Recent changes in mercury deposition and primary productivity inferred from sediments of lakes from the Hudson Bay Lowlands, Ontario, Canada

Environmental Pollution 173 (2013) 52e60

Contents lists available at SciVerse ScienceDirect

Environmental Pollution journal homepage: www.elsevier.com/locate/envpol

Recent changes in mercury deposition and primary productivity inferred from sediments of lakes from the Hudson Bay Lowlands, Ontario, Canada Michelle L. Brazeau a, Alexandre J. Poulain a, Andrew M. Paterson b, Wendel (Bill) Keller c, Hamed Sanei d, Jules M. Blais a, * a

Department of Biology, University of Ottawa, 30 Marie-Curie, Ottawa, Ontario K1N 6N5, Canada Dorset Environmental Science Centre, Ontario Ministry of the Environment, 1026 Bellwood Acres Road, P.O. Box 39, Dorset, Ontario P0A 1E0, Canada Cooperative Freshwater Ecology Unit, Biology Department, Laurentian University, 935 Ramsey Lake Rd., Sudbury, Ontario P3E 2C6, Canada d Geological Survey of Canada, 3303 33rd St., Calgary, Alberta T2L 2A7, Canada b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 30 April 2012 Received in revised form 14 September 2012 Accepted 28 September 2012

Spatial and temporal changes in mercury (Hg) concentrations and organic carbon in lake sediments were examined from the Hudson Bay Lowlands to investigate whether Hg deposition to sediments is related to indicators of autochthonous production. Total organic carbon, “S2” carbon (mainly algalderived OC), C:N and v13C indicators suggest an increase in autochthonous productivity in recent decades. Up-core profiles of S2 concentrations and fluxes were significantly correlated with Hg suggesting that varying algal matter scavenging of Hg from the water column may play an important role in the temporal profiles of Hg throughout the sediment cores. Absence of significant relationship between total Hg and methyl Hg (MeHg) in surficial sediments suggested that inorganic Hg supply does not limit MeHg production. MeHg and OC were highly correlated across lakes in surface and deep sediment layers, indicating that sediment organic matter content explains part of the spatial variation in MeHg concentrations between lakes. Ó 2012 Elsevier Ltd. All rights reserved.

Keywords: Mercury Organic carbon Sediment Hudson Bay Lowlands Rock-Eval

1. Introduction Mercury (Hg) is a ubiquitous contaminant that is emitted through natural and anthropogenic processes, and is identified as a priority chemical by the United Nations Environment Programme (U.N.E.P., 2008). It has a very complex biogeochemical cycle because it is readily transformed between oxidation states through biotic and abiotic reactions, warranting the need to consider it separately from other heavy metals (Morel et al., 1998). Typically, the volatile form, Hg0, has an atmospheric residence time of 1e2 years allowing for a global distribution and the contamination of areas far removed from point sources of Hg (Lamborg et al., 2002). Hg is of particular concern in Arctic and subarctic ecosystems where individuals are exposed to dangerous levels of methylmercury (the potent, bioaccumulative neurotoxic form of Hg), through the consumption of traditional foods (Macdonald and Bewers, 1996). A review of Hg trends in the Arctic demonstrated that Hg bioaccumulation in the Canadian and Greenland regions is significantly increasing when compared to the North Atlantic region

* Corresponding author. E-mail address: [email protected] (J.M. Blais). 0269-7491/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.envpol.2012.09.017

(Riget et al., 2011). Numerous studies have focused on Hg in the Canadian Archipelago and the Canadian Western Arctic, but the Eastern Arctic and Subarctic remain comparatively understudied (Carrie et al., 2010; Kirk et al., 2011). Northern Ontario represents a unique eastern subarctic ecosystem with particular characteristics due to the influence of Hudson Bay (Rouse, 1991). The Hudson Bay Lowlands (HBL) contain countless lakes and rivers that are showing signs of significant ecological changes in recent decades. For example, Gunn and Snucins (2010) reported die-offs of anadromous brook charr (Salvelinus fontinalis) and white suckers (Catostomus commersoni) in the Sutton River, which they attributed to unseasonably warm air temperatures and increased water temperatures during the summer of 2001. Increases in primary productivity in northern lakes present a new challenge for the study of Hg in these systems (Grant et al., 2011). Atmospheric deposition through oxidization of volatile elemental mercury (Hg0) is presumed to be the only source of Hg to environments removed from industrial areas. Hg concentrations in lake sediments are therefore used as historical archives to elucidate atmospheric mercury concentrations for regions where modern monitoring techniques are unavailable. However, as proposed by Outridge et al. (2007), increases in phytoplankton biomass may increase scavenging of Hg from the water column, and accelerate its

M.L. Brazeau et al. / Environmental Pollution 173 (2013) 52e60

53

Fig. 1. Map showing the location of 17 study lakes in the Hudson Bay Lowlands, sampled in July and August of 2009, and August of 2010 and 2011. Sediment cores were obtained from 13 of these lakes (Table 1). The solid line represents the boundary of continuous and discontinuous permafrost.

deposition to the sediment. As a consequence, sediment concentrations of Hg may not accurately reflect mercury delivery via the atmosphere. The authors of this hypothesis argue that Hg concentrations in Arctic and subarctic lakes are steadily increasing despite stable or even declining Hg concentrations in air. In contrast, lakes in temperate areas, where warming impacts are less pronounced, do not display this increase in sediment Hg concentrations. The objective of this study was to determine the history of Hg deposition to lakes of the Hudson Bay Lowlands in Ontario, and to determine the relationship between sedimentary concentrations of mercury and organic matter using a temporal and spatial approach. We also used organic carbon, “S2” carbon (algal-derived carbon), carbon to nitrogen ratios (C:N) and v13C values to track changes in carbon sources (allochthonous versus autochthonous) in lake sediments. Finally, we examined the algal-mercury scavenging hypothesis using Rock-EvalÒ pyrolysis analyses by characterizing the type of organic matter found in the sediments of two of the study lakes.

vegetation includes: tamarack (Larix laricina), Sphagnum, and other mosses and sedges. Chemically, the study lakes are alkaline, moderately tea-stained, and range in trophic status from oligo- to meso-trophic (Table 1). The lakes are either very shallow (less than 3 m) or moderately deep (>10 m), with Aquatuk, Hawley, and North Raft lakes (i.e., the lakes from which we examined full temporal records) falling into the latter category. All of the sites were found to be generally well mixed at the time of sampling, although Hawley Lake showed a 6e7
C temperature difference from top to bottom, and has been shown to stratify in mid-summer in previous years (Gunn and Snucins, 2010). Mean annual temperature in Churchill, MB, on the western shore of Hudson Bay, was 7.2
C between 1960 and 1990. However, a significant increase in temperature Table 1 Average values of key water chemistry variables collected in July or August of 2009, and August of 2010 and/or 2011. Water samples were analysed at the Ontario Ministry of the Environment’s Dorset Environmental Science Centre using standard protocols. The parameters analysed include: Gran alkalinity, calcium, pH, dissolved organic carbon and total phosphorus concentrations. Also shown are the years of sampling, and the measured depth at the sampling locations (which may or may not be equivalent to the maximum depth of the lake). Lake

2. Materials and methods 2.1. Study region The Hudson Bay Lowlands (HBL) area includes a continuous belt of land with an elevation of less than 200 m surrounding the western side of Hudson Bay and James Bay. The area is mostly flat, very wet, with countless lakes and ponds, and is traversed by the southern boundaries of continuous and discontinuous permafrost as well as the Arctic tree line (Rouse et al., 1997). The study lakes are located in a transect that extends from the southern shore of Hudson Bay to w150 km inland. The boundary of continuous to discontinuous permafrost runs approximately eastwest through the study region (Fig. 1), which is covered with a mantle of sedimentary glacial and marine sediments. Exceptions are outcroppings of Precambrian origin (i.e., the Sutton Ridge) which may protrude as high as 140 m above the surrounding lowland. The region is dominated by vast peatlands (bogs and fens), although black spruce (Picea mariana) are common along lake margins, riparian zones, and gravel beach ridges that run parallel to Hudson Bay. Other common

pH DOC TP Years Gran al Ca (mg L1) (mg L1) sampled (mg L1 (mg L1) as CaCO3)

Aquatuk 98.7 Billbear 71.5 Cassie 90.3 Hawley 105.3 Julison 90.7 Kinushseo 38.9 N. Raft 80.0 Oppinagau E. 52.4 Raft 76.4 Sam 82.4 Spruce 63.4 Stuart 79.3 Wolfgang 70.1

32.4 27.1 33.6 35.1 32.6 13.3 28.9 17.4 25.8 27.0 23.0 30.4 24.6

8.1 7.6 7.8 9.0 8.1 10.4 8.1 6.3 8.0 7.5 7.7 7.0 7.9 6.8 7.8 8.8 7.9 10.4 8.2 6.7 7.9 7.9 8.0 6.3 7.9 9.1

13.8 28.8 4.4 6.8 14.8 9.1 6.4 16.4 7.0 8.0 10.8 19.6 15.2

2009e11 2010e11 2010e11 2009e11 2010 2009 2009e11 2009e10 2009 2010e11 2009e11 2010 2010e11

Depth at sampling location (m) 12.1 0.8 0.9 35.2 0.9 2.5 10.5 1.7 1.3 1.5 14.0 0.7 1.1

54

M.L. Brazeau et al. / Environmental Pollution 173 (2013) 52e60

has been observed in the northern HBL since the early 1990s, with mean annual temperature increasing at a rate > 0.5e1.0
C per decade. 2.2. Sample collection Thirteen lakes were sampled in the HBL of Ontario, Canada, from 2009 to 2011, which range in longitude and latitude from 53 to 55
N and 84 to 85
W, respectively. The experimental design was devised to include a spatial and a temporal aspect. Sediment cores were taken using a MiniGlew gravity corer (Glew, 1991) and plexiglass coring tubes. Complete cores were taken from North Raft (NRT), Hawley (HWL) and Aquatuk (AQT) lakes and sectioned into 0.5 cm intervals over the first 10 cm (20 samples) and every 2 cm for the remaining fraction of the core. The surface and bottom 0.5 cm fractions of cores were sampled from 10 other lakes. Samples were placed into Whirl-PakÒ bags, protected from light and kept at 4
C until frozen at 20
C. Coring tubes were scrubbed and rinsed with lake water between sample collections. Surface water for total Hg analyses was collected in triplicate in 100 mL white cap, Teflon lined, acid washed glass bottles and surface water for MeHg analyses was collected in triplicate in 1 L HDPE bottles. All water samples were spiked with 0.5% ultra pure trace metal grade hydrochloric acid. All bottles were rinsed three times with lake water before sampling and samples were kept at 4
C and protected from light. 2.3. Radiometric dating Sediment cores were radiometrically dated using gamma (g) spectrometry and analysed for the activity of 210Pb, 137Cs and 226Ra in an Ortec germanium crystal well detector (DSPec, Ortec, model # GWL-120230) following the method by Appleby (2001). Analysis of 210Pb was performed on 14e18 selected depth intervals of the sediment cores to determine the sediment age and the sediment accumulation rate. Samples were lyophilized, homogenized, hermetically sealed and left to reach secular equilibrium for a minimum of 21 days before being counted for 23 h (82,800 s). The resulting spectrum files showed 210Pb activity with a peak at 46.5 keV, and 137Cs at 662 keV. 226Ra activity was determined by g-ray emissions of its daughter isotope 214Pb, resulting in peaks at 295 and 352 keV. Long-term sedimentation rates were determined for each core using methods described by Appleby and Oldfield (1978). The samples from lakes of which only surface and deep sediments were collected could not be completely radiometrically dated however, 210Pb and 214Bi activities of these sediments allowed us to estimate the age of bottom sediments using the equation tx ¼ k1 LnðC0 $Cx1 Þ where k is the 210Pb decay constant (0.03114 y1), C0 is the unsupported 210Pb activity at surface, and Cx is the unsupported 210Pb activity at the bottom (Appleby, 2001; details in Table S1, Supplemental Material). We note that the estimated age of the bottom sediments is 90  32 years (SD), which is well representative of pre-industrial conditions as they predate the majority of Hg emissions in the 20th Century (Mills et al., 2009). Hence we defined deep sediments (10e30 cm) as pre-industrial and surface sediment as recently deposited. 2.4. Mercury analyses 2.4.1. Total mercury Water samples were preserved with bromium monochloride. Total mercury was analysed in water samples using dual gold trap pre-concentration and cold vapour atomic fluorescence spectroscopy (CV-AFS). The analysis was conducted using a Tekran 2600 system control module equipped with a Tekran 2610 liquid handling module and Tekran model 2620 auto-sampler, following the modified U.S. EPA Method 1631 (EPA, 1996). The instrument was calibrated with a stock standard of mercuric chloride (HgCl2) for a calibration curve with a R2 value of 0.999. Initial precision and recovery (IPR) and ongoing precision and recovery (OPR) were 105  1.2% and 104  4.6%, respectively. Field and travel blanks had a concentration of 0.06  0.03 ng L1. Frozen sediments were transferred from Whirl-PakÒ bags into 50 mL sterile high density polypropylene FalconÒ tubes. Samples were lyophilized for a period of 72 h under a vacuum of 5 atm and homogenized. Total Hg in sediments was analysed with Nippon Instruments Corporation’s Mercury SP-3D Analyzer (CV-AAS) by thermal decomposition with gold trap amalgamation and cold vapour atomic absorption method (Fox et al., 2005; LLC, 2004). The instrument was calibrated with Mercury Reference Solution 1000 ppm  1% (Fisher CSM114-100) and MESS-3 (91  9 ng g1, National Research Council of Canada) was used as reference material. Instrumental blanks were performed and were below 0.02 ng. 2.4.2. Methylmercury Organomercury concentrations in both sediment and water samples were determined by capillary gas chromatography coupled with atomic fluorescence spectrometry (GC-AFS) using the Analytical Mercury System Model PSA 10.723, as described by Cai et al. (1996) and Cai et al. (1997) for water and soil, respectively. Recovery of certified reference material ERM580 was between 93 and 109% (n ¼ 8). Field and travel blanks of water samples were 0.001  0.0001 ng L1 and 0.003  2.7  107 ng L1, respectively.

2.5. Organic geochemistry (Rock-EvalÒ 6 pyrolysis) Rock-EvalÒ 6 (Vinci Technologies, France) pyrolysis analyses were performed by the Geological Survey of Canada in Calgary, AB, on selected depth intervals of NRT and HWL lake sediment cores only, due to logistical constraints. This technique utilizes the degree of thermal degradation of various organic compounds to elucidate the quantity and type of organic matter present within the sample. The first step occurs in a pyrolysis oven in which the S1 and S2 fractions are recorded. S1 is geochemically identified as volatile, free hydrocarbons in the sample corresponding to amorphous organic matter, including pigments and oils/lipid products (Lafargue et al., 1998). S2 is classified as a kerogen fraction which consists mainly of hydrogen-rich aliphatic biomacromolecules that form the cell walls of algal matter; it is therefore referred to as algal-derived OC (Carrie et al., 2012; Sanei et al., 2005). In addition to the various organic matter fractions, the analysis also yields a hydrogen index (HI; mg HC g1 TOC) and an oxygen index (OI; mg CO2 g1 TOC), which represent quality and origin of kerogen (autochthonous algal-rich OM versus land-derived, higher plant OM, respectively). 2.6. Carbon and nitrogen isotopes Elemental and isotopic analyses of all samples were performed at the G.G. Hatch Stable Isotope Laboratory at the University of Ottawa, Ottawa, ON. Prior to the analyses, lyophilized sediment samples were re-hydrated with HPLC water and desiccated with HCl to remove carbonates, and then re-lyophilized. Samples and standards were submitted to an elemental analysis (EA) to determine the elemental composition of organic carbon using the CE EA1110 Elemental Analyzer. Samples were subjected to Dumas combustion with the addition of oxygen (Handley et al., 1991). Ultra-pure helium was used to carry the resulting gases through the column of chemicals to obtain N2, CO2, H2O, and SO2, then through a gas chromatograph column to separate the gases which were measured by a thermoconductivity detector. Eager 200 software for Windows was used to control the EA, as well as process the results from the thermoconductivity detector using either the Linear Fit or K-factor method. The analytical precision (2 sigma) for the analyses is 0.1% for N, H and S and 0.3% for C (Pella, 1990). Amounts needed for the isotopic analyses were based on the results of the EA. Sediments were weighed accordingly into tin capsules (0.5e17 mg) with 2 parts tungsten oxide (WO3). Calibrated internal standards were prepared as a reference with every batch of samples. The isotopic composition of organic carbon and nitrogen was determined by the analysis of CO2 and N2, produced by combustion on a VarioEL III Elemental Analyzer (Elementar, Germany) followed by “trap and purge” separation and on-line analysis by continuous-flow with a DeltaPlus XP Plus Advantage Isotope Ratio Mass Spectrometer coupled with a ConFlo II (Thermo, Germany). The data were reported in Delta notation v, the units are per mil (&) and defined as v ¼ ((Rx  Rstd)/Rstd)*1000 where R is the ratio of the abundance of the heavy to the light isotope, x denotes sample and std is an abbreviation for standard. The routine precision of the analyses was 0.20& (Pella, 1990). 2.7. Flux calculations and statistical analyses Mercury fluxes (HgF), S2 fluxes (S2F), anthropogenic fluxes (DHgF, DS2F), flux ratios (FR) and Hg enrichment factors (EF) were calculated as in Muir et al. (2009). 1. Flux (F) (ng or mg m2 y1) ¼ Concentration (ng g1 or mg g1)  210Pb-derived sedimentation rates (g m2 y1) for each core horizon. 2. Anthropogenic flux (DF) (ng or mg m2 y1) ¼ Frecent e Fpre-ind. 3. Flux ratio (FR) ¼ Frecent/Fpre-ind. 4. Enrichment factor (EF) ¼ recent (post-1990)/pre-industrial (pre-1900) concentrations. Sediment particle focusing factors (FF) were estimated for the sediment cores. The core’s cumulative unsupported 210Pb inventory (Bq m2) was multiplied by the 210 Pb decay constant (0.03114 y1). This observed 210Pb flux (Bq m2 y1) was then divided by the predicted atmospheric 210Pb flux of 150 Bq m2 y1 for this latitude (Omelchenko et al., 1995). Statistical analyses such as general linear models and correlations were performed using Systat Software Inc. 2008. The reported p-values passed the normality (ShapiroeWilk) and constant variance tests for all of the data.

3. Results and discussion 3.1. Radioisotopic dating 210 Pb was used to elucidate the sediment core chronology. The profile of the 210Pb and 226Ra activities converge at depth, confirming that supported and unsupported 210Pb horizons were

M.L. Brazeau et al. / Environmental Pollution 173 (2013) 52e60

55

137Cs (Bq kg-1) Cummulative Dry Mass (g cm-2)

0

20

40

60

80 2010

0

0

120 2009

0

160 2009

2

1971

2

1959

2

1945

4

1903

4

1952

4

1895

6

1934

6

1912

8

1855 800

10

6

8

30

60

90

8

Pb

A

Cs

10 100

200

300

40

80

120

B

10 0

0

Year

0

400

0

200

400

600

C 0

200

400

600

Excess 210Pb (Bq kg-1) Fig. 2. Excess 210Pb (Bq Kg1)  SD and 137Cs (Bq Kg1)  SD versus cumulative dry mass (g cm2) and year in sediment cores from Aquatuk (A), Hawley (B) and North Raft (C) lakes.

captured in all sediment cores. Most excess 210Pb and 137Cs was concentrated in the top 2 cm of accumulated dry mass for all three cores suggesting a very low sediment accumulation rate (Fig. 2). As a result, background 210Pb concentrations were found above the depth of 10 cm. In HWL and NRT cores, peak activity of 137Cs (corresponding to the maximum fallout of this anthropogenic radionuclide in 1963 (Pennington et al., 1973)) occurred in the surface sediment suggesting a very low sedimentation rate, though the 137Cs peak at the very surface suggests some postdepositional mobility of 137Cs in sediment, which has been documented elsewhere (e.g. Blais et al., 1995). In the AQT core, peak 137Cs activity was recorded at 4e4.5 cm depth interval. The constant rate of supply model (CRS) was used to calculate dates and sedimentation rates in the three cores. Study lake locations, sampling depths, 210Pb flux, and focus factors are given in Table 2. 3.2. Spatial Hg distribution A wide range of total mercury (Hg) concentrations was recorded amongst the study lakes. For example, the lowest concentrations were observed in SRT lake (core surface: 6.13  0.7 ng g1 dw and at depth: 4.96  0.2 ng g1 dw) and highest in RFT lake (core surface: 125.4  0.9 ng g1 dw and at depth: 91.4  1.7 ng g1 dw) (Table 3, Fig. S1 in Supplemental Data), emphasizing the importance of physical and chemical parameters as well as in situ processes in the fate of atmospherically deposited Hg. Methylmercury (MeHg) concentrations were lowest in SRT lake surface (0.077 ng g1 dw) and SPC deep sediments (0.0202 ng g1 dw) and highest in WGN surface (1.83 ng g1 dw) and RFT deep (0.466 ng g1 dw) sediments (Table 3, Fig. S1 in Supplemental Material). There was no significant correlation between Hg and MeHg in surface sediment (p ¼ 0.54) underscoring how

Hg concentrations can be a poor predictor of MeHg concentrations and its biological importance in lake sediments (Mason and Lawrence, 1999). In surficial lake water, Hg concentrations varied from 0.66 to 2.7 ng L1 and MeHg varied from 0.02 to 0.06 ng L1which is below the established guidelines for the protection of aquatic wildlife of 26 ng L1 and 4 ng L1, respectively (CCME, 2003). However, these guidelines do not address exposure through food or bioaccumulation to higher trophic levels and may not necessarily protect wildlife that consumes aquatic species (CCME, 2003). When considering Hg concentrations in the environment, it is important to consider the proportion of Hg that is in the methylated form; the Hg fraction that is biologically relevant and of most concern. MeHg represented between 0.78 and 5.2% of the total Hg in all water samples which is once again below the guideline of 10% established by the Canadian Council of Ministers of the Environment (Table 3) (CCME, 2003). In surface sediment, 0.17e2.9% of Hg was present as MeHg while 0.078e0.95% of Hg was in a methylated form in deep sediments. MeHg degradation is common in subsurface sediments which is why decreases in MeHg concentrations and MeHg:Hg proportions in deeper sediments were observed (Rydberg et al., 2008). The Hg enrichment factor (EF) represents the increase in Hg concentrations due to anthropogenic emissions of Hg after the onset of the Industrial Revolution. The mean Hg EFs for the 13 lakes was 2.09  0.8 which is consistent with the global average of 3 and suggests that these lakes receive anthropogenic Hg inputs through atmospheric deposition (Yang et al., 2010). Hg is known to be highly correlated with organic matter (OM) in sediments due to its affinity for sulphur, oxygen and nitrogen groups, which results in the formation of HgeOM complexes (Lindberg and Harriss, 1974). However this correlation can be complicated by several factors including dilution by inorganic clastic material flux, heterogeneity

Table 2 Summary of cored lake characteristics. Name

Sampling year

Latitude longitude

Observed 210Pb flux (Bq m2 y1)

Predicted 210Pb flux (Bq m2 y1)

Focusing factor (FF)

Aquatuk (AQT)

2010

54
210 11.800 84
340 29.800 54
310 38.700 84
370 46.500 54
320 4.000 84
450 21.600

117

150

0.782

Hawley (HWL)

2009

North Raft (NRT)

2009

a b

Average sedimentation rate (g m2 y1) (SD)a 21.3 (7.73)

98.1

150

0.654

120.6 (104.9)

49.9

150

0.333

23.5 (9.23)

According to the constant rate of supply model. Note that the average sedimentation rate is inflated by the sedimentation event in 1950e1960.

Depth at sampling location (m) 12.2

b

35.2 10.5

56

M.L. Brazeau et al. / Environmental Pollution 173 (2013) 52e60

Table 3 Total mercury, methylmercury and mercury present as methylmercury (%) in surface water (ng L1) and surface and deep sediment (ng g1 dw) from lakes in the Hudson Bay Lowlands. AQT ¼ Aquatuk; BLB ¼ Billbear; CSE ¼ Cassie; HWL ¼ Hawley; JLN ¼ Julison; KSO ¼ Kinushseo; NRT ¼ North Raft; OPE ¼ Opinnagau East; RFT ¼ Raft; SAM ¼ Sam; SPC ¼ Spruce; SRT ¼ Stuart; WGN ¼ Wolfgang. Lake

AQT ‘09 AQT ‘10 BLB CSE HWL ‘09 JLN KSO NRT OPE RFT SAM SPC SRT WGN

THg ng L1, ng g1 dw, (SD)

MeHg ng L1, ng g1 dw, (SD)

Hg as MeHg (%)

Surface water

Surface sediment

Deep sediment

Surface water

Surface sediment

Deep sediment

Surface water

Surface sediment

Deep sediment

1.31 0.87 2.7 0.95 1.1 0.93 1.3 1.9 1.7 1.7 0.66 1.4 1.2 0.77

104 110 61.9 68.3 98.2 28.8 93.9 120 92.2 125 45.2 56.2 6.1 62.9

63.1 63.2 24.8 20.8 45.2 13.9 54.5 57.7 47.9 91.4 34.3 25.8 4.96 14.6

0.06 0.025 0.021 0.041 0.056 0.02 0.02 0.052 0.04 0.056 0.031 0.038 0.046 0.03

0.212 0.305 0.755 0.718 0.415 0.513 0.747 0.204 1.12 1.76 0.883 0.185 0.0772 1.83

0.0797 0.0745 0.236 0.0246 0.0938 0.0402 0.171 0.13 0.117 0.466 0.099 0.0202 0.0241 0.0205

4.6 2.8 0.78 4.3 5.2 2.2 1.6 2.7 2.4 3.3 4.6 2.8 3.9 3.9

0.2 0.28 1.2 1.1 0.42 1.8 0.8 0.17 1.2 1.4 1.9 0.33 1.3 2.9

0.13 0.12 0.95 0.12 0.21 0.29 0.31 0.23 0.24 0.51 0.29 0.078 0.49 0.14

(0.13) (0.18) (0.18) (0.061) (0.078) (0.056) (0.15) (0.45) (0.33) (0.036) (0.11) (0.094) (0.027) (0.039)

(4.9) (1.6) (0.96) (0.55) (4.5) (0.82) (3.5) (2.7) (0.24) (0.93) (1.4) (4.6) (0.67) (2.1)

(5.8) (2.6) (1.2) (0.65) (1.4) (1.1) (2.0) (3.4) (1.5) (1.7) (0.5) (5.0) (0.19) (2.6)

(0.0063) (0.0048) (0.0042) (0.0024) (0.013) (0.017) (0.02) (0.029) (0.0053) (0.058) (0.0062) (0.0031) (0.012) (0.011)

in organic source, OM degradation, and sulphide draw down (Outridge et al., 2011). Interestingly, no significant correlation was found between Hg and total organic carbon (OC), either in surface, or in deep sediments, and there was no significant relationship between DOC (dissolved organic carbon) and either Hg or MeHg in lake water (p > 0.1). However, there was a significant correlation between MeHg and OC in surface and deep sediments (p ¼ 0.0001 and 0.016, respectively), suggesting that MeHg production in these lakes is greater in organic-rich sediments. Delongchamp et al. (2010) observed higher MeHg concentrations in sediment of the St. Lawrence River where anaerobic OM decomposition and methane emissions were elevated due to methanogenic bacteria which are known to methylate Hg (Avramescu et al., 2011). 3.3. Temporal Hg distribution The Hg depth profiles of the sediment cores from AQT, HWL and NRT lakes were consistent with those reported by others in the Arctic and subarctic (Kirk et al., 2011). Hg concentrations steadily increased towards the surface of all three cores, with the highest concentrations in the most recently deposited sediments (Fig. 3). In AQT lake, the historical [Hg] was 68.8  5.4 ng g1 (dw) (mean  SD) from 10.25 to 22.25 cm depth and the recent [Hg] was 108.6  1.4 ng g1 (dw) yielding an EF of 1.58 (Fig. 3). Historical concentration of Hg in HWL sediments from 22.25 to 24.25 cm was 44.8  0.54 ng g1 (dw) and the recent sediments had a mean concentration of 92.7  4.8 ng g1 (dw) for an enrichment factor of 2.07. NRT lake historical concentrations were 62.3  3.0 ng g1 (dw) between the depths of 9.25e29.75 cm and reached a maximum of 112.5  11 ng g1 (dw) in recent sediments resulting in an EF of 1.80. We calculated the Hg flux (HgF) which accounts for variations in sedimentation rates. The AQT core has a relatively constant sedimentation rate therefore the Hg and HgF profiles were similar (Fig. 3). The sedimentation rate of the NRT core could only be inferred for the top 10 cm intervals due to the rapid loss of excess 210 Pb therefore we could not calculate the HgF, the FR and the DF for the complete core. However, the HgF peaked at 2770 ng m2 y1 at 6.25 cm depth (1938) then decreased to 1090 ng m2 y1 in 2009. Hg in the HWL lake core diminished between 1950 and 1960 which is likely a dilution due to a large input of allochthonous inorganic matter. The cause of this large allochthonous input from the catchment is currently unknown, but may have been an episode of severe erosion due to a forest fire in the catchment (K. Rühland, Personal Communication, 2011). The sedimentation rate increased

drastically from 5 to 10 cm and TOC and water content decreased, likely due to the dilution by inorganic clays and silts from the catchment (Fig. 3). The HgF was highest in the HWL core when [Hg] was at its lowest (17,600 ng m2 y1 and 28.5 ng g1 dw), suggesting that the allochthonous matter that was introduced into the lake contained elevated levels of Hg. However, the HgF was likely inflated by the drastic rise in sedimentation rate at this depth. The HgF in recent sediments of HWL and NRT is decreasing as opposed to AQT where it is increasing (Fig. 3). The anthropogenic Hg flux (DHgF) and the HgFR was 2510 ng m2 y1 and 2.72 in AQT, 620 ng m2 y1 and 1.69 in HWL, and 778 ng m2 y1 and 2.36 in NRT, respectively. Landers et al. (1998) reported DHgFR across Arctic and Boreal ecosystems ranging from 0.7 to 8.9 and attributed industrial activity as the main source of Hg to the lakes. In contrast to the spatial analyses, Hg was significantly correlated with OC in all three sediment cores (p value: AQT ¼ 0.007; HWL and NRT < 0.001). In all three cores, Hg g1 OC is increasing in recent sediments, suggesting an increasing input of Hg bound to OM. This effect can be expected to intensify in some of the study lakes and other northern lakes as the integrity of the underlying permafrost of the catchment is compromised by progressive climate warming (Zhang et al., 2008). MeHg analyses were limited to 5 depth intervals. As anticipated, MeHg concentrations were highest in surface sediment and decreased steadily with depth in all cores most likely due to MeHg degradation (Fig. 3). MeHg was correlated with OC in all three depth profiles (AQT: R2 ¼ 0.824; p ¼ 0.033; HWL: R2 ¼ 0.866, p ¼ 0.022; NRT: R2 ¼ 0.848, p ¼ 0.026) and as a result the MeHg profiles remained similar once corrected for OC. MeHg was also diluted by the input of allochthonous matter in the HWL core at 5 cm depth (1950e1960). MeHg degradation can be biogenic and is either oxidative or reductive, depending on the final C product (CO2 or CH4). The relative importance of each process in freshwater is very difficult to predict without direct experimentation. Most importantly, MeHg is considered unstable within sediment cores and its sediment half life was estimated to be w2 days by Hintelman et al. (2000). It is therefore very difficult to attempt to explain the exact cause of MeHg degradation over depth; we can only speculate it is biological in nature. 3.4. Algal-derived carbon Rock-Eval analyses performed on HWL and NRT sediment cores measured concentrations of S2 fraction, corresponding mainly to algal-derived OC (Carrie et al., 2012; Sanei et al., 2005). The result

M.L. Brazeau et al. / Environmental Pollution 173 (2013) 52e60

HgF (ng m-2 y-1)

Hg (ng g-1, dw) 20

40

60

80

100

120 0

5000

57

Hg (ng g-1 OC)

10000 15000 20000 800

1200

1600

Water (%)

2000

2400

50

60

70

80

90

0

2010

5

1971

10

1903

15 20

A

25

0

2009

5

1959

10

1952

15

1934

20

1912

B

25

Year

Core Depth (cm)

30

1855

30 0

2009

5

1945

10

1895

15 20 25

Hg MeHg

30 0.05 0.15 0.25 0.35 0.45 0.55 0

MeHg (ng g-1, dw)

Hg g-1 OC TOC

HgF Sedimentation rate 200

400

600

800

Sedimentation Rate (g m-2 y-1)

0

2

4

6

8

10

C

12

TOC (%)

Fig. 3. Mercury concentrations (Hg; ng g1, dw), mercury flux (HgF; ng m2 y1), sedimentation rate (g m2 y1), total organic carbon (TOC; %) and water content (%) according to core depth and CRS year in sediment cores from Aquatuk (A), Hawley (B) and North Raft (C) lakes.

showed S2 varies from 2.84 to 12.4 mg HC g1 (dw) with an average of 6.87  2.3 mg HC g1 (dw) in HWL and 9.43e12.1 mg HC g1 (dw) and an average of 10.2  0.62 mg HC g1 (dw) in NRT. The concentrations were consistent with those found in other subarctic lakes (Kirk et al., 2011). S2 rose sharply towards the surface of both cores and there was a significant positive relationship between S2 and Hg (HWL: R2 ¼ 0.81, p < 0.001; NRT: R2 ¼ 0.44, p ¼ 0.007) supporting the hypothesis that Hg and algal productivity are increasing in tandem (Fig. 4). The S2F in HWL had a similar profile as the HgF, where concentrations increased up to 1997 mg HC m2 y1 between 6 and 8 cm due to the increase in the sedimentation rate. The S2F in NRT also increased slightly at 5 cm depth to 382.3 mg HC m2 y1 due to the slight increase in sedimentation rate. The S2F in both cores was also significantly correlated with the HgF (p < 0.001). The average hydrogen index (HI) and the oxygen index (OI) in the HWL core (205  26 mg HC g1 TOC and 219  14 mg CO2 g1 TOC, respectively) were within close range to those observed for the NRT core (212  4.9 mg HC g1 TOC and 201  11 mg CO2 g1 TOC, respectively). The source and quality of the sedimentary organic matter can be identified by plotting HI versus OI using a pseudo-van Krevelen diagram (Fig. S2 in Supplemental Material). This plot helps to illustrate the source and composition of organic matter by representing the proportion of hydrogen-rich organic matter (HI)

dominantly composed of autochthonous matter relative to oxygenrich OM exhibited by the lignin/cellulosic rich land-derived, higher plant OM in the sediments (Carrie et al., 2012; Langford and BlancValleron, 1990). In both HWL and NRT, the kerogen was Type IIeIII which is characterized as kerogens consisting mainly of organic matter with a relatively high proportion of land-derived OM (Carrie et al., 2009). The temporal variation of HI was inversely proportional to OI throughout the HWL core, showing an upward increase in HI towards the recent sediments, while OI decreased. This indicates that hydrogen-rich kerogen of predominantly algal-derived origin is becoming an increasingly important fraction of the OM in the sediment towards the upper section of the core (Fig. 4B). Most interestingly, we observed a sharp increase in OI associated with a sudden decline of HI at the depth corresponding to the sedimentation event of 1950e1960 (Fig. 4B). Stern et al. (2009) interpreted increases in OI such as this as being caused by factors such as inputs of reworked-oxidized organic carbon, the cellulosic remains of higher terrestrial plants, and/or airborne residues of carbonized organic matter such as char and ash from wildfires, further corroborating the hypothesis of a severe erosion event due to a wildfire in the HWL catchment during this time period. In contrast, NRT sediments showed a consistent temporal trend in HI while a progressive increase in OI towards the upper part of

58

M.L. Brazeau et al. / Environmental Pollution 173 (2013) 52e60

HI (mg HC g-1 TOC)

S2 (mg HC g-1, dw) 0

270 2010

5

1959

10

1952

15

1934

20

1912

25

4

6

8

10

14 150

12

190

A

B

C

D

230

1855

30 0

2009

5

Year

Core Depth (cm)

2

1945

10

1895

15 20 S2 Hg

25

HI OI

30 20

40

60

80

100

120 180

200

-1

220

240

260

OI (mg CO2 g-1 TOC)

Hg (ng g , dw)

Fig. 4. Algal-derived organic carbon (S2; mg HC g1, dw), mercury concentration (Hg; ng g1, dw) hydrogen index (mg HC g1 TOC) and oxygen index (mg CO2 g1 TOC) according to depth and year of deposition in sediment cores from Hawley (A, B) and North Raft (C, D) lakes.

the sediment core is observed. This indicates that NRT received higher proportion of land-derived OM than algal-derived OM in more recent times, possibly due to a shift in the depositional environment and the watershed features (i.e., Sanei et al., 2012). This also reinforces the marked difference between the watershed makeup and depositional environment of HWL and NRT lakes. This is supported by the C:N ratio data (Fig. 5) showing higher values for NRT than HWL and the NRT v13C signature is less depleted in 13C, both supporting a more allochthonous origin of OM in NRT.

3.5. Carbon and nitrogen There was a significant difference between the C:N ratio of the surface and deep sediments (p ¼ 0.01) of the 13 study lakes. The average C:N ratio in deep sediments was 12.9  3.0 and 10.5  1.2 in surface sediments. A decrease in C:N ratio was also observed in recent sediments of all three sediment cores (Fig. 5). The C:N ratios in surface and deep sediments were 9.23 and 11.3 in AQT, 9.21 and 10.20 in HWL, and 10.7 and 13.3 in NRT, respectively. Algae and

C:N

Core Depth (cm)

9

2010

0

10.5 2009

5

1971

5

1959

5

1945

10

1903

10

1952

10

1895

15

15

1934

15

20

20

1912

20

1855

25

25 30 -30.6

10

11

9.0

9.5

10.0

25

δ13C C:N -30.2

12

A -29.8

-29.4

-29.0

30 -31.4

B -31.0

-30.6

-30.2

-29.8

10.5 0

11.5

12.5

13.5 2009

Year

0

C

30 -29.4

-29.1

-28.8

-28.5

-28.2

δ13C Fig. 5. Carbon to nitrogen ratio (C:N) and v13C (average, n ¼ 3  SD) according to depth and year of deposition in sediment cores from Aquatuk (A), Hawley (B) and North Raft (C) lakes from the Hudson Bay lowlands.

M.L. Brazeau et al. / Environmental Pollution 173 (2013) 52e60

phytoplankton are characterized by high protein content and the absence of cellulose, and show low atomic C:N ratios (4e12) whereas terrestrial plants have higher C:N ratios due to their low protein content and abundance of cellulose (Meyers, 1994). Therefore the observed decrease in the C:N ratio in surface sediments provides some support to the hypothesis of increasing algal productivity within these lakes. However, the lability of carbon and nitrogen in oligotrophic lakes remains an outstanding issue. Several studies have shown the presence of active bacterial and archaeal populations in sub-surface sediments which consume various forms of carbon through diverse metabolic pathways (Biderre-Petit et al., 2011; Lay et al., 1996; Scholten et al., 2000). For example, Xia et al. (2008) demonstrated the impact of microbial populations in a bioreactor where the diversity of microbial community structure was positively correlated with C:N ratio. Therefore, the higher C:N ratio in subsurface sediments could be due to microbial consumption of organic carbon; further studies are needed to assess the impact of microbial populations on the C:N ratios in these lakes. Carbon isotopes are fractionated, to varying degrees, during photosynthesis due to the inherent kinetics of enzymatic reactions (Farquhar and Hubick, 1989). The two major terrestrial plant groups, C3 and C4, have v13C of approximately 26 to 28& and 12 to 14&, respectively (Hecky and Hesslein, 1995). A combination of physiological and environmental influences can modify the carbon isotopic composition of plankton communities resulting in a large range in v13C of algae (France, 1997, and references therein). Most of the study lakes were oligotrophic (total phosphorous ¼ 11.3  6.5 mg L1) with moderate DOC values (>5 mg/L), and therefore we expect the plankton community to exhibit depleted v13C signatures (sensu del Giorgio and Peters, 1994; Hessen, 1992). Surface sediment v13C signatures varied between 32.0& and 25.7& with an average of 29.0  2&, whereas deep sediments varied between 30.4& and 24.1& with an average of nbsp;27.9  2&. The temporal depletion of the surface sediment v13C signature is also seen in the three sediment cores. The average v13C of the sediment at surface and depth was 30.4& (0.04) and 29.3& (0.1) in AQT lake, 31.3& (0.04) and 30.6& (0.04) in HWL, and 29.2& (0.1) and 28.5& (0.2) in NRT, respectively. The slight depletion of the v13C signature in the surface sediment compared to deep sediment is not significant (p ¼ 0.12), however it could be an additional indicator of increasing primary productivity in these lakes over recent decades. The only proxy that was available to us in order to predict organic matter lability and therefore its availability to microbes originated from the Rock-Eval data that has been previously discussed. Overall, the significance of microbes (both in terms of magnitude and rate) in altering the pool of carbon within sediment is surprisingly lacking. While surface sediment processes have been well studied, the combined role of bio-diagenetic processes in carbon transformation through depth remains to be thoroughly explored. Our study offers new direction to tackle this question. Further experimental data on carbon processing rates by microbes in our sediments would be needed to include the possible role of microbes in the present discussion. 4. Conclusions Three lake sediment cores, and sediment from surface and deep intervals of 10 additional lakes from the Hudson Bay Lowlands, were analysed to determine the history of Hg deposition and the changes occurring in organic carbon concentrations in lakes of this region. Despite the remote location of the lakes, sediments revealed that they are influenced by atmospheric deposition of Hg. After the onset of the Industrial Revolution, when atmospheric concentrations of Hg began to rise, sedimentary concentrations of Hg increased in

59

all three cores and reached concentrations comparable to those found in other Canadian lakes subjected to anthropogenic Hg inputs (Lockhart et al., 1998). There was also a wide range in Hg concentrations in surface sediments, ranging from 6.13  0.7 to 125.4  0.9 ng g1 dw, which emphasizes the importance of in situ and post-depositional processes in the Hg cycle. Total Hg and MeHg were not correlated in surface sediments, indicating that total Hg is a poor predictor of MeHg, the form of greatest biological concern. There was a significant correlation between Hg and OC in all three sediment depth profiles however there was no significant relationship between Hg and OC in surface sediments across the study lakes. MeHg and OC were positively correlated in surface sediments of the lakes, suggesting that the production of MeHg in these lakes may be limited by the availability of OC. Rock-Eval analyses were applied to two of the sediment cores and significant relationships were found between S2 (algal-derived organic carbon) and Hg, supporting the hypothesis that [Hg] and algal productivity are increasing in tandem. Rock-Eval, OC, C:N ratios and v13C signatures all provide evidence to support an increase in primary productivity in these lakes, which is assumed to be a consequence of a warming climate. Acknowledgements This work was funded by NSERC Discovery grants to AJP and JMB, the Ontario Ministry of the Environment through the Climate Change and Multiple Stressor Research Program at Laurentian University, and the Ontario Graduate Scholarship Program. We thank Albert and Gilbert Chookomolin for their assistance in the field. We also thank Dr. Emmanuel Yumvihoze and Linda Kimpe for their help with analyses. Appendix A. Supplementary material Supplementary material related to this article can be found at http://dx.doi.org/10.1016/j.envpol.2012.09.017. References Appleby, P.G., 2001. Chronostratigraphic techniques in recent sediments. In: Last, W.M., Smol, J.P. (Eds.), Tracking Environmental Changes in Lake Sediments: Physical and Chemical Techniques. Kluwer Academic Publishers, Dordrect. Appleby, P.G., Oldfield, F., 1978. The calculation of lead-210 dates assuming a constant rate of supply of unsupported 210Pb to the sediment. Catena 5, 1e8. Avramescu, M.L., Yumvihoze, E., Hintelmann, H., Ridal, J., Fortin, D., Lean, D.R.S., 2011. Biogeochemical factors influencing net mercury methylation in contaminated freshwater sediments from the St. Lawrence River in Cornwall, Ontario, Canada. Science of the Total Environment 409, 968e978. Blais, J.M., Kalff, J., Cornett, R.J., Evans, R.D., 1995. Evaluation of 210Pb dating in lake sediments using stable Pb, Ambrosia pollen, and 137Cs. Journal of Paleolimnology 13, 169e178. Biderre-Petit, C., Jezequel, D., Dugat-Bony, E., Lopes, F., Kuever, J., Borrel, G., Viollier, E., Fonty, G., Peyret, P., 2011. Identification of microbial communities involved in the methane cycle of a freshwater Meromictic lake. Fems Microbiology Ecology 77, 533e545. Cai, Y., Jaffé, R., Alli, A., Jones, R.D., 1996. Determination of organomercury compounds in aqueous samples by capillary gas chromatography-atomic fluorescence spectrometry following solid-phase extraction. Analytica Chimica Acta 334, 251e259. Cai, Y., Tang, G., Jaffé, R., Jones, R.D., 1997. Evaluation of some isolation methods for organomercury determination in soil and fish samples by capillary gas chromatography e atomic fluorescence spectrometry. International Journal of Environmental Analytical Chemistry 68, 331e345. Carrie, J., Sanei, H., Goodarzi, F., Stern, G., Wang, F.Y., 2009. Characterization of organic matter in surface sediments of the Mackenzie River Basin, Canada. International Journal of Coal Geology 77, 416e423. Carrie, J., Sanei, H., Stern, G., 2012. Standardisation of RockeEval pyrolysis for the analysis of recent sediments and soils. Organic Geochemistry 46, 38e53. Carrie, J., Wang, F., Sanei, H., Macdonald, R.W., Outridge, P.M., Stern, G.A., 2010. Increasing contaminant burdens in an Arctic Fish, Burbot (Lota lota), in a warming climate. Environmental Science & Technology 44, 316e322. CCME, 2003. Canadian Water Quality Guidelines for the Protection of Aquatic Life: Inorganic Mercury and Methylmercury.

60

M.L. Brazeau et al. / Environmental Pollution 173 (2013) 52e60

del Giorgio, P.A., Peters, R.H., 1994. Patterns in planktonic P: R ratios in lakes e influences of lake trophy and dissolved organic carbon. Limnology and Oceanography 39, 772e787. Delongchamp, T.M., Ridal, J.J., Lean, D.R.S., Poissant, L., Blais, J.M., 2010. Mercury transport between sediments and the overlying water of the St. Lawrence River area of concern near Cornwall, Ontario. Environmental Pollution 158, 1487e 1493. EPA, U, 1996. Method 1631: Mercury in Water by Oxidation, Purge and Trap and Cold Vapor Atomic Fluorescence Spectrometry. U.S. Environmental Protection Agency, Office of Water. EPA821-R-96e001. Farquhar, E.J., Hubick, K.T., 1989. Carbon isotope discrimination and photosynthesis. Annual Review of Plant Physiology 40, 503e537. Fox, B.S., Mason, K.J., McElroy, F.C., 2005. Determination of total mercury in crude oil by combustion cold vapor atomic absorption spectrometry (CVAAS). In: Nadkarni, R.A.K. (Ed.), Elemental Analysis of Fuels and Lubricants: Recent Advances and Future Prospects, pp. 196e206. France, R.L., delGiorgio, P.A., Westcott, K.A., 1997. Productivity and heterotrophy influences on zooplankton delta C-13 in northern temperate lakes. Aquatic Microbial Ecology 12, 85e93. Glew, J., 1991. Miniature gravity corer for recovering short sediment cores. Journal of Paleolimnology 5, 285. Grant, R.F., Humphreys, E.R., Lafleur, P.M., Dimitrov, D.D., 2011. Ecological controls on net ecosystem productivity of a mesic arctic tundra under current and future climates. Journal of Geophysical Research-Biogeosciences 116. Gunn, J., Snucins, E., 2010. Brook charr mortalities during extreme temperature events in Sutton River, Hudson Bay Lowlands, Canada. Hydrobiologia 650, 79e84. Handley, L.L., Scrimgeour, C.M., Thornton, S.F., Sprent, J.I., 1991. Determination of the natural abundances of the stable isotopes of N-15 and C-13 by massspectrometry e a simplified manual methods for the preparation of N2 and CO2. Functional Ecology 5, 119e124. Hecky, R.E., Hesslein, R.H., 1995. Contributions of benthic algae to lake food webs as revealed by stable isotope analysis. Journal of the North American Benthological Society 14, 631e653. Hessen, D.O., 1992. Dissolved organic carbon in a humic lake e effects on bacterial production and respiration. Hydrobiologia 229, 115e123. Hintelman, H., Keppel-Jones, K., Evans, R.D., 2000. Constants of mercury methylation and demethylation rates in sediments and comparison of tracer and ambient mercury availability. Environmental Toxicology and Chemistry 19, 2204e2211. Kirk, J.L., Muir, D.C.M., Antoniades, D., Douglas, M.S.V., Evans, M.S., Jackson, T.A., Kling, H., Lamoureux, S., Lim, D.S.S., Pienitz, R., Smol, J.P., Stewart, K., Wang, X., Yang, F., 2011. Climate change and mercury accumulation in Canadian high and subarctic lakes. Environmental Science & Technology 45, 964e970. Lafargue, E., Marquis, F., Pillot, D., 1998. Rock-Eval 6 applications in hydrocarbon exploration, production, and soil contamination studies. Revue De L’Institut Français Du Pétrole 53, 421e437. Lamborg, C.H., Fitzgerald, W.F., O’Donnell, J., Torgerson, T., 2002. A non-steady-state compartmental model of global-scale mercury biogeochemistry with interhemispheric atmospheric gradients. Geochimica et Cosmochimica Acta 66, 1105e1118. Landers, D., Gubala, C., Verta, M., Lucotte, M., Johansson, K., Vlasova, T., Lockhart, W., 1998. Using lake sediment mercury flux ratios to evaluate the regional and continental dimensions of mercury deposition in arctic and boreal ecosystems. Atmospheric Environment 32, 919e928. Langford, F.F., Blanc-Valleron, M.M., 1990. Interpreting Rock-Eval pyrolysis data using graphs of pyrolyzable hydrocarbons vs total organic carbon. American Association of Petroleum Geology Bulletin 74, 799e804. Lay, J.J., Miyahara, T., Noike, T., 1996. Methane release rate and methanogenic bacterial populations in lake sediments. Water Research 30, 901e908. Lindberg, S.E., Harriss, R.C., 1974. Mercury-organic matter associations in estuarine sediments and interstitial water. Environmental Science & Technology 8, 459e 462. LLC, U, 2004. Total Mercury and Mercury Species in Liquid Hydrocarbons. UOP Method 938e00. Lockhart, W.L., Wilkinson, P., Billeck, B.N., Danell, R.A., Hunt, R.V., Brunskill, G.J., Delaronde, J., St Louis, V., 1998. Fluxes of mercury to lake sediments in central and northern Canada inferred from dated sediment cores. Biogeochemistry 40, 163e173. Macdonald, R.W., Bewers, J.M., 1996. Contaminants in the arctic marine environment: priorities for protection. Ices Journal of Marine Science 53, 537e563. Mason, R.P., Lawrence, A.L., 1999. Concentration, distribution, and bioavailability of mercury and methylmercury in sediments of Baltimore Harbor and

Chesapeake Bay, Maryland, USA. Environmental Toxicology and Chemistry 18, 2438e2447. Meyers, P.A., 1994. Preservation of elemental and isotopic source identification of sedimentary organic-matter. Chemical Geology 114, 289e302. Mills, R.B., Paterson, A.M., Blais, J.M., Lean, D.R.S., Smol, J.P., Mierle, G., 2009. Factors influencing the achievement of steady state in mercury contamination among lakes and catchments of south-central Ontario. Canadian Journal of Fisheries and Aquatic Sciences 66, 187e200. Morel, F.M.M., Kraepiel, A.M.L., Amyot, M., 1998. The chemical cycle and bioaccumulation of mercury. Annual Review of Ecology and Systematics 29, 543e 566. Muir, D.C.G., Wang, X., Yang, F., Nguyen, N., Jackson, T.A., Evans, M.S., Douglas, M., Kock, G., Lamoureux, S., Pienitz, R., Smol, J.P., Vincent, W.F., Dastoor, A., 2009. Spatial trends and historical deposition of mercury in eastern and northern Canada inferred from lake sediment cores. Environmental Science & Technology 43, 4802e4809. Omelchenko, A., Lockhart, W.L., Wilkinson, P., 1995. Study of the Depositional Characteristics of the Lake Sediments across Canada with 210-Pb and Cs-137. Outridge, P.M., Sanei, H., Stern, G.A., Goodsite, M., Hamilton, P.B., Carrie, J., Goodarzi, F., Macdonald, R.W., 2011. Comment on climate change and mercury accumulation in Canadian high and subarctic lakes. Environmental Science & Technology 45, 6703e6704. Outridge, P.M., Sanei, H., Stern, G.A., Hamilton, P.B., Goodarzi, F., 2007. Evidence for control of mercury accumulation rates in Canadian High Arctic lake sediments by variations of aquatic primary productivity. Environmental Science & Technology 41, 5259e5265. Pella, E., 1990. Elemental Organic Analysis. Instruction Manual for the EA 1110. Pennington, W., Tutin, T.G., Cambray, R.S., Fisher, E.M., 1973. Observations on lake sediments using fallout Cs-137 as a tracer. Nature 242, 324e326. Riget, F., Braune, B., Bignert, A., Wilson, S., Aars, J., Born, E., Dam, M., Dietz, R., Evans, M., Evans, T., Gamberg, M., Gantner, N., Green, N., Gunnlaugsdottir, H., Kannan, K., Letcher, R., Muir, D., Roach, P., Sonne, C., Stern, G., Wiig, O., 2011. Temporal trends of Hg in Arctic biota, an update. Science of the Total Environment 409, 3520e3526. Rouse, W.R., 1991. Impacts of Hudson-Bay on the terrestrial climate of the HudsonBay Lowlands. Arctic and Alpine Research 23, 24e30. Rouse, W.R., Douglas, M.S.V., Hecky, R.E., Hershey, A.E., Kling, G.W., Lesack, L., Marsh, P., McDonald, M., Nicholson, B.J., Roulet, N.T., Smol, J.P., 1997. Effects of climate change on the freshwaters of arctic and subarctic North America. Hydrological Processes 11, 873e902. Rydberg, J., Galman, V., Renberg, I., Bindler, R., Lambertsson, L., MartinezCortizas, A., 2008. Assessing the stability of mercury and methylmercury in a Varved lake sediment deposit. Environmental Science & Technology 42, 4391e4396. Sanei, H., Outridge, P.M., Dallimore, A., Hamilton, P.B., 2012. Mercury-organic matter relationships in pre-pollution sediments of thermokarst lakes from the Mackenzie River Delta, Canada: the role of depositional environment. Biogeochemistry 107, 149e164. Sanei, H., Stasiuk, L.D., Goodarzi, F., 2005. Petrological changes occurring in organic matter from recent lacustrine sediments during thermal alteration by Rock-Eval pyrolysis. Organic Geochemistry 36, 1190e1203. Scholten, J.C.M., Conrad, R., Stams, A.J.M., 2000. Effect of 2-bromo-ethane sulfonate, molybdate and chloroform on acetate consumption by methanogenic and sulfate-reducing populations in freshwater sediment. Fems Microbiology Ecology 32, 35e42. Stern, G.A., Sanei, H., Roach, P., Dalaronde, J., Outridge, P.M., 2009. Historical interrelated variations of mercury and aquatic organic matter in lake sediment cores from a subarctic lake in Yukon, Canada: further evidence toward the algal-mercury scavenging hypothesis. Environmental Science & Technology 43, 7684e7690. U.N.E.P, 2008. The Global Atmospheric Mercury Assessment: Sources, Emissions and Transport. Xia, S., Li, J., Wang, R., 2008. Nitrogen removal performance and microbial community structure dynamics response to carbon nitrogen ratio in a compact suspended carrier biofilm reactor. Ecological Engineering 32, 256e262. Yang, H., Engstrom, D.R., Rose, N.L., 2010. Recent changes in atmospheric mercury deposition recorded in the sediments of remote Equatorial lakes in the Rwenzori Mountains, Uganda. Environmental Science & Technology 44, 6570e 6575. Zhang, Y., Chen, W.J., Riseborough, D.W., 2008. Transient projections of permafrost distribution in Canada during the 21st century under scenarios of climate change. Global and Planetary Change 60, 443e456.