Science of the Total Environment 506–507 (2015) 430–443

Contents lists available at ScienceDirect

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

Spatial, temporal, and source variations of hydrocarbons in marine sediments from Baffin Bay, Eastern Canadian Arctic Karen L. Foster a,b,⁎, Gary A. Stern a, Jesse Carrie a, Joscelyn N.-L. Bailey a, Peter M. Outridge a,c, Hamed Sanei a,d, Robie W. Macdonald a,e a

Centre for Earth Observation Sciences (CEOS), Department of Environment and Geography, University of Manitoba, Winnipeg, Canada Environmental & Resource Studies Program, Trent University, Peterborough, Canada Geological Survey of Canada, Natural Resources Canada, Ottawa, Canada d Geological Survey of Canada, Natural Resources Canada, Calgary, Canada e Institute of Ocean Sciences, Department of Fisheries and Oceans, Sidney, Canada b c

H I G H L I G H T S • • • • •

Hydrocarbon concentrations and profiles in sediments from northern Baffin Bay Data from surface and historical sediments from each sediment core are given Current hydrocarbon levels are within a factor of 10 of pre-industrial sediments Petrogenic PAH sources to the region are predominate The results are a preliminary characterization of the hydrocarbons in this region

a r t i c l e

i n f o

Article history: Received 30 July 2014 Received in revised form 1 November 2014 Accepted 2 November 2014 Available online 26 November 2014 Editor: F. Riget Keywords: Sediment cores Arctic Polycyclic aromatic hydrocarbons (PAHs) Alkanes Biomarkers Hydrocarbon profiles

a b s t r a c t With declining sea ice conditions in Arctic regions owing to changing climate, the large prospective reservoirs of oil and gas in Baffin Bay and Davis Strait are increasingly accessible, and the interest in offshore exploration and shipping through these regions has increased. Both of these activities are associated with the risk of hydrocarbon releases into the marine ecosystem. However, hydrocarbons are also present naturally in marine environments, in some cases deriving from oil seeps. We have analyzed hydrocarbon concentrations in eleven sediment cores collected from northern Baffin Bay during 2008 and 2009 Amundsen expeditions and have examined the hydrocarbon compositions in both pre- and post-industrial periods (i.e., before and after 1900) to assess the sources of hydrocarbons, and their temporal and spatial variabilities. Concentrations of ΣPAHs ranged from 341 to 2693 ng g−1 dw, with concentrations in cores from sites within the North Water (NOW) Polynya generally higher. Individual PAH concentrations did not exceed concentrations of concern for marine aquatic life, with one exception found in a core collected within the NOW (one of the seven sediment core samples). Hydrocarbon biomarkers, including alkane profiles, OEP (odd-to-even preference), and TAR (terrigenous/aquatic ratios) values indicated that organic carbon at all sites is derived from both terrigenous higher plants and marine algae, the former being of greater significance at coastal sites, and the latter at the deepest sites at the southern boundary of the NOW. Biomarker ratios and chemical profiles indicate that petrogenic sources dominate over combustion sources, and thus long-range atmospheric transport is less significant than inputs from weathering. Presentday and historic pre-1900 hydrocarbon concentrations exhibited less than an order of magnitude difference for most compounds at all sites. The dataset presented here provides a baseline record of hydrocarbon concentrations in Baffin Bay sediments in advance of offshore exploration and increased shipping activities. © 2014 Elsevier B.V. All rights reserved.

⁎ Corresponding author at: Environmental & Resource Studies Program, Trent University, 1600 West Bank Drive, Peterborough, ON K9J 7B8, Canada. Tel.: +1 705 768 2081. E-mail addresses: [email protected] (K.L. Foster), [email protected] (G.A. Stern), [email protected] (J. Carrie), [email protected] (J.N.-L. Bailey), [email protected] (P.M. Outridge), [email protected] (H. Sanei), [email protected] (R.W. Macdonald).

http://dx.doi.org/10.1016/j.scitotenv.2014.11.002 0048-9697/© 2014 Elsevier B.V. All rights reserved.

1. Introduction The Canadian Arctic, including marine areas of Nunavut, in Canada's eastern Arctic, potentially contains large under‐sea oil and natural gas reservoirs. Within Canada, the largest potential oil fields lie under Baffin Bay, with several possible reservoirs containing as much as 10 billion barrels of oil (Gautier et al., 2009). The on‐going reduction of sea‐ice

K.L. Foster et al. / Science of the Total Environment 506–507 (2015) 430–443

coverage in summer due to climate warming (Perovich et al., 2013) makes the exploration of these possible reservoirs more feasible and encourages an increase in ship traffic through Canadian Arctic territorial waters (AMAP, 2010). Both oil and gas exploration/exploitation, and ship traffic, carry with them the potential for accidental releases of crude oil and hydrocarbons into the environment. Crude oils are complex mixtures of thousands of organic compounds, including alkanes (paraffins), cycloalkanes, polycyclic aromatic hydrocarbons (PAHs), and inorganic elements present as part of these compounds, including sulfur, vanadium, uranium, copper, and nickel (AMAP, 2010). PAHs in particular are an emerging chemical class of concern in some Arctic regions. Concentrations of PAHs in fish and invertebrates from the Barents and Norwegian Seas were found to have increased 10 to 30-fold over the past 25 years (De Laender et al., 2011). The chemical structures and profiles (the relative concentrations of compounds) of hydrocarbons, including PAHs, provide useful biomarkers to infer hydrocarbon sources (Peters et al., 2007; Yunker et al., 2002a, 2002b, 2011). Petrogenic (igneous rock-derived) PAHs, including those in petroleum, tend to be more thermodynamically stable than pyrogenic PAHs because they are produced during the long term thermal degradation of organic material, which allows time for slow kinetic processes to proceed (Lima et al., 2005; Yunker and Macdonald, 1995). Petrogenic PAHs tend to contain more alkylated (branched) compounds, whereas pyrogenic PAHs tend to be un-substituted (parent) compounds (Lima et al., 2005; Peters et al., 2007; Yunker and Macdonald, 1995). A number of PAH isomers and/or ratios can also be uniquely characteristic of particular sources. For example, the ratio of the less stable PAH isomer fluoranthene to pyrene (molar mass of 202 g mol−1) is indicative of petrogenic sources if the ratio is low (b0.4), and pyrogenic sources if it is higher (N 0.5) (Yunker et al., 2011). PAH ratios and chemical profiles have been used to assess the sources of hydrocarbons within marine sediments in Arctic (e.g., Boitsov et al., 2013; Yunker and Macdonald, 1995; Yunker et al., 1995, 2011) and other regions (e.g., Yunker et al., 2002a; Liu et al., 2013). Natural oil seeps from under‐sea reservoirs release oil into marine waters; rivers may also contribute hydrocarbons to the near‐shore marine environment with the relative quantities of petrogenic and combustion PAHs depending on fossil carbon composition of soil in the river basin and the degree of atmospheric deposition (Yunker and Macdonald, 1995; Yunker et al., 2002b; Yunker et al., 2011). The only documented natural oil seeps in Baffin Bay that we are aware of are at Scott Inlet, and in west Lancaster Sound (Blasco et al., 2010). The challenge in the event of an accidental oil spill occurring in the future will be to discriminate between natural, background, and contaminant hydrocarbons due to the spill. In this regard, comprehensive spatial baseline data together with time series will be useful in distinguishing between sources and assessing remediation. In addition, time series provide a means to determine whether or not the recent changes in Arctic ice climate (sea ice and permafrost) have impacted hydrocarbon fluxes and sources to the marine environment (Macdonald et al., 2005). Here, we establish the composition and concentrations of hydrocarbons in surface and historical sediments from eleven sites across northern Baffin Bay, prior to oil and gas exploration/exploitation in the area, and provide an assessment of 1) the geographical variability of current hydrocarbon concentrations in this region; 2) predominant sources of hydrocarbons based on their composition; and 3) whether recently accumulated sediment (post-industrial) differs from older (pre-industrial) sediment. This region of the Canadian Arctic has not been characterized for hydrocarbon concentrations and compositions (AMAP, 2010). Thus, the baseline data presented here, prior to oil and gas exploration in the region or major expansion of shipping traffic through the Northwest Passage, is of imminent importance. The sediment data produced by this project complements those available for other regions of Canada such as the southern Beaufort Sea, off southern Baffin Island, the Cambridge Bay area, the Bent Horn field in the Arctic Archipelago

431

and several datasets available for the Greenland side of Davis Strait (AMAP, 2010; Obermajer et al., 2010). 2. Materials and methods 2.1. Study sites and sample collection Duplicate sediment cores were collected by box coring at eleven locations in northern Baffin Bay (including the North Water (NOW) Polynya, Lancaster Sound, Scott Inlet, and Gibbs Fjord) during the ArcticNet 2008 and 2009 cruises aboard the CCGS Amundsen research vessel (Fig. 1A and Table 1). Once the box cores were hauled onboard the ship, plexiglass tubes (10 cm diameter) were pushed into the sediments; cores were between 17 and 36 cm long. The cores were extruded and sectioned on the ship into 0.5 cm intervals for the first 10 cm, then 1 cm intervals thereafter. All sections were stored in Whirlpak® bags and kept frozen at −40 °C until analysis. The NOW in northern Baffin Bay spans up to 80,000 km2 (Wassmann et al., 2004) and receives only light seasonal ice cover, distinguishing it from the remainder of Baffin Bay, which is ice covered for most of the year. The NOW is a nexus for oceanic currents flowing northwards up the east side of Davis Strait, and those flowing eastwards (Lancaster and Jones Sounds) and southwards (Nares Strait) from the Arctic Ocean through the Canadian Arctic Archipelago. These various currents transport and mix water derived from the Atlantic Ocean, the Pacific Ocean, from multiple rivers and from sea-ice melt (Alkire et al., 2010). It is therefore a critical downstream area for monitoring the distribution of future oil pollution in the Canadian Arctic. Six of the sampling sites are located within the NOW region (101, 108, 111, 115, 136, 138). 2.2. Sediment dating The sediment cores were dated using 210Pb profiles as described in detail in Bailey et al. (2013). Given that all cores exhibited a surface mixed layer, the 210Pb profiles were interpreted using an advectivediffusive model, which was then applied to determining the depths in the core below which sediments accumulated prior to 1900, and the depths above which sediments accumulated after 1900 (Bailey et al., 2013). 2.3 . Hydrocarbon analyses Three to four sections per core were analyzed for hydrocarbons; at least one from the surface representing post-1900 sediments and two from pre-1900 sediments deeper in the core. Hydrocarbons were analyzed by AXYS Analytical Services Ltd. of Sidney, British Columbia. Hydrocarbon analyses were based on a modified EPA 8270 method. Briefly, samples were spiked with deuterated surrogate and recovery standards, followed by Soxhlet extraction using dichloromethane, chromatographic clean-up and fractionation (PAH and alkane fractions), and analysis by gas chromatography (GC, RTX-5)/low-resolution mass spectrometry (LRMS). Quantification was done using isotope dilution and a relative recovery factor (RRF) correction determined from a calibration solution analyzed at the beginning and end of each sample batch; twenty samples were run per batch. Sample detection limits (SDLs) were determined for each analyte in each sample as the concentration equivalent to three times the chromatographic noise height. Recoveries of labeled surrogate standards from spiked reference matrix were 105.8% ± 7.0 for aromatic (n = 2) and 94.0% ± 33.0 for alkane (n = 2) analytes. Procedural blanks were run with each batch of samples and measured concentrations were below the SDL for most analytes; 60% of aromatic analytes (n = 2) and 60% of alkane analytes (n = 2). Concentrations were not blank corrected. None of the samples exhibited contamination that can be associated with the intermediate storage of sediments in polyethylene bags (Grosjean and Logan, 2007).

432

K.L. Foster et al. / Science of the Total Environment 506–507 (2015) 430–443

-70° W

-60° W

B

A

76° N 80° N

74° N

75° N

72° N

Depth (m)

70° N

>

Seeps

22 5 20 0 0 15 0 0 10 0 00 75 0 50 0 25 0 10 0 50 0

Sites Wells

70° N -90° W

-80° W

-70° W

Fig. 1. A) Locations of sediment cores collected from Baffin Bay. Basin watershed map source: http://geogratis.cgdi.gc.ca/, bathymetrical data (SRTM30 PLUS) source: http://topex.ucsd. edu/. Greenland exploration well locations: http://www.bmp.gl, B) concentrations of total polycyclic aromatic hydrocarbons (ΣPAHs) in post-1900 sediment.

A total of 108 compounds and homologue groups were measured; 32 alkanes, 66 PAHs, 3 biphenyls, and 7 dibenzothiophenes. Note that where co-eluting hydrocarbons occurred, they were quantified together and are indicated with a “/” separating the two names. Analytes (with abbreviations used here in parentheses) included: n-alkanes (n-C12 through n-C36), and ΣPAHs comprised of naphthalene (N0), alkyl naphthalenes (N1–N4), acenaphthene (Ace), phenanthrene (P0), anthracene (A0), alkyl phenanthrenes/anthracenes (P/A1–P/A4), fluorene (F0), alkyl fluorenes (F1–F3), fluoranthene (Fl), Pyrene (Py), alkyl fluoranthenes/pyrenes (Fl/Py1–Fl/Py4), benz[a]anthracene (BA), chrysene (Ch), alkyl benz[a]anthracenes/chrysenes (BA/Ch1–BA/Ch4), benzo [b]fluoranthene (BbFl), benzo[j,k]fluoranthenes (BjkFl), benzo[e] pyrene (BPy), benzo[a]pyrene, perylene (Per), dibenz[a,h]anthracene, indeno[1,2,3-cd]pyrene (IPy), benzo[ghi]perylene (BPer), retene (Ret), and alkyl benzofluoranthenes/benzopyrenes (BFl/BPy1–BFl/BPy4). US Table 1 GPS coordinates (decimal degrees), water depths, and average sedimentation rates of sediment from eleven sampling locations across Baffin Bay. Site

Latitude

Longitude

Depth (m)

Sed. ratea (g cm−2 y−1)

101 108 111 115 136 138 141 205 233 301 GBF

76.40 76.27 76.31 76.33 74.79 74.94 71.38 77.22 76.73 74.15 70.77

−77.50 −74.60 −73.22 −71.23 −73.68 −69.07 −70.15 −78.89 −71.82 −83.20 −72.27

367 438 597 668 789 1016 258 359 593 699 2

0.046 0.021 0.031 0.027 0.032 0.060 NAb 0.084 0.023 0.038 0.065

a b

Sedimentation rates from Bailey et al. (2013). NA = not analyzed.

EPA probable carcinogens (Σ7PAHs) are the sum of Ch, BA, BbFl, BjkFl, IPy, benzo[a]pyrene, and dibenz[a,h]anthracene. Where peaks were present but did not meet the stringent quantification criteria, maximum possible concentrations were used. Nondetects, including concentrations less than the SDL, were replaced with a randomly generated number between zero and the detection limit. Compounds not detected in over 20% of the samples were not considered in statistical analyses. This resulted in the exclusion of thirteen compounds including acenaphthylene, benzo[a]pyrene, and dibenz[a, h]anthracene. Anthracene with 20% non-detects was the single exception; anthracene values were added to those of phenanthrene for the purposes of figures and statistical analyses. 2.3.1. Principal component analysis (PCA) Concentration data were first scaled by taking the logarithms; thereby accounting for the non-linear significance of magnitude within the dataset (Skillicorn, 2007). PCA was then done on the covariance matrix of the standardized variables (the correlation matrix) using Systat© 13. Variables were standardized to have a mean of zero and variance of one (auto scaled) to ensure that the magnitude of concentration did not give any one compound undue weighting. This yielded principal components with scores that sum to zero (mean centered). PCA analyses were done on PAHs, which included parent PAHs and alkylated PAH groups, and on n-alkanes. 2.3.2. OEP and TAR calculations Odd-to-even preference (OEP) ratios centered around n-C29 were computed as in Scalan and Smith (1970) as:

OEP C29 ¼

C27 þ 6C29 þ C31 4C28 þ 4C30

ð1Þ

K.L. Foster et al. / Science of the Total Environment 506–507 (2015) 430–443

where Ci is the relative weight percent of n-alkanes with i carbons. Terrigenous/aquatic ratios (TARs) were calculated as in Bourbonniere and Meyers (1996) as: TAR ¼

C27 þ C29 þ C31 : C15 þ C17 þ C19

ð2Þ

3. Results and discussion 3.1. Spatial variation of hydrocarbons across Baffin Bay The highest concentrations of ΣPAHs in post-1900 sediments among the Baffin Bay cores were found at sites 111 and 108 within the NOW with concentrations of 2693.0 and 2111.1 ng g−1 dw, respectively (Fig. 1B, Table 2). These sites also had the highest concentration of Σ7PAH, 44.3 and 30.0 ng g−1 dw, respectively. The lowest concentrations of ΣPAH and Σ7PAH were found at Gibbs Fjord (GBF) (341.0 and 0.9 ng g−1 dw, respectively) and at site 141 (382.9 and 3.4 ng g−1 dw, respectively), which is adjacent to GBF and about 2 km from Scott Inlet. The percentage of Σ7PAH relative to ΣPAH in all of the sites ranged from 0.3 to 2.5%, with sites 136 and 115 having the highest percentage of EPA's probable carcinogen PAHs and site GBF having the lowest. Measured concentrations exceeded the interim sediment quality guideline (ISQG) values in only one sample. The concentration of pyrene, a US EPA priority pollutant, exceeded the ISQG in surficial (3–3.5 cm) sediment at site 111 with a concentration of 188.0 ng g−1 dw compared to the ISQG value of 153 ng g−1 dw. PAH concentrations in general were higher in this sample, particularly of the C1–C4 branched phenanthrenes/ anthracenes, which are characteristic of petrogenic sources. Additional sediment sections above and below this slice (intervals 2.0–2.5, 2.5– 3.0, 3.5–4.0, 4.0–4.5 cm) from the site 111 core were analyzed for PAHs for comparison. The PAH profiles in the additional sediment sections were similar and ΣPAH concentrations were among the highest in this study; the concentrations in the additional sections were not as high as in slice 3.0–3.5 cm. Average post-1900 concentrations for site 111 presented here include all of samples for which PAHs were measured. Site 111 is a unique case discussed in greater detail below. Fluxes of ΣPAHs to the sediments (Table 2) were estimated as the product of the concentration of ΣPAHs in ng g−1 dw and the average sedimentation rate (Bailey et al., 2013, Table 1), which assumes that fluxes have been relatively constant over the time period. The range in ΣPAH fluxes was similar between historic and surficial sediments with pre-1900 fluxes ranging from 19.2 to 107.4 ng cm− 2 y− 1 and post1900 fluxes ranging from 20.5 to 108.9 ng cm− 2 y−1. Sites GBF and 233 had the lowest average ΣPAH fluxes; 22.2 and 15.7 ng cm−2 y−1 for post- and pre-1900 GBF sediments, respectively, and 20.5 and 19.2 ng cm−2 y−1, respectively for site 233. The highest ΣPAH fluxes to both post- and pre-1900 sediments were at sites 101, 111, 138, and 205. Surface sediment concentrations of ΣPAH (parent and alkyl) reported across the Arctic Ocean, which range from ~ 113 ng g−1 dw in the Makarov Basin to ~ 2959 ng g−1 dw in the Mackenzie delta (Yunker et al., 2011), were comparable to those reported here for Baffin Bay. Average Baffin Bay surficial sediment concentrations ranged from 341.0 to 2693.0 ng g− 1 dw, although, site 111 had ΣPAH concentrations that ranged from 917.0 to 6939.7 ng g−1 dw; the maximum concentration over two-fold higher than concentrations found in sediments from the Mackenzie delta. 3.2. Temporal variation of hydrocarbons Hydrocarbon concentrations in present-day surficial sediments and in historic, pre-1900 sediments from individual cores were generally similar; within a factor of 10 for most chemicals (Fig. 2). Fig. 2 shows a comparison of pre- and post-1900 sediment concentrations for a

433

representative set of 70 individual hydrocarbons and hydrocarbon groups within each core and for all samples; a total of 2552 comparisons across all eleven cores. The slopes of these comparisons, or the ratio of pre-1900 to post-1900 concentrations, fell between 0.1 and 10 for 99.6% of the slopes indicating that post-1900 concentrations were within a factor of 10 of pre-1900 concentrations. Nine exceptions were noted where pre- and post-1900 concentrations were different by a factor greater than 10. Five of these occurred within the GBF core, where concentrations of benzo[e]pyrene, benzo[j,k] fluoranthenes, C1 -dibenzothiophenes, C 4 -phenanthrenes/anthracenes and C1-benzofluoranthenes/benzopyrenes in the surficial sediment sample were up to 545-fold higher than in the pre-1900's samples. Pristane, primarily produced during the degradation of chlorophyll (Peters et al., 2007), in the post-1900 sediment sample at site 205 had concentrations 17-fold higher than in one of the pre1900 samples. One pre-1900 sample from core 141 also had concentrations of perylene, C1 and C2-benzofluoranthenes/benzopyrenes that were up to 16% higher than in the surficial post-1900 sample. Perylene is associated with the microbial degradation of a variety of terrigenous and marine plants, therefore, concentrations tend to increase with sediment depth (Peters et al., 2007). However, these nine exceptions were noted for only single comparisons of post- and pre-1900 sediment samples; all pre- versus post-1900 comparisons within these cores did not yield these same exceptions. Sites 101, 108, 115, 136, 138, 233, and 301 had the most consistent hydrocarbon concentrations over time, with 89–99% of post-1900 concentrations within a factor of two of pre-1900 concentrations, with post-/pre-1900 slopes falling between 0.5 and 2. Hydrocarbon concentrations at sites 111, GBF, 141 and 205 were less consistent with 53– 79% of pre- to post-1900 concentrations falling within a factor of 2. Thus, general agreement was found for most chemicals between preand post-1900 sediment concentrations, although, some sites were more variable than others. Additional study is necessary to determine if these exceptions are indeed temporal anomalies. 3.3. Sources of hydrocarbons 3.3.1. Terrigenous vs. marine biomass The sum of all measured hydrocarbon compounds in sediments represent a small fraction of the total organic carbon present in these sediments, between 0.02 and 0.19% depending on the site, yet their relative concentrations or chemical signatures are like fingerprints that indicate the source of organic carbon to the sites. Of the samples analyzed here, sites GBF, 233, and 108 appear to have the highest terrigenous plant contributions to organic matter with higher OEP (4.2–4.4) and TAR (1.9–7.3) values; for sites GBF and 233 this is probably a reflection of proximity to the coastline and associated terrigenous run-off and coastal erosion (Fig. 1). Higher OEP and TAR values at site 108 may be associated with nearby immature source rock as discussed below. Conversely, at polynya sites 136 and 138 aquatic plant biomass (algae) is the dominant source of organic matter with TAR values b 1 (0.6 and 0.9, respectively) and among the lowest OEP values reported here. This suggests stronger marine influence at sites 136 and 138 in keeping with the bulk organic carbon properties at these sites (Bailey et al., 2013), probably a consequence of the location, which is far from shore and well within the highly productive NOW polynya (Table 1, Fig. 1). It should be noted that low OEP values, such as those found at sites 136, 138, and 141, may also indicate greater thermal maturity of alkanes. Indeed, site 141 is located ~2 km from a known oil seep at Scott Inlet (Fig. 1). Site 111 also had lower TAR (0.4) and OEP values (2.5). The importance of both marine and terrigenous contributions to organic matter at all sites is also evident in the bimodal pattern of nalkane concentrations in the chemical profiles (Fig. 3). Odd C numbered higher n-alkanes (n-C25, n-C27, n-C29, and n-C31) associated with terrigenous, vascular plant sources of organic matter (Peters et al., 2007) comprised one mode, and lower n-alkanes (n-C15 to n-C21), characteristic of

434 Table 2 Sediment concentrations (ng g−1 dw) of PAHs and alkanes from 11 sites in Baffin Bay for A) surficial post-1900 and B) pre-1900 core sediments. Interim sediment quality guidelines (ISQG) for the protection of marine aquatic life (CCME, 2012), where available, are also shown. Non detectable concentrations are indicated with ND. OEP and TAR bioindicator ratios and ΣPAH fluxes are also given. Values shown are means ± SE, where possible. A)

Site

ISQG 108

111

115

136

138

141

205

233

301

GBF

2

1

5c

1

2

1

1

1

1

1

1

Naphthalene C1-naphthalenes C2-naphthalenes C3-naphthalenes C4-naphthalenes Acenaphthenea Fluorenea Phenanthrenea/anthracenea C1-phenanthrenes/ anthracenes C2-phenanthrenes/ anthracenes C3-phenanthrenes/ anthracenes C4-phenanthrenes/ anthracenes Fluoranthenea/pyrenea

6.8, 7.9 16.6, 19.1 57.0, 61.7 74.5, 85.3 70.9, 84.1 2.5, 2.5 6.2, 7.6 54.2, 64.2 55.9, 88.1

7.4 17.5 67.8 77.7 74.9 1.6 4.3 39.5 126.0

8.1 ± 0.6 12.5 ± 1.7 29.1 ± 4.4 46.3 ± 8.5 37.7 ± 10.2 1.5 ± 0.1 5.3 ± 0.4 50.9 ± 9.2 142.3 ± 50.9

12.4 22.4 45.0 47.7 29.3 2.6 5.2 23.2 45.3

10.0, 12.1 22.5, 27.5 39.7, 45.6 42.4, 41.3 35.7, 28.2 1.0, 1.6 6.4, 4.8 48.8, 30.3 50.2, 35.7

8.7 20.0 32.4 27.3 23.4 1.2 3.2 49.9 122.0

2.5 3.5 10.0 10.3 8.8 0.5 0.8 17.9 36.3

3.8 10.6 129.0 54.5 63.0 1.5 3.7 75.4 81.3

3.9 7.5 70.3 30.2 40.3 0.9 2.7 29.8 100.0

4.4 10.8 21.5 23.8 22.4 0.7 2.1 41.4 48.2

0.7 1.6 32.5 8.1 11.8 0.3 0.7 8.2 28.2

100.0, 118.0

217.0

271.4 ± 126.2

57.2

79.2, 65.5

145.0

51.9

133.0

136.0

105.0

45.4

72.3, 120.0

234.0

376.8 ± 199.6

52.8

54.2, 52.9

90.1

43.6

103.0

97.1

99.1

31.7

168.0, 214.0

383.0

561.4 ± 262.8

115.0

138.0, 99.1

371.0

65.5

163.0

20.3

134.0

52.3

32.5, 39.8

63.3

87.7 ± 39.6

21.3

22.5, 18.5

34.6

13.6

36.6

34.2

31.4

10.0

C1-fluoranthenes/pyrenes C2-fluoranthenes/pyrenes C3-fluoranthenes/pyrenes C4-fluoranthenes/pyrenes Chrysenea,b Benz[a]anthracenea,b Benzo[a]pyrenea,b,h Perylene Benzo[b]fluoranthenea,b Benzo[j,k]fluoranthenesa,b Benzo[e]pyrene Benzo[ghi]perylenea Indeno[1,2,3-cd]pyrenea,b Dibenz[a,h]anthracenea,b,h Retene ΣPAHs

49.5, 67.0 53.5, 62.5 17.3, 17.6 10.7, 10.8 6.0, 6.2 1.9, 2.3 2.7, 2.4 169.0, 219.0 5.8, 5.3 3.0, 3.1 3.7, 3.3 3.9, 4.9 4.7, 4.1 ND, 0.7 42.3, 47.5 1448.6, 1745.5 24.0, 24.1 20.7, 21.2 1193.2, 1975.9 1206.7, 1355.2 3.6, 3.4 1.9, 1.2 66.6, 80.3

108.0 121.0 36.1 12.4 13.8 2.0 2.4 116.0 5.3 2.5 4.3 4.9 4.0 ND 69.8 2111.1

160.8 ± 79.9 227.0 ± 122.3 105.9 ± 60.3 43.5 ± 23.5 30.9 ± 14.9 3.7 ± 1.4 1.3 ± 0.1 70.5 ± 12.7 3.8 ± 0.6 1.9 ± 0.4 3.9 ± 0.8 3.4 ± 0.5 2.3 ± 0.4 0.5 ± 0.1 73.4 ± 37.5 2693.0 ± 1143.9

36.7 43.1 15.7 9.1 7.6 1.9 1.9 87.5 5.7 2.0 4.5 4.8 3.6 0.6 31.5 932.8

34.6, 37.0 36.2, 42.3 16.4, 15.2 6.1, 6.0 5.3, 5.7 1.0, 1.7 0.7, 1.0 12.8, 29.3 6.1, 7.1 1.5, 1.7 3.3, 4.0 3.2, 4.2 2.2, 3.0 0.6, 0.9 37.3, 23.0 1034.7, 829.5

34.3 55.5 44.9 30.0 10.2 1.3 ND 10.9 4.3 1.0 4.1 1.8 1.2 ND 82.0 1453.5

16.4 12.9 7.0 3.0 1.7 0.3 ND 3.3 0.7 0.2 0.3 0.5 0.4 ND 12.9 382.9

53.8 42.4 14.3 4.9 3.5 1.4 0.9 52.9 2.7 0.9 1.5 2.3 1.5 0.3 40.3 1296.5

44.2 33.8 8.6 2.9 3.7 0.6 0.6 21.6 1.7 0.7 1.1 2.2 1.3 0.2 19.0 889.9

48.5 40.7 14.2 6.5 4.9 0.9 0.9 23.1 3.3 1.1 1.9 3.2 1.8 0.4 24.0 874.3

12.7 10.9 4.8 1.3 0.8 0.1 ND 1.0 ND ND ND ND ND ND 9.5 341.0

30.0 12.7 931.7

44.3 ± 17.9 10.2 ± 5.0 5084.0

23.3 19.6 1676.9

17.4, 21.1 18.0 12.0, 15.0 9.1 2575.4, 1853.3 2209.0

3.4 11.2 332.3

11.2 14.5 1651.0

8.7 11.7 601.5

13.3 13.7 1011.3

0.9 6.5 282.2

927.8

918.4

952.6

1186.4, 641.7

1066.0

303.3

1002.5

545.4

810.7

972.5

4.2 2.1 44.3

2.5 0.4 83.5 ± 35.5

2.4 1.1 25.2

2.0, 2.9 0.6, 0.5 33.1, 26.5

1.8 0.9 87.2

1.6 1.4 NAg

2.8 1.0 108.9

4.4 1.9 20.5

2.8 1.2 33.2

4.2 7.3 22.2

n a

Σ7PAHsb % parent PAHs Σ(n-C15–C21) alkanes Σ(n-C25–C31) alkanes OEP C29d TARe ΣPAH flux (ng cm−2 y−1)f

34.6

6.7 21.2 86.7/46.9

113.0/ 153

108.0 74.8 88.8

6.2

K.L. Foster et al. / Science of the Total Environment 506–507 (2015) 430–443

101

B)

Site

ISQG 108

111

115

136

138

141

205

233

301

GBF

2

2

2

3

2

2

3

3

3

2

3

Naphthalenea C1-naphthalenes C2-naphthalenes C3-naphthalenes C4-naphthalenes Acenaphthenea Fluorenea Phenanthrenea/anthracenea C1-phenanthrenes/ anthracenes C2-phenanthrenes/ anthracenes C3-phenanthrenes/ anthracenes C4-phenanthrenes/ anthracenes Fluoranthenea/pyrenea

10.5, 10.6 22.9, 25.4 64.1, 69.0 62.0, 73.6 53.7, 42.7 1.6, 2.1 5.8, 6.5 38.6, 44.1 103.0, 107.0

10.9, 12.9 25.9, 31.8 63.1, 81.9 64.0, 94.6 57.9, 71.8 1.7, 2.5 5.9, 7.1 37.8, 49.8 110.0, 115.0

11.9, 12.3 23.7, 25.1 52.8, 52.0 73.9, 62.4 65.2, 49.5 2.7, 2.3 6.2, 7.6 55.6, 64.1 124.0, 223.0

14.3 ± 0.1 25.8 ± 0.8 47.5 ± 3.9 52.8 ± 7.7 36.1 ± 5.0 3.3 ± 1.0 7.5 ± 3.3 33.6 ± 8.3 62.1 ± 8.5

11.3, 13.8 29.7, 32.9 50.9, 55.3 52.0, 55.7 39.4, 42.7 1.1, 2.4 7.2, 6.4 44.2, 44.5 57.8, 75.9

12.8, 5.1 18.5, 9.0 28.3, 15.9 24.5, 15.6 23.9, 14.9 0.6, 0.7 2.1, 1.8 50.6, 55.8 143.0, 86.4

3.1 ± 1.0 3.7 ± 0.3 11.3 ± 3.1 7.5 ± 1.5 7.2 ± 1.5 0.4 ± 0.3 0.6 ± 0.2 11.5 ± 3.8 21.4 ± 5.0

2.5 ± 0.2 7.2 ± 0.3 32.7 ± 4.4 25.7 ± 1.8 19.5 ± 3.0 0.7 ± 0.1 2.2 ± 0.5 17.0 ± 1.0 33.2 ± 6.6

5.9 ± 0.7 11.3 ± 0.5 32.2 ± 2.8 27.8 ± 3.1 26.0 ± 7.0 0.9 ± 0.1 2.4 ± 0.2 38.3 ± 6.3 62.0 ± 11.8

5.6, 4.9 13.2, 12.8 23.0, 23.2 24.4, 19.6 21.9, 15.4 0.9, 1.0 2.6, 2.6 40.0, 31.7 69.6, 50.8

0.6 ± 0.1 1.0 ± 0.2 10.6 ± 2.3 6.9 ± 1.8 5.3 ± 0.4 0.1 ± 0.1 0.4 ± 0.1 4.5 ± 2.7 11.3 ± 6.5

196.0, 198.0

168.0, 156.0

151.0, 412.0

80.2 ± 19.0

50.9, 69.0

211.0, 130.0

32.9 ± 11.6

57.0 ± 20.6

96.9 ± 23.4

121.0, 85.0

21.4 ± 11.3

246.0, 274.0

168.0, 145.0

166.0, 479.0

73.6 ± 21.5

38.7, 42.0

154.0, 118.0

35.2 ± 14.6

62.0 ± 27.0

88.3 ± 30.2

133.0, 81.1

24.9 ± 14.6

365.0, 489.0

262.0, 295.0

286.0, 787.0

152.3 ± 33.6

110.0, 114.0

387.0, 175.0

92.0 ± 41.3

129.0 ± 53.5

127.2 ± 32.0

218.0, 22.3

51.0 ± 38.5

56.5, 56.0

42.3, 39.3

52.6, 118.1

26.9 ± 3.6

46.1, 33.4

10.4 ± 3.8

18.4 ± 5.9

25.4 ± 5.3

39.3, 26.4

6.8 ± 3.3

C1-fluoranthenes/pyrenes C2-fluoranthenes/pyrenes C3-fluoranthenes/pyrenes C4-fluoranthenes/pyrenes Chrysenea,b Benz[a]anthracenea,b Benzo[a]pyrenea,b,h Perylene Benzo[b]fluoranthenea,b Benzo[j,k]fluoranthenesa,b Benzo[e]pyrene Benzo[ghi]perylenea Indeno[1,2,3-cd]pyrenea,b Dibenz[a,h]anthracenea,b,h Retene ΣPAHs

107.0, 125.0 124.0, 158.0 46.2, 56.2 19.0, 19.3 11.5, 14.9 2.6, 2.5 2.6, 2.3 141.0, 132.0 4.7, 4.9 2.2, 2.2 3.1, 2.9 4.8, 4.8 3.6, 3.1 ND, ND 61.4, 70.9 2016.9, 2233.1 29.4, 32.3 14.2, 12.8 3274.0, 2811.0 1058.5, 1552.3 2.9, 3.1 0.8, 1.5 92.8, 102.7

84.3, 91.4 86.9, 111.0 20.1, 36.1 12.2, 21.1 10.5, 12.3 2.6, 3.2 2.6, 3.5 125.0, 151.0 4.8, 6.7 2.6, 2.9 3.9, 5.1 5.6, 7.3 3.4, 4.3 0.6, 0.8 61.1, 68.1 1647.0, 1883.0 29.5, 36.6 15.6, 16.2 2198.0, 3140.0 1028.8, 1286.9 3.4, 4.3 1.4, 1.2 34.6, 39.5

80.5, 185.0 95.1, 228.0 34.3, 65.5 12.4, 35.0 12.2, 29.0 2.1, 3.9 2.0, 2.5 121.0, 127.0 4.9, 5.3 1.8, 2.8 4.5, 5.2 5.4, 5.6 3.0, 3.6 0.6, 0.7 59.9, 112.0 1825.9, 3464.1

47.7 ± 6.6 61.7 ± 11.4 26.3 ± 4.2 17.1 ± 3.1 8.3 ± 2.6 1.8 ± 0.4 2.3 ± 0.2 117.7 ± 4.6 5.6 ± 0.2 2.5 ± 0.3 4.4 ± 0.03 5.9 ± 0.3 3.8 ± 0.1 0.6 ± 0.1 41.6 ± 6.9 1190.0 ± 170.7

47.9, 41.1 73.4, 39.8 49.0, 19.6 17.0, 6.9 12.2, 5.3 2.2, 1.0 ND,0.2 5.9, 5.2 9.7, 4.3 2.0, 0.5 8.6, 2.4 4.2, 1.9 2.5, 1.6 0.9, 0.6 56.7, 25.1 1574.5, 932.5

21.2 ± 8.1 27.8 ± 9.9 12.4 ± 6.0 6.5 ± 2.8 3.1 ± 3.1 0.5 ± 0.3 0.5 ± 0.3 23.6 ± 24.7 1.1 ± 0.6 0.5 ± 0.4 0.7 ± 0.2 1.2 ± 0.7 0.8 ± 0.3 0.1 ± 0.1 13.4 ± 9.1 409.2 ± 201.4

36.8 ± 10.5 34.7 ± 9.7 13.9 ± 4.7 5.5 ± 0.4 3.5 ± 2.7 1.1 ± 0.2 0.7 ± 0.1 41.2 ± 1.7 2.1 ± 0.1 0.7 ± 0.2 1.3 ± 0.05 2.0 ± 0.1 1.3 ± 0.1 0.3 ± 0.1 25.4 ± 10.6 669.7 ± 276.5

46.5 ± 12.4 39.6 ± 9.4 11.4 ± 2.1 5.2 ± 1.2 4.5 ± 1.2 0.8 ± 0.2 0.7 ± 0.1 34.0 ± 3.7 2.2 ± 0.3 0.7 ± 0.2 1.5 ± 0.1 2.2 ± 0.3 1.6 ± 0.1 0.2 ± 0.03 22.9 ± 5.0 834.3 ± 274.4

64.6, 43.5 60.3, 44.1 28.0, 13.0 9.6, 7.8 7.9, 4.2 1.2, 0.8 1.1, 0.9 22.6, 22.2 3.4, 3.3 0.9, 0.8 2.2, 2.3 3.2, 3.5 1.9, 2.1 0.4, 0.5 29.1, 21.5 1100.8, 659.0

12.2 ± 6.2 13.9 ± 8.7 8.0 ± 4.8 1.9 ± 1.1 2.0 ± 2.6 0.1 ± 0.2 ND 1.4 ± 0.6 ND ND ND ND 0.1 ± 0.05 ND 5.5 ± 5.4 241.1 ± 211.3

29.4, 50.1 15.5, 11.2 3401.0, 3129.0

7.2 ± 5.2 10.7 ± 3.2 12.0 ± 2.1 18.4, 14.7 15.1 ± 6.9 14.9 ± 3.3 15.0 ± 2.8 12.0, 16.0 839.1 ± 171.6 1685.5 ± 612.2 1055.7 ± 270.6 1964.0, 1551.2 552.9 ± 238.6 560.0 ± 61.2 499.0 ± 13.1 869.5, 854.6

2.4 ± 2.8 7.4 ± 2.0 485.2 ± 238.1

1318.9, 1457.1

27.9 ± 3.5 31.4, 38.2 32.8, 14.5 20.0 ± 2.4 17.7, 18.9 10.1, 12.7 3163.7 ± 1680.4 2474.0, 1797.2 3421.4, 2586.0 1207.9 ± 93.8 751.6, 1027.3 1024.3, 602.2

3.4, 3.8 1.3, 1.4 56.6, 107.4

3.3 ± 0.6 1.5 ± 0.9 32.1 ± 2.7

2.3 ± 1.0 1.7 ± 0.8 NAg

4.2 ± 1.2 2.7 ± 1.1 15.7 ± 7.9

Σ7PAHsb % parent PAHs Σ(n-C15–C21) alkanes Σ(n-C25–C31) alkanes OEP C29d e

TAR ΣPAH flux (ng cm−2 y−1)f a b c d e f g h

20.7, 23.7 43.8, 48.7 52.3, 62.4 21.4, 23.2 5.6, 8.6 7.5, 8.8 2.2, 2.5 1.7, 2.3 45.0, 71.3 8.9, 11.4 3.1, 3.4 6.5, 7.5 5.9, 7.2 4.0, 4.9 0.9, 1.2 28.9, 36.0 944.3, 1101.6

3.0, 3.4 0.7, 1.6 30.2, 35.3

1.6, 2.3 0.7, 0.6 94.5, 55.9

2.9 ± 1.0 0.8 ± 0.1 56.3 ± 13.4

4.4 ± 0.5 1.3 ± 0.2 19.2 ± 3.6

3.1, 3.4 1.2, 1.4 41.8, 25.0

34.6

6.7 21.2 86.7/46.9

113.0/ 153

108.0 74.8 88.8

6.2

K.L. Foster et al. / Science of the Total Environment 506–507 (2015) 430–443

101 n

413.7 ± 239.5

US EPA priority pollutants. US EPA probable carcinogens, Σ7PAHs = sum of seven US EPA probable carcinogens (Ch, BA, BbFl, BjkFl, IPy, benzo[a]pyrene, and dibenz[a,h]anthracene). For site 111, n = 1 for alkane analysis. Odd-to-even preference (OEP) centered around C29 was computed as in Scalan and Smith (1970). Terrigenous/aquatic ratios (TARs) calculated as in Bourbonniere and Meyers (1996). Computed as the product of ΣPAH concentration (ng g−1 dw) and sedimentation rate of the core (g cm−2 y−1), see Table 1. NA = sedimentation rate not analyzed. Compounds with N20% NDs that are not included in statistical analyses.

435

436

K.L. Foster et al. / Science of the Total Environment 506–507 (2015) 430–443

10000

Y = 1.82X - 44.82 r2 = 0.86

1000

Alkanes Biphenyls Dibenzothiophenes Heavy PAHs (252-284) Light PAHs (128-184) Medium PAHs (192-244)

100

10

101

1 10000

Y = 0.38X + 46.86 r2 = 0.65

Y = 0.80X + 31.62 r2 = 0.96 1000

100

10

108

1 10000

Y = 2.48X - 107.57 2 r = 0.75

Y = 0.79X + 31.66 r2 = 0.81

Pre-1900s Concentration (ng g-1 dw)

111

1000

100

10

115

1 10000

136 Y = 1.39X + 2.46 r2 = 0.72

Y = 0.91X - 0.85 2 r = 0.96 1000

100

10

141

138

1 10000

Y = 0.68X + 13.14 r2 = 0.64

Y = 0.10X + 38.06 r2 = 0.75 1000

100

10

233

205

1 10000

Y = 0.31X + 9.95 2 r = 0.67

Y = 0.71X + 18.98 r2 = 0.88 1000

100

10

GBF

301

1 1

10

100

1000

10000 1

10

100

1000

10000

1

Post-1900s Concentration (ng g- dw) Fig. 2. Hydrocarbon concentrations in pre- versus post-1900 sediments for 11 sites in Baffin Bay. Diagonal lines indicate perfect agreement between historic and current concentrations (solid line) within a factor of 2 (long dashed lines) and of 10 (dotted lines).

K.L. Foster et al. / Science of the Total Environment 506–507 (2015) 430–443

plankton (Bourbonniere and Meyers, 1996), comprised the other. In general, higher n-alkane profiles of odd chain length formed a rounded peak with n-C27 and n-C29 having the highest concentrations, followed by nC25 and n-C31, then by n-C23 and n-C33. In particular, n-C27, n-C29, and n-C31 are associated with terrigenous higher plant waxes (Peters et al., 2007). The lower n-alkane profiles were dominated by the even chain length n-C14, n-C16 and n-C18 alkanes, which fall within the range of alkanes that are predominant in algae (Bourbonniere and Meyers, 1996). The most notable exception was site 111, where the concentrations of n-C14 through n-C23 dominated the alkane profile with concentrations that were up to 32-fold higher than those measured at the other sites for n-C21, up to 27-fold higher for n-C22, and up to 19-fold higher for n-C18.

437

Both terrigenous and marine organic carbon sources were found to contribute to Baffin Bay sediments; the relative contribution of each varied between sites. Odd-to-even preference (OEP) values were all N1 and ranged from 1.6 to 4.4 indicating that odd carbon numbered n-alkanes, characteristic of terrigenous plant organic matter and immature source rock (Peters et al., 2007), were important at all sites (Table 2). Terrigenous/aquatic ratios (TARs) ranged from 0.4 to 7.3, indicating the greater importance of terrigenous relative to aquatic inputs at most sites. Modeled percent terrigenous organic matter in Baffin Bay sediments, computed as an average of predictions from three bulk organic markers (δ15N, δ13C, C/N) along the length of the cores (pre- and post-1900 sediments), has been found to range from 0 to 47% (Bailey et al., 2013).

A 1000 800

101

600 500

Alkanes Branched PAHs Parent PAHs

400

600

300

400

200

200

100

0 1000

108

0 600 500

800

400 300

400

200

200

100

0 1000

111

0 600 500

800

400

600

300

400

200

200

100

0 1000

115

0 600

PAH Concentration (ng g-1 dw)

Alkane and Retene Concentration (ng g-1 dw)

600

500 800

400

600

300

400

200

200

100

0 1000

136

0 600 500

800

400

600

300

400

200

200

100 0 n-C12 n-C13 n-C14 n-C15 n-C16 n-C17 n-C18 n-C19 n-C20 n-C21 n-C22 n-C23 n-C24 n-C25 n-C26 n-C27 n-C28 n-C29 n-C30 n-C31 Ret N0 N1 N2 N3 N4 P/A0 P/A1 P/A2 P/A3 P/A4 Fl/Py0 Fl/Py1 Fl/Py2 Fl/Py3 Fl/Py4 166 228 252 276 Per

0

Fig. 3. Surficial sediment (post-1900) concentrations of PAHs and alkanes predominantly from plant biomass (green bars), combustion (gray bars), and petrogenic (purple bars) sources from cores collected in Baffin Bay. Shown are n-C12 to C31 alkanes, retene (Ret), naphthalene (N0), C1–4 naphthalenes (N1–4), phenanthrene and anthracene (P/A0), C1–4 phenanthrenes/ anthracenes (P/A1–4), fluoranthene and pyrene (Fl/Py0), C1–4 fluoranthenes/pyrenes (Fl/Py1–4), parent PAHs with molar masses of 166 (fluorene), 228 (benz[a]anthracene and chrysene), 252 (benzo[b]fluoranthene, benzo[j,k]fluoranthenes, and benzo[e]pyrene), and 276 (indeno[1,2,3-cd]pyrene and benzo[ghi]perylene) g mol-1, and perylene (Pery).

438

K.L. Foster et al. / Science of the Total Environment 506–507 (2015) 430–443

B 138

1000

600 500

800

400

600

300

400

200

200

100

0

141

1000

0 600 500

800

400 300

400

200

200

100

0

205

1000

0 600 500

800

400

600

300

400

200

200

100

0

233

1000

0 600 500

800

PAH Concentration (ng g-1 dw)

Alkane and Retene Concentration (ng g-1 dw)

600

400

600

300

400

200

200

100

0

301

1000

0 600 500

800

400

600

300

400

200

200

100

0

GBF

1000

0 600 500

800

400

600

300

400

200

200

100 0 n-C12 n-C13 n-C14 n-C15 n-C16 n-C17 n-C18 n-C19 n-C20 n-C21 n-C22 n-C23 n-C24 n-C25 n-C26 n-C27 n-C28 n-C29 n-C30 n-C31 Ret N0 N1 N2 N3 N4 P/A0 P/A1 P/A2 P/A3 P/A4 Fl/Py0 Fl/Py1 Fl/Py2 Fl/Py3 Fl/Py4 166 228 252 276 Per

0

Fig. 3 (continued).

3.3.2. Petrogenic vs. pyrogenic Hydrocarbon biomarkers consistently indicated a predominance of petrogenic sources of PAHs, as opposed to combustion sources, at all sites and particularly at sites 108 and 111. The percentage of parent PAHs relative to total PAHs in surficial sediments ranged from 7 to 21% depending on the sampling location, indicating a similar, small overall influence of combustion sources (Table 2). Concentrations of

the branched (C1 or greater) PAH homologues were consistently higher than those of the corresponding parent PAHs (C0). Phenanthrenes/anthracenes (P/A1 through P/A4) consistently dominated the PAH chemical profiles with the highest concentrations (Fig. 3). Individual parent PAHs (C0) and branched (C1 to C4) PAH homologues as a percentage of the total concentration of their homologue groups (i.e., N, P/A, and Fl/Py) are shown in Fig. 4. Standardizing

K.L. Foster et al. / Science of the Total Environment 506–507 (2015) 430–443

60

101

N P/A Fl/Py

50 40

439

30 20 10 0 60

108

111

115

136

138

141

205

233

301

GBF

50 40 30 20 10 0 60 50

% PAH Homologue Group

40 30 20 10 0 60 50 40 30 20 10 0 60 50 40 30 20 10 0 60 50 40 30 20 10 0 0

1

2

3

4

0

1

2

3

4

Number of Carbon Substituents Fig. 4. PAH homologue compositions as a percent of total homologue group concentrations for surface sediments at sites across Baffin Bay. Naphthalene (N), phenanthrene/ anthracene (P/A) and fluoranthene/pyrene (Fl/Py) homologue groups are shown with C 0 through to C4 carbon substitutions.

440

K.L. Foster et al. / Science of the Total Environment 506–507 (2015) 430–443

P/A, and Fl/Py homologues was predominated by branched homologues and not parent PAHs. Generally, at all sites the homologue maximum for naphthalenes occurred at either C2 or C3, for P/A at either C2 or C4 and for Fl/Py the maximum concentration occurred at either C1 or C2 (Fig. 4). Therefore, the profiles of all three homologue groups for all eleven sites indicate the predominant origin of PAHs to be petrogenic.

homologue concentrations to the total concentration of the homologue group readily illustrates which number of carbon substituents is dominant within each homologue group. A maximum concentration in a homologue group at C1 or higher is characteristic of mature organic material or petroleum, whereas a maximum at C0 usually indicates combustion sources (Yunker et al., 2002a). The percent composition of N,

Pristane

A n-C16

n-C18

n-C20 n-C19

n-C25

n-C27

n-C23

n-C14

n-C29

Abundance

n-C31

Pristane n-C21

B

n-C22 n-C20 n-C16

Phytane

n-C23

n-C19

n-C25 n-C14 n-C27 n-C29 n-C31

Time (min) Fig. 5. Original extracted ion, m/z 57, chromatograms for Baffin Bay sites A) 205 and B) 111. Unresolved complex mixtures (UCMs) are visible at both sites. Abundances are notably higher at site 111.

K.L. Foster et al. / Science of the Total Environment 506–507 (2015) 430–443

Site

GBF 301 233 205 141 138 136 115 111 108 101 0.20

0.25

0.30

0.35

0.40

0.45

0.50

0.55

Fl/(Fl+Py) GBF 301 233 205 141 138 136 115 111 108 101 0.0

0.1

0.2

0.3

0.4

BA/(BA+Ch) Fig. 6. Ratios for post- (●) and pre- (○) 1900 sediment samples. Ratios typical of petrogenic (purple), mixed combustion (gray), and biomass (green) sources are indicated; ratio boundaries are from Yunker et al. (2011).

Site 111 had an unusual petrogenic PAH signature with average concentrations of branched fluoranthenes/pyrenes and phenanthrenes/anthracenes up to 33- and 28-fold higher for Fl/Py4 and P/A4, respectively, than those of the other sites. For sediment interval 3– 3.5 cm of core 111, concentrations of Fl/Py4 and P/A4 were up to 95and 70-fold higher than concentrations at other sites. Removing interval 3–3.5 cm from consideration, the average of the remaining four post-1900 concentrations of Fl/Py4 and P/A4 for site 111 was up to 18- and 17-fold, respectively, higher than those of other sites and the overall PAH concentrations were similar to those of sites 108 and 138. Higher concentrations of petrogenic compounds at sites 108, 111, and 138 (Fig. 2) are suggestive of a nearby petrogenic source, such as an oil seep. The chromatograms for sites 111 and 205, which are both remote from known oil seeps, do show unresolved complex mixtures (UCMs) (Fig. 5). The abundances of the UCM for site 111 were notably higher. The ratios of Fl/(Fl + Py) and BA/(BA + Ch) also suggest the predominance of petrogenic sources in most samples (Fig. 6), being lower than 0.4 and 0.2, respectively (Yunker et al., 2011). The Fl/(Fl + Py) ratios for Baffin Bay presented here are comparable to those reported for recent Holocene sediments from the Mackenzie River/Delta, and Beaufort Sea in Canada, as well as Nansen Basin (Yunker et al., 2011). However, the ratios were distinct from those of the Laptev Sea, Novaya Zemlya, and SE Barents Seas, which have ratios greater than 0.5 indicating wood, grass or coal combustion as the predominant sources of PAHs (Yunker et al., 2011). Ratios at Baffin Bay sites 136 and 138, which were indicative of mixed PAH sources, were comparable to those measured in the Chukchi Sea and Canada/Makarov Basin, as well as the Greenland Sea and N Barents Sea (Yunker et al., 2011). The degradation of the less stable Fl and BA, has been known to lead to ratio bias in older sediments (Yunker et al., 2002a, 2002b, 2011). Indeed, the two deepest sites, 136 and 138, had ratios that incrementally decreased with core depth, suggesting the gradual decomposition of Fl and BA through the water column and in the older sediments. A number of PAH ratios have been applied to distinguish between petrogenic and pyrogenic sources. Here, the ratios of the concentration of fluoranthene (Fl) to pyrene (Py), computed as Fl/(Fl + Py), and of benz[a]anthracene (BA) to chrysene (Ch), BA/ (BA + Ch), were used because of 100% detection with well-defined peaks in all samples. These are also both established biomarker ratios (Yunker et al., 2002a, 2002b, 2011).

441

3.3.3. Principal component analyses The PAH PCA model found a high degree of correlation between variables with the first two principal components accounting for 86.4% of the total variability (Fig. 7). PC1 (77.8%) grouped samples from polynya sites (101, 108, 111, 115, 136, and 138) separately from non-polynya sites (205, 233, 301, 141, and GBF), with loadings N1 for all PAH variables. Thus, the PC1 axis is an index of PAH concentration, indicating that the sediment samples from polynya sites tended to have higher concentrations of PAHs. Significant differences between polynya and non-polynya samples based on PAH concentrations were found (MANOVA, λ = 0.006, p ≤ 0.0001), and univariate F-tests indicated that the independent PAH variables in all cases were not equal between polynya and non-polynya samples (p ≤ 0.001). A similar grouping of samples from polynya and non-polynya sites for PC1 was also observed for the alkane-only PCA model (not shown). PC2 (8.6%) axis scores distinguished between sites based on PAH signatures, providing an indication of the probable source. The PCA model contrasted light (128–184 g mol−1) and heavy (252–284 g mol−1) with medium (192–244 g mol−1) PAHs. Light and heavy PAHs had positive PC2 axis loadings and projected to the top of the loadings plot. PAHs falling into this region of the loadings plot included most unsubstituted parent PAHs (there were four exceptions: Fl, Py, BA, and Ch, which had negative loadings), all naphthalenes (N0–N4), and all five ring PAHs. PAHs with negative PC2 axis loadings were predominantly medium molar mass PAHs including the four unsubstituted PAHs noted above and branched PAHs and homologue groups with between two and four aromatic rings. Unsubstituted PAHs, especially those with four to six rings, are associated with combustion as opposed to petrogenic sources (Laflamme and Hites, 1978). Thus, negative PC2 axis scores are indicative of petrogenic sources and positive scores are indicative of combustion sources. Most of the sediment core samples grouped around the PC1 and PC2 origins, however, sites 111, 141, and GBF stand out as having distinct PAH signatures. The positive PC1 values group all site 111 samples with the samples from other sites within the NOW, whereas the negative PC1 values group sites 141 and GBF with other non-polynya sites. Three samples from site 111, representing sediment intervals from 3 to 4.5 cm in the core, and the two top GBF samples (1.5–2.0 cm and 5.5–6.0 cm), have the lowest PC2 axis scores, indicating PAH signatures more consistent with petrogenic sources. Site 141, similar to GBF is also located near the known Scott Inlet oil seep (Fig. 1) and also generally had negative PC2 axis scores indicative of petrogenic sources, with one exception in a pre-1900 sample (9.5–10.0 cm). 4. Conclusions The data presented provide a baseline record of hydrocarbon concentrations and chemical profiles in northern Baffin Bay sediments. With so little known about hydrocarbons in northern Baffin Bay, the significance of these results is that they provide a preliminary characterization of the spatial variability in both present-day and historic sediments (pre-1900), and an indication of the relative importance of different natural sources (marine versus terrigenous). In general, concentrations and fluxes of hydrocarbons to sediments were higher within the polynya (101, 108, 111, 115, 136, and 138) than elsewhere (141, 205, 233, 301, and GBF). Whether this indicates a local source of hydrocarbons, particularly petrogenic PAHs, to the NOW or a different mechanism for sequestering hydrocarbons from the water column by suspended particulates and subsequent deposition to the sediments, remains unclear. However, the hydrocarbon biomarkers and chemical profiles in the sediments suggest the predominance of petrogenic, as opposed to combustion, PAHs at all sites included in this study. An important conclusion from these results is that northern Baffin Bay appears to be relatively clean with regard to hydrocarbons. The current burden of PAHs at these eleven sites is predominantly the result of local, diffuse petrogenic sources to Baffin Bay rather than combustion

442

K.L. Foster et al. / Science of the Total Environment 506–507 (2015) 430–443

4

A Sites

PC2 (8.6 %)

2

Polynya 101 108 111 115 136 138 Non-Polynya 141 205 233 301 GBF

0

-2

-4

-6 -20

-10

0

10

PC1 (77.8 %) 0.6

B PC2 Loadings (8.6 %)

0.4

0.2

IPy* BPer* N1 BFl/BPy2 BbFl* N0* Per* Ace* BFl/BPy1 F* BjkFl* N3 N4F1 P/A0* BA/Ch3

BPy* N2 BA/Ch4

Molar Mass (g mol-1)

BA/Ch2

0.0 F2

-0.2

F3 P/A1

BA*

Light (128-184) Medium (192-244) Heavy (252-284)

BA/Ch1 Ret Fl*

Fl/Py4 Fl/Py1 Ch* Fl/Py2

-0.4

-0.6 0.70

P/A4 Fl/Py3 P/A3

0.75

0.80

P/A2 Py*

0.85

0.90

0.95

1.00

PC1 Loadings (77.8 %) Fig. 7. Principal component A) scores and B) loadings for PAH concentrations in Baffin Bay sediment cores for all samples (pre and post-1900). Loadings for parent (un-substituted PAHs) are indicated by ‘*’. See text for site and compound abbreviations.

sources, which are typically associated with industrial activity, the products of which would have to be delivered through long-range atmospheric transport. Surficial sediments from northern Baffin Bay, reflecting present-day levels, were also found to have hydrocarbon concentrations consistently within a factor of 10 of historic, pre-industrial sediments. If recent, post-industrial anthropogenic sources of PAHs are reaching Baffin Bay, they would appear to contribute less than 10% to the natural background and are, therefore, difficult to detect given the natural variability. Acknowledgments The authors are grateful to the Nunavut General Monitoring Plan (NGMP) and ArcticNet, a Canadian Network of Centres of Excellence, for supporting this research. Thank you to the officers and crew of the CCGS ship the Amundsen. Thanks also to Pond Inlet community members, especially Shelly Elverum and the Hamlet Council as well as the Nunavut Research Institute (Iqaluit) for hosting KLF and for their valuable feedback on this research. References Alkire MB, Falkner KK, Boyd T, Macdonald RW. Sea ice melt and meteoric water distributions in Nares Strait, Baffin Bay, and the Canadian Arctic Archipelago. J Mar Res 2010; 68:767–98.

AMAP. AMAP Assessment 2007: oil and gas activities in the arctic — effects and potential effects. Arctic Monitoring & Assessment Programme, Oslo, vols. 1 and 2. ; 2010. Bailey JN-L, Macdonald RW, Sanei H, Outridge PM, Johannessen SC, Hochheim K, et al. Change at the margin of the North Water Polynya, Baffin Bay, inferred from organic matter records in dated sediment cores. Mar Geol 2013;341:1–13. Blasco KA, Blasco SM, Bennett R, MacLean B, Rainey WA, Davies EH. Seabed geologic features and processes and their relationship with fluid seeps and the benthic environment in the Northwest Passage. Geological Survey of Canada, Open File 6438; 2010. (57 pp.). Boitsov S, Petrova V, Jensen HKB, Kursheva A, Litvinenko I, Klungsøyr J. Sources of polycyclic aromatic hydrocarbons in marine sediments from southern and northern areas of the Norwegian continental shelf. Mar Environ Res 2013;87/88:73–84. Bourbonniere RA, Meyers PA. Sedimentary geolipid records of historical changes in the watersheds and productivities of Lakes Ontario and Erie. Limnol Oceanogr 1996;41: 352–9. CCME, 2012. Canadian Council of Ministers of the Environment. http://www.ccme.ca/. Accessed Nov. 23rd, 2012. De Laender F, Hammer J, Hendriks AJ, Soetaert K, Janssen C. Combining monitoring data and modeling identifies PAHs as emerging contaminants in the Arctic. Environ Sci Technol 2011;45:9024–9. Gautier DL, Bird KJ, Charpentier RR, Grantz A, Houseknecht DW, Klett TR, et al. Assessment of undiscovered oil and gas in the Arctic. Science 2009;324:1175–9. Grosjean E, Logan GA. Incorporation of organic contaminants into geochemical samples and an assessment of potential sources: examples from Geoscience Australia Marine Survey S282. Org Geochem 2007;38:853–69. Laflamme RE, Hites RA. The global distribution of polycyclic aromatic hydrocarbons in recent sediments. Geochim Cosmochim Acta 1978;42:289–303. Lima ALC, Farrington JW, Reddy CM. Combustion-derived polycyclic aromatic hydrocarbons in the environment — a review. Environ Forensics 2005;6:109–31. Liu X, Jia H, Wang L, Qi H, Ma W, Hong W, et al. Characterization of polycyclic aromatic hydrocarbons in concurrently monitored surface seawater and sediment along the Dalian coast after oil spill. Ecotoxicol Environ Saf 2013;90:151–6.

K.L. Foster et al. / Science of the Total Environment 506–507 (2015) 430–443 Macdonald RW, Harner T, Fyfe J. Recent climate change in the Arctic and its impact on contaminant pathways and interpretation of temporal trend data. Sci Total Environ 2005;342:5–86. Obermajer M, Dewing K, Fowler MG. Geochemistry of crude oil from Bent Horn field (Canadian Arctic Archipelago) and its possible Paleozoic origin. Org Geochem 2010; 41:986–96. Perovich D, Gerland S, Hendricks S, Meier W, Nicolaus M, Richter-Menge J, et al. Sea ice [in Arctic Report Card 2013]. http://www.arctic.noaa.gov/reportcard, 2013. Peters KE, Walters CC, Moldowan JM. The biomarker guide v.2: biomarkers and isotopes in petroleum exploration and earth history. NY, USA: Cambridge University Press; 2007. Scalan RS, Smith JE. An improved measure of the odd-even predominance in the normal alkanes of sediment extracts and petroleum. Geochim Cosmochim Acta 1970;34: 611–20. Skillicorn D. Understanding complex datasets: data mining with matrix decompositions. Boca Raton, FL, USA: Chapman & Hall/CRC; 2007. Wassmann P, Bauerfeind E, Fortier M, Fukuchi M, Hargrave B, Moran B, et al. Particulate organic carbon flux to the Arctic Ocean sea floor. In: Stein R, Macdonald RW, editors. The organic carbon cycle in the Arctic Ocean. Berlin: Springer-Verlag; 2004.

443

Yunker MB, Macdonald RW. Composition and origins of polycyclic aromatic hydrocarbons in the Mackenzie River and on the Beaufort Sea shelf. Arctic 1995;48:118–29. Yunker MB, Macdonald RW, Veltkamp DJ, Cretney WJ. Terrestrial and marine biomarkers in a seasonally ice-covered Arctic estuary — integration of multivariate and biomarker approaches. Mar Chem 1995;49:1–50. Yunker MB, Macdonald RW, Vingarzan R, Mitchell H, Goyette D, Sylvestre S. PAHs in the Fraser River basin: a critical appraisal of PAH ratios as indicators of PAH source and composition. Org Geochem 2002a;33:489–515. Yunker MB, Backus SM, Graf Pannatier E, Jeffries DS, Macdonald RW. Sources and significance of alkane and PAH hydrocarbons in Canadian arctic rivers. Estuar Coast Shelf Sci 2002b;55:1–31. Yunker MB, Macdonald RW, Snowdon LR, Fowler BR. Alkane and PAH biomarkers as tracers of terrigenous organic carbon in Arctic Ocean sediments. Org Geochem 2011;42:1109–46.