Feathers as a means of monitoring mercury in seabirds: Insights from stable isotope analysis

ENVIRONMENTAL F’OLLUTION Environmental Pollution 101 (1998) 193-200

Feathers as a means of monitoring mercury in seabirds: insights from stable isotope analysis D.R. Thompsona,*, S. Bearhopa, J.R. Speakmanb, R.W. Furness” “Applied Ornithology Unit, Graham Kerr Building, University of Glasgow, Glasgow GI2 8QQ, UK bDepartment of Zoology, University of Aberdeen, Aberdeen AB9 ZTN, UK

Received 4 December 1997; accepted 31 March 1998

Abstract

Mercury concentrations, together with nitrogen and carbon stable isotope signatures, were determined in body feather samples from northern fulmars Fulmarus glacialis and great skuas Catharacta skua, and in different flight feathers from great skuas. There were no significant relationships between trophic status, as defined using isotope analysis, and mercury concentration in the same feather type, in either species. Mercury concentrations in body feather samples were markedly different between fulmars and skuas, reflecting differences in diet, but there was no corresponding difference in trophic status as measured through nitrogen stable isotope signatures. We conclude that mercury concentrations and stable isotope values in feathers are uncoupled, mercury concentrations apparently reflecting the body pool of accumulated mercury at the time of feather growth whilst stable isotope values

reflect the diet at the time of feather growth. There were significant positive correlations between the different flight feathers of great skuas for all three parameters measured. These were strongest between primary 10 and secondary 8, suggesting that these two feathers are replaced at the same time in the moult sequence in great skuas. Stable isotope analysis of different feathers may provide a means of investigating moult patterns in birds. 0 1998 Elsevier Science Ltd. All rights reserved. Keywords:

Mercury monitoring; Stable isotopes; Seabirds; Feathers; Heavy metal pollution

1. Introduction There is a substantial body of published work using seabirds as biomonitors of mercury in a wide range of environments and locations (see Walsh, 1990; Burger, 1993; Monteiro and Fumess, 1995 for recent reviews). Mercury is excreted via growing feathers, and once bound to the feather keratin molecule is relatively physically and chemically stable, being resistant to a variety of rigorous treatments (Appelquist et al., 1984). Thus, feathers afford a non-destructive and relatively noninvasive way in which to monitor mercury contamination in seabirds. Those feathers moulted and replaced towards the beginning of the moult process (usually the large flight feathers in seabirds) exhibit the highest

* Corresponding author. Present Address: National Institute of Water and Atmospheric Research (NIWA), 301 Evans Bay Parade, Greta Point, PO Box 14-901, Kilbimie, Wellington, New Zealand. Tel.: 0064 4 386 0300; fax: 0064 4386 0574; e-mail: niwa.cri.nz

d.thompson@

0269-7491/98/%19.00 0 1998 Elsevier Science Ltd. All rights reserved. PII: SO269-7491(98)00078-5

mercury concentrations, with feathers moulted and replaced later in the moult sequence exhibiting correspondingly lower mercury concentrations (Fumess et al., 1986; Honda et al., 1986; Braune and Gaskin, 1987). These authors inferred that this general pattern arises because during the inter-moult period, mercury ingested in the diet is accumulated within internal tissues. At the onset of moult, this ‘body pool’ of mercury diminishes as mercury is eliminated into growing feathers. However, this hypothesis remains to be tested. The use of feathers as a means of investigating mercury in seabirds has several advantages. Mercury concentrations in feathers have been shown to correlate with those in internal tissues (Furness and Hutton, 1979; Thompson et al., 1991), thus reducing the need to kill individuals. In studies incorporating adult seabirds of known age, mercury concentrations in feathers were unrelated to age (Fumess et al., 1990; Thompson et al., 1991, 1993a), removing this possible confounding factor. The stability of mercury bound to feather keratin and the fact that there are extensive collections of preserved seabirds in museum collections dating back

194

D.R. Thompson et al./Environmental Pollution 101 (1998) 193-200

over 150 years, has made it possible to investigate longterm patterns of mercury contamination using feathers from preserved and contemporary seabirds from a range of habitats (Appelquist et al., 1985, Thompson et al., 1992a, 1993a, b, in press). Mercury concentrations in feathers have been shown to reflect those in the diet in a dose-dependent fashion (Lewis and Furness, 1991), high dietary intake of mercury leading to correspondingly high levels in feathers. Concentrations of mercury vary considerably between birds within a population, and mercury is the only heavy metal for which good evidence exists that bioamplification occurs up food chains (Bryan, 1979). On that basis, one would predict that seabirds feeding at relatively high trophic positions would exhibit elevated mercury concentrations, whilst those seabirds feeding at relatively low trophic positions would exhibit lower mercury concentrations. This could explain part of the intra-population variation in mercury concentrations. However, linking measured mercury concentrations in seabirds with intra- and inter-specific differences in diet has proved difficult beyond a very broad scale approach (Muirhead and Furness, 1988; Lock et al., 1992; Thompson et al., 1992b). In the case of feathers, this problem is compounded due to a paucity of detailed dietary information at the time of feather growth in seabirds, which often takes place away from the breeding colony outwith the breeding season. Furthermore, defining the trophic status of an individual or species requires a comprehensive assessment of its diet (for example, Sanger, 1987), which may not be practical in many seabirds which spend many months of each year away from land. Complimentary to conventional dietary studies, the use of nitrogen and carbon stable isotope analysis has been employed in order to quantify trophic status and dietary overlap of seabirds (for example, Hobson, 1991, 1993; Hobson et al., 1994; Thompson et al., 1995). This approach is based on the fact that stable isotope signatures in proteins of consumers reflect those in their prey in a predictable manner (DeNiro and Epstein, 1978, 1981; Peterson and Fry, 1987). For nitrogen, the ratio 15N:14N (conventionally expressed as Si5N) exhibits a step-wise enrichment of about 4%0 at each trophic level (for example, Minagawa and Wada, 1984; Hobson and Welch, 1992; Hobson, 1993; Hobson et al., 1994). For carbon, the ratio 13C:12C (Si3C) may also provide some trophic information, increasing at each trophic level, but to a lesser degree than nitrogen (for example, Rau et al., 1983; Fry and Sherr, 1984; Hobson and Welch, 1992). Stable isotope signatures of feathers reflect the diet at the time of feather growth (Hobson and Clark, 1992), and have been used to infer temporal variation in the trophic status of free-living seabirds (Thompson and Furness, 1995; Thompson et al., 1995).

The use of stable isotope analysis to quantify trophic status has clear implications for the study of pollutant transfer up food chains. Several recent studies have combined stable isotope analysis with the assessment of a wide range of pollutant burdens (for example, Broman et al., 1992; Cabana and Rasmussen, 1994; Kidd et al., 1995a, b; Kiriluk et al., 1995; Muir et al., 1995; Jarman et al., 1996, 1997). Here we investigate the influence of trophic status, as defined using stable isotope analysis in feather samples, on mercury concentrations in the same feathers from northern fulmars Fulmarus glacialis and great skuas Catharacta skua. Specifically, we test the hypothesis that fulmars and skuas with high mercury concentrations in feather samples occupy correspondingly high trophic positions, defined from the stable isotope signatures. Then we incorporate analyses of different feather types regrown at different times within the moult sequence, to test the hypothesis that mercury concentration and isotope signature varies as the body pool of mercury is depleted through moult.

2. Methods 2.1. Sample collection and preparation Body-feather samples (6-10 small feathers from each individual) were taken from the breast, flanks and dorsal region of great skuas and northern fulmars during the breeding season at Foula, Shetland (60” 08’ N 02” 05’ W). A total of 27 northern fulmars and 25 great skuas were sampled for body feathers, although a small number of great skua samples were subsequently lost during some analyses. Additionally, from 10 individuals that had either been previously shot (under licence) by crofters or found dead at Foula and subsequently stored deep frozen at -20°C the tenth (outermost) primary (lolO), and the first (outermost), eighth and sixteenth secondaries (2”1, 2”8 and 2”16) were removed from the right wing. Skuas moult the innermost primary first, then progress to primary 10, moult of secondaries starting once primary moult is well progressed. Thus all these selected feathers represent the latter part of the moult cycle. All feather samples were stored in polythene bags prior to further treatments. Any surface contamination (primarily blood from birds that had been shot or found dead) was removed by washing feathers in 0.25M sodium hydroxide solution. This relatively mild agent, compared to organic solvents such as acetone, was assumed to have no effect on feathers. Cleaned feathers were dried in an oven at 50°C for 24 h. In order to homogenise samples; all feathers were ground to an extremely fine powder using an impactor mill operating at liquid nitrogen temperature.

D.R. Thompson et al./Environmental Pollution 101 (1998) 193-200

2.2. Mercury analysis All feather samples were analysed for total mercury using a cold vapour technique following standard acid digestion (Furness et al., 1986). All mercury concentrations are expressed on a fresh weight basis as pg gg’ (equivalent to ppm). The limit of detection was 0.01 pg g-i, although no concentration measured in the present study approached this lower limit. Accuracy of mercury measurement was determined by analysing International Atomic Energy Agency horse kidney Reference Material H-8. The results obtained from these analyses fell well within the 95% confidence limits of the published mean for this material (Thompson and Furness, 1989).

195

cross-check consistency. The standard deviations of the means of replicate in-house analyses were f 0.2 for carbon and kO.3 for nitrogen. Stable isotope ratios were expressed in conventional notation whereby: -

Where X is “N or 13C and R is the corresponding ratio R lsN.idN or 13c.12~ standard for 15N and 13C are atmospheric nitrogen (AIR) and Peedee Belemnite (PDB), respectively.

3. Results 2.3. Stable isotope analysis Weighed sub-samples of the powdered feathers were placed into tin buckets (Elemental Microanalysis Ltd) and crimped. Sample size was dependent on the isotope being determined and was approximately 1 pg for carbon analyses and 3mg for nitrogen analyses. Samples (in the crimped tin cups) were then combusted at 1030°C with silvered cobaltous oxide granules and chromium oxide in an elemental analyser (Carlo Erba) using oxygen from an external cylinder. Excess oxygen was removed using copper wire. The combustion products were dried using magnesium perchlorate and in the case of nitrogen analyses the CO2 was removed using carbosorb. The dry gases were transported in a stream of CP grade helium to a gas source isotope ratio mass spectrometer (Micromass ‘Optima’). Each sample of combustion products was measured relative to a working standard gas (CO2 or N2), a sample of which was admitted to the mass spectrometer and measured immediately prior to each experimental sample. The working standard gases were characterized relative to the IAEA Nitrogen and Carbon standards (IAEA 305a and 309a) and an internal reference (leucine) was run at the start and end of all experimental batches to

Nitrogen and carbon stable isotope signatures, together with mercury concentrations, for all feather types from both northern fulmars and great skuas are presented in Table 1. For body feathers, mean S15N values were not significantly different between fulmars (14.96%~; Table 1) and great skuas (14.90%; Table 1; I-test: #=0.23, df= 50, p=O.891), whereas there was a significant difference of 1% between mean S13C values (fulmar, -15.57%; great skua -14.57K, Table 1; t-test: t = -2.21, df = 47, p < 0.05). Mercury concentrations in body feathers were markedly different between the two species, with no overlap in values (Table 1). There were no significant correlations between either S15N or S13C signature and mercury concentration in body feather samples for either species (Fig. 1; Pearson correlation p > 0.05 in each case). For fulmars, there was a significant positive correlation between 615N and S13C values in body feathers (Pearson correlation r = 0.43, n = 27, p < 0.05), although the same was not true in great skuas (Pearson correlation r = -0.12, n = 22, p = 0.607). There were no significant differences in mean 615N and S13C values nor mean mercury concentrations between the four different flight feather types sampled from great skuas (Table 1; l-Way ANOVA: p > 0.05 in

Table I Nitrogen and carbon stable isotope signatures and mercury concentrations in feather samples from northern fulmars and great skuas Species

Northern fulmar (Fulmarus glacialis) Great skua (Catharacta skua)

Feather type

6”N(%)

s’q%)

Mean f SE (sample size)

Range

Mean f SE (sample size)

body

14.96+0.15 (27)

13.6517.19

-15.57*0.34

body primary 10 secondary 1 secondary 8 secondary 16

14.90 f 0.21 (25) 16.03*0.31 (10) 16.21 ItO. (9) 16.09~kO.34(10) 16.01 AO.39 (9)

12.91-17.12 136417.03 15.28-18.35 13.51-17.10 13.94-18.49

-14.57 ZlYO.28 (22) -13.97+0.53 (10) -14.48*0.38 (10) -14.03+0.44 (10) -14.22&0.44 (10)

(27)

f-fg (gg g-‘) Range -21.67-13.21 -16.43-l 1.74 -16.44-11.92 -15.88-12.49 -15.64-11.79 -16.0912.23

Mean f SE (sample size) 1.66kO.14 (27) 11.08 f 1.05 (22) 6.08hO.92 (10) 4.14+0.73 (10) 6.95 f 1.64 (9) 3.91 kOo.66(10)

Range

0.6e3.28 4.68-28.50 2.25-12.50 1.53-9.11 3.4&18.00 2.17-8.91

196

D.R. Thompson et al,/Environmental Pollution 101 (1998) 193-200

each case). Similarly, there were no significant correlations between either 615N or Si3C and mercury values in any of the four flight feather categories, nor between Si5N and 613C in any of the flight feathers (Pearson correlation ~~0.05 in each case). For each variable measured (S15N, 613C and Hg) there were significant and positive correlations between different flight feather categories. However, the strongest correlations were always between values in primary 10 and secondary 8 (Fig. 2; Pearson correlations S15N, I= 0.96, n = 10, p
isotope analysis has been employed to investigate patterns of synthetic organic pollutant accumulation. For example, Broman et al. (1992) reported that the concentrations of some polychlorinated dibenzo-p-dioxin congeners correlated positively with 615N values in Baltic food chains, Kidd et al. (1995b) found similar relationships for hexachlorocyclohexanes, total DDT and total chlorinated bornanes in a subarctic lake and Muir et al. (1995) reported that levels of total PCBs, total DDT, total chlordane and toxaphene all correlated positively with 615N signatures in walrus Odobenus rosmarus.

In the present study we did not find any positive relationships between S15N values and mercury concentrations, in either species or for any feather type (see Section 3 and Fig. 1). The timing and duration of body feather moult has not been studied in detail in fulmars and great skuas. For fulmars, it is thought that body moult begins during July and continues until midFebruary, with two successive generations of body feathers present at any time (Cramp and Simmons, 1977; Ginn and Melville, 1983). Similarly in great skuas, body feather moult is thought to begin around August and continue until mid-March (Cramp and Simmons, 1977; Ginn and Melville, 1983). Hence, for both species, a sample of 6-10 body feathers chosen randomly from an individual bird could be representative of many months’ feather growth. In both species, flight-feather

4. Discussion The usefulness of 615N signatures as a measure of trophic status in studies of pollutant accumulation has been reported recently for a variety of pollutants. In the case of mercury, Cabana and Rasmussen (1994) described positive correlations between 615N values and mercury concentrations in lake trout Salvelinus namaytush in Canadian shield lakes, whilst Kidd et al. (1995a) noted similar patterns for a range of freshwater fish species over a number of lakes. Similarly, Jarman et al. (1996) reported a positive correlation between S15N and mercury concentration in the eggs of seabirds from the Gulf of the Farallones food web. In addition, stable

18

0

17

??Fulmar

Great skua

??

0

0 0

0 0 0 0

0

0

??

Cl

0 0

0

?? 0

??

13

0

0

??

0

12 0

5

10

15 Hg W

20

25

30

g-‘)

Fig. 1. Comparison between mercury levels and SL5Nsignatures in the body feathers of great skuas and northern fulmars.

D.R. Thompson et al./Environmental Pollution 101 (1998) 193-200

12 1

t

13

14

15

16

Primary 10 6%

17

18

(7,)

-16 r ??

??

?? ? ? ?? ??

-.

1

-12

-II

-13

-14

-15

-16

-17

Primary IO t?‘C(“/,)

21 ??

%o 18 -00 i

?? ??

0

’ 0

I

2

4

6

8

10

12

14

Primary 10 Hg level (pg g-‘)

Fig. 2. The relationships between primary 10 and secondary 8 for: (a) 615N, (b) 613Cand (c) Hg.

moult is thought to occur over a similar time-scale (Cramp and Simmons, 1977, 1983; Ginn and Melville, 1983), the body pool of accumulated mercury depleting over this period. Randomly selected body feathers will have been grown at different times during this depletion of the body pool of mercury and as such will have mercury concentrations dependent on how far this depletion had progressed: body feathers moulted and replaced towards the onset of post-breeding moult should have relatively high mercury concentrations, whereas those moulted and replaced towards the end of the moulting process, and when the body pool of mercury will be correspondingly diminished, should have relatively low mercury concentrations. Hobson and

191

Clark (1992), in a study of captive American crows Corvus brachyrhynchos, demonstrated that S13C signatures in growing feathers tracked those of the diet. If S15N signatures in feathers similarly represent those in the diet at the time of feather growth, then isotope values and mercury concentrations in body feathers are effectively ‘uncoupled’, isotope signatures representing the diet at the time of feather growth, with mercury concentrations representing the extent to which the body pool of mercury had diminished when any given body feather was growing. Such an uncoupling of isotopic signatures from mercury burdens in body feathers would account for the lack of any significant relationships found here. In a recent study of mercury concentrations in seabirds at Foula, Stewart et al. (1997) found a significant positive relationship between the amount of seabird prey in the diet (measured through regurgitated pellet analysis) and mercury concentration in body feathers of great skuas. In other words, those skuas feeding at relatively high trophic levels (with a large proportion of regurgitated pellets containing other seabirds) had higher mercury levels in their body feathers. That relationship further highlights that isotope values in body feathers may be an excellent index of diet and trophic status at the time they were grown, but explain little of the variation in mercury concentrations which reflect, in part, accumulation of mercury during the breeding season. That there were no significant relationships between isotope values and mercury concentrations in flight feathers from great skuas (see Section 3) is perhaps more surprising. We chose flight feathers that are likely to have been regrown towards the end of the moult sequence, when the body pool of mercury should be at its lowest, and when dietary inputs of mercury are likely to have relatively more impact on concentrations in growing feathers. The primary feathers are lost and replaced sequentially, starting with primary 1, the innermost primary, and ending with primary 10, the outer-most primary (Ginn and Melville, 1983). Moult of secondary flight feathers in great skuas has received little attention, but is likely to be ascendant (towards the body) starting with secondary 1 (the outer-most secondary). Therefore, particularly primary 10 and secondaries 8 and 16 chosen in the present study can be considered indicative of phases of the moult sequence when body pool levels of mercury should be relatively low, and when dietary inputs of mercury would be relatively more important. However, the lack of any significant relationships between mercury concentrations and isotopic signature in the present study may imply that even these flight feathers are not replaced late enough in the moult sequence for dietary sources of mercury be the principal determining effect. Mean Si5N values for body feathers from fuhnars and great skuas were remarkably similar (Table l), indicating

198

D.R. Thompson et al./Environmental Pollution 101 (1998) 193-200

that these species feed at very similar trophic levels over the time these feathers were replaced. The diet of great skuas at Foula consists mainly of sandeels Ammodytes sp. and discarded whitefish and to a lesser extent other seabirds (Furness and Hislop, 1981; Hamer et al., 1991; Phillips et al., 1997). Fulmars at Foula feed mainly upon sandeels, other, mainly discarded, fish and fish offal (Fumess and Todd, 1984; Thompson et al., 1995; Hamer et al., 1997). On the basis of conventional dietary information, one might predict great skuas to be at a higher trophic level than fulmars since seabirds, which occupy relatively high trophic positions, make up part of the skuas’ diet. Furthermore, this dietary difference should also result in higher mercury levels in great skuas, since seabird prey have been shown to exhibit much higher mercury concentrations compared to fish at Foula (Thompson et al., 1991). In contrast to the similarity in S15N values, mercury concentrations in body feathers were indeed markedly different between these two species (mean 1.66 pg gg’ in fulmars, 11.08 pg g-l in great skuas; Table l), with no overlap in values. It is clear, therefore, that even on an interspecific basis, mercury burdens in body feathers cannot be predicted from corresponding 6i5N values. However, the dietary information above represents the diet during the breeding season; almost nothing is known of the diet of these two species once they leave the colony after breeding. As mentioned previously, body feathers in fulmars and great skuas are replaced Outwith the breeding-season, beginning just after breeding is completed. It may well be, therefore, that isotopic signatures in body feathers in these two species could be indicative of a very similar diet post-breeding, but that mercury concentrations in body feathers reflect breeding-season diet to some extent. The apparent anomaly between trophic status (as defined using 6i5N signatures) and mercury concentrations in body feathers further illustrates that these two parameters are based on two different sources of information. This may go some way to explain why it has been difficult in the past to link interspecific differences in mercury concentrations in body feathers to known diet in seabirds (for example Thompson et al., 1992b). Alternatively, the close agreement in S15N values between northern fulmars and great skuas may be indicative of winter feeding in two separate locations, which had different 615N profiles. Variation in the nitrogen signal at the base of the food chain could produce essentially the same 615N value for different prey at different trophic levels. Hence, fulmars and great skuas may indeed have markedly different diets and trophic positions outwith the breeding season. This difference could in part contribute to the considerable difference in mercury concentrations measured in these two species, but feathers grown at this time may exhibit the same 615N signatures due to specific food chain nitrogen isotope profiles.

Feathers have clear advantages as monitoring tools for mercury contamination in seabirds. However, timing of moulting and of the replacement of specific feathers within the moult sequence, would appear to have a pronounced effect of mercury burdens in feathers (see Furness et al., 1986) extending to body feathers. The diet (and trophic status) at the time of feather growth, as indicated here through stable isotope measurements, would seem to have far less impact on mercury concentrations in feathers. However, since wintering ranges of these species are very large and overlap extensively (Cramp and Simmons, 1977, 1983) the similarity in 6i5N signatures may reflect a corresponding similarity in diet and trophic status at this time. The large difference in the mercury concentrations found in body feathers would appear, therefore, to reflect differences in summer diet (see previously). Whilst we did not find any significant relationships between mercury burdens and isotopic signatures in any feather type, there were significant positive correlations between different flight feathers for both mercury concentration and isotopic signature (see Section 3 and Fig. 2). In particular, mercury concentration, Si5N and 6i3C values were most strongly correlated between primary 10 and secondary 8. Agreement between isotopic signatures in two feathers may be indicative of diets consumed at different times but which were coincidentally very similar isotopically. If this were the case then isotope values would not be useful as a means of elucidating timing of moult and feather growth in birds. Only when isotope values were different between two feathers could one be absolutely certain that those feathers were not grown at the same time. However, the fact that in the present study two isotope signatures and a heavy metal concentration (in this case mercury) were all positively and significantly correlated between two feather types (primary 10 and secondary 8) is perhaps indicative that these feathers may have been regrown together. Since mercury concentrations in feather represent the level of depletion of the ‘body pool’ of mercury at the time of feather growth (see previously), this alone would tend to suggest that similar mercury concentrations represent concurrent position in the moult sequence. The fact that two isotope values also concur supports this since it is clear that variation exists in isotope signatures between different flight feathers in great skuas (Table l), a pattern previously noted for northern fulmars (Thompson and Fumess, 1995) indicating that these species’ diets can vary during the course of flight feather replacement. The application of stable isotope analysis to different feathers may prove to be a useful approach to the elucidation of moult patterns in seabirds, which would otherwise be difficult or impossible to study as much of the moulting process in seabirds takes place away from land following the completion of breeding.

D.R. Thompson et aLlEnvironmental Pollution IO1 (1998) 193-200

Acknowledgements We thank the Holboum family for permission to work on Foula and Peter Thompson for his technical assistance with isotope analyses. S. Bearhop was supported by a Glasgow University Scholarship and a Glasgow University Initiative Grant to RWF funded this study.

References Appelquist, H., Asbirk, S., Drabaek, I., 1984. Mercury monitoring: mercury stability in bird feathers. Marine Pollution Bulletin 15,22-24. Appelquist, H., Drabaek, I., Asbirk, S., 1985. Variation in mercury content of guillemot feathers over 150 years. Marine Pollution Bulletin 16, 244-248. Braune, B.M., Gaskin, D.E., 1987. Mercury levels in Bonaparte’s gulls (Larus Philadelphia) during autumn molt in the Quoddy Region, New Brunswick. Canada. Archives of Environmental Contamination and Toxicology 16,539-549. Broman, D., Naf, C., Rolff, C., Zebuhr, Y., Fry, B., Hobbie, J., 1992. Using ratios of stable nitrogen isotopes to estimate bioaccumulation and flux of polychlorinated dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs) in two Baltic food chains from the northern Baltic. Environmental Toxicology and Chemistry 11, 331-345. Bryan, G.W., 1979. Bioaccumulation of marine pollutants. Philosophical Transactions of the Royal Society of London B 286,483-505. Burger, J., 1993. Metals in avian feathers: bioindicators of environmental pollution. Reviews in Environmental Toxicology 5, 203-311. Cabana, G., Rasmussen, J.B., 1994. Modelling food-chain structure and contaminant bioaccumulation. Nature 372,255257. Cramp, S., Simmons, K.E.L., 1977. Handbook of the Birds of Europe, the Middle East and North Africa: Volume 1. Oxford University Press, Oxford. Cramp, S., Simmons, K.E.L., 1983. Handbook of the Birds of Europe, the Middle East and North Africa: Volume 3. Oxford University Press, Oxford. DeNiro, M.J., Epstein, S., 1978. Influence of diet on the distribution of carbon isotopes in animals. Geochimica et Cosmochimica Acta 42,495506. DeNiro, M.J., Epstein, S., 1981. Influence of diet on the distribution of nitrogen isotopes in animals. Geochimica et Cosmochimica Acta 45,341-351. Fry, B., Sherr, E., 1984. SL3Cmeasurements as indicators of carbon flow in marine and freshwater systems. Contributions to Marine Science 27, 1347. Furness, R.W., Hislop, J.R.G., 1981. Diets and feeding ecology of great skuas Catharacta skua during the breeding season. Journal of Zoology 195, 1-23. Furness, R.W., Hutton, M., 1979. Pollutant levels in the great skua Catharacta skua. Environmental Pollution 19,261-268. Furness, R.W., Todd, C.M., 1984. Diets and feeding of fulmars Fulmarus glacialis during the breeding season: a comparison between St Kilda and Shetland colonies. Ibis 126, 379387. Furness, R.W., Muirhead, S.J., Woodbum, M., 1986. Using bird feathers to measure mercury in the environment: relationships between mercury content and moult. Marine Pollution Bulletin 17,27-30. Furness, R.W., Lewis, S.A., Mills, J.A., 1990. Mercury levels in the plumage of red-billed gulls L.urus novaehollandiae scopulinus of known sex and age. Environmental Pollution 63, 33-39. Ginn, H.B., Melville, D.S., 1983. Moult in Birds. British Trust for Ornithology, Tring, Hertfordshire. Hamer, K.C., Fumess, R.W., Caldow, R.W.G., 1991. The effects of changes in food availability on the breeding ecology of great skuas Catharacta skua in Shetland. Journal of Zoology 223, 175-188.

199

Hamer, K.C., Thompson, D.R., Gray, C., 1997. Spatial variation in the feeding ecology, foraging ranges and breeding energetics of northern fulmars in the north-east Atlantic Ocean. ICES Journal of Marine Science 54,645653. Hobson, K.A., 1991. Stable isotopic determinations of the trophic relationships of seabirds: preliminary investigations of alcids from coastal British Columbia. In: Montevecchi, W.A. and Gaston, A.J. (Eds.). Studies of high-latitude seabirdr. I. Behavioural. energetic and oceanographic aspects of seabird feeding ecology, pp. 16-20. Canadian Wildlife Service Occasional Paper 68. Canadian Wildlife Service, Ottawa, Canada. Hobson, K.A., 1993. Trophic relationships among high Arctic seabirds: insights from tissue-dependent stable-isotope models. Marine Ecology Progress Series 95, 7-18. Hobson, K.A., Clark, R.G., 1992. Assessing avian diets using stable isotopes I: turnover of Si3C in tissues. Condor 94, 181-188. Hobson, K.A., Welch, H.E., 1992. Determinations of trophic relationships within a high Arctic marine food web using 6i3C and St5N analysis. Marine Ecology Progress Series 84, 9-18. Hobson, K.A., Piatt, J.F., Pitocchelli, J., 1994. Using stable isotopes to determine seabird trophic relationships. Journal of Animal Ecology 63,786798. Honda, K., Nasu, T., Tatsukawa, R., 1986. Seasonal changes in mercury accumulation in the black-eared kite, Milvus migrans lineatus. Environmental Pollution (A) 42, 325334. Jarman, W.M., Hobson, K.A., Sydeman, W.J., Bacon, C.E., McLaren, E.B., 1996. Influence of trophic position and feeding location on contaminant levels in the Gulf of the Farallones food web revealed by stable isotope analysis. Environmental Science and Technology 30,654660. Jarman, W.M., Sydeman, W.J., Hobson, K.A., Bergqvist, P.-A., 1997. Relationship of polychlorinated dibenzo-p-dioxin and polychlorinated dibenzofuran levels to stable-nitrogen isotope abundance in marine birds and mammals in coastal California. Environmental Toxicology and Chemistry 16, lOlf&1013. Kidd, K.A., Hesslein, R.H., Fudge, R.J.P., Hallard, K.A., 1995a. The influence of trophic level as measured by 6i5N on mercury concentrations in freshwater organisms. Water Air and Soil Pollution 80, 101l-1015. Kidd, K.A., Schindler, D.W., Hesslein, R.H., Muir, D.C.G., 1995b. Correlation between stable nitrogen isotope ratios and concentrations of organochlorines in biota from a freshwater food web. Science of the Total Environment 160/161, 381-390. Kiriluk, R.M., Servos, M.R., Whittle, D.M., Cabana, G., Rasmussen, J.B., 1995. Using ratios of stable nitrogen and carbon isotopes to characterize the biomagnification of DDE, Mirex and PCB in a Lake Ontario pelagic food-web. Canadian Journal of Fisheries and Aquatic Sciences 52, 26-2674. Lewis, S.A., Fumess, R.W., 1991.Mercury accumulation and excretion in laboratory reared black-headed gull Larus ridibundus chicks. Archives of Environmental Contamination and Toxicology 21, 316320. Lock, J.W., Thompson, D.R., Furness, R.W., Bartle, J.A., 1992. Metal concentrations in seabirds of the New Zealand region. Environmental Pollution 75, 289-300. Minagawa, M., Wada, E., 1984. Stepwise enrichment of SiSN along food chains: further evidence and the relation between S15N and animal age. Geochimica et Cosmochimica Acta 48, 11351140. Monteiro, L.R., Furness, R.W., 1995. Seabirds as monitors of mercury in the marine environment. Water Air and Soil Pollution 80, 851-870. Muir, D.C.G., Segstro, M.D., Hobson, K.A., Ford, C.A., Stewart, R.E.A., Olpinski, S., 1995. Can seal eating explain elevated levels of PCBs and organochlorine pesticides in walrus blubber from eastern Hudson Bay (Canada)? Environmental Pollution 90, 335-348. Muirhead, S.J., Fumess, R.W., 1988. Heavy metal concentrations in the tissues of seabirds from Gough Island, south Atlantic Ocean. Marine Pollution Bulletin 19. 278-283.

200

D.R. Thompson et al./Environmental Pollution 101 (1998) 193-200

Peterson, B.J., Fry, B., 1987. Stable isotopes in ecosystem studies. Annual Review of Ecology and Systematics 18,293-320. Phillips, R.A., Catty, P., Thompson, D.R., Hamer, K.C., Fumess, R.W., 1997. Inter-colony variation in diet and reproductive performance of great skuas Catharacta skua. Marine Ecology Progress Series 152,285-293. Rau, G.H., Meams, A.J., Young, D.R., Olson, R.J., Schafer, H.A., Kaplan, I.R., 1983. Animal 13C/‘2C correlates with trophic level in pelagic food webs. Ecology 64, 1314-1318. Sanger, G.A., 1987. Trophic levels and trophic relationships of seabirds in the Gulf of Alaska. In: Croxall, P. (Ed.). Seabirds: feeding ecology and role in marine environments, pp. 229-257. Cambridge University Press, Cambridge, UK. Stewart, F.M., Phillips, R.A., Catry, P., Fumess, R.W., 1997. Influence of species, age and diet on mercury concentrations in Shetland seabirds. Marine Ecology Progress Series 151,237-244. Thompson, D.R., Furness, R.W., 1989. Comparisons of the levels of total and organic mercury in seabird feathers. Marine Pollution Bulletin 20, 577-579. Thompson, D.R., Furness, R.W., 1995. Stable isotope ratios of carbon and nitrogen in feathers indicate seasonal dietary shifts in northern fulmars. Auk 112,493-498. Thompson, D.R., Hamer, KC., Furness, R.W., 1991. Mercury accumulation in great skuas Catharacta skua of known age and sex, and its effects upon breeding and survival. Journal of Applied Ecology 28,672-684.

Thompson, D.R., Furness, R.W., Walsh, P.M., 1992a. Historical changes in mercury concentrations in the marine ecosystem of the north and north-east Atlantic Ocean as indicated by seabird feathers. Journal of Applied Ecology 29, 74-84. Thompson, D.R., Furness, R.W., Barrett, R.T., 1992b. Mercury concentrations in seabirds from colonies in the northeast Atlantic. Archives of Environmental Contamination and Toxicology 23,383389. Thompson, D.R., Furness, R.W., Lewis, S.A., 1993a. Temporal and spatial variation in mercury concentrations in some albatrosses and petrels from the sub-Antarctic. Polar Biology 13, 239-244. Thompson, D.R., Becker, P.H., Furness, R.W., 1993b. Long-term changes in mercury concentration in herring gulls Lorus argentatus and common terns Sterna hiruna’o from the German North Sea coast. Journal of Applied Ecology 30, 3 16-320. Thompson, D.R., Furness, R.W., Lewis, S.A., 1995. Diets and longterm changes in Si5N and 6i3C values in northern fulmars Fulmarus glacialis from two northeast Atlantic colonies. Marine Ecology Progress Series 125, 3-l 1. Thompson, D.R., Furness, R.W., Monteiro, L.R., in press. Seabirds as biomonitors of mercury inputs to epipelagic and mesopelagic marine food chains. Science of the Total Environment (in press). Walsh, P.M., 1990. The use of seabirds as monitors of heavy metals in the marine environment. In: Furness, R.W. and Rainbow, P.S. (Eds.). Heavy metals in the marine environment, pp. 183-204. CRC Press, Boca Raton, FL.