2b-1: Sources and cycling of mercury in freshwater ecosystems

Mercury (Hg) methylation often occurs at the active redox boundary between oxic and anoxic conditions in sediment and the water column of lakes. Previous studies have suggested that sediment methylation rates are highest when the overlying water is oxygenated and that during periods of stratification the predominant zone of methylation activity switches from the sediment to the water column. However, studies that simultaneously measured methylmercury (MeHg) production in the water column and sediment remain limited. Understanding the relative importance of sediment versus water column methylation and how this changes in response to seasonal stratification has important implications for management strategies aimed at reducing MeHg production, such as hypolimnion oxygenation. Our study measured Hg methylation and demethylation rates using inorganic Hg and MeHg stable isotope tracers in sediments and water of the littoral zones and deeper central waters of a reservoir in Californias central coastal range. Measurements were conducted during well-mixed/oxygenated conditions (winter) as well as at the beginning (spring) and end of thermal stratification (late-summer). We also evaluated how variations in the isotope spike solution (using DI water, ambient water, carbon enriched water) affected the methylation rates. These results found the lowest methylation rates were associated with the DI isotope spike solution, suggesting that microbial community prefers Hg bound to a carbon source. The results from the field sampling showed that the ambient MeHg concentrations were very high in the hypolimnetic waters (up to 7.5 ng/L representing up to 79% MeHg/THg) during periods of stratification. During the late summer, the sediment in the littoral zones underlying oxic water had significantly higher methylation rates (2.4 %/day) compared to the sediments underlying anoxic water in the hypolimnion (1.3 %/day). Sediment demethylation rates were much higher in sites under the anoxic hypolimnetic waters (9.8%/day), compared to the littoral sediments with overlying oxic water (2.2 %/d). These results highlight the importance of net MeHg production in the littoral zones. Following turnover, methylation rates in this deep section of the reservoir did not increase when the overlying water became oxygenated. Under anoxic conditions, water-column methylation rates were of similar magnitude to the sediment (3.0 %/day) and were insignificant in oxygenated water. Taken together, these results suggest that methylation in the water column provides a significant source of MeHg to the reservoir, and that following turnover when water-column methylation ceases, there is not a concomitant shift towards higher production in the sediments.

The aim of this study is to understand the correlation between MMHg concentrations in fish and trophic states of different reservoirs. For this purpose, we investigated MMHg concentrations in 466 samples of four common fish species (barbel steed, largemouth bass, leopard mandarin, and bluegill) and trophic states of 14 artificial reservoirs in South Korea. The trophic state index (TSI) of each reservoir was determined using empirical equations based on the monthly chlorophyll-a, total phosphorus, and Secchi depth collected over a three-year period. The length-normalized MMHg concentrations in fish (MMHgadj) showed a negative correlation with the TSI based on total phosphorus (r2 =0.75), which might be a result of particle dilution of MMHg in surface waters. A log-normal correlation was found between the fish MMHgadj and the TSI based on chlorophyll-a (r2 =0.76), attributable to low chlorophyll-a concentrations, despite of the high particle densities, in the oligotrophic reservoirs. This study is the first to reveal MMHg accumulation in fish via the relationship with TSI of the lakes, suggesting that the measurement of TSI based on chlorophyll-a and total phosphorus is an effective way to predict MMHg bioaccumulation across diverse reservoirs.

The Three Gorges reservoir (TGR) is the largest hydroelectric reservoir in China, with a total water surface area of 1080 km2 and a storage capacity of 39.3 billion m3. Hydroelectric regulation results in a water-level fluctuation zone (WLFZ) with a total area of 306.3 km2 that undergoes annual flooding and drying alternating cycles. The alternating redox conditions in this WLFZ make it a very interesting area for the cycling of mercury (Hg) and sulfur (S), although the effect of the S redox cycle on Hg methylation is not clear. Sulfide may reduce the availability of Hg through immobilizing mercuric ions as sparingly soluble HgS(s). On the other hand, the dissolved mercury-sulfide (Hg-S) complexes formed in a sulfidic environment may facilitate Hg methylation. Here we present the results from a year-long monitoring of the WLFZ, as well as a flooding and drying simulation experiment with elemental sulfur (S0) addition. Field studies showed that the concentrations of both methylmercury (MeHg) and reduced S in the WLFZ were significantly higher than that in non-flooded soils. Significant positive correlations were found between MeHg and S0 (p < 0.01), acid volatile sulfide (AVS) (p < 0.01), pH (p < 0.05) and soil organic matter (p < 0.01) in soil/sediment of the WLFZ. In the flooding and drying simulation experiment, S0, which is an intermediate product of sulfate reduction and sulfide re-oxidation, showed a higher correlation (r = 0.74, p < 0.05) with MeHg when compared to field samples. In the S0 added anaerobic sediment, the correlation coefficient of S0 and MeHg was even higher (r = 0.86, p < 0.01) when the content of dissolved organic matter was held constant. These relationships indicate that S0 in the redox alternating WLFZ environment could promote MeHg production especially under reducing conditions. This could be due to the formation of polysulfides from reactions of S0 and sulfide, which are known to increase the concentrations of dissolved Hg from the solid HgS pool in the soil. The periodic regulation of the water level in the TGR and the resulting redox cycling of inorganic sulfur in the WLFZ may thus enhance the production of MeHg in both aquatic and terrestrial ecosystems.

Mercury (Hg), a neurotoxin with a global atmospheric distribution, has been shown to readily deposit onto northern landscapes where it can enter freshwater systems and bioaccumulate in sport and subsistence fish. However, the biogeochemical cycle of Hg, including bioaccumulation, is complex and influenced by many physicochemical and ecological factors. While several recent reviews have been published on the factors that affect aquatic Hg cycling in the Arctic, reviews on Hg research in the global boreal zone are lacking. The main goal of this review was to assess which physicochemical parameters (e.g., DOC, pH, Lake size) significantly affect Hg concentrations in boreal waters, sediments, peatlands, invertebrates, and/or fish in 5 key boreal regions: North America, Fennoscandia, Russia, China, and other countries (e.g., Estonia). A total of 673 correlative relationships were found between various physicochemical parameters and abiotic and biotic Hg concentrations in freshwater systems across these regions. No major spatial differences were found in the direction of the relationships reported between Hg concentrations and a given physicochemical parameter. Consistent positive relationships were frequently reported between Hg concentrations in water, sediments, and biota and aqueous organic matter (OM) content as well as certain watershed characteristics (e.g., % forested or wetland area) of freshwater boreal systems, suggesting that terrestrial Hg is transported into freshwater systems by OM complexes. Conversely, base cations were negatively correlated with Hg concentrations in boreal sediments, invertebrates, and fish, potentially due to an inhibition of Hg methylation and increased competition for binding sites on OM complexes. In addition to assessing these correlative relationships, we compiled data on Hg in fish muscle from 92 studies across the boreal region and found that perch (Perca flavescens in N. America and Perca fluviatilis in Fennoscandia/Russia) had significantly lower length-standardized (at 14 cm) total Hg concentrations in Fennoscandian lakes (0.176±1.097 ppm, wet weight) compared to North American (0.356±1.108 ppm, wet weight) and Russian (0.471±1.320 ppm, wet weight) lakes. Results from this review will provide valuable insight into which physical and chemical factors control Hg cycling across the boreal zone. We also discuss the implications of these results in the context of anticipated changes to boreal landscapes due to various types of industrial development (e.g., forestry, mining, and hydropower), which are expanding across the boreal.

Anthropogenic emissions combined with long-range transport and atmospheric deposition have contributed to a state-wide fish consumption advisory for mercury (Hg) in Michigan. Lake sediment cores in Michigans Upper Peninsula (UP) indicate that Hg accumulation increased approximately 3-fold above background rates. There are conflicting predictions as to the future trajectory for Hg deposition and for the recovery of lakes from that deposition. Concentrations of Hg in Michigan rivers and lakes increased over the past two decades during which time atmospheric deposition declined. Furthermore, the geographic pattern of concentrations in lakes and fish does not match the pattern of atmospheric deposition. Accurate prediction of the future recovery of lakes requires an understanding of the magnitude and timing of mercury runoff from terrestrial catchments as well as understanding of the rate of cycling of mercury within the lake and food web.

This study applied both a statistical and a mechanistic modeling approach to meet the two objectives:
1) to determine the causes of variability for mercury in lakes of the UP, and 2) to predict the time required for fish from UP lakes to be safe for unlimited consumption. We examined the spatial distribution of mercury in lakes, fish, rivers and atmospheric deposition across the Upper Peninsula (UP) of Michigan. Multiple data bases were combined to estimate the magnitude of mercury runoff from catchments, the best predictor variables for mercury concentrations in lakes, and the factors best explaining methyl mercury (MeHg) concentrations in fish. The statistical analyses enabled us to create categories of lakes having similar fish mercury, and to apply a Hg cycling model to predict the timeline for recovery under different scenarios for Hg deposition.

This study indicates that the magnitude of mercury runoff from catchments has been much less (5-15% of atmospheric deposition) than estimated by previous studies (~25%). Large variations in dissolved Hg concentrations as well as mercury content in fish among lakes in Michigans Upper Peninsula (UP) suggest that runoff may be highly variable among catchments, but the major rivers in the UP show remarkable uniformity in runoff. While DOC is an important explanatory variable for dissolved mercury in lakes, the DOC is not well predicted by wetland areas. For mercury in fish, lake pH is the most significant predictor variable, but lake size is also important. The time to recovery depends primarily on the regulations imposed on emissions and on the timescale for run-off from catchments.

Over the last decade, a rapidly evolving interest in mercury’s (Hg) isotopic composition have shed light on the pathways and chemical forms Hg undertakes, along with identification of contributing sources to the complex global Hg-cycle, yet many questions remain regarding source and trophic transfer of Hg in natural systems. Marine and fresh-water food webs, and the Hg isotopic composition within them, are often investigated separately and treated as two distinctly different ecosystems with little overlap between them. An exception to this is fish farms, where fresh-water fish is bred and raised on a man-made and atypical high energy diet, which main protein source is based on fishery products of marine origin. Yet, the effect on natural ecosystems of non-native fish, raised on a marine protein based diet, in a geochemical perspective in general, and in the scope biomagnification of Hg and Hg-isotopes specifically, is still unknown. Using a combined geochemistry (THg) and stable isotope (N, C, Sr and Hg) approach, we here show that the introduction of farmed brown trout (Salmo Trutta Fario) to a high-altitude catchment leads to a small, yet relevant, input of directly bioavailable MeHg and to a recycling of marine Hg, which potentially clouds the isotopic signature of Hg in the natural system. We found that farmed trout show Hg-isotopic signatures directly comparable to that of marine and off-coastal biota whereas wild trout indicated Hg-isotopic signatures related to atmospherically derived Hg and photochemical processes. The stocked, farm-rearing fish, showed Hg-isotopic signatures more similar to the marine and coastal biota, but as the trout evolved and shifted its diet to a higher trophic level, gaining an overall higher THg-concentration, the marine farmed rearing signature was diluted by the atmospherically derived Hg, rendering in lower MDF (mass dependent fractionation, δ202Hg) and higher MIF (mass independent fractionation, Δ199Hg). In contrast to the current status quo on global Hg-cycle, i.e. that bioaccumulated MeHg biomagnifies up the food chain we here claim that in watersheds subject to stocking of farmed fish, Hg does not only bioaccumulate and biomagnify, but also re-biomagnifies, thus leading to inherent difficulties when deciphering the ecological pathway of Hg transfer on global scales, but also locally i.e. through the food web within an aquatic system.

Mercury (Hg) concentration trends in top predator fish (lake trout and walleye) of the Great Lakes (GL) from 2004 to 2015 were determined by Kendall−Theil robust regression with a cluster-based age normalization method to control for the effect of changes in lake trophic status. When data from the GLs (except Lake Erie) are combined, a significant decreasing trend in the lake trout Hg concentrations was found between 2004 and 2015 with an annual decrease of 4.1% per year, consistent with the decline in regional atmospheric Hg emissions and water Hg concentrations. However, a breakpoint was detected with a significant decreasing slope (−8.1% per year) before the breakpoint (2010), and no trend after the breakpoint. When the lakes are examined individually, Lakes Superior and Huron, which are dominated by atmospheric Hg inputs and are more likely than the lower lakes to respond to declining emissions from areas surrounding the GL, have significant decreasing trends with rates between 5.2 and 7.8% per year from 2004 to 2015. These declining trends appear to be driven by decreasing regional atmospheric Hg emissions although they may be partly counterbalanced by other factors, including increasing local emissions, food web changes, eutrophication, and responses to global climate change. Lakes Michigan, Erie and Ontario may have been more impacted by these other factors and their trends changed from decreasing to non-decreasing or increasing in recent years.

Photo-reduction and photo-oxidation are fundamental mechanisms controlling mercury volatilization and accumulation in freshwaters. In all surface waters dissolved gaseous mercury (DGM) is produced as a net result of the reduction of reducible mercury, which is believed to be primarily divalent mercury (Hg(II)) bound to specific carbon-based ligands, and the oxidation of elemental mercury (Hg(0)). These two processes control the amount of DGM available for evasion across the water-air interface; however, determination of the fundamental rate constants and mechanisms of these reactions in freshwaters are still areas that require more research. In particular, the total amount of photoreducible mercury is emerging as a key variable that requires more exploration.

Here, we review the progress our group has made in this field over the past 10 years; we present rate constants as well as temporal dynamics in total reducible mercury derived from two recent projects that examined water samples from a series of freshwater lakes in Kejimkujik National Park, Nova Scotia, Canada. We examined the hypothesis that gross photoreduction and photooxidation rates would be significantly different in lake water. Another hypotheses was that the amount of mercury available for reaction with solar radiation (i.e. reduction of Hg(II) to gaseous Hg(0)) in surface waters would significantly change over a summer. A Luzchem photo-reactor was used to irradiate 200 mL water samples in quartz beakers continuously exposed to ultraviolet radiation for 24 h with concurrent Hg(0) analysis to derive pseudo-first order gross reduction rate constants and batch experiments were used to derive net reduction rates (and gross photooxidation by difference).

Results showed that the net photo-oxidation rates for freshwaters were low, with mercury reduction and oxidation reactions very close to being in balance. We also found that the amount of total reducible Hg(II) changed significantly in three of the lakes over several sampling months. Dissolved organic carbon concentration was a key factor positively correlated with these results. This research provides the first quantitative measurements of gross photooxidation and photoreduction rates as well as total photo-reducible mercury over a season in surface freshwater lakes.