Article pubs.acs.org/est
Transformation of Mercury at the Bottom of the Arctic Food Web: An Overlooked Puzzle in the Mercury Exposure Narrative Monika Pućko,*,† A. Burt,† W. Walkusz,‡ F. Wang,† R. W. Macdonald,†,§ S. Rysgaard,†,∥,⊥,# D. G. Barber,† J.-É. Tremblay,▽ and G. A. Stern† †
Centre for Earth Observation Science, University of Manitoba, 460 Wallace Building, 125 Dysart Road, Winnipeg, R3T 2N2, Canada Institute of Oceanology, Polish Academy of Sciences (IOPAS), Powstańców Warszawy 55, 81-712 Sopot, Poland § Institute of Ocean Sciences, Department of Fisheries and Oceans, 9860 West Saanich Road, Sidney, British Columbia, V8L 4B2, Canada ∥ Department of Geological Sciences, University of Manitoba, 240 Wallace Building, 125 Dysart Road, Winnipeg, R3T 2N2, Canada ⊥ Greenland Climate Research Centre, Greenland Institute of Natural Resources, Kivioq 2, 3900 Nuuk, Greenland # Arctic Research Centre, Aarhus University, Building 1110, C. F. Møllers Alle 8, 8000 Aarhus, Denmark ▽ Québec-Océan, Département de Biologie, Université Laval, Pavillon Alexandre-Vachon, 1045 av. de la Médecine, Québec, Québec, G1 V 0A6, Canada ‡
S Supporting Information *
ABSTRACT: We show 2008 seasonal trends of total and monomethyl mercury (THg and MeHg, respectively) in herbivorous (Calanus hyperboreus) and predatory (Chaetognaths, Paraeuchaeta glacialis, and Themisto abyssorum) zooplankton species from the Canadian High Arctic (Amundsen Gulf and the Canadian Beaufort Sea) in relation to ambient seawater and diet. It has recently been postulated that the Arctic marine environment may be exceptionally vulnerable to toxic MeHg contamination through postdepositional processes leading to mercury transformation and methylation. Here, we show that C. hyperboreus plays a hitherto unrecognized central role in mercury transformation while, itself, not manifesting inordinately high levels of THg compared to its prey (pelagic particulate organic matter (POM)). Calanus hyperboreus shifts Hg from mainly inorganic forms in pelagic POM (>99.5%) or ambient seawater (>90%) to primarily organic forms (>50%) in their tissue. We calculate that annual dietary intake of MeHg could supply only ∼30% of the MeHg body burden in C. hyperboreus and, thus, transformation within the species, perhaps mediated by gut microbial communities, or bioconcentration from ambient seawater likely play overriding roles. Seasonal THg trends in C. hyperboreus are variable and directly controlled by species-specific physiology, e.g., egg laying and grazing. Zooplankton that prey on species such as C. hyperboreus provide a further biomagnification of MeHg and reflect seasonal trends observed in their prey.
■
up to 50% of the total mercury (THg).7−9 Methylation of mercury in the Arctic Ocean is primarily a microbial process performed by sulfate-reducing10 or iron-reducing11 bacteria with regulatory mechanisms not yet clearly understood.12 Once incorporated into the food web, the percentage of MeHg in THg increases with trophic level (TL), ranging from 3).6,13,14 Even though the reapportionment of MeHg between the first and the third trophic level appears crucial to the ultimate exposure of methylated Hg in top predators, this component of
INTRODUCTION
Mercury is a major contaminant of concern in the Arctic due to the toxicity of monomethylmercury (MeHg).1,2 Organic mercury biomagnifies in Arctic aquatic food webs, eventually reaching levels that exceed toxicity thresholds in many fish, birds, and top predators, such as ringed seals (Phoca hispida), beluga whales (Delphinapterus leucas), and narwhal (Monodon monoceros).2−4 These animals constitute a substantial part of traditional diet in the North and, thus, play a conveying role in mercury transfer from the ocean to the Northern people.2,5 Mercury enters the Arctic Ocean primarily in inorganic forms; i.e., elemental gaseous mercury from the atmosphere (Hg0) or oxidized divalent mercury (HgII) from snow, permafrost, soil, and freshwater.4,6 Within Arctic waters, mercury exists primarily as HgII (>90%), except in the deep ocean and oxycline, where methylated mercury (methylated Hg, sum of MeHg and dimethylmercury − DMeHg) contribute © 2014 American Chemical Society
Received: Revised: Accepted: Published: 7280
October 30, 2013 June 5, 2014 June 5, 2014 June 5, 2014 dx.doi.org/10.1021/es404851b | Environ. Sci. Technol. 2014, 48, 7280−7288
Environmental Science & Technology
Article
Teflon Niskin bottles (General Oceanics, Miami, U.S.A.), and seawater for depth profiles was collected from 10 L Teflon Niskin bottles mounted on a Rosette sampler/CTD system SBE 911plus (Sea-Bird Electronics, Bellevue, U.S.A.). Sampling was performed in accordance with clean-hands−dirty-hands protocol.20 Duplicate samples were analyzed for both THg and methylated Hg at the Class 100 cleanroom, Portable In-situ Laboratory for Mercury Speciation (PILMS) onboard the ship. THg was analyzed using cold vapor atomic fluorescence spectroscopy (CVAFS) on a Model 2600 Hg Analyzer (Tekran, Toronto, Canada) following U.S. EPA 1631.21 Total bulk concentration of methylated Hg (sum of MeHg and DMeHg in unfiltered seawater) was analyzed following U.S. EPA 163022 on an automated Hg analyzer (MERX, Brooks Rand, Seattle, U.S.A.) using gas chromatographic separation followed by CVAFS detection. Method detection limits (MDLs), defined as 3 times standard deviation (SD) of the mean laboratory blank (n = 7), were 0.05 and 0.008 ng/L for THg and methylated Hg, respectively. All field blanks were below MDL. Duplicate samples differed by less than 10%. POM Collection and Hg Analysis. As described fully in ref 23, the bottom 10 cm of ice cores were collected using a 9 cm inner diameter Mark II coring system (Kovacs Enterprises, Lebanon, U.S.A.) following the clean-hands−dirty-hands protocol.20 Ice samples scraped with Hg-clean porcelain blades were then stored and shipped at −20 °C to the laboratory at the Department of Fisheries and Oceans, Winnipeg for further analysis. There, samples were freeze-dried, and the remaining particulate carefully resuspended using a heavy metal-free synthetic seawater solution (formulated from Milli-Q water to salinity of 32; Fisher Scientific, Toronto, Canada). The resulting solutions were then filtered onto precombusted and weighed glass fiber filters (GF/F) with a nominal pore size of 0.7 μm (Whatman, Kent, U.K.), oven-dried at 60 °C, and weighed prior to analysis. Pelagic POM was collected either from the surface or from subsurface chlorophyll a maximum (chl a max) whenever present (determined by a sensor mounted on the Rosette/ CTD sampler onboard the ship). Details on chl a max depth and overall system productivity in relation to Julian day throughout the duration of this study can be found elsewhere.24 Ice-free surface water and chl a max water samples were collected following the same procedure as described above, and surface water on ice-covered stations was collected with an under-ice pump or a modified 3.5 L Trident suction (slurp) gun from ∼5 cm below the ice bottom by SCUBA divers as described elsewhere.25 Seawater samples were subsequently filtered onto precombusted and weighed glass fiber filters (GF/ F) with a nominal pore size of 0.7 μm (Whatman, Kent, U.K.), oven-dried at 60 °C, and weighed prior to analysis. THg in the POM samples was extracted using a modified hot acid aqua regia digest26 and analyzed using Cold Vapor Atomic Absorption Spectroscopy (CVAAS) following U.S. EPA 245.2.27 Reagent blanks, filter blanks, duplicates, and Certified Reference Materials (CRMs; LUTS-1, SRM 1573a, TORT-2, DORM-3; the National Institute of Standards and Technology and the National Research Council of Canada) were used for quality assurance and control. MDL was 0.001 μg/g dw (dry weight), and filter blanks were subtracted from each sample. MeHg in POM was analyzed by Flett Research Ltd. (Winnipeg, Canada) using Cold Vapor Atomic Fluorescence Spectrometry (CVAFS) with a detection limit of 0.000 15 μg/g dw, following U.S. EPA 1630.28
the food web remains vastly under-researched. Developing an understanding of the nature and seasonality of speciation, biomagnification, and transformation at this stage is, arguably, the most important step toward a better understanding of mercury fate in Arctic marine food webs where the vast majority of fish, birds, and mammals rely ultimately on herbivorous zooplankton (copepods) as a common early step in their food chain.15−18 Here, we investigate seasonal trends in concentrations and biomagnification factors (BMFs) of THg and MeHg in primary producers, and herbivorous (Calanus hyperboreus) and predator (Chaetognaths, Paraeuchaeta glacialis and Themisto abyssorum) zooplankton species (TL ≈ 1−3) in Amundsen Gulf and Canadian Beaufort Sea with emphasis on environmental and physiological controls. We pay particular attention to the role of the herbivorous copepods in supporting BMFs and controlling seasonal concentration trends in higher trophic levels.
■
EXPERIMENTAL SECTION Samples were collected onboard the CCGS Amundsen between February (33rd Julian day) and July (187th Julian day) of 2008 as part of the International Polar Year (IPY)−Circumpolar Flaw Lead (CFL) system study in the Amundsen Gulf and the Canadian Beaufort Sea19 (Figure 1 and Supporting Information
Figure 1. Cruise track as a function of Julian days in 2008 on a surface water THg concentration map (A; number indicates a Julian day), and typical vertical profiles of THg in the region, measured close to the Franklin Bay on day 138 (closed markers), and near Banks Island on day 75 (open markers) (B); BS, Beaufort Sea, AG, Amundsen Gulf.
(SI) 1S). Samples of water (surface and depth profiles) and lower trophic levels of biota (bottom-ice and pelagic particulate organic matterPOM; herbivorous zooplankton species− copepod Calanus hyperboreus; predator zooplankton species Chaetognaths including Sagitta elegans and Eukhronia hamata, copepod Paraeuchaeta glacialis, and amphipod Themisto abyssorum) were collected for measurements of total mercury (THg) and methylated forms of mercury (methylated Hg for water and MeHg for biota) at stations distributed throughout Amundsen Gulf and in the Beaufort Sea (Figure 1A). Seawater Collection and Hg Analysis. Details on seawater collection and analysis can be found elsewhere.9 Briefly, surface seawater was sampled with acid-precleaned 5 L 7281
dx.doi.org/10.1021/es404851b | Environ. Sci. Technol. 2014, 48, 7280−7288
Environmental Science & Technology
Article
Table 1. Mean ± SD of THg and MeHg at Different Trophic Levels (TL) in the Food Chain Measured for Four Different Time Periodsa TL taxa Pelagic POM
n
1.0 ± 0.0
n/a
Calanus hyperboreus
2.0 ± 0.7
16
Chaethognaths
2.6 ± 0.6
8
Paraeuchaeta glacialis
Themisto abyssorum
a
Julian days
mean ± SD
2.9 ± 0.7
2.8 ± 0.4
7
5
THg [μg/g dw] mean ± SD
160 mean ± SD 160 mean ± SD 160 mean ± SD 160 mean ± SD 160 mean ± SD
n/a 0.018 0.018 0.061 0.036 0.012 0.016 0.015 0.012 0.014 0.020 0.026 0.026 0.025 0.024 0.055 0.051 0.070 0.053 0.056 0.116 0.206 0.169 0.191 0.159
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
0.013 0.013 0.021 0.027 0.004 0.003 0.008 0.003 0.004 0.003 0.004 0.006 0.008 0.005 0.015 0.018 0.029 0.037 0.022 0.048 0.140
± 0.031 ± 0.103
MeHg [μg/g dw] n
mean ± SD
% THg
n
0 12 4 10 26 18 29 9 9 65 15 23 8 9 55 15 19 8 6 48 15 13 1 3 31
n/a
Transformation of Mercury at the Bottom of the Arctic Food Web: An Overlooked Puzzle in the Mercury Exposure Narrative Monika Pućko,*,† A. Burt,† W. Walkusz,‡ F. Wang,† R. W. Macdonald,†,§ S. Rysgaard,†,∥,⊥,# D. G. Barber,† J.-É. Tremblay,▽ and G. A. Stern† †
Centre for Earth Observation Science, University of Manitoba, 460 Wallace Building, 125 Dysart Road, Winnipeg, R3T 2N2, Canada Institute of Oceanology, Polish Academy of Sciences (IOPAS), Powstańców Warszawy 55, 81-712 Sopot, Poland § Institute of Ocean Sciences, Department of Fisheries and Oceans, 9860 West Saanich Road, Sidney, British Columbia, V8L 4B2, Canada ∥ Department of Geological Sciences, University of Manitoba, 240 Wallace Building, 125 Dysart Road, Winnipeg, R3T 2N2, Canada ⊥ Greenland Climate Research Centre, Greenland Institute of Natural Resources, Kivioq 2, 3900 Nuuk, Greenland # Arctic Research Centre, Aarhus University, Building 1110, C. F. Møllers Alle 8, 8000 Aarhus, Denmark ▽ Québec-Océan, Département de Biologie, Université Laval, Pavillon Alexandre-Vachon, 1045 av. de la Médecine, Québec, Québec, G1 V 0A6, Canada ‡
S Supporting Information *
ABSTRACT: We show 2008 seasonal trends of total and monomethyl mercury (THg and MeHg, respectively) in herbivorous (Calanus hyperboreus) and predatory (Chaetognaths, Paraeuchaeta glacialis, and Themisto abyssorum) zooplankton species from the Canadian High Arctic (Amundsen Gulf and the Canadian Beaufort Sea) in relation to ambient seawater and diet. It has recently been postulated that the Arctic marine environment may be exceptionally vulnerable to toxic MeHg contamination through postdepositional processes leading to mercury transformation and methylation. Here, we show that C. hyperboreus plays a hitherto unrecognized central role in mercury transformation while, itself, not manifesting inordinately high levels of THg compared to its prey (pelagic particulate organic matter (POM)). Calanus hyperboreus shifts Hg from mainly inorganic forms in pelagic POM (>99.5%) or ambient seawater (>90%) to primarily organic forms (>50%) in their tissue. We calculate that annual dietary intake of MeHg could supply only ∼30% of the MeHg body burden in C. hyperboreus and, thus, transformation within the species, perhaps mediated by gut microbial communities, or bioconcentration from ambient seawater likely play overriding roles. Seasonal THg trends in C. hyperboreus are variable and directly controlled by species-specific physiology, e.g., egg laying and grazing. Zooplankton that prey on species such as C. hyperboreus provide a further biomagnification of MeHg and reflect seasonal trends observed in their prey.
■
up to 50% of the total mercury (THg).7−9 Methylation of mercury in the Arctic Ocean is primarily a microbial process performed by sulfate-reducing10 or iron-reducing11 bacteria with regulatory mechanisms not yet clearly understood.12 Once incorporated into the food web, the percentage of MeHg in THg increases with trophic level (TL), ranging from 3).6,13,14 Even though the reapportionment of MeHg between the first and the third trophic level appears crucial to the ultimate exposure of methylated Hg in top predators, this component of
INTRODUCTION
Mercury is a major contaminant of concern in the Arctic due to the toxicity of monomethylmercury (MeHg).1,2 Organic mercury biomagnifies in Arctic aquatic food webs, eventually reaching levels that exceed toxicity thresholds in many fish, birds, and top predators, such as ringed seals (Phoca hispida), beluga whales (Delphinapterus leucas), and narwhal (Monodon monoceros).2−4 These animals constitute a substantial part of traditional diet in the North and, thus, play a conveying role in mercury transfer from the ocean to the Northern people.2,5 Mercury enters the Arctic Ocean primarily in inorganic forms; i.e., elemental gaseous mercury from the atmosphere (Hg0) or oxidized divalent mercury (HgII) from snow, permafrost, soil, and freshwater.4,6 Within Arctic waters, mercury exists primarily as HgII (>90%), except in the deep ocean and oxycline, where methylated mercury (methylated Hg, sum of MeHg and dimethylmercury − DMeHg) contribute © 2014 American Chemical Society
Received: Revised: Accepted: Published: 7280
October 30, 2013 June 5, 2014 June 5, 2014 June 5, 2014 dx.doi.org/10.1021/es404851b | Environ. Sci. Technol. 2014, 48, 7280−7288
Environmental Science & Technology
Article
Teflon Niskin bottles (General Oceanics, Miami, U.S.A.), and seawater for depth profiles was collected from 10 L Teflon Niskin bottles mounted on a Rosette sampler/CTD system SBE 911plus (Sea-Bird Electronics, Bellevue, U.S.A.). Sampling was performed in accordance with clean-hands−dirty-hands protocol.20 Duplicate samples were analyzed for both THg and methylated Hg at the Class 100 cleanroom, Portable In-situ Laboratory for Mercury Speciation (PILMS) onboard the ship. THg was analyzed using cold vapor atomic fluorescence spectroscopy (CVAFS) on a Model 2600 Hg Analyzer (Tekran, Toronto, Canada) following U.S. EPA 1631.21 Total bulk concentration of methylated Hg (sum of MeHg and DMeHg in unfiltered seawater) was analyzed following U.S. EPA 163022 on an automated Hg analyzer (MERX, Brooks Rand, Seattle, U.S.A.) using gas chromatographic separation followed by CVAFS detection. Method detection limits (MDLs), defined as 3 times standard deviation (SD) of the mean laboratory blank (n = 7), were 0.05 and 0.008 ng/L for THg and methylated Hg, respectively. All field blanks were below MDL. Duplicate samples differed by less than 10%. POM Collection and Hg Analysis. As described fully in ref 23, the bottom 10 cm of ice cores were collected using a 9 cm inner diameter Mark II coring system (Kovacs Enterprises, Lebanon, U.S.A.) following the clean-hands−dirty-hands protocol.20 Ice samples scraped with Hg-clean porcelain blades were then stored and shipped at −20 °C to the laboratory at the Department of Fisheries and Oceans, Winnipeg for further analysis. There, samples were freeze-dried, and the remaining particulate carefully resuspended using a heavy metal-free synthetic seawater solution (formulated from Milli-Q water to salinity of 32; Fisher Scientific, Toronto, Canada). The resulting solutions were then filtered onto precombusted and weighed glass fiber filters (GF/F) with a nominal pore size of 0.7 μm (Whatman, Kent, U.K.), oven-dried at 60 °C, and weighed prior to analysis. Pelagic POM was collected either from the surface or from subsurface chlorophyll a maximum (chl a max) whenever present (determined by a sensor mounted on the Rosette/ CTD sampler onboard the ship). Details on chl a max depth and overall system productivity in relation to Julian day throughout the duration of this study can be found elsewhere.24 Ice-free surface water and chl a max water samples were collected following the same procedure as described above, and surface water on ice-covered stations was collected with an under-ice pump or a modified 3.5 L Trident suction (slurp) gun from ∼5 cm below the ice bottom by SCUBA divers as described elsewhere.25 Seawater samples were subsequently filtered onto precombusted and weighed glass fiber filters (GF/ F) with a nominal pore size of 0.7 μm (Whatman, Kent, U.K.), oven-dried at 60 °C, and weighed prior to analysis. THg in the POM samples was extracted using a modified hot acid aqua regia digest26 and analyzed using Cold Vapor Atomic Absorption Spectroscopy (CVAAS) following U.S. EPA 245.2.27 Reagent blanks, filter blanks, duplicates, and Certified Reference Materials (CRMs; LUTS-1, SRM 1573a, TORT-2, DORM-3; the National Institute of Standards and Technology and the National Research Council of Canada) were used for quality assurance and control. MDL was 0.001 μg/g dw (dry weight), and filter blanks were subtracted from each sample. MeHg in POM was analyzed by Flett Research Ltd. (Winnipeg, Canada) using Cold Vapor Atomic Fluorescence Spectrometry (CVAFS) with a detection limit of 0.000 15 μg/g dw, following U.S. EPA 1630.28
the food web remains vastly under-researched. Developing an understanding of the nature and seasonality of speciation, biomagnification, and transformation at this stage is, arguably, the most important step toward a better understanding of mercury fate in Arctic marine food webs where the vast majority of fish, birds, and mammals rely ultimately on herbivorous zooplankton (copepods) as a common early step in their food chain.15−18 Here, we investigate seasonal trends in concentrations and biomagnification factors (BMFs) of THg and MeHg in primary producers, and herbivorous (Calanus hyperboreus) and predator (Chaetognaths, Paraeuchaeta glacialis and Themisto abyssorum) zooplankton species (TL ≈ 1−3) in Amundsen Gulf and Canadian Beaufort Sea with emphasis on environmental and physiological controls. We pay particular attention to the role of the herbivorous copepods in supporting BMFs and controlling seasonal concentration trends in higher trophic levels.
■
EXPERIMENTAL SECTION Samples were collected onboard the CCGS Amundsen between February (33rd Julian day) and July (187th Julian day) of 2008 as part of the International Polar Year (IPY)−Circumpolar Flaw Lead (CFL) system study in the Amundsen Gulf and the Canadian Beaufort Sea19 (Figure 1 and Supporting Information
Figure 1. Cruise track as a function of Julian days in 2008 on a surface water THg concentration map (A; number indicates a Julian day), and typical vertical profiles of THg in the region, measured close to the Franklin Bay on day 138 (closed markers), and near Banks Island on day 75 (open markers) (B); BS, Beaufort Sea, AG, Amundsen Gulf.
(SI) 1S). Samples of water (surface and depth profiles) and lower trophic levels of biota (bottom-ice and pelagic particulate organic matterPOM; herbivorous zooplankton species− copepod Calanus hyperboreus; predator zooplankton species Chaetognaths including Sagitta elegans and Eukhronia hamata, copepod Paraeuchaeta glacialis, and amphipod Themisto abyssorum) were collected for measurements of total mercury (THg) and methylated forms of mercury (methylated Hg for water and MeHg for biota) at stations distributed throughout Amundsen Gulf and in the Beaufort Sea (Figure 1A). Seawater Collection and Hg Analysis. Details on seawater collection and analysis can be found elsewhere.9 Briefly, surface seawater was sampled with acid-precleaned 5 L 7281
dx.doi.org/10.1021/es404851b | Environ. Sci. Technol. 2014, 48, 7280−7288
Environmental Science & Technology
Article
Table 1. Mean ± SD of THg and MeHg at Different Trophic Levels (TL) in the Food Chain Measured for Four Different Time Periodsa TL taxa Pelagic POM
n
1.0 ± 0.0
n/a
Calanus hyperboreus
2.0 ± 0.7
16
Chaethognaths
2.6 ± 0.6
8
Paraeuchaeta glacialis
Themisto abyssorum
a
Julian days
mean ± SD
2.9 ± 0.7
2.8 ± 0.4
7
5
THg [μg/g dw] mean ± SD
160 mean ± SD 160 mean ± SD 160 mean ± SD 160 mean ± SD 160 mean ± SD
n/a 0.018 0.018 0.061 0.036 0.012 0.016 0.015 0.012 0.014 0.020 0.026 0.026 0.025 0.024 0.055 0.051 0.070 0.053 0.056 0.116 0.206 0.169 0.191 0.159
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
0.013 0.013 0.021 0.027 0.004 0.003 0.008 0.003 0.004 0.003 0.004 0.006 0.008 0.005 0.015 0.018 0.029 0.037 0.022 0.048 0.140
± 0.031 ± 0.103
MeHg [μg/g dw] n
mean ± SD
% THg
n
0 12 4 10 26 18 29 9 9 65 15 23 8 9 55 15 19 8 6 48 15 13 1 3 31
n/a
