G7 (I, II, III) Environmental Biogeochemistry: Environmental Approaches

Monday, 25 July, 2011

MG7-P1 — 11:00-12:00 and 17:30-18:30
Authors: BRADLEY, Paul1, JOURNEY, Celeste A.1, CHAPELLE, Francis H.1, CONRADS, Paul A.1, LOWERY, Mark A.1
(1) USGS, pbradley@usgs.gov

Mercury (Hg) burdens in top-predator fish differ substantially between adjacent South Carolina Coastal Plain river basins with similar wetlands coverage. In the Congaree River, floodwaters frequently originate in the Blue Ridge and Piedmont regions, where wetlands coverage and surface-water dissolved methylmercury (MeHg) concentrations are low. Piedmont-driven flood events can lead to downward hydraulic gradients in the Coastal Plain riparian wetland margins, inhibiting MeHg transport from wetland sediments, and decreasing MeHg availability in the Congaree River habitat. In the adjacent Edisto River basin, floodwaters originate only within Coastal Plain sediments, maintaining upward hydraulic gradients even during flood events, promoting MeHg transport to the water column, and enhancing MeHg availability in the Edisto River habitat. These results indicate that flood hydrodynamics contribute to the variability in Hg vulnerability between Coastal Plain rivers and that comprehensive regional assessment of the relationship between flood hydrodynamics and Hg risk in Coastal Plain streams is warranted.

MG7-P2 — 11:00-12:00 and 17:30-18:30
Authors: TSUI, Martin Tsz Ki1, FINLAY, Jacques C.2, BALOGH, Steven J.3, NATER, Edward A.4
(1) Department of Geological Sciences, University of Michigan, mtktsui@umich.edu; (2) Department of Ecology, Evolution & Behavior, University of Minnesota; (3) Metropolitan Council Environmental Services; (4) Department of Soil, Water & Climate, University of Minnesota.

In temperate streams, litter fall represents an important source of allochthonous carbon but also a potentially significant input of inorganic mercury (Hg) into the stream ecosystems. In this study, we performed in-situ litterbag decomposition experiments on maple litter (Acer spp.) in ten Minnesota streams that varied widely in land cover and water chemistry. We monitored total-Hg, methyl-Hg and a suite of trace elements and nutrients in both water and leaf litter from late fall of 2009 through early spring of 2010. In all streams, [total-Hg]litter increased over time; the highest [total-Hg]litter (91 ng/g d.w. compared to 30 ng/g d.w. in the original litter) were observed in two highly urbanized streams in Minneapolis and St. Paul, Minnesota. We found that [Fe]litter were significantly elevated (~ 400 times the original [Fe]litter) in these two streams, implying that deposition and incorporation of inorganic suspended particles onto the decomposing litter may partly contribute to the observed elevation of [total-Hg]litter. On the other hand, the highest [methyl-Hg]litter (6.2 ng/g d.w. compared to 0.9 ng/g d.w. in the original litter) were found in streams having the highest average [methyl-Hg]water, especially those draining forested and wetland landscapes, suggesting potential contribution due to the incorporation of aqueous MeHg into decomposing litter. Overall, there are substantial changes in Hg concentrations and speciation associated with leaf litter decomposition in stream ecosystems, that may influence Hg incorporation into stream food webs. Our study therefore provides an indication that these chemical changes of decomposing litter can be controlled by particle deposition (for total-Hg) and/or aqueous species partitioning (for methyl-Hg). Therefore, Hg pools within decomposing litter appear to be dynamic and influenced by in-situ water characteristics.

MG7-P3 — 11:00-12:00 and 17:30-18:30
Authors: TSUI, Martin Tsz Ki1, SEBESTYEN, Stephen D.2, FINLAY, Jacques C.3, NATER, Edward A.4, JEREMIASON , Jeff D. 5, JACOBSON, Meghan3
(1) Department of Geological Sciences, University of Michigan, mtktsui@umich.edu; (2) USDA Forest Service, Northern Research Station; (3) Department of Ecology, Evolution & Behavior, University of Minnesota; (4) Department of Soil, Water & Climate, University of Minnesota; (5) Department of Chemistry, Gustavus Adolphus College;

Wetlands such as peatlands are well known as sources of highly toxic methylmercury (MeHg) to downstream ecosystems. Aqueous mercury concentrations (i.e., [total-Hg] and [MeHg]) and yields from peatland watersheds are known to vary with hydrological conditions, and therefore any major shifts in climatic regimes (e.g., changing intensity and frequency of rainfall) may substantially influence MeHg export from these MeHg hotspots. The current study was aimed to explore the relationships between Hg export and climate, and to elucidate potential mechanisms mediating these relationships. Our study site is the S2 watershed that includes a 3.2-ha bog and a 6.5-ha upland at the Marcell Experimental Forest located in northern Minnesota. In both 2009 and 2010, we examined filtered [total-Hg], [MeHg], and a suite of chemical tracers in runoff waters that flow through upland mineral soils, pore waters in the lagg zone interface between bog organic soils and upland mineral soils, and the stream that flows from the watershed. Samples were collected twice per week when the stream was flowing, with more samples collected during 2010 due to the more frequent rainfall in the summer. Annual yields of water and total-Hg did not differ remarkably between 2009 and 2010 from the study watershed. However, we found that total annual export of MeHg was significantly higher in 2010. Interestingly, we found that [total-Fe] positively correlated with [MeHg] in streamflow, and [total-Fe] were significantly higher in streamflow during 2010. The export of Fe should be associated with the anoxic reduction of Fe within the bog zone. Therefore, we hypothesize that Fe-reducing bacteria may be important methylators of Hg in this sulfate-deficient peatland ecosystem, and this may result in increases in both Fe and MeHg exports during the wetter year of 2010. Our findings also suggest greater inter-annual variation in annual MeHg export with climate change when more extreme weather is projected in the coming decades.

MG7-P4 — 11:00-12:00 and 17:30-18:30
Authors: MURPHY, Julie W.1, GUENTZEL, Jane L.1
(1) Coastal Carolina University, jwmurphy@coastal.edu

The Waccamaw is a blackwater river located along the eastern coastline of South Carolina. The fish in this river contain concentrations of mercury that have resulted in the issuance of fish consumption advisories. Large areas of the shoreline contain solid masses of water hyacinths during the warmer months (April-December). When hyacinths grow they form dense floating mats on the surface which can alter water chemistry and microbial populations below the mats. These mats can block light and oxygen penetration creating low to no oxygen environments near the surface that are conducive to the transformation of inorganic mercury to methylmercury. This study quantified the chemical and biological parameters of the water column near and under these mats that may promote the formation of methylmercury and quantified the amount of total and methylmercury in these plants as they mature. Water samples were collected monthly (May 2010 to November 2010) from underneath a water hyacinth mat and from an open water site. Dissolved oxygen levels under the mat ranged from 1.9-4.1 mg/L (25.6-41.7 % saturation) and from 3.8-7.7 mg/L (47.4-75.3 % saturation) at the open water site. Sulfate reducing bacteria (SRB) ranged from 0-100 cells/mL at the open water site and 0-10,000 cells/mL in the water under the mat. The SRB on the plant roots ranged from 0-603,500 cells/wet gram of root. Total Hg in the water underneath the mat ranged from 3.2-8.9 ng/L with 10-31% MeHg relative to the Total Hg. Total Hg concentrations at the open water site ranged from 2.9-7.6 ng/L with 9-23% MeHg relative to the Total Hg. Plant samples were collected monthly (May 2010 to November 2010). Total Hg in the roots ranged from 30.1-71.9 ng/g dry with 25-52% MeHg relative to the Total Hg. Total Hg in the shoots ranged from 2.2-58.7 ng/g dry with 2-63% MeHg relative to the Total Hg. The amount of Hg in a square meter of water hyacinth ranged from 8,947-52,420 ng Total Hg/m2 dry and 2,759-15,790 ng MeHg/m2 dry. The low oxygen levels, the presence of SRB, and the high percentages of methylmercury in the water and plants suggest that these mats may serve as a site for mercury methylation. These findings also suggest that water hyacinth plants bioaccumulate mercury and may be a source of mercury to organisms that consume these plants.

MG7-P5 — 11:00-12:00 and 17:30-18:30
Authors: GUENTZEL, Jane1, MURPHY, Julie W.1, GRAVES, David2, GLOVER, James2, LASORSA, Brenda3, GILL, Gary A.3
(1) Coastal Carolina University, jguentze@coastal.edu; (2) South Carolina Department of Health and Environmental Control; (3) Pacific Northwest National Laboratory;

Mercury exists in many different physical and chemical forms in the environment and it is the interconversions between these species that mediate its distribution patterns and biogeochemical cycling. The most widely known conversion is the biological transformation of inorganic Hg (II) to methyl Hg and its subsequent biomagnification in piscivorous fish, which poses a risk to higher trophic level organisms and humans who consume these fish. The discovery of high levels of Hg (>0.25 ppm) in fish from water bodies in South Carolina (SC) has resulted in the issuance of many fish consumption advisories throughout the state. The highest fish tissue concentrations have been found in unregulated blackwater rivers in the Southeastern Plain and the Middle Atlantic Coastal Plain ecoregions, with lower fish Hg concentrations found in the Southern Coastal Plain, Blue Ridge, and Piedmont ecoregions. Investigating the geochemical cycling of Hg in aquatic systems is a necessary precursor to understanding the mechanisms responsible for its accumulation in higher trophic levels. The SC Clean metals project was initiated in 2010 and includes 4 quarterly sampling events. The first objective of this study is to determine levels of total and methyl Hg in water and sediments from sites along a geographical gradient (west-east) within SC. Sites within differing ecoregions were chosen to identify potential factors governing the spatial variability of Hg levels in water and sediment throughout the state and how they may relate to the concentration of Hg in fish tissue. A second objective of the study is to examine how the mercury levels in the water and sediment may change on a seasonal basis. We will report data from the first (November 2010) and second (February 2010) sampling events.

MG7-P6 — 11:00-12:00 and 17:30-18:30
Authors: NAIK, Avani P.1, HAMMERSCHMIDT, Chad R.1
(1) Wright State University, avani.p.naik@gmail.com

Many natural watersheds are increasingly affected by changes in land use associated with suburban sprawl and such alterations may influence concentrations, partitioning, and fluxes of toxic trace metals in fluvial ecosystems. We investigated the cycling of mercury (Hg), monomethylmercury, cadmium, copper, lead, nickel, and zinc in three watersheds at the urban fringe of Dayton, Ohio, over a 13-month period. Metal concentrations were related positively to discharge in each stream, with each metal having a high affinity for suspended particles and Hg also having a noticeable association with dissolved organic carbon. Although not observed for the other metals, levels of Hg in river water varied seasonally and among streams. Yields of Hg from two of the catchments were comparable to that predicted for runoff of atmospherically deposited Hg (~25% of wet atmospheric flux), whereas the third watershed had a significantly greater annual flux associated with greater particle-specific and filtered water Hg concentrations, presumably from a point source. Fluxes of metals other than Hg were similar among each watershed and suggestive of a ubiquitous source, which could be either atmospheric deposition or weathering. Results of this study indicate that, with the exception of Hg being increased in one watershed, processes affecting metal partitioning and loadings are similar among southwest Ohio streams. Relative differences in land use, catchment area, and presence or absence of waste water treatment facilities had little or no detectable effect on most trace metal concentrations and fluxes. This suggests that suburban encroachment on agricultural and undeveloped lands has either similarly or not substantially impacted trace metal cycling in streams at the urban fringe of Dayton and, by extension, other comparable metropolitan areas.

MG7-P7 — 11:00-12:00 and 17:30-18:30
Authors: TABATCHNICK, Melissa D.1, NOGARO, Geraldine1, HAMMERSCHMIDT, Chad R.1
(1)Wright State University, melissa.tabatchnick@gmail.com

Cycling of mercury (Hg) and monomethylmercury (MMHg) in forest ecosystems can affect exposures of terrestrial and aquatic wildlife within the watershed. Litterfall has been posited to be a major source of MMHg and Hg to the forest floor; however, the origin of MMHg associated with tree foliage is largely unknown. We tested the hypothesis that leaf MMHg is controlled by root uptake and thereby proportional to levels in soil. Fresh leaves and associated soil samples were sampled from nine tree species (deciduous and coniferous) at 30 locations spanning a 1100 km2 area in southwest Ohio, a region presumed to have relatively homogeneous atmospheric deposition of Hg and MMHg. Concentrations of Hg species in tree leaves were unrelated to those in soil. In contrast, tree genera and trunk diameter were dominant variables influencing Hg levels in tree foliage. The fraction of total Hg as MMHg in leaves was relatively constant among all genera and averaged 0.4%. Results of this study suggest that uptake of gaseous Hg0 from the atmosphere is the dominant source of total Hg in foliage and that MMHg is formed by in vivo transformation of Hg(II) in proportion to the concentration accumulated. Via litterfall, it appears that processes occurring within tree leaves are a major source of Hg and MMHg to the forest floor.

MG7-P8 — 11:00-12:00 and 17:30-18:30
Authors: RODRIGUES, Sónia1, HENRIQUES, Bruno G.1, REIS, Ana T.1, DUARTE, Armando C.1, PEREIRA, Eduarda1, RÖMKENS, Paul F.A.M.2
(1) Universidade de Aveiro, smorais@ua.pt; (2) Alterra - Wageningen UR.

Understanding the transfer of mercury (Hg) from soil to crops is crucial to derive soil quality criteria that are effective to ensure feed and food safety.

We performed a field study in urban, agricultural, mining and industrial areas in Portugal to assess uptake of Hg by Italian ryegrass, ryegrass, orchard grass, collard greens and rye. In total, 160 soils and 153 plants were analysed, originating from both clean and polluted sites. Total Hg concentrations in soils varied from 0.01 to 98 mg kg-1. Mercury levels ranged from 0.01 to 6.9 mg kg-1 in shoots and from 0.01 to 36 mg kg-1 roots. Levels of organic Hg were less than 0.1% of the total levels in both soil and plants. The elevated levels of Hg in crops suggest that plants are capable of enhancing the transfer of inorganic Hg from metal enriched soils into the above ground biomass and hence the food chain.

To predict Hg levels in plants Freundlich-type empirical models were derived from the combined soil – plant data using total soil Hg concentrations and soil properties as explanatory variables. For ryegrass, orchard grass and collard greens the soil to root and shoot transfer of Hg were controlled both by the total Hg in soils and, partly, by amorphous Al or Fe oxides. The latter observation is possibly related to the influence of such oxides on the solid: solution partitioning of Hg thereby controlling the availability of Hg for root uptake. For Italian ryegrass however, total Hg levels in soil alone explained 98% and 95% of the variance in root and shoot Hg levels despite differences in soil properties.

Freundlich-type empirical models were used to back-calculate threshold concentrations of total Hg in soils from EC feed quality limits (for green fodder). Threshold concentrations were 1.7 mg Hg kg-1 for Italian ryegrass, and 1.9 to 2.8 mg Hg kg-1 for ryegrass depending on the Al oxide content (50 mmol kg-1 and 100 mmol kg-1 respectively).

Soil-plant Hg transfer functions allow to correct for regional differences in soil properties in order to obtain realistic estimates of Hg crop levels. Also, soil-plant transfer functions can form the basis of a coherent set of soil quality criteria in order to meet feed and food quality demands.

MG7-P9 — 11:00-12:00 and 17:30-18:30
Author: LAMBORG, Carl1
(1) Woods Hole Oceanographic Institution, clamborg@whoi.edu

In collaboration with on-going studies of nutrient and trace metal geochemistry in the Massachusetts Military Reserve wastewater plume, we have determined the mercury concentration and speciation in groundwater collected from monitoring wells arrayed longitudinally along the plume. In these often anoxic waters, and most notably in locations where iron reduction was evident, speciation shifted toward production of the elemental form, enhancing subsurface mobility of mercury. Enhancements of methylated mercury was also observed in these samples, especially in the dimethylated form. Highly elevated concentrations of total mercury were found in groundwater directly beneath the point where wastewaters were allowed to infiltrate the aquifer. These findings are examined in the wider context of residential wastewater management on Cape Cod where private septic systems are the norm, and where groundwater discharge into coastal marine ecosystems is a significant local biogeochemical process.

MG7-P10 — 11:00-12:00 and 17:30-18:30
Authors: GITTINGS, Michael 1, CARPI, Anthony1
(1) John Jay College, mgittings87@gmail.com

Intact forests and soils within these systems serve as a sink for mercury, diminishing or delaying its movement in the environment and entry into the human food chain. Deforestation can release mercury from these ecosystems as a result of the burning of biomass or following the burn event, as emissions from exposed soils. Margarelli and Fostier (2005) have shown that soil mercury emissions from a deforested site in the Negro River Basin were statistically higher than soil emissions from an intact forested area. Thus, the liberation of mercury stored in forest soils may be an important source of mercury in regional ecosystems. To further investigate the impact of deforestation on soil mercury emissions, we conducted a two week study in the Blackrock Forest in Cornwall, NY. An intact site, covered with European and Japanese Barberry (Berberis vulgaris and Berberis thunbergil), was studied for one week with two Teflon dynamic flux chambers connected to a Tekran Mercury Vapor Analyzer to obtain continuous background mercury flux readings. At the start of the second week, vegetation was removed from the plot by cutting, and emissions were monitored for an additional week. Mercury flux during the pre-clearing period averaged -1.88ng/m2/hr and showed peak emissions of 8.3ng/m2/hr. However, post-clearing flux values averaged 10.62ng/m2/hr and showed peak emissions of 107ng/m2/hr. The maximum value seen is significantly greater than previously reported. Contributing factors will be discussed.

MG7-P11 — 11:00-12:00 and 17:30-18:30
Author: NOH, Seam1
(1)gwangju institute of science and technology, seam@gist.ac.kr

The estuarine high turbidity zone (HTZ) is well known to provide a temporary trap for riverine and marine particles, and these particles undergo a diagenetic reaction in sediments. However, the involvement of Hg speciation in the estuarine HTZ is not yet fully understood. In the current study, concentrations of Hg species, as well as compositions of suspended particles and sediments were investigated in both the low turbidity zone (freshwater) and the HTZ (brackish water) of the Tieu Estuary located in the Mekong River. As freshwater particles with high organic content (5.3±2.1%) were mixed with suspended sediments in the HTZ, particulate organic matter in suspended loads decreased in the HTZ (1.7±0.17%). The total Hg concentrations in unfiltered surface waters were higher in the HTZ (120±81 pM) than in the freshwater (28±16 pM) due to increased suspended loads in the HTZ. In contrast, concentrations of unfiltered monomethylmercury (MMHg) in surface water decreased in the HTZ (0.24±0.073 pM) than in the freshwater zone (0.12±0.045 pM). The decreased unfiltered MMHg concentration in the HTZ was a result of decreased particulate MMHg associated with the low particulate organic matter content. Regarding sediment Hg speciation, the total Hg and MMHg contents were similar between the freshwater zone and the HTZ. Large amounts of acid volatile sulfide (AVS) detected in the HTZ sediments (3.6±2.6 mmol g-1) appears to limit the active methylation of inorganic Hg(II) in those sediments. Overall, despite of intense organic degradation, increased AVS in sediment and decreased POM in SPM impeded active production and water column transfer of MMHg, respectively, in an estuarine HTZ.

MG7-P13 — 11:00-12:00 and 17:30-18:30
Authors: KONKLER, Matt J.1, HAMMERSCHMIDT, Chad R.1
(1) Wright State University, konkler.2@wright.edu

Emissions from coal-fired utilities are the major anthropogenic source of mercury (Hg) to the atmosphere. Because some of the emitted Hg may be atmospherically deposited near the source, there are concerns over potential impacts on levels of toxic monomethylmercury (MMHg) in local biota. Recent investigations have suggested a link between atmospheric deposition of inorganic Hg and the accumulation of MMHg in aquatic biota. MMHg concentrations in mosquitoes (Diptera: Culicidae), which are ubiquitous and have aquatic life stages, have been shown to be a useful and sensitive indicator of Hg loadings, including those derived from atmospheric deposition. We investigated the potential impact of a large Hg-emitting (> 500 kg Hg y–1) coal-fired electrical utility on MMHg levels in mosquitoes near the Conesville power station in central Ohio. Adult, host-seeking female mosquitoes were sampled with CO2-baited traps at 23 locations within a 60 km radius of the utility during a two-week period in July 2010 as well as two reference sites distant from known combustion sources of Hg. MMHg levels in mosquitoes sampled at locations within a 30-km radius of the plant (n = 12) were comparable to those at locations within a 30–60 km radius (n = 11) and the two control sites. These preliminary findings suggest that either 1) little of the Hg emitted from the utility is deposited locally or 2) near-source deposition of Hg emissions from the plant does not have a significant impact on MMHg residues in mosquitoes and, by extension, potentially other organisms in the local food web.

MG7-P14 — 11:00-12:00 and 17:30-18:30
Authors: HARRIS, Reed C.1, HUTCHINSON, David H.1, BEALS, Don I.2
(1)Reed Harris Environmental Ltd., reed@reed-harris.com; (2) Beals and Associates.

A regression model has been developed to predict the relative increase in fish mercury (Hg) concentrations following reservoir creation, based on the extent of flooding and flow. Fish Hg levels in reservoirs on the Canadian Shield increase following flooding, often exceeding levels associated with consumption advisories. Peak fish Hg levels tend to occur 5-15 years after flooding, then decline towards background levels 2-3 decades post-flood. The primary cause of increased fish Hg in reservoirs is flood-zone decomposition, stimulating bacterial production of methylmercury (MeHg), the dominant Hg form in fish. Observed relative increases in fish Hg concentrations (peak/baseline) vary significantly among Shield reservoirs, ranging from roughly 1.5 to 7 times above background levels in adult sportfish. The purpose of the model is to predict whether future reservoirs will result in low, moderate or high relative increases in fish Hg levels. The model approach is empirical but has mechanistic underpinnings. It is assumed that the primary source of MeHg in new reservoirs is flooding, represented by the fraction of the total reservoir area that is flooded. Removal mechanisms in the model include outflow and a second term aggregating processes that are effectively area based (sedimentation, photochemical and biological degradation). The model has the following form: PIF = a * (Af/(Q+b*At)) + c, where PIF = peak relative increase factor above baseline (dimensionless), Af and At = flooded and total area respectively (km2), Q = mean annual flow (km3 yr-1), and a, b and c are calibration coefficients. Peak fish Hg concentrations can be predicted by multiplying relative increases by pre-flood baseline levels. The model has been calibrated for northern pike and walleye using observations from 11 reservoirs on the Canadian Shield, reasonably fitting observations. Results improve when reservoirs in series are treated as a single reservoir increasing in size travelling downstream, subject to the reservoirs being flooded within a decade and being within 200 km of each other, with no major additional water inputs in between. Model results suggest that flow can be an important factor moderating the relative increase in fish Hg levels in reservoirs. Run of the river situations (high flow, low flooding) are still predicted however to result in a 50-100% increase in fish Hg. This could be due to flow dilution of MeHg in the water column but not in sediments, which are also sources of MeHg to aquatic food webs.

MG7-P15 — 11:00-12:00 and 17:30-18:30
Authors: DEWILD, John1, THOMPSON, Charles1, DAVID, Krabbenhoft1, KNAUER, Doug 2, GEORGE, Pelkola2
(1) US Geological Survey, jfdewild@usgs.gov; (2) Michigan Department of Natural Resources and Environment;

For many years there has been a well recognized inverse relationship between the abundance of mercury (Hg) and methylmercury (MeHg) to selenium (Se) in aquatic food webs. Twenty years ago Swedish scientists demonstrated pronounced declines in fish Hg concentrations upon dosing with Se and considered it a possible treatment method for Hg impacted lakes. While the response of Se loading on Hg bioaccumulation is clear, the mechanistic reasons remain relatively poorly defined. To further explore this phenomenon, the U.S. Geological Survey initiated a study on a Se-enriched lake (Goose Lake) in the Upper Peninsula of Michigan. Attention was initially drawn to Goose Lake when prey fish results showed abnormally low Hg levels. This revelation was especially surprising given the lake’s sediment Hg concentrations are significantly (~5-10x) above local background levels. This study is designed to assess how elevated Se levels lead to pronounced suppression of bioaccumulted Hg in prey fish from Goose Lake.

Field efforts involved the collection of sediment, water column profiles, and fish. Because Goose Lake sediments are elevated in both Se and Hg, similar samples were collected at three nearby lakes to assess the effects of Se on MeHg production and bioaccumulation. Reference lake 1 (Deer Lake) sediments are elevated in Hg withSe levels near background. Reference lake 2 (Cedar Lake) sediments have background levels of both Hg and Se (each about 10x lower than Goose Lake). Reference lake 3 (South Bay) is a former bay of Goose Lake now isolated by a train bed, with background levels of Hg and Se. Results from sediment analysis (top 5 cm) reveal that similar levels of MeHg are present in Goose and Deer lakes (>5 ng/g, dry weight); whereas Cedar Lake and South Bay have comparatively less (<1 ng/g). Methylation assays using stable Hg isotopes reveal 12 hour production rates of less than 0.1 percent for Goose and Deer Lakes; whereas rates at Cedar Lake and South Bay were 0.1 to 0.3 percent. Results from the water column were more revealing of Se effects. Aqueous MeHg levels in Goose Lake are 6-12x lower than Deer and South Bay, but similar to Cedar. However, MeHg/THg ratios and the fraction of dissolved MeHg at Goose lake is significantly lower (2% and 22%, respectively) compared to the other three lakes (9-24% and 45-68%).

MG7-P16 — 11:00-12:00 and 17:30-18:30
Authors: CAI, Yong1, LIU, Guangliang1, NAJA, G. Melodie2, KALLA, Peter3, SCHEIDT, Dan3, TACHIEV, Georgio1, ROELANT, David1
(1) Florida International University, cai@fiu.edu; (2) Everglades Foundation; (3) U.S. Environmental Protection Agency, Region 4;

Development of mass inventories of total Hg (THg) and methylmercury (MeHg), which include inputs, outputs, and storage in each ecosystem component (e.g., water, soil, flocculent detrital organic matter (floc), periphyton, macrophyte, and fish), is important for evaluating the magnitude of mercury (Hg) contamination in the Florida Everglades, a large freshwater wetland ecosystem with subareas connected by water flows. The data obtained through the U.S. Environmental Protection Agency (EPA) Everglades Regional Environmental Monitoring and Assessment Program (R-EMAP), along with other datasets which include the USGS Aquatic Cycling of Mercury in the Everglades (ACME), the Mercury Deposition Network (MDN), and the South Florida Water Management District (SFWMD) DBHYDRO database, were utilized to construct mass inventories of THg and MeHg for different subareas of the Everglades. Masses of THg storage in Water Conservation Area 1 (WCA 1), WCA 2, WCA 3, and the Everglades National Park (ENP) are 914, 1138, 4931, and 7602 kg, respectively, with soil being the largest sink. The current annual Hg inputs (primarily from atmospheric deposition) account for 1-2% of the legacy THg storage. Masses of MeHg storage in WCA 1, WCA 2, WCA 3, and ENP are 15, 6.8, 32, and 51 kg respectively, with soil still being the largest sink but a considerable fraction of MeHg being entrapped in floc (in particular in WCAs), indicating the importance of floc in MeHg cycling. Albeit in a minor amount, water flow can transport THg or MeHg across subareas, with seasonal transport amounting to 1% or less of THg or MeHg storage, except for WCA 2 where inflow inputs can contribute 4% of total MeHg storage. The potential influence of the ongoing Everglades restoration (e.g., increasing water deliveries to the wetland) on the across-subarea Hg transport and subsequent ecosystem-scale Hg cycling should be carefully evaluated.

MG7-P17 — 11:00-12:00 and 17:30-18:30
Author: KELLY, Mark D1
(1) Queen’s University, 8mdk@queensu.ca

Ontario provincial records indicate that fish total mercury (THg) has significantly declined since the 1970’s, yet THg in precipitation has remained relatively stable during this period. With three likely hypotheses; i) changes in food web structure, ii) changes in fish growth rates or iii) declines in acid sulphur (S) deposition, known links to bioavailability directed an assessment of acid precipitation as an influence on fish THg. We examined temporal trends for THg and S in sediment cores with fish tissues from two lakes that differ in S-reduction rates (SRR), Kahshe Lake and Mountain Lake, in south-central Ontario. Hypothesizing that lakes with anoxic hypolimnion increase SRR, we expect increasing Hg bioavailability. Conversely, lakes with oxic hypolimnion may reflect trends of mercury deposition. Between-lake variability of fish THg is compared to describe changes related to S accumulation in sediments. Paleolimnological records for Kahshe Lake show distinctive trends of increasing S from 1952 to 1990 followed by declines to present, whereas Mountain Lake S remains unchanged over 150 years. Preliminary results of fish THg in Kahshe Lake appear to support our hypothesis. Archived fish samples provide fish THg to be paired with sediment records through an unprecedented 84 years to further evaluate this relationship.

MG7-P18 — 11:00-12:00 and 17:30-18:30
Authors: GRAYDON, Jennifer A. 1, EMMERTON, Craig A. 1, GAREIS, Jolie A. L.2, NAFZIGER, Jennifer1, LESACK, Lance F. W. 2, HICKS, Faye1
(1) University of Alberta, jgraydon@ualberta.ca; (2) Simon Fraser University;

Concentrations of toxic monomethyl mercury (MeHg) in the tissues of marine mammals living in Canada’s Beaufort Sea (Arctic Ocean) can be above those recommended for human consumption. Several factors likely contribute to MeHg contamination of these mammals, however the large volume of sediment-rich water delivered by the Mackenzie River cannot be overlooked as a potential source of mercury (Hg) contamination to this region. Currently, it is not known how much Hg is delivered from the Mackenzie River each year and therefore the river remains an important, yet unmeasured, source of Hg to the Beaufort Sea. We have been sampling surface waters from the lower Mackenzie River and its delta channels for total Hg (THg; all forms of Hg in a sample) and MeHg since 2004 to help establish a long-term dataset of Hg delivery from the Mackenzie River. There are two overarching goals of our ongoing research and monitoring program. First, we are measuring the amount of THg and MeHg delivered annually by the Mackenzie River to the Beaufort Sea using extensive ice-influenced and open-water season field sampling campaigns and a dynamic ice-jam flow model (Mackenzie Delta Hydrodynamic Model) through a partnership with the International Polar Year Project Study of Canadian Arctic River-delta Fluxes (IPY-SCARF). This model is used to provide flow-distribution data, and to enhance our understanding of how the hydraulics of the delta may change under global change and resource development scenarios. Secondly, we are quantifying the influence of the ecologically-rich, yet rapidly changing, floodplain of the Mackenzie River Delta (MRD) on the delivery of Hg from the Mackenzie River. Here we present results from our sampling program and modeling including multi-year THg and MeHg concentrations and export from several locations throughout the MRD.

MG7-P19 — 11:00-12:00 and 17:30-18:30
Authors: BRAVO, Andrea1, GUEDRON, Stephane2, ZOPFI, Jokob3, AMOUROUX, David4, WILDI, Walter5, DOMINIK, Janusz5
(1) Geneva University , andrea.garcia@unige.ch; (2) LGIT Grenoble; (3) Neuchatel University; (4) univ-pau.fr; (5) Geneva University;

Mercury (Hg) is one of the most hazardous contaminants in sediments of Vidy Bay, which receives treated domestic and industrial wastewater and is recognized as the most contaminated area of Lake Geneva. High concentrations of iron have been found in Vidy Bay sediments due to the addition of Fe(III)-oxides as de-phosphatation treatment in the Waste Water Treatment Plant (WWTP). Recent studies suggested that part of the biological Hg methylation in freshwater sediment may be attributed to the activity of iron-reducing bacteria (IRB). The purpose of this study is focussed on methylation process in sediment cores of Vidy Bay.

A detailed study was carried out on sediment cores collected in May 2007 and in August 2008. We examined sediments from two sites situated at 190 m (site CP) and 530 m (site FP), respectively, from the current location of the outlet pipe discharging treated sewage from the city of Lausanne.

Total Hg concentrations were similar for both sites. MMHg profiles in cores showed pronounced maxima at 1-2 cm depth with the highest level reaching 5–6% of the THg at site CP. Methylation and demethylation yileds were determined by incubating sediment slices (2 cm thick) with 199IHg and 201MMHg for 24h. The methylation yields were in the same order of magnitude and followed the same trend than MMHg profiles, with maxima in the top 2 cm for both sites. However, the absolute values were in the same order of magnitude but higher for site CP.

For both sampling sites, most probable number counts of IRB in the uppermost 2 cm of sediment were two orders of magnitude higher than for sulphate-reducing bacteria (SRB). In this layer MMHg concentrations and methylation yields reached their maxima. In addition, Fe2+ concentrations in overlying water of the CP core were close to the detection limit but increased strongly at 1 cm depth.

Results revealed that wastewater discharge from the WWTP changed geochemical properties of sediments enhancing Hg methylation. This research represents one of the first studies of in situ confirmation of an overlap between MMHg production zone, reduction of Fe(III)-oxides and the presence of IRB, suggesting IRB as important Hg methylators in this sediment.

MG7-P20 — 11:00-12:00 and 17:30-18:30
Authors: BALOGH, Steven J.1, BLUMENTRITT, Dylan J.2, ENGSTROM, Daniel R.3, NOLLET, Yabing H.1, TRIPLETT, Laura D.4
(1) Metropolitan Council Environmental Services, St. Paul, MN 55106, steve.balogh@metc.state.mn.us; (2) Department of Geology and Geophysics, University of Minnesota, Minneapolis, MN 55455; (3) St. Croix Watershed Research Station, Science Museum of Minnesota, Marine on St. Croix, MN 55047; (4) Department of Geology, Gustavus Adolphus College, St. Peter, MN 56082.

Lake Pepin is a large natural floodplain lake on the Upper Mississippi River located about 80 km downstream of the Minneapolis/St. Paul (Twin Cities) metropolitan area. The lake is an effective trap for sediments carried by the river, and sediment cores there record the quantitative history of Hg and sediment loads in the river dating back to local Euro-American settlement (c. 1830). 210Pb-dated sediment cores collected in 1995-1996 showed large inputs of Hg to the river during the middle part of the 20th century as development of the metropolitan area upstream accelerated, but Hg loads have decreased since about 1970 due to reduced inputs from wastewater and other sources. Recently (2006-2008), new sediment cores were collected at the same locations as 10 of the previous cores using precise GPS coordinates. Analysis of these dated cores provides a measure of changing Hg and sediment loads in the river since 1996. Post-1996 chronologies were calculated based on a precise alignment of prominent magnetic susceptibility peaks evident in both the original and repeat cores. This method accounted for varying degrees of sediment compaction among the cores resulting from differences in sediment accumulation rates at each site. Our results indicate a 23% reduction in the whole-basin Hg accumulation rate over the past 10-12 years, from 110 kg yr-1 to 85 kg yr-1. Sediment accumulation rates declined by 12% (876 ktons yr-1 to 770 ktons yr-1), and the whole-basin mean Hg concentration decreased from 125 ng g-1 to 110 ng g-1. Reduced loads of Hg in the Upper Mississippi River at Lake Pepin over the past decade probably reflect further improvements in wastewater treatment and reduced direct and indirect (i.e. atmospheric deposition) inputs, but may also indicate declining contributions from contaminated legacy sediments eroded from areas within and downstream of the Twin Cities. The recent Hg accumulation rate in Lake Pepin remains well above the pre-settlement level there (3 kg yr-1) and also far above the current rate observed in Lake St. Croix (7 kg yr-1), a similar natural impoundment on the nearby St. Croix River. While Hg accumulating in the sediments of Lake St. Croix is derived primarily from atmospheric inputs to the relatively undeveloped watershed, the much higher Hg accumulation rate in Lake Pepin reflects continuing large erosional inputs from row-crop agricultural areas in the Minnesota River sub-basin and lingering inputs from the dense urban area upstream.

MG7-P21 — 11:00-12:00 and 17:30-18:30
Authors: HE, Feng1, ZHENG, Wang1, LIANG, Liyuan1, GU, Baohua1
(1)Oak Ridge National Laboratory, hef2@ornl.gov

Photochemical reduction of mercury (as Hg(II) species) is one of the most important pathways of elemental Hg(0) production in natural and contaminated open waters and plays a key role in Hg cycling in the environment. This redox transformation is affected by aqueous ionic species, among which the aromatic moieties of naturally dissolved organic matter (DOM) are thought to be the most important because of their light-absorbing characteristics. However, the exact mechanism by which DOM mediates the photochemical reduction of Hg(II) is not yet clear owing to our incomplete understanding of the structural arrangement and stoichiometry of reactive functional groups in DOM. In this study, DOM analogs including salicylic acid, 4-hydrobenzoic acid, anthranilic acid, 4-aminobenzoic acid, and phthalic acid were used to systematically investigate the role of functional groups and their steric arrangements in photochemical reduction of Hg(II). We show that the photochemical reduction rates of Hg(II) are influenced not only by the neighboring functional groups but also their positioning on the aromatic benzene ring structure. The reduction rate was found to increase in the order of anthranilic acid > salicylic acid > phthalic acid by varying neighboring functional groups from amine, hydroxyl, to carboxyl on benzoic acid. The substitution position of the functional groups affected reduction rates in the order: anthranilic acid > 4-aminobenzoic acid and salicylic acid > 4-hydroxybenzoic acid. Analyses of ultraviolet (UV) light absorbance of these DOM analogs indicate that the Hg(II) reduction rate is positively correlated to their UV absorption. The reduction rate was also found to depend on the concentration of the organic compounds, suggesting that the secondary reaction mechanism is mainly responsible for the photochemical reduction of Hg(II) in these systems. These observations shed additional light to the understanding of the role of DOM in Hg photoredox transformation and geochemical cycling in the environment.

MG7-P22 — 11:00-12:00 and 17:30-18:30
Authors: WANG, Yanping1, WIATROWSKI, Heather2, YEE, Nathan1, BARKAY, Tamar1, LIN, Chu-Ching3
(1) Rutgers University, yawang@rci.rutgers.edu; (2) Clark University; (3) National Central University.

The contamination of groundwater with mercury (Hg) is an increasing problem worldwide. However, little is known about the interactions between Hg and microorganisms in subsurface environments, particularly under nitrate-reducing conditions as nitrate often appears as a co-pollutant in groundwater. Because the behavior of Hg in the environment strongly depends on its chemical forms, we conducted laboratory studies to investigate the mechanisms that may underpin Hg redox transformation mediated by dissimilatory nitrate-reducing enrichment cultures derived from the Oak Ridge Integrated Field Research Challenge site, where nitrate is a major contaminant.

Data from Hg toxicity experiments showed an inverse relationship between Hg concentration and onsets of nitrate reduction in enrichment cultures; higher Hg concentrations irreversibly inhibited denitrification. A decline in the number of species in the various enrichment cultures, as indicated by the number of 16S rRNA gene TRFLP peaks, occurred with increasing Hg concentrations. This loss in community complexity resulted in only a single TRFLP peak being detected for a treatment with 312 nM Hg. This single peak corresponded to a Bradyrhizobium sp., which reduced Hg(II) to Hg(0) trapped in a solution consisting of strong acids (i.e., nitric acid and sulfuric acid) and oxidants (i.e., permanganate and persulfate). The strain also contained a merA gene with 97% identity to merA of the Proteobacteria and Firmicutes.

Results from time-course experiments on Hg(II) reduction showed that while formation of gaseous Hg(0) was observed after overnight incubation, nitrate reduction and cell growth did not follow Hg(II) reduction; rather, a decrease of Hg(0) in trapping solution occurred in parallel with an increase of Hg in the cultures upon the onset of cell growth. Furthermore, when incubated with the spent medium obtained from different growth phases, Hg(0) was oxidized. Taken together, these results suggest that in addition to the impact of Hg on denitrifying community structure, Hg-denitrifier interactions can result in Hg redox cycling and thus may affect Hg mobility in anoxic environments.

MG7-P23 — 11:00-12:00 and 17:30-18:30
(1) Universidade Federal do rio de Janeiro, torquato@biof.ufrj.br; (2) Universidade Federal de Rondonia;

The creation of reservoirs in the Amazon contributes to the mobilization of mercury from> the soil followed by its release into the water column with subsequent conversion of inorganic mercury to its organic form methylmercury. In this study we evaluated the distribution of methylmercury in the water column for the tropical reservoir in the Jamari river in the western Amazon. Samuel Reservoir flooding began in 1989. The reservoir is in the Jamari river localized to 52km from the city of Porto Velho. Samples were collected in five stations from the reservoir during dry season in July 2010. The M01 station is located 1000 meters upstream the dam. The M04 station is located in the center of the reservoir. The M05 station and São Marcos are located in the tributaries of the Jamari river. The M08 station is a section located 80km upstream the dam being the passage for the main channel of the lake. Water column samples (n=23) were collected using a Van Dorn and acid-cleaned polypropylene tubing. Vinyl gloves were used for handling operations. Samples were collected in acid-clean borosilicate bottles. Aliquots of unfiltered water samples for MMHg determinations were acidified to 0.5% (v/v) with HCl. The aqueous solutions were analyzed for MMHg by gas chromatography after ethylation, separation and detection by atomic fluorescence spectrometric CG-AFS (MERX™ – Brooks Rand). Limnological parameters such temperature, pH, dissolved oxygen, conductivity, deepth and redox were recorded in situ with a Horiba U20 multiparameter probe. The water column shows anoxic layer starting at 10 meters of the depth. MeHgT concentration in the water column showed average values of 0.066 ng.l-1. The lowest concentration of MeHgT was found in surface of the water. The M05 station showed metylmercury concentation in the anoxic layer 18 times higher (0.312 ng.l-1) than concentrations found in the superficial layer. Our preliminary results suggest that methylmercury may be produced in anoxic layer of the Samuel dam.

MG7-P24 — 11:00-12:00 and 17:30-18:30
Authors: YAN, Haiyu1, FENG, Xinbin1, LIU, Bian2, WANG, Cuiping3, LI, Qiuhua4, DOMINIK, Janusz2
(1) Chinese Academy Of Sciences, yanhaiyu@vip.skleg.cn; (2) University of Geneva; (3) Guizhou University; (4) Guizhou normal University;

The combination of control the pollution sources and dredging the sediments are common ways to clean up the contaminants such as mercury (Hg) in an aquatic system. However, dredging can often cause a sharp pulse of increased transfer of contaminants from the sediment to the water column, which requires a relatively long time for the system to reach the new equilibrium with desired results. Here, based on our six field studies during 2000-2010 in a Hg-contaminated reservoir (Baihua Reservoir), we suggest that control the sources of Hg pollution without dredging is a better option for the recovery of the reservoir.

Baihua Reservoir is historically contaminated with Hg discharged from Guizhou Organic Chemical Plant (GOCP) from 1970 to 1997. During the past 20 years, actions that have been taken to reduce the Hg pollution include updating the technology of GOCP and controlling surrounding pollution sources without any treatment of the reservoir sediment. Over the ten years of our study period, we found that the concentrations of mercury species in the water column and the sediment showed a sharp decreasing trend with the highest total Hg concentration in the water column changing from 153ng/L in 2000 to 5ng/L in 2010. In addition, we found low methylation rate and low fish mercury level (< 0.1mg/kg wet weight). Our data suggests that in this reservoir mercury in the water column tends to settle into the sediment and the transfer of mercury from the sediment into the water column is limited. Therefore, the most effective remediation method for Baihua Reservoir appears to be the control of the discharge sources and dredging of the sediment is not recommended.

MG7-P25 — 11:00-12:00 and 17:30-18:30
(1) CIEMAT, rocio.millan@ciemat.es

Valdeazogues River flows through the Almadén mercury mining district (Spain) crossing one of the areas with the highest mercury concentration in the world. The experimental work was performed in a section along the Valdeazogues River banks. Soil and plant samples (Nerium oleander L.) were taken in the springtime of three consecutives years (2007-2009).

Mercury in the different soil fractions was determined by using a six-step sequential extraction procedure developed at Ciemat (Spain). Furthermore, the relationship between the percentage of organic matter in soil and the percentage of mercury associated with the exchangeable and oxidizable fractions was established. Once the available mercury was determined in soil, the absorption and distribution of mercury in oleander plants was studied.

The results show that total mercury concentrations in soil range from 116.7 ± 24.3 up to 350.9 ± 68.6 mg kg -1. However, the soluble and exchangeable fractions are about 0.15% of the total mercury measured in all samples. Soluble mercury is less than 0.037 mg kg-1, and therefore, the leaching process and transport of mercury to river water is very low. Comparing the different soil fractions, Hg was found mainly in the most resistant soil fractions (crystalline Fe-Mn oxyhydroxides, organic matter absorbed and the final residue).

Mercury distribution in N. oleander plants is not homogeneous throughout the aerial part, but is not considered a high content in any case. In general, the concentration is significantly higher in the leaves followed by stems and fruits (leaves: 0.282 ± 0.014 – 1.022 ± 0.110 mg kg-1; stems: 0.087 ± 0.011 – 0.354 ± 0.046 mg kg-1; fruits: 0.030 ± 0.003 – 0.077 ± 0.009 mg kg-1). This evergreen shrub has a high biomass and with its high toxicity it is non-edible to animals, and therefore this specie is appropriate in a phytotechnology process to be applied within flooded river areas.

MG7-P26 — 11:00-12:00 and 17:30-18:30
(1) CIEMAT, mj.sierra@ciemat.es

Once the decision is taken on the close down of mining exploitation in an area, the selection and implementation of the most appropriate socio-economic alternatives have to be planned. These options also include alternative land uses. It is the case of Almadén mercury (Hg) mining district, where research work about evaluating the possible cultivation of lavender (Lavandula stoechas L.) as a safe alternative land use, has been performed. After preliminary studies carried out under controlled conditions, field studies have been performed in this work. Lavender plants were sampled on a wide range of soils with different land use in Almadén district where these plants grow spontaneously and have significant presence. These sample areas are situated both in old mining activities zones and surroundings (total mercury in soil ranges from 6.3±0.8 mg kg-1 to 55±5 mg kg-1) and in zones far from these pollution sources (total mercury in soil ranges from 1.6±0.2 mg kg-1 to 7.8±2.1 mg kg-1). In each sampled point of lavender, soil that was right next to plant root (rhizosphere) was sampled. Furthermore, surrounding soil (0 - 20 cm in depth) that remained when the lavender sample was taken (bulk soil) was sampled.

The aim of this work is to evaluate different edaphic parameters (pH, Cation exchange capacity, electrical conductivity, organic matter, soluble ions) in rhizosphere and in bulk soil and their influence on mercury behaviour in soil-lavender system. According to a discriminant analysis, among the different edaphic parameters, the most important in the differentiation is electrical conductivity.

Furthermore, root uptake of mercury by lavender and its distribution through the plant have been studied in this work. Lavender seems to uptake mercury, corresponding to the available fraction, of the rhizosphere. The more available mercury in rhizosphere soil the more mercury is translocated to aerial part and less mercury is retained by root. In this context, the mercury concentration in root was higher than in aerial part where not significant differences among mercury concentrations in the different fractions were found.

MG7-P27 — 11:00-12:00 and 17:30-18:30
Authors: WINDHAM-MYERS, Lisamarie1, MARVIN-DIPASQUALE, Mark1, STRICKER, Craig1, FLECK, Jacob1, ALPERS, Charles N.1, ACKERMAN, Josh T.1
(1) U.S. Geological Survey, lwindham@usgs.gov

Rice agriculture is a significant wetland type globally and in California, where rice acreage is at least equivalent to naturally vegetated wetland acreage. A variety of factors appear to contribute to the relatively high rates of methylmercury (MeHg) production in California rice fields (>10 pg gdry sed-1 d-1), including: 1) seasonally pulsed high productivity and subsequent decomposition, 2) biannual flooding and dessication associated with planting and harvesting cycles, and 3) elevated concentrations of legacy mercury (Hg) from historic gold-mining. In order to protect ecological and human health, there is an identified need in Central Valley wetlands to limit MeHg export to the Sacramento-San Joaquin watershed, the major freshwater source to the San Francisco Estuary, and control MeHg bioaccumulation in situ. The roles of vegetation and sediment biogoechemistry, plant physiology, and community structure were examined in 3 types of agricultural wetlands (white rice, wildrice, and fallow fields) and adjacent managed natural wetlands (cattail- and bulrush-dominated) with respect to their potential influences on MeHg production, export and bioaccumulation. A vegetated:devegetated paired plot experiment demonstrated that the presence of live plants enhanced rates of mercury methylation by an average of 380%, primarily through enhanced pools of labile carbon in the rhizosphere (also about 4-fold greater in vegetated plots), but also potentially through transpiration-driven re-oxidation of iron pools. Canopy shading at peak biomass (leaf area index range= 0.8-2.6) significantly reduced aquatic exposure to photosynthetically active radiation by 40-80%, which may have limited rates of photodemethylation, thus allowing greater MeHg export. To account for storage terms in the MeHg export calculations, biomass and litter MeHg concentrations and pools were measured and modeled over a full annual crop cycle. Whereas total mercury concentrations declined in leaves over the growing season, MeHg concentrations and pools in rice and non-rice vegetation generally increased as the growing season progressed. Rice and wildrice seeds reached mean MeHg concentrations of 4.2 and 6.2 ng gdw-1, respectively, with seed husks reaching concentrations as high as 9 ng gdw-1. Seed MeHg concentrations were correlated with root MeHg concentrations (r=0.90), suggesting a direct relationship between sediment MeHg production and tissue accumulation. Plant biomass was an insignificant sink of MeHg with respect to seasonal export from fields, but the apparent redistribution of MeHg into seeds may be a significant problem for migrating and resident waterfowl with a rice seed-based diet. In sum, the relative role of vegetation varied among field types but generally enhanced MeHg production, export, and potential bioaccumulation.

MG7-P28 — 11:00-12:00 and 17:30-18:30
Authors: MONPERRUS, Mathilde1, AMOUROUX, David1, SHARIF, Abubaker1, ALANOCA, Lucia2, POINT, David2
(1) LCABIE-IPREM UMR5254 UPPA-CNRS, mathilde.monperrus@univ-pau.fr; (2) GET UMR5563 IRD-UPS-CNRS;

Bolivian lacustrine ecosystems are exposed since several centuries to intense mining activities and act as an important sink for sediments, mining effluents and associated trace contaminants. The Uru Uru Lake (214 km2 at 3693m of elevation), part of the Titicaca-Desaguadero-Poopó-Salar lagoonal system, is characterized by contrasted seasonal hydrologic regimes, shallow waters, strong physico-chemical gradients, intense anthropogenic pressure and high primary productivity. Only scarce data on Hg distribution are up to now available for such ecosystems. High altitude lakes could act as a highly dynamic biogeochemical reactor for Hg transformation and transfer mechanisms. Moreover there is a need of knowledge concerning the Hg cycle in that particular environment, due to the risk of exposition of the local populations through the Hg bioaccumulation along the food web, as well as for the preservation of endemic aquatic species.

The aim of this work was thus to investigate the biogeochemical reactivity of Hg by coupling mercury species measurements in the sediment, pore waters and water column continuum with in situ incubation experiments. Gaseous (Hg°, DMeHg), dissolved and particulate mercury species (Hg(II), MeHg) have been measured at 10 stations along a transect from the north to the south. Diurnal reactivity has been evaluated by determining the different mercury species concentrations every 2 hours during a 24 hours cycle, from which gaseous Hg fluxes at the water-atmosphere interface were derived. In situ incubation experiments using isotopically enriched species (199Hg(II) and Me201Hg) have been achieved for water samples, sediments and periphyton in order to quantify mercury species transformations (i.e. reduction, demethylation and methylation) and to evaluate their contribution to the net MeHg formation in the lacustrine system.

MG7-P29 — 11:00-12:00 and 17:30-18:30
Authors: THOMPSON, Charles1, DEWILD, John2, KRABBENHOFT, David 2, KNAUER, Doug 3, PELKOLA, George 3
(1)US Geological Survey, cdthomps@usgs.gov; (2) U.S. Geological Survey, Reston, VA ; (3) Michigan Department of Natural Resources and Environment ;

Dating back to dosing studies in the 1980s, it has long been known that selenium (Se) has an antagonistic effect on methylmercury (MeHg) bioaccumulation in aquatic organisms, and more recently there is evidence to suggest that the co-occurrence of Se and mercury (Hg) within biological systems may have an amelioratory effect on the toxicity of both elements. Therefore, while our scientific understanding of Hg-Se interactions has increased substantially over the past twenty years, there remains a significant understanding gap on the effects of Se on biogeochemical cycling and bioaccumulation of Hg and MeHg in aquatic food webs. With that in mind, the U.S. Geological Survey (USGS) initiated a study on a Se-enriched lake (Goose Lake) in the Upper Peninsula of Michigan to further our understanding of the antagonistic effects of Se on Hg and MeHg bioaccumulation in aquatic food webs.

For this project we collected benthic invertebrates, algae and prey fish (Walleye and Northern Pike) from two nearby lakes (Goose Lake and Deer Lake) in the Upper Peninsula of Michigan. Sediment Hg and MeHg levels in the two lakes are similar, although Se levels in Goose Lake are about 12 times greater than Deer Lake (30 vs. 2.5 µg/g, dry wt). Prey fish mercury levels in Deer Lake are correspondingly about 30 times greater (4100 vs. 130 µg/g), supporting the originally proposed antagonistic effect of Se on Hg bioaccumulation. As noted above, MeHg levels in sediment are similar in the two lakes, suggesting that a sediment-water transfer of MeHg to pelagic fish may be occurring. Results from the benthic invertebrates show that Se enrichment also has an inhibitory effect on MeHg bioaccumulation in these organisms. Results from sites with background levels of Se compared to sites near Goose Lake with enriched Se in sediment show a 2.0-3.4 µg/g difference in MeHg content – notably lower than the difference observed for pelagic fish. In addition, benthic invertebrate MeHg/Hg ratios at Se contaminated sites are significantly lower than background sites, likely resulting from the inhibitory effect of Se on MeHg production. Overall, data from this project suggests a negative relationship between MeHg, total Hg, and the MeHg/Hg ratio and selenium abundance near Goose Lake, Michigan. This relationship holds for both benthic and pelagic portions of the aquatic food web, and it appears a more pronounced difference is exhibited for pelagic organisms.

MG7-P30 — 11:00-12:00 and 17:30-18:30
Authors: ROTHENBERG, Sarah1, FENG, Xinbin1
(1) Institute of Geochemistry, Guiyang, China, rothenberg.sarah@gmail.com

The effects of root exudates on mercury (Hg) methylation were investigated in a flooded rice paddy in Guizhou province, China, where three-fourths of the paddy was planted with rice while one-fourth remained fallow. In both regions, time series pore water pH data were collected every 10 minutes for 63 days. Depth profiles for Hg species were established from sediment cores, which were extracted every 2 weeks and sectioned at 2-cm intervals (n=4 time points). On two occasions, pore water (0-2 cm depth) was collected every 3 hours for 24 hours to investigate diurnal Hg cycling. In the planted region, pH was lower (Planted: pH 6.05 ± 0.153; Fallow pH 6.95 ± 0.286, n=8967 for both sites), and MeHg levels were higher (Planted: 6.3 ± 4.2 ng/L, n=36; Fallow: 3.7 ± 1.9, n=33). Surface water MeHg levels were approximately equal at both sites (<1 ng/L), suggesting differences between regions were possibly due to root exudates. In both regions, pore water total sulfide was <10 uM, while in the planted region pore water MeHg and sulfate levels were positively correlated (r2 = 0.16, p<0.05, Sulfate: 14-490 uM, n=26; when one observation removed, r2=0.33, p<0.01), but were uncorrelated in the unplanted region (r2 = 0.08, p>0.15, Sulfate: 12-2100 uM, n=25). Rice roots leak oxygen (O2) into the rhizosphere, a byproduct of photosynthesis. Higher O2 in the rhizosphere results in replenishment of electron acceptors (e.g., sulfate), which increases substrate for sulfate-reducing bacteria and reduces the buildup of sulfide, leading to higher bioavailability of Hg-sulfidic species to sulfate-reducing bacteria. In the planted region, enhanced cycling of sulfate/sulfide likely promoted MeHg production. Rhizosphere pH is typically 1-2 log units lower compared to the bulk soil (i.e., unplanted region), partially due to O2 leakage from the roots. Lower pH is associated with higher MeHg concentrations, which was consistent with results. However, diurnal cycling of pore water pH was not observed in either region, while strong diurnal cycling was observed for MeHg, with highest MeHg levels observed at midnight on two occasions, suggesting O2 was an important control on the cycling of Hg, more so than pH.

MG7-P31 — 11:00-12:00 and 17:30-18:30
Authors: HOGGARTH, Cameron G J1, HALL, Britt D.1
(1)University of Regina, hoggartc@uregina.ca

>In aquatic ecosystems the majority of MeHg is likely produced through methylation of Hg(II) by sulfate-reducing and iron reducing bacteria. Wetland sediments may be significant sites of MeHg production due to the combination of anoxic conditions and the availability of organic matter for microbial activity. Sediment total mercury (THg) concentrations may provide an indication of the amount of mercury available for methylation, which provides the base of the aquatic food chain with neurotoxic MeHg. In 2008 and 2010, sediment cores were collected from prairie wetlands located in the St. Denis National Wildlife Area (SDNWA), ~40 km east of Saskatoon in central Saskatchewan. The SDNWA contains more than 200 wetlands and has been a site of prairie wetland research, including waterfowl and hydrological research, since 1968. Prairie Pothole Region wetlands are valuable for groundwater recharge and support 50-80% of North American waterfowl. However, little information is available on THg concentrations in prairie wetland sediments. The top two centimetres of 22 sediment cores collected in fall 2010 from 12 wetlands in the SDNWA were analyzed for THg concentrations and organic content. Preliminary sediment THg concentrations (2010 only) ranged between 15.94 and 88.07 ng/g, with a mean of 50.40±17.30ng/g. Wetland sediment organic carbon content as measured by loss on ignition, ranged between 3.51 and 21.85%, with a mean of 7.87±3.74%. Sediment methylmercury analysis and THg analysis of 2008 sediments are in progress.

Sediment organic matter may influence the amount of THg available for methylation, and potentially the amount of MeHg in aquatic organisms. However, no significant relationship between sediment organic content and THg concentration was observed (R2 = 0.12, df = 21, p = 0.113). Water content and THg concentrations had a significant linear relationship R2 = 0.51, df = 21, p < 0.001), which may be a result of high THg concentrations in sediment pore waters relative to dry sediments. Future research will include measurement of wetland sediment methylation rates through the injection of an inorganic mercury stable isotope into wetland sediment cores, followed by incubation of the sediment cores and measurement of sediment MeHg stable isotopes in the incubated cores. Determining which factors influence THg concentrations in prairie wetland sediments may assist in identifying factors contributing to the availability of THg for biotic methylation of THg into the bioaccumulating neurotoxin, MeHg.

MG7-P32 — 11:00-12:00 and 17:30-18:30
Author: TATE, Michael T.1
(1)US Geological Survey, mttate@usgs.gov

During the summer of 2007, USEPA’s National Lakes Assessment (NLA) project conducted a national-scale survey of the “health” of lakes and reservoirs across the continental United States. A total of 1,070 lakes, representing six size classes and distributed relatively uniformly across the lower 48 states, were included in the survey. Although the NLA emphasized measures of lake health (nutrient levels and water chemistry, biological parameters, habitat surveys, and land use data), mercury (Hg) and methylmercury (MeHg) content of surficial (0–2cm) sediment was also assessed through the collection of intact cores from the middle of each lake. The scope of this paper is to assess the spatial distribution of total Hg and MeHg in lake sediments and their controlling mechanisms across the U.S.

Sediment total Hg concentrations show a distinct sub-regional distribution, in which there are clusters of lakes with elevated concentrations – although, scattered individual lakes with elevated total Hg concentrations also exist. The primary controlling co-factor of total Hg concentration was determined to be the trophic status (defined by chlorophyll a concentration) and, to a lesser, degree proximity to known point (mining) sources (i.e., Lahonton Reservoir, NV and Coure d’Alene, ID). There was no obvious relation to differences in atmospheric Hg deposition patterns that are published annually by the Mercury Deposition Network (MDN). Spatial analysis shows lakes with the highest total Hg concentration are predominately oligotrophic and are located in New England, northern Florida, northern Wisconsin and Minnesota, and the Olympic Peninsula in Washington. Overall, lakes in the Mississippi River basin exhibit generally lower total Hg concentration, which is again related to the higher trophic status of lakes in this region.

The spatial distribution of MeHg concentrations is very complex, exhibiting very little regional structure or obvious spatial trends. Also, MeHg concentration shows little correspondence to the total Hg concentration. A linear correlation analysis (r2=0.27) reveals total Hg is a poor driver of MeHg in lake sediments, suggesting total Hg in sediment is not an important driver. Lake-quality factors do influence the MeHg/total Hg ratio, which has been shown to be a good indicator of net methylation activity. Specifically, the greatest MeHg/total Hg ratios are associated with the more highly eutrophic conditions. MeHg/total Hg ratios in sediment show peaks at 1 mg/L and 10 mg/L for total Nitrogen and Sulfate in water, respectively.

MG7-P33 — 11:00-12:00 and 17:30-18:30
Authors: ULANOWSKI, Tom1, BRANFIREUN, Brian1
(1) University of Western Ontario, tulanows@uwo.ca

The second largest peatland complex in the world is found in the Hudson Bay Lowlands (HBL) of Northern Canada, with a total area of over 320 000 sq. km. Current and future climate and land use impacts on the hydrology and mercury dynamics of these extensive peatlands necessitates the collection of baseline biogeochemical data for environmental monitoring and policy-making. The sheer size, heterogeneity and complexity of these large northern peatland ecosystems demands rigorous statistical sampling in order to obtain a representative sample of the distribution of mercury and ancillary pore water data, particularly when such information is limited.

This study assessed the small scale spatial (1 m) and temporal (21 day) variability of surface peat and pore water chemistry in a Sphagnum-dominated ombrotrophic raised bog and Carex-dominated intermediate fen in the James Bay Lowlands. Systematic sampling of a 7x7 m grid containing forty-nine 1 m2 units for in situ pH and water temperature, total mercury (THg), dissolved organic carbon (DOC), and major anions and cations at the water table revealed distinct biogeochemical environments between wetland types and variability in both space and time in these environmental parameters. In the intermediate fen, microtopography was found to have an influence on in situ temperature, pH and DOC concentrations that generally increased from deep hollow to lawn. Changes to solute chemistry, particularly base cations (Na+, Ca2+ and Mg2+) over a three week period were statistically significant (ANOVA) in the intermediate fen, likely a result of a dynamic connection to mineral-rich groundwater. In the raised bog, solute chemistry was less variable, but DOC concentrations in July exhibited unexpectedly large variability within the grid, ranging from 12.31 to 30.44 mg L-1. In both the bog and fen, sulphate concentrations showed spatial variability, and a significant decrease over the early summer period. Coupled to THg data in the solid peats, the implications of DOC and sulphate dynamics will be discussed. Variability at this within-ecosystem scale will be used to estimate sampling size requirements to generate accurate data within a range of margin of error. The variability in peatland pore water chemistry over such small spatial and temporal scales has implications for the characterization of peatland type, as well as studies attempting to establish baseline biogeochemical conditions in these important northern ecosystems.

MG7-P34 — 11:00-12:00 and 17:30-18:30
Author: HEIM, Wesley1
(1)Moss Landing Marine Laboratories, wheim@mlml.calstate.edu

There is widespread mercury (Hg) contamination in the Sacramento-San Joaquin Delta (Bay-Delta). The Hg species of greatest concern to human and wildlife health is monomethylmercury (MeHg). Wetland habitat is well suited for the conversion of inorganic Hg into the more toxic MeHg. As current and future restorations increase total acreage of wetland habitat in the Bay-Delta MeHg contamination in connected adjacent waterways may worsen. The goal of this study was to develop best management practices (BMPs) to limit MeHg loads discharged off wetlands in the Yolo Wildlife Area (YWA) located within the Bay-Delta complex. Two experiments were conducted: 1) measured MeHg concentrations in seven ponds during seasonal flooding, and 2) measured MeHg as seasonal wetlands were drained into permanent ponds. Results show concentrations of MeHg in most of the seasonal ponds studied have an initial peak occurring in fall followed by a decrease and a leveling off after January. Concentrations are high (6-15 ng/L) relative to supply water indicating seasonal wetlands produce MeHg. Permanent ponds had lower MeHg concentrations than supply water from seasonal wetlands. Reduction in MeHg concentration occurs once tail water from a seasonal wetland enters a permanent pond. Permanent pond S5 received water from a seasonal pond high in MeHg (average 11.3 ng/L) and concentration was reduced by an order of magnitude (average 1.18 ng/L). Permanent ponds acted as treatment ponds and removed an average of 81% of the MeHg that entered the ponds. Further development of a BMP using permanent ponds to treat seasonal pond tail water is warranted from data presented in this study. Several factors in permanent ponds may be important in removal of MeHg and should be investigated to develop the BMP to implementation stage. These factors include pond design, biological communities, photodemethylation, hydraulic residence time, and particle settling. Development of BMPs to reduce MeHg loads from seasonal wetlands is relevant to wetland restorations as they limit an increase in Hg contamination to surrounding waterways.

MG7-P35 — 11:00-12:00 and 17:30-18:30
Authors: KRONBERG, Rose-Marie 1, TJERNGREN, Ida1, SKYLLBERG, Ulf1, BJÖRN, Erik2
(1) Swedish University of Agricultural Sciences, rose-marie.kronberg@slu.se; (2) Umeå University.

In forested landscapes wetlands are generally recognized as major sources for methyl mercury (MeHg), whereas uplands and certain types of lakes are major sinks. This picture is, however, too simplistic. To be able to counteract negative effects of e.g. forestry activities like clear-cutting on MeHg production and export, we need to understand more in detail factors in control of methylation and demethylation processes and how these vary in the landscape. This paper builds on the results from four years of input/output studies of Hg and MeHg in an Alnus glutinosa swamp in southern Sweden. The Alnus swamp was shown to be a substantial sink of MeHg. Studies of the soil showed that absolute and relative concentrations of MeHg, as well as the total mass per ha, decreased along a gradient running from the inlet to the outlet of the swamp. Potential demethylation rates increased along the gradient, whereas potential methylation rates showed no clear trend. There was a clear seasonality in methylation and demethylation rates, the former with a climax during summer and the latter with its maximum during the colder months. This seasonal pattern was also reflected by %MeHg in soil. In order to test the generalisability of our results, eight additional Alnus glutinosa swamps were studied by means of one-day input-output budgets during early summer 2010. Five sites showed reduction in MeHg, two sites showed no change and one site showed an increase in MeHg. We conclude that Alnus glutinosa swamps seem to be sinks for MeHg, at least when located downstream sites acting as major sources of MeHg production. The results favor policies to restore formerly drained Alnus glutinosa swamps (which often are associated with high biodiversity), in order to mitigate negative effects of MeHg in the landscape. A next step would be to investigate the effect of Alnus glutinosa swamps as buffer strips along streams after clear-cut. An issue high up on the agenda for forest authorities in Sweden.

MG7-P36 — 11:00-12:00 and 17:30-18:30
Authors: DEMERS, Jason D.1, YAVITT, Joseph B.2, DRISCOLL, Charles T.3
(1) University of Michigan, jdemers@umich.edu; (2) Cornell University; (3) Syracuse University.

Elevated atmospheric deposition of mercury since the mid-1800s industrial revolution has led to a large legacy of mercury accumulation in forest and wetland soils. However, the mechanisms influencing the retention and biogeochemical cycling of mercury within the soils of forests and different wetland types remains unclear. In this study, mercury pool size and stoichiometry with C, N, and S are quantified across upland-wetland transects and among wetlands of different hydrogeologic settings in the Adirondack region of New York State. The pool size of mercury varied across the upland-wetland interface, among wetland types, across individual wetland transects, and along depth profiles in soils of forests and wetlands of the Sunday Lake watershed. Total mercury pool size was greater in upland soils than in wetland soils (to 50 cm depth, p<0.01). Shallow peat riparian wetlands had greater mercury pool size (to 50 cm depth) than deep peat riparian or headwater wetlands (p<0.099, p<0.046). There was greater accumulation of mercury at the upland-wetland interface in deep peat riparian and headwater wetlands as compared to the overall wetland transect (p=0.008). In mineral horizons of the uplands and shallow peat riparian wetlands, mercury was strongly correlated with carbon (p=0.002, r2=0.73), and nitrogen (p<0.001, r2=0.82), but not sulfur. In deep peat riparian and headwater wetlands, there was no relationship between mercury and carbon or nitrogen; however, there was a strong correlation between mercury and sulfur in the peat of headwater wetlands (p<0.0001, r2=0.60). Hydrogeologic setting influences decomposition processes, biogeochemical cycling of mercury, and hydrologic transport that, in turn, govern the size, distribution, and stoichiometry of mercury pools across the upland-wetland interface, along peat depth profiles, and among different wetland types. The absence of strong correlation between dissolved mercury (Hgd) and dissolved organic carbon (DOC) in the peat porewater and outlet stream water of the headwater wetlands suggests that export of mercury from these headwater wetlands is limited by the supply of mercury rather than by the availability of Hg-DOC binding sites. In contrast, it appears that the quantity of mercury transported in riparian streams is more limited by DOC binding and mobilization, and reflects the strong correlation between mercury and carbon observed in the upland mineral soil horizons and shallow peat riparian wetlands.

MG7-P37 — 11:00-12:00 and 17:30-18:30
Authors: MIRANDA, Marcio R.1, GUIMARAES, Jean Remy D.2, COELHO-SOUZA, Sergio Augusto1, BASTOS, Wanderley R.3, ALMEIDA, Ronaldo3, CARVALHO, DP de3, HOLANDA, IBB3
(1) Inpetam, topo@biof.ufrj.br ; (2) IBCCF/Universidade Federal do Rio de Janeiro; (3) UNIR;

The Madeira River Basin accounts for 20% of the drainage area of the Amazon Basin. In the last 30 years the Rio Madeira basin has been impacted by anthropogenic activities, including alluvial gold extraction, deforestation for agriculture projects, and more recently damming for a hydroelectric power plant. Damming rivers for hydropower generation causes different environmental and health impacts, among which the increase in fish Hg levels. Reservoirs can act as total Hg traps, due to sedimentation of suspended sediments, and as MeHg sources to the reservoir itself and waters downstream. The aim of this study was to investigate mercury methylation sites at the direct influence area of the Santo Antônio hydroelectric reservoir before its construction in the Madeira River (Rondônia, Brazil). The following factors were evaluated: i) in situ methylmercury production in the rhizosphere and soil from Terra Firme (upland) Forest; ii) methylmercury production in the rhizosphere and soil stimulated by seasonal flooding pulse; iii) methylmercury production In bottom sediment. Samples were incubated with 203HgCl2 to evaluate methylmercury production. Results from Terra Firme Forest confirmed the in situ methylmercury formation in rhizosphere (0.38±0.16 %MeHg/g dw/24h, rainy season; 0.56±0.35 %MeHg/g dw/24h, dry season) and soil samples (0.08±0.03 %MeHg/g dw/24h, rainy season; 0.31±0.18 %MeHg/g dw/24h, dry season). Regarding the Flooded Forest samples, the flooding pulse does not seem to influence methylmercury formation in rhizosphere (0.40±0.19 %MeHg/g dw/24h, rainy season; 0.74±0.43 %MeHg/g dw/24h, dry season) nor in soil samples (0.13±0.07 %MeHg/g dw/24h, rainy season; 0.58±0.31 %MeHg/g dw/24h, dry season). Mercury methylation potentials in Madeira River sediments were 0.043±0.012 %MeHg/g dw/24h in the rainy season and 0.12±0.097 %MeHg/g dw/24h in the dry season. In tributaries, mercury methylation potentials were 0.21±0.046 %MeHg/g dw/24h for the rainy season and 0.29±0,254 %MeHg/g dw/24h for the dry season. Observed differences between sampling sites and seasons were not statically different (p = 0,534). The results indicate that the area that will be affected by the construction of the reservoir presents an expressive methylmercury formation in comparison to previous studies in other Amazonian basins.

MG7-P38 — 11:00-12:00 and 17:30-18:30
Authors: MIRANDA, Marcio R.1, GUIMARAES, Jean Remy D.2, ALMEIDA, Ronaldo3, BASTOS, Wanderley R.3, CARVALHO, DP de3, MARINHO, Claudio C.4, SANTOS, JP 3, MENEZES, JM3
(1) Inpetam, topo@biof.ufrj.br ; (2) IBCCF/Universidade Federal do Rio de Janeiro; (3) UNIR; (4) IB/UFRJ;

One of the several environmental impacts of creating reservoirs is the increased methylmercury production and its accumulation in fishes. It can be a result of the major entry of organic matter when the area is flooded. Since biological methylation is a result of microbial activity, this input of organic matter can fuel the methylmercury formation as a result of degradation and mineralization. Samuel Reservoir on the Jamari river (Rondônia, Brazil), built in 1986 is one of the most important hydroelectric power plants in the Amazonian Region. The sediment samples were collected in 4 different sites of the reservoir: P1 - near the dam, P2 - 40 km upstream the dam, P3 - Igarapé São Marcos and P4 - Igarapé Japiin. Samples were incubated with radioactive mercury (203Hg) for 24hs in the dark. Radioactive methylmercury formed (Me203Hg) was measured in liquid scintillation counter after selective extraction in toluene and results were expressed as mercury methylation potential (%). Tributaries presented higher mean values than Jamari River sites (p = 0.0067). The methylation potential measured in tributaries varied from 0.8-1.97 %MeHg/g dry weight/24h, and in the Jamari River varied from 0.23-0.63 %MeHg/g dry weight/24h. Samuel reservoir is an old tropical hydroelectric power plant (20 years), however methylmercury production in its sediments is still high values when compared to the Madeira River bottom sediments.

MG7-P39 — 11:00-12:00 and 17:30-18:30
Authors: OGOREK, Jacob1, KRABBENHOFT, David1, DE WILD, John1, SABIN, Thomas1, TATE, Michael1, THOMPSON, Charles1
(1) US Geological Survey, jmogorek@usgs.gov

Elevated levels of methylmercury (MeHg) are observed in sport and commercial fish across the Great Lakes basin. Although mercury (Hg) is recognized as a primary issue for the Great Lakes, our level of understanding of driving factors (sources, internal cycling, invasive species, and climate change) remains poorly understood. One of the primary reasons for this understanding gap is general lack of data, especially for the Great Lakes themselves. In 2010, the U.S. Geological Survey (USGS) and U.S. Environmental Protection Agency (EPA) initiated a multi-year project to evaluate Hg and MeHg abundance and distribution in water, sediment, and biota from all five Great Lakes and their tributaries on the US side. Acquiring data at this time is critical to establish a baseline for quantifying the potential response of the Great Lakes to pending national (US) Hg emission regulations; to define the relative importance of MeHg sources (tributaries, atmospheric deposition, internal generation); and, to help guide restoration plans and anticipate responses to ecosystem restoration changes: nutrient load reductions, wetland restoration, invasive species, and climate change.

During August 2010, water, sediment, and benthos were collected during a month-long cruise aboard the EPA’s research vessel, Lake Guardian. During that cruise, water samples were collected in profile at 37 stations (4 in Michigan; 4 in Huron; 9 in Erie; 8 in Ontario; and, 12 in Superior); sediment samples from 52 stations (15 in Michigan; 9 in Huron; 12 in Erie; 7 in Ontario; and, 9 in Superior), and benthos from 13 stations (7 in Michigan; 2 in Huron; and, 4 in Erie). Water column profiles were acquired using a remotely operated sampling rosette equipped with 12 Teflon-lined Niskin bottles and an onboard water quality sensor. Samples were collected at specific water depths: mid-epilimnion, mid-thermocline, deep chlorophyll layer, mid-hypolimnion, and the bottom. In addition, water samples were acquired from 59 tributaries in the fall of 2010. Water, biota, and sediment samples are being analyzed for Hg and MeHg, which allows us to examine whether MeHg formation is occurring in the water column, sediments, or both; and, to provide mass flux estimates from streams. Results to date show top-to-bottom declining Hg concentrations at most of the water column sampling sites, suggesting the importance of atmospheric deposition. Methylmercury profiles most frequently show maximum concentrations at the thermocline or deep chlorophyll layer, suggesting water-column methylation.

MG7-P40 — 11:00-12:00 and 17:30-18:30
Authors: CARRIE, Jesse D1, STERN, Gary A.2, SANEI, Hamed3
(1) University of Manitoba, carrie@cc.umanitoba.ca; (2) Freshwater Institute, Dept. of Fisheries and Oceans Canada; (3) Geological Survey of Canada-Calgary.

Temperature-programmed desorption (TPD) of mercury (Hg) is a method that can determine broad classes of Hg-binding species, such as Hg(0), HgX2, humic/organic-bound Hg, HgS and HgO and the relative proportion of each specie in a soil or sediment matrix, in lieu of operationally defined speciation typical of sequential extractions. It also has the benefit of only needing minute amounts of sample (~10-100 mg) for analysis when coupled with a sensitive detector such as atomic fluorescence. We have used TPD to analyse a series of sediment samples from freshwater systems, ranging from surface sediment from rivers from the Mackenzie River Basin in northwestern Canada to lake sediments from numerous sites across Arctic and sub-Arctic Canada. Our results show that samples derived from mountainous tributaries of the Mackenzie River contain Hg predominantly as HgS (likely from weathering of widespread sphalerite (ZnS) deposits), while samples from Arctic lakes show a predominantly humic/organic signal and sub-Arctic lakes show a mix of both humic/organic and HgS. The mountainous tributary sites are concordant with previous sequential extraction data showing sulfide-bound Hg and illustrate the importance of geogenic processes on the Hg in the Mackenzie River Basin. As sample size did not allow for sequential extractions of the Arctic/sub-Arctic lake sites, TD was effective in corroborating the correlations seen between Hg and organic matter fractions from Rock-Eval Analyses (i.e., S2 carbon). This method is also useful to determine relative bioavailability of Hg in sediments and soil.

MG7-P41 — 11:00-12:00 and 17:30-18:30
Authors: MIRANDA, Marcio R.1, GUIMARAES, Jean Remy D.2, BRITO, EF1, ALMEIDA, Ronaldo3, MARINHO, Claudio C.4, COELHO-SOUZA, Sergio A.1, AZEVEDO-SILVA, CE5, CARVALHO, DP de6, MANZATO, AG3, BASTOS, Wanderley R.3, OSTERREICHER, Patricia7
(1) Inpetam, topo@biof.ufrj.br; (2) IBCCF/Universidade Federal do Rio de Janeiro; (3) UNIR; (4) IB/UFRJ; (5) IBCCF/UFRJ; (6) UNIR/Inpetam; (7) PUC-Rio.

Organic matter decomposition can stimulate mercury methylation. The Amazon Region hydrological cycle is dominated by annual flooding events, known as Flood Pulses. Large areas are flooded for long periods, accelerating organic matter decomposition. Since detrital microbial metabolism is one of the major energy flows in all ecosystems, we investigated methylmercury production and microbial activity during the in-situ decomposition of fresh leaf litter in a stream of the Madeira river basin. Leaf packs were placed in the stream (igarapé) and leaves sampled 3, 7, 17 and 51 and 83 days after the beginning of the experiment. Samples (approximately 2g ww) were incubated in situ with 203HgCl2 for 24 hs. Formed Me203Hg was extracted in toluene and radioassayed by liquid scintillation. Bacterial carbon production was estimated from incorporation of radio-labeled leucine (3H or 14C) into bacterial protein over time. The incubations were done in 2 mL microcentrifuge tubes containing a leaf disc (0.5 cm diameter) and 1 mL of filtered (0.22 µm) stream water. The 60-min. incubations were done in the dark at 25°C and interrupted with formaldehyde (5% final concentration). Bacterial proteins were extracted using the trichloroacetic acid method combined with microcentrifugation. Eh, pH, dissolved oxygen, conductivity, temperature and suspended solids were measured during the sampling. Here we show that high mercury methylation potentials (10.63±6.53 %MeHg/g dry weight/24h) and bacterial carbon production (0.13±0.02 ng C/cm2/h) were found in decomposing Chamaecrista negrensis leaves in the first week of exposure. After two weeks, mercury methylation and bacterial production decreased. Mercury methylation and bacterial production showed a significative linear regression (R2 = 0.8041, p = 0,0392). This finding evidences the importance of the microbiological community activity on methylmercury formation. The results indicate that the leaf litter has an enormous potential for methylmercury formation. This potential is related with several environmental variables. Among these, the quality of organic matter to the aquatic system and its subsequent degradation are likely to be the most important ones.

MG7-P42 — 11:00-12:00 and 17:30-18:30
Authors: OLIVEIRA, Diane C1, SILVA, Raquel Rose C.2, GUIMARAES, Jean remy D.2
(1) IBCCF/Universidade Federal do Rio de Janeiro, jeanrdg@biof.ufrj.br; (2) IBCCF/UFRJ;

Mangroves occupy extensive areas of tropical and subtropical coast lines, possess high ecological importance concerning its productivity and fish nursery functions and are usually located in areas of high anthropic pressure and pollution. Although mercury pollution generates environmental and health concerns, there are few studies regarding its accumulation and behavior in this ecosystem. The study was conducted in a mangrove located near Rio de Janeiro, Brazil, with the objective of analyzing MMHg formation in sediments under different vegetation covers (mangrove trees and aquatic macrophytes), at different depths, incubation periods and salinities. Adjacent litter and Spartina sp. root samples were also analyzed for MMHg formation. The experiments consisted of incubation with 203Hg, MMHg extraction (MM203Hg) and counting by liquid scintillation. Results suggest that mangrove sediment is capable of significant Hg methylation (up to 14% of MMHg), the latter being higher in superficial sediment fractions. Incubations of up to 10 days showed that MMHg formation leveled off after 1-2 days.. No significant correlations were found between Eh, pH, salinity gradients (1.5 to 28‰) and MMHg formation. Sediment is not the only mangrove substrate in which the process can occur, as Spartina sp. roots and litter methylated up to 3% and 1.5% respectively. MMHg formation is present in mangrove ecosystems and needs to be further studied since this metal represents a contamination risk to local biota and human populations that feed on them.

MG7-P43 — 11:00-12:00 and 17:30-18:30
Authors: DAVIS, Eric1, BOWDEN, William 1
(1) University of Vermont, Eric.Davis@uvm.edu

The adverse effects of mercury contamination on wildlife are well documented and include inhibited performance, maintenance, and reproduction. Anoxic conditions in aquatic systems facilitate the transformation of atmospherically deposited, inorganic mercury to bioavailable methylmercury. Increasing concentrations of dissolved organic carbon (DOC) and decreasing pH are positively correlated with increased methylation. Vernal pools may be efficient methylators of mercury as these ecosystems are characterized by these conditions. Mercury is of particular concern in vernal pools as they provide critical breeding habitat for amphibian populations. In addition to the direct detrimental effects mercury contamination has on these globally pressured populations, amphibians are a high quality food source and comprise significant biomass. Their importance as a trophic link represents a potential vector for methylmercury to move out of vernal pools and biomagnify up trophic levels. Vernal pool hydrology is distinctly different from other aquatic systems in the variability of inundation and the extent of drying. We hypothesize that this affects methylation dynamics.

Eight vernal pools have been selected for this study, four in each of Marsh Billings National Historical Park in Woodstock, VT and Saratoga National Historical Park in Saratoga, NY. Sites were selected along a morphological gradient of deep, well confined sites to shallow, extensive sites to include pools with varying hydrologic regimes. Duration of inundation and variation of inundation were quantified by deploying pressure transducers that continuously logged water level. High-resolution elevation maps will be constructed using RTK GPS surveying equipment allowing water level data to be extrapolated to surface area inundated. Mercury samples were collected once at each pool in fall 2010 and three times throughout spring 2011. Additionally, two pools were sampled at a greater frequency during the spring 2011 season to better capture temporal variation in these systems. Each time a mercury sample was collected, additional samples were collected and analyzed for chemical parameters that have been shown to affect methylation in aquatic systems. These parameters include pH, temperature, conductivity, DOC, and sulfate concentrations. Methylation efficiency (MeHg/THg) was calculated for each water sample. The characteristics of the hydrologic regime will be combined with the chemical parameters as independent variables in a multivariate analysis in which the total mercury, methylmercury, and methylation efficiencies are the dependent variables to determine how hydrology may influence methylation.

MG7-P44 — 11:00-12:00 and 17:30-18:30
Authors: RICHARDSON, Murray1, BRANFIREUN, Brian2
(1) Carleton University, murray_richardson@carleton.ca; (2) University of Western Ontario.

Throughout Canada’s northern forested regions, the fate and transport of terrestrially sequestered atmospheric mercury and its potential to prolong recovery of Hg impacted aquatic ecosystems is a topic of ongoing concern. It remains unclear whether contemporary rates of atmospheric mercury are necessarily a first-order control on Hg accumulation in freshwater biota given the large terrestrial supply of sequestered Hg now residing in terrestrial ecosystems, and its potential sensitivity to interannual variability in weather and longer term climate shifts. In Precambrian Shield landscapes, seasonal and inter-annual variability in terrestrial watershed processes result in large variations in terrestrial aquatic fluxes of solutes linked to watershed Hg transport and speciation, including sulphate, dissolved organic carbon (DOC) and hydronium ion. The purpose of this study was to determine the extent to which such variations can influence biotic Hg uptake in headwater lakes.

An analysis of historical monitoring data of streamflow discharges, hydrochemistry, and Hg concentrations in yearling perch in two contrasting headwater lake basins in the south-central Precambrian Shield region of Ontario, Canada was conducted. For the period between 1981-1989, annual terrestrial aquatic fluxes of DOC, sulphate, hydronium ion and natural colour to Harp and Dickie Lakes were calculate based on the historical data record provided by the Ontario Ministry of the Environment Dorset Environmental Science Centre. Correlation analyses were conducted between watershed loading rates of the key solutes and corresponding in-lake concentrations and physico-chemical conditions. Also included in the correlation analyses were late august concentrations of Hg in yearling perch. At Dickie lake, a shallow, humic lake with persistent summertime hypolimnetic anoxia, Hg concentrations in yearling perch were strongly and positively correlated to water year runoff (r=0.75, p<0.01), sulphate flux (r=0.83, p<0.01) and hydronium flux (r=0.96, p<0.001), and to a lesser extent in-lake temperature (r=0.75, p<0.1). At Harp Lake which is a deeper, cooler, oligotrophic lake, Hg concentrations in yearling perch were completely decoupled from terrestrial-aquatic fluxes of water and solutes. The results underscore the critical need for whole-ecosystem monitoring within physiographically distinct watersheds and regions in order to refine our understanding of how landscape characteristics, basin morphometry, climate and weather interact and co-influence uptake of mercury by aquatic biota in northern forested regions.

MG7-P45 — 11:00-12:00 and 17:30-18:30
Author: WILDMAN, Rich1
(1) Harvard University Center for the Environment and Harvard School of Public Health, rwildman@hsph.harvard.edu

Methylation of mercury, which increases mercury accumulation in fish and health risks to humans and wildlife, is understood to occur when anoxic conditions stimulate the growth and metabolism of anaerobic bacteria. Anoxia occurs in stratified lakes and reservoirs primarily in hypolimnia or sediment when loads of settling particulate organic carbon are high and aerobic metabolism occurs too rapidly for dissolved oxygen to be replenished through photosynthesis or diffusion from the atmosphere.

The physical circulation of surface water is often a primary driver of dissolved oxygen concentrations. In lakes, this is primarily influenced by seasonal changes in surface water temperature. Heating of surface water in summer promotes stratification and possibly anoxia in bottom waters, whereas cooling during autumn leads to sinking of surface water, turning over of the water column, and increased dissolved oxygen in bottom water. Circulation of reservoirs is usually considerably more complicated for several reasons. First, lateral flow from the inflow region towards the dam is non-negligible (i.e., a river is constantly entering a reservoir and exiting out the dam) and can include seasonal overflow or underflow density currents. Second, water is often released out dams from below the thermocline, which can imply that overflow currents are forced downwards near the dam or that underflow currents never return to the surface as they would in a lake. Third, water residence time varies spatially and temporally due to varying depth and dam operations.

These hydrodynamic processes have been shown to affect dissolved oxygen concentrations, and we are testing a hypothesis that they can also affect mercury speciation and transport through Grand Lake, which is a reservoir of moderate size (~80 km long, ~40 m deep at the dam) in northeast Oklahoma, USA. We are currently measuring circulation (T, SC), water chemistry (DO, Fe(aq), Mn(aq), pH), and mercury speciation (dissolved and particulate total mercury, dissolved and particulate methylmercury, and dissolved gaseous mercury) during sampling campaigns in autumn 2010, winter 2011, spring 2011, summer 2011, and autumn 2011 to assess the effect of seasonal variations in reservoir circulation on production of methylmercury and export of total and methylmercury downstream. The creation of reservoirs has long been understood to promote mercury methylation due to the flooding of riparian wetlands, but, if confirmed, our hypothesis will suggest an important additional pathway for mercury methylation that is relevant for reservoirs long after their period of initial flooding has passed.

MG7-P46 — 11:00-12:00 and 17:30-18:30
Authors: HINTELMANN, Holger1, DIMOCK, Brian1, ZHU, Joy1, GREENFIELD, Ben2, HARROLD, Katie 2, RIDOLFI, Kat2, SLOTTON, Darell3
(1)Trent University, hhintelmann@trentu.ca; (2) San Francisco Estuary Institute; (3) UC Davis.

The performance of new DGT sentinels was explored as a monitoring tool to compare levels of methylmercury (MMHg) in the San Francisco Bay area. Samples were collected at 44 sites according to a semi-random protocol originally developed for the Small Fish sampling project. Stations in the spatial survey were sorted into i) open regions, ii) enclosed bays and wetlands, iii) sites with historically elevated mercury or MMHg concentrations, iv) sites with industrial influence, and v) locations downstream of Waste Water Treatment Plants.

The DGT sentinels identified the open bay regions to be consistently the least exposed to MMHg. In contrast, enclosed bays show greater variability among sites. Stations with industrial influence were on average much higher in MMHg compared to open sites. Legacy sites, despite their large variation in MMHg were also significantly elevated. The situation downstream of the Waste Water Treatment Plants is inconclusive. When evaluating all data on the spatial scale, the highest levels of MMHg were generally observed in the South Bay, and the by far highest concentrations were measured in the Alviso area. A clear gradient towards the central Bay follows. Also the eastern area of the Northern Bay shows a few very high spots of MMHg, potentially indicating influence of Sacramento River discharge. Compared to the other Bay regions San Pablo Bay was generally much lower in MMHg. A third hot spot was found around Point Isabel in the Central Bay, which was otherwise low in MMHg. This general pattern suggest formation of and enhanced presence of MMHg from historic mining activities, which still export elevated levels of Hg to both ends of the Bay (Guadalupe and Sacramento rivers). In place generated MMHg is subsequently diluted by exchange with ocean water, leading to background levels of MHHg near the Golden Gate Bridge area, which experiences the highest flushing rates.

MMHg uptake into the DGTs was also compared to MMHg concentration in small fish sampled through the Small Fish project. On a large scale, the fish monitoring program revealed similar south to north gradient with lowest Hg in fish levels found in San Pablo Bay. This suggests that both monitoring tools identify the same general trends of MMHg exposure. However, smaller scale, site-by-site comparisons provided only weak correlations between fish and DGT data.

MG7-P47 — 11:00-12:00 and 17:30-18:30
Authors: KASCHAK, Elisabeth1, KÖNIG, H.1
(1)Institute of Microbiology and Wine Research, Johannes Gutenberg-University Mainz, Germany, Kaschak@uni-mainz.de

Mercury and its derivates are widely distributed in the biosphere and are mainly released by anthropogenic activities such as stationary combustion like coal fired power plants, waste disposal and cement production. These industrial processes are responsible for the current debit of our environment with these toxic compounds [1]. Especially alkylated mercury species are potent neurotoxins due to their ability to cross the blood-brain barrier. They are involved in the most serious human poisonings by organometals. Biomagnifications in higher trophic levels is a critical aspect in environmental cycles.

Biotic methylation of mercury by aquatic microorganisms is well studied. In contrast, there is less knowledge about such reactions in soil and the features of the involved microorganisms. We studied the accumulation of methylmercury by the earthworm Eisenia foetida after feed with soil enriched with mercurychloride. In order to investigate whether methylmercury in worm tissue originates from soil or gut bacteria we isolated bacteria from Eisenia foetida. These microorganisms were cultivated on different media and tested for their capability of mercury transformation. For further characterization, microorganisms were identified by denaturating gradient gel electrophoresis (DGGE) and 16S rDNA-analysis. Resistance to inorganic and organic mercury compounds in Gram-positive or Gram-negative bacteria is typically mediated by a mercuric reductase enzyme (MerA), a protein (MerT) for Hg0 uptake and MerR, a metal responsive regulator [2]. For further comprehension of mechanisms of mercury transformation we screened bacterial cultures for presence of MerA, MerT and MerR. Of special interest is the speciation of anorganic and alkylated mercury at trace levels, because toxicity and bioavailability depends on chemical species. The detection of mercury compounds in complex samples like animal tissue or bacterial culture media was performed by gas chromatography atomfluorescence spectrometry (GC-AFS) and gas chromatography inductively coupled plasma mass spectrometry with isotope dilution method (ID-GC-ICP-MS). Sample preparation included derivatization with sodium tetra-(n-propyl)-borate and liquid extraction with hexane. Our experiments indicate that Enterobacteriaceae, Pseudomonadaceae and Clostridiaceae are responsible for mercury methylation in the gut of invertebrates such as earthworms.

[1] P. Craig, Organometallic Compounds in the Environment (2003), Wiley.
[2] T. Barkay, Bacterial mercury resistance from atoms to ecosystems, FEMS Microbiology Reviews 27 (2003) 355-384.

MG7-P48 — 11:00-12:00 and 17:30-18:30
Authors: GHSEMPOURIA, Seyed Mahmoud 1, SARI, Abbas Esmaili1, ABDOLLAHZADEHB, Esmail 2, SHOOSHTARIA , Sharif Joorabian 1
(1) Department of Environment, Faculty of Natural Resources and Marine Sciences, Tarbiat Modares University; (2) Department of Fisheries, Faculty of Natural Resources and Marine Sciences, Tarbiat Modares University;

Toxicological studies confirm that mercury, specially methyl mercury, is a highly toxic component. This review article covers major studies on mercury levels in several taxa contain human, plants, wildlife and aquatic animals, which have been done in Iran during the last decade. For this purpose, data regarding 83 treatments of 23 studies from a six-years period were collected and were re-evaluated.
Animal tissues which studied including hair, feather, nails, muscle, liver, kidney, spleen, gonad, skin and gills, various plant tissues (e.g. roots, stems, leaves and seeds) also were investigated in different geographical regions and nine provinces were sampled during the studies period.
Oriental sole with 5.61 µg/g total Hg in muscle had shown the most pollution among the Persian Gulf fishes. For avifauna, 9.25 µg/g in kidney in great cormorant was the most and the highest mercury level among mammals observed in Caspian seal hair (10.12 µg/g dw). Minimum and maximum levels of mercury in hair in different group of Iranian people, like pregnant women, dentists, students were 0.07 µg/g and 53.56 µg/g.

MG7-P49 — 11:00-12:00 and 17:30-18:30
Author: SHANG, Lihai1
(1) Institute of Geochemistry, Chinese Academy of Sciences, shanglihai@vip.skleg.cn

Reservoirs are considered as sensitive aquatic systems for mercury since large quantity of organic matter (OM) and mercury are leached from the impounded vegetation and soil. In these systems, fish often pose a health threat because of theirhigh methylmercury concentrations. Sediment records the past history of both mercury and OM inputs. Lignin is an organic macromolecule specific to terrigenous vegetation. Lignin derived biomarkers signature allow a characterization of organic matter sources and degradation state. Honfeng and Wanfeng reservoirs are located in SW China, with impoundment dating back about 50 years and 13 years respectively. To identify mercury sources, sediment cores were collected from several sites in both reservoirs. Total Hg (THg) contents were analyzed by Lumex pyrolysis method. Organic matter concentrations were measured on a PE2400-II elemental analyzer whereas lignin derived phenols were measured by GC-MS after a mild copper oxidation (CuO). By tracing the different vegetation sources of organic matter as well as describing its state of degradation, we aim at identifying if impoundment is responsible for increase in Hg inputs in the aquatic system and how it impacts methylation processes in such man made subtropic water bodies.

Monday, 25 July, 2011