Aquatic Sciences (2018) 80:40 https://doi.org/10.1007/s00027-018-0591-2 Aquatic Sciences RESEARCH ARTICLE Breeding eider ducks strongly influence subarctic coastal pond chemistry Matthew P. Duda1 H. Grant Gilchrist3 · Kathryn E. Hargan2,5 · Neal Michelutti1 · Linda E. Kimpe2 · Nik Clyde3 · Mark L. Mallory4 · Jules M. Blais2 · John P. Smol1 · Received: 17 April 2018 / Accepted: 13 August 2018 © Springer Nature Switzerland AG 2018 Abstract Arctic freshwater ponds are typically pristine and oligotrophic, however seabird biovectors can markedly alter water quality via enrichment with marine-derived nutrients and bioaccumulated metals. These ornithogenic inputs can be the dominant factor structuring aquatic biota and the surrounding island flora. Here, we measured a suite of limnological water chemistry variables and sediment geochemistry from 21 freshwater ponds influenced by Common Eiders (Somateria mollissima) in Hudson Strait, near the northern communities of Cape Dorset (Nunavut) and Ivujivik (Quebec). Nest counts and sedimentary δ15N values were used as proxies of bird abundance. Nutrient-rich guano from the nesting eiders visibly promoted the growth of catchment vegetation. Elevated metal (Al, Cd, Zn), metalloid (Se), and nutrient concentrations (N, P) in the water of eider-affected sites were recorded (Sign test; p = 0.004), but the proximity of many sites to the coast meant that variables related to ocean spray (conductivity, Na+, Mg2+, Cl−, Sr) confounded the effects of birds on pond water chemistry. In contrast, sediment geochemistry appeared to more clearly characterize sites according to the level of eider activity in their catchments by tracking Pb, Cd, N, and P sedimentary concentrations (Sign test; p = 0.02). These results have direct implications for reconstructing historical eider population trends using sediment archives, which is necessary to inform effective conservation management strategies. Keywords Ornitholimnology · Biovector · Common Eider · Arctic · Nutrients · Metals Introduction Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00027-018-0591-2) contains supplementary material, which is available to authorized users. * Matthew P. Duda mattpduda@gmail.com 1 Paleoecological Environmental Assessment and Research Lab (PEARL), Department of Biology, Queen’s University, Kingston, ON K7L 3N6, Canada 2 Department of Biology, University of Ottawa, 30 Marie Currie Pvt., Ottawa, ON K1N 6N5, Canada 3 Wildlife Research Division, Science and Technology Branch, Environment and Climate Change Canada, Ottawa, ON K1S 5B6, Canada 4 Department of Biology, Acadia University, 33 Westwood Ave., Wolfville, NS B4P 2R6, Canada 5 Present Address: Keck Sciences Department, Claremont McKenna College, 925 N. Mills Ave., Claremont, CA 91711, USA The study of Arctic lakes and ponds is logistically challenging, yet limnological research at high latitudes remains active primarily due to the sensitivity of Arctic freshwaters to global change (Smol and Douglas 2007; Rühland et al. 2008; Kaufman 2009; Smol 2016). One emerging field of study, ornitholimnology (Hurlbert and Chang 1983), investigates the influence of bird colonies on freshwater systems. Seabirds are potent biovectors, meaning they are capable of transporting and focusing nutrients (and contaminants) from their marine feeding grounds to their terrestrial nesting sites largely via their nutrient-rich feces and other deposits, eggshells and carcasses (Bildstein et al. 1992; Blais et al. 2007). In naturally oligotrophic Arctic regions, the nutrients and contaminants transported by biovectors greatly modifies the recipient terrestrial and aquatic ecosystems. For example, Blais et al. (2005) describe a 25-fold enrichment of total Hg and a 60-fold enrichment dichlorodiphenyltrichloroethane (DDT) to the environment derived from a Northern Fulmar 13 Vol.:(0123456789) 40 M. P. Duda et al. Page 2 of 16 (Fulmarus glacialis) colony in Cape Vera, Nunavut. Seabirds are one of the most prolific biovectors on the planet, with breeding seabirds producing and concentrating an estimated global excrement input onto land of 591 Gg N/y and 99 Gg P/y (Otero et al. 2018) and a total global input from all seabirds of 3800 Gg N/y and 631 Gg P/y (Otero et al. 2018). The total seabird input of N and P is comparable to estimates of the total nutrient input inland from the sea via global fisheries (N = 3700 Gg/y, Maranger et al. 2008, P = 320 Gg/y; Mackenzie et al. 1993). The effect of N and P loading cannot be understated as avian guano is a driving factor in shaping many biological communities, including floral (Zwolicki et al. 2013; Otero et al. 2015) and faunal (Sánchez-Piñero and Polis 2000; Zhu et al. 2015) assemblages. Because seabirds markedly alter the nutrient composition of the water chemistry and sediment geochemistry of their environment (Evenset et al. 2004; Blais et al. 2005; Keatley et al. 2009), paleolimnology can be used to infer population presence (Stewart et al. 2015), trends (Luoto et al. 2014), and dynamics through time (Keatley et al. 2011). By determining the historical population dynamics of a species, more effective conservation strategies can be implemented to sustain extant populations. Common Eiders (Somateria mollissima) are the largest duck in the Northern Hemisphere, inhabiting coastal marine Arctic and subarctic environments (Goudie et al. 2000). Most Common Eiders in the Arctic are migratory, although some may reside year-round and principally feed on molluscs and crustaceans (Goudie et al. 2000). Eiders nest in colonies, reusing previously built nests (Goudie et al. 2000). Once eggs are laid, female eiders spend > 99% of their time on the nest (Mallory et al. 2015), presumably to provide warmth and to prevent egg predation. However, eiders must drink freshwater, therefore it is advantageous to build nests near inland ponds to minimize time away from the nest (Fast 2006). Due to their gregarious nature, as well as the proximity of their nests to freshwaters, Common Eiders are ideal study organisms in ornitholimnological investigations. Here, we focus on the Northern Common Eider (S. m. borealis; hereafter eider), a subspecies that breeds along the coastline of the eastern Canadian Archipelago and winters along southwest Greenland through Newfoundland and Labrador (Mosbech et al. 2006; Steenweg et al. 2017). This subspecies prefers to build its nests on small, exposed, lowlying flat islands with small amounts of cover (Schmutz et al. 1983; Goudie et al. 2000). Many of these islands support small, freshwater ponds, which were the targets for our investigation. However, given the proximity of marine waters to the ponds and nest sites, we expected that ocean spray variables (conductivity, major ions and Sr, a trace metal highly associated with ocean spray) may confound variables related to eider influence, therefore we determined if other elements in the sediments more effectively track 13 eider presence. Annual eider inputs to the same islands and nesting sites, including guano, eggshells, moulted feathers and carcasses, continually fertilize the soil around their nests with nutrients and potentially metals and other contaminants, which inadvertently also fertilizes the nearby pond catchment, water, and sediments (Mallory et al. 2006; Brimble et al. 2009a; Keatley et al. 2009; Mallory 2015; Clyde 2016). To quantify the influence of eider presence on the subarctic islands, we measured limnological water chemistry and sediment geochemistry of 21 ponds on islands in Hudson Strait influenced to varying degrees by eiders. The amount of eider activity in the catchments was estimated by direct counts of the active nests on each island and by using stable isotopes of nitrogen (δ15N) in surface sediments, a wellestablished proxy for tracking marine nutrients in freshwaters (Minagawa and Wada 1984; Kelly 2000; Michelutti et al. 2009). Additionally, the guano of eiders was analyzed to determine the direct elemental inputs and isotopic signature. Principal component analysis (PCA) was used to graphically visualize the main patterns of variation in the study sites in relation to the water chemistry variables and sediment geochemistry. We hypothesized that eiders act as biovectors and ecological engineers (like dovekies Alle alle, González-Bergonzoni et al. 2017) and that their breeding activity alters the terrestrial and freshwater habitats. Consequently, we predicted that ornithogenic markers (e.g., δ15N, Pb, Zn, P) would be enriched in water and/or sediment samples of the high influence ponds compared to low influence sites. This study is the first ornitholimnological investigation in the remote Hudson Strait region and has direct implications towards reconstructing long-term eider population dynamics using sediment records from the ponds. Study area and site selection Cape Dorset (Nunavut, Canada) and Ivujivik (Quebec, Canada) are remote communities in the westerly arm of Hudson Strait (Fig. 1). Both areas have several nearby small uninhabited islands on which eiders nest. The two study regions exhibit similar geography and topography, both categorized as the Meta Incognita ecoregion (Sanborn-Barrie et al. 2008). The area is characterized by continuous permafrost and rugged bedrock, with minor amounts of colluvial soils (3vGeomatics Inc. 2011). Natural vegetation is dwarfed due to high winds, frigid temperatures and poor soils (Ricketts et al. 1999). Sampling locations were given unofficial names by the monitoring program at Environment Canada and Climate Change (ECCC). Sites A036 to A136 were ponds near Cape Dorset, and sites D003 to D022 were ponds near Ivujivik (Fig. 1). Sampling locations were the main pond of each island, and islands were selected on the basis of previous eider colony Breeding eider ducks strongly influence subarctic coastal pond chemistry Page 3 of 16 40 Fig. 1 Map showing the locations of the study ponds. Gull and ocean spray reference ponds are represented by black stars; low influence ponds (active nests ≤ 10; δ15N ≤ 6‰) are represented by open circles; moderate influence ponds (δ15N < 10‰) are represented by grey circles; high influence ponds (δ15N > 10‰) are represented by black triangles surveys, attempting to encompass a range of eider abundance (Clyde 2016). Site D018 was a nearly abandoned eider colony with few remaining eiders, taken over by a large number of gulls, outnumbering the number of eider nests of the other sampled islands. Large Arctic gulls occupy high trophic levels and are opportunistic feeders that consume a variety of foods, including fish, eggs and carrion (Mallory and Braune 2012). Since site D018 had a larger population of the gulls than eiders (which occupy lower trophic levels), site D018 was expected to have higher concentrations of metals (Portnoy 1990; Michelutti et al. 2009). Site D019 was particularly close to the ocean and received large amounts of ocean spray, and therefore was expected to have higher levels of aqueous marine ions, such as Na+, Mg2+, Cl−, as well as Sr compared to the study sites located farther inland (Côté et al. 2010; ChaguéGoff 2010; Hargan et al. 2017). Certain physical characteristics of the sample islands were measured post-hoc, including surface area using Google Earth ©2018 and distance to shore, which was measured from the centre of the pond to the nearest shoreline, accurate to 10 m (Government of Canada 2016). Materials and methods Sample collection Ivujivik surface sediments and water chemistry samples were collected from nine island ponds between July 25–30, 2014, and Cape Dorset samples were collected from 12 island ponds between July 17–25, 2015. Approximately 1 L of surface water (< 1 m depth) and a single sediment core were collected from each pond, and sampling occurred once per pond. For metal analysis, approximately 300 mL of water was filtered using Sartorius© 47 mm polycarbonate filters with 0.4 µm pore size and kept in the dark at 4 °C until analysis. For chlorophyll-a determination, 47 mm Whatman© glass microfibre filters were pre-ashed at 450 °C for 12 h, then 500 mL of water was filtered. The filters were transferred to Petri dishes, wrapped in aluminum foil, and then stored frozen in the dark until sent to National Laboratory for Environmental Testing (NLET; Burlington, ON). Upon return to the lab, the water filtered for metal analysis was acidified to a pH < 2 and all water samples were sent to NLET to be analyzed for major ions, nutrients, and trace metals using standard procedures (Environment Canada Manual of Analytical Methods 1994a, b). Specific conductivity and pH measurements were taken on-site using a calibrated YSI meter (Xylem, USA), and Hanna® handheld pH meter (USA), respectively. Temperature and pH were not collected in 2015 due to meter malfunction and therefore omitted from analyses (see Supplementary Table S1). Concurrently, sediment cores were retrieved using a highresolution push corer (Glew and Smol 2016) and sectioned onsite at 0.5 cm intervals using a Glew (1989) extruder. Only the surface 0.5 cm interval, representing the most recent conditions, was used in sedimentary geochemistry and δ15N 13 40 M. P. Duda et al. Page 4 of 16 analyses. Eider guano samples were collected from eiders that defecated while being handled by ornithologists and analyzed independently. Elemental analysis Samples were prepared following standard methodologies (SGS Canada Inc. 2014). Briefly, freeze-dried sediment and guano samples from two eiders were pulverized in an agate bowl and then subjected to an aqua regia digestion to extract the environmentally relevant metals while preserving the silicate matrix. Inductively coupled plasma mass spectrometry was used to analyze 30 elements of the sediment and guano: Al, As, Ba, Be, Bi, Ca, Cd, Co, Cr, Cu, Fe, K, Li, Mg, Mn, Mo, Na, Ni, P, Pb, Sb, Se, Sn, Sr, Ti, Tl, U, V, Y, and Zn. Quality assurance and control were ensured by running certified reference material, internal standards, blanks, and duplicates after every batch of 20 samples. Analyses were carried out by SGS Canada in Lakefield, Ontario, with results certified by the Canadian Association for Laboratory Accreditation Inc. (CALA). δ15N stable isotopes Nitrogen elemental and isotopic analyses of all samples were performed at the G.G. Hatch Stable Isotope Laboratory at the University of Ottawa, Ottawa, ON. For elemental %N analysis, sediment samples and standards were analyzed using a Vario EL III Elemental Analyzer (Elementar, Germany) following methods described in Brazeau et al. (2013). Sediment amounts needed for the δ15N isotopic analyses were determined based on the results of the elemental analysis and weighed accordingly into tin capsules with two parts tungsten trioxide (WO3). The isotopic composition of nitrogen was determined by the analysis of N2, produced by combustion on a Vario EL III Elemental Analyzer (Elementar, Germany) followed by “trap and purge” separation and online analysis by continuous-flow with a DeltaPlus XP Plus Advantage Isotope Ratio Mass Spectrometer coupled with a ConFlo II (Thermo, Germany). Our δ15N data were reported using delta (δ) notation in parts per thousand (‰) enrichments or depletions relative to common standards. Isotope data were normalized using previously calibrated internal standards, and analytical precision was ± 0.2% (Pella 1990). Quantifying eider influence To evaluate the influence of eiders on each pond, we used the number of occupied nests on the island, as well as δ15N values from the surface sediments of each site. Prior to fieldwork, the number of nests on each island were determined from unpublished field surveys conducted by ECCC and were used to approximate the number of eiders inhabiting 13 the island (Clyde 2016). Due to the inconsistency of survey years, ranging from 1997 to 2012, the possibility of eiders abandoning their nests between years, and the opportunity of migratory birds utilizing the islands as stopovers, sedimentary δ15N was used in conjunction with nest counts. Sedimentary δ15N has been shown repeatedly to be elevated in soils and sediments receiving marine-derived nitrogen, and a faithful indicator of seabird inputs (Mizutani et al. 1986; Blais et al. 2005; Brimble et al. 2009a). δ15N is enriched by approximately 3.4‰ per trophic level (Minagawa and Wada 1984), and therefore molluscivorous eiders introduce elevated levels of δ15N via their guano to the otherwise low δ15N freshwater environment (Michelutti et al. 2010). Further, due to biological processes, ocean water has a higher natural abundance of δ15N than terrigenous or freshwater habitats (Montoya 2007), therefore ocean spray will elevate pond δ15N. Based on the above qualifiers of active nest counts and δ15N, sites were divided into five categories: low, moderate and high influence, gull reference and ocean spray reference. Low influence sites had few active nests (≤ 10 per island) and low δ15N (≤ 6‰). All other sites were split into moderate or high influence ponds, dependant on whether they were higher or lower than δ15N = 10‰, the average level of eider guano (Clyde 2016). Based on this categorization, increasing number of nests was generally related to increasing δ15N. The main exception was A038 that was categorized as high influence (active eider nests = 7, δ15N = 12.3), which may have elevated δ15N due to migratory gull and/ or goose populations, evidenced by droppings around the catchment of the pond. This highlights the importance of using an independent ornithological measurement, such as δ15N, to confirm nest or bird counts. As explained earlier, the gull and ocean spray reference sites, D018 and D019, were determined to qualify the effects of ornithological and ocean spray influence. Statistical techniques Principal component analysis (PCA) was used to summarize the relationship of study sites to measured environmental variables. Water chemistry variables that had > 10% below the detection limit (DL) were removed. Otherwise, if < 10% of measured variables fell below DL, they were approximated using DL of variable/square root(2) (Hornung and Reed 1990). Variables that were right skewed were normalized with a log(x + 1) transformation. Parameters were then tested for normality with a Shapiro–Wilk test (α = 0.05) (Shapiro and Wilk 1965), and nonparametric variables were eliminated. Redundant variables were identified with a Pearson correlation matrix (p ≤ 0.05) with Bonferroni-adjusted probabilities using SPSS (see Supplementary Tables S2, S3) (IBM SPSS Breeding eider ducks strongly influence subarctic coastal pond chemistry Statistics for Windows, Version 19, 2010). A PCA of the correlated variables was used to determine a single variable that explained the largest amount of variation along the primary axis to represent the group. All ordination analyses were performed in Canoco, version 5.0 (ter Braak and Šmilauer 2012). A sign test is a nonparametric analysis that was used to determine if ornithological tracers (i.e. Al, As, Cd, Pb, Se, Zn, P, TN-F) were elevated in high influence ponds compared to low influence ponds. This test is well suited for small datasets to determine if the difference in medians is zero (Conover 1999). Page 5 of 16 40 Statistical limitations There were statistical limitations in this study due to the difficulty of locating and sampling islands with ponds with varying degrees of eider presence. This, compounded with finding sites that had similar geology and pond morphology, resulted in a limited sample size of 21 ponds, with a narrow range of eider populations. For this reason, there were too few samples to normalize several variables for PCA. However, of the removed variables, Cl−, K+, Mg2+, Na+ are all major ions correlated to conductivity, which was represented on the water PCA. Additionally, the removed ornithogenic markers (As, Cd, TP-UF) were represented by other bird influence markers, including Pb and chlorophyll-a. Data treatment Water chemistry ordination Results and discussion Arsenic, Cd, Cl −, K +, Mg 2+, Na + and total phosphorus unfiltered (TP-UF) were eliminated from the ordinal analysis due to skewness from a normal unimodal distribution. Other than dissolved inorganic carbon (DIC), Cu and Pb, all variables were normalized with a log(x + 1) transformation. Conductivity, major ions (Ca2+, SO42−) and Sr were correlated (see Supplementary Table S2) and represented by conductivity, which explained the largest amount of variation. Calcium and SO42− are major ions that contribute to conductivity, and Sr is a trace metal highly associated with ocean spray (Chen et al. 1997; Wang and Zhai 2008; Chagué-Goff 2010). To constrict ordination results to reflect only the distribution of eider sites relative to water chemistry and sediment geochemistry, the gull and ocean spray reference ponds, D018 and D019, were plotted passively. Guano and physical characteristics Sediment geochemistry ordination Six sedimentary variables (As, Bi, Li, Sb, Se, Sn) were removed due to a large amount of below detection values. All variables other than Cd and Pb were normalized with a log(x + 1) transformation, however Be, Ca2+, Mg2+, Mo, Na+, Th and U were eliminated from further analysis due to skewness. Trace metals, including Al, Ba, Co, Cr, Cu, Fe, Ni, Ti and V, were all correlated (see Supplementary Table S3), and therefore grouped using Al, which explained the largest amount of variation. Similarly, Cd and Zn were also correlated due to their binary relationship (Tang et al. 2014) and represented by Zn. Zinc has stable geochemistry in sediments (Boyle 2001) and has been correlated with seabird inputs (Brimble et al. 2009b; Foster et al. 2011). Similar to the water ordination, the gull and ocean spray reference ponds, D018 and D019, were plotted passively. Given the proximity of the study ponds to each other and their similar morphometries (Table 1), variability related to site-specific differences such as geology, climate, and atmospheric deposition was minimal. Thus, the main factor influencing pond chemistries was presumed to be the varying abundances of eiders at each site. As has been shown previously, seabird guano fertilizes and modifies freshwater ponds and their catchments (Mallory et al. 2006; Brimble et al. 2009a; Côté et al. 2010). Guano is high in P, constituting 0.9–17% of excrement total mass (Otero et al. 2015). The guano samples analyzed from two eiders had P concentrations of 980 ± 170 µg/g (Table 2), which likely fertilized the catchment because sites that had high eider abundance also had visibly more catchment vegetation compared to the catchments of unaffected sites, which were bare rock (Fig. 2). Eiders prefer to build their nests in vegetation, using plant material and their down feathers to insulate their eggs and avoid harsh conditions (Goudie et al. 2000). This likely forms a mutualistic feedback loop, in which eiders return to vegetated areas, which they ultimately fertilized, and continue to support plant material via nutrient-rich waste products. Ornithogenic elements analyzed in the guano, including Al, Cd, Pb and Se, were more elevated in the water and sediment chemistry of high influence ponds relative to low influence ponds (Tables 3, 4), emphasizing the pronounced ornitholimnological effect of eiders. Eiders bioaccumulate trace metals due to their preference to feed on molluscs and benthic crustaceans, which are typically enriched in Al, Cd, Pb and Zn (Szefer et al. 2006) or their propensity to acquire Pb from hunting activities (Hicklin and Barrow 2004; Falk et al. 2006; Johansen et al. 2006). In eiders, elements bioaccumulate due to persistence and their inability 13 40 M. P. Duda et al. Page 6 of 16 Table 1 Locations and physical variables of the 21-pond set, as well as the number of active Common Eider nests on each island in the most recent survey year, with corresponding surface sedimentary δ15N Pond Official island name Latitude (N) Longitude (W) Gull ref. Ocean spray ref. Low D018 D019 N/A N/A 62°19′50″ 62°20′17″ 78°09′16″ 78°09′30″ 70 30 0.027 0.034 21 (2012) 11 (2012) 23.0 10.6 A036 D007 A045 A054 A056 A114 D003 D004 D012 D022 A038 A044 A083 A085 A108 A135 A136 D013 D016 Neta Islands North Skerries Tunitjuak Island Putaguk Island Qasigijjat Salunnaqtuuq North Skerries North Skerries South Skerries North Skerries Neta Islands Simikutak Tatsiumajukallak Inugiavvik Qalirusilik Qujjautaq N/A South Skerries Île Pikiulik 64°15′12″ 62°26′43″ 64°17′08″ 64°19′04″ 64°16′59″ 64°16′06″ 62°25′52″ 62°26′37″ 62°23′01″ 62°26′37″ 64°14′51″ 64°17′52″ 64°19′20″ 64°17′32″ 64°20′23″ 64°03′44″ 64°05′07″ 62°22′56″ 62°19′20″ 76°18′21″ 78°08′08″ 75°46′56″ 75°45′17″ 75°44′25″ 74°11′20″ 78°10′17″ 78°09′02″ 78°11′06″ 78°09′02″ 76°13′45″ 75°47′14″ 74°40′01″ 74°38′56″ 74°22′33″ 73°31′56″ 73°30′44″ 78°11′32″ 78°10′33″ 250 50 110 100 190 170 130 110 70 110 50 80 60 180 130 170 60 50 60 0.54 0.040 1.2 0.096 0.28 0.39 0.24 0.19 0.11 0.068 0.010 0.078 0.037 0.17 0.15 0.25 0.12 0.058 0.054 10 (1997) 8 (2012) 15 (2012) 243 (2012) 434 (2012) 197 (2012) 228 (2012) 212 (2012) 230 (2012) 101 (2012) 7 (2011) 141 (2012) 259 (2010) 225 (2010) 234 (2010) 127 (2012) 444 (2012) 74 (2012) 367 (2012) 5.42 3.08 7.14 8.65 6.30 9.64 7.72 7.68 9.04 9.82 12.3 12.2 11.6 11.6 12.2 10.4 13.6 10.6 10.3 Mod. High Table 2 Concentration and standard deviation (SD) of relevant elements in Common Eider (n = 2) guano Element Concentration (µg/g dry weight ± SD) As Ca Cd Cu P Pb Se Sr Zn 2.15 ± 1.20 175 ± 7 0.17 ± 0.021 5.15 ± 0.49 980 ± 170 1.05 ± 0.07 1.35 ± 0.071 770 ± 85 11.5 ± 0.7 to biodegrade (Cardwell et al. 2013). The elevated elements are released back to the ecosystem through guano, eggshells and carcasses. Pond water chemistry Elevated major ions and ornithogenic elements were recorded in the water of the eider-affected sites. In our 21-pond dataset, the gull colony reference pond, D018, had the highest measured metals of ornithogenic influence, 13 Distance to shore (m) Island surface area (km2) δ15N (‰) Influence Nests (year) as also described by Brimble et al. (2009b), including: As (8.45 µg/L), Cd (0.199 µg/L), Se (1.6 µg/L) and Zn (17 µg/L) (Table 3). Additionally, D018 had the highest concentrations of variables linked to eutrophication, and thus eider and/or gull influence, including: chlorophyll-a (429 µg/L), TP-F (3800 µg/L) and TP-UF (7100 µg/L) (Table 3). In our dataset, the ocean spray reference pond, D019, had the highest concentrations of major ions including: conductivity (13,000 µS/cm), Ca2+ (133 mg/L), K+ (116 mg/L), Mg2+ (357 mg/L), Na+ (3350 mg/L) and SO42− (862 mg/L) (Table 3). Moreover, D019 had the highest Sr (3180 µg/L), which is strongly associated with ocean spray (Chen et al. 1997; Wang and Zhai 2008; Chagué-Goff 2010). The ornithogenic bioaccumulated elements were elevated in the water of the high eider-influence ponds compared to low influence ponds. Examples of such metals and metalloids include Al (low influence 28.2 ± 11.8 µg/L, high influence 93.8 ± 70.7 µg/L), Cd (low influence 0.031 ± 0.031 µg/L, high influence 0.047 ± 0.032 µg/L), Se (low influence 0.19 ± 0.16 µg/L, high influence 0.23 ± 0.19 µg/L) and Zn (low influence 1.7 ± 0.57 µg/L, high influence 2.7 ± 1.1 µg/L). Though comparisons of individual elements were not statistically significant due to a limited sample size, mean values of all eight identified Breeding eider ducks strongly influence subarctic coastal pond chemistry Page 7 of 16 40 Fig. 2 On the left is an image of a low-influence pond, A036, with minimal vegetation, and on the right, is a high-influence pond, D016, with lush vegetation ornithogenic tracers (Al, As, Cd, Pb, Se, Zn, P, TN-F; Mallory et al. 2004; Michelutti et al. 2010) were greater in the waters of the moderate and high influence ponds compared to low influence (Sign test; p = 0.004). Some of these elements were highlighted by Brimble et al. (2009b) as ornithogenic metals of concern, and potentially toxic. However, at the time of measurement, the elements were below the chronic level of concern in freshwater as per the Water Quality Guidelines for the Protection of Aquatic Life set by the Canadian Council of Ministers of the Environment (CCME 1999) (Cd 0.09 µg/L; Se 1 µg/L; Zn 30 µg/L). Further, these metals are considered guano-derived, and would not normally be elevated in pristine, unaffected ponds (Roberts et al. 2017), additionally supporting the marked influence of eiders on their environment. Similar to metals, nutrients were elevated in eider-affected sites relative to control ponds (Table 3). As expected, measured phosphorus in the water was high, similar to other Arctic seabird studies (e.g. Keatley et al. 2009; Côté et al. 2010). In a 2001 study of nutrients in the Canadian Arctic Archipelago, the mean TP-UF of unimpacted lakes was 12 ± 18 µg/L (Hamilton et al. 2001). Although the low influence ponds had a mean TP-UF of 102 ± 1.4 µg/L, the high influence sites recorded a mean value four times greater at 416 ± 408 µg/L. The elevated P concentration in the low influence sites indicates that there is likely an additional source of phosphorus. Although the control sites currently do not have a large eider population, it is possible that seasonal migratory geese or gull populations land at the islands for short stopovers, evidenced by geese droppings across the islands. Geese are known to release a substantial amount of P in the Arctic via their feces (Mariash et al. 2018), from which the detectable reintroduction can last several decades (Søndergaard et al. 2003). Additionally, high δ15N sites with few eider nests (e.g. A038) support the possibility of transient bird populations. Total filtered nitrogen (TN-F) was also measured to determine the amount of organic N in the water, which has been shown to increase due to biovectors (Marion et al. 1994; Zwolicki et al. 2013). We measured a TN-F of 1.61 ± 1.06 mg/L in the low influence ponds and 3.98 ± 7.52 mg/L in high influence ponds (Table 3). Nitrogen, along with P, is associated with the growth and abundance of algal communities (Smith 1982) and can be linked to elevated chlorophyll-a concentrations in the eider ponds (low influence 19.3 ± 1.13 µg/L; high influence 22.0 ± 25.8 µg/L). The elevated TN-F in the eider affected ponds further highlights the ability of eiders to shape their environment. In the Arctic, it is difficult to find control ponds with no bird influence because suitable habitat is limited, therefore every island is likely occupied and influenced by some bird presence during either migration or breeding seasons. Even sites with few active eider nests (e.g. A036, A038, D007) had higher P and δ15N than would be expected if there were no birds present (Brimble et al. 2009a). This highlights the importance of independent proxies, such as δ15N, to corroborate nest or bird counts. Without controls, biogenic enrichment factors (Brimble et al. 2009b) to quantify the enrichment of affected sites due to guano subsidies were not developed. This limitation made comparisons between high and low influence sites statistically imperfect. The PCA axes 1 and 2 explained 48.4% of the total variation with eigenvalues of λ1 = 0.294, and λ2 = 0.190 (Fig. 3). Axis 1 was strongly characterized by a Pb and dissolved organic carbon (DOC) gradient, which we consider to track the influence of eiders. As seen in other studies, eiders commonly have elevated Pb in their liver (Mallory et al. 2004), 13 40 Lake CHL-a (µg/L) DIC (mg/L) DOC (mg/L) TN (mg/L) TP-F (µg/L) TP-UF (µg/L) Cond. (µS/ cm) Ca2+ (mg/L) K+ (mg/L) Mg2+ (mg/L) Na+ (mg/L) SO42− (mg/L) Gull ref. Ocean spray ref. Low D018 D019 429 8.00 14.8 10.1 47.1 14.2 7.90 1.17 3800 30 7100 93.0 1066 13,000 14.0 133 16.7 116 13.0 357 177 3350 52.5 862 A036 D007 Mean ± SD Moderate A045 A054 A056 A114 D003 D004 D012 D022 Mean ± SD High A038 A044 A083 A085 A108 A135 A136 D013 D016 Mean ± SD 18.5 20.1 19.3 ± 1.13 6.00 0.100 0.100 4.20 128 25.3 166 3.50 41.7 ± 66.3 10.6 76.5 47.9 3.60 3.10 33.8 14.2 7.20 1.20 22.0 ± 25.8 8.60 10.2 9.40 ± 1.10 10.7 20.8 14.3 4.00 3.50 12.0 18.0 20.6 12.9 ± 6.80 13.3 4.90 8.20 9.30 9.90 9.20 9.60 10.6 26.6 11.3 ± 6.10 10.2 19.8 15.0 ± 6.8 11.9 15.4 8.80 14.0 18.8 20.8 35.6 16.3 17.7 ± 8.20 10.7 14.6 7.40 17.9 13.5 14.8 14.7 21.6 40.2 17.3 ± 9.50 0.867 2.36 1.61 ± 1.06 2.64 1.54 0.865 0.817 1.04 1.71 3.61 1.89 1.76 ± 0.962 1.69 1.04 0.838 1.46 1.39 1.38 1.44 2.57 24.0 3.98 ± 7.52 65 34 50 ± 22 38 79 47 180 41 66 88 170 89 ± 57 350 69 40 510 230 71 49 52 1300 290 ± 401 101 103 102 ± 1.40 75.0 102 65.0 230 249 549 1260 283 352 ± 399 397 294 130 594 308 304 161 126 1430 416 ± 408 172 5000 2590 ± 3410 292 438 324 75.0 325 650 1270 346 465 ± 363 342 112 1490 157 332 117 198 414 750 435 ± 445 8.95 46.0 28.0 ± 26.0 13.6 40.1 19.9 2.12 7.35 20.4 42.8 24.8 21.4 ± 14.4 12.5 5.33 17.7 12.1 9.21 10.1 12.1 27.2 32.0 15.4 ± 8.80 2.14 39.1 20.6 ± 26.1 2.70 3.37 4.16 0.53 2.09 4.96 9.83 2.47 3.76 ± 2.79 2.21 1.67 9.08 0.94 2.50 0.51 0.84 3.66 5.80 3.02 ± 2.81 2.36 112 57.2 ± 77.5 4.44 5.46 4.54 1.31 4.34 7.59 19.8 5.95 6.68 ± 5.59 6.04 1.62 25.0 2.18 5.64 1.70 3.70 6.34 8.85 6.79 ± 7.26 17.3 1060 539 ± 737 34.2 37.5 33.8 10.6 52.7 89.9 223 45.4 65.9 ± 67.3 43.2 12.2 209 16.8 43.6 12.0 17.5 60.3 85.9 55.6 ± 62.8 5.80 328 167 ± 228 17.0 40.5 17.5 2.60 16.2 31.0 44.3 13.4 22.8 ± 14.3 10.5 2.20 54.6 3.70 13.4 4.00 12.9 21.9 31.3 17.2 ± 16.9 Influence Lake Al (µg/L) As (µg/L) Cd (µg/L) Cr (µg/L) Cu (µg/L) Fe (µg/L) Pb (µg/L) Se (µg/L) SiO2 (mg/L) Sr (µg/L) V (µg/L) Zn (µg/L) Gull ref. Ocean spray ref. Low D018 D019 52.3 28.6 8.45 1.72 0.199 0.041 0.34 0.42 5.22 2.31 196 76.7 0.165 0.047 1.6 0.34 1.6 0.38 189 3180 3.3 2.0 17 1.7 0.24 2.02 1.13 ± 1.26 0.009 0.22 0.053 0.15 0.031 ± 0.031 0.19 ± 0.05 A036 36.5 D007 19.8 Mean ± SD 28.2 ± 11.8 9.34 91.3 6.32 114 7.83 ± 2.14 103 ± 16.1 0.171 0.07 0.39 27.5 0.68 0.083 0.30 0.72 833 1.3 0.127 ± 0.062 0.19 ± 0.16 0.56 ± 0.23 430.3 ± 570.0 0.99 ± 0.44 1.3 2.1 1.7 ± 0.57 M. P. Duda et al. Influence Page 8 of 16 13 Table 3 Water variables of the 21-pond set Influence Lake Moderate A045 A054 A056 A114 D003 D004 D012 D022 Mean ± SD A038 High A044 A083 A085 A108 A135 A136 D013 D016 Mean ± SD Al (µg/L) As (µg/L) Cd (µg/L) Cr (µg/L) Cu (µg/L) Fe (µg/L) Pb (µg/L) Se (µg/L) SiO2 (mg/L) Sr (µg/L) V (µg/L) Zn (µg/L) 46.8 121 14.4 154 125 167 73.2 48.8 93.8 ± 55.6 22.1 148 56.8 103 93.4 249 27.3 49.6 94.7 93.8 ± 70.7 0.66 1.35 3.91 0.32 0.47 1.45 3.15 0.79 1.51 ± 1.32 0.79 0.36 0.71 0.62 0.88 0.78 0.44 1.02 4.34 1.10 ± 1.23 0.048 0.075 0.006 0.022 0.024 0.078 0.039 0.037 0.041 ± 0.025 0.049 0.123 0.046 0.021 0.050 0.032 0.011 0.038 0.056 0.047 ± 0.032 0.29 0.88 0.10 0.23 0.29 0.64 0.34 0.18 0.37 ± 0.26 0.10 0.36 0.18 0.25 0.44 0.19 0.47 0.35 0.31 0.29 ± 0.12 5.14 5.19 6.91 3.39 4.83 6.19 0.91 4.96 4.69 ± 1.84 1.25 6.25 3.64 1.52 12.4 2.75 5.33 2.79 3.53 4.38 ± 3.41 203 547 60.6 375 210 463 668 111 330 ± 218 271 90.2 154 257 363 295 1370 275 1130 467 ± 455 0.303 0.130 0.043 0.129 0.262 0.371 0.410 0.167 0.227 ± 0.130 0.467 0.140 0.129 0.195 0.402 0.201 0.026 0.601 0.263 0.269 ± 0.185 0.10 0.17 0.08 0.10 0.23 0.36 0.35 0.24 0.20 ± 0.11 0.14 0.20 0.16 0.16 0.20 0.10 0.10 0.38 0.62 0.23 ± 0.19 0.84 0.35 0.31 0.06 1.5 2.0 1.3 2.0 1.1 ± 0.77 0.81 0.91 0.05 0.02 1.7 0.53 0.14 0.13 2.2 0.72 ± 0.78 75.3 216 97.1 17.4 52.4 128 312 113 126 ± 95.2 89.9 40.6 229 87.5 82.1 78.1 75.5 171 215 119 ± 68 1.3 1.3 0.74 1.2 1.0 7.8 7.1 0.95 2.7 ± 2.9 0.71 1.1 0.88 1.3 1.8 2.1 0.57 6.2 9.8 2.7 ± 3.2 1.1 3.4 0.5 4.1 4.3 4.3 2.8 3.4 3.0 ± 1.5 1.6 5.2 3.0 1.9 3.1 2.6 1.5 2.1 3.1 2.7 ± 1.1 Breeding eider ducks strongly influence subarctic coastal pond chemistry Table 3 (continued) Chlorophyll-a is expressed as CHL-a; dissolved inorganic carbon is expressed as DIC; dissolved organic carbon is expressed as DOC; total nitrogen filtered is expressed as TN-F; total phosphorus is expressed as TP, and is either filtered, -F, or unfiltered, -UF; conductivity is expressed as Cond. Mean and standard deviation are calculated with the gull and ocean spray reference ponds omitted Page 9 of 16 40 13 40 Lake Ca2+ (mg/g) K+ (mg/g) Mg2+ (mg/g) Na+ (mg/g) P (mg/g) Al (µg/g) As (µg/g) Ba (µg/g) Be (µg/g) Cd (µg/g) Co (µg/g) Cr (µg/g) Cu (µg/g) Gull ref. Ocean spray ref. Low D018 D019 56 34 6.7 2.7 6.3 4.7 16 14 16 0.98 1400 1800 6.9 0.9 37 37 0.03 0.03 0.86 0.09 2.8 3.2 7.2 16 10 5.7 A036 D007 Mean ± SD A045 A054 A056 A114 D003 D004 D012 D022 Mean ± SD A038 A044 A083 A085 A108 A135 A136 D013 D016 Mean ± SD 15 11 13 ± 2.8 8.5 16 16 8.3 29 16 14 60 21 ± 17 11 16 39 12 8 10 13 18 12 15 ± 9.3 3.8 11 7.4 ± 5.1 4.5 3.4 3.8 1.3 2.7 4.8 3.8 4.6 3.6 ± 1.2 5.2 2.1 1.5 3.1 2.9 2.0 2.9 5.4 2.0 3.0 ± 1.4 3.1 26 15 ± 16 3.8 3.3 4.6 2.2 4.4 5.9 3.0 5.7 4.1 ± 1.3 3.3 9.6 2.8 1.6 4.0 1.5 2.6 5.0 2.3 3.6 ± 2.5 2.6 98 51 ± 67 0.75 0.76 0.97 0.59 1.7 1.3 2.3 1.2 1.2 ± 0.57 2.9 0.61 1.6 0.37 1.2 0.29 2.0 2.3 0.96 1.4 ± 0.91 3.7 0.26 2.0 ± 2.4 5.6 4.3 3.0 2.7 2.4 4.2 4.8 4.7 4.0 ± 1.1 7.0 3.3 0.98 4.5 2.0 4.2 5.6 5.3 3.1 4.0 ± 1.9 1700 530 1100 ± 830 2600 2300 3500 3700 4000 3600 710 3100 2900 ± 110 1100 14,000 1900 690 2400 1100 2100 3300 1600 3100 ± 4200 < 0.5 3.0 1.7 ± 1.7 1.3 1.6 19 0.8 1.0 2.2 5.7 3.0 4.3 ± 6.1 0.8 < 0.5 < 0.5 0.7 < 0.5 2.0 1.6 2.7 2.8 1.3 ± 1.0 34 27 31 ± 4.9 78 150 400 62 45 94 18 89 120 ± 120 9.0 38 38 9.2 53 7.5 36 47 24 29 ± 17 0.05 < 0.02 0.03 ± 0.02 0.13 0.05 0.26 0.16 0.11 0.06 0.03 0.11 0.11 ± 0.07 0.04 0.40 0.05 0.03 0.03 0.04 0.15 0.09 < 0.02 0.09 ± 0.12 0.72 0.04 0.38 ± 0.48 0.66 2.5 1.5 4.1 2.2 1.0 1.6 1.4 1.9 ± 1.1 1.8 2.9 0.28 3.3 0.44 3.4 2.4 0.84 2.9 2.0 ± 1.2 2.3 1.4 1.9 ± 0.63 5.1 8.7 22 10 3.8 4.9 1.8 3.5 7.5 ± 6.5 0.90 13 2.2 1.4 5.3 1.7 4.6 3.2 3.7 4.0 ± 3.7 15 3.6 9.3 ± 8.1 26 21 80 25 17 22 12 18 28 ± 22 5.9 62 14 7.7 24 7.4 30 12 9.0 19 ± 18 240 4.4 120 ± 170 83 91 280 130 74 66 20 49 99 ± 80 13 100 6.1 25 46 45 180 16 45 53 ± 55 Influence Lake Fe (mg/g) Li (µg/g) Mo (µg/g) Ni (µg/g) Pb (µg/g) Se (µg/g) Sr (µg/g) Ti (µg/g) Th (µg/g) U (µg/g) V (µg/g) Y (µg/g) Zn (µg/g) Gull ref. Ocean spray ref. Low D018 D019 5.9 17 3 4 0.4 0.8 7.2 5.9 1.4 0.88 3.2 0.7 310 190 270 400 0.02 0.05 0.42 0.25 12 48 2.3 5.9 81 20 3 5 4±1 2 0.3 1±1 13 5.3 9.2 ± 5.4 12 0.60 6.3 ± 8.1 1.4 < 0.7 0.95 ± 0.49 53 190 120 ± 97 180 150 170 ± 21 0.17 11 < 0.02 0.12 0.10 ± 0.11 5.6 ± 7.7 13 4.0 8.5 ± 6.4 4.1 0.54 2.3 ± 2.5 55 10 33 ± 32 Moderate High A036 4.7 D007 1.4 Mean ± SD 3.1 ± 2.3 M. P. Duda et al. Influence Page 10 of 16 13 Table 4 Sedimentary variables of the 21-pond set Influence Lake Fe (mg/g) Li (µg/g) Mo (µg/g) Ni (µg/g) Pb (µg/g) Se (µg/g) Sr (µg/g) Ti (µg/g) Th (µg/g) U (µg/g) V (µg/g) Y (µg/g) Zn (µg/g) Moderate A045 A054 A056 A114 D003 D004 D012 D022 Mean ± SD A038 A044 A083 A085 A108 A135 A136 D013 D016 Mean ± SD 8.7 9.3 12 8.1 9 13 5.8 9.2 9.4 ± 2.2 4.3 17 8.7 3.1 16 2.5 31 13 4.3 11 ± 9.3 6 8 28 3 4 5 <2 6 7.8 ± 8.4 3 35 5 <2 4 <2 <2 4 <2 7 ± 11 0.8 0.3 1 1 0.3 0.3 < 0.1 0.3 0.5 ± 0.4 0.4 0.2 0.6 0.4 0.4 0.4 0.5 0.6 0.4 0.4 ± 0.1 40 86 250 56 54 15 8.2 13 65 ± 79 3.2 29 5.9 8.7 15 20 48 6 11 16 ± 14 12 2.4 10 16 12 5.1 5.0 4.1 8.3 ± 4.8 9.2 7.3 1.2 4 3.4 2.8 3.8 10 2.6 4.9 ± 3.1 49 93 170 160 150 84 90 230 130 ± 59 85 130 250 92 49 62 92 88 76 102 ± 60 480 320 1230 260 360 680 130 470 490 ± 340 180 840 360 64 630 93 150 550 300 350 ± 270 0.11 0.13 0.3 0.07 0.09 0.12 0.04 0.1 0.12 ± 0.08 0.05 0.11 0.07 0.04 0.09 < 0.02 0.07 0.08 0.05 0.06 ± 0.03 0.9 0.21 1.9 2.3 3.8 1.1 0.21 0.66 1.4 ± 1.2 1.0 0.53 0.35 0.39 0.61 3.0 0.60 0.73 0.39 0.84 ± 0.83 18 14 51 23 12 25 13 19 22 ± 13 6.0 105 16 4.0 33 5.0 19 31 13 26 ± 32 3 1.3 3.4 15 7.9 5.9 1.3 5.4 5.4 ± 4.5 1.5 4.7 7.2 1.2 5.1 2.5 4.8 6.0 1.7 3.9 ± 2.2 53 70 120 330 130 68 75 72 110 ± 91 62 140 26 120 44 120 130 70 150 96 ± 46 High 1.3 1.5 2.8 5.3 3.5 2.4 2.9 3.2 2.9 ± 1.3 1.3 2.1 < 0.7 2.3 < 0.7 < 0.7 2.5 2.4 3.2 1.8 ± 0.93 Breeding eider ducks strongly influence subarctic coastal pond chemistry Table 4 (continued) Mean and standard deviation are calculated with the gull colony, D018, and ocean spray pond, D019, omitted. Bismuth and Sb were below detection for all sites and therefore omitted Page 11 of 16 40 13 40 M. P. Duda et al. Page 12 of 16 which is attributed to shot birds that survive and ingested lead shot leftover from hunting (Flint and Grand 1997; Franson et al. 2000; Grand et al. 2002). Additionally, eiders are molluscivorous, feeding on mussels and other benthic crustaceans (Goudie et al. 2000). Since molluscs are filter feeders, they concentrate Pb from the sediments (Szefer et al. 2006), which may be a minor contributor to Pb elevation in eiders, despite the inability of Pb to biomagnify (Cardwell et al. 2013) and evidence for biominification (Jenkins 1980). As the Pb is not metabolized, it is eventually excreted into the catchments of the study ponds (Michelutti et al. 2010). In alkaline waters (pH > 7.5), like the water in this study, Pb complexes to form the insoluble PbCO3, which is very stable and persistent in sediments (Long and Angino 1977). Dissolved organic carbon in water is largely related to vegetation (Neff and Hooper 2002), and thus aqueous DOC would be largely allochthonous in origin. There are also autochthonous sources of DOC via aquatic algal cell death and senescence, grazing, viral lysis and extracellular release (Bertilsson and Jones 2003). The study ponds had high productivity as evidenced by the elevated chlorophyll-a concentration compared to other Canadian Archipelago ponds (mean = 0.55 µg/L; Hamilton et al. 2001). Therefore, DOC in the highly influenced eider ponds will be elevated likely due to allochthonous eider guano fertilization and enhanced growth of mosses, as well as autotrophic algae. Importantly, increased DOC may influence the aquatic biota composition by limiting primary production (Carpenter et al. 1998), affecting both epilimnetic (Hanson et al. 2003) and hypolimnetic respiration (Houser et al. 2003) and increasing metal Fig. 3 Principal components analysis (PCA) of the 21-pond set. On the left is a PCA of water variables; on the right is a PCA of sedimentary variables. Gull and ocean spray reference ponds are represented by black stars; low influence ponds (active nests ≤ 10; δ15N ≤ 6‰) are represented by open circles; moderate influence 13 toxicity (Evans et al. 2005). The secondary axis was characterized by a gradient of conductivity, which is explained primarily by ocean spray. Ocean spray can strongly influence the water chemistry of nearshore Arctic ponds (Rühland and Smol 1998; Michelutti et al. 2002; Antoniades et al. 2003). Sediment geochemistry The sediment geochemistry of the study ponds also tracked the effects of eiders. The gull pond reference site, D018, had the highest measured P (16 mg/g) and As (6.9 µg/g), both of which are elements associated with high bird influence (Bildstein et al. 1992; Brimble et al. 2009b). Ocean spray variables that were elevated in the water chemistry (Ca2+, Mg2+, Na+ and SO42−) were not increased in the sedimentary geochemistry of the ocean spray reference pond, D019. Most ions (Mg 2+, Na +, SO 42−) are likely elevated in the sediment due to the underlying geology as opposed to water chemistry and allochthonous sources (Lent 1994), indicating that the effects of ocean spray on tracking biovectors in water do not translate to the sediments. Calcium in sediments may be elevated from guano and uneaten mussel shells (Öst and Kilpi 1998; Ebert et al. 2013). Potentially toxic trace metals and elements known to bioaccumulate in seabirds (Braune et al. 1999; Brimble et al. 2009b) were elevated in high influence relative to the low influence sites, including Al (low influence 110 ± 830 µg/g; high influence 3100 ± 4200 µg/g), Cd (low influence 0.38 ± 0.48 µg/g; high influence 2.0 ± 1.2 µg/g), Se (low inf luence 1.1 ± 0.49 µg/g; high inf luence ponds (δ15N < 10‰) are represented by grey circles; high influence ponds (δ15N > 10‰) are represented by black triangles. Dissolved organic carbon is expressed as DOC; dissolved inorganic carbon is expressed as DIC; chlorophyll-a is expressed as CHL-a Breeding eider ducks strongly influence subarctic coastal pond chemistry 1.8 ± 0.93 µg/g), and Zn (low influence 33 ± 32 µg/g; high influence 96 ± 46 µg/g) (Table 4). As with the water chemistry, these comparisons were not statistically significant for each element due to low sample sizes. However mean values of all ornithogenic tracers (Al, As, Cd, Pb, Se, Zn, P) were higher in sediments of high and moderate influence ponds relative to low influence ponds (Sign test; p = 0.02). Additionally, many of these elements (As, Cd, Se, Zn, P) are considered guano-derived and would not normally be found elevated in the environment (Roberts et al. 2017). As with the water chemistry, sedimentary metals were far below the probable effect level of concern in sediment set by the CCME (1999). Given that eiders occupy a relatively low trophic level, feeding on filter-feeding molluscs, there is minor metal accumulations compared to other Arctic marine birds. Our results are consistent with earlier work describing species-specific differences in the effect of Arctic avian biovectors, dependent on their trophic position (Michelutti et al. 2010). Considering the increased concentrations of ornithogenic tracers in high influence sites compared to low influence sites, and the high metal concentrations in the guano, we conclude that sediment geochemistry is tracking the ornithogenic inputs. Elemental phosphorus was also elevated in the sediment of the ponds highly affected by eiders (low influence 2.0 ± 2.4 mg/g; high influence 4.0 ± 1.9 mg/g). Phosphorus in seabird guano constitutes 0.9–17% of total excrement mass (Otero et al. 2015). Following this trend, we measured high P in eider guano, which no doubt contributes to the long-term eutrophication and fertilization of the island and ponds (see Fig. 2). Axes 1 and 2 of the PCA explained 51.6% of the total variation with eigenvalues λ 1 = 0.311, and λ 2 = 0.205 (Fig. 3). The first axis was characterized by a combination of eider influence (K+) and ocean spray (Sr). Potassium is both bioenriched in sediments due to birds (Brimble et al. 2009a), as well as a major ion in ocean water. The secondary axis showed a strong gradient of Sr, which is associated with sea spray (Chen et al. 1997; Wang and Zhai 2008; ChaguéGoff 2010). Importantly, ornithogenic variables oriented together, including δ15N, Pb, Zn and P (Fig. 3). This reflects the elevated δ15N and metal concentrations typical of seabird guano. As discussed above in “Pond water chemistry”, Pb would be expected to be elevated due to hunting and in part prey preference. Zinc is an essential element that has the potential to be toxic in waterfowl (Beyer et al. 2004) and may be elevated in eiders (Burger and Gochfeld 2008; Lovvorn et al. 2013; Mallory et al. 2014). Zinc is a robust ornithogenic tracer due to its stability in sediments (Boyle 2001) and strong independent correlation to seabird inputs (Brimble et al. 2009b; Foster et al. 2011). Finally, P, as described above, was oriented with other ornithogenic variables. This Page 13 of 16 40 is due to the high P concentrations in eider guano, washing into the pond from the nest. Conclusions Our study characterized the water chemistry and sediment geochemistry of 21 ponds influenced by eiders inhabiting Hudson Strait and demonstrated a pronounced ornitholimnological influence of the eiders. Proxies of ornithogenic influence, including Pb, bioaccumulated high-trophic metals (Al, Cd, Zn), metalloid (Se), and nutrient concentrations (N, P), were all higher in ponds with larger eider abundances than ponds with few or no eiders, consistent with earlier studies on the influence of various bird species on pond chemistry (Brimble et al. 2009a, b; Michelutti et al. 2009; González-Bergonzoni et al. 2017; Roberts et al. 2017). Consequently, these ornithogenic variables support the hypothesis that eiders act as ecological engineers and biovectors, thus having the potential to structure the biota of the islands, as was reflected in much greater chlorophyll-a levels. Also, we determined that sediment geochemistry appears to better record ornithogenic variables than water chemistry because it tracks major ornithogenic proxies, including δ15N, Pb, Zn, and P. Since sedimentary geochemistry appears to faithfully track ornithogenic enrichment, this research provides additional evidence that downcore paleolimnological analyses are key tools for garnering insights into the timing of seabird arrival, colony growth and possible extirpation; information critical for conservation management. These data are key in understanding population and colony dynamics in areas for which data are typically sparse or logistically difficult to acquire. Acknowledgements Thank you to Michael Janssen and Jake RussellMercier (Wildlife Research Division, ECCC) for logistical support. Thank you to the Aiviq Hunters and Trappers Organizations’ and Ivujivik Hunters and Trappers Association for their support in our research. We especially thank Xiaowa Wang for her careful attention to the water samples. This research was funded by Environment and Climate Change Canada, the W. Garfield Weston Foundation, the Natural Sciences and Engineering Research Council of Canada (NSERC), the Pew Charitable Trusts, the Nunavut General Monitoring Plan (NGMP), and ArcticNet Network Centres of Excellence Canada. Finally, thank you to two anonymous reviewers for their very helpful feedback and comments. References 3vGeomatics Inc. (2011) Nunavut terrain and soil analysis. Government of Nunavut, Kugluktuk, Rep. No. 2010-69, pp 1–51 Antoniades D, Douglas MSV, Smol JP (2003) Comparative physical and chemical limnology of two Canadian High Arctic regions: Alert (Ellesmere Island, NU) and Mould Bay (Prince Patrick 13 40 Page 14 of 16 Island, NWT). Arch Hydrobiol 158(4):485–516. https ://doi. org/10.1127/0003-9136/2003/0158-0485 Bertilsson S, Jones JB (2003) Supply of dissolved organic matter to aquatic ecosystems: autochthonous sources. In: Findlay SEG, Sinsabaugh RL (eds) Aquatic ecosystems: interactivity of dissolved organic matter. Academic Press, San Diego, pp 3–24. https://doi. org/10.1016/B978-012256371-3/50002-0 Beyer WN, Dalgarn J, Dudding S, French JB, Mateo R, Miesner J, Sileo L, Spann J (2004) Zinc and lead poisoning in wild birds in the Tri-State Mining District (Oklahoma, Kansas, and Missouri). Arch Environ Contam Toxicol 48(1):108–117. https://doi. org/10.1007/s00244-004-0010-7 Bildstein KL, Blood E, Frederick P (1992) The relative importance of biotic and abiotic vectors in nutrient transport. Estuaries 15(2):147–157. https://doi.org/10.2307/1352688 Blais JM, Kimpe LE, McMahon D, Keatley BE, Mallory ML, Douglas MSV, Smol JP (2005) Arctic seabirds transport marine-derived contaminants. Science 309(5733):445. https://doi.org/10.1126/ science.1112658 Blais JM, Macdonald RW, Mackay D, Webster E, Harvey C, Smol JP (2007) Biologically mediated transport of contaminants to aquatic systems. Environ Sci Technol 41(4):1075–1084. https:// doi.org/10.1021/es061314a Boyle J (2001) Redox remobilization and the heavy metal record in lake sediments: a modelling approach. J Paleolimnol 26(4):423–431. https://doi.org/10.1023/A:1012785525239 Braune B, Muir D, DeMarch B, Gamberg M, Poole K, Currie R, Dodd M, Duschenko W, Eamer J, Elkin B, Evans M, Grundy S, Hebert C, Johnstone R, Kidd K, Koenig B, Lockhart L, Marshall H, Reimer K, Sanderson J, Shutt L (1999) Spatial and temporal trends of contaminants in Canadian Arctic freshwater and terrestrial ecosystems: a review. Sci Total Environ 230(1–3):145–207. https://doi.org/10.1016/S0048-9697(99)00038-8 Brazeau ML, Poulain AJ, Paterson AM, Keller W, Sanei H, Blais JM (2013) Recent changes in mercury deposition and primary productivity inferred from sediments of lakes from the Hudson Bay Lowlands, Ontario, Canada. Environ Pollut 173:52–60. https:// doi.org/10.1016/j.envpol.2012.09.017 Brimble SK, Blais JM, Kimpe LE, Mallory ML, Keatley BE, Douglas MSV, Smol JP (2009a) Bioenrichment of trace elements in a series of ponds near a northern fulmar (Fulmarus glacialis) colony at Cape Vera, Devon Island. Can J Fish Aquat Sci 66(6):949–958. https://doi.org/10.1139/F09-053 Brimble SK, Foster KL, Mallory ML, Macdonald RW, Smol JP, Blais JM (2009b) High Arctic ponds receiving biotransported nutrients from a nearby seabird colony are also subject to potentially toxic loadings of arsenic, cadmium, and zinc. Environ Toxicol Chem 28(11):2426–2433. https://doi.org/10.1897/09-235.1 Burger J, Gochfeld M (2008) Comparison of arsenic, cadmium, chromium, lead, manganese, mercury and selenium in feathers in bald eagle (Haliaeetus leucocephalus), and comparison with common eider (Somateria mollissima), glaucous-winged gull (Larus glaucescens), pigeon guillemot (Cepphus columba), and tufted puffin (Fratercula cirrhata) from the Aleutian Chain of Alaska. Environ Monit Assess 152(1–4):357–367. https://doi.org/10.1007/ s10661-008-0321-7 Canadian Council of the Ministers of the Environment (CCME) (1999) Canadian environmental quality guidelines. Canadian Council of the Ministers of the Environment, Winnipeg Cardwell RD, DeForest DK, Brix KV, Adams WJ (2013) Do Cd, Cu, Ni, Pb, and Zn biomagnify in aquatic ecosystems? Rev Environ Contam Toxicol 226:101–122. https ://doi. org/10.1007/978-1-4614-6898-1_4 Carpenter SR, Cole JJ, Kitchell JF, Pace ML (1998) Impact of dissolved organic carbon, phosphorus, and grazing on phytoplankton 13 M. P. Duda et al. biomass and production in experimental lakes. Limnol Oceanogr 43(1):73–80. https://doi.org/10.4319/lo.1998.43.1.0073 Chagué-Goff C (2010) Chemical signatures of palaeotsunamis: a forgotten proxy? Mar Geol 271(1–2):67–71. https ://doi. org/10.1016/j.margeo.2010.01.010 Chen Z, Chen Z, Zhang W (1997) Quaternary stratigraphy and traceelement indices of the Yangtze Delta, Eastern China, with special reference to marine transgressions. Quat Res 47(2):181–191. https ://doi.org/10.1006/qres.1996.1878 Clyde N (2016) Marine nutrient subsidies to the terrestrial environment of Common Eider nesting colonies in the Canadian Arctic. Dissertation, Carleton University Conover WJ (1999) The sign test. In: Practical nonparametric statistics, 3rd edn. Wiley, New York, pp 157–176 Côté G, Pienitz R, Velle G, Wang X (2010) Impact of geese on the limnology of lakes and ponds from Bylot Island (Nunavut, Canada). Int Rev Hydrobiol 95(2):105–129. https://doi.org/10.1002/ iroh.200911151 Ebert D, Hottinger JW, Pajunen VI (2013) Unsuitable habitat patches lead to severe underestimation of dynamics and gene flow in a zooplankton metapopulation. J Anim Ecol 82(4):759–769. https ://doi.org/10.1111/1365-2656.12044 Environment Canada Manual of Analytical Methods (1994a) Major ions and nutrients, vol 1. National Laboratory for Environmental Testing, Canadian Centre for Inland Waters, Burlington Environment Canada Manual of Analytical Methods (1994b) Trace metals, vol 2. National Laboratory for Environmental Testing, Canadian Centre for Inland Waters, Burlington Evans CD, Monteith DT, Cooper DM (2005) Long-term increases in surface water dissolved organic carbon: observations, possible causes and environmental impacts. Environ Pollut 137(1):55–71. https://doi.org/10.1016/j.envpol.2004.12.031 Evenset A, Christensen GN, Skotvold T, Fjeld E, Schlabach M, Wartena E, Gregor D (2004) A comparison of organic contaminants in two high Arctic lake ecosystems, Bjørnøya (Bear Island), Norway. Sci Total Environ 318(1–3):125–141. https:// doi.org/10.1016/S0048-9697(03)00365-6 Falk K, Merkel F, Kampp K, Jamieson SE (2006) Embedded lead shot and infliction rates in common eiders Somateria mollissima and king eiders S. spectabilis wintering in southwest Greenland. Wildl Biol 12(3):257–265. https://doi.org/10.2981/09096396(2006)12%5B257:ELSAIR%5D2.0.CO;2 Fast PL (2006) Nest site selection by common eiders: relationships with habitat features, microclimate and incubation success. Dissertation, University of Saskatchewan Flint PL, Grand JB (1997) Survival of spectacled eider adult females and ducklings during brood rearing. J Wildl Manage 61(1):217– 221. https://doi.org/10.2307/3802430 Foster KL, Kimpe LE, Brimble SK, Liu H, Mallory ML, Smol JP, Macdonald RW, Blais JM (2011) Effects of seabird vectors on the fate, partitioning, and signatures of contaminants in a high Arctic ecosystem. Environ Sci Technol 45(23):10053–10060. https://doi. org/10.1021/es202754h Franson JC, Hollmén T, Poppenga RH, Hario M, Kilpi M (2000) Metals and trace elements in tissues of Common Eiders (Somateria mollissima) from the Finnish archipelago. Ornis Fenn 77:57–63 Glew JR (1989) A new trigger mechanism for sediment samplers. J Paleolimnol 2:241–243 Glew JR, Smol JP (2016) A push corer developed for retrieving highresolution sediment cores from shallow waters. J Paleolimnol 56(1):67–71. https://doi.org/10.1007/s10933-015-9873-z González-Bergonzoni I, Johansen KL, Mosbech A, Landkildehus F, Jeppesen E, Davidson TA (2017) Small birds, big effects: the little auk (Alle alle) transforms high arctic ecosystems. Proc R Soc B 284:20162572. https://doi.org/10.1098/rspb.2016.2572 Breeding eider ducks strongly influence subarctic coastal pond chemistry Goudie RI, Robertson GJ, Reed A (2000) Common Eider (Somateria mollissima), version 2.0. In: Rodewald PG (ed) The birds of North America. Cornell Lab of Ornithology, Ithaca. https://doi. org/10.2173/bna.546 Government of Canada, Natural Resources Canada, Earth Sciences Sector, Canada Centre for Mapping and Earth Observation (2016) The Atlas of Canada—Toporama. http://atlas.gc.ca/toporama/en/ index.html. Accessed 1 July 2017 Grand JB, Franson JC, Flint PL, Petersen MR (2002) Concentrations of trace elements in eggs and blood of spectacled and common eiders on the Yukon–Kuskokwim Delta, Alaska, USA. Environ Toxicol Chem 21(8):1673–1678. https://doi.org/10.1002/etc.5620210819 Hamilton PB, Gajewski K, Atkinson DE, Lean DRS (2001) Physical and chemical limnology of 204 lakes from the Canadian Arctic Archipelago. Hydrobiologia 457:133–148. https ://doi. org/10.1023/A:1012275316543 Hanson PC, Bade DL, Carpenter SR, Kratz TK (2003) Lake metabolism: relationships with dissolved organic carbon and phosphorus. Limnol Oceanogr 48(3):1112–1119. https://doi.org/10.4319/ lo.2003.48.3.1112 Hargan KE, Michelutti N, Coleman K, Grooms C, Blais JM, Kimpe LE, Gilchrist G, Mallory M, Smol JP (2017) Cliff-nesting seabirds influence production and sediment chemistry of lakes situated above their colony. Sci Total Environ 576:85–98. https://doi. org/10.1016/j.scitotenv.2016.10.024 Hicklin PW, Barrow WR (2004) The incidence of embedded shot in waterfowl in Atlantic Canada and Hudson Strait. Waterbirds 27(1):41–45. https ://doi.org/10.1675/15244695(2004)027%5B0041:TIOESI%5D2.0.CO;2 Hornung RW, Reed LD (1990) Estimation of average concentration in the presence of nondetectable values. Appl Occup Environ Hyg 5(1):46–51. https://doi.org/10.1080/1047322X.1990.10389587 Houser JN, Bade DL, Cole JJ, Pace ML (2003) The dual influences of dissolved organic carbon on hypolimnetic metabolism: organic substrate and photosynthetic reduction. Biogeochemistry 64(2):247–269. https://doi.org/10.1023/A:1024933931691 Hurlbert SH, Chang CCY (1983) Ornitholimnology: effects of grazing by the Andean flamingo (Phoenicoparrus andinus). Proc Natl Acad Sci USA 80(15):4766–4769. https://doi.org/10.1073/ pnas.80.15.4766 IBM Corp (2010) IBM SPSS statistics for windows, version 19.0.0. IBM Corp, Armonk Jenkins DW (1980) Biological monitoring and surveillance. Biological monitoring of toxic metals, vol 1. US EPA Office of Research and Development, Rep. No. 600/3-80-089 Johansen P, Pedersen HS, Asmund G, Riget F (2006) Lead shot from hunting as a source of lead in human blood. Environ Pollut 142(1):93–97. https://doi.org/10.1016/j.envpol.2005.09.015 Kaufman DS (2009) An overview of late Holocene climate and environmental change inferred from Arctic lake sediment. J Paleolimnol 41(1):1–6. https://doi.org/10.1007/s10933-008-9259-6 Keatley BE, Douglas MSV, Blais JM, Mallory ML, Smol JP (2009) Impacts of seabird-derived nutrients on water quality and diatom assemblages from Cape Vera, Devon Island, Canadian High Arctic. Hydrobiologia 621(1):191–205. https://doi.org/10.1007/ s10750-008-9670-z Keatley BE, Blais JM, Douglas MSV, Gregory-Eaves I, Mallory ML, Michelutti N, Smol JP (2011) Historical seabird population dynamics and their effects on Arctic pond ecosystems: a multiproxy paleolimnological study from Cape Vera, Devon Island, Arctic Canada. Fundam Appl Limnol 179(1):51–66. https://doi. org/10.1127/1863-9135/2011/0179-0051 Kelly JF (2000) Stable isotopes of carbon and nitrogen in the study of avian and mammalian trophic ecology. Can J Zool 78(1):1–27. https://doi.org/10.1139/z99-165 Page 15 of 16 40 Lent RM (1994) Sources and cycling of major ions and nutrients in Devil’s Lake, North Dakota (Water-Resources Investigations Report, Rep. No. 94-4171). U.S. Geological Survey, Bismarck Long DT, Angino EE (1977) Chemical speciation of Cd, Cu, Pb, and Zn in mixed freshwater, seawater, and brine solutions. Geochim Cosmochim Acta 41(9):1183–1191. https://doi.org/10.1016/00167037(77)90064-3 Lovvorn JR, Raisbeck MF, Cooper LW, Cutter GA, Miller MW, Brooks ML, Grebmeier JM, Matz AC, Schafer CM (2013) Wintering eiders acquire exceptional Se and Cd burdens in the Bering Sea: physiological and oceanographic factors. Mar Ecol Prog Ser 489:245–261. https://doi.org/10.3354/meps10439 Luoto TP, Brooks SJ, Salonen V-P (2014) Ecological responses to climate change in a bird-impacted High Arctic pond (Nordaustlandet, Svalbard). J Paleolimnol 51(1):87–97. https://doi. org/10.1007/s10933-013-9757-z Mackenzie FT, Ver LM, Sabine C, Lane M, Lerman A (1993) C, N, P, S global biogeochemical cycles and modeling of global change. In: Wollast R, Mackenzie FT, Chou L (eds) Interactions of C, N, P and S biogeochemical cycles and global change, vol 4. Springer, Berlin, pp 1–61 Mallory ML (2015) Site fidelity, breeding habitats, and the reproductive strategies of sea ducks. In: Savard J-PL, Derksen DV, Esler D, Eadie JM (eds) Ecology and conservation of North American sea ducks. CRC Press, New York, pp 299–324 Mallory ML, Braune BM (2012) Tracking contaminants in seabirds of Arctic Canada: temporal and spatial insights. Mar Pollut Bull 64(7):1475–1484. https ://doi.org/10.1016/j.marpo lbul.2012.05.012 Mallory ML, Braune BM, Wayland M, Gilchrist HG, Dickson DL (2004) Contaminants in common eiders (Somateria mollissima) of the Canadian Arctic. Environ Rev 12(4):197–218. https://doi. org/10.1139/a05-004 Mallory ML, Fontaine AJ, Smith PA, Wiebe Robertson MO, Gilchrist HG (2006) Water chemistry of ponds on Southampton Island, Nunavut, Canada: effects of habitat and ornithogenic inputs. Arch Hydrobiol 166(3):411–432. https ://doi. org/10.1127/0003-9136/2006/0166-0411 Mallory ML, Braune BM, Robertson GJ, Gilchrist HG, Mallory CD, Forbes MR, Wells R (2014) Increasing cadmium and zinc levels in wild common eiders breeding along Canada’s remote northern coastline. Sci Total Environ 476–477:73–78. https ://doi. org/10.1016/j.scitotenv.2013.12.102 Mallory ML, Mahon L, Tomlik MD, White C, Milton GR, Spooner I (2015) Colonial marine birds influence island soil chemistry through biotransport of trace elements. Water Air Soil Pollut 226:31. https://doi.org/10.1007/s11270-015-2314-9 Maranger R, Caraco N, Duhamel J, Amyot M (2008) Nitrogen transfer from sea to land via commercial fisheries. Nat Geosci 1:111–112. https://doi.org/10.1038/ngeo108 Mariash HL, Smith PA, Mallory M (2018) Decadal response of arctic freshwaters to burgeoning goose populations. Ecosystems. https ://doi.org/10.1007/s10021-017-0215-z Marion L, Clergeau P, Brient L, Bertru G (1994) The importance of avian contributed nitrogen (N) and phosphorus (P) to Lake Grand-Lieu, France. Hydrobiologia 279(1):133–147. https://doi. org/10.1007/BF00027848 Michelutti N, Douglas MSV, Lean DRS, Smol JP (2002) Physical and chemical limnology of 34 ultra-oligotrophic lakes and ponds near Wynniatt Bay, Victoria Island, Arctic Canada. Hydrobiologia 482(1–3):1–13. https://doi.org/10.1023/A:1021201704844 Michelutti N, Keatley BE, Brimble S, Blais JM, Liu H, Douglas MSV, Mallory ML, Macdonald RW, Smol JP (2009) Seabird-driven shifts in Arctic pond ecosystems. Proc R Soc B 276:591–596. https://doi.org/10.1098/rspb.2008.1103 13 40 Page 16 of 16 Michelutti N, Blais JM, Mallory ML, Brash J, Thienpont J, Kimpe LE, Douglas MSV, Smol JP (2010) Trophic position influences the efficacy of seabirds as metal biovectors. Proc Natl Acad Sci USA 107(23):10543–10548. https://doi.org/10.1073/pnas.1001333107 Minagawa M, Wada E (1984) Stepwise enrichment of 15N along food chains: Further evidence and the relation between δ15N and animal age. Geochim Cosmochim Acta 48(5):1135–1140. https://doi. org/10.1016/0016-7037(84)90204-7 Mizutani H, Hasegawa H, Wada E (1986) High nitrogen isotope ratio for soils of seabird rookeries. Biogeochemistry 2(3):221–247. https://doi.org/10.1007/BF02180160 Montoya JP (2007) Natural abundance of 15N in marine planktonic ecosystems. In: Michener R, Lajtha K (eds) Stable isotopes in ecology and environmental science, 2nd edn. Blackwell Publishing Ltd, Oxford, pp 176–201. https://doi.org/10.1002/9780470691854.ch7 Mosbech A, Gilchrist G, Merkel F, Sonne C, Flagstad A, Nyegaard H (2006) Year-round movements of Northern Common Eiders Somateria mollissima borealis breeding in Arctic Canada and West Greenland followed by satellite telemetry. Ardea 94(3):651–665 Neff JC, Hooper DU (2002) Vegetation and climate controls on potential CO2, DOC, DON production in northern latitudes. Glob Change Biol 8(9):872–884. https ://doi.org/10.104 6/j.1365-2486.2002.00517.x Öst M, Kilpi M (1998) Blue mussels Mytilus edulis in the Baltic: good news for foraging eiders Somateria mollissima. Wildl Biol 4(2):81–89. https://doi.org/10.2981/wlb.1998.004 Otero XL, Tejada O, Martín-Pastor M, De La Peña S, Ferreira TO, Pérez-Alberti A (2015) Phosphorus in seagull colonies and the effect on the habitats. The case of yellow-legged gulls (Larus michahellis) in the Atlantic Islands National Park (Galicia-NW Spain). Sci Total Environ 532:383–397. https://doi.org/10.1016/j. scitotenv.2015.06.013 Otero XL, De La Peña-Lastra S, Pérez-Alberti A, Ferreira TO, HuertaDiaz MA (2018) Seabird colonies as important global drivers in the nitrogen and phosphorus cycles. Nat Commun. https://doi. org/10.1038/s41467-017-02446-8 Pella E (1990) Elemental organic analysis, part 1. Am Lab 22(2):116–125 Portnoy JW (1990) Gull contributions of phosphorus and nitrogen to a Cape Cod kettle pond. Hydrobiologia 202(1–2):61–69. https:// doi.org/10.1007/bf00027092 Ricketts TH, Dinerstein E, Olson DM, Loucks CJ, Eichbaum W, DellaSala DA, Kavanagh K, Hedao P, Hurley P, Carney K, Abell R, Walters S (1999) Terrestrial ecoregions of North America: a conservation assessment, 1st edn. Island Press, Washington, DC Roberts SJ, Monien P, Foster LC, Loftfield J, Hocking EP, Schnetger B, Pearson EJ, Juggins S, Fretwell P, Ireland L, Ochyra R, Haworth AR, Allen CS, Moreton SG, Davies SJ, Brumsack H-J, Bentley MJ, Hodgson DA (2017) Past penguin colony responses to explosive volcanism on the Antarctic Peninsula. Nat Commun. https:// doi.org/10.1038/ncomms14914 Rühland K, Smol JP (1998) Limnological characteristics of 70 lakes spanning Arctic treeline from Coronation Gulf to Great Slave Lake in the Central Northwest Territories, Canada. Int Rev Hydrobiol 83(3):183–203. https://doi.org/10.1002/iroh.19980830302 Rühland K, Paterson AM, Smol JP (2008) Hemispheric-scale patterns of climate-related shifts in planktonic diatoms from North American and European lakes. Glob Change Biol 14(11):2740–2754. https://doi.org/10.1111/j.1365-2486.2008.01670.x Sanborn-Barrie M, St-Onge MR, Young MD, James DT (2008) Bedrock geology of southwestern Baffin Island, Nunavut: expanding the tectonostratigraphic framework with relevance to mineral resources. Geological Survey of Canada, Current Research 2008-6 13 M. P. Duda et al. Sánchez-Piñero F, Polis GA (2000) Bottom-up dynamics of allochthonous input: direct and indirect effects of seabirds on islands. Ecology 81(11):3117–3132. https://doi.org/10.2307/177405 Schmutz JK, Robertson RJ, Cooke F (1983) Colonial nesting of the Hudson Bay eider duck. Can J Zool 61(11):2424–2433. https:// doi.org/10.1139/z83-324 SGS Canada Inc. (2014) Rocks to results: a practical guide to laboratory operations, 5th edn. SGS Canada Inc., Ontario Shapiro SS, Wilk MB (1965) An analysis of variance test for normality (complete samples). Biometrika 52(3–4):591–611. https://doi. org/10.1093/biomet/52.3-4.591 Smith VH (1982) The nitrogen and phosphorus dependence of algal biomass in lakes: an empirical and theoretical analysis. Limnol Oceanogr 27(6):1101–1112. https ://doi.org/10.4319/ lo.1982.27.6.1101 Smol JP (2016) Arctic and Sub-Arctic shallow lakes in a multiple stressor world: a paleoecological perspective. Hydrobiologia 778(1):253–272. https://doi.org/10.1007/s10750-015-2543-3 Smol JP, Douglas MSV (2007) Crossing the final ecological threshold in high Arctic ponds. Proc Natl Acad Sci 104(30):12395–12397. https://doi.org/10.1073/pnas.0702777104 Søndergaard M, Jensen JP, Jeppesen E (2003) Role of sediment and internal loading of phosphorus in shallow lakes. Hydrobiologia 506(1–3):135–145. https://doi.org/10.1023/B:HYDR.0000008611 .12704.dd Steenweg RJ, Crossin GT, Kyser TK, Merkel FR, Gilchrist HG, Hennin HL, Robertson GJ, Provencher JF, Flemming JM, Love OP (2017) Stable isotopes can be used to infer the overwintering locations of prebreeding marine birds in the Canadian Arctic. Ecol Evol 7(21):8742–8752. https://doi.org/10.1002/ece3.3410 Stewart EM, Michelutti N, Shenstone-Harris S, Grooms C, Weseloh C, Kimpe LE, Blais JM, Smol JP (2015) Tracking the history and ecological changes of rising double-crested cormorant populations using pond sediments from islands in Eastern Lake Ontario. PLoS ONE 10(7):e0134167. https://doi.org/10.1371/journal.pone.01341 67 Szefer P, Fowler SW, Ikuta K, Osuna FP, Ali AA, Kim B-S, Fernandes HM, Belzunce M-J, Guterstam B, Kunzendorf H, Wołowicz M, Hummel H, Deslous-Paoli M (2006) A comparative assessment of heavy metal accumulation in soft parts and byssus of mussels from subarctic, temperate, subtropical and tropical marine environments. Environ Pollut 139(1):70–78. https://doi.org/10.1016/j. envpol.2005.04.031 Tang L, Qiu R, Tang Y, Wang S (2014) Cadmium–zinc exchange and their binary relationship in the structure of Zn-related proteins: a mini review. Metallomics 6(8):1313–1323. https://doi. org/10.1039/c4mt00080c ter Braak CJF, Šmilauer P (2012) CANOCO reference manual and users guide: software for ordination (version 5.0). Microcomputer Power, Ithaca, pp 1–496 Wang G-P, Zhai Z-L (2008) Geochemical data as indicators of environmental change and human impact in sediments derived from downstream marshes of an ephemeral river, Northeast China. Environ Geol 53(6):1261–1270. https://doi.org/10.1007/s0025 4-007-0714-x Zhu R, Shi Y, Ma D, Wang C, Xu H, Chu H (2015) Bacterial diversity is strongly associated with historical penguin activity in an Antarctic lake sediment profile. Sci Rep. https://doi.org/10.1038/ srep17231 Zwolicki A, Zmudczyńska-Skarbek KM, Iliszko L, Stempniewicz L (2013) Guano deposition and nutrient enrichment in the vicinity of planktivorous and piscivorous seabird colonies in Spitsbergen. Polar Biol 36(3):363–372. https://doi.org/10.1007/s0030 0-012-1265-5