Impacts of Climate and Feeding Conditions on the Annual Accumulation

ARTICLE pubs.acs.org/est

Impacts of Climate and Feeding Conditions on the Annual Accumulation (1986
2009) of Persistent Organic Pollutants in a Terrestrial Raptor Jan O. Bustnes,*,† Nigel G. Yoccoz,†,‡ Georg Bangjord,§ Dorte Herzke,|| Lutz Ahrens,^ and Janneche U. Skaare# †

)

Norwegian Institute for Nature Research, FRAM—High North Research Centre on Climate and the Environment, NO-9296 Tromsø, Norway ‡ Department of Arctic and Marine Biology, University of Tromsø, NO-9037 Tromsø, Norway § Oddatunet, NO-7057 Jonsvatnet, Norway Norwegian Institute for Air Research, FRAM—High North Research Centre on Climate and the Environment, NO-9296 Tromsø, Norway ^ Environment Canada, Science and Technology Branch, Toronto, ON, M3H 5T4, Canada # Norwegian School of Veterinary Science (NSVS), P.O. Box 8146 Dep., N-0033 Oslo, Norway ABSTRACT: The relationships between climate variability, feeding conditions, and the annual accumulation of polychlorinated biphenyls (PCB), 1,1-dichloro-2, 2-bis (p-chlorophenyl) ethylene (p,p0 -DDE) and polybrominated diphenyl ethers (PBDE) in eggs (n = 168) of tawny owls (Strix aluco) were assessed over a 24-year period (1986
2009) in Central Norway. Winter climate variables included the North Atlantic Oscillation (NAO) and snow conditions, whereas feeding conditions were assessed using vole abundance. The concentrations of all compounds declined between 79% (p,p0 -DDE) and 86% (PBDE) over the time period. For PCB and p,p0 -DDE, the accumulation was positively associated with snow depth, and negatively associated with NAO (i.e., high accumulation in cold and dry winters) when vole abundance was low, suggesting trade-offs between winter severity and feeding conditions. That is, females mobilize more body fat reserves in cold winters when feeding conditions are poor, which results in increased circulating contaminant concentrations and maternal transfer to the eggs. Owls may also have been forced to feed on prey with higher contaminant loads due to restricted prey availability. For the recently banned PBDEs, the accumulation was high when NAO was low, while snow depth was positively associated with PBDE accumulation only when feeding conditions were good. This suggests somewhat different dynamics of PBDE in the environment or in the owls, compared to PCB and p,p0 -DDE. However, climate and feeding conditions explained as much of the annual variation in concentrations of all POPs, as the overall decreasing trend over the 24 years. Hence, such factors should be considered in monitoring programs for POPs. Moreover, to better understand the mechanisms of climate effects on POP accumulation, future studies should measure pollutants in different components of the food chain.

’ INTRODUCTION Cessation of production and use of many chlorinated persistent organic pollutants (POPs) resulted in declining environmental concentrations on a global scale.1
3 However, the changes in the accumulation of such contaminants in biota is not determined solely by the time elapsed since they were banned, and although they have declined there are still great spatial and temporal variations in concentrations.2
6 For example, in Great Lake herring gulls (Larus argentatus), both climatic factors (winter severity) and diet variation were related to temporal fluctuations in the levels of polychlorinated biphenyls (PCB) in eggs.4,7 Environmental and biological factors thus complicate the monitoring of persistent organic pollutants; i.e., pollution trends may go undetected or false trends could be established due to environmental changes.4,8 Moreover, other r 2011 American Chemical Society

anthropogenic influences, for example climate change, are affecting the biodiversity and food web structures of ecosystems.9 This might halt declining trends of contaminants, for example in top predators, due to changes in the lower food chain.10,11 This study focuses on the tawny owl (Strix aluco) in Central Norway, a common terrestrial raptor which feeds mainly on voles. However, in years with low vole abundance, it may switch to alternative prey such as passerine birds,12
14 which may increase their exposure to contaminants.15 A recent analysis from our study population showed that the concentrations of various Received: May 26, 2011 Accepted: July 26, 2011 Revised: July 25, 2011 Published: August 15, 2011 7542

dx.doi.org/10.1021/es201786x | Environ. Sci. Technol. 2011, 45, 7542–7547

Environmental Science & Technology organochlorines and brominated flame retardants (polybrominated diphenyl ethers [PBDEs]) in tawny owl eggs declined (60
95%) between 1986 and 2004.16 However, interannual variations in POP concentrations were great, and the aim of this work was to unveil how variability in climate and food resources was related to the annual accumulation of POPs in the period between 1986 and 2009. To cover the variation among different POPs and to reduce the number of statistical analyses, we selected compounds based on the variation observed in the previous analyses;16 i.e., PCB (the sum of 7 polychlorinated biphenyl congeners), and p, p0 -DDE (1,1-dichloro-2,2-bis (p-chlorophenyl) ethylene) as representatives for legacy persistent organic pollutants (POPs), and the sum of 5 BDE (polybrominated diphenyl ether congeners [PBDE]) for brominated flame retardants. Egg formation occurs in winter and we used two measurements of winter climate variability: the North Atlantic Oscillation index (NAO: an index of air flow and low pressure activity), and snow depth. For owls, much snow impairs vole hunting 17
19 which results in poor feeding conditions (i.e., low vole availability) and may also force birds to feed on alternative prey. Moreover, cold weather prior to egg laying will lead to higher energy expenditure. This may force owls to spend more stored body lipid reserves to produce eggs, and POPs embedded in these lipids will be sequestered into the eggs.20
22 However, such relationships might differ with variation in feeding conditions. That is, when abundance of voles is high, the probability of energy shortfalls due to adverse climatic conditions is likely to be reduced.

’ MATERIALS AND METHODS Study Area. The work was carried out in the area surrounding the city of Trondheim (63.42 N, 10.23E) in Sør-Trøndelag County, Central Norway.16 Tawny owl eggs were collected between 1986 and 2009. Sample Collection. More than 100 tawny owl nest boxes have been deployed, and annually each nest box was visited twice; the first visit in early April, shortly after egg laying. The nest boxes were revisited in the first half of May and all nonhatching eggs were collected. Eggs were frozen shortly after collection. In this analysis, 168 eggs were selected for contaminants analysis. The number of eggs sampled per year varied between one (1990) and 18 (2004), except for 1988 when no eggs were found. For details of egg collections see Bustnes et al.16 Climate Variables. Climate variation may be measured in a multitude of ways (temperature, wind, precipitation, etc.). These variables can further be divided, e.g., by seasons, and it is therefore necessary to make a priori choices concerning the variables to include in analyses.23 Furthermore, it has been demonstrated that large scale climate indexes such as the NAO (North Atlantic Oscilliation) often better predict variation in ecological processes than local climate variables such as temperatures.24,25 The NAO can be indexed by the sea-level pressure variability between the Azores and Iceland and NAO summarizes, on a large scale, a number of climate variables in Central Norway including temperature and precipitation.25 Data on the winter NAO index (December-March) for all relevant years were obtained from the Web site of Jim Hurrell of the National Centre for Atmospheric Research (http://www.cgd.ucar.edu/cas/jhurrell/ indices.html). To assess the extent to which NAO represented the local climate, we estimated the relationship between winter NAO and mean winter temperature and precipitation in our study area. We used all weather stations located in the area and for which

ARTICLE

data were available from the Norwegian Meteorological Institute at the eklima Web site (http://sharki.oslo.dnmi.no/portal/page?_ pageid=73,39035,73_39049&_dad=portal&_schema=PORTAL). For temperature, only one station at Trondheim airport (Værnes) had measurements for the whole period, whereas 5 stations could be used for snow depth and precipitation. As temporal variation in temperature is much more synchronized spatially than variations in precipitation or snow depth,26 using more stations for precipitation and snow will not affect the comparison. For temperature and precipitation, we defined winter as the sum over the three months January to March, whereas for snow depth, we used the maximum value. As expected temperature and precipitation were strongly and positively related to NAO (r = 0.66 and 0.51 respectively, n = 25, 1985
2009), whereas maximum snow depth was not related to NAO (r = 0.05). Maximum snow depth was not directly related to temperature or precipitation either (r < 0.05), because snowy winters in this area are a result of unusual combination of wet and cold winters. Hence, NAO was a good index of local climate except for snow. Feeding Conditions. To assess the vole availability in different years, we calculated a vole index by using the number of dead voles (mostly field vole [Microtus agrestis]: G. Bangjord unpublished data) found in all nest boxes where females were caught, including boxes with no nonhatching eggs. The index was the mean number of voles found in each nest box each year. Voles are by far the most important prey of the tawny owls in our study area; i.e., of 8515 determined prey items from 100 nest boxes, ∼80% were voles (G. Bangjord and J. Obuch, unpublished data). This might be even higher during egg formation in winter because the most important migrating birds in the owl diets have not arrived yet, e.g., fieldfare (Turdus pilaris). Moreover, there is strong evidence that the vole index predicts reproductive performance of the owls.27 The sample of nest-boxes varied among years between 8 (2001) and 72 (2007); a total of 850 (mean per year = 37) with 1164 voles (mean per year = 50.6). Chemical Analysis. All information regarding the chemical analyses of the eggs collected between 1986 and 2004 can be found in Bustnes et al.16 For this period, the analyses were carried out by the National Veterinary Institute (NVI) in Oslo. For the period 2005
2009, the Norwegian Institute for Air Research (NILU) in Tromsø conducted the analyses, for which methods are described in Herzke et al.28 In these analyses, NIST Standard Reference Material 1588b was used. Recovery was between 77% and 84%. To ensure that the analyses from the two laboratories were comparable, the NILU reanalyzed pooled samples from the years 1987, 1991, 1993, 1994, 1997, 1999, 2001, 2003, and 2004 and compared the results with the mean concentrations reported in the earlier study.16 When comparing the mean concentrations from NVI for these years with the mean concentrations from the NILU pools, the deviations were only 16% (R2 = 0.69), 22.5% (R2 = 0.70) and 16% (R2 = 0.65), for the PCB, p,p0 -DDE and PBDE, respectively. This suggests a good accordance between the two laboratories. The following PCB-congeners were determined from both laboratories; CB-101,
99,
118,
153,
138,
187,
180, while the following PBDE congeners were analyzed: BDE-47,
99,
100,
153,
154. Because of the similar trends for the different PCB and PBDE congeners and strong correlation between the congeners,16 we used their sum in the analysis. In addition, p,p0 -DDE was used in this analysis because the trends differed somewhat from other legacy POPs. Lipid content was 7543

dx.doi.org/10.1021/es201786x |Environ. Sci. Technol. 2011, 45, 7542–7547

Environmental Science & Technology

ARTICLE

Table 1. Concentrations (μg/g, lipid weight) of Different Persistent Organic Pollutants in Tawny Owl Eggs from Central Norway Collected between 1986 and 2009a years 1986
2009

1986
1995

2001
2009

N

mean

median

min

max

PCB

168

2.32

1.32

0.09

38.88

p,p0 - DDE

168

2.51

1.47

0.06

43.29

PBDE

168

0.16

0.08

ND

5.27

PCB

60

4.07

1.88

0.38

38.88

p,p0 - DDE PBDE

60 60

3.63 0.25

2.05 0.09

0.36 0.01

43.29 5.27

PCB

71

1.09

0.81

0.09

4.19

p,p0 - DDE

71

1.58

1.16

0.65

10.52

PBDE

71

0.08

0.06

ND

0.54

Concentrations are given for the whole period and for the first and last 10 years of the period.

a

determined gravimetrically, and lipid normalized concentrations were used in the analyses Statistical Analyses. The data were analyzed using linear and additive models in R.29,30 The goodness of fit of linear models was assessed using partial residual plots and influence values,31 and additive models indicated that simple linear models gave an adequate fit compared to more complicated nonlinear models. Compound concentrations were log-transformed to achieve constancy of residual variance. Note that such models predict the expected log-transformed concentrations, which differ substantially from the log of the expected concentrations because the variance of concentrations was large. We used models on logtransformed concentrations, however, because the residual distribution was symmetric on the log scale, and therefore the mean was close to the median and reflected the overall changes much better than the expected values on the original scale. To evaluate different hypotheses, we tested four models for each of the compounds containing the following predictor variables: winter NAO and snow depth in the current year, and the temporal component (year). In addition, the interactions between NAO and vole abundance (log[vole]), and snow and vole abundance (log[vole]) were examined. We chose the most parsimonious statistical models based on the Akaike’s Information Criterion corrected for small sample size (AICc). More specifically, we calculated the relative likelihood of each model using AICc weights derived from difference in AICc values between the best model (lowest AICc) and other models.23

’ RESULTS AND DISCUSSION Summaries of the concentrations of the different compounds can be found in Table 1. There were negative temporal relationship over the 24 year period, and PCB, p,p0 -DDE and PBDE declined by 83%, 79%, and 86%, respectively, after including climate and feeding conditions in the statistical models (Figure 1, Table 2; these values were calculated using the predicted values for the first and last year of the study, and additive models showed similar changes [results not shown]). Hence, the declines seem to have continued after 2004, up to which the results were published in Bustnes et al.16 The best models for accumulation of PCB and p,p0 -DDE in tawny owl eggs included positive effects of snow, and the interactions between NAO and vole abundance (Table 2, Figure 1). That is, in cold and dry winters (low NAO index)

with low abundance of voles, the concentrations of PCB and p,p0 DDE were high, whereas in years with high vole abundance there were no such relationships. To be able to survive unpredictable winter conditions at high latitudes, tawny owls accumulate large body fat reserves in autumn.20 These fat reserves contain lipidsoluble POPs that are released to the blood when they are mobilized,32
34 due to poor feeding conditions. Hence, the positive relationship between legacy POPs and snow depth is probably caused by increased lipid mobilization, since snow impairs vole hunting.17
19 In addition, cold winters results in higher probability of energy shortfalls and more mobilization of body fat reserves in tawny owls,14,20,35,36 which also increases the circulating POP concentrations. Hence, more pollutants will be sequestered into the eggs when females are in poor condition since eggs normally contain concentrations of POPs proportional to female blood concentrations.21,33 However, the lack of the association between NAO and the concentrations of PCB and p,p0 -DDE when vole availability was high, suggests that the effects of cold winters was canceled out when the females had a stable food source prior to egg laying. In addition, in good vole years, clutches were larger,27 suggesting that pollutant loads were spread in more eggs, and each egg may have lower concentrations than in small clutches.37 In essence, PCB and p,p0 -DDE behaved similarly, but the relationship with time was weaker for p,p0 -DDE which seems to be caused by a reduced decline in p,p0 -DDE concentrations after the early 1990s, while PCB continued to decrease over the whole period.16 Second, the interaction between NAO and the vole abundance was stronger for p,p0 -DDE than for PCB; i.e., when voles were low the NAO effect was very strong (Figure 1, Table 2). Unlike PCBs, DDT is still being used in tropical regions and may reach Norway via the atmosphere, which possibly explains the differences between the compounds. PBDEs were discovered in the environment as late as 1981,38 and penta- and octa-BDE products were not banned in the EU and Norway until 2004.39 In this work, they behaved somewhat differently from the legacy POPs, which have been banned in Europe for 30
40 years. The best model for PBDEs included a negative NAO effect (lower concentrations in mild years), and the interaction between snow and vole abundance; i.e., when both vole abundance and snow increased, so did PBDE concentrations (Table 2, Figure 1). The negative NAO effect may be expected from the relationships between cold winters and mobilization of body fat reserves. However, it was unexpected that PBDEs only were positively related to snow depth when vole abundance was high. Norway is a recipient of long-transported legacy POPs and PBDE,40 but the transport potential of different BDE congeners varies greatly. For example, Wania and Dugani 41 found that low- and intermediate-brominated BDE congeners (including penta-BDE) have a long-range transport potential comparable to highly chlorinated PCBs, whereas higher brominated congeners were not subject to significant long-range transport. However, highly brominated PBDEs can be transported on the surface of airborne particles, and it has been reported that due to their high porosity, falling snowflakes are effective at scavenging atmospheric particles.42,43 Hence, snow will have PBDE profiles enriched in the more brominated congeners (BDE 99 and 100), which are mainly associated to atmospheric particulate matter.44 In this work, nearly 80% of the loads of PBDEs were made up by the congeners
99,
100,
153, and
154,16 which has also been found in other raptor species in Scandinavia.45,46 However, the cause of the association between snow and PBDE during high vole abundance is not known. It 7544

dx.doi.org/10.1021/es201786x |Environ. Sci. Technol. 2011, 45, 7542–7547

Environmental Science & Technology

ARTICLE

Figure 1. Relationships between the concentrations (μg/g, lipid weight) of PCB, p,p0 -DDE and PBDE in eggs of tawny owls and the following predictors: NAO (North Atlantic Oscillation), vole abundance, snow depth and year. The plots are partial residuals based on the most parsimonious models in Table 2; i.e., the effects of all predictors have been controlled for each others. Annual vole abundance is a continuous variable and increasing color intensity denotes increasing vole abundance. Lines show predicted effects of NAO (PCB, p,p0 -DDE) or snow (PBDE) for high vole abundance (mean +2 SD; dotted line) and low vole abundance (mean
2 SD; solid line), respectively.

may be possible that more of the PBDEs in this study originate from local sources than do the legacy PCBs, since PBDEs have been in active use for most of the study period, also in Norway.47 This long-term work provides evidence that climate and feeding conditions strongly influence the accumulation of lipidsoluble POPs in eggs of tawny owls. As the yearly trend, climate, and feeding conditions could explain only variation between

years, and not within years, we used average yearly concentrations to explore the relative importance of those two variables. The year effect alone explained between 23 (PBDE) and 52% (PCB), while the climate and feeding variables combined explained between 26 (p,p0 -DDE) and 48% (PCB) of the annual variation. Year and climate/feeding variables explained different components of the variation as they were nearly uncorrelated, 7545

dx.doi.org/10.1021/es201786x |Environ. Sci. Technol. 2011, 45, 7542–7547

Environmental Science & Technology

ARTICLE

Table 2. Selection among Different Models Explaining the Annual Accumulation of Different Persistent Organic Pollutants in Tawny Owl Eggs from Central Norway between 1986 and 2009 compound models PCB

rank

variable estimates 2

K

R

adj

AICc

ΔAICc

(s.e.) for rank 1 model

AICc wt.

1

snow + NAO  log(voles) + year

7

0.320

382.81

0.00

0.47

snow

2

snow + NAO + log(voles) + year

6

0.310

383.97

1.16

0.26

NAO


0.13 (0.039)

3 4

(snow + NAO) x log(voles) + year snow  log(voles) + NAO + year

8 7

0.316 0.307

384.93 385.89

2.12 3.07

0.16 0.10

log(voles) NAO  log(voles)


0.096 (0.032) 0.060 (0.034)

year


0.078 (0.012)

0.014 (0.0033)

0

p,p -DDE 1

snow + NAO  log(voles) + year

7

0.230

415.02

0.00

0.59

snow

2

(snow + NAO)  log(voles) + year

8

0.225

417.22

2.20

0.20

NAO


0.15 (0.042)

3

snow + NAO + log(voles) + year

6

0.212

417.67

2.65

0.16

log(voles)


0.069 (0.10)

4

snow  log(voles) + NAO + year

7

0.208

419.72

4.70

0.06

NAO  log(voles) year

0.014 (0.0037)

0.080 (0.037)
0.067 (0.014)

PBDE 1

snow  log(voles) + NAO + year

7

0.198

482.95

0.00

0.66

snow

2

(snow + NAO)  log(voles) + year

8

0.193

485.14

2.19

0.22

NAO


0.19 (0.052)

3

snow + NAO + log(voles) + year

6

0.173

486.97

4.01

0.09

log(voles)


0.59 (0.27)

4

snow + NAO  log(voles) + year

7

0.168

489.11

6.16

0.03

snow  log(voles) year

and between 58% (PBDE) and 81% (PCB) of the yearly variation was explained by these two sets of variables. That is, climate and feeding conditions explained as much of the annual variation in concentrations of all POPs, as the overall decreasing trend over the 24 years. This work thus adds to the growing body of evidence that accumulation of POPs in biota is a complicated process influenced both by factors in the physical environment such as climate and feeding conditions,7,8,48 highlighting the importance that studies of temporal trends take into account such factors. In addition, other physical and biological factors may also add variation in our study species, and should be taken into consideration. For example, the feeding conditions of tawny owls may be difficult when winds are strong, since they depend on hearing for catching voles; i.e., during strong winds the noise may impair the hearing (G. Bangjord, Pers. obs.). Furthermore, we collected eggs remaining in the nest after the adults have left the nest. This means the sample may contain both unfertilized eggs, addled eggs, and eggs in various stages of incubation. The state of embryo development does add variation to POP concentrations measured in eggs.49 In addition, features of the egg laying female, such as possible diet specialization 50 may be important. To reduce this sampling noise, one may collect eggs at the same incubation stage; e.g., freshly laid eggs. However, in rare and threatened species such as raptors, this may not be feasible. Finally, to improve the understanding of how different abiotic and biotic environmental factors influence the accumulation of POPs in terrestrial raptors it will be a great advantage to measure POPs in different compartments of the food chain.

’ AUTHOR INFORMATION Corresponding Author

*Phone: +47 77 75 04 07; fax: +47 77 75 04 01; e-mail: Jan.O. [email protected].

0.019 (0.0046)

0.010 (0.0041)
0.085 (0.017)

’ ACKNOWLEDGMENT We are grateful to Kine Bæk and Anuschka Polder for carrying out the laboratory analyses at NVI. We thank three anonymous reviewers for comments that greatly improved the manuscript. This work was financed by the Norwegian Research Council (Project no. 159435/S30). ’ REFERENCES (1) Loganathan, B. G.; Kannan, K. Global organochlorine contamination trends: An overview. Ambio 1994, 23, 187–191. (2) Hebert, C. E.; Norstrom, R. J.; Weseloh, D. V. C. A quarter century of environmental surveillance: the Canadian Wildlife Service’s Great Lakes herring gull monitoring program. Environ. Rev. 1999, 7, 147–166. (3) Bignert, A.; Olsson, M.; Persson, W.; Jensen, S.; Zakrisson, S.; Litz
en, K.; Erikson, U.; H€aggberg, L.; Alsberg, T. Temporal trends of organochlorines in northern Europe, 1967
1995. Relation to global fractionation, leakage from sediments and international measures. Environ. Pollut. 1998, 99, 177–198. (4) Hebert, C. E.; Hobson, K. A.; Shutt, J. L. Changes in food web structure affect rate of PCB decline in herring gull (Larus argentatus) eggs. Environ. Sci. Technol. 2000, 34, 1609–1614. (5) Newton, I.; Wyllie, I.; Asher, A. Long-term trends in organochlorine and mercury residues in some predatory birds in Britain. Environ. Pollut. 1993, 79, 143–151. (6) Wegner, P; Kleinstauber, G; Baum, F; Schilling, F. Long-term investigation of the degree of exposure of German peregrine falcons (Falco peregrinus) to damaging chemicals from the environment. J. Ornithol. 2005, 146, 34–54. (7) Hebert, C. E. Winter severity affects migration and contaminant accumulation in northern Great Lakes herring gulls. Ecol. Appl. 1998, 8, 669–679. (8) Hebert, C. E.; Weseloh, D. V. C.; Idrissi, A.; Arts, M. T.; O’Gorman, R.; Gorman, O. T.; Locke, B.; Madenjian, C. P.; Roseman, 7546

dx.doi.org/10.1021/es201786x |Environ. Sci. Technol. 2011, 45, 7542–7547

Environmental Science & Technology E. F. Restoring piscivorous fish populations in the Laurentian Great Lakes causes seabird dietary change. Ecology 2008, 89, 891–897. (9) Hooper, D. U.; Chapin, F. S.; Ewel, J. J.; Hector, A.; Inchausti, P.; Lavorel, S.; Lawton, J. H.; Lodge, D. M.; Loreau, M.; Naeem, S.; Schmid, B.; Setala, H.; Symstad, A. J.; Vandermeer, J.; Wardle, D. A. Effects of biodiversity on ecosystem functioning: A consensus of current knowledge. Ecol. Monogr. 2005, 75, 3–35. (10) Hebert, C. E.; Weseloh, D. V. C. Adjusting for temporal change in trophic position results in reduced rates of contaminant decline. Environ. Sci. Technol. 2006, 40, 5624–5628. (11) Noyes, P. D.; McElwee, M. K.; Miller, H. D.; Clark, B. W.; Van Tiem, L. A.; Walcott, K. C.; Erwin, K. N.; Levin, E. D. The toxicology of climate change: Environmental contaminants in a warming world. Environ. Int. 2009, 35, 971–986. (12) Petty, S. J. Diet of tawny owls (Strix aluco) in relation to field vole (Microtus agrestis) abundance in a conifer forest in northern England. J. Zool. 1999, 248, 451–465. (13) Petty, S. J.; Lambin, X.; Sherratt, T. N.; Thomas, C. J.; Mackinnon, J. L.; Coles, C. F.; Davison, M.; Little, B. Spatial synchrony in field vole Microtus agrestis abundance in a coniferous forest in northern England: The role of vole-eating raptors. J. Appl. Ecol. 2000, 37, 136–147. (14) Solonen, T.; Karhunen, J. Effects of variable feeding conditions on the tawny owl Strix aluco near the northern limit of its range. Ornis Fenn. 2002, 79, 121–131. (15) Blus, L. J. Effects of pesticides on owls in North America. J. Raptor Res. 1996, 30, 198–206. (16) Bustnes, J. O.; Yoccoz, N. G.; Bangjord, G.; Polder, A.; Skaare, J. U. Temporal trends (1986
2004) of organochlorines and brominated flame retardants in tawny owl eggs from northern Europe. Environ. Sci. Technol. 2007, 41, 8491–8497. (17) Sonerud, G. A Effect of snow cover on seasonal changes in diet, habitat, and regional distribution of raptors that prey on small mammals in boreal zones of Fennoscandia. Holarctic Ecol. 1986, 9, 33–47. (18) Je-drzejewski, W.; Je-drzejewska, B.; Zub, K.; Ruprecht, A. L.; Bystrowski, C. Resource use by tawny owls Strix aluco in relation to rodent fluctuations in Biazowie_za National Park, Poland. J. Avian Biol. 1994, 25, 308–318. (19) Halonen, M.; Mappes, T.; Meri, T.; Suhonen, J. Influence of snow cover on food hoarding in Pygmy Owls Glaucidium passerinum. Ornis Fenn. 2007, 84, 105–111. (20) Overskaug, K.; Sunde, P.; Kristiansen, E. Subcutaneous fat accumulation in Norwegian owls and raptors. Ornis Fenn. 1997, 74, 29–37. (21) Drouillard, K. G.; Norstrom, R. J. Quantifying maternal and dietary sources of 2,2 0 ,4,4 0 ,5,5 0 -hexachlorobiphenyl deposited in eggs of the ring dove (Streptopelia risoria). Environ. Toxicol. Chem. 2001, 20, 561–567. (22) Van den Burg, A. B. Limitation of owl reproduction in the wild: Is there a role of food quality besides quantity? Ardea 2009, 97, 609–614. (23) Burnham, K. P.; Anderson, D. R. Model Selection and Multimodel Inference. A Practical Information-Theoretic Approach; SpringerVerlag: New York; 2002. (24) Stenseth, N. C.; Mysterud, A.; Ottersen, G.; Hurrell, J. W.; Chan, K.-S.; Lima., M. Ecological effects of climate fluctuations. Science 2002, 297, 1292–1296. (25) Hurrell, J. W.; Kushnir, Y.; Ottersen, G.; Visbeck, M. The North Atlantic Oscillation: Climate significance and environmental impact. Geophys. Monogr. Ser. 2003, 134, 279. (26) Daly, C.; Halbleib, M.; Smith, J. I.; Gibson, W. P.; Doggett, M. K.; Taylor, G. H.; Curtis, J.; Pasteris, P. P. Physiographically sensitive mapping of climatological temperature and precipitation across the conterminous United States. Int. J. Climatol. 2008, 28, 2031–2064. (27) Yoccoz, N. G.; Bustnes, J. O.; Bangjord, G.; Skaare, J. U. Reproduction and survival of tawny owls in relation to persistent organic pollutants. Environ. Int. 2009, 35, 107–112. (28) Herzke, D.; Nygard, T.; Berger, U.; Huber, S.; Rov, N. Perfluorinated and other persistent halogenated organic compounds in European shag (Phalacrocorax aristotelis) and common eider (Somateria

ARTICLE

mollissima) from Norway: A suburban to remote pollutant gradient. Sci. Total Environ. 2009, 408, 340–348. (29) R Development Core Team. R: A language and environment for statistical computing. In. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3
900051
07
0, URL http://www.R-project. org., 2011. (30) Wood, S. N. Generalized Additive Models: An Introduction with R. Taylor & Francis, CRC Press, London, 2006. (31) Venables, W. N.; Ripley, B. D. Modern Applied Statistics with S; Berlin: Springer-Verlag; New York, 2002. (32) Henriksen, E. O.; Gabrielsen, G. W.; Skaare, J. U. Validation of the use of blood samples to assess tissue concentrations of organochlorines in glaucous gull Larus hyperboreus. Chemosphere 1998, 37, 2627–2643. (33) Drouillard, K. G. Modeling the toxicokinetics and biomagnificantion of polychlorinated bihenyls (PCBs) in birds. Ph.D. Thesis, Trent University: Peterborough, ON, Canada, 2001. (34) Bustnes, J. O.; Moe, B.; Herzke, D.; Hanssen, S. A.; Nordstad,  T.; Sagerup, K.; Gabrielsen, G. W.; Borga, K. Strongly increasing blood concentrations of lipid-soluble organchlorines in high arctic common eiders during incubation fast. Chemosphere 2010, 79, 320–325. (35) Solonen, T. Are vole-eating owls affected by mild winters in southern Finland? Ornis Fenn. 2004, 81, 65–74. (36) Solheim, R. Age, sex and size of dead tawny owls Strix aluco found during winter famine 2006. Ardea 2009, 97, 597–601. (37) Van den Steen, E.; Eens, M.; Jaspers, V. L.B.; Covaci, A; Pinxten, R. Effects of laying order and experimentally increased egg production on organic pollutants in eggs of a terrestrial songbird species, the great tit (Parus major). Sci. Total Environ. 2009, 407, 4764–4770. € Blomkvist, G. Polybrominated aromatic pollu(38) Andersson, O.; tants found in fish in Sweden. Chemosphere 1981, 10, 1051–1060. (39) De Wit, C. A.; Alaee, M.; Muir, D. C. G. 2006. Levels and trends of brominated flame retardants in the Arctic. Chemosphere 2006, 64, 209
233. (40) Schuster, J. K.; Gioia, R.; Brevik, K.; Steinnes, E.; Scheringer, M.; Jones, K. Trends in European air reflects reduction in primary emissions of PCBs and PBDEs. Environ. Sci. Technol. 2010, 44, 6760–6766. (41) Wania, F.; Dugani, C. B. Assessing the long-range transport potential of polybrominated diphenyl ethers: a comparison of four multimedia models. Environ. Toxicol. Chem. 2003, 22, 1252–1261. (42) Franz, T. P.; Eisenreich, S. J. Snow scavenging of polychlorinated biphenyls and polycyclic aromatic hydrocarbons in Minnesota. Environ. Sci. Technol. 1998, 32, 1771–1778. (43) Quiroz, R.; Arellano, L.; Grimalt, J. O.; Fernandez, P. Analysis of polybrominated diphenyl ethers in atmospheric deposition and snow samples by solid-phase disk extraction. J. Chromatogr. A 2008, 1192, 147–151. (44) Hoh, E.; Hites, R. A. Brominated flame retardants in the atmosphere of the east-central United States. Environ. Sci. Technol. 2005, 39, 7794–7802. (45) Lindberg, P.; Sellstr€om, U.; H€aggberg, L.; de Wit, C.A.. Higher brominated diphenyl ethers and hexabromocyclododecane found in eggs of peregrine falcons (Falco peregrinus) breeding in Sweden. Environ. Sci. Technol. 2004, 38, 93–96. (46) Herzke, D.; Berger, U.; Kallenborn, R.; Nygard, T.; Vetter, W. Brominated flame retardants and other organobromines in Norwegian predatory bird eggs. Chemosphere 2005, 61, 441–449. (47) Miljøstatus Norge (State of the Environment Norway). Available at: http://www.miljostatus.no/ (48) Bustnes, J. O.; Gabrielsen, G. W.; Verreault, J. Climate variability and temporal trends of persistent organic pollutants in the Arctic: A study of glaucous gulls. Environ. Sci. Technol. 2010, 44, 3155–3161. (49) Drouillard, K. G.; Nostrom, R. J.; Fox, G. A.; Gilman, A.; Peakall, D. B. Development and validation of a herring gull emryo toxicokinetic model for PCBs. Ecotoxicology 2003, 12, 55–68. (50) Bolnick, D. I.; Svanback, R.; Fordyce, J. A.; Yang, L. H.; Davis, J. M.; Hulsey, C. D.; Forister, M. L. The ecology of individuals: Incidence and implications of individual specialization. Am. Nat. 2003, 161, 1–28.

7547

dx.doi.org/10.1021/es201786x |Environ. Sci. Technol. 2011, 45, 7542–7547