Mercury photoreduction and photooxidation in lakes: Effects of filtration and dissolved organic carbon concentration.

Mercury is a globally distributed, environmental contaminant. Quantifying the retention and loss of mercury is integral for predicting mercury-sensitive ecosystems. There is little information on how dissolved organic carbon (DOC) concentrations and particulates affect mercury photoreaction kinetics in freshwater lakes. To address this knowledge gap, samples were collected from ten lakes in Kejimkujik National Park, Nova Scotia (DOC: 2.6-15.4mg/L). Filtered (0.2µm) and unfiltered samples were analysed for gross photoreduction, gross photooxidation, and net reduction rates of mercury using pseudo first-order curves. Unfiltered samples had higher concentrations (p=0.04) of photoreducible divalent mercury (Hg(II)RED) (mean of 754±253pg/L) than filtered samples (mean of 482±206pg/L); however, gross photoreduction and photooxidation rate constants were not significantly different in filtered or unfiltered samples in early summer. DOC was not significantly related to gross photoreduction rate constants in filtered (R2=0.43; p=0.08) and unfiltered (R2=0.02; p=0.71) samples; DOC was also not significantly related to gross photooxidation rate constants in filtered or unfiltered samples. However, DOC was significantly negatively related with Hg(II)RED in unfiltered (R2=0.53; p=0.04), but not in filtered samples (R2=0.04; p=0.60). These trends indicate that DOC is a factor in determining dissolved mercury photoreduction rates and particles partially control available Hg(II)RED in lake water. This research also demonstrates that within these lakes gross photoreduction and photooxidation processes are close to being in balance. Changes to catchment inputs of particulate matter and DOC may alter mercury retention in these lakes and could partially explain observed increases of mercury accumulation in biota.

The Estimated Six-Year Mercury Dry Deposition Across North America.

Dry deposition of atmospheric mercury (Hg) to various land covers surrounding 24 sites in North America was estimated for the years 2009 to 2014. Depending on location, multiyear mean annual Hg dry deposition was estimated to range from 5.1 to 23.8 µg m-2 yr-1 to forested canopies, 2.6 to 20.8 µg m-2 yr-1 to nonforest vegetated canopies, 2.4 to 11.2 µg m-2 yr-1 to urban and built up land covers, and 1.0 to 3.2 µg m-2 yr-1 to water surfaces. In the rural or remote environment in North America, annual Hg dry deposition to vegetated surfaces is dominated by leaf uptake of gaseous elemental mercury (GEM), contrary to what was commonly assumed in earlier studies which frequently omitted GEM dry deposition as an important process. Dry deposition exceeded wet deposition by a large margin in all of the seasons except in the summer at the majority of the sites. GEM dry deposition over vegetated surfaces will not decrease at the same pace, and sometimes may even increase with decreasing anthropogenic emissions, suggesting that Hg emission reductions should be a long-term policy sustained by global cooperation.

Impact of closing Canada's largest point-source of mercury emissions on local atmospheric mercury concentrations.

The Flin Flon, Manitoba copper smelter was Canada's largest point source of mercury emissions until its closure in 2010 after ~80 years of operation. The objective of this study was to understand the variables controlling the local ground-level air mercury concentrations before and after this major point source reduction. Total gaseous mercury (TGM) in air, mercury in precipitation, and other ancillary meteorological and air quality parameters were measured pre- and postsmelter closure, and mercury speciation measurements in air were collected postclosure. The results showed that TGM was significantly elevated during the time period when the smelter operated (4.1 ± 3.7 ng m(-3)), decreased only 20% during the year following its closure, and remained ~2-fold above background levels. Similar trends were observed for mercury concentrations in precipitation. Several lines of evidence indicated that while smelter stack emissions would occasionally mix down to the surface resulting in large spikes in TGM concentrations (up to 61 ng m(-3)), the largest contributor to elevated TGM concentrations before and after smelter closure was from surface-air fluxes from mercury-enriched soils and/or tailings. These findings highlight the ability of legacy mercury, deposited to local landscapes over decades from industrial activities, to significantly affect local air concentrations via emissions/re-emissions.

Abiotic production of methylmercury by solar radiation.

Methylmercury [MeHg(I) in the aerobic surface water of lakes is thought to be rapidly degraded, but contrary to expectations, we show that MeHg(I) concentrations often increase during sunlight hours or remain relatively constant. We hypothesized that there were water column processes that generated MeHg(I) and that these processes were linked to dissolved organic matter (DOM) and solar radiation. A 2-day diurnal pattern of MeHg(I) in surface water with corresponding bottled controls was assessed for two contrasting lakes in Kejimikujik, Nova Scotia, Canada. Following this study, a tangential ultrafiltrator was used to size-fractionate and generate a concentration gradient of DOM from four different lakes located near Lac Berthelot, Quebec, Canada. The watersheds of two of these lakes were not substantially logged whereas the other two had been extensively logged. Different size fractions of DOM as well as different concentrations of DOM were exposed to sunlight for varying periods of time. We observed that, in Keiimikujik, the concentration of MeHg(I) in surface waters peaked in the early afternoon. Furthermore, this also occurred in bottled water for one of the lakes, Puzzle, eliminating the possibility that in-lake mixing played a role in this pattern. The formation of MeHg(I) was found to be dependent on the size fraction and amount of DOM present in the water. Specifically, DOM less than 5 kDa or between 30 and 300 kDa generated MeHg(I) when exposed to sunlight, but larger fractions did not. Furthermore, although data are limited, we found that water from lakes with logged watersheds generated MeHg(I) when exposed to sunlight, whereas water from lakes with low levels of logging in the undisturbed watersheds did not. Our results demonstrate that MeHg(I) can be formed in freshwaters of certain lakes in response to solar radiation. This photoproduction of MeHg(I) is dependent on DOM concentrations and type, with the importance of water chemistry not yet clear. The significance of this process to freshwater lakes and the mechanism responsible for MeHg(I) photoproduction is still unclear, but a correction in the conventional wisdom that MeHg(I) is rapidly photodegraded is timely.