US Geological Survey


Date of this Version



Published in U.S. Geological Survey Professional Paper 1717, 1-35, (2007)


Reconnaissance investigations have been conducted to identify how geochemical techniques can be applied to biological studies to assist wildlife management in and near Yellowstone National Park (the Park). Many elements (for example, As, B, Be, Ce, Cl, Cs, F, Hg, K, Li, Mo, Rb, S, Sb, Si, and W) are commonly enriched in (1) thermal waters in the Yellowstone area, (2) rocks altered by these waters, (3) sinter and travertine deposits, and (4) soils and stream sediments derived from these rocks. Some of these elements, such as As, F, Hg, and Mo, may be toxic to wildlife and could be passed up the food chain to many species of animals.

Three investigations are described here. The first discusses the abundance and distribution of selected elements in the scat (feces) of bison (Bison bison), elk (Cervus elaphus), and moose (Alces alces) collected in and near the Park from areas underlain by both unaltered and hydrothermally altered rock. As compared to mean values for stream-sediment analyses, those of scat analyses collected in the Yellowstone area show relatively high concentrations for 12 elements. This suite of elements comprises (1) hydrothermally related elements (As, Br, Cs, Mo, Sb, and W), (2) essential major elements for plants (Ca and K) and some trace elements (Ba, Rb, and Sr) that commonly proxy (substitute) for Ca or K, and (3) zinc. The behavior of zinc is not understood. It is an essential element for plants and animals but does not normally proxy for either Ca or K. Zinc is also not related to hydrothermal activity. This unique behavior of zinc is discussed in other parts of this investigation.

Six elements (Cr, Hg, Ni, Pb, Se, and U) that can be toxic to wildlife are present in low concentrations in scat, reflecting their generally low concentrations in rock and stream-sediment samples collected throughout the Park.

The chemistry of large-animal scat provides information on the feeding habits of large animals in the Park. Scat chemistry shows a high spatial correlation with fossil or active thermal areas or with areas immediately downstream from thermal areas. The longer that animals forage in these localities, the more likely it is that they may ingest significant amounts of potentially toxic elements such as arsenic.

A second investigation describes the concentration levels of hydrothermal mercury and other elements in cutthroat trout (Oncorhynchus clarki bouvieri) and lake trout (Salvelinus namaycush). These elements are derived from sublacustrine hot springs and their habitats in Yellowstone Lake, and this study demonstrates that mercury can be used as a tracer in animal ecology studies. Mercury concentrations are significant in the muscle (average 0.9 ppm, dry weight for both) and liver (average cutthroat = 1.6 ppm, dry weight; average lake trout = 2.1 ppm) of cutthroat and lake trout populations. The mercury levels in fish are believed to be related to mercury introduced to the lake by sublacustrine hot springs, which have dissolved mercury concentrations of as much as 0.170 ppb. Methylation of mercury in thermal waters is probably carried out by methanogenic or sulfate-reducing bacteria that live around sublacustrine hot springs and are consumed by crustaceans such as amphipods, which are a major food source for the cutthroat trout. The mercury levels in the cutthroat trout are transferred to lake trout and to land animals that eat trout. For example, hair of grizzly bears that have been collected near Yellowstone Lake have high mercury levels (0.6–1.7 ppm, dry weight), whereas hair of bears sampled at more remote areas in the greater Yellowstone ecosystem have low mercury contents (0.006–0.09 ppm, dry weight). This observation provides strong evidence that mercury in grizzly bears is derived from feeding on spawning cutthroat trout in the spring and early summer. Studies of mercury and metal contents in other grizzly bear food sources (plants and animals) show that only cutthroat trout are strongly enriched in mercury. These data can potentially be used to quantify the percentage of the bear population that eats cutthroat trout and to determine how far individual bears travel to Yellowstone Lake to eat them.

A third investigation describes carbon-, nitrogen-, and sulfur-isotope compositions in grizzly bears and in some of their foods and describes how these data can be applied to studies of grizzly bear demographics. δ13C values in the Yellowstone ecosystem range from –21.7 ‰ to –30.4 ‰, a range that reflects the influence of C3 plants on the carbon reservoir and probably the effect of elevation on physiological processes. δ15N values range from –2.3 ‰ to 11.0 ‰ and show classical trophic enrichments with respect to most grizzly bear food sources. Cutthroat trout δ15N values (8.3±1.0 ‰) may reflect the importance of sublacustrine hydrothermal springs to the food chain in Yellowstone Lake. Lake trout have even larger δ15N values (11.0±0.4 ‰) that are consistent with their feeding on cutthroat trout. Grizzly bear δ15N values range from 7.0 ‰ to 8.8 ‰. Although grizzly bears are known to eat cutthroat trout, trophic enrichment in δ15N above values found in trout is not apparent in analyses of bear hair. This discrepancy occurs because δ15N values are averaged over one year and include the significantly lower δ15N values of vegetable food sources consumed by bears while their hair is growing. δ34S values in the ecosystem range from –3.1 ‰ to 11.1 ‰. δ34S values of fish (1.2±0.5 ‰) are nearly the same as those in sulfate from thermal springs. Vegetation (clover, cow parsnip, and spring beauty), ungulates (deer, elk, and bison), and moths show a greater range of δ34S values (–3.3 ‰ to 3.2 ‰). However, bears show higher δ34S values (3.2 ‰ to 5.4 ‰ in muscle and 6.1 ‰ to 8.7 ‰ in hair) that are consistent with the consumption of whitebark pine nuts (δ34S = 8.3 ‰ to 11.4 ‰). δ34S values in bears and their food sources seem to be constrained by the major sources of sulfate and sulfide sulfur in the igneous and sedimentary rocks that underlie much of the Park. The large δ34S values found in bear tissues are consistent with the documented fact that most grizzly bears eat substantial amounts of whitebark pine nuts when available. This consumption occurs during hair growth in the fall, thus providing an isotopic marker that may be useful in quantifying nut consumption in individual bears.

These three studies show some different ways that geochemical techniques can be applied to biologic issues. The results suggest that integration of geochemistry into specific biologic studies may help address issues of interest to wildlife managers in Yellowstone National Park and the greater Yellowstone ecosystem.