Pan-Arctic River Fluxes of Polychlorinated Biphenyls - Environmental

ARTICLE pubs.acs.org/est

Pan-Arctic River Fluxes of Polychlorinated Biphenyls € Daniel Carrizo and Orjan Gustafsson* Department of Applied Environmental Science (ITM), Stockholm University, 106 91 Stockholm, Sweden

bS Supporting Information ABSTRACT: Observations of polychlorinated biphenyls (PCB) concentrations in fluvial surface sediments near the mouths of the six Great Arctic Rivers (GARs; Ob, Yenisey, Lena, Indigirka, Kolyma, and Mackenzie) were combined with annual dissolved organic carbon (DOC) and particulate organic carbon (POC) loadings and hydraulic discharge to estimate the pan-Arctic river flux of PCBs. The highest total-phase fluxes of ∑13PCB were found for the Ob River, with 184 kg/yr and the smallest for the Indigirka River with 3.9 kg/yr. Consistent with a continent-scale trend among the Eurasian GARs of increasing POC concentrations eastward, which is extending to the North American Mackenzie River, a general shift in the estimated PCB partitioning from dissolved to particle-associated flux was found toward the east. Pentachlorinated and hexachlorinated PCBs constituted the majority (>70%) of the total PCB fluxes in the Eurasian Rivers. In contrast, trichlorinated and tetrachlorinated congeners were the most abundant in the Mackenzie (≈ 75%). The total ∑13PCB fluxes from the pan-Arctic rivers are here estimated to be ∼0.4 tonne/yr. This is geochemically consistent with the inventory of total PCBs in the Polar Mixed Layer of the entire Arctic Ocean (0.39 tonne) and about a factor 2 less than two new estimates of the PCB settling export to Arctic subsurface waters. Hence, the yearly Great Arctic River PCB fluxes only represent 0.001% of the historical PCB emission into the global environment. To our knowledge, this is the first estimate of circum-Arctic river flux of any organic pollutant based on a comprehensive investigation of the pollutants in several rivers and it contributes toward a more complete understanding of largescale contaminant cycling in the Arctic.

’ INTRODUCTION Persistent organic pollutants (POPs) such as polychlorinated biphenyls (PCBs) have physicochemical properties and structures that render them semivolatile, lipophilic and resistant to degradation. As a consequence of these characteristics, PCBs are dispersing through the global environment 1,2 and tend to accumulate in biota (e.g., refs 3 7). Diverse studies suggest that PCBs continue to be one of the compound classes of highest ecotoxicological concern for the Arctic Ocean ecosystem (e.g., refs 7 and 8). Since uptake in the base of the aquatic food web is governed by PCB loadings in the abiotic compartments, where the vast majority of the PCBs in the Arctic reside, it is of interest to improve our understanding of the magnitude of the different delivery pathways to the Arctic of POPs such as PCBs. The Arctic Ocean receives input of PCBs through tropospheric circulation, surface ocean currents and fluvial discharge by the many northward-draining rivers emptying across the Arctic rim. None of these three PCB flux vectors have to our knowledge been quantitatively assessed on the pan-Arctic scale using real observations. This study addresses the pan-Arctic river flux of PCBs. Freshwater river discharge is a characteristic feature of the Arctic Ocean (AO). While the AO only constitutes 1% of the World Ocean volume, it receives 11% of the global input of freshwater and organic matter.9 The pan-Arctic drainage area of the six GARs combined is 10.4 106 km2, which is more than twice the area of the European Union. Their combined hydraulic r 2011 American Chemical Society

discharge of 1.9 103 km3/yr makes the pan-Arctic catchment one of the most important river systems in the World. While there have been a few pioneering studies of PCB discharge into the Arctic from single rivers,10 12 there is a large paucity of information on both the quantitative magnitude and congener distribution of this putatively important vector of PCB delivery on the pan-Arctic scale, particularly for the Siberian Arctic rivers. The objective of this investigation is to provide an empirical estimate of the discharge of PCBs into the Arctic Ocean from the six Greatest Arctic Rivers (the GARs; Ob, Yenisey, Lena, Indigirka, Kolyma, and Mackenzie). This is achieved by combining actual measurements of PCB congener abundance near the mouth of each river with information on the annual hydraulic discharge and its load of dissolved and particulate organic carbon (DOC and POC) within a three-phase partitioning framework for individual congeners.

’ MATERIALS AND METHODS Study Area and Biogeochemical Characteristics of the Great Arctic Rivers. This assessment includes the five largest Received: May 24, 2011 Accepted: August 24, 2011 Revised: August 19, 2011 Published: August 24, 2011 8377

dx.doi.org/10.1021/es201766z | Environ. Sci. Technol. 2011, 45, 8377–8384

Environmental Science & Technology

ARTICLE

Figure 1. PCB river fluxes (kg/yr) partitioned into three phases (circle diagram); POC-associated (blue), DOC-associated (red) and dissolved (green). The size of the circle is proportional to the total-phase PCB discharge flux (number is given on white background). The thickness of the blue arrows is proportional to hydraulic water discharge and the thickness of the brown arrows is proportional to river POC discharge. The shaded/transparent blue areas are marking the drainage basins of the rivers.

Siberian Arctic rivers (i.e., Ob, Yenisey, Lena, Indigirka, and Kolyma) and the largest North American Arctic river (i.e., Mackenzie) (Figure 1). The Eurasian study sites stretch over a climosequence with decreasing annual precipitation and surface air temperatures from west to east across northern Siberia.13 Average summer temperatures are similar throughout the Eurasian study areas (+7 °C to +9 °C) whereas average winter temperatures reach differences of 20 degrees between West Siberia and North America ( 20 °C) versus that in East Siberia ( 40 °C). Based on varying meteorological conditions, topography and mountainous orology the Eurasian area can be divided in two regions influenced by two separate atmospheric circulations patterns. The area located to the west of the Lena basin is under the influence of Atlantic air masses (coming across Europe and western Russia), while the area east of the Lena is influence primarily by Pacific air masses (coming across eastern Russia and China) (e.g., refs 14 and 15). The Mackenzie region is under the influence of air masses primary originated in the Pacific and North American sectors, although other air masses (i.e., back trajectories from China and Russia) can here also have a minor influence. This pattern is predominant during the year and

sharpest during the colder months. It has been shown that air masses originating in the Russian sector have a major proportion of tetra-CBs, penta-CBs, and hexa-CBs whereas the opposite pattern occurs when the air masses originate in the North American/Pacific sector with di-CBs and tri-CBs as the major contributors.16 The six Great Arctic Rivers (GARs) have a combined freshwater discharge around 1935 km3/yr, by far the greatest portion of the total freshwater discharge to the Arctic Ocean. The six rivers have a combined catchment area of 10.4 106 km2, with different drainage basin areas, hydrology and water discharge characteristics (Supporting Information Table S1). The watersheds of the two west Siberian Rivers (Ob and Yenisey), draining the world’s largest wetland, the West Siberian Peatland17 are mainly located in the region of nonpermafrost and/or (discontinuous) permafrost islands and flow into Kara Sea. The East Siberian Rivers are mainly located in the continuous permafrost region that covers approximately half of the territory of Russia and flow into the Laptev Sea (Lena) and the East Siberian Sea (Indigirka and Kolyma). The Kolyma and Indigirka rivers have the smallest drainage areas of all the GARs, their headwater is 8378

dx.doi.org/10.1021/es201766z |Environ. Sci. Technol. 2011, 45, 8377–8384

Environmental Science & Technology almost entirely within the Arctic and drain almost exclusively continuous permafrost regions. In contrast, Ob, Yenisey, Lena, and Mackenzie begin their flow far south of the Arctic Region (Figure 1). These vastly different permafrost coverages affect the hydrological flowpaths, with the potential for more subsurface flow in the discontinuous zone compared to a flow predominantly in the active surface layer as would be expected for the East Siberian rivers (e.g., refs 18 and 19). The Arctic Rivers become free of ice in early summer and discharge about 90% of the annual delivery from May to July. Organic matter concentrations in the Arctic Rivers are among the highest of the world’s major rivers. The dissolved organic carbon (DOC) concentrations generally exceed particulate organic carbon (POC), with the exception of the Mackenzie River, which is exceptionally rich in suspend matter and has comparable concentrations of DOC and POC20 (Supporting Information Table S1). There is also a clear trend of increasing POC:DOC ratio eastward among the Siberian GARs consistent with drainage reliefs of the rivers. The large export of DOC and POC in the GARs reflects organic-matter-rich drainage areas of the pan-Arctic.21 Sampling Procedure. Surface sediment samples from the Lena, Indigirka, and Kolyma Rivers were collected using a stainless steel van Veen grab sampler (dimensions 20 30 cm), during the R/V Ivan Kireev September 2004 expedition in the Laptev and East Siberian Seas. Typically, kg-sized sediment samples were collected (for compound-specific radiocarbon analysis performed in parallel projects); for this reason, in the Indigirka river, four samples were combined. Surface sediment samples from the Ob and Yenisey rivers were similarly obtained during the R/V Ivan Kireev September 2005 expedition in the Kara Sea. The sediments were all taken from a central location of the near-coastal river plume. The pooled samples from the Indigirka River were taken from 8 to 11 m water depths while all the others river samples were taken at 1 2 m water depths. In all the cases sediment integrity was first visually inspected for an undisturbed sediment-water interface before approximately 0 2 cm was carefully subsampled manually, put into precombusted glass jars and then kept frozen at 20 °C until analysis in the laboratory. A detailed description of sampling and sediment characteristics for the Siberian rivers can be found in van Dongen et al.19 and Elmquist et al.22 One surface sediment sample from the Mackenzie were collected and sub sampled from the riverbed about 5 10 m away from the riverbank using a stainless steel hand shovel during June 2004, as described elsewhere.22,23 Extraction and Analysis. Concentrations of 15 PCB congeners (PCB IUPAC numbers 18, 28, 52, 70, 90/101, 110, 118, 105, 149, 153, 138, 180, 199, and 194) were determined in the surface sediment samples following previously described methods (e.g., ref 24 and references therein). Briefly, internal standards in the form of seven 13C-labeled PCB congeners were added to each sample prior to 24 h Soxhlet extraction in preextracted cellulose thimbles with toluene (glass-distilled quality; Burdick & Jackson, Fluka Chemie AG, Buchs, Switzerland) using a Dean Stark trap for collection of water. All extracts were eluted on an open silica column, containing three layers of modified silica (SiO2/H2SO4 3 cm, SiO2/KOH 3 cm, and SiO2/H2O 3 cm) with hexane as a mobile phase. Activated copper was added to this extract to remove sulfur and the suspension was left overnight. Finally, samples were quantified on a HP6890 (HewlettPackard, Avondale, PA) gas chromatograph (GC) equipped with a PTE-5 capillary column (30 m 0.25 mm i.d., 0.25 μm film thickness; Supelco Inc., Bellefonte, PA) with a high-resolution

ARTICLE

mass spectrometer (HRMS) (Autospec Ultima; Micromass, Altrincham, UK) operated in the electron impact mode. A 13C labeled standard (PCB 153) was added to all samples before injection on the GC-HRMS in order to check the sensitivity of the instrument during the sample injections. Flux Calculations. Annual river fluxes (kg/yr) for individual PCB congeners were calculated for the dissolved, DOC associated and POC-associated as well as for the total phases for each of the six GARs (equations listed in the Supporting Information). The coastal particle water distribution of PCBs determined for the pan-Arctic shelf seas and reported elsewhere25 were used for the ambient PCB distribution coefficients between POC-seawater (Kpoc) and DOC-seawater (Kdoc) in these calculations. A previous year-round study in boreal to subArctic surface waters of the open Baltic Sea yielded near-constant values of Kpoc for the different seasons.26 The PCB river fluxes were then obtained by combining these distribution coefficients with actual OC-normalized concentrations of individual congeners from the riverine deposit in surface sediments off each river mouth (pg-PCB/gdw-OC; Supporting Information Table S2), together with literature-based information on hydraulic discharge with associated DOC and POC loadings for each river (Supporting Information Table S1). This approach thus seeks to overcome the large seasonal variability by either probing matrices that integrates over supra-annual scales (coastal sediments) and/ or integrated monthly discharge data over a full year. While logistically quite challenging, future studies may attempt to obtain monthly resolved ambient three-phase partitioned PCB concentrations synoptically in each Great Arctic River to combine with monthly riverine DOC, POC, and hydraulic discharge. Consideration of Uncertainties. We sought to estimate the uncertainties involved in this first assessment of the pan-Arctic PCB river fluxes by considering a number of factors. The analytical uncertainties (based on >70 separate analyses for PCBs in Arctic samples) were combined with uncertainties in the employed values for hydraulic discharge, DOC and POC loadings. The uncertainties of those biogeospheric parameters for each river were based on either estimates from the literature, when available, or based on expert judgment. We assume an analytical uncertainty of 5% (rsd) for the POC, 10% (rsd) for the DOC of and 10% (rsd) for the water discharge values.9,15,19 The propagated uncertainties for these phase-specific PCB fluxes for each river are detailed in the Supporting Information Tables S2 S5. It is quite likely that uncertainties are largest in our understanding of the natural system such as in the seasonal variability in river geochemistry and particle transport and how that may impact the partitioning and fluxes of the PCBs. For instance, the uncertainty associated with the assumption that the OC-normalized PCB concentration in the surface sediment floc outside the river is representative of OC-normalized PCB concentration in riverexported POC is challenging to constrain quantitatively. The here measured OC-normalized PCB concentrations for the circumArctic is in the same range as other reports for individual systems (Supporting Information Table S7 and discussion below).

’ RESULTS AND DISCUSSION Characteristics and PCB Content of Riverine Surface Sediments. The organic matter in the top surface sediments col-

lected off or near the mouths of these Arctic rivers are overwhelmingly of riverine-terrestrial origin. This is evidenced by their composition of both lipid biomarkers and carbon isotopes 8379



ARTICLE

Figure 2. The relative congener contributions (grouped by chlorination degree) to the PCB fluxes of the six Great Arctic Rivers.

(δ13C and Δ14C).18,19,22,27 The estuaries of large rivers act as natural filters to scavenge riverine DOC and POC and their associated constituents into the underlying sediments.28,29 The polyelectrolytic negatively charged riverine DOC and POC flocculate in the higher ionic strength coastal waters and settle as the current speeds decrease. Hence, the surface sediments off rivers provide a useful matrix for studying the composition of river-exported substances such as PCBs (e.g., refs 30 32). The river-derived sediment organic carbon (SOC) content measured off the six Arctic river mouths, ranged from 9.2 and 5.7 mg OC/ gdw off the Ob and Lena River, respectively, to somewhat higher values in the other rivers, yet contained within a narrow range of 18.1 20.2 mg OC/gdw (Supporting Information Table S1). The OC-normalized sediment PCB concentrations ranged from 17 to 67 ng/gOC ∑13PCBs, with the highest OC-normalized value for the OC-lean sediments off the Lena and the lowest for the Indigirka River (Supporting Information Table S2). To assess the representativeness of these observations, these were compared with other reports of OC-normalized PCB concentrations for coastal surface sediments around the Arctic (SI Table S7). There were good agreement between the PCB concentration reports, which thus supports the validity of using the current observations as a basis for the GAR flux estimations. Distinctly different congener patterns were found in the Eurasian rivers compared with the North American Mackenzie River (Figure 2). While the Eurasian GARs were dominated by penta- to

hexachlorinated congeners (maximum concentration value for PCB#153 and PCB#138), the Mackenzie was to three-quarters composed of tri- to tetrachlorinated PCBs (i.e., PCB#28 and PCB#52). (Figure 2; Supporting Information Table S2). Riverine PCB Fluxes to the Arctic Ocean. The total annual ∑13PCB fluxes of the six Great Arctic Rivers were 415 kg/yr (Supporting Information Table S3). The phase-specific and total PCB fluxes for the different rivers are in part related to the hydraulic and geochemical characteristics of the different rivers. The highest total-phase ∑13PCB fluxes were found for the Ob (183 kg/yr) and Lena (113 kg/yr). These two rivers also have the highest annual water discharge. Furthermore, their extensive catchment areas hold important population/industrial areas. The Ob is also the GAR that is geographically closest to centralnorthern Europe, which is one of the hotspots of historical PCB consumption.33 The Mackenzie (60 kg/yr) and the Yenisey (45 kg/yr) followed in PCB river flux magnitude (Figure 1). Our estimates suggest that the smallest PCB river flux contributions among the six GARs come from the Kolyma and Indigirka rivers (10 and 3.9 kg/yr, respectively). This is consistent with these two rivers being located in the remote northeast Siberia, draining relatively small drainage basins that are further away from population/industrial centers than for the other GARs. The total annual PCB fluxes entering the Arctic Ocean by the Great Arctic Rivers only correspond to (∼0.001%) of the total PCBs that have been historically emitted to the environment.34 8380


Environmental Science & Technology There is a clear trend in the phase-distribution of the total PCB flux across the Eurasian GARs. The dissolved fraction goes from three-quarters in the west Siberian rivers Ob and Yenisey to about one-half in the Lena to drop to about one-fourth in the east Siberian rivers Indigirka and Kolyma. This naturally follows from the inverse continent-scale trend in DOC and particularly POC concentrations that are the highest in the east. This, again, follows from the steeper relief of the eastern drainage basins, draining the east Siberian Highlands compared to the West Siberian Lowlands (Figure 1). The Mackenzie delivers about half of the total PCB river flux in the dissolved fraction. Individual congener and phase-specific (dissolved, DOC and POC) fluxes for each river are detailed in Supporting Information Tables S4 S6. The highest fluxes were in the dissolved-associated phase for the Ob, Yenisey, and Lena rivers, but this trend changed to the POC-associated phase for the Kolyma, Indigirka, and Mackenzie rivers. Congener Distribution of the Estimated Pan-Arctic Rivers PCB Flux. There are three spatially distinct patterns in congener contributions (Figure 2). The first one corresponds to the western Eurasian rivers (i.e., Ob and Yenisey), where the distribution is skewed toward higher contribution of the penta-CBs (≈50%) and hexa-CBs (≈30%). The second pattern corresponds to the Eastern Eurasian Rivers (i.e., Lena, Indigirka, and Kolyma) with a high proportion of the hexa-CBs (≈60%) and penta-CBs (≈30%). The third congener distribution pattern corresponds to the Mackenzie River with equal proportion of triCBs and tetra-CBs (≈35% each). The higher-chlorinated congener contribution (hexa-CB and penta-CB) in the Eurasian rivers may reflect the composition of technical PCB mixtures produced and heavily used in the former Soviet Union (USSR), represented by around 100 000 tonnes of Sovol (≈50% pentaCBs) and 25 000 tonnes of Trichlorodiphenyl (mainly tri-CBs).35 A similar congener pattern has been reported for air samples of the coastal Russian Arctic16 when the air masses originated in the Russian sector. Moreover, in air samples taken in the Lake Baikal region, a congener pattern corresponding to a mixture of these two Soviet technical mixtures has been found.35 37 The higher proportion of lighter PCB congeners in the Mackenzie River can be explained by atmospheric deposition after long-range transport to this Arctic region. A similar low-chlorinated PCB profile was observed in early work by Stern et al.16 on air samples and by Macdonald et al.,12 on air samples from the Canadian Arctic. This congener pattern was presumably related to air masses originating in the North American and Pacific sectors. We have previously reported that the PCB congener distribution in the North Pole area is dominated by tri- to tetrachlorinated PCBs24 and suggested that this is consistent with vaporpressure driven preferential scavenging of higher-chlorinated PCBs during long-range atmospheric transport. An alternative or complementary process becoming apparent from the present work is preferential removal of higher-chlorinated DOC- and POC-associated PCBs during estuarine scavenging, leaving lowerchlorinated congeners, relatively more partitioned to the dissolved phase, to be preferentially transported with the Transpolar Drift to the central basins of the Arctic Ocean. Hence, a low-chlorinated congener profile in the Arctic Ocean may result from either atmospheric or riverine delivery modes. Comparison with Other Estimates of PCB Fluxes in Arctic Rivers. There is a scarcity of previous reports on PCB river fluxes in the Arctic to compare the current results with. Carroll et al.10 reported a mean flux of dissolved ∑10PCBs of 63 kg/yr for the Ob

ARTICLE

and Yenisei rivers combined into the Kara Sea. The present study estimated dissolved ∑13PCB fluxes for Ob of 144 kg/yr and Yenisey of 32 kg/yr. In spite of that study using a completely different approach, the fluxes were quantified as a product of annual river discharge and the dissolved PCB concentration at zero salinity that can be highly variable, the two estimates are within a factor of 3. In contrast, Kimstach and Dutchak11 reported a total-phase ∑14PCB river export flux of 3760 kg/yr (1250 kg/yr dissolved and 2510 kg/yr particulate) for the Yenisei River alone, which is a factor 120 higher than the present study. In that study annual fluxes were calculated using hydrological discharge measurements and water sampling at stations all along the river (11 stations) during four typical hydrological regimes. Analytical data of PCBs concentration were obtained in 11 pooled water samples and 11 pooled suspend matter samples over 3-day periods. That paper provides no information about QA/QC protocols during sampling and handling. Finally, Macdonald et al. 200012 estimated a ∑69PCB flux of 394 kg/yr for all the Canadian rivers flowing to the Arctic Ocean. In that work the fluxes were calculated as a product of PCBs concentration per unit mass of suspended sediment, suspended sediments mass concentration, PCBs dissolved concentration and volume flow rate. As in the Carroll et al. study, the terms in this equation are highly variable and the estimated fluxes are only approximate because measurements taken during high flow have been used and the amounts as well as the proportions transported as dissolved or suspended phases vary. The present study estimates a somewhat lower flux albeit just for the Mackenzie River and for a summation of much fewer congeners (60 kg/yr). An additional reason for the discrepancy may be that the Macdonald et al. study relied on PCB concentration data from the early 1990s when the PCB concentrations may have been higher than in year 2005 when our Mackenzie sample was collected. Pan-Arctic River Fluxes in View of Other Estimates of Large-Scale PCB Fluxes and Inventories in the Arctic. To allow comparison with other estimates of large-scale inventories and fluxes in Arctic Ocean compartments (Table 1), the river flux in the present study was recalculated on the ∑7-ICES PCB basis and found to be 0.30 tonne/yr for the six Pan-Arctic GARs. The pan-Arctic river fluxes were then compared with burial fluxes of ∑7-ICES PCBs in the pan-arctic shelf sediments. J€onsson et al.2 collated a large database of PCB concentrations in the mixed surface sediment of the pan-arctic shelf seas (the world’s largest continental shelf system) and used geophysical information to estimate a burial flux below the mixed layer of 2.6 tonne/yr. In this perspective it should be realized that the current pan-arctic river fluxes are not extrapolated to the total flux from all panarctic rivers. Second, the sediment burial fluxes are operating over a longer time scale (integrating over the several decades long residence time in the mixed sediment layer) and are thus expected to be higher as it includes the periods when the PCB levels in the environment were (much) higher. The river ∑7-ICES PCB flux (0.30 tonne/yr) can also be compared with the ∑7-ICES PCB inventory of 0.26 tonne estimated for the surface seawater compartment of the entire Arctic Ocean (Polar Mixed Layer-PML). The PML inventory was estimated based on trace-clean sampling, handling and analytical protocols for dissolved and particulate PCBs from the seven major Arctic shelf seas and the Central Arctic Ocean Basin obtained on three basin-wide expeditions in the 2000s.25 The PML receives input also via long-range atmospheric and ocean 8381



ARTICLE

Table 1. Inventories and Fluxes of PCB in Global and Arctic Compartments (∑7 ICES Congeners) global emissions (tonne)

29 319

ref 33

global shelf sediment inventories (tonne)

5794

ref 2

Arctic shelf sediment inventories (tonne)

136

ref 2

Arctic shelf sediment burial flux (tonne/yr)

2.8

ref 2

model estimated Arctic Ocean flux to deep

7 70.7

ref 39

Arctic Ocean PML inventory (tonne)

0.26

ref 25

observation-based estimate of Arctic Ocean flux to deep waters (tonne/yr)

0.74

ref 25,39a

observation-234 Th based estimate of Arctic Ocean

0.5

ref 25,38

0.30

this work

waters (tonne/yr)

flux to deep waters (tonne/yr) Arctic river flux (tonne/yr) a

Using observational data of POC-associated PCB from ref. 25 in equations of ref. 40.

current transport and, at the same time, particle-associated PCBs are efficiently scavenged to sub-PML strata. Nevertheless, the PML inventory does not appear inconsistent with a GAR river input of PCB of the estimated magnitude. Another geochemical/system consistency test is to compare the GAR river fluxes with the vertical settling flux out of the Arctic Ocean (AO) PML. Using a typical AO surface water ∑7-ICES PCB concentration in the particulate fraction of 1 pg/L,25 the upper ocean (PML) settling export flux can be estimated from estimates of first-order settling rate constants for shelf regimes (e.g., 0.1 d 1; ref 38), the vertical extent of the mixed surface ocean (e.g., PML of 15 m; ref 25) and the areal extent of the AO (about 1 107 km2; ref 39). This gives an annual settling flux of ∑7-ICES PCB on the order of 0.5 tonne/yr, which is of the same order as the river fluxes. Again, this other PCB Arctic flux vector is on the same scale as the PCB GAR flux and thus not inconsistent with a circum-arctic GAR PCB export of 0.3 tonne/yr. Using Wania and Daly40 model framework and parametrization yields an estimate of the ∑7-ICES PCB settling flux from the AO-PML to deeper strata of 7 70.7 tonnes/yr, which is 10 100 times larger than both the above estimate of the upper Arctic Ocean settling flux (0.5 tonne/yr) and the current study estimate for the amount ∑7-ICES PCB entering the AO-PML from rivers (0.3 tonne/yr). One reason for this discrepancy is likely that the Wania and Daly40 model calculations assumed seawater particulate PCB concentrations of 10 100 ng/g OC. This contrasts with recently reported actual measurements of particulate PCB concentrations in the AO in the range of 0.2 to 3 ng/g OC.25 By instead using our/these measured POC-associated seawater concentration in the same calculations as Wania and Daly,40 we derive a PCB settling flux to subsurface waters in the AO interior basins of 0.74 tonne/yr. This value is more consistent with our other estimates of PCB upper ocean export flux, river export flux and shelf sediment burial fluxes. Taken together, the current estimates of pan-arctic river fluxes of PCBs add to a growing understanding of the major components of the PCB mass balance of the Arctic Ocean. There is still a need for a synthesizing mass balance model of PCBs for the Arctic, based on quality-assured observations of PCB distributions combined with geophysical and biogeochemical system functioning. Such a picture would aid in understanding what are the major PCB compartments and what are the major transport pathways of these substances in the vulnerable Arctic

Ocean system. The current work provides the first comprehensive assessment of PCB fluxes for the great Pan-Arctic Rivers emptying into the Arctic Ocean. Distinctly different congener distributions were found in the North American Arctic (∼ 70% of ∑13PCB made up of trichlorinated and tetrachlorinated congeners) versus in the Eurasian sector where pentachlorinated and hexachlorinated congeners were more abundant than trichlorinated counterparts. The herein estimated PCB river fluxes provide information on the importance of this vector to the overall Arctic Ocean contaminant load, yet the PCB river flux to the Arctic Ocean holds only a tiny portion (∼0.001%) of the PCBs that have been historically emitted to the environment. See Table 1.

’ ASSOCIATED CONTENT

bS

Supporting Information. A table of geochemical characteristics of the Great Arctic Rivers, a table of congener-specific PCB concentrations in river mouth sediments, tables of estimated river fluxes for total PCBs, dissolved PCB, DOC-associated PCB and POC-associated PCB congener fluxes, a table with comparison with other published PCB sediment concentrations and text outlining the equations used in the flux estimations. This material is available free of charge via the Internet at http:// pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Phone: +46-703247317; e-mail: [email protected].

’ ACKNOWLEDGMENT Colleagues at the Far Eastern Branch of the Russian Academy of Sciences (FEBRAS): Igor Semiletov, Oleg Dudarev, and Alexander Charkin are gratefully acknowledged for collecting the Siberian Arctic river sediments. Similarly, Laodong Guo (Univ. of Mississippi) is gratefully acknowledged for collecting the river mouth sediments from the Mackenzie. Hanna Gustafsson is acknowledged for performing the PCB sediment analyses. This work is in part funded by the Swedish Research Council (VR contract nr 621-2007-4631) and the EU 7 FP project ArcRisk € (contract nr 1346810). O.G. also received support as an Academy Researcher from the Royal Swedish Academy of Sciences through a grant from the Knut and Alice Wallenberg Foundation (grant no 629-2002-2309). ’ REFERENCES (1) Meijer, S. N.; Ockenden, W. A.; Steinnes, E.; Corrigan, B. P.; Jones, K. C. Spatial and temporal trends of POPs in Norwegian and UK back-ground air: Implications for global cycling. Environ. Sci. Technol. 2003, 37, 454–461. € Axelman, J.; Sundberg, H. Global (2) J€onsson, A.; Gustafsson, O.; accounting of PCBs in the continental shelf sediments. Environ. Sci. Technol. 2003, 37, 245–255. (3) Gobas, F.; Wilcockson, J. B.; Russell, R. W.; Haffner, G. D. Mechanism of biomagnifications in fish under laboratory and field conditions. Environ. Sci. Technol. 1999, 33, 133–141. € Passive (4) Sobek, A.; Cornelissen, G.; Tiselius, P.; Gustafsson, O. partitioning of polychlorinated biphenyls between seawater and zooplankton, a study comparing observed field distributions to equilibrium sorption experiments. Environ. Sci. Technol. 2006, 40, 6703–6708. (5) Sobek, A.; McLachlan, M. S.; Borga, K.; Asplund, L.; Lundstedt€ A comparison of PCB Enkel, K.; Polder, A.; Gustafsson, O. 8382


Environmental Science & Technology bioaccumulation factors between an Arctic and a temperate marine food web. Sci. Total Environ. 2010, 408, 2753–2760. (6) Bidelman, T. F., Macdonald, R. W., Stow, J. P.Sources, occurrence, trends and pathways of contaminants in the Arctic.Sci. Total Environ. 2005, 342. (7) AMAP. Assessment Report: Arctic Pollution Issues;Arctic Monitoring and Assessment Program (AMAP); Oslo, Norway, 1998. (8) Macdonald, R. W.; Barrie, L. A.; Bidleman, T. F.; Diamond, M. L.; Gregor, R. G.; Semkin, R. G.; et al. Contaminants in the Canadian Arctic: 5 years of progress in understanding sources, occurrence and pathways. Sci. Total Environ. 2000, 254, 93–234. (9) Dittmar, T.; Kattner, G. The biogeochemistry of the river and shelf ecosystem of the Arctic Ocean: Areview. Mar. Chem. 2003, 83, 103–120. (10) Carroll, J.; Savinov, V.; Savinova, T.; Dahle, S.; Mccrea, R.; Muir, D. PCBs, PBDEs and pesticides released to the Arctic Ocean by the Russian rivers Ob and Yenisei. Environ. Sci. Technol. 2008, 42, 69–74. (11) Kimstach, V., Dutchak, S. Persistant toxic substances (PTS) sources and pathways. In Persistent Toxic Substances, Food Security and Indigenous Peoples of the Russian North, nal Report, Chapter 4; Kimstach, V., Ed.; Arctic Monitoring and Assessment Programme (AMAP): Oslo, Norway, 2004; pp 33 80. (12) Macdonald, R. W.; Barrie, L. A.; Bidelman, T. F.; Diamond, M. L.; Gregor, D. J.; Semkin, R. G.; Strachan, W. M. J.; Li, Y. F.; Wania, F.; Alaee, M.; Alexeeva, L. B.; Backus, S. M.; Bailey, R.; Bewers, J. M.; Gobeil, C.; Halsall, C. J.; Harner, T.; Hoff, J. T.; Jantunen, L. M. M.; Lockhart, W. L.; Mackay, D.; Muir, D. C. G.; Pudykiewicz, J.; Reimer, K. J.; Smith, J. N.; Stern, G. A.; Schroeder, W. H.; Wagemann, R.; Yunker, M. B. Contaminants in the Canadian Arctic: 5 years of progress in understanding sources, occurrence and pathways. Sci. Total Environ. 2000, 254, 93–234. (13) Impacts of a Warming Arctic: Arctic Climate Impact Assessment (ACIA); Cambridge University Press: Cambridge, UK, 2005, 1042 pg. (14) Barry, R. G., Serreze, M. C. Atmospheric components of the Arctic Ocean freshwater balance and their interannual variability. In The Freshwater Budget of the Arctic Ocean, NATO meeting/NATO ASI Series; Lewis, E. L., Ed.; Kluwer Acad. Publishers: Dordrecht, 2000; pp 45 56. (15) Semiletov, I. P., Savalieva, N. I., Weller, G. E., Pipko, I. I., Pugach, S. P., Yu Gukov, A., Vasilevskaya, L. N. The dispersion of Siberian River flows into coastal waters: meteorological, hydrological and hydrochemical aspects. In The Freshwater Budget of the Arctic ocean, NATO meeting/NATO ASI Series; Lewis, E. L., Ed.; Kluwer Acad. Publishers: Dordrecht, 2000; pp 323 367. (16) Stern, G. A.; Halsall, C. J.; Barrie, L. A.; Muir, D. C. G.; Fellini, P.; Rosenberg, B.; Rovinsky, Ya.; Ya, F.; Kononov, E.; Pastuhov, B. Polychlorinated biphenyls in Arctic air. 1. Temporal and spatial trends: 1992 1994. Environ. Sci. Technol. 1997, 31, 3619–3628. (17) Kremenetski, K. V.; Velichko, A. A.; Borisova, O. K.; MacDonald, G. M.; Smith, L. C.; Frey, K. E.; Orlova, L. A. Peatlands of the western Siberian lowlands: Current knowledge on zonation, carbon content and Late Quaternary history. Quat. Sci. Rev. 2003, 22, 703–723. (18) van Dongen, B. E., Semiletov, I., Weijers, J. W. H., Gustafsson, € Contrasting lipid biomarker composition of terrestrial organic matter O. exported from across the Eurasian Arctic by five great Russian Arctic rivers. Global Biogeochem. Cycles. 2008, 22, GB1011, DOI: 10.1029/ 2007/GB002974. € van Dongen, B. E., Vonk, J. E., Dudarev, O. V., (19) Gustafsson, O., Semiletov, I. P. Widespread release of old carbon across the Siberian Arctic echoed by its large rivers. Biogeosciences 2011, 8, 1737 1743. DOI: 10.5194/bg-8-1737-2011. (20) Raymond, P. A.; McClelland, J. W.; Holmes, R. M.; Zhulidov, A. V.; Mull, K.; Peterson, B. J.; Striegl, R. G.; Aiken, G. R.; Gurtovaya, T. Y. Flux and age of dissolved organic carbon exported to the Arctic Ocean: A carbon isotopic study of the five largest arctic rivers. Global Biogeochem. Cycles 2007, 21, GB4011. (21) Tarnocai, C., Canadell, J. G., Schuur, E. A. G., Kuhry, P., Mazhitova, G. Zimov S. Soil organic carbon pools in the northern

ARTICLE

circumpolar permafrost region, Glob. Biogeochem. Cycles, 2009. 23, GB2023, DOI: 10.1029/2008GB003327. € Pan-Arctic (22) Elmquist, M., Semiletov, I., Guo, L., Gustafsson, O. patterns in black carbon sources and fluvial discharges deduced from radiocarbon and PAH source apportionment markers in estuarine surface sediments. Global Biogeochem. Cycles. 2008, 22, GB2018, DOI: 10.1029/2007/GB002994. (23) Guo, L. D., Ping, C. L., Macdonald, R. W. Mobilization pathways of organic carbon from permafrost to arctic rivers in a changing climate, Geophys. Res. Let., 2007, 34, L13603, DOIi: 10.1029//GL030689. € Latitudinal fractionation of poly(24) Sobek, A.; Gustafsson, O. chlorinated biphenyls in surface seawater along a 62°N-89°N transect from the North Sea to the North Pole area. Environ. Sci. Technol. 2004, 38, 2746–2751. € Distribution and inventories (25) Carrizo, D.; Gustafsson, O. of polychlorinated biphenyls in the Polar Mixed Layer of the panArctic shelf seas and interior basins. Environ. Sci. Technol. 2011, 45, 1420–1427. € Hajdu, A.; Larsson, U. Particle-water (26) Sobek, A.; Gustafsson, O.; partitioning of PCBs in the photic zone: a 25-month study in the open Baltic Sea. Environ. Sci. Technol. 2004, 38, 1375–1382. € Selective (27) Vonk, J. E., van Dongen, B. E., Gustafsson, O. preservation of old organic carbon fluvially released from sub-Arctic soils. Geophys. Res. Let., 2010, 37, L11605, DOI: 10.1029// 2010GL042909. (28) Hedges, J. L.; Keil, R. G.; Benner, R. What happens to terrestrial organic matter in the ocean? Org. Geochem. 1997, 27, 195–212. (29) Bianchi, T. S.; Allison, M. A. Large-river delta-front estuaries as natural “recorders” of global environmental change. PNAS. 2009, 106, 8085–8092. (30) Tolosa, I.; Bayona, J. M.; Albaiges, J. Spatial and temporal distribution, fluxes and budgets of organochlorinated compounds in Northwest Mediterranean sediments. Environ. Sci. Technol. 1995, 20, 2519–2527. (31) Cid Montanes, J. F.; Risebrough, R. W.; De Lappe, B. W.; Marino, M. G.; Albaiges, J. Estimated inputs of organochlorines from the river Ebro into the Northwestern Mediterranean. Mar. Pollut. Bull. 1990, 21, 518–523. (32) Fernandez, M. A.; Alonso, C.; Gonzalez, M. J.; Hernandez, L. M. Occurrence of organochlorine insecticides, PCBs and PCB congeners in waters and sediments of the Ebro River (Spain). Chemosphere 1999, 38, 33–43. (33) Breivik, K.; Sweetman, A.; Pacyna, J. M.; Jones, K. C. Towards a global historical emission inventory for selected PCB congeners-a mass balance approach: 1. Global production and consumption. Sci. Total Environ. 2002, 290, 181–198. (34) Breivik, K.; Sweetman, A.; Pacyna, J. M.; Jones, K. C. Towards a global historical emission inventory for selected PCB congenersa mass balance approach: 2. Emissions. Sci. Total Environ. 2002, 290, 181–224. (35) Ivanov, V.; Sandell, E. Characterization of polychlorinated biphenyl isomers in Sovol and trichlorodiphenyl formulations by highresolution gas chromatography with electron capture detection and high-resolution gas chromatography-mass spectrometry techniques. Environ. Sci. Technol. 1992, 26, 2012–2017. (36) McConnell, L.; Kucklick, J. R.; Bidleman, T. F.; Ivanov, G. P.; Chernyak, S. M. Air-water gas exchange of organochlorine compounds in Lake Baikal, Russia. Environ. Sci. Technol. 1996, 30, 2975–2983. (37) Iwata, H.; Tanabe, S.; Ueda, K.; Tatsukawa, R. Persistent organochlorines residues in air, water, sediments and soils from the Lake Baikal region, Russia. Environ. Sci. Technol. 1995, 29, 792–801. € Gschwend, P. M.; Buesseler, K. O. Settling (38) Gustafsson, O.; removal rates of PCBs into the Northwestern Atlantic derived from the 238U-234Th disequilibria. Environ. Sci. Technol. 1997, 31, 3544–3550. (39) Jakobsson, M., et al. Bathymetry and physiography of the Arctic Ocean and its constituent seas, In Arctic Ocean Organic Carbon Cycle: 8383



ARTICLE

Present and Past; Stein, R., Macdonald, R. W., Eds., Springer-Verlag: Berlin Heidelberg, 2004; pp 1 6. (40) Wania, F.; Daly, G. L. Estimating the contribution of degradation in air and deposition to deep sea to global loss of PCBs. Atmos. Environ. 2002, 36, 5581–5593.

8384


Pan-Arctic River Fluxes of Polychlorinated Biphenyls - Environmental

Recommend Documents