Trace element concentrations and gastrointestinal parasites of Arctic terns breeding in the Canadian High Arctic

Science of the Total Environment 476–477 (2014) 308–316

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Trace element concentrations and gastrointestinal parasites of Arctic terns breeding in the Canadian High Arctic J.F. Provencher a,⁎, B.M. Braune b, H.G. Gilchrist b, M.R. Forbes a, M.L. Mallory c a b c

Department of Biology, Carleton University, Ottawa, Ontario K1S 5B6, Canada Environment Canada, Science and Technology Branch, Raven Road, Carleton University, Ottawa, Ontario K1S 5 B6, Canada Biology Department, Acadia University, 33 Westwood Avenue, Wolfville, Nova Scotia B4P 2R6, Canada

H I G H L I G H T S • • • •

Bismuth, selenium, and mercury in Arctic terns were high compared with other published values. Selenium, mercury and arsenic concentrations varied across the time periods. Selenium concentrations were significantly associated with the presence of gut parasites. High bismuth concentrations were associated with the absence of gut parasites.

a r t i c l e

i n f o

Article history: Received 4 October 2013 Received in revised form 3 January 2014 Accepted 6 January 2014 Available online 25 January 2014 Keywords: Terns Mercury Arctic Parasites Selenium Breeding stage

a b s t r a c t Baseline data on trace element concentrations are lacking for many species of Arctic marine birds. We measured essential and non-essential element concentrations in Arctic tern (Sterna paradisaea) liver tissue and brain tissue (mercury only) from Canada's High Arctic, and recorded the presence/absence of gastrointestinal parasites during four different phases of the breeding season. Arctic terns from northern Canada had similar trace element concentrations to other seabird species feeding at the same trophic level in the same region. Concentrations of bismuth, selenium, lead and mercury in Arctic terns were high compared to published threshold values for birds. Selenium and mercury concentrations were also higher in Arctic terns from northern Canada than bird species sampled in other Arctic areas. Selenium, mercury and arsenic concentrations varied across the time periods examined, suggesting potential regional differences in the exposure of biota to these elements. For unknown reasons, selenium concentrations were significantly higher in birds with gastrointestinal parasites as compared to those without parasites, while bismuth concentrations were higher in Arctic terns not infected with gastrointestinal parasites. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Anthropogenic activities release numerous environmental contaminants, including trace elements, which can lead to an increase in environmental levels as compared with naturally occurring concentrations. This includes elements such as mercury (Hg) and lead (Pb), which are known to be toxic to biota (Franson and Pain, 2011; Weiner et al., 2003). Once metals and trace elements are in the environment they have the potential to bioaccumulate and biomagnify leading to high levels in top predators (Atwell et al., 1998; Campbell et al., 2005), which might negatively affect wildlife and their populations (Dietz et al., 2013; Sonne-Hansen et al., 2002; Sonne, 2010).

⁎ Corresponding author. Tel.: +1 613 998 8433. E-mail address: [email protected] (J.F. Provencher). 0048-9697/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.scitotenv.2014.01.016

Trace elements in biota are classified as either essential or nonessential elements. Essential elements are those required for metabolic processes, such as calcium (Ca) and copper (Cu), while non-essential elements have no known organismal function, such as Hg (GarciaBarrera et al., 2012; Hoffman et al., 2003). All elements are naturally occurring in varying amounts in the environment. Although biota is naturally exposed to many elements, all elements may become toxic at high enough levels, even those that are required by the body (Garcia-Barrera et al., 2012; Puls, 1994). Importantly, trace elements can accumulate in tissues that have different turnover rates (Hobson and Clark, 1992), thus examination of trace elements in multiple tissues is important when assessing possible exposure patterns and effects. Additionally, the levels of some elements may vary by sex, so it is important to consider possible gender influences on trace element concentrations in tissues (Robinson et al., 2011). Tracking of trace elements in wildlife is important to understand how elements may cycle through biota and ecosystems, as well as

J.F. Provencher et al. / Science of the Total Environment 476–477 (2014) 308–316

how elements may accumulate within wildlife to potentially toxic levels (Metcheva et al., 2010). Although the tissue concentrations of essential trace elements have been intensively studied for some groups of animals (e.g. livestock; Puls, 1994), less is known about levels in wildlife populations, especially those occurring in remote areas such as polar regions (Metcheva et al., 2010). In the Arctic, seabirds are top predators that have been useful in the study of trace elements in wildlife (Braune and Scheuhammer, 2008; Burger et al., 2007). Like most top predators, seabirds are exposed to trace elements through their diet, and high concentrations of trace elements may negatively influence wildlife health (Braune et al., 2006; Braune et al., 2012; Mallory et al., 2006). Although few studies exist that define what level of trace elements are safe for wild birds, some field studies have linked high concentrations of trace elements in wildlife with poor body condition (Wayland et al., 2002), decreased immune system function (Wayland et al., 2002), decreased survival (Wayland et al., 2008), and increased parasite burdens (Sagerup et al., 2009b). In addition, studies have documented some trace elements in apparently healthy breeding wild birds (e.g. Franson et al., 2004; Savinov et al., 2003). Although any single element may affect biota, there are many interactions that may occur within the body between trace elements. Some elements have negative impacts on biota (e.g. Hg), but there are also combinations of elements which may reduce the toxicity of single elements (Shore et al., 2011). Alternatively, some mixtures of trace elements can have a stronger negative effect than those that occur independently (Sarigiannis and Hansen, 2012). Therefore, it is important to also consider trace elements in relation to one another. Trace elements may also interact with other biological factors such as parasites to influence host health. Possible interactions between trace elements and parasites, and how these interactions may affect wildlife, are unclear as studies examining the interactions have shown that cumulative effects can be additive, synergistic or even antagonistic in both controlled laboratory and field experiments (Bergey et al., 2002; Bustnes et al., 2006; Coors and Du Meester, 2008; McFarland et al., 2012). Interactions between trace elements and parasites are of particular concern in the Arctic as the exposure of wildlife to both is expected to change in the coming decades (Davidson et al., 2011; Munthe et al., 2011). The Arctic tern (Sterna paradisaea) is a small seabird (approximately 112 g) which breeds in the circumpolar Arctic (Hatch, 2002). Arctic terns have the longest known migration of any animal, as they travel between the Arctic and the Antarctic each year (Egevang et al., 2010), spending two to three months in their Arctic breeding range (Hatch, 2002). Although Arctic terns are not of conservation concern globally (Hatch, 2002), declines of breeding Arctic terns in northern Canada (Gilchrist and Robertson, 1999), as well as in Greenland have been reported (Burnham et al., 2005; Egevang and Fredericksen, 2011). Importantly, the long migration route of Arctic terns ensures that this species crosses multiple marine ecosystems during migration, thus possibly exposing them to multiple sources of anthropogenic pollution. In this study, we examine levels of trace elements and endoparasite burdens in Arctic terns breeding in the Canadian High Arctic. In general, we expected Arctic terns to have comparable trace element concentrations as other seabirds feeding at a similar trophic level in the same region of Arctic Canada. Using Arctic terns as our study species, our specific goals were to: a) determine the liver tissue concentrations of trace elements and compare values to other seabirds breeding in northern Canada and the circum-Arctic region; b) examine how hepatic trace element concentrations relate to breeding stage and sex; c) examine the relationship between Hg concentrations in the liver tissue and brain tissue; d) examine the relationship between hepatic Hg and selenium (Se); and, e) determine whether incidence of gastrointestinal parasites was related to trace element concentrations or whether either were affected by season and sex.

309

2. Methods 2.1. Field collections Arctic terns were collected from Nasaruvaalik Island (75°49′N, 96°18′W), just north of Cornwallis Island, Nunavut (Fig. 1; Permits WL000952, NUN-SCI-06-01). Each bird was shot with a 12 gauge shotgun, using #6 steel bird shot. Immediately upon collection birds were examined for ectoparasites (but none were found) and bagged individually (to preserve any missed ectoparasites) and frozen within 2 h of collection. Adult Arctic terns were collected at four different breeding stages in 2007; pre-breeding (27–28 June; n = 11), early incubation (7 July; n = 10), late incubation (17–19 July; n = 10), and chick-rearing (4 August; n = 10). Collections were done across the breeding season to determine if breeding stage influenced either trace element concentrations or parasitism as has been found in other birds (Spakulova et al., 1991; Wayland et al., 2005). 2.2. Bird dissections Arctic tern carcasses were shipped frozen to the Long Point Waterfowl and Wetlands Research Fund Avian Energetics Laboratory, where they were thawed and dissected using standard protocols (e.g. Mallory and Forbes, 2008). Each bird was visually examined for ectoparasites and here again, none were found. They were then weighed, measured for head width, and dissected using acetone/hexane washed instruments and vessels. The brain and liver tissues were removed, and the gastrointestinal tract was examined for the presence of macroscopic endoparasitic helminths. Known tissue turnover rates suggest that the hepatic values

Fig. 1. Map of the Arctic Archipelago indicating the location of Nasaruvaalik Island, and the Northwater Polynya where seabirds have been sampled in 2007 for trace elements in the High Arctic.

310

J.F. Provencher et al. / Science of the Total Environment 476–477 (2014) 308–316

Table 1 Trace elements detected in Arctic terns collected from a High Arctic colony (n = 41). Geometric means, 95% confidence interval, minimum detected levels and maximum detected levels are presented, along with the geometric mean in wet weights are given. The number of birds with levels above the detection limit is represented by n. NA indicates values not available due to limited sample sizes or results that were below the detection limits, and dw indicates where values are given in dry weight. Arctic tern levels reported are compared with existing published reports⁎. n

Detection limit (μg/g dw, unless otherwise stated)

Geometric mean (95% CI) μg/g dw

Minimum value (μg/g dw)

Maximum value (μg/g dw)

Geometric mean (95% CI) μg/g wet weight

Levels compared to reported levels

Essential elements Aluminum (Al) Calcium (Ca) Cobalt (Co) Copper (Cu) Chromium (Cr) Iron (Fe) Magnesium (Mg) Manganese (Mn) Molybdenum (Mo) Nickel (Ni) Potassium (K) Selenium (Se) Sodium (Na) Vanadium (V) Zinc (Zn)

41 41 6 41 35 41 41 41 41 37 41 41 41 0 41

1.9094 0.3014 0.0827 0.3234 0.2020 0.0169 0.1401 0.0065 0.0604 0.1413 2.9165 0.5908 1.2233 0.2724 0.0230

137 (1.60) 442 (3.32) NA 24.9 (0.05) 0.79 (0.01) 791 (2.22) 817 (0.84) 11.5 (0.03) 2.66 (0.01) 0.61 (0.01) 9511 (5.44) 15.77 (0.04) 4158 (4.13) NA 119 (0.17)

35 181 0.10 14.1 0.49 406 678 5.1 1.5 0.20 8776 10.0 3278 NA 84.79

777 1833 0.23 45.9 1.56 1459 1032 16.9 3.9 2.08 11481 26.7 5070 NA 157.00

44.6 (0.5) 143 (1.03) NA 8.09 (0.02) 0.26 (0.01) 257 (0.78) 265 (0.28) 3.74 (0.01) 0.86 (0.01) 0.19 (0.01) 3088 (1.97) 5.12 (0.01) 1350 (1.33) NA 38.8 (0.06)

Above normald Above normald Normald Normald Normald Normald

Non-essential elements Arsenic (As) Barium (Ba) Beryllium (Be) Bismuth (Bi) Boron (B) Cadmium (Cd) Lead (Pb) Strontium (Sr) Titanium (Ti) Total mercury (Hg) in the liver Total mercury (Hg) in the brain

36 20 0 41 41 41 4 41 1 41 41

0.3850 0.0467 0.0057 0.8748 0.1098 0.0716 0.27 0.0253 0.2031 0.0002 0.00016 mg/kg ww

2.72 (0.01) 0.15 (0.01) NA 8.30 (0.02) 1.42 (0.01) 14.4 (0.05) NA 1.11 (0.02) NA 4.03 (0.01) 0.32 (0.01)

1.03 0.09 NA 4.11 0.51 7.65 1.00 0.22 0.28 2.14 0.63

6.38 0.31 NA 15.7 3.12 25.52 86.29 11.69 0.28 6.33 1.81

0.88 (0.01) 0.04 (0.01) NA 2.69 (0.01) 0.46 (0.01) 4.68 (0.02) NA 0.36 (0.01) NA 1.31 (0.01) 0.98 (0.01)

Below Toxicd

Normald Above Normald Above normald Highb,c,d

Normald

Highd Normald Above normald,f Higha,d

Above normale Highe

⁎ (aFranson and Pain, 2011; bHeinz, 1996; cOhlendorf and Heinz, 2011; dPuls, 1994; eShore et al., 2011; fWayland and Scheuhammer, 2011).

we report reflect exposure levels during the week before collection, while the brain tissue likely sample reflects a longer period of up to several weeks before the collection (Bauchinger and McWilliams, 2009; Hobson and Clark, 1992). A small (~2 g) piece of breast muscle was also removed for stable isotope analysis. 2.3. Laboratory analysis Frozen liver samples were sent to the University of Windsor (Great Lakes Institute for Environmental Research) laboratory for analysis of 25 trace elements (Table 1; also provides detection limits). The brain tissue was only analyzed for Hg. Prior to trace element analyses, tissues were homogenized. All samples were analyzed according to CALA-accredited procedures (EnvironmentCanada, 1989). An approximate 2 g sample of the liver was digested in 5 mL of 1:1 H2SO4:HNO3 acid, purified, and then trace elements were analyzed by Inductively Coupled Plasma Optical Emission Spectrophotometry (IRIS #701776, Thermo Jarrell Ash Corporation), and the sample response compared against that generated for a standard calibration curve. Total Hg was determined from the liver tissue and brain tissue samples digested in 2:1 H2SO4:HNO3 acid, purified, and then analyzed by atomic absorption spectrometry–cold vapor generation using an Atomic Absorption Spectrophotometer (300; Varian) and a vapor generation accessory unit (VGA-76; Varian). Quality assurance/quality control (QA/QC) procedures included analysis of three method blanks (purified water), three certified biological reference tissues (DORM2, LUTS-1 and DOLT-2; National Research Council, Canada) and two randomly selected duplicate samples per batch of 25 samples. All QA/QC measures were in compliance with the normal laboratory operating procedures at the time of analysis. All trace element concentrations are presented in dry weight μg/g, unless otherwise indicated for comparison purposes (Table 1).

Breast muscle samples were shipped to the Nature Laboratory at the University of New Brunswick, where they were dried, ground into a homogenous powder using a mortar and pestle and weighed (~1 mg) into tin cups for stable isotope analysis. Samples were combusted at 1800 °C in a Robo-Prep elemental analyzer, and gasses were sent to a continuous flow isotope ratio mass spectrometer (CFIRMS) with one laboratory standard run for every 10 samples analyzed. Stable nitrogen and stable carbon abundance were expressed in δ notation as the deviation from standards in parts per thousand (‰) according to the following equation: δX ¼

h


 i Rsample =Rstandard −1  1000

where δX is the isotope of interest (in this case, δ15N or δ13C), R is the ratio of the abundance of the heavy to the light isotope (15N/14N or 13C/12C), with Rsample being the ratio within the given sample, and Rstandard the ratio of heavy to light isotope within the international standard (Vienna Pee Dee Belemnite (V-PDB) scale for carbon and atmospheric nitrogen). Analytical precision, as determined by replicate measurements of the internal laboratory standards, was ±0.2‰. 2.4. Statistical analysis Trace elements were treated as essential and non-essential elements. Essential elements examined included aluminum (Al), Ca, cobalt (Co), Cu, chromium (Cr), iron (Fe), magnesium (Mg) manganese (Mn), molybdenum (Mo), nickel (Ni), potassium (K), Se, sodium (Na), and zinc (Zn). Non-essential elements examined were arsenic (As), boron (B), barium (Ba), beryllium (Be), bismuth (Bi), cadmium (Cd), Pb, Hg, strontium (Sr), titanium (Ti) and vanadium (V) (Garcia-Barrera et al., 2012). The geometric mean for all trace elements is presented, along

J.F. Provencher et al. / Science of the Total Environment 476–477 (2014) 308–316

with the 95% confidence interval, the minimum and maximum values. When values less than the detection limit occurred, values were set at 10% below the detection limit to complete the above calculations (i.e. all geometric means presented are n = 41). In the case where less than 75% (n N 30) of the individuals were above the non-detectable limit, only minimum and maximum values are given. All statistical analyses were completed using R statistical software (R-Development-Core-Team, 2008). We used a general linear model (GLM) approach to examine all data with the level of significance set at 0.05. In some cases, hepatic trace element results were reported as below the detection limit. As in the above statistics when values less than the detection limit occurred, values were set at 10% below the detection limit to complete the statistical analysis which required numeric values. Elements were only included in multivariate and univariate analysis when more than 75% of the individuals were above the non-detectable limit. Levene's test was used to check for homogeneity of variance and the Shapiro–Wilk test was used to test the data for a normal distribution. Elements that did not show equal variances or normal distributions were log-transformed to achieve these assumptions for multivariate analysis. When assumptions were still violated, we compared levels in different groups using non-parametric, Kruskal– Wallis sum rank tests. Arctic tern sex and size were used as a possible co-variate in multivariate and univariate statistical analysis. Size was calculated as mass divided by the total head length as a size–mass corrected value. For the multivariate analysis of trace elements, we excluded six elements with b30 individuals above the detection limits (Ba, Be, Co, Pb, Ti, V). Of the remaining elements several did not meet the assumptions for parametric tests and were log transformed (Al, B, Cd, Cu, Ni, Sr and the brain Hg). Transformations did not normalize the distributions for Ca, Cr, K, Mg and Se, thus these were not used in the multivariate analysis. This left 14 trace elements (Al, As, B, Bi, Cd, Cu, Fe, Hg, Mn, Mo, Na, Ni, Sr, and Zn) to be examined for patterns across all individuals sampled. A multivariate analysis was used to examine the eight essential element concentrations (Al, Cu, Fe, Mn, Mo, Ni, Na and Zn). A separate multivariate analysis was used to examine a subset of the concentrations of three non-essential elements in relation to each other (As, B, Bi, Cd, Hg and Sr). For both the essential and non-essential multivariate analyses we used a principle component analysis (PCA) to first examine the data for a smaller number of trace elements that represented most of the variation observed. Based on the PCA results, those elements that contributed the most variability, Fe and Na for the essential elements and As, Bi, Hg and Sr for nonessential elements, were included in a multivariate MANOVA to examine how trace element concentrations varied with breeding stage and sex. Any significant results were then examined using a general linear model (GLM) to investigate the relationships between those single element values and independent variables (sex and breeding stage). In addition we also investigated how Se varied with breeding stage and sex using a non-parametric analysis (Kruskal– Wallis) due to Se being of specific concern in marine birds (Ohlendorf and Heinz, 2011). To investigate elemental interactions between Hg and Se, a GLM was also used to compare the molar ratio. All trace element concentrations were inspected to examine any relationships between element pairings using a correlation matrix and Spearman rank values. Correlations with values greater than 0.75 were examined using GLM when data met the parametric assumptions, or with a Poisson generalized linear model (GLzM) when data were not normally distributed. We compared Hg and Se hepatic concentrations specifically due to the physiological link between these elements (Dietz et al., 2000). We used the above described MANOVA-GLM approach to test for relationships between the presence or absence of endoparasites in Arctic terns in relation to essential and non-essential element concentrations. When an element did not meet the assumptions of the parametric GLM a Kruskal–Wallis test was used (e.g. Se). An unpaired t-test, Welch

311

corrected for small sample sizes, was used to compare our results to published findings. 3. Results We collected 41 Arctic terns (22 females, 19 males) with an overall mean mass of 106 ± 7 g (SD), a slightly lower mean than reported for breeding birds from maritime Canada and western Russia (Hatch, 2002). Male terns (103 ± 5 g) had a lower body mass than females (108 ± 8 g) at each breeding stage, and the relationship was significant across the breeding season (GLM; F1,39 = 6.79, p = 0.013). No significant difference was found for mass among breeding stages, even when sex was considered (p N 0.05). Male terns had similar mean δ15N (12.93 ± 0.19SE, n = 19) and δ13C (− 18.08 ± 0.09) as female terns (δ15N — 13.08 ± 0.15SE, n = 21; δ13C — −18.00 ± 0.08; t-tests, both p = 0.52). 3.1. Essential elements in tern hepatic tissue Trace elements were above detection limits in all samples except for Co, Cr, Ni, As, Ba, Be, Pb, Ti and V (Table 1). Of the 14 essential elements examined, hepatic tissue levels of all except Se were ≤“normal” (below the high threshold) reported for poultry, ducks and waterfowl (Puls, 1994; see Table 1). Only Se levels (geometric mean 15.6 μg/g dw in females, n = 22) were above the range considered to cause impairment to egg-laying females in some species (N 10 μg/g dw), but within levels considered background for some marine birds (20–75 μg/g dw; Ohlendorf and Heinz, 2011). Arctic terns from northern Canada had significantly higher concentrations of Se compared with Arctic terns in the Barents Sea [mean = 7.3 ± 0.67 μg/g dw (Savinov et al., 2003); Welch's approximate t41 = 11.72, p b 0.0001]. 3.2. Non-essential elements in terns Of the 11 non-essential elements examined, concentrations of hepatic Bi, Hg, Pb and Cd, and the brain Hg were at levels considered high for birds (Table 1Franson and Pain, 2011; Puls, 1994; Shore et al., 2011; Wayland and Scheuhammer, 2011). 3.3. Multivariate analysis of essential elements PCA analysis found that 99.8% of the variation in the eight essential elements examined was driven by Fe and Na. When Fe and Na were examined together in a MANOVA with both size and sex as predictor variables, only sex was significant, but it co-varied with size, thus only sex was considered further. Both sex and breeding stage had significant effects on Fe and Na concentrations in terns (MANOVA; Sex F2 = 3.79, p = 0.033; breeding stage F6 = 2.27, p = 0.046), with no significant interaction effect found (p N 0.16). When examined individually, neither Na nor Fe showed significant patterns when sex and breeding stage were included (GLM; p N 0.05). In a separate analysis, pre-breeding terns had higher hepatic Se (χ23 = 10.49, p = 0.01) than the other stages (pre-breeding — 20.78 ± 5.24 μg/g; all other stages combined mean 14.68 ± 2.85 μg/g; Fig 2A). 3.4. Multivariate analysis of non-essential elements When As, B, Bi, Cd, Hg and Sr were examined using a PCA, 97% of the variation came from the distribution of As, Bi, Hg and Sr; these were included in the subsequent MANOVA. Both breeding stage and sex had a significant influence on hepatic As, Bi, Hg and Sr concentrations when considered together (MANOVA; breeding stage F4 = 4.33, p = 0.007; MANOVA; sex F12 = 2.93, p = 0.002). When hepatic As was examined hepatic concentrations differed significantly when both breeding stage and sex were both considered (GLM; F7,33 = 2.76, p = 0.02). Male

312

J.F. Provencher et al. / Science of the Total Environment 476–477 (2014) 308–316

terns had significantly higher levels of As than females (p = 0.04; females — 2.8 ± 1.5 μg/g; males — 3.4 ± 1.6 μg/g; Fig. 3C). In addition, hepatic concentrations of As were significantly higher in both sexes during late incubation as compared to the other breeding stage (p = 0.03; late incubation — 4.0 ± 1.6 μg/g; all other stage combined mean 2.7 ± 1.4 μg/g; Fig. 3A). Both Bi and Sr did not vary significantly with sex or breeding stage when considered independently (GLM; p N 0.05). Male Arctic terns had significantly higher concentrations of hepatic Hg compared to females (GLM; F7,33 = 2.66, p = 0.026; males — 4.56 ± 0.89 μg/g, females — 3.77 ± 0.957 μg/g; Fig. 3D), and in both sexes, birds in the early incubation period had significantly lower concentrations than those from the other time periods (p = 0.04; Fig. 3B). Hepatic total Hg was also highly correlated with concentrations of Hg in the brain tissue (GLM; F1,39 = 22.17, p b 0.0001). However, unlike hepatic Hg, there was little variation in the brain total Hg across the breeding season or between the sexes (GLM; p = 0.15).

3.5. Elemental interactions The molar mass of Hg and Se were not significantly correlated across individual terns (GLM; p = 0.92). However, terns with higher hepatic Bi also had higher levels of Fe (GLM; F1,39 = 651.8, p b 0.0001; Fig. 4B). Similarly, terns with higher concentrations of Ca had higher levels of Sr (GLzM; Wald χ2 = 60.5, p b 0.0001). All other Spearman rank correlation values for the remaining trace element combinations were b0.75.

3.6. Parasites No ectoparasites were found on any of the Arctic terns examined. Seven of 38 (18%) of terns had unidentified cestodes, with the proportions of terns infested with cestodes declining through the breeding season (60% of pre-breeders, 12% of early incubators, 0% of late incubation birds and 0% of chick-rearing adults). In general, levels of essential and non-essential elements in terns with and without gastrointestinal parasites were similar (all p N 0.05). However, birds infected with parasites had significantly higher hepatic Se (21.0 ± 4.2 μg/g) compared with uninfected birds (15.4 ± 4.1 μg/g; χ21 =8.67, p = 0.003; Fig. 3B). Similarly, terns that did not have gastrointestinal parasites had higher concentrations of Bi (8.8 ± 2.4 μg/g) than infected terns (7.0 ± 2.2 μg/g; GLM, F1,39 = 3.19, p = 0.055; Fig. 4A).

4. Discussion 4.1. Trace element concentrations in Arctic terns Marine bird tissues often contain relatively higher levels of trace elements than terrestrial or freshwater species (Furness and Camphuysen, 1997), and our findings support this observation for many trace elements. It is important to note that threshold value for trace element concentrations in wild birds is an area of study with considerable debate in the published literature. Many threshold values are based on domestic aquatic species leaving many knowledge gaps for many species, and especially for marine birds (i.e. Puls, 1994; Shore et al., 2011). With these limitations in mind we present our findings for trace element concentrations in Arctic terns from the Canadian High Arctic, and compare concentrations to what is known in other species. Trace element concentrations in hepatic tissue of Arctic terns were at or above the normal range for most essential elements examined as compared with reported levels for birds, including marine birds where data were available. Selenium concentrations were the only exception. Arctic terns had liver Se concentrations that were above typical background levels, and above levels known to cause reproductive impairment in egg-laying females in some sensitive species (N 10 μg/g dw; Heinz, 1996; Ohlendorf and Heinz, 2011). However, in pre-breeding Arctic terns, tissue concentrations of Se (range = 13–27 μg/g ww) were within levels reported as toxic for some species, although a range of sensitivities to Se concentrations is recognized within wild birds (Heinz, 1996; Ohlendorf and Heinz, 2011). Importantly, our results show that concentrations in a single element may change from potentially toxic levels to non-toxic levels rapidly, so timing of sampling is an important component in any interpretation. This finding indicates that the breeding stage of migratory birds may be an important factor when considering levels of trace elements as either a risk to wildlife health, or as part of a monitoring program. When the non-essential elements were examined, most hepatic tissue concentrations in Arctic terns were within normal reported ranges for waterfowl and poultry, but three elements (Bi, Pb and Hg) were at high or potentially toxic levels. Hepatic Bi concentrations were high (Puls, 1994). Bismuth enters the environment as a refining by-product, but also as bird shot for hunting (though stainless steel was used for this study), which can cause inflated hepatic Bi concentrations when embedded in tissue (Sanderson et al., 1998; Tsuji et al., 2004). Terns are not hunted and presumably would spend relatively

Fig. 2. Hepatic selenium levels (μg/g dry weight) in Arctic terns in 2007 from the Canadian High Arctic. Concentrations are presented across the breeding season (A) and individuals infected with gastro-intestinal parasites and uninfected (B). Stars indicate significant differences in either stages in the breeding season that have as compared to some other stages, or differences between birds infected or uninfected with gastro-intestinal parasites.

J.F. Provencher et al. / Science of the Total Environment 476–477 (2014) 308–316

313

Fig. 3. Hepatic concentrations of arsenic and mercury (μg/g dry weight) in Arctic terns in 2007 from the Canadian High Arctic. Concentrations are presented across the breeding season (A and B) and between females and males (C and D). Stars indicate significant differences in either stages in the breeding season that have as compared to some other stages, or differences between concentrations between the sexes. See text for explanation of differences.

little time in areas near hunting activity (Egevang et al., 2010), so the source of high Bi merits further investigation. Only four birds had Pb levels over the detection limit. One apparently healthy, breeding tern was attending a nest and yet had a hepatic Pb concentration which was a magnitude of order higher than reported levels which induced toxicity (Franson and Pain, 2011). Consequently,

like Bi, the sources of Pb and lethality of this element to terns merits additional study. The liver Hg concentrations in Arctic terns were below levels known to cause impairment in waterfowl (Puls, 1994). However, this is the first study to examine levels of Hg in the brain tissue of an Arctic marine bird in Canada, and the results merit concern. The brain Hg concentrations

Fig. 4. Hepatic Bi concentrations (μg/g dry weight) in Arctic terns from the Canadian High Arctic, in relation to gastro-intestinal parasites (A) and Fe (B).

314

J.F. Provencher et al. / Science of the Total Environment 476–477 (2014) 308–316

were within the range considered high (Puls, 1994), and with a geometric mean concentration of 0.32 μg/g ww tern, the brain Hg concentrations fall within the range where neurochemical effects of Hg have been observed in marine wildlife (i.e. polar bears; Dietz et al., 2013). Since the brain tissue levels are more indicative of long-term exposure, these levels may indicate that Arctic terns are exposed to higher levels of Hg than previously thought. Additionally, overall tern exposure to Hg may be at levels known to affect bird health and reproduction even though the liver Hg concentrations are not within levels of known concern. We also found that hepatic Hg concentrations increased through the breeding season, indicating that Arctic terns may have their highest Hg exposure in their Arctic breeding grounds. Similarly, common terns (Sterna hirundo) in southern Canada had higher Hg concentrations in tissues grown on the breeding grounds as compared with tissues grown on the wintering areas (Baird, 2013). Differential Hg exposure between regions is known to occur (Aubail et al., 2011), and our results suggest that terns may be exposed to the highest levels of Hg exposure on their high north Canadian breeding grounds. As Hg can have a negative effect on reproduction (Dietz et al., 2013), and is predicted to continue increasing in the coming decades (Jaffe et al., 2005), this non-essential trace element may pose high risk to Arctic marine birds in the future, specifically in the high north. 4.2. Comparisons to seabirds in Canada and other regions In general, Arctic tern levels of As, Cd and Hg appear to be similar to other species feeding at the same trophic level in the Canadian High Arctic. Although Arctic tern eggs in Arctic Canada have been studied for Hg (Akearok et al., 2010), adult terns have not been previously examined for trace elements, although there are studies reporting on levels in Arctic tern from the Barents Sea (Savinov et al., 2003) and Northern Siberia (Kim et al., 1996). Levels of Cd, Cu and Zn in terns were similar across these regions, while differences between the regions are seen in other elements (Table 2). Canadian Arctic terns had lower concentrations of As and significantly higher concentrations of Se than birds reported for the Barents Sea (Savinov et al., 2003). Arctic terns from northern Canada also had similar Hg concentrations as terns in Northern Siberia, but higher levels than those sampled in the Barents Sea. These patterns of Hg concentrations most likely reflect differences in Hg deposition across the Arctic region. Interestingly, Hg concentrations in Arctic terns at a colony in southern Canada were among the highest of the marine birds examined (Bond and Diamond, 2009). Taken with our results and compared to other Arctic marine birds (Braune and Scheuhammer, 2008), it appears that terns may be particularly susceptible to accumulating high Hg levels in their tissues. Consequently, this species may warrant special concern as a bioindicator of Hg pollution in the coming decades. Currently, Arctic terns from the High Arctic have hepatic Hg burdens similar to breeding common terns where reproduction was found to be unhindered (Hoffman et al., 2011). This suggests that Arctic terns from northern Canada are currently not experiencing reductions in reproductive output due to Hg exposure. Importantly, common terns did experience impaired reproduction associated with methylmercury exposure when average hepatic Hg levels were 3.65 μg/g ww (Hoffman et al.,

2011). Although levels in Arctic terns in this study were below this level, it highlights the potential for terns to be exposed to levels that affect bird health and ultimately reproductive output under increasing Hg levels in the Canadian Arctic (Braune, 2007). 4.3. Trace element interactions When the molar ratio between Hg and Se was examined, no significant pattern was evident; similar to what has been observed in other bird species (Ohlendorf, 2003). Although correlations between Hg and Se have been found more consistently in mammals, our results agree with those of Ohlendorf and Heinz (2011) that no clear pattern between these two elements exists in birds. However, a strong positive relationship between Bi and Fe was found (Fig 4). The sources of Bi and Fe are potentially linked in the environment (Tsuji et al., 2004), or they may be linked physiologically, or both. 4.4. Trace elements related to breeding season and sex Although little variation was observed in most essential elements across the breeding season or between the sexes, we did find patterns in As, Hg and Se (Figs. 2 and 3). Hepatic Hg and As were higher later in the breeding season, suggesting that dietary exposure to these elements is likely higher on the breeding grounds as compared to other areas used by the terns throughout the year. This pattern supports other findings that some Arctic regions accumulate relatively high levels of Hg (Borga et al., 2006). Arsenic and Hg were both significantly lower in females as compared to males, which suggests that female Arctic terns may be depurating these elements into their eggs, a pattern observed in some other marine birds (Wayland et al., 2005). Sexbiased differences in trace elements are unlikely due to dietary differences between males and females, because both sexes had nearly identical stable isotope values, suggesting that they were consuming similar prey (Robinson et al., 2011). In contrast, Se showed a different pattern with levels in both males and females decreasing over the breeding season, which has also been observed in Forster's tern (Sterna forsteri) in California (Ackerman and Eagles-Smith, 2009). The decline of Se in both males and females suggests that egg formation does not play a large part in how Se is expelled from the body. In contrast, Wayland et al. (2005) found no change in Se levels in northern common eider ducks (Somateria mollissima borealis) between pre-nesting birds and nesting birds. This lack of difference in Se concentrations across the breeding season in eiders may be attributed to eiders having a relatively short migration between the Arctic and the North Atlantic as compared with the Arctic tern migration that spans multiple oceans and continents (Egevang et al., 2010), where trace elements are potentially acquired. The declining trend in Se concentrations indicates that Arctic terns may be potentially exposed to lower levels of Se in their local breeding environment compared to their wintering grounds or on their migration routes. 4.5. Trace elements and endoparasites Along with a decreasing pattern over the breeding season, Se levels were significantly higher in terns infected with gastrointestinal parasites,

Table 2 Reported trace element concentrations in Arctic terns (as determined from the liver tissue) from the circumpolar region by Kim et al. (1996)ŧ and Savinov et al. (2003)⁎. Location Franz Josef Land⁎ ŧ

Year

Figure

As

Cd

Cu

Hg

Se

Zn

1992

Mean SD Mean SD Mean SD

5.7 2.2 NA

16.3 4.7 15.3 7.7 15.2 4.9

28.4 3.0 25.1 5.7 25.5 5.5

1.1 0.2 4.8 3.7 4.2 1.0

7.3 0.7 NA

130.0 14.9 82.6 20.6 120.0 17.0

Northern Siberia

1993

Nasaruvaalik Island (this study)

2007

3.1 1.5

16.3 4.5

J.F. Provencher et al. / Science of the Total Environment 476–477 (2014) 308–316

which could be caused by a variety of interactions. First, Se and endoparasites both decreased over the breeding season, thus the relationship between Se and endoparasites could be simply correlative, where the birds are exposed to both higher Se levels and higher parasite encounter rates in their wintering areas or migration routes, and then shed both while breeding in the Arctic. Alternatively, birds with high levels of Se may be more susceptible to accumulating endoparasites when they are encountered, or worse at shedding endoparasites as compared with their counterparts with lower levels of Se. A positive relationship between Se levels and cestodes was also found in glaucous gulls (Larus hyperboreus; same sub-order of birds) from Svalbard (Sagerup et al., 2009a), suggesting that the relationship between Se and endoparasites in larids might be more common than currently reported. High levels of Se, coupled with changes across the breeding season and its relationship with parasites suggest that Se may play an important role in Arctic tern biology. Although we found relationships between Se, Bi, and parasites, we acknowledge that more detailed information is required on parasite incidence and abundance in order to test relationships more rigorously. The general lack of parasites on Arctic terns supports the idea that high latitude hosts have fewer parasites (Dobson et al., 2008). It also appears that Arctic terns in the High Arctic may gradually lose their gastrointestinal parasites over the breeding season, although more study is required for confirmation. This trend may be caused by changes in their diet during their annual cycle. For example, Arctic tern parasite communities may be found in geographic areas visited during migration and absent near the breeding colonies. Another possible explanation is that those birds that carry heavy parasite burdens do not complete migration or may not attempt to breed if parasite prevalence or intensity is high (Poulin, 2007). Overall, our novel baseline data on Arctic terns from a remote colony in the Canadian High Arctic indicate that terns have high levels of several non-essential trace elements, but reproductive effort and success at this colony remains high (M. L. Mallory, unpubl. data). Although few parasites were found in the terns examined, terns with endoparasites had higher hepatic concentrations of both Se and Bi. There is great potential among Arctic marine birds to test questions regarding how parasites and contaminants interact in the wild. We recommend that future studies should focus on species that have an abundant parasite fauna, as well as a well-known contaminant profile. Acknowledgments We thank Kelly Boadway, Josh Boadway and Al Fontaine for assistance with field work. We also thank Kerrie Wilcox (Long Point Bird Observatory) for conducting dissections, and Ken Drouillard (University of Windsor) for completing the chemical analyses. Field work and collections were completed under permits issued by the Government of Nunavut, Environment Canada, and Aboriginal Affairs and Northern Development Canada. Financial support was provided by Environment Canada (CWS, S&T), the Natural Resources Canada (PCSP), and the Natural Resources and Engineering Research Council (NSERC to JFP and MLM). References Ackerman JT, Eagles-Smith CA. Selenium bioaccumulation and body condition in shorebirds and terns breeding in San Francisco Bay, California, USA. Environ Toxicol Chem 2009;28:2134–41. Akearok JA, Hebert CE, Braune BM, Mallory ML. Inter- and intraclutch variation in egg mercury levels in marine bird species from the Canadian Arctic. Sci Total Environ 2010;408:836–40. Atwell L, Hobson KA, Welch HE. Biomagnification and bioaccumulation of mercury in an Arctic marine food web: insights from stable nitrogen isotope analysis. Can J Fish Aquat Sci 1998;55:1114–21. Aubail A, Teilmann J, Dietz R, Riget F, Harkonen T, Karlsson O, et al. Investigation of mercury concentrations in fur of phocid seals using stable isotopes as tracers of trophic levels and geographical regions. Polar Biol 2011;34:1411–20.

315

Baird C. Mercury and stable isotopes in common terns (Sterna hirundo) from the St. Lawrence River: a comparison between breeding and winter habitats. Kingston, Canada: Department of Biology. MSc. Queen's University; 201382. Bauchinger U, McWilliams S. Carbon turnover in tissues of a passerine bird: allometry, isotopic clocks, and phenotypic flexibility in organ size. Physiol Biochem Zool 2009;82:787–97. Bergey L, Weis J, Weis P. Mercury uptake by the estuarine species Palaemonetes pugio and Fundulus heteroclitus compared with their parasites, Probopyrus pandalicola and Eustrongylides sp. Mar Pollut Bull 2002;44:1046–50. Bond AL, Diamond AW. Total and methyl mercury concentrations in seabird feathers and eggs. Arch Environ Contam Toxicol 2009;56:286–91. Borga K, Campbell L, Gabrielsen GW, Norstrom RJ, Muir DCG, Fisk AT. Regional and species specific bioaccumulation of major and trace elements in Arctic seabirds. Environ Toxicol Chem 2006;25:2927–36. Braune BM. Temporal trends of organochlorines and mercury in seabird eggs from the Canadian Arctic, 1975–2003. Environ Pollut 2007;148:599–613. Braune BM, Mallory ML, Gilchrist HG. Elevated mercury levels in a declining population of ivory gulls in the Canadian Arctic. Mar Pollut Bull 2006;52:978–82. Braune BM, Scheuhammer AM. Trace element and metallothionein concentrations in seabirds from the Canadian Arctic. Environ Toxicol Chem 2008;27:645–51. Braune BM, Scheuhammer AM, Crump D, Jones S, Porter E, Bond D. Toxicity of methylmercury injected into eggs of thick-billed murres and Arctic terns. Ecotoxicology 2012;21:2143–52. Burger J, Gochfeld M, Sullivan K, Irons D. Mercury, arsenic, cadmium, chromium lead, and selenium in feathers of pigeon guillemots (Cepphus columba) from Prince William Sound and the Aleutian Islands of Alaska. Sci Total Environ 2007;387:175–84. Burnham W, Burnham KK, Cade TJ. Past and present assessments of bird life in Uummannaq District, West Greenland. Dan Ornithol Foren Tidsskr 2005;99:196–208. Bustnes JO, Erikstad KE, Hanssen SA, Tveraa T, Folstad I, Skaare JU. Anti-parasite treatment removes negative effects of environmental pollutants on reproduction in an Arctic seabird. Proc R Soc B Biol Sci 2006;273:3117–22. Campbell LM, Norstrom R, Hobson KA, Muir D, Backus S, Fisk A. Mercury and other trace elements in a pelagic Arctic marine food web (Northwater Polynya, Baffin Bay). Sci Total Environ 2005;351–352:247–63. Coors A, Du Meester L. Synergistic, antagonistic and additive effects of multiple stressors: predation threat, parasitism and pesticide in Daphnia magna. J Appl Ecol 2008;45: 1820–8. Davidson R, Simard M, Kutz SJ, Kapel CMO, Hamnes IS, Robertson LJ. Arctic parasitology: why should we care? Trends Parasitol 2011;27:238–44. Dietz R, Riget F, Born EW. An assessment of selenium to mercury in Greenland marine animals. Sci Total Environ 2000;245:15–24. Dietz R, Sonne C, Basu N, Braune B, O'Hara T, Letcher RJ, et al. What are the toxicological effects of mercury in Arctic biota? Sci Total Environ 2013;443:775–90. Dobson A, Lafferty KD, Kuris AM, Hechinger RF, Jetz W. Homage to Linnaeus: how many parasites? How many hosts? Proc Natl Acad Sci U S A 2008;105:11482–9. Egevang C, Fredericksen M. Fluctuating breeding of Arctic terns (Sterna paradisaea) in Arctic and high-Arctic colonies in Greenland. Waterbirds 2011;34:107–11. Egevang C, Stenhouse IJ, Phillips RA, Petersen A, Fox JW, Silk JRD. Tracking of Arctic terns Sterna paradisaea reveals longest animal migration. Proc Natl Acad Sci U S A 2010;107:2078–81. EnvironmentCanada. Analytical methods manual. Ottawa, Canada: Water Quality Branch Environment Canada; 1989. Franson JC, Hollmen TE, Flint PL, Grand JB, Lanctot RB. Contaminants in molting long-tailed ducks and nesting common eiders in the Beaufort Sea. Mar Pollut Bull 2004;48:504–13. Franson JC, Pain DJ. Lead in birds. In: Beyer W, Meador J, editors. Environmental contaminants in biota. New York: CRC Press, Taylor & Francis Publishers; 2011. p. 563–94. Furness RW, Camphuysen KCJ. Seabirds as monitors of the marine environment. ICES J Mar Sci 1997;54:726–37. Garcia-Barrera T, Gomez-Ariza J, Gonzalez-Fernandez M, Morena F, Garcia-Sevillano M, Gomez-Jacinto V. Biological responses related to agonistic, antagonistic and synergistic interactions of chemical species. Anal Bioanal Chem 2012;403:2237–53. Gilchrist HG, Robertson GJ. Population trends of gulls and Arctic terns nesting in the Belcher Islands, Nunavut. Arctic 1999;52:325–31. Hatch J. Arctic tern (Sterna paradisaea). In: Poole A, editor. Birds of North America (online). Ithaca, New York: Cornell Lab of Ornithology; 2002. Heinz G. Selenium in birds. In: Beyer W, Heinz G, Redmon-Norwood A, editors. Environmental contaminants in wildlife: tissues concentrations. CRC, Boca Raton, FL, USA: SETAC Special Publications Series; 1996. p. 447–58. Hobson KA, Clark RG. Assessing avian diets using stable isotopes I: turnover of 13C in tissues. Condor 1992;94:181–8. Hoffman DJ, Eagles-Smith CA, Ackerman JT, Adelsbach TL, Stebbins KR. Oxidative stress response of Forster's terns (Sterna forsteri) and Caspian terns (Hydroprogne caspia) to mercury and selenium bioaccumulation in liver, kidney, and brain. Environ Toxicol Chem 2011;30:920–9. Hoffman DJ, Rattner BA, Burton GAJ, Cairns JJ. Handbook of ecotoxicology. New York: Lewis Publishers; 2003. Jaffe D, Prestbo E, Swartzendruber P, Weiss-Penzias P, Kato S, Takami A, et al. Export of atmospheric mercury from Asia. Atmos Environ 2005;39:3029–38. Kim EY, Ichihashi H, Saeki K, Atrashkevich G, Tanabe S, Tatsukawa R. Metal accumulation in tissues of seabirds from Chaun, Northeast Siberia, Russia. Environ Pollut 1996;92: 247–52. Mallory ML, Braune B, Forbes MR. Contaminant concentrations in breeding and non-breeding northern fulmars (Fulmarus glacialis L.) from the Canadian High Arctic. Chemosphere 2006;64:1541–4.

316

J.F. Provencher et al. / Science of the Total Environment 476–477 (2014) 308–316

Mallory ML, Forbes MR. Costly pre-laying behaviours and physiological expenditures by northern fulmars in the High Arctic. Ecoscience 2008;15:545–54. McFarland CA, Talent LG, Quinn MJ, Bazar MA, Wilbanks MS, Nisanian M, et al. Multiple environmental stressors elicit complex interactive effects in the western fence lizard (Sceloporus occidentalis). Ecotoxicology 2012;21:2372–90. Metcheva R, Yurukova L, Bezrukov V, Beltcheva M, Yankow Y, Dimitriv K. Trace and toxic elements accumulations in food chain representatives at Livingston Island (Antarctica). Int J Biol 2010;2:155160. Munthe J, Goodsite M, Berg T, Chetelat J, Cole A, Dastoor A, et al. Where does the mercury in the Arctic environment come from, and how does it get there? In: Outridge P, Dietz R, editors. AMAP assessment 2011: mercury in the Arctic. Arctic monitoring and assessment program (AMAP). Norway: Oslo; 2011. Ohlendorf H. Ecotoxicology of selenium. In: Hoffman DJ, Rattner BA, Burton GAJ, Cairns JJ, editors. Handbook of ecotoxicology. New York: Lewis Publishers; 2003. Ohlendorf H, Heinz G. Selenium in birds. In: Beyer W, Meador J, editors. New York: CRC Press, Taylor and Francis Group; 2011. p. 669–702. Poulin R. Evolutionary ecology of parasites. Princeton University Press: Princeton; 2007. Puls R. Mineral levels in animal health. Sherpa International: Clearbrook BC; 1994. R-Development-Core-Team. R: a language and environment for statistical computing. Vienna: R Foundation for Statistical Computing; 2008. Robinson SA, Forbes MR, Hebert CE, Scheuhammer AM. Evidence for sex differences in mercury dynamics in double-crested cormorants. Environ Sci Technol 2011;45: 1213–8. Sagerup K, Helgason LB, Polder A, Strom H, Josefsen TD, Skare JU, et al. Persistent organic pollutants and mercury in dead and dying glaucous gulls (Larus hyperboreus) at Bjornoya (Svalbard). Sci Total Environ 2009a;407:6009–16. Sagerup K, Savinov V, Savinova T, Kuklin V, Muir DCG, Gabrielsen GW. Persistent organic pollutants, heavy metals and parasites in the glaucous gull (Larus hyperboreus) on Spitsbergen. Environ Pollut 2009b;157:2282–90. Sanderson G, Anderson W, Foley G, Havera S, Skowron L, Brawn J, et al. Effects of lead, iron, and bismuth alloy shot embedded in the breast muscle of game-farm mallards. J Wildl Dis 1998;34:688–97. Sarigiannis DA, Hansen U. Considering the cumulative risk of mixtures of chemicals — a challenge for policy makers. Environ Heal 2012;11.

Savinov VM, Gabrielsen GW, Savinova TN. Cadmium, zinc, copper, arsenic, selenium and mercury in seabirds from the Barents Sea: levels, inter-specific and geographical differences. Sci Total Environ 2003;306:133–58. Shore R, Pereira M, Walker L, Thompson D. Mercury in non-marine birds and mammals. In: Beyer W, Meador J, editors. Environmental contaminants in biota. New York: CRC Press, Taylor and Francis Group; 2011. p. 609–26. Sonne-Hansen C, Dietz R, Leifsson P, Hyldstrup L, Riget FF. Cadmium toxicity to ringed seals (Phoca hispida): an epidemiological study of possible cadmium-induced nephropathy and osteodystrophy in ringed seals (Phoca hispida) from Qaanaaq in Northwest Greenland. Sci Total Environ 2002;295:167–81. Sonne C. Health effects from long-range transported contaminants in Arctic top predators: an integrated review based on studies of polar bears and relevant model species. Environ Int 2010;36:461–91. Spakulova M, Birove V, Macko J. Seasonal changes in the species composition of nematodes and acanthocephalans of ducks in East Slovakia. Biol (Bratislava) 1991;46:119–28. Tsuji L, Wainman B, Jayasinghe R, Karagatzides J, Van Spronsen E, Nieboer E. Tissue-bismuth levels of game birds harvested with bismuth shotshell: policy implications. Bull Environ Contam Toxicol 2004;72:128–34. Wayland M, Drake KL, Alisauskas RT, Kellett DK, Traylor J, Swoboda C, et al. Survival rates and blood metal concentrations in two species of free-ranging North American sea ducks. Environ Toxicol Chem 2008;27:698–704. Wayland M, Gilchrist HG, Marchant T, Keating J, Smits JE. Immune function, stress response, and body condition in arctic-breeding common eiders in relation to cadmium, mercury, and selenium concentrations. Environ Res 2002;90:47–60. Wayland M, Gilchrist HG, Neugebauer E. Concentrations of cadmium, mercury and selenium in common eider ducks in the eastern Canadian arctic: influence of reproductive stage. Sci Total Environ 2005;351:323–32. Wayland M, Scheuhammer AM. Cadmium in birds. In: Beyer W, Meador J, editors. Environmental contaminants in biota. New York: CRC Press, Taylor and Francis Group; 2011. p. 645–66. Weiner J, Krabbenhoft D, Heinz G, Scheuhammer AM. Ecotoxicology of mercury. In: Hoffman DJ, Rattner BA, Burton GAJ, Cairns JJ, editors. Handbook of ecotoxicology. New York: CRC Press; 2003. p. 409–63.