by increasing MeHg partitioning to the solid phase (i.e. reduction in pore water MeHg concentration), rather than by reducing the bioavailability of inorganic Hg for methylation. The next step in evaluating this tool in the Penobscot would be larger-scale, longer-term plot studies. These studies should include evaluation of long-term changes in MeHg accumulation in soils, as well as studies of animal bioaccumulation, food web structure, plant community structure, marsh productivity and Hg/MeHg flux. Large plot studies would also provide a better estimate of the cost of treatment.

Inorganic Hg, predominantly from widespread atmospheric deposition, but also from point sources including mine and industrial sites, is transformed to methylmercury (MeHg) by anaerobic bacteria in oxygen-poor water and sediment. The profundal zone of productive lakes commonly exhibits summertime anoxia and associated accumulation of MeHg in bottom waters. MeHg in bottom water bioaccumulates in pelagic biota when it diffuses upwards across the thermocline or when MeHg-rich bottom water mixes into the upper water column. With the shortcomings of conventional Hg control strategies, such as dredging, capping, watershed controls and source control, managers are increasingly interested in developing in situ strategies to repress Hg bioaccumulation in managed aquatic ecosystems. One potential strategy is to enhance the redox potential at the sediment-water interface to repress MeHg efflux to overlying water. In this study we present a conceptual model of MeHg cycling at the profundal sediment-water interface and summarize redox-mediated mechanisms that enhance MeHg efflux, including MeHg production by sulfate-reducing bacteria, release of MeHg from metal hydroxides in sediment, demethylaing capacity of aerobic and anaerobic bacteria, and enhanced MeHg mobility under sulfidic conditions. We then present highlights of experimental sediment-water chamber incubations from a number of California reservoirs. Sediment generally showed enhanced efflux of MeHg under anoxic conditions relative to oxic conditions. Typical anoxic release rates of MeHg ranged from 20-150 ng/m2/d. Some chambers showed a correlation between sulfate uptake and MeHg production, and some chambers showed a distinct drop in MeHg accumulation corresponding with the disappearance of sulfate. One interesting observation common to most experiments was a loss of MeHg from chamber water near the end of the anoxic phase of the incubations, suggesting either a shift in the relative importance of methylation and demethylation and/or a sorption sink, perhaps via iron sulfide formation under sulfidic conditions. At one contaminated study site, oxic conditions, while lowering MeHg efflux, resulted in enhanced efflux of ionic mercury, sulfate and iron, suggesting enhanced oxidative dissolution of sulfide minerals with the co-release of mercury. With a more comprehensive understanding of MeHg cycling at the profundal sediment-water interface, managers will be better able to develop effective management strategies aimed at repressing MeHg bioaccumulations in lakes and reservoirs.

Background/Objectives. Heavy metals such as Mercury (Hg) as well as Methyl Mercury (MeHg) are widely distributed and persistent pollutants in the environment. Although all contaminated sites are unique and a site-specific approach to remediation is often required, heavy metal contamination creates even more complex challenges. Remediation methods in general use include isolation, immobilization, toxicity reduction, physical separation and extraction. The ecological and human health effects of Hg, and MeHg are generally related to the environmental transformations. For example, inorganic mercury can transform to the more toxic monomethylmercury (MeHg) in anaerobic environments such as contaminated sediments. Biomagnification of MeHg in the aquatic food web and consumption of fish and shellfish contaminated with MeHg is the primary route of human exposure to Hg(II). MeHg is toxic to both humans and fish. Almost all mercury in fish is MeHg. MeHg is formed largely in anoxic sediments from inorganic mercury methylation mediated primarily by sulfate reducing bacteria. The bottom sediments are the main reservoirs of mercury and to the extent that this mercury is available to overlying water is a sensitive indicator of risk to the aquatic ecosystem. Effective remediation of such sediments to reduce the release of mercury is essential to minimize the contamination of fish and shellfish with MeHg. Experimental results will be presented that demonstrate the efficacy of Mn-based adsorbents for immobilization of Hg (II) and demethylation of MeHg to Hg(II).

Results. A laboratory investigation was conducted to assess the feasibility of immobilizing Hg and demethylating MeHg using iron (Fe) and manganese (Mn) mixed metal oxides. Batch sorption screening test results indicate >99% removal of Hg from aqueous batch reactors. Anaerobic batch microcosms were conducted with Hg-spiked sediments amended with Mn-based oxides to evaluate demethylation as a result of manipulating the redox conditions from reducing to oxidizing. The results of these batch studies will be presented as well as preliminary results from a pilot-scale test.

This work presents findings on the developments of a photochemical treatment method for Hg removal from organic-rich secondary process chlor-alkali wastewater (i.e., conventionally treated with sulfide) that resulted in removal of dissolved ppb Hg down to low ppt Hg levels (<<40.0 ppt Hg). The method, termed ultraviolet activated chelation (UVAC), uses high intensity ultraviolet irradiation at 254 nm (UV-C) followed by sub-micron filtration (0.45 m). Hg removal results are presented as unfiltered and filtered Hg effluent following UVAC treatment with variable pH and UV-C irradiation periods. Low-level ppt Hg results in filtered effluent are shown to occur by chlor-alkali UVAC photo-chemical processes in both Hg wastewater and Hg solutions of DI water and variable IHSS dissolved organic matter and sodium sulfide prepared under open atmosphere. The findings lead to a discussion of theoretical explanations for the observed filterable Hg formation, including the importance of sulfide and photo-activated organic binding ligand action in photochemical valence transfer within soluble Hg complexes. Possible future work follows to include experimental measurements that could further develop UVAC as a viable Hg removal technology for wastewater with variable halide and organic chemistry.

High Hg concentrations in New Hampshire fish have been attributed to atmospheric deposition associated with global anthropogenic emissions initiated during the industrial revolution. However, significant local sources of Hg may exist from mine tailings associated with abandoned lead-zinc mines in eastern New Hampshire and western Maine. The Madison Lead Mine in New Hampshire, most active in the early 20th century, produced a tailings fan (0.7 ha) that extends into Cooks Pond (14.3ha). Pond sediments have concentrations of Pb (>6,400ppm) Zn (>11,200ppm) and Hg (>450ppb) that are significantly higher than sediments from surrounding lakes. Mercury in the mine tailings is likely derived from a variety of brown sphalerite that is found within the ore body that contains up to 11,000ppb Hg.

Eight sediment cores (90mm diameter) were collected from Cooks Pond together with cores from downstream Silver Lake and nearby (2km) Durgin Pond using a Uwitec gravity corer. Cores were collected in water depths ranging from 4.5m to 7.5m and were 60-100cm in length. Fish (pickerel, yellow perch, sunfish) were collected by seine net and hand line. Cores were scanned using an ITRAX core scanner and were subsampled at 3-50mm intervals. All samples of fish and sediment were analyzed for total Hg using a Teledyne Leeman Labs Hydra C mercury analyzer.

The concentration of Hg in sediment cores from Cooks Pond, co-varies with Pb and Zn, with the highest concentration (460ppb Hg) occurring approximately 7cm below the sediment water interface. Concentrations in the sediment also increase with proximity to the onshore tailings deposit. If Hg was only derived from atmospheric deposition, its concentration would be expected to decrease with proximity towards the source of the tailings due to sediment dilution. In nearby Durgin Pond, the highest Hg concentration (320ppb) occurs just 1cm below the sediment water interface reflecting Hg derived primarily from atmospheric deposition.

Higher Hg concentrations would be expected in fish from Cooks Pond given the higher concentrations in the sediment, however it was found that Durgin Pond fish of similar length, weight, and species had consistently higher Hg concentrations (350ppb vs 90ppb). This is likely due to two factors. First, much of the mine tailings Hg is mineral bound and not bioavailable, and second, there are more wetlands surrounding Durgin Pond allowing for more Hg methylation. Therefore we conclude that despite adding Hg to the ecosystem, mining has not led to higher concentrations of Hg in fish.

The California Bay-Delta region is impacted by mercury contamination from historical mercury and gold mining upstream. Microbial activity in downstream aquatic environments can convert inorganic mercury into methylmercury (MeHg), a toxic and readily bioaccumulated form of mercury. The Delta Mercury Control Program (DMCP) was established in 2011 to protect human and wildlife health from MeHg exposure. It requires the California Department of Water Resources to conduct control studies to evaluate whether operational changes or other practices could be implemented to reduce ambient MeHg loads to Delta open waters and the Yolo Bypass, a floodplain area inundated by managed flows. A mechanistic model of mercury cycling is being applied in the Yolo Bypass as part of a multi-agency modeling, field data collection, and experimental program to identify options to reduce MeHg supply and meet DMCP regulatory requirements. The Dynamic Mercury Cycling Model (D-MCM) is a time-dependent 1D-3D mechanistic aquatic mercury cycling model that includes inorganic Hg(II), methylmercury and elemental mercury in water, sediments and a food web. Hydrodynamic inputs for D-MCM are generated with TUFLOW, a high resolution hydrodynamic model. A coarser 47 cell grid for mercury simulations was developed based on 5 criteria: land use, disked/non-disked land, THg concentration in sediments, percent time wet, and wet/dry cycling frequency. Model simulations span 1997-2012. D-MCM is currently being used to predict concentrations and mass balances of inorganic and methylmercury in the Yolo bypass, considering various influences on methylmercury production and concentration, including hydrology, particle dynamics, vegetation, and inorganic mercury contamination. Predicted exports of inorganic Hg and MeHg from the Yolo Bypass are being used as inputs to a companion mercury modeling effort for the downstream Sacramento San Joaquin RIver Delta system.

Mercury (Hg) sequestered in soils and sediments in riverine systems can be mobilized as a result of flooding and precipitation events, acting as a secondary source of contamination to the river. Remediation of this Hg is often complex, greatly depending on the scale of Hg distribution and the influence of Hg transport and deposition processes. At some contaminated sites, partitioning of Hg to reactive amendments may be implemented to stabilize Hg in situ, rendering it relatively immobile and less bioavailable. Saturated column experiments were conducted to evaluate the effect of several inorganic and carbon-based amendment materials on Hg immobilization and transport. Columns were filled with Hg-contaminated sediment collected from the South River, VA, amended with a selected reactive material, and were continuously flushed with low-Hg river water from the South River for more than 6 months. The inorganic amendment columns included sediment blended with a 2% dosage (dry weight) of limestone, unmodified clay or sulfidized clay. The carbon-based amendment columns consisted of a 2% (dry weight) dosage of unmodified biochar, HNO3-modified biochar, sulfidized biochar or a 5% dosage of sulfidized biochar. A control experiment including a column packed with unamended sediment was run to establish baseline Hg concentrations and geochemical conditions. The results of the experiments indicate that the addition of limestone, unmodified biochar and HNO3-modified biochar to contaminated sediment had little effect on Hg transport under saturated conditions. Leaching of both filter-passing (<0.45 mm) and particulate Hg from these amended systems was similar in magnitude to the sediment control. In contrast, extensive Hg mobilization was observed in the sediment amended with unmodified clay. Cumulative filter-passing Hg measured in the effluent of the clay-amended sediment increased by more than 200% when compared to the control, suggesting destabilization of Hg previously sequestered in sediment and colloid-facilitated transport in the presence of the clay amendment. Mercury transport was greatly reduced when sediment was blended with either sulfidized clay or sulfidized biochar. The cumulative release of filtered Hg measured in the effluent of both the sulfidized clay-amended sediment and the sulfidized biochar-amended sediment (2% dosage) was 24% of the control and 29% of the control, respectively. A further reduction of Hg leaching was observed when the amendment dosage of sulfidized biochar increased from 2% to 5%. The strong interactions between Hg and S may have resulted in increased binding of Hg to the S-modified materials, effectively sequestering Hg within the sediment.

Clay particles (< 0.002 mm) can dominate trace element sorption in soils because of their large surface area per unit mass. However, the smaller clay size fractions are also potentially mobile allowing for leaching and loss of particulate Hg from the soil profile. To investigate the extent to which clay particles dominate Hg sorption we size fractionated eight soil samples collected from a contaminated floodplain soil that had been impacted by a chlor-alkali facility. The study site is located in Berlin, New Hampshire; multiple soil cores were collected on the floodplain at increasing distance from the original Chlor-alkali facility; selected samples from different cores and of varying total Hg content and depth in the profile were used for size fractionation. Initially the soil sample (ca. 10g) was dispersed in 50 ml of deionized water by sonication. Four clay fractions were then separated by gravitational settling (2-1 µm) or centrifugation (1-0.5 µm), 0.5-0.1 µm, 0.1-0.05 µm. The final supernatant was retained to study the dissolved partitioning of Hg by size exclusion chromatography-ICP-MS. The separated fractions were dried and total Hg was determined by acid digestion and ICP-MS analysis. With the exception of one soil sample (23% clay) the other soils were low in clay content (1-4%). The soil samples ranged from 0.07 – 8.2 mg/kg total Hg and, as expected, Hg concentration was increased in the clay fractions compared to the bulk and increased with decreasing particle with the smallest fraction having up to 18 times the corresponding bulk soil Hg concentration. Excluding the high clay soil sample, the clay fractions made up, on average, 2% of the total soil mass but contained 17% of the total Hg. The two larger clay fractions, despite being lower in Hg concentration, contained the majority of Hg because they constituted most of the clay mass. Size exclusion chromatography-ICP-MS revealed that Hg in the aqueous phase after the final clay separation was bound in nanoparticle complexes. The average percent of the total soil Hg in the < 0.5 µm was 8% and indicated that particulate transport of Hg from these soils was likely an important process.

The Berrys Creek Study Area (BCSA) is a tidal tributary of the Hackensack River with a long history of human impacts. The BCSA encompasses the 6.5-mile-long Berrys Creek, its tributaries, and approximately 1,100 acres of tidal wetlands dominated by the marsh reed Phragmites. The BCSA is the subject of an ongoing remedial investigation focused on mercury (Hg) and polychlorinated biphenyls (PCBs) as primary contaminants of interest, their potential for bioaccumulation, and remedial options for reducing their potential bioaccumulation. This presentation focuses on the development of laboratory test methods used to examine the bioaccumulation of Hg in benthic invertebrate test organisms and the effects of activated carbon-based amendments on reducing Hg biouptake.

Preliminary studies were conducted to optimize methods to assess the bioavailability of Hg and methyl mercury (MeHg) in BCSA marsh sediments. Several challenges needed to be overcome in order to successfully conduct the bioaccumulation tests. For example, because the test organisms could not be tested with unprocessed marsh sediment samples, which are made up of sediment mixed into a dense matt of Phragmites roots, the field samples were homogenized before the test. This facilitated survival of the test organisms and reduced the time required to process samples and retrieve sufficient biomass of test organisms for chemical analysis. However, collection and processing of sediment is known to disrupt the geochemical conditions in natural sediment, increase available organic matter and microbial activity, and disrupt the balance of methylmercury production and degradation. Therefore, preliminary studies measured the concentrations of MeHg in sediment and porewater, along with other parameters (e.g., redox, sulfides) over time, to demonstrate that redox and net methylation conditions were reestablished before initiating the bioaccumulation phase of the test. We also tested several approaches to mixing and dilution of field samples.

Subsequent 28-day bioaccumulation tests were conducted to determine if treatment of BCSA sediment with activated carbon (AC) reduces the bioavailability and biouptake of total Hg and MeHg by the amphipod Leptocheirus plumulosus. Concentrations of MeHg in porewater were significantly reduced in AC-treated sediments in comparison to untreated sediments at most time points analyzed. Bioaccumulation of MeHg was significantly reduced in test organisms exposed to BCSA sediment treated with activated carbon-based amendments after 7 to 14 days exposure.

Biofilms are well known to be involved in the fate of Hg in lotic systems, by their accumulation, transformations and involvement in its trophic transfer. In contrast, only few are known about Hg impacts on biofilms. The present study aimed thus to examine long-term effects of Hg towards biofilms. To that end, biofilms were grown on microscope glass slides using microcosms filled with Geneva Lake water. Three microcosms were spiked with inorganic IHg at concentrations of 13 ± 2 pM, 131 ± 16 pM and 1.48 ± 0.08 nM (precisely measured), whereas one microcosm was not contaminated and used as a control. After 7 weeks of colonization, biofilms were analyzed for their total THg and intracellular, Hgint, (determined after a washing step with cysteine) IHg and CH3Hg content. Their compositions were examined based on their ash free dry mass, chlorophyll a content, percentage of abiotic and biotic fractions and taxonomic composition. Grown biofilms were further exposed to 2 nM IHg for 24 h and analyzed for their modification of microorganism cell membrane permeability and oxidative stress.

Biofilms grown in control microcosms were found to contain 0.33 ± 0.01 nmol THg/gdw and 0.13 nmol Hgint/gdw. Exposure to 131 ± 6 pM IHg increased by about 2- and 4- fold THg and Hgint content, respectively. At the highest studied Hg concentration, THg content reached 18.7 ± 6.1 nmol/gdw and Hgint content, 1.9 ± 0.3 nmol/gdw, representing only 10% of the total Hg content. That increase in accumulation was accompanied with a 1.5-fold decrease of chlorophyll a content in biofilms grown in 1.48 ± 0.08 nM IHg. Taxonomic analysis also revealed a shift in algal and bacterial species upon exposure to Hg. The modification of membrane permeability was found to be less important in biofilms grown in IHg than in biofilms grown in the control microcosm. In contrast, pre-exposure to IHg did not prevent biofilms to exhibit oxidative stress. Long-term Hg exposure induced an accumulation of IHg in biofilms as well as a change in microbial communities. Biofilms might thus represent useful bioindicators of Hg effects in natural waters.

Reservoirs typically have elevated fish mercury (Hg) levels compared to natural lakes and rivers. A unique feature of reservoirs is water-level management which can results in sediment exposure to the air. The objective of this study is to identify how reservoir water-level fluctuations impact Hg cycling, particularly the formation of the more toxic and bioaccumulative methylmercury (MeHg). Total-Hg (THg), MeHg, stable isotope methylation rates and several ancillary parameters were measured in reservoir sediments (including some in porewater and overlying water) that are seasonally and permanently inundated. The results showed that sediment and porewater MeHg concentrations were over 3-times higher in areas experiencing water-level fluctuations compared to permanently inundated sediments. Sulfate cycling which is often associated with MeHg production was enhanced in the seasonally inundated sediments; however, statistical analysis showed that the main factors correlated with porewater MeHg concentrations were porewater THg, porewater DOC, and sediment-porewater THg partition coefficients (log Kd). The THg log Kd values showed distinct relationships with sediment organic carbon depending on whether the sediments were permanently are seasonally inundated. Overall, our results suggest that sediment exposure to the air increases organic matter breakdown which promotes the partitioning of THg and carbon into the porewater phase where it enhanced methylation.

Streambank soils within the East Fork Poplar Creek (EFPC) watershed have elevated concentrations of mercury (Hg) as a result of historic use and discharge of Hg into the system associated with activities at the Y-12 National Security Complex. Mercury can leach from these soils as a result of water level fluctuations and erosion of stream banks resulting in the soils acting as a source of Hg to the creek. Laboratory experiments were conducted to examine the amount of Hg leaching from two soil samples collected from EFPC and a soil sample collected from Hinds Creek, an uncontaminated creek used as a reference site. For these batch laboratory studies the amount of mercury leaching from the soils when exposed to artificial creek water or water from EFPC was examined for two weeks with samples collected every 3-4 days. After two weeks, solid phase sorbents were added to some of the samples and the amount of Hg leaching from the soils with and without sorbent addition was examined for an additional two weeks. The concentration of Hg in the reference soil was 0.028 g/g and the two soils from EFPC had Hg concentrations 100 and 10,000 times greater than the reference soil. After the two week leaching period with artificial creek water, the concentration of Hg in the aqueous solution was 1.6 ng/L in the soils from the reference site and approximately 100 and 4000 times greater in the aqueous phase of the contaminated soil treatments. The solid phase sorbents that were tested included Thiol-SAMMS(r), OrganoclayTM PM-199, OrganoclayTM MRM, SediMiteTM and a biochar. A reduction in the amount of Hg leaching from the soils was only observed when SediMite was added The Hg partitioning coefficients for the three soil samples were similar (log Kd= 4.3-4.7) and these Kd values were greater than the Kds measured with the sorbents and a Hg-DOM solution (no soil present). The results from this laboratory study will help evaluate the usefulness of solid phase sorbent materials as a remedial option for reducing the amount of Hg leaching from soils into EFPC.

Elevated concentrations of mercury are found in stream bank soils and sediments in the East Fork Poplar Creek watershed in east Tennessee, where erosion of stream banks and leaching of mercury contributes to mercury flux into the downstream environment. The use of sorbent technologies is being investigated as a soil or sediment remediation approach to minimize mercury flux and/or mercury available for microbial methylation by sequestering mercury within a high affinity sorbent matrix. The majority of sorbent studies target ex situ removal of inorganic mercury from industrial waste streams. However, for the in situ treatment of a contaminated aquatic ecosystem, sorbents need to be evaluated in the presence of competing ligands, such as natural organic matter (NOM), mineral surfaces, and other constituents present in soils or sediments. These competing ligands can limit the effectiveness of sorbents deployed in situ. We conducted a series of equilibrium sorption batch experiments to evaluate the impact of NOM complexation on sorbent performance. Sorption isotherms were developed both with and without a standardized NOM reference material (Suwannee River NOM) for the following sorbents: Thiol-SAMMS(r), OrganoclayTM PM-199, OrganoclayTM MRM, SediMiteTM, pine wood biochar, lignin-based carbon materials, brass and bone apatite. The presence of NOM significantly decreased the sorption efficiency of inorganic mercury to the sorbents, quantified by the partition coefficients, which were highest for Thiol-SAMMS(r) followed by carbon-based sorbents. The release of solutes from sorbents was also evaluated (e.g., SO42-, NO3-, Cl-), as the leaching of sulfate can result in changes to the pore water chemistry and may enhance methylation of mercury by anaerobic microorganisms. Additionally, the successful use of a sorbent technology hinges on the assumption that mercury bound to a sorbent is no longer available for microbial methylation. To test this assumption, mercury bioavailability was investigated via a series of bioassays employing a pure culture of D. desulfuricans ND132 as a model organism for mercury methylation. The extent and rates of mercury methylation were evaluated in the presence and absence of sorbent materials loaded with inorganic mercury. The results collected to date suggest in situ treatment using sorbents may be a viable option for reducing mercury flux from distributed point-source locations in the EFPC watershed.

A new method was investigated for treatment of Hg-contaminated sediments by addition of either pyrolusite (Mn4+O2) or potassium birnessite (K end member of (K, Na, Ca)x (Mn4+, Mn3+)2O4.1.5H2O). The addition of Mn-oxide to Hg-contaminated sediment is hypothesized to buffer redox potential at a level higher than is favorable for Hg methylation. Long-term, sediment tank mesocosm experiments investigated changes in Mn-oxide mineralogy over time and its control on system reduction-oxidation potential. Manganese oxide amendments were either granular pyrolusite (80% MnO2, commercially available) or synthetic powdered birnessite (from KMnO4). Mesocosms consisted of Mn-oxide amendment mixed into the upper 5 cm of wet sediment or applied as a thin-layer sand cap with overlying natural water. Amended sediments were sampled between 4 and 16 months of mesocosm operation, and characterized by X-ray absorption spectroscopy (XAS), powder X-ray diffraction (XRD), and scanning electron microscopy (SEM) to examine changes in Mn solid phases and Mn oxidation state. Manganese in unreacted sediment (bulk concentration 1.3 wt. %) was a mixture of mostly Mn2+-carbonate and sorbed or aqueous Mn2+. Characterization of unreacted birnessite amendment by SEM and XRD indicated poorly crystalline material and particle sizes of a few micrometers. Reducing conditions in the sediment microcosms resulted in transformation of Mn4+-oxides with time. For birnessite-amended sediments, a hydrated Mn3+ oxyhydroxide phase was present after 4 months. With further reaction, Mn was transformed to a combination of Mn2+ carbonate (rhodocrosite) and aqueous or sorbed Mn2+. For pyrolusite-amended sediments, the original Mn4+-oxide was first altered to a mixture of Mn3+ oxyhydroxide and oxide phases with minor Mn2+ carbonate, followed by the transient formation of mixed (Mn3+, Mn2+) oxides, and a progressive increase in the Mn2+carbonate fraction with time. After 16 months of reaction, Mn in solid phases was dominantly Mn2+ carbonate. Slow conversion of Mn4+ oxide to Mn3+ hydrated oxides or mixed-valence (Mn3+, Mn2+)-oxides creates a long-term system redox buffer that makes microbial sulfate reduction less energetically favorable and inhibits methylation of inorganic mercury to methylmercury. Longevity of the amendment treatment to sediments could be controlled to some extent by adjustment of the mass and type of Mn4+ oxide applied, degree of mineral crystallinity, and particle size.

Mercury (Hg) contamination in riverine systems is often widespread in sediments and floodplain soils due to erosion, transport and deposition processes, creating a challenge for remediation. When methylated, Hg becomes much more toxic and can bioaccumulate easily. The primarily biotic process of methylation is often independent of the total Hg concentration, requiring a robust treatment method that can decrease both total Hg and methyl Hg (MeHg). Aqueous Hg leached from contaminated soils and sediments can be present in different oxidation states (e.g., Hg(0), Hg(II)), bound to many sulfur-containing ligands, and complexed with dissolved organic carbon and chloride. The capability of a reactive material to broadly treat these many forms of Hg is key. A series of saturated flow-through column experiments was conducted that evaluated hardwood biochar for removal of Hg derived from a contaminated floodplain soil. These column experiments tested the effectiveness of the biochar with different layer thicknesses and placement geometries relative to contaminated soil, and tracked the impact of the biochar material on MeHg concentrations. Mercury concentrations were tracked both temporally in the effluent and spatially along the column length to determine Hg removal efficiency. Solid samples of the biochar material were collected during and after the experiment from select columns and examined using synchrotron-based confocal micro x-ray fluorescence (CXMFI) spectroscopy techniques. Both the total aqueous Hg and MeHg concentrations declined across the biochar treatment layer in all column experiments. In the column experiments where the biochar was placed immediately adjacent to the contaminated soil, Hg concentrations were observed to increase in the soil pore water, though still declined after passing through the biochar layer, suggesting an interaction between the biochar material and the Hg in the contaminated soil. The results of the CXMFI analysis suggest an increase in Hg loading onto the biochar particles over the course of the column experiments, with Hg observed to be distributed evenly over the particles and adjacent to pore spaces. Overall, the results of the column experiments suggest that hardwood biochar is effective when placed as a treatment layer over contaminated sediment under saturated dynamic flow conditions.

Industrial uses of mercury in the 1950s and 1960s resulted in contamination of multiple watersheds in Oak Ridge, Tennessee. Methylmercury concentrations in fish exceed human and ecological risk benchmarks, necessitating implementation of a number of remedial and abatement approaches over the years to try to address the mercury issue. Early actions in the mid-1980s focused on pollution control that positively affected mercury discharges to one local stream, and included consolidation and elimination of untreated discharges, sanitary sewer relining, the construction of a point-source pollution control facility, and replacement of a contaminated settling basin. The 1990s were an intensive period of facility remedial actions designed to improve stream water quality and mercury release. Actions included dechlorinating cooling water discharges, an additional phase of storm sewer relining, the addition of uncontaminated flow from a nearby reservoir, the addition of two small-scale mercury treatment systems, the bypass of stream flow around the replacement settling basin, and the removal of high mercury-contaminated floodplain soils. As a result of almost two decades of facility actions, water concentrations in the receiving stream near the facility decreased steadily from a high of approximately 1700 ng/L to a low of 400 ng/L in 2000, and fish concentrations declined commensurately from a high around 2 mg/kg to a low of 0.6 mg/kg. The positive responses to these actions led to a continued remedial focus on point-source treatment, with two additional facilities going on line (at two separate facilities and watersheds) in the 2000s. Although these facilities have been successful in decreasing water concentrations in the creeks, research suggested that fish concentrations would not decline until water concentrations were below the Tennessee ambient water quality criterion of 51 ng/L. Further, mercury concentrations in fish farther downstream of one facility did not respond to water concentration decreases, and at some sites fish concentrations even increased over historical levels. Although the remedial strategy in Oak Ridge will continue to focus in the near-term on upstream source reduction, recent research suggests that a better understanding of mercury transport, methylation, and bioaccumulation processes in the downstream environment is essential to reduce fish mercury. Since 2014, mercury remediation research and technology development activities have been underway to develop potential remedial solutions for the downstream environment. The Oak Ridge mercury story is a valuable case study for fluvial systems that can inform our national-level understanding of potential mercury cleanup actions and their likely environmental response.

Remediation of diffuse mercury source zones poses a unique challenge at a wide range of the 3000 mercury-contaminated sites globally. The existence of diffuse sources is particularly challenging in remediating a low-order stream system (i.e., East Fork Poplar Creek [EFPC]) located in Oak Ridge, Tennessee. The EFPC ecosystem received large point-source discharges during the 1950 and 1960s. Although upstream mercury discharges to EFPC have declined, mercury release persists from point and diffuse sources within the industrial facility where mercury was used and from diffuse downstream sources, such as contaminated bank soils. Previous studies identified the presence of mercury sulfide (HgS) in EFPC bank soils, but the processes that govern HgS formation remain unclear. In this presentation, we report the results from high-resolution electron microscopy and secondary ion mass spectrometry measurements to systematically describe the processes that may lead to the formation of HgS enriched particles in soils. Soil samples were collected from EFPC stream banks and analyzed to identify mercury-enriched particles and to determine their size, elemental composition, and sulfur isotopic ratio. Results from the energy-dispersive X-ray spectroscopy data, confirmed that mercury is generally collocated with sulfur in mercury-enriched particles in EFPC bank soils. Further analysis of the microscopy images indicates that smaller HgS particles hundreds of nanometers in size, aggregate to form the larger micron-sized HgS clusters of 0.15 µm to 4.2 µm in diameter with an average size of 1.4 ± 1.1 µm. We suggest that, these nanometer-sized HgS particles are formed as a result of the precipitation of mercury with microbially produced sulfide. Nanoparticulate or colloidal HgS is widely recognized as a potential source of bioavailable mercury for methylating bacteria. Understanding the mobility and bioavailability of these nanometer-scale particles is an important step in predicting MeHg production in ecosystems.

According to the acceptance of the Minamata convention on mercury in 2013, final disposal of unused mercury will be requested in the near future. Engineered landfill sites will be one of feasible options for final disposal where dry permanent storage in deep and impermeable underground is geologically impossible. Environmental impact assessment of mercury disposal in engineered landfill sites is necessary to gain public acceptance of mercury final disposal. Numerical simulation using mercury environmental fate models is useful to estimate mercury emissions from landfill sites and its risk. Because the model must include many physical, chemical, and biological reactions like diffusion, adsorption, methylation, and bioaccumulation, it will use many parameters. These parameters strongly depend on the local environment and their values have been reported in the ranges of several orders of magnitude.

The purpose of this study is to assess the impact of model parameter uncertainty on mercury emission from a landfill site and methylmercury exposures to human bodies via polluted fish. A landfill simulation model including mercury reactions was used to assess the impact of model parameter uncertainty on mercury emission from a landfill site. This model includes rainfall penetration into a landfill site, unsaturated water percolation, mercury diffusion, mercury transfer, adsorption, desorption, methylation, demethylation, chemical reduction to elemental mercury, gaseous elemental mercury diffusion in a landfill site, mercury emissions via landfill leachate and gaseous mercury emission from landfill surface. Methylmercury exposures were evaluated using a closed-lake simulation model. This model assumes constant mercury input from the outside and considers mercury reactions including methylmercury biomagnificaiton via food-chain. Precipitation conditions mainly controlled mercury emission. When possible variations of annual precipitation based on 135-year weather records in Japan were considered, mercury emissions at 100-year later since disposal completion was simulated to range from 1.2 to 31 mg-Hg/m2/yr. On the other hand, methlymercury exposures varied much more greatly owing to large uncertainty of model parameters like biomagnification factor. When log-normal distribution were assumed for major model parameters, methlymercury exposures at 10000 years later ranged from 0.122 to 38896 μg-Hg/week/person (65 kg body weight). These results suggest that mercury emissions and exposures to human body, simulated using mercury environmental fate models, inevitably include large uncertainty owing to model parameter uncertainty. Environmental risk of mercury final disposal in a landfill site must be assessed with considering large and unavoidable uncertainty of model simulation results.

Mercury (Hg) contamination of watersheds due to the release of mercury compounds from industrial activities is a world-wide concern. Mercury compounds once released to the environment can contaminate aquatic systems and provide a long-term non-point source of contamination to downstream and surrounding environments. Mercury stabilization in fluvial settings is challenging due to frequent changes in environmental conditions, including geo-chemical conditions such as redox potential, and physical conditions associated with groundwater discharge/recharge, precipitation, and flooding. A robust remediation system that can stabilize Hg under flooding and drainage conditions without producing unintended consequences is critical to ensure long-term management of contaminated sites in fluvial settings. In this study, hardwood biochar prepared from oak wood was evaluated as a potential reactive media for use in passive reactive mats to stabilize Hg from contaminated river bank sediment and floodplain soil collected near the South River, VA. The purpose of this study is to evaluate the effectiveness of the hardwood biochar on Hg transport and net methylmercury (MeHg) production as a passive reactive mat under environmentally relevant flooding and drainage conditions. Two sets of columns were packed with 50 % v/v of hardwood biochar and quartz sand in an experiment designed to simulate periodic flooding and drainage conditions. The cycling experiment started with dry air and water saturated air, followed by the addition of influent solutions collected from the leachates from contaminated sediment and floodplain soil. After 100 weekly cycles, more than 80% of the Hg present in the sediment and floodplain soil leachates were retained on the biochar, with limited formation of MeHg in both aqueous effluent and the solid materials. The Hg retained on the biochar is likely to present within the porous structure in a form of Hg-S as indicated by micro X-ray absorption spectroscopy (-XAS) and micro X-ray fluorescence (-XRF) maps. A disappearance of sulfoxide functional groups, indicated by sulfur K-edge X-ray near edge absorption spectroscopy (XANES), was observed for the biochar collected from the treatment columns at termination of the experiment. The synchrotron-related analyses suggest that the removal of Hg may involve both filtration by the porous structure and complexation with functionalities or ligands on the biochar. The study results suggest that hardwood biochar may be an effective media for stabilizing Hg under cyclic flooding and drainage conditions without promoting an increase in Hg methylation in both aqueous and solid phases.

An estimated 60,000 kg of mercury was discharged into the liquid waste system at the Savannah River Site located near Aiken, South Carolina USA. The mercury is isolated within process vessels and storage tanks, with minimal releases to the surrounding environment. Inorganic mercury was used as a catalyst to aid in the dissolution of aluminum fuel and target assemblies associated with separation processes that supported the US nuclear stockpile. The typical concentration of total mercury in the system is on the order of 100 mg/L. These levels are approximately six orders of magnitude higher than the concentrations that have been studied in most environmental systems.

Liquid waste is a concentrated sodium nitrate solution containing numerous radioactive and non-radioactive elements and ions along with residual inorganic and organic process constituents. The system currently stores approximately 37 million gallons of waste containing approximately 287 million curies of radioactivity. Phases present include an alkaline, high ionic strength liquid phase that is maintained at a high pH (salt solution), a solid phase containing precipitated salts and other solids species (sludge), and a vapor phase associated with the headspace of process vessels. Mercury reactions in the complex-alkaline environment has resulted in the presence of solid, liquid (elemental), vapor, and dissolved mercury species. Over the past two years, there has been an intense effort to understand the chemistry of mercury within the process to determine impacts on the final waste forms. These results revealed that the mercury is present in the elemental state, solid phase, soluble ionic mercury complexes, and methylated species in both the vapor and aqueous phases. Within the aqueous phase organic mercury is the predominant form in several of the process vessels.

Analytical results from the salt solution and sludge indicate that as the high level radioactive sludge is vitrified into glass for final disposal, mercury is currently being recycled within the process and non-volatile mercury species are concentrated by the evaporator systems. Consequently, mercury concentrations are increasing in the salt solution. This is impacting the final radioactive waste form of this material - a solid concrete matrix called saltstone. The presence of organomercury significantly reduces the effectiveness of mercury removal processes built into the system and increases the leachability of mercury from the final disposition waste form. Strategies for the management of organic mercury are being developed to prevent long-term impacts to the disposal of the liquid waste.