Drivers of Mercury in Top-predator Lake Fish from Southwest Alaska Parklands

Krista K. Bartz, Tammy L. Wilson, and Daniel B. Young, National Park Service
Ryan F. Lepak, Jacob M. Ogorek, and David P. Krabbenhoft, US Geological Survey

Lake trout collected for mercury analysis from Katmai National Park and Preserve.

NPS/EVAN BOOHER

The Southwest Alaska Network (SWAN) is part of the National Park Service (NPS) Inventory & Monitoring Division, which was created in 1998 to quantify the status and trends of select indicators or “vital signs.” One of the vital signs monitored by SWAN is resident lake fish (i.e., non-migratory freshwater species). Resident lake fish are key indicators of ecosystem health due to their continuous exposure to lake waters, which integrate conditions from the surrounding land and air (Schindler 2009). For this reason, and also because resident lake fish are important for subsistence and recreational fishing in Alaska, it is crucial to better understand their contamination levels and sources.

From 2005 to 2015, NPS researchers sampled more than 300 resident lake fish for contaminant analyses. These fish included 9 species from 13 lakes in 2 southwest Alaska parks. Results indicated that some samples had elevated concentrations of mercury, the majority of which was methylmercury, a potent neurotoxin and endocrine disrupter. Compared with resident lake fish sampled in 21 parks from 10 monitoring networks in the western United States, mercury concentrations in fish from southwest Alaska were among the highest, although considerable variation existed among lakes (Eagles-Smith et al. 2014). Why do these fish—that inhabit some of the most remote and supposedly pristine waters in North America—have such elevated mercury levels? Answering this question requires an understanding of mercury cycling, or the processes by which mercury moves through the environment (Figure 1).

Figure 1. Simplified diagram of mercury cycling, highlighting processes of interest in southwest Alaska. Letters correspond to factors listed in Table 1. Arrow widths do not correspond to mercury load.

Art credit: Tracey Saxby and Dylan Taillie, IAN Image Library

Mercury is a toxic element with no known essential biological function. It occurs naturally as a solid in various minerals, and as a gas from volcanic eruptions and human activities, particularly mining and coal burning. Gaseous elemental mercury can be transported atmospherically to areas far from original sources, after which it can be oxidized and deposited on the earth’s surface in the form of rain or snow. Under anaerobic conditions, typical of wetland sediments, bacteria can convert this deposited inorganic mercury to organic methylmercury. Methylmercury may then accrue in aquatic organisms through the intake of both water and food. In fish, methylmercury readily crosses biological membranes, excretes slowly relative to its rate of uptake, and accumulates to concentrations vastly exceeding those in surface waters (Wiener et al. 2003).

Factors driving mercury concentrations in fish can be grouped into four broad categories: loading, methylation, bioaccumulation, and biomagnification (Table 1). Loading, or the amount of mercury entering an ecosystem, quantifies the sources of mercury, which can be distant or local. Methylation, or the process by which mercury is converted to methylmercury, is determined by local ecological conditions. Bioaccumulation, the increase in methylmercury over time in an organism, and biomagnification, the increase in methylmercury at higher levels of the food web, are driven by local fish biology and food web differences, respectively. Based on our understanding of mercury cycling, we hypothesize that at least seven key factors could be at play in southwest Alaska lakes (Table 1).

Table 1. Distant and local factors hypothesized to drive fish mercury concentrations in SWAN parklands. The first five factors (A-E) are also depicted in Figure 1.

Scale	Category	Factor	Explanation
Distant	Loading	Atmospheric transport (A)	Mercury-rich emissions from coal-fired power production are transported and deposited to distant habitats.
Local	Loading	Underlying geology (B)	Bedrock and surface soils may be enriched in mercury where natural deposits or active volcanoes exist.
Local	Loading	Melting glaciers (C)	Glaciers contain latent reservoirs of atmospherically deposited mercury, as well as entrained geologic materials.
Local	Loading	Spawning salmon (D)	Adult salmon import not only marine-derived nutrients but also bio-available mercury to lake food webs.
Local	Methylation	Wetland cover (E)	Wetlands provide favorable conditions for methylation and yield compounds that increase mercury bioavailability.
Local	Methylation	Water quality	Water quality conditions, such as pH and turbidity, are believed to control the bioavailability of inorganic mercury.
Local	Biomagnification	Fish age and size	Mercury absorption into tissues outpaces removal, so mercury levels increase over time in individual fish that are older and larger.
Local	Biomagnification	Fish trophic position	Piscivorous top predators tend to have elevated mercury because concentrations increase toward the top of the food web.

Figure 2. Map of 13 study lakes spanning 2 parks: Lake Clark National Park and Preserve (LACL) and Katmai National Park and Preserve (KATM), shown in green. Study lakes are shown in blue with red numbers indicating lake names:

(1) Telaquana, (2) Turquoise, (3) Snipe, (4) Lachbuna, (5) Crescent, (6) Kijik, (7) Kontrashibuna, (8) Kukaklek, (9) Nonvianuk, (10) Kulik, (11) Hammersly, (12) Grosvenor, and (13) Brooks.

To discern the relative importance of these factors, we partnered with the US Geological Survey (USGS) Mercury Research Laboratory in Wisconsin. The ensuing study focused on a long-lived piscivorous species (lake trout; Salvelinus namaycush) in two parks (Katmai and Lake Clark national parks and preserves). The first two years of the study involved sampling water, sediment, plankton, and fish—both lake trout and sockeye salmon (Oncorhynchus nerka). Samples were collected from 13 lakes with a range of glacier, wetland, and salmon influences (Figure 2). They were then analyzed for mercury, methylmercury, and other analytes including stable isotopes of mercury, carbon, and nitrogen. Our goal was to answer two main questions: what drives the relatively high mercury levels observed in resident lake fish from southwest Alaska and what controls among-lake differences in those mercury levels?

Figure 3. Boxplots of mercury in lake trout and sockeye salmon, organized geographically from top (north) to bottom (south) in the diagram. The midline of each box signifies the median value of 10 lake trout or 3 sockeye salmon per lake.

The gray line represents the upper limit for unlimited human consumption of Alaska-caught fish, specifically for women of childbearing age, nursing mothers, and children under the age of 12 (Hamade 2014).

Preliminary results indicate that: (1) lake trout exhibit a wide range of total mercury levels, both among parks and among lakes; (2) the median value of lake trout total mercury in 6 of 13 lakes is above the State of Alaska’s guidance for unlimited fish consumption by at-risk groups; and (3) by comparison, sockeye salmon total mercury levels are consistently low and exhibit little variability (Figure 3). Ongoing work involves developing quantitative models to relate mercury in lake trout with potential drivers. These models include variables measured at both the “fish-level” (i.e., variables recorded once per individual fish, such as age) and the “lake-level” (i.e., variables recorded once per lake, such as wetland cover in the watershed). The fish-level variables that best explain lake trout mercury concentrations are age, body condition, and diet specialization. Specifically, mercury tends to be higher in older, skinnier lake trout that feed offshore. The lake-level variables, either individually or in combination, do not improve the models’ abilities to explain lake trout mercury concentrations. Notably absent in our results to date is a significant effect of sockeye salmon on lake trout mercury concentrations. Hence, sockeye salmon are unlikely to be a major driver of differences in mercury among these lakes.

Results from this study are being summarized in two complementary manuscripts. One, led by USGS, focuses on the mercury stable isotopes; the other, led by SWAN, describes the quantitative models. Both will be submitted to peer-reviewed journals in 2020 or 2021 (for updates, see Freshwater Contaminants). Also in 2020, we plan to launch a related project covering five parks in Alaska. Like the earlier study, this project will use lake trout as a focal species, but will also include associated prey species and primary producers to assess the bioaccumulation of mercury through the food web and to determine whether specific energy pathways (e.g., offshore vs. nearshore) promote greater buildup.

Cite this article as:

Bartz, K. K., T. L. Wilson, D. B. Young, R. F. Lepak, J. M. Ogorek, and D. P. Krabbenhoft. 2020. Drivers of mercury in top-predator lake fish from southwest Alaska parklands. Alaska Park Science 19(1): 80-85.

References

Eagles-Smith, C. A., J. J. Willacker, and C. M. Flanagan Pritz. 2014.
Mercury in fishes from 21 national parks in the Western United States: Inter- and intra-park variation in concentrations and ecological risk. US Geological Survey Open-File Report 2014-1051, Reston, Virginia.

Hamade, A. K. 2014.
Fish consumption advice for Alaskans: A risk management strategy to optimize the public’s health. State of Alaska, Section of Epidemiology. Anchorage, Alaska.

Schindler, D. W. 2009.
Lakes as sentinels and integrators for the effects of climate change on watersheds, airsheds, and landscapes. Limnology and Oceanography 54(6): 2349-2358.

Wiener, J. G., D. P. Krabbenhoft, G. H. Heinz, and A. M. Scheuhammer. 2003.
Ecotoxicology of mercury. Pages 409-463 in D. J. Hoffman, B. A. Rattner, G. A. Burton, Jr., and J. Cairns, Jr., editors. Handbook of ecotoxicology, 2^nd edition. CRC Press, Boca Raton, Florida.