Mercury Speciation and Mobilization in a Wastewater-Contaminated

Article pubs.acs.org/est

Mercury Speciation and Mobilization in a Wastewater-Contaminated Groundwater Plume Carl H. Lamborg,†,* Doug B. Kent,‡ Gretchen J. Swarr,† Kathleen M. Munson,† Tristan Kading,† Alison E. O’Connor,§ Gillian M. Fairchild,∥ Denis R. LeBlanc,⊥ and Heather A. Wiatrowski# †

Woods Hole Oceanographic Institution, Department of Marine Chemistry and Geochemistry, Woods Hole, Massachusetts 02543, United States ‡ United States Geological Survey, Menlo Park, California 94025, United States § Oberlin College, Oberlin, Ohio 44074, United States ∥ Virginia Institute of Marine Science, Gloucester Point, Virginia 23062-1346, United States ⊥ United States Geological Survey, Northborough, Massachusetts 01532, United States # Clark University, Worcester, Massachusetts 01610, United States S Supporting Information *

ABSTRACT: We measured the concentration and speciation of mercury (Hg) in groundwater down-gradient from the site of wastewater infiltration beds operated by the Massachusetts Military Reservation, western Cape Cod, Massachusetts. Total mercury concentrations in oxic, mildly acidic, uncontaminated groundwater are 0.5−1 pM, and aquifer sediments have 0.5−1 ppb mercury. The plume of impacted groundwater created by the wastewater disposal is still evident, although inputs ceased in 1995, as indicated by anoxia extending at least 3 km downgradient from the disposal site. Solutes indicative of a progression of anaerobic metabolisms are observed vertically and horizontally within the plume, with elevated nitrate concentrations and nitrate reduction surrounding a region with elevated iron concentrations indicating iron reduction. Mercury concentrations up to 800 pM were observed in shallow groundwater directly under the former infiltration beds, but concentrations decreased with depth and with distance down-gradient. Mercury speciation showed significant connections to the redox and metabolic state of the groundwater, with relatively little methylated Hg within the iron reducing sector of the plume, and dominance of this form within the higher nitrate/ammonium zone. Furthermore, substantial reduction of Hg(II) to Hg0 within the core of the anoxic zone was observed when iron reduction was evident. These trends not only provide insight into the biogeochemical factors controlling the interplay of Hg species in natural waters, but also support hypotheses that anoxia and eutrophication in groundwater facilitate the mobilization of natural and anthropogenic Hg from watersheds/aquifers, which can be transported down-gradient to freshwaters and the coastal zone.

■

We have noted previously 9 that watersheds on Cape Cod, and by extension perhaps a number of other coastal systems, appear to behave in a way that is consistent with much higher rates of mobilization. At that time, we hypothesized that the relatively low organic carbon content of the sandy aquifer sediments resulted in uniquely low retention. However, Cape Cod is a location where septic leach fields are in common use for wastewater management, and this results in large amounts of macronutrients and organic carbon entering local groundwater that cause subsurface anoxia and coastal eutrophication.10

INTRODUCTION

The loading of mercury (Hg) to freshwater and nearshore marine ecosystems comes from both direct atmospheric deposition and mobilization from watersheds. Typically, the areal watershed contribution divided by the contemporaneous areal atmospheric deposition rate is about 0.25 in temperate systems.1 This implies that about 25% of the Hg deposited to a watershed is mobilized to receiving waters, although the actual mobilization rates are probably much lower and characterized by long lag times.2,3 In previous studies, it has also been noted that much of the mobilized Hg leaves the watershed as Hg(II) complexed to dissolved organic carbon (DOC) compounds through overland flow and to a lesser extent groundwater flow.4−8 © 2013 American Chemical Society

Received: Revised: Accepted: Published: 13239

May 31, 2013 October 18, 2013 November 4, 2013 November 4, 2013 dx.doi.org/10.1021/es402441d | Environ. Sci. Technol. 2013, 47, 13239−13249

Environmental Science & Technology

Article

Figure 1. Map of the study region. Sampled multilevel sampling wells (MLS) are labeled in black by well number (e.g., F605). Two additional sites, marked in red, are sites where solid-phase samples were taken. The region denoted as the plume represents the largest horizontal extent of impacted groundwater at the time of the end of wastewater disposal.

Such a situation was investigated by Barringer and colleagues,11 who noted that high dissolved Hg concentrations, some exceeding the US EPA drinking-water standard, were found in the Kirkwood-Cohassey aquifer in southern New Jersey in locations where septage was released to groundwater, suggesting that either septage is high in Hg or that an indirect effect of aquifer eutrophication is enhanced Hg mobilization. Black and colleagues 12 also noted a circumstantial connection between elevated Hg concentrations in groundwater and the presence of wastewater in California. However, previous measurements of Hg in septage have suggested that it is not necessarily a strong direct source of Hg11,13 Thus, wastewatermediated mobilization of Hg from aquifers appears to be a plausible alternate hypothesis for elevated Hg concentrations and fluxes in Cape Cod groundwater and perhaps in other instances of large nutrient loadings to watersheds. The mechanism by which wastewater might facilitate Hg mobilization from aquifers is unclear. Barringer and colleagues11 hypothesized that production of elemental Hg

(Hg0), perhaps through reduction by Fe(II), might be the cause. However, they lacked speciated Hg measurements to examine this mechanism further. To test this hypothesis and to examine the relationship between Hg speciation and biogeochemical conditions in groundwater, we made use of the extensive network of sampling wells installed by the United States Geological Survey (USGS) in and around the Massachusetts Military Reservation (MMR) and stretching southward along the general alignment of a geographic low known locally as the Ashumet Valley (Figure 1). This current study focused on a plume of impacted groundwater within the Ashumet Valley created by the discharge of effluent from the MMR’s wastewater treatment plant located on the southern end of the property, near the borders of the towns of Falmouth, Mashpee and Sandwich, MA. We report here the results of Hg speciation measurements from these wells, as well as those geochemical and water quality measurements made as part of the observational and experimental program being conducted at 13240

dx.doi.org/10.1021/es402441d | Environ. Sci. Technol. 2013, 47, 13239−13249

Environmental Science & Technology

Article

Unfortunately, it was beyond the financial scope of this current study to collect new sediments from the well sites in order to assess contemporary aqueous-solid partitioning. Instead, we made use of archived samples collected near some of the MLS sampled. The use of older sediments raises potential problems related to either (1) concentrations in the sediment have changed at these sites during the intervening years and (2) some change in the concentration during storage. We discuss the potential implications of these changes on our findings below. Sediment samples were collected from two upgradient sites, F605 (6/19/2009) and S467 (9/12/2000) and one site, S317 (6/20/1991) within the near-source transect (Figure 1). The aquifer sediments consist of unconsolidated coarse sand and gravel to a depth of about 30 m and are underlain by finer sands and silts that extend to bedrock at approximately 70 m. Closer to the coast, there is a freshwater/ saltwater interface at depth, creating a subterranean estuary (e.g., ref 25), but the locations sampled in this work are greater than 6 km from the shoreline and were all in fresh groundwater. The average porosity of aquifer sediments in the study area is 0.39, with high hydraulic conductivity (ca. 100 m d−1), and a horizontal advection rate determined to be about 200 m yr−1.24 Quartz is the dominant mineral, comprising greater than 90% of the sand-sized and smaller grains.26 Sorption and many other chemical properties of the sediments in this aquifer are controlled by nanometer-scale aluminum-substituted goethite and micrometer-scale detrital chlorite and Illite, the dominant constituents of mineral-grain coatings.27 Dissolved Mercury Collection and Analysis. Samples for Hg were generally collected at the same time as those for the other parameters. All samples collected for Hg were filtered through a 0.45-μm pore-sized polyethersulfone membrane capsule filter (Millipore GWSC04501), allowing about 1 L to pass through the capsule before collection commenced. No attempt was made to preserve the redox poise of the samples during collection, though sample bottles (2-L Teflon) were filled to the top when possible to limit gas exchange. The collection bottles were rigorously acid-cleaned prior to use and handled using trace metal clean protocols (e.g., refs 28 and 29). Total Hg and Hg species (including Hg0, (CH3)2Hg, and CH3Hg+) were determined using new and established methods.29−35 In brief, within a few hours following collection, a 200 mL portion of the sample was poured off into a glass IChem bottle and preserved with 20 mM BrCl for later analysis of total Hg by purge and trap/cold vapor atomic fluorescence spectrometry (CVAFS; ref 36). The remaining sample volume was immediately purged with ultrahigh purity N 2 to preconcentrate the two dissolved gaseous species (Hg0 and (CH3)2Hg). Preconcentration of these two forms was achieved by loading onto Tenax and gold-coated sand traps in series (e.g., ref 29). The (CH3)2Hg was determined by CVAFS following heating of the Tenax, isothermal GC separation (on 15-cm column of 5% OV-3 on Chromosorb W-DMCS), and pyrolysis to Hg0. The dissolved Hg0 in the sample was determined as the Hg released from the Au trap upon heating. Blank contributions from this method were assessed by immediately repurging some of the samples and determining the Hg0 and (CH3)2Hg collected. The concentration of CH3Hg+ was determined by acid extraction with H2SO4, neutralization and direct ethylation, followed by purge and trap/detection of the formed CH3HgCH2CH3 in a way analogous to that of (CH3)2Hg.33 Sulfide can interfere with this analytical approach, but

the site by USGS (e.g., ma.water.usgs.gov/MMRCape/; e.g., refs 14−16).

■

METHODS MMR and Sampling Sites. Wastewater generated on the MMR was treated at the Reservation’s treatment plant and then discharged to infiltration beds covering a total area of 4.6 ha through a system of pipes designed to distribute the flow among the beds.17 Discharge to the beds began in 1936 and continued to 1995. Flows for the period 1936 to 1980 were estimated to total 8 billion gallons (30 GL; ref 17). A great deal of research has been conducted on contaminant-transport processes in this plume and natural restoration following the end of disposal (toxics.usgs.gov/bib/bib-cape-cod.html). The plume stretching southward from the beds and into the Ashumet Valley 18 contained wastewater that was comparable to ordinary municipal waste. Wastewater discharge resulted in the development of distinct biogeochemical zones down-gradient of the beds.19,20 Ambient groundwater above and below the plume is mildly acidic with low concentrations of dissolved salts. During active discharge, the upper region of the plume had low concentrations of dissolved oxygen (2 mm), which generally comprises less than 10% by weight of the sediments, was removed by dry-sieving.26 Total Hg on the solid-phase was determined by automated pyrolysis/Au preconcentration/atomic absorbance spectrometry (US EPA Standard Method 7473; ref 38) using a Milestone DMA-80 in the Bothner Laboratory of USGS. The instrument could only accommodate small amounts of sample. In order to achieve reproducible measurements with these small sample sizes, the mud-size fraction (silt plus clay, i.e., less than 0.063 mm) was separated from the sand-size fraction (0.063−2 mm) by dry sieving, and the two fractions were analyzed separately for each sample. Water Quality Parameters. Numerous other parameters are routinely determined at these sites by the USGS as recently described by Savoie and colleagues.39 Of that suite of analytes, we report here results for pH, specific conductance, dissolved O2, dissolved Fe, Mn, NO3−, NH4+, and DOC. In brief, pH and conductance were measured in the field by using electrodes (Orion), dissolved O2 was determined in the field for samples whose concentrations were below 31 μM by using a colorimetric method (CHEMetrics K-7553) and in the laboratory within a few hours of collection in BOD bottles by electrode for concentrations greater than 31 μM. The remaining analytes mentioned here were measured in 0.45μm filtered water, following preservation (pH < 2 with nitric

■

RESULTS AND DISCUSSION Solution-Phase Hg Concentration and Speciation. The concentrations of Hg species in relation to the predominant biogeochemical conditions are shown in Figures 2 and 3, arranged into transects through the plume near the source and in the Ashumet Valley. Results are also summarized in Supporting Information, SI, Table S1. Some of the groundwater chemical conditions along the near-source transect have changed as a result of the cessation of the wastewater disposal in late 1995. As demonstrated by Repert and colleagues,14 the concentrations of the weakly sorbing constituents in the groundwater nearest the beds decreased to near uncontaminated levels relatively quickly because of the rapid advection of water through the aquifer (∼200 m y−1). For example, boron concentrations have decreased to background values throughout the near source transect as a result of flushing with ambient groundwater.14 Nitrate concentrations near the source have also decreased but remain greater than background concentrations near the upper boundary of the wastewatercontaminated zone owing to continued supply from nitrification of ammonium and organic nitrogen bound to vadose-zone sediments in the abandoned disposal beds (not shown). However, the size of the anoxic the core of the plume has only slightly diminished since cessation as the result of persistent biological oxygen demand that continues to be fed by organic carbon stocks built up during the time of active disposal. This is also evident in the persistence of dissolved Fe in the groundwater at sites F343 and F566 (Figure 2). 13242

dx.doi.org/10.1021/es402441d | Environ. Sci. Technol. 2013, 47, 13239−13249

Environmental Science & Technology

Article

Figure 3. Concentrations of total dissolved Hg, N species, and O2, as well as the percent of total Hg as CH3Hg+ in the treated-wastewater plume in the Ashumet Valley. The graphs are arranged in the direction of groundwater flow from up-gradient (left) to down-gradient (right).

wastewater disposal beds (S474) is located under a pad formerly used to dry sludge from the wastewater treatment plant. Recharge through sludge-contaminated sediments likely accounts for the relatively high concentration (22.4 pM) observed near the water table (Figure 2). Otherwise, the concentrations of Hg just up-gradient of the infiltration beds and in the deepest, oldest groundwater were very low ( 4), close to theoretical values for Fe oxides (log Kd = 2) in the absence of organic ligands, providing circumstantial support for Fe phases as being important in Hg sorption in the plume. Low log Kd values could also be caused an increase in the concentration of dissolved Hg-binding ligands that are less sorptive than bulk DOC. Hsu-Kim and Sedlak have documented that wastewaters appear enriched in strong Hg-binding ligands, for example, which could provide such a qualitative change in the DOC pool.49 Regardless of the cause for the difference in Kd values, the dramatically lower Kd values in the plume indicate that under these conditions, Hg is less prone to sorption and therefore more mobile than in uncontaminated regions of the aquifer. As noted in the Methods section, there are potential artifacts to our Hg measurements of the solids due to the length of their storage as well as any possible real changes that occurred in local sediments between the time of sediment collection and our more recent water collections. Long-term storage of sediment material for Hg analysis has not been examined but is likely unimportant due to the relatively high concentration of the soils (unlike the situation with natural waters). As to the real changes that may have occurred in the contaminated sediments since their collection, it is most likely that if recovered today soils from these same sites would show a lower concentration of Hg as a result of cessation. Therefore, Kd values for contemporary samples from the contaminated sites are likely even lower than we present here, which strengthens the argument for sorption to nonorganic phases. Mercury Attenuation and Mobilization. As noted, the concentration of total dissolved Hg is dramatically higher immediately under the infiltration beds (S469). The concentration measured at the shallowest depths greatly exceeds those typically reported for precipitation (∼50−100 pM; ref 3) even if it were further enhanced by dry deposition.50−52 As this elevated Hg is still observable 15 years after cessation and

(farther down-gradient), the concentration of total Hg returned to lower levels, with some enhancements that were commonly associated with redox transition conditions. Speciation differences in Hg were also evident in the two plume regimes. Near-source, increases in Hg0 were associated with increased dissolved Fe concentrations. Figure 4 shows the

Figure 4. The correlation between dissolved Hg0 and dissolved Fe(II) in the northern end of the plume.

correlation of Hg0 with dissolved Fe for those wells/depths where both values were available and Fe was detectable. This correlation was evident for those samples where Hg0 comprised a substantial (>10%) fraction of total Hg. For some samples collected under the beds, the absolute amount of Hg0 was high, but a low percent of total Hg, suggesting little in situ production. Dissolved Fe concentrations are supported by large concentrations of sorbed Fe,22 but the correlation suggests progressive release of Hg during dissimilatory Fe reduction. Within the Fe reducing region of the plume, there is a notable decrease in the percentage of Hg present in the CH3Hg+ form (Figure 5). Above and below the plume, the CH3 Hg + concentration is similar to what is observed in other freshwaters and comprises just a few percent of total dissolved Hg, but within the plume, CH3Hg+ concentration drop to below detection. This is in stark contrast to the sites along the valley transect, where CH3Hg+ represented >10% (sometimes exceeding 100% due to analytical uncertainty) of total dissolved Hg at many depths within the low-oxygen region of the plume (Figure 5, SI Table S1). Such high percentages of total Hg as CH3Hg+ are unusual in natural waters except for organic rich sediment porewaters (e.g., ref 46). As these aquifer sediments are in general very organic-poor compared to most fresh- and saltwater sediments, this speciation distribution is striking. Solid-Phase Concentrations and Sediments As a Reservoir for the Solution-Phase. The Hg concentrations in uncontaminated sediments from the “pristine” (here defined as not contaminated with wastewater) and up-gradient sites (F605 and S467, respectively) are 0.6−1.2 ppb (SI Table S2), considerably lower than typical estimates of the average abundance of Hg in the upper crust of about 50 ppb.47 Although these solid-phase concentrations are low, sedimentbound Hg constitutes a significant potential source of Hg to groundwater if the sorption characteristics of the sediments or aqueous chemical conditions were to change to favor release of Hg. On the basis of an average solid−liquid ratio in the aquifer of 4140 g/L,48 0.6−1.2 ppb sediment-bound Hg, if entirely released into groundwater, would yield dissolved Hg concentrations in the range 12 000−25 000 pM. Not all of this Hg is 13244

dx.doi.org/10.1021/es402441d | Environ. Sci. Technol. 2013, 47, 13239−13249

Environmental Science & Technology

Article

Figure 5. The percentage of total dissolved Hg as CH3Hg+ at each of the MLS sampled.

concentrations of sediment-bound mercury in wastewatercontaminated vadose- and shallow groundwater-zone sediments greatly exceed background sediments, it is likely that the sediments of the vadose zone (between 0 and 15 m below the surface) contain a substantial burden of Hg that was emplaced during the 60 years of wastewater application at the site. Thus, the presence of Hg in this zone could act as an ongoing source of Hg to the aquifer below.

Although no Hg concentration data exist for this treatment plant, the loading to the beds can be estimated by referring to other wastewater effluent measurements of Hg. Balcom et al.,53 for example, found between 64 and 97 pM dissolved and particulate total Hg in Connecticut wastewater treatment facilities, and between 1.3 and 1.8 pM CH3Hg+. This suggests 75 pM as a reasonable first order estimate for the amount of Hg loaded into the infiltration beds at the MMR. Extrapolating flow data from 1936 to 198017 to cessation suggests that about 13245

dx.doi.org/10.1021/es402441d | Environ. Sci. Technol. 2013, 47, 13239−13249

Environmental Science & Technology

Article

Geobacter59,60). Non-δ-proteobacteria Fe reducers (e.g., Shewanella), however, do not appear to methylate Hg.66 Thus, our data from the wastewater plume suggest that some combination of community composition or abiotic reduction driven by Fe2+ may explain the distribution of Hg0 within the northern end of the plume. This trend is consistent with the hypotheses of Barringer and colleagues, who proposed that production of Hg0 in situ might be the cause for great Hg mobility observed in the Kirkwood-Cohassey aquifer (e.g., ref 11). At the southern end of the plume, where nitrate and ammonium are more abundant, Hg0 is a minor species, but CH3Hg+ comprises a surprisingly large percentage of total Hg. This co-occurrence is seen at the broad scale (southern end of the plume version the north), but not always seen on a well-bywell and depth-by-depth basis. There are no reports that we are aware of linking N-cycling with net methylation of Hg, though Barringer and colleagues noted high concentrations of methylated Hg in the few samples test.61 To the contrary, there is at least one report of denitrification leading to net demethylation of Hg in a lake,62 and at least some denitrifying bacteria are known Hg(II) reducers58 and CH3Hg+ demethylators.58,63 This surprising observation could be caused by the action of some N-related metabolism that has yet to be tested for Hg cycling activity. Eutrophic, anoxic groundwater can be the site of some biogeochemical processes which have not received scrutiny by the Hg community, including ANAMMOX and DNRA and ANAMMOX is thought to be occurring in this stretch of the plume.43 There is also evidence from other locations for a coupling of nitrate reduction to sulfide oxidation (e.g., refs 64 and 65), which could conceivably mask some amount of sulfate reduction occurring within this zone of the plume and which could in turn explain the Hg methylation that is apparently occurring there, although there is not much evidence for net nitrate depletion in the southern end of the plume. Transport times to the center of the plume at F350 are 29−38 years.66 Persistence of nitrate for this long shows that nitrate metabolic activity is extremely slow, consistent with likely carbon-limitation to metabolic processes in this region of the plume.21 Microbial processes leading to methylation and/or demethylation of mercury may also be occurring slowly and the products accumulating in groundwater as it is transported down-gradient. Therefore, we cannot use these observations to definitely argue for methylation of Hg by nitrate-reducers/ ANAMMOXers, but the observation of high percentages of total Hg as CH3Hg+ at this end of the plume is strongly suggestive of some connection to N-cycling. If true, then this would add N-cycling to the growing list of microbial metabolisms that are implicated in environmentally significant Hg methylation (e.g., ref 67), although not yet observed in culture. In summary, being generally organic carbon-limited, wastewater loading to the aquifer drives changes in aquifer biogeochemistry, which in turn appears to cause shifts in Hg speciation to the less strongly sorbing CH3Hg+ and Hg0. Inputs of wastewater-derived dissolved organic compounds with strong Hg(II)-binding properties and, possibly, changes in Hg(II) sorption properties of sediments driven by wastewater inputs might also have contributed to the overall lower Kd values observed. Immediately under the disposal beds, the very high concentrations of Hg resulted in sorption and immobilization of much of the wastewater-associated Hg, while further down-gradient, the trends in speciation and sorption have resulted in net mobilization of naturally occurring

42 billion liters of wastewater were disposed of at the site, and therefore contributing a total of about 3.1 mols of total Hg to the unsaturated zone beneath the beds. The concentration of Hg in the effluent is not particularly high when compared to rainwater,3 but the loading of wastewater to the beds (which occupy about 4.6 ha) increased total water loading by about 10× over typical precipitation at the site. Thus, the high Hg concentration directly below the beds is largely due to focused discharge of wastewater over a confined area. Our data suggest that this ongoing source of Hg to the aquifer is rapidly attenuated during transport down-gradient. For example, the concentration of dissolved total Hg drops from a maximum of ∼750 pM at the top of the aquifer under the beds to ∼20 pM at the very next down-gradient MLS sampled, 140 m away (F343). As the horizontal advection rate has previously been estimated to be about 200 m y−1, this suggests that the lifetime (e-folding) of Hg in the aquifer with respect to sediment scavenging is approximately 70 days. This short residence time in the aquifer suggests that relatively little of the dissolved Hg observed in the anoxic plume is directly derived from the wastewater input, but rather is Hg mobilized from the aquifer sediments by complexation with organic compounds (as noted above), as well as a shift in speciation away from Hg2+ and into a form which is less prone to solid sorption (i.e., Hg0 and CH3Hg+). This conclusion can be tested by comparing the total load of Hg estimated above (3.1 mols) to the amount of Hg within the plume which appears anomalously high when compared to the concentrations in sediments directly above and below the plume or the reference site at F605 (perhaps 2−5 pM higher in solution-phase in the northern end of the plume; 1−3 pM for the southern end of the plume). The range of Kd values estimated for the aquifer predict that 3.1 mols of added Hg could explain these elevated concentrations for a maximum of 500 m down-gradient from the beds (with an estimated plume dimensions of 10-m thick and 500-m wide). This is an upper estimate of the distance down-gradient that wastewater-associated Hg has penetrated as much of the added Hg likely remains in the vadose zone underneath the beds. Therefore, the enhanced dissolved Hg not far from the beds and farther down-gradient is likely material liberated from the aquifer sediments due to the altered biogeochemistry. Mercury Speciation Changes and Biogeochemistry. The dramatic shifts in Hg speciation within the wastewater plume suggest that different microbial communities, performing different anaerobic metabolisms are the primary control on Hg cycling in the aquifer. The correlation between Hg0 and Fe(II) suggests that Fe-reducing bacteria drive a net reduction of Hg(II). Iron(II) is a strong enough reducing agent to convert Hg2+ to Hg0. Rates of Hg(II) reduction to Hg0 by magnetite,54 and aqueous Fe(II) or Fe(II) onto goethite are fast enough to account for the increases in Hg0 observed in the plume.55,56 However, a number of dissimilatory Fe reducing bacteria have been shown to reduce Hg(II) directly as well.57,58 Thus, Hg(II) reduction could be caused by the metabolism of a dissimilatory iron reducing microbe directly or indirectly by some metabolic byproduct (such as dissolved Fe(II) or magnetite). It is also notable that in the Fe-reducing zone of the plume, Hg methylation is not observed and indeed net demethylation is implied by the lower concentration of methylated Hg in the core of the plume as compared to above and below. This is in contrast to recent work demonstrating that some (but not all) Fe-reducing δ-proteobacteria can be Hg methylators (e.g., 13246

dx.doi.org/10.1021/es402441d | Environ. Sci. Technol. 2013, 47, 13239−13249

Environmental Science & Technology

Article

(6) Krabbenhoft, D. P.; Benoit, J. M.; Babiarz, C. L.; Hurley, J. P.; Andren, A. W. Mercury cycling in the Allequash Creek Watershed, Northern Wisconsin. Water Air Soil Pollut. 1995, 80 (1−4), 425−433. (7) Babiarz, C. L.; Hurley, J. P.; Krabbenhoft, D. P.; Gilmour, C.; Branfireun, B. A. Application of ultrafiltration and stable isotopic amendments to field studies of mercury partitioning to filterable carbon in lake water and overland runoff. Sci. Total Environ. 2003, 304 (1−3), 295−303. (8) Bradley, P. M.; Journey, C. A.; Lowery, M. A.; Brigham, M. E.; Burns, D. A.; Button, D. T.; Chapelle, F. H.; Lutz, M. A.; MarvinDiPasquale, M. C.; Riva-Murray, K. Shallow groundwater mercury supply in a coastal plain stream. Environ. Sci. Technol. 2012, 46 (14), 7503−7511. (9) Bone, S. E.; Charette, M. A.; Lamborg, C. H.; Gonneea, M. E. Has submarine groundwater discharge been overlooked as a source of mercury to coastal waters? Environ. Sci. Technol. 2007, 41, 3090−3095. (10) Valiela, I.; Teal, J. M.; Volkmann, S.; Shafer, D.; Carpenter, E. J. Nutrient and particulate fluxes in a salt-marsh ecosystemTidal exchanges and inputs by precipitation and groundwater. Limnol. Oceanogr. 1978, 23 (4), 798−812. (11) Barringer, J. L.; Szabo, Z.; Schneider, D.; Atkinson, W. D.; Gallagher, R. A. Mercury in ground water, septage, leach-field effluent, and soils in residential areas, New Jersey coastal plain. Sci. Total Environ. 2006, 361 (1−3), 144−162. (12) Black, F. J.; Paytan, A.; Knee, K. L.; De Sieyes, N. R.; Ganguli, P. M.; Gary, E.; Flegal, A. R. Submarine groundwater discharge of total mercury and monomethylmercury to central California coastal waters. Environ. Sci. Technol. 2009, 43 (15), 5652−5659. (13) Lamborg, C. H.; Fitzgerald, W. F.; Skoog, A.; Visscher, P. T. The abundance and source of mercury-binding organic ligands in Long Island Sound. Mar. Chem. 2004, 90, 151−163. (14) Repert, D. A.; Barber, L. B.; Hess, K. M.; Keefe, S. H.; Kent, D. B.; LeBlanc, D.; Smith, R. L. Long-term natural attenuation of carbon and nitrogen within a groundwater plume after removal of the treated wastewater source. Environ. Sci. Technol. 2006, 40, 1154−1162. (15) Kent, D. B.; Fox, P. M. The influence of groundwater chemistry on arsenic concentrations and speciation in a quartz sand and gravel aquifer. Geochem. Trans. 2004, 5 (1), 1−12. (16) Kent, D. B.; Wilkie, J. A.; Curtis, G. P.; Davis, J. A. Modeling the influence of variable pH on zinc contamination in a quartzsand aquifer. Geochim. Cosmochim. Acta 2002, 66 (15A), A394−A394. (17) LeBlanc, D. Sewage Plume in a Sand and Gravel Aquifer, Cape Cod, Massachusetts. U.S.G.S. Water-Supply Paper 2218; U.S. Department of the Interior: Washington, DC, 1984; pp 1−28. (18) BarbaroJ. R.WalterD. A.LeblancD. R.Transport of Nitrogen in a Treated-Wastewater Plume to Coastal Discharge Areas, Ashumet Valley, Cape Cod, Massachusetts. USGS Scientific Investigations Report 2013− 5061; USGS: Massachusetts, 2013; p 37. (19) Kent, D. B.; Davis, J. A.; Anderson, L. C. D.; Rea, B. A.; Waite, T. D. Transport of chromium and selenium in the suboxic zone of a shallow aquiferInfluence of redox and adsorption reactions. Water Resour. Res. 1994, 30 (4), 1099−1114. (20) Abrams, R. H.; Loague, K.; Kent, D. B. Development and testing of a compartmentalized reaction network model for redox zones in contaminated aquifers. Water Resour. Res. 1998, 34 (6), 1531−1541. (21) Smith, R. L.; Howes, B. L.; Duff, J. H. Denitrification in nitratecontaminated groundwater - occurrence in steep vertical geochemical gradients. Geochim. Cosmochim. Acta 1991, 55 (7), 1815−1825. (22) Höhn, R.; Isenbeck-Schroter, M.; Kent, D. B.; Davis, J. A.; Jakobsen, R.; Jann, S.; Niedan, V.; Scholz, C.; Stadler, S.; Tretner, A. Tracer test with As(V) under variable redox conditions controlling arsenic transport in the presence of elevated ferrous iron concentrations. J. Contam. Hydrol. 2006, 88 (1−2), 36−54. (23) LeBlanc, D. R.; Hess, K. M.; Kent, D. B.; Smith, R. L.; Barber, L. B.; Stollenwerk, K. G.; Campo, K. W. Natural Restoration of a Sewage Plume in a Sand and Gravel Aquifer, Cape Cod, Massachusetts; Water Resources Investigation Report 99−4018C; 1999; pp 245−259. (24) Leblanc, D. R.; Garabedian, S. P.; Hess, K. M.; Gelhar, L. W.; Quadri, R. D.; Stollenwerk, K. G.; Wood, W. W. Large-scale natural

Hg off the aquifer sediments. It is this net mobilization which we hypothesize is the ultimate cause of elevated Hg inputs to the nearby coastal zone as documented by Bone9 since inground wastewater disposal is prevalent on Cape Cod. This conceptual model is consistent with the hypothesized mobilization of Hg by septic discharges in the KirkwoodCohassey aquifer11 and suggests that elevated Hg concentrations in groundwater may be commonplace, particularly where aquifer sediments are depleted in organic carbon. Furthermore, groundwater anoxia caused by nutrient additions is a common occurrence in many locations such as agricultural fields and these situations should be examined for their potential for Hg mobilization.

■

ASSOCIATED CONTENT

S Supporting Information *

The full geochemical data presented in this manuscript (Table S1) and concentration of Hg in sediments from near the MMR (Table S2). This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: 508 289 2556; e-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by Woods Hole Sea Grant, the Woods Hole Oceanographic Coastal Ocean Institute, and the USGS through the Toxic Substances Hydrology, National Water Quality Assessment, and Hydrologic Research Programs. We thank Judy McDowell and Chris Reddy for their support. Thanks to Mike Bothner for use of his DMA-80. We thank Aria Amirbahman, David Bessinger, Matthew Charette, Richard L. Smith, and John A. Colman for their helpful discussions, Christopher Conaway for a presubmission review, and Karen Johannesson and two anonymous reviewers for their insightful comments.

■

REFERENCES

(1) Swain, E. B.; Engstrom, D. R.; Brigham, M. E.; Henning, T. A.; Brezonik, P. L. Increasing rates of atmospheric mercury deposition in midcontinental North America. Science 1992, 257, 784−787. (2) Harris, R. C.; Rudd, J. W. M.; Amyot, M.; Babiarz, C. L.; Beaty, K. G.; Blanchfield, P. J.; Bodaly, R. A.; Branfireun, B. A.; Gilmour, C. C.; Graydon, J. A.; Heyes, A.; Hintelmann, H.; Hurley, J. P.; Kelly, C. A.; Krabbenhoft, D. P.; Lindberg, S. E.; Mason, R. P.; Paterson, M. J.; Podemski, C. L.; Robinson, A.; Sandilands, K. A.; Southworth, G. R.; Louis, V. L. S.; Tate, M. T. Whole-ecosystem study shows rapid fishmercury response to changes in mercury deposition. Proc. Natl. Acad. Sci. U.S.A. 2007, 104 (42), 16586−16591. (3) Fitzgerald, W. F.; Lamborg, C. H. Geochemistry of Mercury in the Environment. In Treatise on Geochemistry; Holland, H. D., Turekian, K. K., Eds.; Pergamon: Oxford, 2003; Vol. 9, Chapter 4, pp 1−47. (4) Brigham, M. E.; Wentz, D. A.; Aiken, G. R.; Krabbenhoft, D. P. Mercury cycling in stream ecosystems. 1. Water column chemistry and transport. Environ. Sci. Technol. 2009, 43 (8), 2720−2725. (5) Branfireun, B. A.; Krabbenhoft, D. P.; Hintelmann, H.; Hunt, R. J.; Hurley, J. P.; Rudd, J. W. M. Speciation and transport of newly deposited mercury in a boreal forest wetland: A stable mercury isotope approach. Water Resour. Res. 2005, 41, (6), Art. No. W06016. 13247

dx.doi.org/10.1021/es402441d | Environ. Sci. Technol. 2013, 47, 13239−13249

Environmental Science & Technology

Article

gradient tracer test in sand and gravel, Cape Cod, Massachusetts 1. Experimental design and observed tracer movement. Water Resour. Res. 1991, 27 (5), 895−910. (25) Charette, M. A.; Sholkovitz, E. R., Oxidative precipitation of groundwater-derived ferrous iron in the subterranean estuary of a coastal bay. Geophys. Res. Lett. 2002, 29, (10). (26) Coston, J. A.; Fuller, C. C.; Davis, J. A. Pb2+ and Zn2+ adsorption by a natural aluminum-bearing and iron-bearing surface coating on an aquifer sand. Geochim. Cosmochim. Acta 1995, 59 (17), 3535−3547. (27) Zhang, S. L.; Kent, D. B.; Elbert, D. C.; Shi, Z.; Davis, J. A.; Veblen, D. R. Mineralogy, morphology, and textural relationships in coatings on quartz grains in sediments in a quartz-sand aquifer. J. Contam. Hydrol. 2011, 124 (1−4), 57−67. (28) Hammerschmidt, C. R.; Bowman, K. L.; Tabatchnick, M. D.; Lamborg, C. H. Storage bottle material and cleaning for determination of total mercury in seawater. Limnol. Oceanogr.-Methods 2011, 9, 426− 431. (29) Lamborg, C. H.; Hammerschmidt, C. R.; Gill, G. A.; Mason, R. P.; Gichuki, S. An intercomparison of procedures for the determination of total mercury in seawater and recommendations regarding mercury speciation during GEOTRACES cruises. Limnol. Oceanogr.-Methods 2012, 10, 90−100. (30) Fitzgerald, W. F.; Gill, G. A. Subnanogram determination of mercury by two-stage gold amalgamation applied to atmospheric analysis. Anal. Chem. 1979, 51, 1714−1720. (31) Gill, G. A.; Fitzgerald, W. F. Picomolar mercury measurements in sea water and other materials using stannous chloride reduction and two-stage gold amalgamation with gas phase detection. Mar. Chem. 1987, 20, 227−243. (32) Bloom, N.; Fitzgerald, W. F. Determination of volatile mercury species at the picogram level by low-temperature gas chromatography with cold-vapor atomic fluorescence detection. Anal. Chim. Acta 1988, 208, 151−161. (33) Bowman, K.; Hammerschmidt, C. R. Extraction of monomethylmercury from seawater for low-femtomolar determination. Limnol. Oceanogr.-Methods 2011, 9, 121−128. (34) Tseng, C. M.; Lamborg, C.; Fitzgerald, W. F.; Engstrom, D. R. Cycling of dissolved elemental mercury in Arctic Alaskan lakes. Geochim. Cosmochim. Acta 2004, 68 (6), 1173−1184. (35) Mason, R. P.; Fitzgerald, W. F. The distribution and biogeochemical cycling of mercury in the Equatorial Pacific Ocean. Deep Sea Res. Part I: Oceanogr. Res. Pap. 1993, 40 (9), 1897−1924. (36) USEPA. Method 1631, Revision E: Mercury in Water by Oxidation, Purge and Trap, and Cold Vapor Atomic Fluorescence Spectrometry; EPA-821-R-02-019; U.S. EPA Office of Science and Technology: Washington, D.C., 2002. (37) Zapico, M. M.; Vales, S.; Cherry, J. A. A wireline piston core barrel for sampling cohesionless sand and gravel below the water-table. Ground Water Monit. Remediat. 1987, 7 (3), 74−82. (38) USEPA. Method 7473, Mercury in solids and solutions by thermal decomposition, amalgamation, and atomic absorption spectrophotometry. Revision 0.; U.S. EPA Office of Science and Technology: Washington, D.C., 1998. (39) Savoie, J. G.; Leblanc, D. R.; Fairchild, G. M.; Smith, R. L.; Kent, D. B.; Barber, L. B.; Repert, D. A.; Hart, C. P.; Keefe, S. H.; Parsons, L. A. Groundwater-Quality Data for a Treated-Wastewater Plume near the Massachusetts Military Reservation, Ashumet Valley, Cape Cod, Massachusetts, 2006−2008; 2012, p 11. (40) Savoie, J. G.; Kent, D. B.; Smith, R. L.; Leblanc, D. R.; Hubble, D. W.Changes in Groundwater Quality near Two Granular-Iron Permeable Reactive Barriers in a Sand and Gravel Aquifer, Cape Cod, Massachusetts, 1997−2000, 2004. (41) Lee, R. W. Effects of carbon dioxide variations in the unsaturated zone on water chemistry in a glacial-outwash aquifer. Appl. Geochem. 1997, 12 (4), 347−366. (42) Kroeger, K. D.; Cole, M. L.; York, J. K.; Valiela, I. Nitrogen loads to estuaries from waste water plumes: Modeling and isotopic approaches. Ground Water 2006, 44 (2), 188−200.

(43) Bohlke, J. K.; Smith, R. L.; Miller, D. N. Ammonium transport and reaction in contaminated groundwater: Application of isotope tracers and isotope fractionation studies. Water Resour. Res. 2006, 42, (5). (44) Krabbenhoft, D. P.; Babiarz, C. L. The role of groundwater transport in aquatic mercury cycling. Water Resour. Res. 1992, 28 (12), 3119−3128. (45) Johannesson, K. H.; Neumann, K. Geochemical cycling of mercury in a deep, confined aquifer: Insights from biogeochemical reactive transport modeling. Geochim. Cosmochim. Acta 2013, 106 (0), 25−43. (46) Hammerschmidt, C. R.; Fitzgerald, W. F.; Lamborg, C. H.; Balcom, P. H.; Visscher, P. T. Biogeochemistry of methylmercury in sediments of Long Island Sound. Mar. Chem. 2004, 90 (1−4), 31−52. (47) Rudnick, R.; Gao, S. The role of lower crustal recycling in continent formation. Geochim. Cosmochim. Acta 2003, 67 (18), A403− A403. (48) Kent, D. B.; Abrams, R. H.; Davis, J. A.; Coston, J. A.; Leblanc, D. R. Modeling the influence of variable pH on the transport of zinc in a contaminated aquifer using semiempirical surface complexation models. Water Resour. Res. 2000, 36 (12), 3411−3425. (49) Hsu, H.; Sedlak, D. L. Strong Hg(II) complexation in municipal wastewater effluent and surface waters. Environ. Sci. Technol. 2003, 37 (12), 2743−2749. (50) Barber, L. B.; Thurman, E. M.; Runnells, D. D. Geochemical heterogeneity in a sand and gravel aquiferEffect of sediment mineralogy and particle-size on the sorption of chlorobenzenes. J. Contam. Hydrol. 1992, 9 (1−2), 35−54. (51) Graydon, J. A.; Louis, V. L. S.; Hintelmann, H.; Lindberg, S. E.; Sandilands, K. A.; Rudd, J. W. M.; Kelly, C. A.; Hall, B. D.; Mowat, L. D. Long-term wet and dry deposition of total and methyl mercury in the remote boreal ecoregion of Canada. Environ. Sci. Technol. 2008, 42 (22), 8345−8351. (52) Engle, M. A.; Tate, M. T.; Krabbenhoft, D. P.; Schauer, J. J.; Kolker, A.; Shanley, J. B.; Bothner, M. H., Comparison of atmospheric mercury speciation and deposition at nine sites across central and eastern North America. J. Geophys. Res.-Atmos. 2010, 115. (53) Balcom, P. H.; Fitzgerald, W. F.; Vandal, G. M.; Lamborg, C. H.; Rolfhus, K. R.; Langer, C. S.; Hammerschmidt, C. R. Mercury sources and cycling in the Connecticut River and Long Island Sound. Mar. Chem. 2004, 90 (1−4), 53−74. (54) Wiatrowski, H. A.; Das, S.; Kukkadapu, R.; Ilton, E. S.; Barkay, T.; Yee, N. Reduction of Hg(II) to Hg(O) by magnetite. Environ. Sci. Technol. 2009, 43 (14), 5307−5313. (55) Charlet, L.; Bosbach, D.; Peretyashko, T. Natural attenuation of TCE, As, Hg linked to the heterogeneous oxidation of Fe(II): An AFM study. Chem. Geol. 2002, 190 (1−4), 303−319. (56) Amirbahman, A.; Kent, D. B.; Curtis, G. P.; Marvin-DiPasquale, M. C. Kinetics of homogeneous and surface-catalyzed mercury(II) reduction by iron(II). Environ. Sci. Technol. 2013, 47 (13), 7204−7213. (57) Wiatrowski, H. A.; Ward, P. M.; Barkay, T. Novel reduction of mercury(II) by mercury-sensitive dissimilatory metal reducing bacteria. Environ. Sci. Technol. 2006, 40 (21), 6690−6696. (58) Kritee, K.; Blum, J. D.; Barkay, T. Mercury stable isotope fractionation during reduction of Hg(II) by different microbial pathways. Environ. Sci. Technol. 2008, 42 (24), 9171−9177. (59) Kerin, E. J.; Gilmour, C. C.; Roden, E.; Suzuki, M. T.; Coates, J. D.; Mason, R. P. Mercury methylation by dissimilatory iron-reducing bacteria. Appl. Environ. Microbiol. 2006, 72 (12), 7919−7921. (60) Fleming, E. J.; Mack, E. E.; Green, P. G.; Nelson, D. C. Mercury methylation from unexpected sources: molybdate-inhibited freshwater sediments and an iron-reducing bacterium. Appl. Environ. Microbiol. 2006, 72 (1), 457−464. (61) Barringer, J. L.; Szabo, Z.; Reilly, P. A. Occurrence and Mobility of Mercury in Groundwater. In Current Perspectives in Contaminant Hydrology and Water Resources Sustainability; Bradley, P. M., Ed.; InTech: New York, 2013; pp 117−149. (62) Todorova, S. G.; Driscoll, C. T.; Matthews, D. A.; Effler, S. W.; Hines, M. E.; Henry, E. A. Evidence for regulation of monomethyl 13248

dx.doi.org/10.1021/es402441d | Environ. Sci. Technol. 2013, 47, 13239−13249

Environmental Science & Technology

Article

mercury by nitrate in a seasonally stratified, eutrophic lake. Environ. Sci. Technol. 2009, 43 (17), 6572−6578. (63) Kritee, K.; Barkay, T.; Blum, J. D. Mass dependent stable isotope fractionation of mercury during mer mediated microbial degradation of monomethylmercury. Geochim. Cosmochim. Acta 2009, 73 (5), 1285−1296. (64) Brunet, R. C.; GarciaGil, L. J. Sulfide-induced dissimilatory nitrate reduction to ammonia in anaerobic freshwater sediments. FEMS Microbiol. Ecol. 1996, 21 (2), 131−138. (65) Cardoso, R. B.; Sierra-Alvarez, R.; Rowlette, P.; Flores, E. R.; Gomez, J.; Field, J. A. Sulfide oxidation under chemolithoautotrophic denitrifying conditions. Biotechnol. Bioeng. 2006, 95 (6), 1148−1157. (66) Shapiro, S. D.; LeBlanc, D.; Schlosser, P.; Ludin, A. Characterizing a sewage plume using the H-3-He-3 dating technique. Ground Water 1999, 37 (6), 861−878. (67) Hamelin, S.; Amyot, M.; Barkay, T.; Wang, Y.; Planas, D. Methanogens: Principal methylators of mercury in Lake Periphyton. Environ. Sci. Technol. 2011, 45 (18), 7693−7700.

13249

dx.doi.org/10.1021/es402441d | Environ. Sci. Technol. 2013, 47, 13239−13249