Spatial variation of mercury bioaccumulation in bats of ...

Science of the Total Environment 626 (2018) 668–677

Contents lists available at ScienceDirect

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

Spatial variation of mercury bioaccumulation in bats of Canada linked to atmospheric mercury deposition John Chételat a,⁎, M. Brian C. Hickey b, Alexandre J. Poulain c, Ashu Dastoor d, Andrei Ryjkov d, Donald McAlpine e, Karen Vanderwolf e,f, Thomas S. Jung g, Lesley Hale h, Emma L.L. Cooke a,1, Dave Hobson i, Kristin Jonasson j, Laura Kaupas k, Sara McCarthy l, Christine McClelland a, Derek Morningstar m, Kaleigh J.O. Norquay n, Richard Novy o,2, Delanie Player p, Tony Redford q, Anouk Simard r, Samantha Stamler s, Quinn M.R. Webber n,3, Emmanuel Yumvihoze c, Michelle Zanuttig a a

Environment and Climate Change Canada, National Wildlife Research Centre, Ottawa, Ontario K1A 0H3, Canada River Institute, Cornwall, Ontario K6H 4Z1, Canada Biology Department, Faculty of Science, University of Ottawa, Ottawa, Ontario K1N 6N5, Canada d Air Quality Research Division, Environment and Climate Change Canada, Dorval, Quebec H9P 1J3, Canada e New Brunswick Museum, Saint John, New Brunswick E2K 1E5, Canada f Canadian Wildlife Federation, Kanata, Ontario K2M 2W1, Canada g Yukon Department of Environment, Whitehorse, Yukon Territory Y1A 2C6, Canada h Ontario Ministry of Natural Resources & Forestry, Peterborough, Ontario K9J 8M5, Canada i Alberta Environment and Parks, Edson, Alberta T7E 1T2, Canada j Department of Biology, Western University, London, Ontario N6A 5B7, Canada k Department of Biological Sciences, University of Calgary, Calgary, Alberta T2N 1N4, Canada l Wildlife Division, Fisheries and Land Resources, Goose Bay, Newfoundland and Labrador A0P 1E0, Canada m Myotistar, Cambridge, Ontario N3C 0B4, Canada n Department of Biology, University of Winnipeg, Winnipeg, Manitoba R3B 2G3, Canada o Golder Associates Ltd., Calgary, Alberta T2A 7W5, Canada p Matrix Solutions Inc., Calgary, Alberta T2R 0K1, Canada q Animal Health Centre, BC Ministry of Agriculture, Abbotsford, British Columbia V3G 2M3, Canada r Direction de l'expertise sur la faune terrestre, l'herpétofaune et l'avifaune, Ministère des Forêts, de la Faune et des Parcs, Québec, Quebec G1S 4X4, Canada s Alberta Environment and Parks, Edmonton, Alberta T6H 4P2, Canada b c

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Mercury in bat fur was related to concentrations in brain, liver, and kidney. • Bat species differed in their fur mercury concentrations. • Fur mercury was higher in adult than juvenile bats, but not related to sex. • Mercury in fur of adult little brown bats was higher in eastern Canada. • Atmospheric deposition explained geographic variation of mercury in bats.

⁎ Corresponding author. E-mail address: [email protected] (J. Chételat). Current address: Ocean Sciences, Memorial University, St. John's, Newfoundland and Labrador A1C 5S7, Canada. 2 Current address: Environmental Resources Management, Minneapolis, MN 55402, USA. 3 Current address: Cognitive and Behavioural Ecology Interdisciplinary Program, Memorial University of Newfoundland, St. John's, Newfoundland and Labrador A1B 3X7, Canada. 1

https://doi.org/10.1016/j.scitotenv.2018.01.044 0048-9697/Crown Copyright © 2018 Published by Elsevier B.V. All rights reserved.

J. Chételat et al. / Science of the Total Environment 626 (2018) 668–677

a r t i c l e

i n f o

Article history: Received 8 November 2017 Received in revised form 5 January 2018 Accepted 6 January 2018 Available online xxxx Editor: Mae Sexauer Gustin Keywords: Chiroptera Bioaccumulation Little brown bat Fur Brain

669

a b s t r a c t Wildlife are exposed to neurotoxic mercury at locations distant from anthropogenic emission sources because of long-range atmospheric transport of this metal. In this study, mercury bioaccumulation in insectivorous bat species (Mammalia: Chiroptera) was investigated on a broad geographic scale in Canada. Fur was analyzed (n = 1178) for total mercury from 43 locations spanning 20° latitude and 77° longitude. Total mercury and methylmercury concentrations in fur were positively correlated with concentrations in internal tissues (brain, liver, kidney) for a small subset (n = 21) of little brown bats (Myotis lucifugus) and big brown bats (Eptesicus fuscus), validating the use of fur to indicate internal mercury exposure. Brain methylmercury concentrations were approximately 10% of total mercury concentrations in fur. Three bat species were mainly collected (little brown bats, big brown bats, and northern long-eared bats [M. septentrionalis]), with little brown bats having lower total mercury concentrations in their fur than the other two species at sites where both species were sampled. On average, juvenile bats had lower total mercury concentrations than adults but no differences were found between males and females of a species. Combining our dataset with previously published data for eastern Canada, median total mercury concentrations in fur of little brown bats ranged from 0.88–12.78 μg/g among 11 provinces and territories. Highest concentrations were found in eastern Canada where bats are most endangered from introduced disease. Model estimates of atmospheric mercury deposition indicated that eastern Canada was exposed to greater mercury deposition than central and western sites. Further, mean total mercury concentrations in fur of adult little brown bats were positively correlated with site-specific estimates of atmospheric mercury deposition. This study provides the largest geographic coverage of mercury measurements in bats to date and indicates that atmospheric mercury deposition is important in determining spatial patterns of mercury accumulation in a mammalian species. Crown Copyright © 2018 Published by Elsevier B.V. All rights reserved.

1. Introduction Mercury is a contaminant that can travel long distances through the atmosphere and be deposited onto terrestrial and aquatic ecosystems far from major emission sources (Fitzgerald et al., 1998; Lamborg et al., 2002; Lindberg et al., 2007). Human activities, particularly coal combustion, have released large quantities of mercury into the atmosphere, resulting in an estimated 3.2 fold increase in the atmospheric mercury pool since 1850 (Streets et al., 2017). Following deposition, inorganic mercury is transformed (principally by microbes) into methylmercury, which bioaccumulates in organisms, biomagnifies through food chains, and is toxic at elevated levels of exposure (Scheuhammer et al., 2012). Recognizing the global nature of mercury pollution and the need for international cooperation, the United Nations Environment (UNE) led the establishment of the Minamata Convention on Mercury (entered into force in August 2017) to protect human and environmental health through the reduction of global anthropogenic mercury emissions. The mercury cycle is complex, and many processes contribute to its transport, chemical transformations, and bioaccumulation in the environment (Lehnherr, 2014; Travnikov et al., 2017). Experimental evidence clearly demonstrates that increased mercury loading to aquatic ecosystems results in greater bioaccumulation in food chains (Harris et al., 2007; Orihel et al., 2006), and large-scale spatial variation in mercury concentrations of fish and aquatic invertebrates is related to atmospheric mercury deposition in North America (Hammerschmidt and Fitzgerald, 2005a; Hammerschmidt and Fitzgerald, 2005b). However, ecosystems also vary in their sensitivity to inorganic mercury loading, and local biogeochemical conditions that control methylmercury production and bioavailability can play an important role in determining the extent of methylmercury contamination in biota (Eagles-Smith et al., 2016; Munthe et al., 2007). Recent studies indicate that bats (Mammalia: Chiroptera) in eastern North America can contain elevated levels of methylmercury in their tissues (Little et al., 2015a; Little et al., 2015b; Yates et al., 2014). Greater mercury exposure has been found for bats collected close to contaminated sites with known mercury emissions (Nam et al., 2012; Wada et al., 2010; Yates et al., 2014), and close to remote acidic lakes in which higher methylmercury production has been observed (Little et al., 2015b). Limited information is available, however, on the geographic scope of elevated mercury levels in North American bats and

more broadly on the toxicological risks to bat health (Zukal et al., 2015), although methylmercury is a known neurotoxin that can impair the reproduction, growth, and health of wildlife at elevated environmental exposure (Fuchsman et al., 2017; Scheuhammer et al., 2015). Nam et al. (2012) showed that little brown bats (Myotis lucifugus) at a contaminated site (in the eastern USA) had neurochemical biomarker responses associated with mercury concentrations that exceeded toxicity thresholds for other mammalian species. Big brown bats (Eptesicus fuscus) collected at the same contaminated site also had elevated mercury concentrations, but showed no adrenocortical response compared to bats from a reference area (Wada et al., 2010). Vampire bats in Central America (Belize) with higher total mercury levels were found to have impaired immune function (Becker et al., 2017). Several species of bats have been designated as endangered or threatened by federal authorities in Canada (ECCC, 2017) and the United States (U.S. Fish and Wildlife Service, 2017), including little brown bats and northern long-eared bats (Myotis septentrionalis) which have suffered large population declines from white-nose syndrome in eastern North America (Ingersoll et al., 2016; Vanderwolf et al., 2016). More information is needed to characterize exposure of bats because of the potential for mercury to be an environmental stressor in addition to introduced disease, habitat loss, pesticide contamination, and wind turbine fatalities (Jones et al., 2009). Mammals, including bats, are exposed to methylmercury primarily through their diet (Wiener et al., 2003). Insectivorous bats consume a variety terrestrial and aquatic species, most of which are captured insects in flight. For example, diet characterization using molecular tools has shown that the eastern red bat (Lasiurus borealis) in southern Ontario (Canada) consumed 127 species of insects, mostly terrestrial prey in the lepidoptera order (Clare et al., 2009). Similarly, little brown bats in southern Ontario were found to consume 61 species of insect prey (mostly from aquatic environments) and five species of arachnids (Clare et al., 2011). The consumption of insects with aquatic life stages or spiders that prey on aquatic insects connects bats to aquatic food chains and can enhance their mercury bioaccumulation (Becker et al., 2018). Aerial insectivores have only recently begun to receive more attention in investigations of mercury bioaccumulation (Cristol et al., 2008; Jackson et al., 2011b; Rimmer et al., 2005; Whitney and Cristol, 2018), perhaps in part because they feed at a lower trophic position than predators found at the top of food chains and were assumed to have lower exposure as a result. There is increasing recognition that

670


aerial insectivores may also be at risk of toxicological effects from mercury exposure (Hartman et al., 2013; Jackson et al., 2011a; Nam et al., 2012; Whitney and Cristol, 2018). We investigated mercury bioaccumulation in bats from across Canada, with a focus on endangered little brown bats. The objectives of this study were to: (1) compare total and methylmercury levels in fur and internal tissues from a small subset of bats; (2) identify biological influences on mercury concentrations, specifically inter-species, sex, and age variation; (3) characterize geographic patterns of mercury bioaccumulation in bats across Canada; and (4) examine the role of atmospheric mercury deposition in geographic patterns of mercury in bats. 2. Methods 2.1. Bat tissue collections Dorsal fur is often collected as a non-lethal measure of mercury bioaccumulation in bats (Hickey et al., 2001; Little et al., 2015a; Syaripuddin et al., 2014). The mercury concentration in fur, which is inert, reflects exposure at the time when the hair was grown. A collection of 1178 samples of bat fur was assembled from 43 locations across nine Canadian provinces or territories covering 20° of latitude (42.58° to 62.20° N) and 77° of longitude (60.38 to 137.04° W) (Fig. 1). The dataset was made possible through contributions from bat biologists in government, university, non-governmental, and private sector organizations who conducted their own ecological field programs. The majority of samples were obtained from locations in Quebec, Ontario, Alberta and the Yukon (132–350 samples per province/territory). Smaller sample numbers were obtained from Newfoundland and Labrador, New Brunswick, Manitoba, British Columbia, and the Northwest Territories (6–83 samples per province/territory). Mercury concentrations in fur of little brown bats from three eastern provinces (Prince Edward Island, Nova Scotia, and Newfoundland and Labrador) have been well characterized by Little et al. (2015a, 2015b), and those published data were examined in this manuscript (see Section 2.5). The majority of fur samples in our study were of little brown bats (n = 719) and big brown bats (n = 314), although smaller numbers of northern long-eared bats (n = 70),

silver-haired bats (Lasionycteris noctivagans; n = 72), and hoary bats (Lasiurus cinereus; n = 3) were also obtained. The median sample size for little brown bats and big brown bats was 19 per location (range = 4–70). Sample sizes for each species and location are provided in Supplementary Table S1. The majority of fur samples (approximately two-thirds) were collected from live bats by biologists as part of ecological studies that were conducted between 2012 and 2016. Additional fur samples were obtained from archived (frozen) bat carcasses, the majority of which were from a collection at the New Brunswick Museum (Saint John, New Brunswick). Most archived bats were collected between 2000 and 2011. For live collections, the sampling location was very specific (i.e. a summer roost). For sampling of archived carcasses, the locations were more general (i.e. a town or city). In a few cases (5 sites), the site represented a series of pooled locations within a small geographic range of 20–40 km. In one case (Nanaimo, British Columbia) where there was low sample size for the area, fur samples were pooled over a 135 km range. Dorsal fur was obtained by clipping hair between the scapulae using scissors. Where possible, ancillary information on each bat was recorded: species, collection date and location, sex, age (juvenile or adult), forearm length, and mass (g). Fur samples were stored in clean cryovials or small Whirl-Pak® bags. 2.2. Tissue comparisons of total and methyl mercury Prior to initiating the geographic survey in 2014, a small number of archived bat carcasses from eastern Ontario (Cornwall, 40 km range) were examined to compare total and methylmercury concentrations in fur, liver, kidney, and brain. Those measurements provided background information to assist with interpreting total mercury concentrations in fur for the geographic survey. Eighteen big brown bats and four little brown bats (total n = 22) were dissected and analyzed for mercury at the Laboratory for the Analysis of Natural and Synthetic Environmental Toxins (LANSET) at the University of Ottawa (Ottawa, Ontario). The same tissue sample was analyzed for both total mercury and methylmercury, although for one of the 22 bats, there was insufficient fur sample for analysis. Total mercury was determined on a MA3000

Fig. 1. Locations of collection sites for fur of little brown bats (LBB), big brown bats (BBB) and northern long-eared bats (NLEB) across Canada. Sites from this study are square symbols and sites with published data from Little et al. (2015a) are circles. AB = Alberta, BC = British Columbia, MB = Manitoba, NB = New Brunswick, NL = Newfoundland and Labrador, NS = Nova Scotia, NT = Northwest Territories, ON = Ontario, PE = Prince Edward Island, QC = Quebec, SK = Saskatchewan, YT = Yukon Territory.


mercury analyzer (Nippon Instruments Corporation, College Station, Texas, USA) using the principle of thermal decomposition, gold amalgamation, and atomic absorption. Methylmercury was analyzed by an Agilent 6890 capillary gas chromatography (GC) coupled with a Tekran 2500 (Tekran Instruments Corporation, Toronto, Ontario, Canada) cold vapor atomic fluorescence spectrometer (CV-AFS) as described by Cai et al. (1997). Following initial strong alkaline digestion and acid leaching, samples were subjected to sodium thiosulfate clean-up and organomercury species were isolated as their bromide derivatives by acidic KBr and CuSO4, with subsequent extraction into a small volume of dichloromethane. Total mercury recoveries for duplicate analyses of DORM-4 (fish protein) and DOLT-4 (dogfish liver) reference materials (National Research Council of Canada) were 93 ± 5% and 101 ± 3%, respectively. Methylmercury recoveries for duplicate analyses of DORM-4 and DOLT-4 reference materials were 95 ± 11% and 98 ± 6%, respectively. Brain, liver, and kidney concentrations are presented on a dry weight basis, while fur concentrations are presented on a fresh weight basis.

671

deposition) and direct deposition to the Earth's surface (i.e. dry deposition) were the removal mechanisms employed in the model. Model simulations for the North American domain were conducted for 2010 to 2015 at a 10 km (latitude-longitude) model resolution using spatially-distributed terrestrial and oceanic emissions of mercury to air from contemporary anthropogenic and natural sources as well as legacy mercury deposition source (i.e. reemission of previously deposited mercury). Canadian anthropogenic mercury emissions were based on the 2013 National Pollutant Release Inventory (NPRI, https:// www.ec.gc.ca/inrp-npri/) for major point sources, and 2010 Air Pollutant Emission Inventory (APEI, http://www.ec.gc.ca/Pollution/) for transportation and area source emissions. Anthropogenic mercury emissions in the United States were based on the 2011 National Emissions Inventory (NEI, https://www.epa.gov/air-emissions-inventories/ 2011-national-emissions-inventory-nei-data). Transport of mercury from global sources to the North American domain was driven by a global-scale simulation of GEM-MACH-Hg using a global anthropogenic emission dataset from AMAP for 2010 (AMAP/UNEP, 2013) at the horizontal model resolution of 1° × 1°.

2.3. Total mercury measurements of bat fur Fur from the geographic survey (typically 1–2 mg) was analyzed for total mercury using a Direct Mercury Analyzer 80 (Milestone Inc., Shelton, Connecticut, USA) at the National Wildlife Research Centre (Environment and Climate Change Canada, Ottawa, Ontario). Tests were conducted to determine if a pre-cleaning treatment was required to remove exogenous dirt (and mercury) from the fur. Sixteen fur samples were split in two, and total mercury results for the untreated portion were compared to fur cleaned with ultra-pure water and trace-metal grade acetone. There was variability between treated and untreated samples (RSD = 20 ± 21%, n = 16) although the bias was positive for about half the samples and negative for the other half. On average, there was no significant difference between total mercury concentrations of treated and untreated samples (paired t-test, p = 0.83, n = 32). We concluded that exogenous mercury contributed little to the total concentration, and therefore, fur samples were not pre-cleaned with organic solvent prior to analysis. The same conclusion was reached by Little et al. (2015b) following a similar test. In cases where there was visible dirt or grease on the fur, it was dabbed clean with a Kim Wipe before analysis. Analytical precision was evaluated through replication of every tenth sample with a mean relative standard deviation (RSD) of 10 ± 12% (n = 166). Total mercury recoveries of two certified reference materials from the National Research Council of Canada were 105 ± 3% (n = 105) for TORT-3 (lobster hepatopancreas) and 107 ± 4% (n = 28) for DOLT-4. A few fur samples (n = 11) were below the method detection limit of 0.174 ng and those results were calculated using half the detection limit. Fur total mercury concentrations are presented on a fresh weight basis. Site means of fur total mercury by bat species are provided in Supplemental Table S2. 2.4. Modelling of atmospheric mercury deposition at bat collection sites Spatially-distributed total mercury deposition rates to terrestrial and aquatic ecosystems in Canada were simulated by the mercury version of Environment and Climate Change Canada's (ECCC) operational airquality forecast model GEM-MACH (Global Environmental Multi-scale, Modelling Air quality and CHemistry model), GEM-MACH-Hg. GEMMACH represents and simulates the physicochemical processes of meteorology and photochemistry of the atmosphere; whereas, GEMMACH-Hg, in addition, includes mercury physicochemical processes of three atmospheric mercury species, Hg0, Hg2+ (gas phase), and Hg2+ (particle phase), based on the ECCC's former mercury model, GRAHM (Global/Regional Atmospheric Heavy Metals model; Dastoor et al., 2015; Dastoor et al., 2008; Durnford et al., 2012). Scavenging of gaseous and particulate Hg species into atmospheric precipitation (i.e. wet

2.5. Data analysis We examined the relationship between mercury concentrations in fur with those in internal tissues by correlation analysis. Pearson correlation coefficients were calculated for covariance of fur total mercury and methylmercury concentrations with those in brain, liver, and kidney. The analysis was conducted on non-transformed data and the level of significance was adjusted for the experiment-wise error rate using Holm's procedure. We tested the influence of biological characteristics, specifically whether species, sex, and age class (juvenile versus adult) affected the total mercury concentrations of fur. The dataset was unbalanced with regard to those factors; for many locations only one sex was sampled (more often females than males), and usually only data were available for one species and maturity type (adults more often) from a location. Given the large geographic area sampled and the potential for different mercury exposure among locations, several models were generated that tested subsets of the data where the effects of biological characteristics could be compared within locations. Data for a site were included in the statistical analysis if the site had a minimum of 5 measurements for each element of the factor being tested (e.g., species, sex). Since data from multiple sites were available, linear mixed models were used to test for fixed effects (e.g., species, sex, age class) while controlling for site as a random variable. Linear mixed modelling was conducted with the software R (R Core Team, 2012) using the packages Linear Mixed Effects Models (lme4) (Bates et al., 2017) and Linear and Nonlinear Mixed Effects Models (nlme) (Pinheiro and Bates, 2017). The Multi-Model Inference package (MuMIn) (Barton, 2016) was used to obtain marginal r2 values for the proportion of variation explained by fixed effects in the models. Linear mixed models were tested on data that were pooled across years, assuming minimal inter-annual variation over the study period. Fur total mercury concentrations were log-transformed to improve normality and homoscedasticity of the model residuals. To investigate geographic patterns of fur mercury in little brown bats from across Canada, previously published data from Little et al. (2015a) were included. That dataset was comprised of 334 measurements of total mercury in fur of adult female little brown bats collected between 2001 and 2012 from 25 Atlantic Canada sites: 10 sites in Nova Scotia, 3 sites in Prince Edward Island and 12 sites from Newfoundland and Labrador. The mean concentrations of study sites from Little et al. (2015a) were included in a comparison of total mercury concentrations by province or territory, as well as a linear regression analysis relating atmospheric mercury deposition with site-mean fur total mercury concentrations of adult little brown bats (see Supplemental Table S3 for the dataset).

672


Table 1 Means (±standard deviation) and ranges of total mercury and methylmercury concentrations in tissues of 22 big brown bats and little brown bats. Concentrations are in μg/g on a dry weight basis except for fur which is as fresh weight. Tissue

Fura Liver Kidney Brain a

Total mercury

Methylmercury

Mean ± SD

Range

Mean ± SD

Range

10.61 ± 13.38 4.22 ± 5.63 3.53 ± 2.84 0.95 ± 0.82

1.65–54.23 0.34–21.33 0.59–11.04 0.27–3.71

6.31 ± 5.84 1.39 ± 0.87 1.22 ± 1.01 0.66 ± 0.50

1.18–20.10 0.33–3.13 0.09–4.89 0.13–2.20

n = 21 for fur.

3. Results 3.1. Comparisons of mercury concentrations in bat tissues Total mercury and methylmercury concentrations were several-fold higher in fur than in liver, kidney, and brain (Table 1). Intermediate concentrations of mercury were found in liver and kidney, while brain had the lowest concentrations. On average, the total mercury and methylmercury concentrations in brain were 18 ± 4% and 13 ± 3% (mean ± 1 standard error), respectively, of the total mercury in fur. Most of the total mercury in both brain and fur was in the form of methylmercury (74 ± 4% and 75 ± 4%, respectively), while lower percentages were found for liver (64 ± 7%) and kidney (49 ± 6%). Brain methylmercury concentrations ranged an order of magnitude from 0.13 to 2.20 μg/g among the 22 bats examined in the tissue comparison. Mercury concentrations in fur were a good indicator of mercury accumulation in internal tissues. Both total mercury and methylmercury concentrations in fur were positively correlated with those in brain (Pearson r = 0.60–0.75, Holm's p b 0.05, n = 21; Fig. 2). Methylmercury concentrations in fur were positively correlated with those in kidney (Pearson r = 0.59, Holm's p b 0.05, n = 21) and liver (Pearson r = 0.56, Holm's p b 0.05, n = 21) whereas correlations between total mercury in fur and liver (Pearson r = 0.19, Holm's p N 0.05, n = 21) or fur and kidney (Pearson r = 0.43, Holm's p N 0.05, n = 21) were not significant. Strong positive correlations were also observed among methylmercury concentrations of internal tissues (i.e. between brain, liver, and kidney; Pearson r = 0.74–0.93, Holm's p b 0.001, n = 22). Similar correlation coefficients were obtained when only big brown bats were examined (n = 17 or 18) and four little brown bats were excluded. To further examine potential error associated with estimating brain concentrations using fur, we conducted a linear regression analysis between fur and brain total mercury using our dataset (see Supplemental Table S4 for more detail). The predicted total mercury concentrations of brain in bats with 1, 10, 30 and 50 μg/g of mercury in fur were estimated at 0.50, 0.92, 1.86 and 2.81 μg/g, according to the regression model. The 95% prediction error on the estimates was approximately ±1.25 μg/g.

Thus, while there was a statistically significant tendency for bats with higher mercury in their fur to also contain higher mercury levels in their brain, the regression model was limited in terms of predicting (with confidence) total mercury levels in brain based on fur measurements. Therefore, small differences in fur total mercury concentrations should be interpreted with caution. 3.2. Species variation in fur total mercury Total mercury accumulation differed among species, with higher concentrations in fur of northern long-eared bats and big brown bats compared with little brown bats (Table 2). For the five locations in the dataset where both little brown bats and big brown bats were sampled, big brown bats had (on average) 1.7 ± 0.2 times more total mercury in their fur. Similarly, northern long-eared bats had 1.5 ± 0.2 times more total mercury in their fur compared with little brown bats at five locations where both species were sampled. However, the variance explained by species differences was low (marginal r2 b 0.10) compared with the full model r2 that accounted for location effects (conditional r2 = 0.21–0.42), suggesting that site-specific exposure was more important. Two other species, silver-haired bats and hoary bats, were only sampled at one to three sites, where species-specific mean concentrations of fur total mercury were found to be low (4.63 ± 0.41 μg/g and 1.64 ± 0.58 μg/g, respectively) (see Supplemental Table S2 for sitespecific data). 3.3. Sex and age influences on fur total mercury Juvenile bats had lower mercury concentrations in fur than adults, but there was no difference between males and females (Table 3). For little brown bats at six locations where both age classes and sexes were sampled, there was no significant difference in fur total mercury between males and females, but juveniles had on average 0.4 ± 0.04 times less total mercury than adults. Similar to the species models in Table 2, the variance explained by age was low (marginal r2 b 0.14) compared with the full model r2 that accounted for location effects (conditional r2 = 0.64), suggesting that site-specific exposure had a stronger influence (Table 3). No age-class information was available for most big brown bats and northern long-eared bats, although tests of sex effects for these species at sites where both males and females were sampled did not show a significant difference (Table 3). 3.4. Geographic patterns of fur total mercury Median total mercury concentrations in fur of little brown bats ranged from 0.88–12.78 μg/g among provinces and territories in Canada, including published concentrations for little brown bats from Little et al. (2015a) (Fig. 3). This data comparison included both juvenile

Fig. 2. Correlations of total mercury or methylmercury concentrations in fur and brain for a subset of 21 big brown bats and little brown bats (left panel, total mercury: Pearson r = 0.75, Holm's p b 0.001; right panel, methylmercury: Pearson r = 0.60, Holm's p b 0.05).


673

Table 2 Linear mixed models to test species effects (fixed factor) on fur total mercury concentrations (μg/g), while controlling for location (random effect). Models were tested on a subset of sites where both species in the comparison were sampled at the same sites. Modelled mean total mercury concentrations (and 95% confidence intervals) are back-transformed values from the log-models. Species

Total mercury (μg/g)

Fixed factor significance

Fixed factor (marginal) r2

Full model (conditional) r2

Model 1 (5 sites, n = 234) Big brown bats Little brown bats

3.00 (2.20–4.09) 1.81 (1.47–2.23)

t-Value = −4.81 p b 0.001

0.09

0.21

Model 2 (5 sites, n = 177) Northern long-eared bats Little brown bats

4.18 (3.31–5.27) 2.79 (1.63–4.79)

t-Value = 3.43 p b 0.001

0.04

0.42

and adult bats. The highest median concentrations were found in eastern Canada, particularly in New Brunswick (7.97 μg/g) and Nova Scotia (12.78 μg/g). The lowest concentrations (median b 3 μg/g) were found in the central and western provinces (Quebec, Ontario, Manitoba, Alberta) and two northern territories (Yukon and the Northwest Territories). No data were available for Saskatchewan and the sample size for British Columbia (on the Pacific coast) was very low (n = 6). A guideline of 10 μg/g of total mercury in fur has been suggested as a threshold for increased risk of sub-lethal neurochemical effects of mercury exposure for bats (Little et al., 2015a; Little et al., 2015b; Nam et al., 2012; Yates et al., 2014). The provinces in eastern Canada were the only regions where a substantial portion of concentrations exceeded the 10 μg/g threshold (Fig. 3). In our dataset, 31% of little brown bats from one site in New Brunswick exceeded the threshold. Only 0–2% of bats sampled at sites in Alberta, Manitoba, Ontario, Quebec, Yukon and Northwest Territories had fur total mercury concentrations N 10 μg/g. Measurements (included here) by Little et al. (2015a) showed that 62%, 26%, and 11% of bats sampled in Nova Scotia, Newfoundland, and Labrador, respectively, exceeded the 10 μg/g threshold. These results indicate that the Atlantic region in eastern Canada is an area of greater bioaccumulation of mercury for little brown bats. We examined whether atmospheric loadings of mercury to terrestrial and aquatic ecosystems could explain the observed geographic pattern of bioaccumulation in little brown bats. GEM-MACH-Hg estimates of annual net deposition of mercury (averaged from 2010 to 2015) indicated that sampling sites in eastern Canada tended to receive greater atmospheric deposition of mercury compared with southern Ontario, southern Quebec, and the western provinces (Fig. 4). Further, sitespecific estimates of atmospheric mercury deposition were positively correlated with mean concentrations of total mercury in fur of adult little brown bats (r2 = 0.43, n = 48 sites, p b 0.001) (Fig. 5). Atmospheric mercury deposition patterns explained a significant portion of the geographic variation in mercury bioaccumulation.

4. Discussion 4.1. Fur as an indicator of mercury bioaccumulation Fur was a relevant non-lethal indicator of mercury concentrations in internal tissues, based on positive correlations of total and methylmercury in fur with concentrations in the brain for a small subset of bats. There were also positive correlations of methylmercury values in fur with those in liver and kidney. Other studies have similarly concluded that concentrations in bat fur reflect their mercury exposure. Nam et al. (2012) observed significant positive correlations between fur, liver, and brain concentrations of mercury in control and mercury-contaminated bats. Wada et al. (2010) and Yates et al. (2014) found that fur and blood concentrations of mercury were positively correlated in several species of bats. The correlation coefficients for associations between mercury in fur and internal tissues in this study (Pearson r b 0.76) were similar to those reported in other studies (e.g., r = 0.81 for fur vs. brain in Nam et al., 2012; r = 0.87 for fur vs. blood in Wada et al., 2010). Given the prediction error associated with correlation coefficients of around 0.8, small differences in fur total mercury should be interpreted with caution when fur is being used to assess internal accumulation. Some of the variability between mercury concentrations in fur and internal tissues may be related to the timing of exposure. Fur is inert, and the mercury concentration reflects exposure at the time of hair growth. In contrast, there is continual uptake and elimination of mercury in internal tissues and, as a result, their concentrations change over time. New fur growth in North American bats typically occurs in June to August, and therefore, the chemical composition of bat fur is likely representative of summer conditions (Fraser et al., 2013). If the bats were collected at times outside of the fur growth period and there was a change in dietary mercury exposure, the mismatch in timing could influence the relationship between fur and internal mercury levels.

Table 3 Linear mixed models to test age and/or sex effects (fixed factors) on fur total mercury concentrations (μg/g) for three different bat species, while controlling for location (random factor). Only a subset of sites was included in the analysis where juveniles and adults and/or males and females were sampled. Modelled means (and 95% confidence intervals) are back-transformed values of log-models for each species. Note that age-class determinations were not available for big brown bats and northern long-eared bats at most sites. Species

Model 3 Little brown bats (6 sites, n = 200)

Model 4 Big brown bats (10 sites, n = 243) Model 5 Northern long-eared bats (2 sites, n = 34)

Sex Male

Female

Juvenile: 1.06 (0.70–1.60) n = 39 Adult: 2.42 (1.96–2.99) n = 30 3.56 (2.92–4.33) n = 123 4.54 (2.35–8.75) n = 23

Juvenile: 1.06 (0.86–1.29) n = 43 Adult: 2.42 (1.29–4.54) n = 88 3.94 (3.02–5.14) n = 120 5.24 (1.21–22.76) n = 11

Fixed factor significance

Fixed factor (marginal) r2

Full model (conditional) r2

Sex: t-value = 0.01, p = 0.989 Age: t-value = −8.07, p b 0.001

0.14

0.64

Sex: t-value = −1.02, p = 0.31

0.004

0.18

Sex: t-value = −0.45 p = 0.66

0.002

0.54

674


Fig. 3. The distribution of fur total mercury concentrations of little brown bats by province or territory in Canada. The horizontal reference line is a threshold of 10 μg/g of total mercury in fur for increased risk of neurochemical effects from mercury exposure in bats. The data are for individual bats (juvenile and adult), and sample sizes are in parentheses on the x-axis. Published site means of fur total mercury from Little et al. (2015a) are included for Nova Scotia, Prince Edward Island and Newfoundland and Labrador. Abbreviations for provinces and territories are provided in Fig. 1.

4.2. Brain methylmercury concentrations in bats The brain is a primary target organ for methylmercury toxicity, and sub-lethal exposure in mammals can cause behavioural and neurological effects (Scheuhammer et al., 2007). The brains of big and little brown bats measured in this study had methylmercury concentrations that

ranged an order of magnitude (0.13–2.20 μg/g dry weight), although the values were lower than those reported for little brown bats that exhibited neurochemical effects at a contaminated site (mean = 4.7 μg/g, range = 0.37–17.0 μg/g) (Nam et al., 2012). A threshold of 3–5 μg/g dry weight of methylmercury in brain was suggested in Dietz et al. (2013) for sub-lethal neurochemical effects, based on earlier studies of

Fig. 4. Modelled atmospheric deposition of mercury in Canada. Values are estimates of annual deposition rates (μg/m2) averaged over six years (2010 to 2015). Collection sites with fur total mercury concentrations for adult little brown bats (n = 48 sites) are identified on the map with circle symbols. The interior colour of each symbol reflects the site mean concentration of fur total mercury (μg/g) as per the legend for atmospheric deposition rates (offset by 1.5X + 20 μg/g to have a contrasting colour from background).

Fur Total Mercury of Little Brown Bats (µg/g)


675

sources) and the trophic transfer of methylmercury from important groups of arthropods consumed by different bat species. No differences were found for fur total mercury between males and females, similar to findings of Korstian et al. (2018). Yates et al. (2014) found significantly higher mercury concentrations in females than males but this difference was only for impacted point-source sites. No sex effect was found for their non-point source sites (similar to our study), and it was unclear what process may have been driving the higher mercury in females at point-source sites. In agreement with Korstian et al. (2018) and Yates et al. (2014), we found that adult little brown bats had higher mean mercury concentrations in fur compared with juveniles. The lower levels in juveniles may reflect differences in diet from adults (Hamilton and Barclay, 1998) or the relatively short period of mercury exposure for young-of-the year bats relative to adults.

30 25 20 15 10 5 0 0

10

20

30

40

50

60

4.4. Geographic patterns of mercury bioaccumulation in Canada

2

Mercury Net Deposition (µg/m ) Fig. 5. Linear regression (with 95% confidence interval) relating mean annual atmospheric deposition of mercury to site mean concentration of total mercury in fur of adult little brown bats (r2 = 0.43, p b 0.001, n = 48 sites).

neurochemical biomarkers in mammals and birds (Basu et al., 2007; Basu et al., 2006; Scheuhammer et al., 2008). Given that brain methylmercury concentrations in bats are approximately 10% of total mercury concentrations in fur (Section 3.1), the threshold for fur should perhaps be higher (30–50 μg/g) than the 10 μg/g of total mercury that has been previously applied (e.g., Little et al., 2015a; Little et al., 2015b; Yates et al., 2014). Regardless, it remains unclear whether the observed methylmercury exposure can be high enough to induce sub-lethal toxicological effects in bats. Further research is recommended to investigate the risk of behavioural and neurological effects at realistic environmental exposure levels, particularly those in eastern Canada (New Brunswick, Nova Scotia, Newfoundland) and the northeastern United States (Yates et al., 2014), where higher bioaccumulation was observed.

4.3. Influence of biological characteristics on mercury bioaccumulation On average, big brown and northern long-eared bats had higher fur total mercury concentrations than little brown bats. It remains unclear what factors may be driving the inter-species differences. Size variation is likely not important, given the similar size of little brown bats and northern long-eared bats (Supplemental Fig. S1). Diet variation among species is likely important for mercury exposure (Becker et al., 2018). Little brown bats typically have broad diets that include the consumption of dozens of insect species, largely from aquatic ecosystems (Clare et al., 2011; Lee and McCracken, 2004). A recent study by Broders et al. (2014) demonstrated using carbon and nitrogen stable isotopes that little brown bats in Atlantic Canada have different and broader diets than northern long-eared bats. Northern long-eared bats tended to feed more on larger beetles and on moths than little browns in a study area in Indiana (Lee and McCracken, 2004), and they may feed more in forested areas (Jung et al., 1999). Big brown bats are also generalist insectivores, but tend to consume more terrestrial arthropods such as coleoptera, lepidoptera, and hemiptera than little brown bats (Agosta and Morton, 2003; Hamilton and Barclay, 1998; Long and Kurta, 2014). It should be noted that the diet of insectivorous bats likely varies seasonally and geographically depending on local prey availability and may also be influenced by climate (Moosman Jr et al., 2012). Little information is available on the methylmercury content of terrestrial arthropods in Canada and typically, measurements of methylmercury on aquatic insects have been done on larval stages rather than the emergent adults consumed by bats. Further research is recommended to investigate the role of diet variation (e.g., terrestrial versus aquatic food

Concentrations of total mercury in fur of little brown bats varied geographically within Canada and were greatest in eastern provinces, associated with higher atmospheric mercury deposition. However, our study was somewhat limited by a lack of samples from the Pacific Coast, where annual atmospheric mercury deposition can reach high levels (≥50 μg/m2). Our findings are consistent with other large-scale studies of mercury in fish and piscivorous birds in North America, which have shown greater bioaccumulation in southeastern Canada and the northeastern United States (Depew et al., 2013; Evers et al., 2003; Kamman et al., 2005; Scheuhammer et al., 2016). The spatial pattern of mercury bioaccumulation in little brown bats likely reflects multiple processes. Atmospheric deposition of mercury explained 43% of the variation of fur total mercury among collection sites in our study. This observation is consistent with other research demonstrating the importance of mercury deposition in controlling bioaccumulation in aquatic food chains (Hammerschmidt and Fitzgerald, 2006; Harris et al., 2007). The model used to estimate atmospheric deposition simulates the direct deposition to the terrestrial and aquatic surfaces and represents the mercury deposition pattern on a regional scale. However, spatial distribution of mercury loadings on a watershed-scale is, in addition, shaped by post-depositional terrestrial mercury processes within the catchment which are governed by the biogeochemical characteristics of the watershed. For example, lakes in watersheds with high catchment to lake area ratios receive greater proportion of mercury loadings from the catchment stream inflow compared to the direct atmospheric deposition of mercury to the lake surface. Given the heterogeneity of terrestrial biogeochemistry in Canada, direct atmospheric deposition alone is not expected to explain all the variability in bat fur mercury concentrations in ecosystems across Canada. Lake acidity was likely another influencing factor on a large scale because of the preponderance of poorly buffered soils in eastern Canada and decades of acidification during the 20th century (Depew et al., 2013; Scheuhammer et al., 2016). On a regional scale, Little et al. (2015b) demonstrated greater mercury bioaccumulation in little brown bats was related to the acidity of lakes adjacent to collection sites in Nova Scotia. Little brown bats feed extensively on emergent aquatic insects, whose bioaccumulation of methylmercury is enhanced in lower pH lakes (Clayden et al., 2014). 4.5. Global context of findings on mercury bioaccumulation in bats The highest mercury levels in bat fur reported in the literature (i.e. N100 μg/g) are from sites adjacent to known point-source emitters of mercury in the northeastern United States (Nam et al., 2012; Yates et al., 2014). Lower though still elevated levels (10–50 μg/g) have been observed in bats from non-point source sites in northeastern United States (Yates et al., 2014), central United States (Korstian et al., 2018), and Atlantic Canada (Little et al., 2015a; this study). There is very little information on mercury accumulation in bats from other

676


parts of the world, with the exception of low concentrations reported in fur of insectivorous bats in Spain (≤2.3 μg/g; Lisón et al., 2017) and frugivorous and insectivorous bats in Malaysia (≤ 14 μg/g; Syaripuddin et al., 2014), as well as three-order of magnitude variation in fur mercury among 22 bat species in Belize (0.04–145 μg/g; Becker et al., 2018). Walker et al. (2007) measured total mercury in kidney of insectivorous bats in Britain at concentrations (0.93–5.50 μg/g) comparable to those in this study (Table 1), while Allinson et al. (2006) observed lower values of total mercury in liver (≤ 0.5 μg/g) of an insectivorous species in southern Australia. The diet (and trophic level) of bat species plays an important role in their mercury exposure, with tropical frugivorous and sanguivorous bats accumulating less mercury than insectivores (Becker et al., 2018; Syaripuddin et al., 2014). As demonstrated in this study, atmospheric mercury deposition also plays a key role for insectivorous bats, and future integration of mercury bioaccumulation measurements of bats on a large scale will need to account for ecosystem exposure to mercury. 5. Conclusions To our knowledge, this dataset provides the broadest geographic coverage of mercury levels in bats to date and the first empirical demonstration of a link between atmospheric mercury deposition and spatial patterns of mercury accumulation in a mammalian species. Our research has identified broad-scale geographic variation and key biological and environmental determinants of mercury levels in bats that will provide the basis for long-term tracking of their mercury exposure in Canada. Further investigation of mercury levels in bats from regions of Canada with high (≥50 μg/m2) annual atmospheric mercury deposition, such as the Pacific Coast and large offshore islands (e.g., Vancouver Island), may be warranted. Moreover investigation of the toxicological effects of high mercury exposure on bats and the trophic pathways that link mercury deposition with dietary exposure (particularly in terrestrial environments) will allow us to better evaluate the risk of this environmental stressor on bat populations, especially for endangered species. Big brown bats may serve as a bioindicator of potential mercury exposure for recovering populations of endangered little brown bats and northern long-eared bats, particularly in Atlantic Canada where population losses due to white nose syndrome have been particularly severe, and where our data show that fur total mercury is highest. Further, non-lethal sampling of insectivorous bats may provide a useful bioindicator for international monitoring initiatives to track the success of global efforts to reduce anthropogenic mercury emissions under the UNE Minamata Convention. Acknowledgements This study was funded primarily by Environment and Climate Change Canada, with additional financial support for field collections by the respective provincial and territorial government agency, university, and industry affiliations of the co-authors. Support for E.L.L. Cooke was provided from a Discovery Grant of the Natural Science and Engineering Research Council to John Chételat (2015-03846). We thank Lesley Howes for assisting with national coordination among field biologists during the initial stages. We also thank Piia Kukka (Yukon Department of Environment), Ariane Massé (Ministère des Forêts, de la Faune et des Parcs) and Helen Schwantje (British Columbia Ministry of Forests, Lands, Natural Resource Operations, and Ministry of Agriculture) who were instrumental in the field collection or compilation of samples from some locations. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2018.01.044.

References Agosta, S.J., Morton, D., 2003. Diet of the big brown bat, Eptesicus fuscus, from Pennsylvania and western Maryland. Northeast. Nat. 10, 89–104. Allinson, G., Mispagel, C., Kajiwara, N., Anan, Y., Hashimoto, J., Laurenson, L., et al., 2006. Organochlorine and trace metal residues in adult southern bent-wing bat (Miniopterus schreibersii bassanii) in southeastern Australia. Chemosphere 64, 1464–1471. AMAP/UNEP, 2013. Technical Background Report to the Global Atmospheric Mercury Assessment. Arctic Monitoring and Assessment Programme (AMAP), and United Nations Environment Programme (UNEP) Chemicals Branch, Oslo, Norway and Geneva, Switzerland http://www.amap.no/documents/download/1265. Barton, K., 2016. Multi-model Inference (MuMIn). https://cran.r-project.org/web/packages/MuMIn/index.html. Basu, N., Scheuhammer, A.M., Rouvinen-Watt, K., Grochowina, N., Klenavic, K., Evans, R.D., et al., 2006. Methylmercury impairs components of the cholinergic system in captive mink (Mustela vison). Toxicol. Sci. 91, 202–209. Basu, N., Scheuhammer, A.M., Rouvinen-Watt, K., Grochowina, N., Evans, R.D., O'Brien, M., et al., 2007. Decreased N-methyl-D-aspartic acid (NMDA) receptor levels are associated with mercury exposure in wild and captive mink. Neurotoxicology 28, 587–593. Bates, D., Maechler, M., Bolker, B., Walker, S., 2017. Linear Mixed Effects-models Using ‘Eigen’ and S4 (lme4). R-Core https://cran.r-project.org/web/packages/lme4/index. html. Becker, D.J., Chumchal, M.M., Bentz, A.B., Platt, S.G., Czirják, G.Á., Rainwater, T.R., et al., 2017. Predictors and immunological correlates of sublethal mercury exposure in vampire bats. R. Soc. Open Sci. 4, 170073. Becker, D.J., Chumchal, M.M., Broders, H.G., Korstian, J.M., Clare, E.L., Rainwater, T.R., et al., 2018. Mercury bioaccumulation in bats reflects dietary connectivity to aquatic food webs. Environ. Pollut. 233, 1076–1085. Broders, H.G., Farrow, L.J., Hearn, R.N., Lawrence, L.M., Forbes, G.J., 2014. Stable isotopes reveal that little brown bats have a broader dietary niche than northern long-eared bats. Acta Chir. 16, 315–325. Cai, Y., Tang, G., Jaffé, R., Jones, R., 1997. Evaluation of some isolation methods for organomercury determination in soil and fish samples by capillary gas chromatography - atomic fluorescence spectrometry. Int. J. Environ. Anal. Chem. 68, 331–345. Clare, E.L., Fraser, E.E., Braid, H.E., Fenton, M.B., Hebert, P.D.N., 2009. Species on the menu of a generalist predator, the eastern red bat (Lasiurus borealis): using a molecular approach to detect arthropod prey. Mol. Ecol. 18, 2532–2542. Clare, E.L., Barber, B.R., Sweeney, B.W., Hebert, P.D.N., Fenton, M.B., 2011. Eating local: influences of habitat on the diet of little brown bats (Myotis lucifugus). Mol. Ecol. 20, 1772–1780. Clayden, M.G., Kidd, K.A., Chételat, J., Hall, B.D., Garcia, E., 2014. Environmental, geographic and trophic influences on methylmercury concentrations in macroinvertebrates from lakes and wetlands across Canada. Ecotoxicology 23, 273–284. Cristol, D.A., Brasso, R.L., Condon, A.M., Fovargue, R.E., Friedman, S.L., Hallinger, K.K., et al., 2008. The movement of aquatic mercury through terrestrial food webs. Science 320, 335. Dastoor, A.P., Davignon, D., Theys, N., Van Roozendael, M., Steffen, A., Ariya, P.A., 2008. Modeling dynamic exchange of gaseous elemental mercury at polar sunrise. Environ. Sci. Technol. 42, 5183–5188. Dastoor, A., Ryzhkov, A., Durnford, D., Lehnherr, I., Steffen, A., Morrison, H., 2015. Atmospheric mercury in the Canadian Arctic. Part II: insight from modeling. Sci. Total Environ. 509–510, 16–27. Depew, D.C., Burgess, N.M., Campbell, L.M., 2013. Spatial patterns of methylmercury risks to common loons and piscivorous fish in Canada. Environ. Sci. Technol. 47, 13093–13103. Dietz, R., Sonne, C., Basu, N., Braune, B., O'Hara, T., Letcher, R.J., et al., 2013. What are the toxicological effects of mercury in Arctic biota? Sci. Total Environ. 443, 775–790. Durnford, D., Dastoor, A., Ryzhkov, A., Poissant, L., Pilote, M., Figueras-Nieto, D., 2012. How relevant is the deposition of mercury onto snowpacks?-part 2: a modeling study. Atmos. Chem. Phys. 12, 9251–9274. Eagles-Smith, C.A., Wiener, J.G., Eckley, C.S., Willacker, J.J., Evers, D.C., Marvin-DiPasquale, M., et al., 2016. Mercury in western North America: a synthesis of environmental contamination, fluxes, bioaccumulation, and risk to fish and wildlife. Sci. Total Environ. 568, 1213–1226. ECCC, 2017. Species At Risk Public Registry. Environment and Climate Change Canada (ECCC) https://www.registrelep-sararegistry.gc.ca. Evers, D.C., Taylor, K.M., Major, A., Taylor, R.J., Poppenga, R.H., Scheuhammer, A.M., 2003. Common loon eggs as indicators of methylmercury availability in North America. Ecotoxicology 12, 69–81. Fitzgerald, W.F., Engstrom, D.R., Mason, R.P., Nater, E.A., 1998. The case for atmospheric mercury contamination in remote areas. Environ. Sci. Technol. 32, 1–7. Fraser, E.E., Longstaffe, F.J., Fenton, M.B., 2013. Moulting matters: the importance of understanding moulting cycles in bats when using fur for endogenous marker analysis. Can. J. Zool. 91, 533–544. Fuchsman, P.C., Brown, L.E., Henning, M.H., Bock, M.J., Magar, V.S., 2017. Toxicity reference values for methylmercury effects on avian reproduction: critical review and analysis. Environ. Toxicol. Chem. 36, 294–319. Hamilton, I.M., Barclay, R.M.R., 1998. Diets of juvenile, yearling, and adult big brown bats (Eptesicus fuscus) in Southeastern Alberta. J. Mammal. 79, 764–771. Hammerschmidt, C.R., Fitzgerald, W., 2005a. Methylmercury in freshwater fish linked to atmospheric mercury deposition. Environ. Sci. Technol. 40, 7764–7770. Hammerschmidt, C.R., Fitzgerald, W.F., 2005b. Methylmercury in mosquitoes related to atmospheric mercury deposition and contamination. Environ. Sci. Technol. 39, 3034–3039.

J. Chételat et al. / Science of the Total Environment 626 (2018) 668–677 Hammerschmidt, C.R., Fitzgerald, W.F., 2006. Methylmercury in freshwater fish linked to atmospheric mercury deposition. Environ. Sci. Technol. 40, 7764–7770. Harris, R.C., Rudd, J.W.M., Amyot, M., Babiarz, C.L., Beaty, K.G., Blanchfield, P.J., et al., 2007. Whole-ecosystem study shows rapid fish-mercury response to changes in mercury deposition. Proc. Natl. Acad. Sci. U. S. A. 104, 16586–16591. Hartman, C.A., Ackerman, J.T., Herring, G., Isanhart, J., Herzog, M., 2013. Marsh wrens as bioindicators of mercury in wetlands of great salt lake: do blood and feathers reflect site-specific exposure risk to bird reproduction? Environ. Sci. Technol. 47, 6597–6605. Hickey, M.B.C., Fenton, M.B., MacDonald, K.C., Soulliere, C., 2001. Trace elements in the fur of bats (Chiroptera: Vespertilionidae) from Ontario and Quebec, Canada. Bull. Environ. Contam. Toxicol. 66, 699–706. Ingersoll, T.E., Sewall, B.J., Amelon, S.K., 2016. Effects of white-nose syndrome on regional population patterns of 3 hibernating bat species. Conserv. Biol. 30, 1048–1059. Jackson, A.K., Evers, D.C., Etterson, M.A., Condon, A.M., Folsom, S.B., Detweiler, J., et al., 2011a. Mercury exposure affects the reproductive success of a free-living terrestrial songbird, the carolina wren (Thryothorus ludovicianus). Auk 128, 759–769. Jackson, A.K., Evers, D.C., Folsom, S.B., Condon, A.M., Diener, J., Goodrick, L.F., et al., 2011b. Mercury exposure in terrestrial birds far downstream of an historical point source. Environ. Pollut. 159, 3302–3308. Jones, G., Jacobs, D.S., Kunz, T.H., Wilig, M.R., Racey, P.A., 2009. Carpe noctem: the importance of bats as bioindicators. Endanger. Species Res. 8, 93–115. Jung, T.S., Thompson, I.D., Titman, R.D., Applejohn, A.P., 1999. Habitat selection by forest bats in relation to mixed-wood stand types and structure in central Ontario. J. Wildl. Manag. 63, 1306–1319. Kamman, N.C., Burgess, N.M., Driscoll, C.T., Simonin, H.A., Goodale, W., Linehan, J., et al., 2005. Mercury in freshwater fish of northeast North America - a geographic perspective based on fish tissue monitoring databases. Ecotoxicology 14, 163–180. Korstian, J.M., Chumchal, M.M., Bennett, V.J., Hale, A.M., 2018. Mercury contamination in bats from the central United States. Environ. Toxicol. Chem. 37, 160–165. Lamborg, C.H., Fitzgerald, W.F., Damman, W.H., Benoit, J.M., Balcom, P.H., Engstrom, D.R., 2002. Modern and historic atmospheric mercury fluxes in both hemispheres: global and regional mercury cycling implications. Glob. Biogeochem. Cycles 16, 1104–1115. Lee, Y.F., McCracken, G.F., 2004. Flight activity and food habits of three species of Myotis bats (Chiroptera: Vespertilionidae) in sympatry. Zool. Stud. 43, 589–597. Lehnherr, I., 2014. Methylmercury biogeochemistry: a review with special reference to Arctic aquatic ecosystems. Environ. Rev. 22, 229–243. Lindberg, S., Bullock, R., Ebinghaus, R., Engstrom, D., Feng, X.B., Fitzgerald, W., et al., 2007. A synthesis of progress and uncertainties in attributing the sources of mercury in deposition. Ambio 36, 19–32. Lisón, F., Espín, S., Aroca, B., Calvo, J.F., García-Fernández, A.J., 2017. Assessment of mercury exposure and maternal-foetal transfer in Miniopterus schreibersii (Chiroptera: Miniopteridae) from southeastern Iberian Peninsula. Environ. Sci. Pollut. Res. 24, 5497–5508. Little, M.E., Burgess, N.M., Broders, H.G., Campbell, L.M., 2015a. Distribution of mercury in archived fur from little brown bats across Atlantic Canada. Environ. Pollut. 207, 52–58. Little, M.E., Burgess, N.M., Broders, H.G., Campbell, L.M., 2015b. Mercury in little brown bat (Myotis lucifugus) maternity colonies and its correlation with freshwater acidity in Nova Scotia, Canada. Environ. Sci. Technol. 49, 2059–2065. Long, B.L., Kurta, A., 2014. Activity and diet of bats in conventional versus organic apple orchards in southern Michigan. Can. Field-nat. 128, 158–164. Moosman Jr., P.R., Thomas, H.H., Veilleux, J.P., 2012. Diet of the widespread insectivorous bats Eptesicus fuscus and Myotis lucifugus relative to climate and richness of bat communities. J. Mammal. 93, 491–496. Munthe, J., Bodaly, R.A., Branfireun, B.A., Driscoll, C.T., Gilmour, C.C., Harris, R., et al., 2007. Recovery of mercury-contaminated fisheries. Ambio 36, 33–44. Nam, D.H., Yates, D., Ardapple, P., Evers, D.C., Schmerfeld, J., Basu, N., 2012. Elevated mercury exposure and neurochemical alterations in little brown bats (Myotis lucifugus) from a site with historical mercury contamination. Ecotoxicology 21, 1094–1101.

677

Orihel, D.M., Paterson, M.J., Gilmour, C.C., Bodaly, R.A., Blanchfield, P.J., Hintelmann, H., et al., 2006. Effect of loading rate on the fate of mercury in littoral mesocosms. Environ. Sci. Technol. 40, 5992–6000. Pinheiro, J., Bates, D., 2017. Linear and Nonlinear Mixed Effects Models (nlme). R-Core https://cran.r-project.org/web/packages/nlme/index.html. R Core Team, 2012. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria http://www.R-project.org/. Rimmer, C.C., McFarland, K.P., Evers, D.C., Miller, E.K., Aubry, Y., Busby, D., et al., 2005. Mercury concentrations in Bicknell's thrush and other insectivorous passerines in montane forests of northeastern North America. Ecotoxicology 14, 223–240. Scheuhammer, A.M., Meyer, M.W., Sandheinrich, M.B., Murray, M.W., 2007. Effects of environmental methylmercury on the health of wild birds, mammals, and fish. Ambio 36, 12–18. Scheuhammer, A.M., Basu, N., Burgess, N.M., Elliott, J.E., Campbell, G.D., Wayland, M., et al., 2008. Relationships among mercury, selenium, and neurochemical parameters in common loons (Gavia immer) and bald eagles (Haliaeetus leucocephalus). Ecotoxicology 17, 93–101. Scheuhammer, A.M., Basu, N., Evers, D.C., Heinz, G.H., Sandheinrich, M., Bank, M.S., Bank, M.S., 2012. Toxicology of mercury in fish and wildlife: recent advances. Mercury in the Environment: Pattern and Process. University of California Press, Berkley. Scheuhammer, A., Braune, B., Chan, H.M., Frouin, H., Krey, A., Letcher, R., et al., 2015. Recent progress on our understanding of the biological effects of mercury in fish and wildlife in the Canadian Arctic. Sci. Total Environ. 509–510, 91–103. Scheuhammer, A.M., Lord, S.I., Wayland, M., Burgess, N.M., Champoux, L., Elliott, J.E., 2016. Major correlates of mercury in small fish and common loons (Gavia immer) across four large study areas in Canada. Environ. Pollut. 210, 361–370. Streets, D.G., Horowitz, H.M., Jacob, D.J., Lu, Z., Levin, L., Ter Schure, A.F.H., et al., 2017. Total mercury released to the environment by human activities. Environ. Sci. Technol. 51, 5969–5977. Syaripuddin, K., Kumar, A., Sing, K.W., Halim, M.R.A., Nursyereen, M.N., Wilson, J.J., 2014. Mercury accumulation in bats near hydroelectric reservoirs in Peninsular Malaysia. Ecotoxicology 23, 1164–1171. Travnikov, O., Angot, H., Artaxo, P., Bencardino, M., Bieser, J., D'Amore, F., et al., 2017. Multi-model study of mercury dispersion in the atmosphere: atmospheric processes and model evaluation. Atmos. Chem. Phys. 17, 5271–5295. U.S. Fish & Wildlife Service, 2017. Endangered Species Database. U.S. Fish & Wildlife Service, Ecological Services https://www.fws.gov/endangered/. Vanderwolf, K.J., Malloch, D., McAlpine, D.F., 2016. Fungi on white-nose infected bats (Myotis spp.) in Eastern Canada show no decline in diversity associated with Pseudogymnoascus destructans (Ascomycota: Pseudeurotiaceae). Int. J. Speleol. 45, 43–50. Wada, H., Yates, D.E., Evers, D.C., Taylor, R.J., Hopkins, W.A., 2010. Tissue mercury concentrations and adrenocortical responses of female big brown bats (Eptesicus fuscus) near a contaminated river. Ecotoxicology 19, 1277–1284. Walker, L.A., Simpson, V.R., Rockett, L., Wienburg, C.L., Shore, R.F., 2007. Heavy metal contamination in bats in Britain. Environ. Pollut. 148, 483–490. Whitney, M.C., Cristol, D.A., 2018. Impacts of sublethal mercury exposure on birds: a detailed review. In: de Voogt, P. (Ed.), Reviews of Environmental Contamination and Toxicology. Vol. 244. Springer International Publishing, Cham, pp. 113–163. Wiener, J., Krabbenhoft, D., Heinz, G., Scheuhammer, A., 2003. Ecotoxicology of mercury. In: Hoffman, D., Rattner, B., Burton, G., Cairns, J. (Eds.), Handbook of Ecotoxicology. Lewis Publishers, Boca Raton, USA. Yates, D.E., Adams, E.M., Angelo, S.E., Evers, D.C., Schmerfeld, J., Moore, M.S., et al., 2014. Mercury in bats from the northeastern United States. Ecotoxicology 23, 45–55. Zukal, J., Pikula, J., Bandouchova, H., 2015. Bats as bioindicators of heavy metal pollution: history and prospect. Mamm. Biol. 80, 220–227.