Assessing the utility of sulfur isotope values for understanding mercury concentrations in water and biota from high Arctic lakes

Methylmercury (MeHg) biomagnifies through aquatic food webs resulting in elevated concentrations in fish globally. Stable carbon and nitrogen isotopes are frequently used to determine dietary sources of MeHg and to model its biomagnification. However, given the strong links between MeHg and sulfur cycling, we investigated whether sulfur isotopes (δ34S) would improve our understanding of MeHg concentrations ([MeHg]) in Arctic lacustrine food webs. Delta34S values and total mercury (THg) or MeHg were measured in water, sediments, and biota from six lakes near Resolute Bay, NU, Canada. In two lakes impacted by historical eutrophication, aqueous sulfate δ34S was ∼8‰ more positive than sedimentary δ34S, suggestive of bacterial sulfate reduction in the sediment. In addition, aqueous δ34S showed a significant positive relationship with aqueous [MeHg] across lakes. Within taxa across lakes, [THg] in Arctic char muscle and [MeHg] in their main prey, chironomids, were positively related to their δ34S values across lakes, but inconsistent relationships were found across entire food webs among lakes. Across lakes, nitrogen isotopes were better predictors of biotic [THg] and [MeHg] than δ34S within this dataset. Our results suggest some linkages between Hg and S biogeochemistry in high Arctic lakes, which is an important consideration given anticipated climate-mediated changes in nutrient cycling.

. Furthermore, since the δ 34 S values 91 in a predator generally reflect that of its diet (McCutchan et al. 2003), this isotope allows one to Information file (SI file; Figure SI -5). This region has a polar desert climate, with low mean 119 annual precipitation and temperature (150 mm and -16.5°C;Antoniades et al., 2011). Detailed 120 water chemistry and lake morphological characteristics are reported in Lescord et al. (2015a). 121 Briefly, these lakes range in size from 39 to 130 ha and in maximum depth from 8 to 27 m, and 122 they are ultra-oligotrophic, with mean chlorophyll-a, total phosphorus (TP), and dissolved 123 organic carbon (DOC) concentrations ranging from 1.28 to 2.23 µg/L, 3.6 to 8.0 µg/L, and 0.5 to 124 1.9 mg/L, respectively. Dissolved oxygen and temperature profiles showed no thermal or oxygen 125 stratification in the water columns of these systems (Lescord et al. 2015a). Mean pH was similar 126 across the lakes, ranging from 7.9 to 8.1, whereas mean aqueous SO 4 2in surface waters were 127 low but varied by approximately 6-fold (0.23 to 1.35 mg/L across lakes). The food webs have 128 low biodiversity; Arctic char are the only fish species, while the benthic invertebrate community 129 consists mainly of chironomids and the zooplankton community is dominated by copepods 130 (Chételat and Amyot 2009). The lakes from this study (and therefore the char that reside in them) D r a f t Page | 7 2011) that then flows into Resolute Lake. Although recovery from this eutrophication has 137 occurred, these inputs affected both the chemical and microbial characteristics of Meretta Lake 138 (Antoniades et al. 2011). Furthermore, the proximity of both Meretta and Resolute Lakes to the 139 local airport has altered their organic contaminant profiles when compared to more remote lakes 140 in the area (Lescord et al. 2015b;Cabrerizo-Pastor et al. 2018).  Lescord et al. (2015a). Briefly, water samples were collected weekly above the deepest part of 147 each lake, approximately 1 m below the surface using a Niskin bottle. Larval and adult 148 chironomids were collected by kick sweeping shore lines and aspiration from the ice surface, 149 respectively. Periphyton scrapings were taken from littoral rocks in triplicates. Bulk zooplankton 150 samples were collected by towing a net (80 μm mesh) for 3-5 minutes on each lake; these 151 samples were then size filtered at <250 μm, 250-500 μm, and >500 μm. Large char (>18 cm total 152 length) were collected using gill nets and small char (<18 cm total length) were electrofished 153 along shorelines; these size categories of fish were kept separate for all laboratory and statistical 154 analyses. Fish sampling was done using protocols approved by the Animal Care Committee of 155 the University of New Brunswick (#2010/11-1D-01). For δ 34 S analysis, water samples were 156 collected from the surface and deep point of each lake (n = 1 replicate per lake per depth, in each D r a f t Page | 8 159 Continental Shelf (PCSP) base by ion-exchange as barium sulfate (Carmody et al. 1998). Surface 160 sediment samples from the Arctic lakes were taken from the top 0.5 cm slice of cores (n = 18, 3 161 replicates per lake in 2011) that were collected in the littoral zone of each lake as part of other 162 studies (Lescord et al. 2015a;Lescord et al. 2015b      three of the remote lakes (0.1-0.12%), but not 9-Mile Lake, which had sediments with >0.4% 273 (see Table SI-1). Changes to the microbial community can alter sulfur cycling and different 274 strains of SRB fractionate S isotopes at different rates (Wing and Halevy 2014;Bradley et al. 275 2016). While the anthropogenic changes to bacterial communities in Meretta have been 276 relatively well-studied, any changes to the microbial community in Resolute Lake have not.

277
Furthermore, the historical eutrophication of Meretta Lake may have created more favorable 278 redox condition in its sediments, enhancing sulfate reduction (Holmer and Storkholm 2001).

287
However, recent studies suggest that the rates and degree of fractionation of S isotopes due to 288 SRBs vary among bacterial strains/species (Bradley et al. 2016). Little is known about SRB 289 communities in these high Arctic lakes where aqueous sulfate concentrations are very low (i.e., Although δ 13 C values indicated differences in the habitat use of zooplankton and 297 chironomids in our study lakes (see Lescord et al. 2015a), there was no consistent difference in 298 δ 34 S values between these groups (Figure 2A) which precluded the use of isotope mixing models 299 to assess their relative importance in the diets of Arctic char. Delta 34 S values were also not useful 300 in distinguishing between benthic invertebrates and zooplankton as prey for fish in temperate 301 coastal lakes (Clayden et al. 2017). In our Arctic lakes, δ 13 C values were more negative in 302 pelagic zooplankton than benthic invertebrates in each lake ( Figure 2B) and mixing models 303 indicated that most or all of the carbon (≥76% across lakes) in the diet of char in these systems is 304 derived from benthic rather than pelagic sources (Lescord et al. 2015a

337
(LME p = 0.129; Figure 3). However, this relationship is largely influenced by Meretta Lake and 338 when removed, the negative relationship between aqueous THg concentrations and δ 34 S became 339 significant (LME p = 0.040, r 2 = 0.506; not shown). Because the number of lakes examined in D r a f t Page | 16 340 this study was low, we were not able to determine whether Meretta Lake's influential data were 341 indeed statistical outliers. Studies have shown that the lighter δ 34 S of DOM complexes are 342 reflective of a terrestrial-based carbon source and to aerobic conditions in riparian soils and 343 limited S isotope fractionation by shoreline plants (Giesler et al. 2009). DOM complexes are 344 important transporters of Hg in aquatic systems; therefore terrestrial organic matter increases 345 aqueous [THg] in boreal streams and lakes presumably due to enhanced transport from riparian 346 areas (Teisserenc et al. 2011(Teisserenc et al. , 2014Eklöf et al. 2012). The negative relationship between δ 34 S 347 values of sulfate and aqueous THg in the current study may therefore reflect higher Hg input 348 with greater terrestrial DOM. However, sulfur compounds, DOM, and Hg have complex 349 biogeochemical interactions (e.g., Haitzer et al. 2002;Ravichandran 2004) and further study is  (Schmitt et al., 2011), no relationships between these variables across lakes (Ethier et al., 366 2008;Clayden et al., 2017), or that δ 34 S was a strong positive predictor of [THg] in fishes in 367 estuarine wetlands (Willacker et al. 2017). In the latter study, the authors attribute the positive 368 relationship to the influence of marine S values as well as the presumed higher rates of Hg 369 methylation and 34 S enrichment due to SRB activity in impounded sites (Willacker et al. 2017). and show that the [Hg]-δ 34 S relationships in food webs can differ between similar lake 390 ecosystems within a relatively small geographic area (i.e., ~12 km). It is possible that various 391 physical and chemical factors (e.g., stratification and anoxic habitat) may contribute to these 392 differences, and more work is needed to investigate these effects. Interestingly, the two lake food 393 webs that showed the strongest relationships between [Hg] and δ 34 S had considerably different 394 morphology: Meretta Lake was one of the smallest (0.39 km 2 ) and shallowest lakes (max depth = 395 9.2 m), while Resolute Lake was the largest (1.3 km 2 ) and one of the deepest systems (14.7 m) in 396 our dataset.  suggests increased Hg uptake in individuals with more benthic feeding.

421
Overall, these results imply that [Hg] in Arctic biota are most strongly related to an    (Michelutti et al. 2007) and changes to microbial S cycling have been reported (Drevnick et al. 438 2010).  Table 1: A summary of LASSO regression results of isotopic and elemental predictors of log 10 THg or MeHg concentrations in biota across six lake food webs near Resolute, Nunavut, Canada. Variables included in the models were carbon, nitrogen, and sulfur isotopic and elemental measures (δ 34 S, δ 15 N adj , δ 13 C, %S, %N, %C). The results presented are considered the best-fit model parameters and were selected based on Mallow's Cp values at a given level of the penalty parameter (λ).  ) and sedimentary sulfur (S) from six lakes near Resolute Bay, Nunavut, Canada. Lakes are arranged on the y-axis in order of increasing mean SO 4 2concentration in surface water (i.e., 1.35±0.22, 0.85±0.10, 0.56±0.14, 0.31±0.12, 0.71±0.09, and 0.23±0.05 mg/L, respectively; Lescord et al. 2015a) within groups (remote versus impacted). Total sulfur content in North Lake was too low to obtain δ 34 S values. MeHg percentages and concentrations (mean ± SD) in surface waters of these lakes are also plotted (data from Lescord et al. 2015a).