Mercury Cycling in Northern Alaskan Tundra Ecosystems
AuthorOlson, Christine L.
Civil and Environmental Engineering
AltmetricsView Usage Statistics
Atmospheric mercury (Hg), emitted from both natural and anthropogenic sources, has a long atmospheric lifetime of 6-24 months, which allows it to be transported over great distances to remote regions such as the Arctic where it deposits. The Arctic in modern times has experienced increased atmospheric Hg inputs and increased methylmercury production compared to pre-industrialized periods, with significant consequences for humans and wildlife living in this region. Mercury concentrations in Arctic sediments show an enrichment of about 3-5 times since the industrial revolution. Currently, little information is known about Hg cycling and storage in Arctic upland tundra ecosystems that represent a large receptor area for atmospheric Hg deposition due to their large surface area. These upland tundra ecosystems serve as main sources of Hg to Arctic lakes, rivers, and the Arctic Ocean. The tundra biome covers about 6% of the land surface area globally, but Hg dynamics, concentrations, origins, and mass balances are not well quantified. This research aimed to improve our understanding of terrestrial Hg cycling in the Arctic tundra in northern Alaska. I selected three main topical research areas of Hg cycling that address different comportments of tundra ecosystems: Hg dynamics in Arctic plants, Hg dynamics in tundra soils, and to understand the potential mobilization of Hg in soils to the aqueous phase. Specifically, the objectives of this study were to (1) understand how Hg was distributed spatially across different tundra ecosystems and areas of Alaska, in particular in soils and plants, to determine areas most affected by atmospheric deposition; (2) determine origins of Hg accumulation across this remote area where almost no Hg concentration and mass data is available; (3) quantify the pools of Hg stored in tundra ecosystems to understand if tundra ecosystems serve as important repositories on a global scale, potentially storing large amounts of current and past atmospheric Hg pollution that has been transported to the Arctic and removed from active cycling; and (4) study the link between soil Hg concentrations and Hg in aqueous phase (soil solution and adjacent streams) to assess if and how runoff is linked to upland soil concentrations and understand potential controls for transport and mobilization of this pollutant, particularly as perturbations in climate and permafrost conditions may affect mobilization of current Hg pools in tundra soils.The main study site was located at the Toolik Field Station (TFS), approximately 200 km inland from the Arctic Ocean coast. Additional sampling sites were located along a large latitudinal transect extending from TFS to Prudhoe Bay nearby the Arctic ocean consisting of seven additional sampling sites, near Denali National Park in central Alaska, and at the Noatak National Preserve in northwestern Alaska.Results of this study showed that (1) tundra soils contain high soil Hg concentrations and pool sizes across the remote Alaskan tundra, with soil Hg concentrations having an average of 89 µg kg-1 and a range of 30 to 226 µg kg-1. Stable Hg isotope analysis and geochemical elemental ratios showed that soil Hg was largely derived from atmospheric deposition of elemental Hg (Hg(0)), with additional sources of divalent Hg and little indication for photochemical Hg reduction and reemission of Hg in plants and soils. A mass balance estimation of observed soil Hg pools suggest that deposition of atmospheric Hg must have occurred over millennia, and that this long-term deposition along with little photochemical reemissions allowed soil sequestration of up to 400 Gg of Hg in northern tundra and boreal soils. (2) Surprisingly high vegetation Hg concentrations and standing biomass Hg pools, similar to levels observed in temperate sites, were observed in tundra ecosystems, in spite of short Arctic growing seasons, which I expected to limit vegetation uptake of atmospheric Hg. Across nine sites, bulk vegetation Hg concentrations in the tundra averaged 46±7 µg kg-1, with spatial differences across sites. Different functional plant types strongly differed in tissue Hg concentrations, with highest concentrations observed in feather mosses (on average 58±6 µg kg-1), and brown and white lichen (41±2 µg kg-1 and 34±2 µg kg-1, respectively), showing Hg concentrations three to six times those of vascular plant tissues (e.g., 8±1 µg kg-1 in dwarf birch leaves and 9±1 µg kg-1 in tussock grass). The high representation of nonvascular vegetation in standing aboveground biomass resulted in substantial Hg mass contained in tundra aboveground vegetation (29 µg m-2), which subsequently can be transferred to tundra soils via plant senescence and litterfall contributing to atmospheric deposition. (3) In-site field measurements of Hg in Arctic waterways show a gradient in the order of lakes (1.2±0.1 ng L-1) < rivers (1.6±0.1 ng L-1) < in situ field soil solution (2.8±0.4 ng L-1) < wetland (4.4±0.6 ng L-1) samples. Hg and dissolved organic matter (DOM) were not well correlated at the local, watershed scale, however, scaling up these data points with field data from other Arctic studies greatly improved the linear correlation between Hg and DOM (P=0.00; r2=0.94). Experimental procedures to determine aqueous-phase mobility of dissolved Hg in active-layer and permafrost cores showed low mobilization (average: 2.7 ± 0.4 ng L-1) accounting for 0.04 ± 0.01 % of total soil Hg. Dissolved Hg (DHg) concentrations were two times higher in active-layer soils (3.2 ± 0.5 ng L-1, or 0.04 ± 0.01 % of total soil Hg) compared to permafrost (2.0 ± 0.7, or 0.02 ± 0.01% of total soil Hg). DHg concentrations followed the order of total soil Hg concentrations, with highest DHg concentrations in A-horizons (4.8 ± 2.1 ng L-1) followed by O-horizons (4.0 ± 1.5 ng L-1), B-horizons (2.8 ± 0.4 ng L-1), and permafrost (0.9 ± 0.3 ng L-1). While no individual factor explained samples with higher DHg concentrations, soil samples with higher mobilization of Hg generally exhibited low concentrations of soil C, high Hg/C ratios, and generally higher concentrations of Fe, Al, and Mn.