Hexachlorocyclohexanes (HCH) in ringed seal (Phoca hispida) from Ulukhaktok (Holman), NT: Trends from 1978 to 2006

Science of the Total Environment 407 (2009) 5139–5146

Contents lists available at ScienceDirect

Science of the Total Environment j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c i t o t e n v

Hexachlorocyclohexanes (HCH) in ringed seal (Phoca hispida) from Ulukhaktok (Holman), NT: Trends from 1978 to 2006 R.F. Addison a,⁎, D.C.G. Muir b, M.G. Ikonomou c, L. Harwood d, T.G. Smith e a

1705 Eagle View Place, Duncan, BC, Canada V9L 6R1 Environment Canada, Aquatic Ecosystem Protection Research Division, Burlington, ON, Canada L7R 4A6 Department of Fisheries and Oceans, Institute of Ocean Sciences, P.O. Box 6000, Sidney, BC, Canada V8L 4B2 d Department of Fisheries and Oceans, 101 5204, 50th Avenue, Yellowknife, NT, Canada X1A 1E2 e EMC Eco Marine Corporation 5694 Camp Comfort, Garthby, PQ, Canada G0Y 1B0 b c

a r t i c l e

i n f o

Article history: Received 27 February 2009 Received in revised form 21 May 2009 Accepted 26 May 2009 Available online 27 June 2009 Keywords: Hexachlorocyclohexanes (HCH) Arctic Ringed seal (Phoca hispida) Trends Ulukhaktok (Holman)

a b s t r a c t Trends in α-, β-, and γ-hexachlorocyclohexane (HCH) concentrations were examined in blubber lipid of ringed seals (Phoca hispida) from Ulukhaktok (Holman), NT (Canada) sampled at intervals between 1978 and 2006. α-HCH usually represented approximately 90% of the total HCH isomers. α-HCH and γ-HCH concentrations showed no change over the sampling interval, but β-HCH concentrations increased significantly, about 8–10-fold in females and 4–5-fold in males. Residue concentrations showed no dependence on age. Concentrations (all data as ng/g lipid, GM (range)) of α-HCH were significantly higher (Pb 0.001 by t-test) in males (217 (93.9–517), n=37) than those in females (138 (40.9–402), n=38). β-HCH concentrations did not differ between the sexes. Concentrations of γ-HCH were significantly higher (Pb 0.05) in males (6.74 (0–46.7)) than in females (4.35 (0–19.0)). Although global emissions of both α-HCH and β-HCH have declined since the early 1980's, the “signal” of HCH emission changes has not yet resulted in a “response” in ringed seal residue concentrations. In the light of our current understanding of the dynamics of HCH in the Arctic, we conclude that any such response may not be detected by retrospective analyses of the sort describe here at least for another decade or so, because of the longevity of the seals. © 2009 Elsevier B.V. All rights reserved.

1. Introduction Hexachlorocyclohexanes (HCHs) are among the legacy chlorinated hydrocarbon pesticides found as contaminants in Arctic biota. Three isomers are usually reported: α-HCH, β-HCH and γ-HCH (lindane) which differ in the conformation of the Cl substituents on the cyclohexane molecule. Technical HCH is usually comprised of 60–70% α-HCH, 5–12% β-HCH, 10–12% γ-HCH (the most insecticidally potent of the isomers) and 10–15% (together) of the δ-and ε-isomers (Willett et al., 1998). Several factors control the distribution of HCHs and other persistent organic pollutants (POPs) in Arctic ecosystems. Briefly, these can be summarised as: (i) source functions, i.e., scale of production, use and emission patterns, and transport to the Arctic. Global emissions of α-HCH and β-HCH reached a maximum during the 1970's–early 1980's and then began to decline to near zero by 2000 (Li and Macdonald, 2005). However, the distribution of HCHs in western North American Arctic ecosystems may be more directly affected by regional use of these

⁎ Corresponding author. Tel./fax: +1 250 597 1558. E-mail address: [email protected] (R.F. Addison). 0048-9697/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2009.05.049

chemicals (e.g., Wania and Mackay, 1999; Muir et al., 2000). α-HCH and γ-HCH were introduced to the Arctic mainly through atmospheric transport until the late 1990's but β-HCH was, and is, transported mainly by ocean currents via the Bering Strait (Li et al., 2002; Li and Macdonald, 2005). (ii) physico-chemical properties of HCH isomers (Henry's Law constant H, which controls air–sea exchange, depending on ice cover, and indices such as octanol–water (Kow) or organic carbon partition coefficients (Koc) which control partitioning into biota or on to particulate material). The much lower value of H for β-HCH than for α-HCH or γ-HCH probably explains its tendency to enter the Arctic through oceanic advection rather than atmospheric transport (Li and Macdonald, 2005). Bio-concentration factors (estimated for aquatic animals) of the three isomers are approximately similar (Willett et al., 1998). (iii) chemical and biological degradation, probably depending mainly or initially on microbial degradation, but including processes in higher biota. Microbial degradation has been proposed as the main process of removal of HCH from water, and is faster and more significant than hydrolysis or sedimentation in Arctic marine systems (Harner et al., 1999, 2000); it is also significant in freshwater systems (e.g., Helm et al., 2000). Ringed seals and some seabirds can also probably degrade some HCH components (e.g., Moisey et al., 2001;

5140

R.F. Addison et al. / Science of the Total Environment 407 (2009) 5139–5146

Table 1 Comparison of analyses of standard reference materials (SRM) by Axys Laboratories in 1995 and by CCIW in 2008.

NIST 1588 Certified value Found (Axys 1995) NIST 1588b Certified value Found (CCIW 2008)

α-HCH

γ-HCH

86 ± 19 (unspecified) 83 ± 11 (6)

24 ± 6 (unspecified) 17 ± 2.3 (6)

99 ± 15 (unspecified) 98.8 ± 5.47 (10)

23.3 ± 1.7 (unspecified) 19 ± 2.12 (10)

Concentrations in ng/g; data as mean ± s.d. (no. of observations).

Fisk et al., 2002; Hoekstra et al., 2003b). Photolysis, hydrolysis and isomerisation are not considered to be significant pathways for degradation of any of the HCH isomers under Arctic conditions (UNEP, 2006, 2007a,b). In an analysis of trends in persistent organochlorines in blubber from ringed seal (Phoca hispida) collected at Holman, NWT (now Ulukhaktok, NT), at intervals from 1972 to 1991, Addison and Smith (1998) noted that between 1981 and 1991, concentrations of α-HCH and γ-HCH showed no significant changes, even though global HCH emissions had declined dramatically during the 1980's. This was in contrast to the DDT-group of insecticides and the polychlorinated biphenyls (PCBs) both of whose concentrations responded fairly rapidly, within a decade or so, to reductions in global emissions during the 1970's. Addison and Smith (1998) proposed that the relatively slow response of HCH concentrations in seals during the 1981–1991 interval to global declines in HCH emissions might reflect the higher water solubility of HCHs relative to that of the DDT group or the PCBs; this might control the adsorption of HCHs to small particulate material which is the probable first step in the bioaccumulation process. They also predicted that if the over-saturated α-HCH concentrations in Arctic surface waters began to “outgas” around 1990, as implied by the data of Jantunen and Bidleman, (1995), and subsequently interpreted as a response to the decline in α-HCH emissions at that time (Li and Bidleman (2003), then α-HCH concentrations in Arctic ringed seals should begin to decline some time during the 1990's. The ringed seal population at Ulukhaktok has been sampled at intervals since 1972. In this paper, we describe the trends in HCH concentrations in this population sampled at intervals between 1978 and 2006, and interpret these in the light of our current understanding of HCH behaviour in the Arctic ecosystem. 2. Materials and methods Ringed seals which forage throughout the southern Beaufort Sea were collected during subsistence hunts during May and June at Ulukhaktok, NT (77° 44′ N, 117° 43′ W) in 1978, 1981, 1991, 1996, 2001 (female samples only) and 2006. General sampling conditions are described elsewhere (Addison and Smith 1998; Harwood et al., 2000; Muir et al., 2007). Blubber samples were taken from the mid-dorsal region, and sex and blubber thickness at the sternum of the samples

were recorded. Age determinations were made by reading the dentinal annuli of tooth cross sections under transmitted light (Smith 1973). For 1978, 1996, 2001 and 2006 samples, which had not been analysed previously, sub-samples of blubber tissue minus skin and muscle were homogenized with sodium sulfate and extracted (Soxhlet) with dichloromethane–hexane (1:1). HCHs were isolated by gel permeation chromatography (GPC) followed by silica gel cleanup, and analysed by gas chromatography with electron capture detection (GC-ECD). Appropriate blank and quality control samples were run. Standard reference materials (SRM) (NIST 1588b, a cod liver oil) were included with every 12 samples to assure inter-comparability. The analyses were conducted at Environment Canada labs. in Canada Centre for Inland Waters (CCIW). 1981 and 1991 samples had previously been analysed by Axys Laboratories (Sidney, BC) in 1995, using approaches generally comparable to those described above, and using SRM NIST 1588 (cod liver oil) as a standard. To confirm consistency between the two sets of analyses, six archived samples each from 1981 and 1991 were re-analysed at CCIW and compared to the previous data. Data were expressed as concentrations in extractable lipid and were analysed statistically using Statistica release 8 (Statsoft, Tulsa, OK). 3. Results and discussion 3.1. QA/QC; consistency among analyses Detection limits for α-, β-and γ-HCH were usually in the range of 0.05–0.1 ng/g lipid (Axys analyses) and 0.01 ng/g lipid (CCIW analyses). Concentrations of β-HCH and γ-HCH in the samples were usually at least 50-fold, and of α-HCH, N1000-fold above these limits. A major difficulty in comparing data collected by different laboratories over long intervals is the consistency of analyses. We have approached the issue by relying on the analysis of SRM to ensure general data quality, and by re-analysing archived samples previously analysed elsewhere to assess the comparability of current and previous analyses. Table 1 summarises the performance of the two labs. in analysing SRM; both labs. were very close to the target value for α-HCH but tended to under-estimate γ-HCH concentrations. (No certified value for β-HCH was provided.) Table 2 shows the results of re-analysis by CCIW of 12 samples previously analysed by Axys, and examined statistically by t-test for paired samples. Determinations of β-HCH and of γ-HCH did not differ significantly (P N 0.05) but mean concentrations of α-HCH differed by about 23%. Furthermore, variance among the Axys analyses (SD expressed as percentage of the mean) was wider than in the later CCIW analyses. Since both laboratories had analysed the SRMs accurately for α-HCH, we cannot easily explain the discrepancy in α-HCH estimations. Ratios of CCIW re-analysis: Axys original analysis for α-HCH and γ-HCH did not differ between 1981 and 1991 samples (1-way ANOVA: data not shown) which suggests

Table 3 Age and blubber thickness (sternum) in Ulukhaktok, (Holman) NT, ringed seals (P. hispida) sampled between 1978 and 2006. Table 2 Concentrations of α-HCH, β-HCH and γ-HCH in 12 blubber samples from Ulukhaktok, (Holman) NT, ringed seals (P. hispida) analysed by Axys Laboratories in 1995 and re-analysed by CCIW in 2008. Laboratory

Axys (1995)

CCIW (2008)

α-HCH β-HCH γ-HCH

192 ± 94.5 (12) 14.3 ± 10.5 (10) 9.54 ± 6.38 (11)

237 ± 74.6 (12) 15.4 ± 8.85 (10) 9.16 ± 4.87 (11)

t = 2.46, P = 0.03 t = 0.63, P = 0.55 t = −0.47, P = 0.64

Data as mean ± s.d. (no. of samples). Data were analysed statistically by t-test for paired samples.

Year

1978 1981 1991 1996 2001 2006

Females

Males

Age (year)

Blubber thickness (cm)

Age (year)

Blubber thickness (cm)

5.67 ± 4.93a (3) 10.8 ± 5.20a (8) 14.3 ± 7.34a (6) 11.5 ± 2.35a (6) 14.5 ± 5.96a (10) 20.8 ± 3.96b (5)

4.70 ± 0.92a (3) 4.57 ± 0.67a (7) 4.20 ± 1.02a (6) 4.10 ± 1.89a (6) 2.90 ± 0.61a (10) 3.20 ± 0.91a (5)

3.50 ± 4.51a (4) 9.50 ± 3.41a (10) 12.3 ± 3.55b (11) 12.2 ± 3.06b (6) No sample 10.4 ± 3.13a (5)

3.96 ± 0.51ab (5) 4.66 ± 0.60a (10) 3.45 ± 0.90b (11) 3.98 ± 0.65ab (6) No sample 2.80 ± 0.76b (5)

Data are presented as means ± s.d. (no. of samples). Data in the same column with the same superscript do not differ significantly (P N 0.05) by one-way ANOVA.

R.F. Addison et al. / Science of the Total Environment 407 (2009) 5139–5146

5141

Table 4 Concentrations (ng/g lipid) of α-HCH, β-HCH and γ-HCH in blubber of male and female ringed seals (P. hispida) from Ulukhaktok (Holman) NT between 1981 and 1996. Sample/year (no. of samples)

α-HCH

β-HCH

γ-HCH

Total HCH isomers

Females 1978 (n = 3) 1981 (n = 6) 1991 (n = 5) 1996 (n = 6) 2001 (n = 10) 2006 (n = 6)

197 (114–402)a 126 (63–250)a 176 (110–280)a 156 (90.5–288)a 117 (57.9–347)a 113 (40.9–231)a

5.97 (3.58–13.0)ab 4.41 (0–6.20)a 11.6 (8.20–18.0)bc 18.9 (10.8–35.2)cd 27.3 (21.2–45.5)de 45.1 (34.4–96.6)e

3.74 (1.59–11.6)a 6.27 (0–9.50)a 5.76 (3.10–19.0)a 5.27 (4.10–8.02)a 4.49 (1.79–16.2)a 6.11 (2.61–11.8)a

209 (119–417)a 132 (63.0–263)a 194 (123–311)a 184 (117–320)a 153 (86.3–386)a 169 (79.1–283)a

Males 1978 (n = 5) 1981 (n = 10) 1991 (n = 11) 1996 (n = 6) 2006 (n = 5)

295 (141–401)a 190 (97–260)a 178 (100–230)a 326 (151–517)a 194 (93.9–474)a

7.19 (3.82–30.2)ab 6.73 (2.80–10.0)a 17.7 (9.10–40.0)bc 37.8 (21.9–96.6)c 29.8 (20.7–40.0)c

7.86 (4.75–12.4)a 10.5 (0–19.0)a 6.08 (2.70–20.5)a 11.5 (5.12–46.7)a 4.06 (2.76–5.84)a

310 (149–422)a 206 (101–286)a 203 (126–501)a 383 (198–661)a 231 (125–508)a

Data are presented as geometric means (ranges) (no. of samples). Within each sex, data in the same column with the same superscript do not differ significantly by one-way ANOVA (P N 0.05) of log-transformed data (see text).

that the laboratory differences were not related to sample storage over that interval and that there was no deterioration of residues during storage. For the following discussion, we have ignored the 23% discrepancy in analyses of α-HCH in archived samples, since (as we discuss further below) it does not affect our general conclusions. 3.2. Biological variables and HCH concentrations in Ulukhaktok samples, 1978–2006 Table 3 summarises age and blubber thickness in the samples. Data were distributed normally (P for Shapiro–Wilk W N 0.05 for both variables within each sex). One-way ANOVA showed that among the females, only the 2006 samples were significantly older than the others; among the males, the 1991 and 1996 samples were significantly older than the rest. Blubber thickness declined throughout the sampling period; regression of blubber thickness (cm) on year yielded slopes significantly different (P b 0.01) from zero in both females (− 0.067 ± 0.018 [mean ± SE], n = 37) and males (−0.054 ± 0.014, n = 37). This downward trend in seal body condition appears related to earlier break-up of the land fast ice, as discussed by Harwood et al. (2000, 2008). Table 4 summarises concentrations of α-HCH, β-HCH and γ-HCH, separated by sex. In all samples, α-HCH predominated, usually representing 90% or more of the sum of the three isomers. When aggregated within sexes (since sample sets in any 1 year were too small to allow discrimination between normal and log-normal distribution) data were usually distributed log-normally (P for Shapiro–Wilk W calculated for log residue concentrations N0.05, except for β-HCH in females). Data are therefore presented as geometric means (GM) and ranges (or 95% CL in the Figures) and the following statistical comparisons were made on ln-transformed data. HCH concentrations measured here are similar to those reported in other ringed seal samples from the Alaskan Beaufort or Chukchi Seas (e.g., Hoekstra et al., 2003a). When all males and all females (regardless of sampling dates) were treated as two groups (since there were no significant differences within sexes and between samples from different years: Table 4) α-HCH, γ-HCH and ΣHCH were all significantly higher in males than in females (Table 5). β-HCH concentrations did not differ significantly by t-test between the sexes, presumably because of the large variance introduced by the steady increase in β-HCH concentrations over the sampling period (see below). The generally lower concentrations of HCHs in females probably reflect their clearance during lactation, as is the case with the DDT-group and PCBs (Addison, 1989). Residue concentrations in the four subsets which had N8 individuals (1981 males and females, 1991 males and 2001 females) did not increase significantly with age, except for β-HCH in 2001 females (data not shown).

3.3. Trends in HCH isomer concentrations, 1978–2006 Figs. 1 and 2 show trends in HCHs in blubber of female and male ringed seals, respectively, from 1978 to 2006. In females, α-HCH, γ-HCH and ΣHCH concentrations showed no significant changes over the sampling period. In males, ΣHCH varied significantly (1-way ANOVA: Pb 0.05) and α-HCH almost significantly (1-way ANOVA, P=0.054) over the interval. However, regressions of α-HCH and ΣHCH concentrations on sampling year showed no statistically significant trends, i.e., slopes of regressions were not significantly different from zero (PN 0.05; data not shown). If α-HCH concentrations were reduced by 23% in 1978, 1996, 2001 and 2006 samples (see Section 1 above) neither α-HCH nor ΣHCH in either sex varied significantly between 1978 and 2006 by 1-way ANOVA (data not shown). β-HCH concentrations in both females and in males increased significantly over the sampling interval, by about 8–10-fold in females and 4–5-fold in males. We now discuss these data in the light of our current understanding of HCH dynamics in the Arctic. Consider first α-HCH. Global emissions of α-HCH rose steadily from the mid-1940's and reached a peak in the early 1980's after which they fell sharply (Li and Macdonald, 2005). Model calculations (Li et al., 2004a) suggest that α-HCH was delivered to the Arctic mainly by atmospheric transport until the late 1990's, when ocean current advection became the main route. Concentrations of α-HCH in Arctic air sampled at various locations were well correlated with emissions throughout the 1980's and 1990's, which implies a rapid (b1 year) transport from source (Li and Bidleman, 2003). Although the emission peak, and its subsequent decline, are postulated to have been reflected within a year or so in surface water concentrations of α-HCH in the North American Arctic Ocean, atmosphere–surface water exchange was probably hindered to some extent by ice cover (Li and Macdonald, 2005). There has been no consistent monitoring of surface water HCH concentrations in the region but the sporadic data available suggest that there has been no clear decline in α-HCH concentrations in surface waters of the Chukchi

Table 5 Concentrations (ng/g lipid) of α-HCH, β-HCH, γ-HCH and ΣHCH (sum of all three isomers) in female and male ringed seals from Ulukhaktok, (Holman) NT, aggregated among years as described in the text. Residue

Females (n = 38)

Males (n = 37)

t

P

α-HCH β-HCH γ-HCH ΣHCH

138 (40.9–402) 12.4 (0–96.6) 4.35 (0–19.0) 165 (63.0–417)

217 45.3 6.74 244

− 3.99 − 0.7 − 2.32 − 3.74

b0.001 0.48 b0.05 b0.001

(93.9–517) (2.80–96.6) (0–46.7) (101–661)

Data are presented as geometric means (ranges). Statistical comparisons are by t-test on log-transformed data. P is probability that concentrations differ significantly between sexes.

5142

R.F. Addison et al. / Science of the Total Environment 407 (2009) 5139–5146

Fig. 1. Trends in concentrations (ng/g lipid) of α-, β-, γ-HCH and total HCH (sum of isomers) in blubber from female ringed seals (P. hispida) from Ulukhaktok (Holman), NT. Data presented as GM (bars) and 95% CL (whiskers) as described in text.

Sea–Beaufort Sea region between about the late 1970's and the late 1990's (Jantunen and Bidleman, 1995; Bidleman et al., 2007). This implies that changes in α-HCH contamination of Beaufort–Chukchi Seas surface water were essentially averaged over a period of a several years, perhaps up to a decade, and any changes in α-HCH contamination of the lower trophic levels of the Beaufort Sea marine food web in response to a decline in (global) emissions would therefore probably not have begun until the mid-1990's. As Table 3 and Figs. 1 and 2 show, there was no indication of any decline in α-HCH concentrations in any of the Ulukhaktok ringed seal samples taken between 1991 and 2006. Since the estimated ages of individual animals are available, α-HCH concentrations can be plotted against year of birth, i.e., year when exposure to the Beaufort Sea food web began. (Exposure to contamination in the Beaufort Sea of course preceded an animal's birth date, since it inherited contaminants from its mother [Addison, 1989] but since there was no significant temporal decline in α-HCH tissue concentrations over the period “sampled” by these animals (Figs. 1 and 2) we assume that all individuals were born with similar residue burdens.) Fig. 3(a) and (b) show such plots for female and male seals, respectively There was no statistically significant trend (i.e., slopes of the regressions did not differ significantly from zero). If the data were divided into pre-1983 and post-1982 samples (since the maximum of Arctic air concentrations occurred in 1982: Li and Bidleman, 2003), the pre-1983 samples again showed no statistically

significant trend. Post-1982 female samples had slightly a negative slope, but this was not statistically significant. The absence of any clear temporal trend in α-HCH concentrations in the Ulukhaktok ringed seal blubber, at least from the 1980's onwards, is therefore generally consistent with Chukchi Sea–Beaufort Sea surface water data. The absence of any clear increasing trend in blubber α-HCH concentrations before the 1980's is less easy to explain, but may be attributable to (a) the time-averaging of atmosphere–surface water exchange of α-HCH, (b) the time required for the processes of adsorption of α-HCH to small particulate material (the probable “point of entry” to the food web) and (c) time required for subsequent bioaccumulation processes. In other words, the ratedetermining steps in the availability of α-HCH to the seals are presumably, first, the input (either by air–sea exchange or by water advection) and second, the transfer of α-HCH through the seals' food web to the trophic level at which they prey. The β-HCH trend differs from that of α-HCH. β-HCH global emissions followed those of α-HCH, though on a different scale, reaching a maximum in the early 1980's (Li and Macdonald, 2005). However, the relatively low H of β-HCH is predicted to have led to its rapid partitioning into surface waters south of the Bering Strait, which appears to act as a bottleneck in its transport to the southern Arctic Ocean and so has delayed the arrival of the emission peak there until the early 1990's (Li and Macdonald, 2005). Fig. 4(a) and (b) show the

R.F. Addison et al. / Science of the Total Environment 407 (2009) 5139–5146

5143

Fig. 2. Trends in concentrations (ng/g lipid) of α-, β-, γ-HCH and total HCH (sum of isomers) in blubber from male ringed seals (P. hispida) from Ulukhaktok (Holman), NT. Data presented as GM (bars) and 95% CL (whiskers) as described in text.

regressions of β-HCH concentrations on year of birth in female and male seals (analogous to the data in Fig. 3); concentrations increased consistently from the mid-1960's until the mid-1990's. The faster increase in of β-HCH concentrations in females than in males is consistent with a faster turnover (due to birth and lactation) followed by replenishment of β-HCH during more recent feeding when residue concentrations were higher. This trend is again consistent with the expected trend in surface water β-HCH concentrations in the southern Arctic. The absence of any decline in β-HCH concentrations in ringed seal blubber even by the mid-1990's and possibly beyond, suggests that any change in β-HCH concentrations in the seals' prey has lagged a change in surface water concentrations by at least 2–5 years. Another factor which may contribute to the difference between α-HCH and β-HCH accumulation in seals is the difference in octanol/air partition coefficient (Koa) which is about 10 times higher in β-HCH than in α-HCH (Shoeib and Harner, 2002); even though Kow for the two isomers are similar, the higher Koa of β-HCH would tend to increase its accumulation by the seals (cf. Kelly et al., 2007). We also note that the scatter of the β-HCH data is much lower than that of either the α-HCH or γ-HCH data (apart from one clear outlier in 1978). This may indicate that the supply of β-HCH to the southern Arctic food web via (mainly) water advection is more consistent than that of α-HCH or γ-HCH via (mainly) air–sea exchange (at least until the late 1990's) which may have varied from year to year as a result of changes in ice-cover. Whatever the processes of bioaccumulation of α-HCH and β-HCH, similar increases in β-HCH concentrations have been recorded in Lake Baikal seals (P. sibirica)

between 1992 and 2002 (Tsydenova et al., 2004) and in central Arctic sea-birds between 1975 and 2003 (Braune et al., 2001; Braune 2007). Fig. 5(a) and (b) show the regressions of γ-HCH concentrations in blubber on year of birth. As in the case of α-HCH, there is no evidence of any temporal trend. No estimates of global γ-HCH emissions appear to have been published, but in the circum-polar northern hemisphere, use of γ-HCH in the former Soviet Union reached a peak of approximately 18,000 tonnes in 1965 and then declined to about 4000 tonnes by the late 1980's (Li et al., 2004c); γ-HCH use in Europe declined from about 8000 tonnes/year in the early 1970's to less than 5000 tonnes/year by the mid-1990's (Breivik et al., 1999); and in Canada, lindane use reached a peak by 1994 of 558 tonnes, with total use between 1970 and 2000 estimated to have been about 9000 tonnes (Li et al., 2004b). Combining these three use patterns suggests that γ-HCH emissions from these regions have declined steadily but slowly since the 1970's. Two other data sets describe trends in HCH concentrations in ringed seals, from Lancaster Sd. in the central Arctic (Lohmann et al., 2007) and from western Greenland (Rigét et al., 2008). In the Lancaster Sd. ringed seals, α-HCH and β-HCH concentrations were approximately similar to those from Ulukhaktok, but α-HCH concentrations appeared to decline roughly synchronously with global emission patterns. α-HCH in the Lancaster Sd. ecosystem probably arrives by atmospheric transport from the south, water advection via the Arctic Ocean outflow and some more easterly sources via the Nares Strait. We are unable to explain differences between α-HCH

5144

R.F. Addison et al. / Science of the Total Environment 407 (2009) 5139–5146

or the more varied regional sources. β-HCH concentrations in Lancaster Sd. ringed seals appeared to increase from the mid-1970's to about 1990, and then to decline. The Lancaster Sd. ecosystem is likely to be supplied with β-HCH by advection from general circumpolar sources, as opposed to the southern Beaufort region which is probably supplied mainly via the Bering Strait; it is possible that the different β-HCH trends reflect these sources and pathways. However, we note that in contrast to the Lancaster Sd. ringed seal trend, seabird eggs of some species from the region have β-HCH concentrations which have been generally increasing from 1975 until 2003 (Braune, 2007). In western Greenland ringed seals (Rigét et al., 2008), concentrations of α-HCH were well below those from Ulukhaktok, those of β-HCH slightly lower, and those of γ-HCH roughly similar. Concentrations of α-HCH in the west Greenland ringed seals have been declining since 1994, but concentrations of β-HCH and γ-HCH have been fairly constant. The west Greenland ecosystem experiences generally southward air movements, as does the southern Beaufort Sea (Macdonald et al., 2005) but appreciably

Fig. 3. (a) α-HCH concentrations regressed on year of birth in female ringed seals (P. hispida) from Ulukhaktok (Holman), NT. Pre-1983 data are shown as filled points and post-1982 data as empty points. The regression using all data points (ln transformed) shown by the solid regression line had the form: y = ð−0:0104 F 0:0096Þx + 25:5; r = 0:18; n = 38; P = 0:28 where y is α-HCH concentration and x is year of birth. Pre-1983 data (ln transformed, filled points, dashed regression line) had the form: y = ð0:0153 F 0:0164Þx − 25:1; r = 0:20; n = 22; P = 0:36 and post-1982 data (ln transformed, empty points, dotted regression line) had the form: y = ð−0:032 F 0:032Þx + 68:1; r = 0:25; n = 16; P = 0:34 (b) α-HCH concentrations regressed on year of birth in male ringed seals (P. hispida) from Ulukhaktok (Holman) NT, using conventions as in (a). The regression (all data, ln transformed) had the form: y = ð0:0063 F 0:0098Þx − 7:18; r = 0:10; n = 37; P = 0:52; The pre-1983 regression equation had the form: y = ð0:0124 F 0:0206Þx − 19:2; r = 0:12; n = 27; P = 0:55 and the post-1982 regression equation had the form:

Fig. 4. (a) β-HCH concentrations (ln transformed) regressed on year of birth in female ringed seals (P. hispida) from Ulukhaktok (Holman) NT. Regression has the form:

y = ð0:0026 F 0:032Þx + 70:328; r = 0:03; n = 10; P = 0:94:

y = ð0:069 F 0:014Þx − 134; r = 0:66; n = 34; Pb0:001 using the nomenclature in Fig. 3(a). (b) β-HCH concentrations (ln transformed) regressed on year of birth in male ringed seals (P. hispida) from Ulukhaktok (Holman) NT. Regression has the form:

trends in Ululhaktok and Lancaster Sd. ringed seals, other than to suggest that they may reflect either differences in ice cover and therefore fluxes of α-HCH via air–sea exchange (Jantunen et al., 2008)

y = ð0:0514 F 0:0:0143Þx − 99:0; r = 0:52; n = 37; Pb0:01 using the nomenclature in Fig. 3(a).

R.F. Addison et al. / Science of the Total Environment 407 (2009) 5139–5146

Fig. 5. (a) γ-HCH concentrations (ln transformed) regressed on year of birth in female ringed seals (P. hispida) from Ulukhaktok (Holman) NT. Regression has the form: y = ð−0:0123 F 0:0129Þx + 26:0; r = 0:17; n = 34; P = 0:35 (b) γ-HCH concentrations (ln transformed) regressed on year of birth in male ringed seals (P. hispida) from Ulukhaktok (Holman) NT. Regression has the form: y = ð−0:032 F 0:0:014Þx + 64:9; r = 0:35; n = 36; P = 0:03:

less ice cover (cf. Harwood et al., 2000; Hvidegaard et al., n.d.), and this may allow a faster response to atmospheric emission changes, especially in α-HCH. Water advection is predominantly northward via the West Greenland current, and the β-HCH input probably arises more from northern European and eastern Russian sources than is the case in the Ulukhaktok samples. In the discussions above, we have not considered the possible contribution of metabolism (either isomerisation or degradation) to any of the trends we have observed. All three HCH isomers appear to be susceptible to microbial degradation under both aerobic and anaerobic conditions, but the processes are relatively slow and β-HCH is considered to be the most recalcitrant isomer (UNEP, 2007a). Nevertheless, microbial degradation, particularly of α-HCH, appears to be the main process of HCH removal from Arctic Ocean surface waters (Harner et al., 2000). Some metabolism of α-HCH in seals can be inferred from the fact that there may be an excess of the (+) enantiomer in ringed seal blubber (Hoekstra et al., 2003b) though other studies (Wiberg et al., 2000; Moisey et al., 2001; Fisk et al., 2002) showed ringed seal blubber α-HCH to be essentially racemic. α-HCH is converted in rats to β-pentachlorocyclohexene and eventually to trichlorophenols (Koransky et al., 1975; Parzefall et al., 1980); some of

5145

this is apparently mediated via the hepatic cytochrome P-450 system, but some may also occur via abiotic processes. γ-HCH is also reported to be metabolised by rats to unspecified polar metabolites (Chadwick et al., 1971). It would not be surprising to find that some capacity exists in ringed seal for degradation of at least α-HCH and γ-HCH though it is difficult at this stage to estimate its significance. The absence of any appreciable accumulation of HCHs with age in male seals also suggests that there may be some metabolic degradation by the seals, though this may also be explained by the relatively low Kow of most of these compounds (e.g., Kelly et al., 2007). Finally, we note that the absence of any response in seal blubber α-HCH and β-HCH concentrations to emission reductions in the early 1980's is in contrast to the behaviour of the polychlorinated biphenyls (PCBs) and (perhaps) the DDT-group of insecticides. The global restrictions on manufacture and use of the PCBs in the early to mid-1970's (Addison, 1983; De Voogt and Brinkmann, 1989) resulted in a decline in PCB concentrations in the Ulukhaktok ringed seals as early as 1981, i.e., within a decade of emissions being reduced (Addison et al., 1986). However, PCB concentrations in the seals did not change appreciably after that initial decline (Addison and Smith, 1998). Although global DDT-group emissions declined after the early 1970's, DDT-group concentrations in Ulukhaktok seals did not begin to decline until after the 1980's, which probably indicated some continuing use of DDT during the 1970's (Addison et al., 1986; Addison and Smith, 1998). This (relatively) fast response of DDT-group and PCB concentrations in Ulukhaktok ringed seals to global emission trends probably reflects the rapid atmospheric transport of these compounds to the southern Arctic Ocean, and subsequent relatively rapid accumulation in the regional food web to a trophic level at which the seal prey; both processes seem to have occurred within a fairly short period of b10 years, and probably arise from the relatively high value of H for these groups of compounds, their low water solubilities and high Kow. Compared to the DDT-group and the PCBs, the delay in any response of the Ulukhaktok seals to changes in the emissions of α-HCH which took place in the early 1980's probably arises mainly from the less efficient air–sea exchange of α-HCH (driven by its low H, higher water solubility and lower Kow). The continuing increase in β-HCH concentrations in Ulukhaktok seal blubber until the present is consistent with continuously increasing advection into the southern Arctic Ocean from south of the Bering Strait “bottleneck” (Li and Macdonald, 2005). Trends in the accumulation of POPs in the Ulukhaktok ringed seals therefore reflect the contribution of (or competition between) several processes which include at least: (a) delivery to the Arctic by atmospheric transport, oceanic advection, or both, depending on H, Kow and water solubility; (b) atmosphere–surface water exchange, especially for atmospherically-borne compounds with relatively high H and Kow; (c) microbial degradation (significant for HCHs, but probably less so for the DDT-group and PCBs); (d) transfer through lower trophic levels of the seals' food web, which is probably more efficient as H and Kow increase; and (e) “scavenging” by sedimenting particles, again probably more significant for compounds with high H and high Kow (cf. Hargrave et al., 1988). Since these interactions are complex, and may be slow, detection of a response in Ulukhaktok ringed seals to changes in HCH emissions will probably have to await future analyses of seals born 10–20 years after the mid 1990's. Acknowledgements We thank John and Emma Alikamik, and the seal hunters of Ulukhaktok, for sampling/providing access to harvested seals. We thank Jessica Epp and Camilla Teixeira of Environment Canada, Burlington, ON, for sample extraction and GC analysis of HCH isomers, and Xiaowa Wang for project coordination. An anonymous reviewer provided perceptive comments which we hope have improved the manuscript. The Fisheries Joint Management Committee (FJMC)

5146

R.F. Addison et al. / Science of the Total Environment 407 (2009) 5139–5146

provided core project funding in 2001 and 2006. Finally, we thank the Northern Contaminants Programme (Indian and Northern Affairs, Canada) for its continued support of this work. References Addison RF. PCB replacements in dielectric fluids. Environ Sci Technol 1983;17:486A–94A. Addison RF. Organochlorines and marine mammal reproduction. Can J Fish Aquat Sci 1989;46:360–8. Addison RF, Smith TG. Trends in organochlorine residue concentrations in ringed seal (Phoca hispida) from Holman, NWT, 1972–1991. Arctic 1998;51:253–61. Addison RF, Zinck ME, Smith TG. PCBs have declined more than DDT group residues in Arctic ringed seals (Phoca hispida) between 1972 and 1981. Environ Sci Technol 1986;20:253–6. Bidleman TF, Kylin H, Jantunen LM, Helm PA, Macdonald RW. Hexachlorocyclohexanes in the Canadian Archipelago. 1. Spatial distribution and pathways of α-, β-, and γ-HCHs in surface water. Environ Sci Technol 2007;41:2688–95. Braune B. Temporal trends of organochlorines and mercury in seabird eggs from the Canadian Arctic, 1975–2003. Environ Pollut 2007;148:599–613. Braune BM, Donaldson GM, Hobson KA. Contaminant residues in seabird eggs from the Canadian Arctic. Part I. Temporal trends 1975–1998. Environ Pollut 2001;114:39–54. Breivik K, Pacyna JM, Münch J. Use of α-, β-and γ-hexachlorocyclohexane in Europe, 1970–1996. Sci Total Environ 1999;239:151–63. Chadwick RW, Cranmer MF, Peoples AJ. Comparative stimulation of yHCH metabolism by pretreatment of rats with yHCH, DDT and DDT+ yHCH. Toxicol. Appl Pharmacol 1971;18:685–95. De Voogt P, Brinkmann UATh. Production, properties and usage of polychlorinated biphenyls. In: Kimbrough RD, Jensen AA, editors. Halogenated biphenyls, terphenyls naphthalenes dibenzodioxins and related products. 2nd ed. Amsterdam: Elsevier; 1989. p. 3–45. Fisk AT, Holst M, Hobson KA, Duffe J, Moisey J, Norstrom RJ. Persistent organochlorine contaminants and enantiomeric signatures of chiral pollutants in ringed seals (Phoca hispida) collected in the east and west side of the Northwater Polynya, Canadian Arctic. Arch Environ Contam Toxicol 2002;42:118–26. Hargrave BT, Vass WP, Erickson PE, Fowler BR. Atmospheric transport of organochlorines to the Arctic Ocean. Tellus 1988;40B:480–93. Harner T, Kylin H, Bidleman TF, Strachan WMJ. Removal of α-and γ-hexachlorocyclohexane and enantiomers of α-hexachlorocyclohexane in the Eastern Arctic Ocean. Environ Sci Technol 1999;33:1157–64. Harner T, Jantunen LMM, Bidleman TF, Barrie LA. Microbial degradation is a key elimination pathway of hexachlorocyclohexanes from the Arctic Ocean. Geophys Res Lett 2000;27:1155–8. Harwood LA, Smith TG, Melling H. Variation in reproduction and condition of the ringed seal (Phoca hispida) in Prince Albert Sound, N.T., Canada, as assessed through a community-based sampling program. Arctic 2000;53:422–31. Harwood LA, Melling H, Alikamik J. Assessing reproduction and body condition of the ringed seal near Sachs Harbour and Ulukhaktok, NT, Canada, through a harvestbased sampling program, 1992–2007. Presented at: US minerals management service information transfer meeting, Anchorage, Alaska; 2008. October 28. Helm PA, Diamond ML, Semkin R, Bidleman TF. Degradation as a loss mechanism in the fate of α-hexachlorocyclohexane in Arctic watersheds. Environ Sci Technol 2000;34:812–8. Hoekstra PF, O'Hara TM, Fisk AT, Borgå K, Solomon KR, Muir DCG. Trophic transfer of persistent organochlorine contaminants (OCs) within an Arctic marine food web from the southern Beaufort–Chukchi Seas. Environ Pollut 2003a;124:509–22. Hoekstra PF, O'Hara TM, Karlsson H, Solomon KR, Muir DCG. Enantiomer-specific biomagnification of α-hexachlorocyclohexane and selected chiral chlordanerelated compounds within an Arctic food web. Environ Toxicol Chem 2003b;22:2482–91. Hvidegaard, S.M., Forsberg, R., Hanson, S., Toudal, L., (no date). Ice studies off West Greenland 2006. www.geus.dk/ghexis/is/Is-samlet_vers3.ppt (accessed Feb. 2009). Jantunen LM, Bidleman TF. Reversal of air–water gas exchange direction of hexachlorocyclohexanes in the Bering and Chukchi Seas: 1983 versus 1988. Environ Sci Technol 1995;29:1081–9. Jantunen L, Helm PA Kylin H, Bidleman TF. Hexachlorocyclohexanes (HCH) in the Canadian Archipelago. 2. Air–water gas exchange of α-and γ-HCH. Environ Sci Technol 2008;42:465–70.

Kelly BC, Ikonomou MG, Blair JD, Morin AE, Gobas FACP. Food web specific biomagnifications of organic pollutants. Science 2007;317:236–9. Koransky W, Münch G, Noack G, Portig J, Sodomann S, Wirsching M. Biodegradation of alpha-hexachlorocyclohexane. V. Characterization of the major urinary metabolites. Naunyn Schmiedebergs Arch Pharmacol 1975;288:65–78. Li YF, Bidleman TF. Correlation between global emissions of α-hexachlorocyclohexane and its concentrations in the Arctic air. J Environ Informatics 2003;1:52–7. Li YF, Macdonald RW. Sources and pathways of selected organochlorine pesticides to the Arctic and the effect of pathway divergence on HCH trends in biota: a review. Sci Total Environ 2005;342:87–106. Li YF, Macdonald RW, Jantunen LMM, Harner T, Bidleman TF, Strachan WMJ. The transport of β-hexachlorocyclohexane to the western Arctic Ocean: a contrast to α-HCH. Sci Total Environ 2002;291:229–46. Li YF, Macdonald RW, Ma JM, Hung H, Vankatesh S. Historical α-HCH budget in the Arctic Ocean: the Arctic Mass Balance Box Model (AMBBM). Sci Total Environ 2004a;324:115–39. Li YF, Struger J, Waite D, Ma J. Gridded Canadian lindane usage inventories with 1/6° X 1/4° latitude and longitude resolution. Atmos Environ 2004b;38:1117–21. Li YF, Zhulidov AV, Robarts RD, Korotova LG. Hexachlorocyclohexane use in the former Soviet Union. Arch Environ Toxicol 2004c;48:10–5. Lohmann R, Breivik K, Dachs J, Muir D. Global fate of POPs: current and future research directions. Environ Pollut 2007;150:150–65. 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. Moisey J, Fisk AT, Hobson KA, Norstrom RJ. Hexachlorocyclohexane (HCH) isomers and chiral signatures of α-HCH in the Arctic marine food web of the Northwater Polynya. Environ Sci Technol 2001;35:1920–7. Muir DCG, Rigét F, Cleeman M, Skaare J, Kleivane L, Nakata H, et al. Circumpolar trends of PCBs and organochlorine pesticides in the Arctic marine environment inferred from levels in ringed seals. Environ Sci Technol 2000;34:2431–8. Muir DCG, Kwan M, Evans M, Small J, Sverko E, Barresi E, et al. Temporal trends of persistent organic pollutants and metals in ringed seals from the Canadian Arctic. Synopsis of research conducted under the 2006–2007 Northern Contaminants Program, Indian and Northern Affairs Canada, Ottawa; 2007. Parzefall W, Münster J, Schulte-Hermann R. A comparative study on the effects of αhexachlorocyclohexane and its metabolite β-pentachlorocyclohexene on growth and monooxygenase activities in rat liver. Biochem Pharmacol 1980;29:2169–78. Rigét F, Vorkamp K, Dietz R, Muir DCG. Levels and temporal trends of HCH isomers in ringed seals from West and East Greenland. J Environ Monit 2008;10:935–40. Shoeib M, Harner T. Using measured octanol–air partition coefficients to explain environmental partitioning of organochlorine pesticides. Environ Toxicol Chem 2002;21:984–90. Smith TG. Population dynamics of the ringed seal in the Canadian eastern Arctic. Bull Fish Res Board Can 1973;181:viii+55. Tsydenova O, Minh TB, Kajiwara N, Batoev V, Tanabe S. Recent contamination by persistent organochlorines in Baikal seal (Phoca sibirica) from Lake Baikal, Russia. Mar Pollut Bull 2004;48:749–58. UNEP. Lindane risk profile; 2006. UNEP/POPS/POPRC.2/17/Add.4; 25pp. UNEP. Beta hexachlorocyclohexane risk profile; 2007a. UNEP/POPS/POPRC. 3/20/ Add.9; 23pp. UNEP. Alpha hexachlorocyclohexane risk profile; 2007b. UNEP/POPS/POPRC. 3/20/ Add.8; 24pp. Wania F, Mackay D. Global chemical fate of α-hexachlorocyclohexane. 2. Use of a global distribution model for mass balancing, source apportionment and trend prediction. Environ Toxicol Chem 1999;18:1400–7. Wiberg K, Letcher RJ, Sandau CD, Norstrom RJ, Tysklind M, Bidleman TF. The enantioselective bioaccumulation of chiral chlordane and α-HCH contaminants in the polar bear food chain. Environ Sci Technol 2000;34:2668–74. Willett KL, Ulrich EM, Hites RA. Differential toxicity and environmental fates of hexachlorocyclohexane isomers. Environ Sci Technol 1998;32:2197–207.