Environ. Sci. Technol. 2010, 44, 6869–6874
Preliminary Assessment of Avian Stomach Oils: A Vector of Contaminants to Chicks and Potential for Diet Analysis and Biomonitoring K A R E N L . F O S T E R , * ,† S H I W A Y W . W A N G , ‡ DON MACKAY,§ MARK L. MALLORY,| AND JULES M. BLAIS† Program for Chemical and Environmental Toxicology, Department of Biology, University of Ottawa, Ottawa, Ontario K1N 6N5, Canada, Sedna Ecological, Inc., P.O. Box 74280, Fairbanks, Alaska 99707, Centre for Environmental Modelling and Chemistry, Trent University, Peterborough, Ontario K9J 7B8, Canada, and Canadian Wildlife Service, Environment Canada, Iqaluit, Nunavut X0A 0H0, Canada
Received March 29, 2010. Revised manuscript received July 10, 2010. Accepted July 15, 2010.
Bird species from the order Procellariiformes or petrels, including the northern fulmar (Fulmarus glacialis), produce high lipid and high energy content stomach oils from the prey they consume, which enables them to exploit distant marine food sources. Stomach oils are also used as a food source for chicks and for defensive purposes. Samples of stomach oils from two Arctic colonies, St. George Island Alaska, USA and Cape Vera, Devon Island Nunavut, Canada, were collected and analyzed for organochlorine contaminants. ΣPCB concentrations ranged from 13 to 236 ng g-1 wet weight (ww) and ΣDDT concentrations from 5 to 158 ng g-1 ww and were similar in both sites, though differences in chemical signatures were apparent. Stomach oils are a rich energy source; however, they may also provide a higher dose of contaminants per unit energy than the direct consumption of prey items, as illustrated using mass and energy balance calculations to estimate chick exposure to ΣDDT for hypothetical stomach oil and whole prey diets. The results of this study suggest that stomach oils are an important vector of organochlorine contaminants to chicks and should be considered in future risk assessments of northern fulmars and other species of petrels. To our knowledge this is the first study of stomach oils as an overlooked vector of organochlorine contaminants to chicks and as a potentially valuable medium for dietary analysis and noninvasive biomonitoring both of petrel dietary exposure and of marine contaminant concentrations.
distant, often pelagic, food supplies and return to the nesting site with a source of energy both for themselves and their nest-bound young (1, 2). Stomach oils are produced by both male and female adults and chicks of most species from the order Procellariiformes, commonly referred to as tubenoses or petrels. This order contains 108 species and includes a broad range of seabirds including diving petrels, storm petrels, albatrosses, shearwaters, and fulmars - the northern fulmar (Fulmarus glacialis) is the subject of this study. Petrels are the most widely distributed order of birds, with species occurring in all oceans from the Arctic to the Antarctic (1, 2). Stomach oils are produced from prey items in an organ, the proventriculus, or glandular stomach (Figure 1 3-7). In the proventriculus the oil fraction of the partially digested prey (chyme) is concentrated and retained longer than the aqueous fraction due to slower rates of gastric emptying, enabling oils to be accumulated (8, 9). Components in the chyme are partitioned between the oil and aqueous phases based on relative water-lipid solubilities (8). This process concentrates the energy density of stomach oils by a factor of 5 to 35 over that of the prey items from which they were produced, making them a highly concentrated and readily transported food source (7, 10). Indeed, the average energy density of over 30 stomach oil samples representing 11 different species of petrel was 40 kJ g-1 (39.2 kJ g-1 for northern fulmars (10)), consistent with the energy density of pure fat (40 kJ g-1 (11)). In addition to being a source of energy to feed both themselves and their young, stomach oils are also used for defensive purposes. For example, fulmars will expel stomach oils during agonistic interactions with conspecifics for nest sites and as a defense against predators (12, 13). In some cases the oil-laden plumage of the intruder will render it unable to fly, decrease plumage water repellency, or decrease its ability to thermoregulate possibly resulting in death by drowning or hypothermia (7, 14, 15). The benefits to chick health and development of consuming stomach oils provided by the parents have been well documented (7, 16). Similarly, the composition of stomach oil has been well-studied and is known to consist of a mixture of hydrocarbons, wax esters, various acylglycerols, alcohols, cholesterols, and free fatty acids (3-5, 10, 17, 18). The fatty acid signatures of stomach oils correspond to those of prey items, thus enabling diets to be determined. This use of stomach oil fatty acids as a qualitative indicator of diet has been investigated and applied to northern fulmars and other petrel species (19-22). However, in addition to concentrated
Introduction Avian stomach oils are a readily transportable form of energyrich food that enables certain species of seabirds to exploit * Corresponding author e-mail:
[email protected]. † University of Ottawa. ‡ Sedna Ecological, Inc. § Trent University. | Environment Canada. 10.1021/es1009983
2010 American Chemical Society
Published on Web 08/13/2010
FIGURE 1. Location of the avian proventriculus, the site of stomach oil formation. VOL. 44, NO. 17, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6869
energy, oils, and fatty acids, stomach oils may also contain concentrated levels of contaminants. Boersma (23) observed that the stomach oils of birds nesting close to large petroleum oil spills in the Pacific Ocean contained alkanes characteristic of petroleum oil. Boersma (23) proposed that stomach oils could serve as ideal biomonitors of oceanic water quality because petrels are widely distributed, and they forage at the ocean surface where hydrophobic pollutants such as petroleum oil concentrate and thus are ingested by the birds along with their prey items. Stomach oils can also be collected without harming the birds. Bogan and Bourne (24, 25) measured a concentration of 1.3 ppm of total polychlorinated biphenyls (PCBs) in a sample of northern fulmar stomach oil from the North Atlantic near Scotland. Since the concentration of PCBs in the fat of the bird was 32 times higher than in the stomach oil, they concluded that the high contamination levels found in the fat were not being passed from adult to chick. However, to our knowledge the contaminant levels in stomach oils and their significance to the exposure of chicks has not been studied since. Here we provide an initial assessment of the concentrations of PCBs and organochlorine pesticides (OCPs) in the stomach oils of northern fulmars nesting at two Arctic colonies, to demonstrate that in addition to being a source of valuable energy to chicks, stomach oils may also be a major source of contaminants. The accumulation of dietderived contaminants in stomach oils suggests that these oils may be valuable biomonitors of ocean water concentrations and can assist in the diet analysis of wild birds.
Methods Sample Collection. Stomach oils were collected from 10 adult northern fulmars nesting at a colony on St. George Island, Alaska, USA (56°35′N, 170°35′W) in the eastern Bering Sea in June 2004. Full details of the collection are given elsewhere (22). Briefly, birds were captured from their nests using either a modified dip net, noose-pole or by hand. The stomach oils, which the birds regurgitate as a defense mechanism during handling, were collected directly into Whirl-Pak bags, transferred to glass vials with Teflon lined caps and kept frozen until analysis. Stomach oils were also collected from 2 northern fulmars harvested from the colony at Cape Vera, on Devon Island, Nunavut, Canada (76°13′N, 89°14′W) in 2007. Stomach oils were collected directly from the mouth of the birds into solvent washed Falcon tubes and kept upright and frozen until analysis. Sample Preparation and Lipid Determination. Samples from the St. George Island fulmars were centrifuged for 10 min at 2500 rpm to separate the oil from any particulate or aqueous phases. As available, 0.5 or 1 g of oil was mixed with dichloromethane (DCM), spiked with recovery standards (PCBs 30 and 205, 1,3,5-tribromobenzene (1,3,5-TBB), 1,2,4,5tetrabromobenzene (1,2,4,5-TTBB), δ-hexachlorocyclohexane (δ-HCH), endrin ketone), and filtered through a 0.2 µm PTFE disposable syringe filter. Lipid was separated from the organochlorine analytes in the samples by liquid chromatography on an Agilent 1200 series preparative LC with two columns (Waters 19 × 150 mm and 19 × 300 mm), a PDA detector monitoring 254 wavelength, DCM carrier phase, and a flow rate of 7 mL min-1. Two collections were taken, one for the lipid and one for the analytes. The solvent was evaporated from the lipid extract, and the mass fraction lipid content of the samples was determined gravimetrically. Further cleanup and fractionation of the analyte extract was done on activated silica gel/sodium sulfate columns. Fraction A was eluted with hexane and contained PCBs and less polar OCPs, and then fraction B was eluted with 1:1 hexane:DCM and contained the remainder of the OCPs. Samples were 6870
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 17, 2010
brought down to a final volume of 500 or 200 µL depending on the amount of oil extracted and spiked with internal standard. Stomach oils from Cape Vera fulmars were extracted using an Accelerated Solvent Extractor (ASE 200, Dionex), first with hexane at 100 °C and a pressure of 2000 psi held static for 5 min before a 60% flush and second cycle. Then with DCM at 100 °C, a pressure of 2000 psi held static for 8 min before a 60% flush and second cycle. The mass fraction lipid content was determined gravimetrically by evaporating the solvent from a known percentage of the total extract. Lipid was removed from the analyte extract using size exclusion chromatography; the extract was passed through a calibrated column packed with 60 g of BioBeads using 1:1 hexane:DCM as the mobile phase. The extract was further fractionated and cleaned using silica/sodium sulfate columns as described above. Contaminant Analysis. All samples were analyzed on a Hewlett-Packard 6890 series II gas chromatograph with a 63 Ni electron capture detector (GC-ECD) and a 60 m × 0.25 mm (0.25 µm film) column. St. George Island samples were analyzed on an Agilent 19091J-436 column with hydrogen carrier gas at a flow rate of 20.5 psi. The Cape Vera samples were analyzed on a J&W Scientific DB-5MS column with helium carrier gas at a flow rate of 2.0 mL min-1. PCBs measured were IUPAC congeners: 5/8, 18, 29, 31/28, 52, 49, 44, 66, 101, 99, 87, 110, 149, 118, 146, 153, 132, 105, 163, 138, 187, 183, 128, 156, 201, 157, 180, 170, 195, 194, 206, 209. OCPs measured were as follows: 1,2,3-trichlorobenzene, 1,2,3,4tetrachlorobenzene, pentachlorobenzene, R-hexachlorocyclohexane (R-HCH), hexachlorobenzene (HCB), γ-hexachlorocyclohexane, heptachlor, aldrin, heptachlor epoxide (HE), γ-chlordane, R-endosulfan/R-chlordane, p,p′-DDE, dieldrin, endrin, β-endosulfan, p,p′-DDD, o,p′-DDT, p,p′-DDT, methoxychlor, and mirex. Mean recoveries for all samples including blanks were 95% ( 17 for PCB 30, 88% ( 16 for PCB 205, 71% ( 14 for 1,3,5-TBB, 76% ( 16 for 1,2,4,5-TTBB, 101% ( 22 for δ-HCH, and 54% ( 39 for endrin ketone. Concentrations were blank corrected but not recovery corrected. Method detection limits, calculated as the student’s t value appropriate for a 95% confidence level (tn-1,0.05 ) 2.015) times the standard deviation of the averaged blank concentrations (n ) 6) were 1.8 ng g-1 for ΣPCBs and 2.5 ng g-1 for ΣOCPs for a typical 1 g sample. Two samples of triolein spiked with 16 PCB congeners were analyzed with the St. George Island samples, and on average the recoveries were 75%. Standard reference material (mussel tissue NIST 2978) was analyzed with the Cape Vera samples (n ) 7), results for the concentrations of 20 PCB congeners were on average 87% ( 17 of the certified concentration, and the results for 7 OCPs were 76% ( 20 (i.e., within 13% and 24% of certified concentrations, respectively). Similarly, using the preparative LC/silica method concentrations measured for certified reference material (mussel tissue NIST 2977) were on average 92.6% ( 23 for PCBs, and 66% ( 16 for OCPs, of the certified concentrations.
Results Wet weight concentrations (ww) of PCBs and OCPs in stomach oils from two fulmar colonies are summarized in Table 1. Concentrations of ΣPCB and ΣDDT were the highest among the analyte groups, with mean concentrations of 72.6 ( 21.1 ng g-1 ww and 44.0 ( 14.3 ng g-1 ww, respectively, for St. George Island and 31.6 ng g-1 ww and 12.6 ng g-1 ww for Cape Vera birds. Concentrations of PCBs and OCPs were similar between the two colonies; generally within a factor of 2 to 4. Some differences in chemical compositions (i.e., signatures) between stomach oils from the two colonies were suggested, though the small n for Cape Vera samples
TABLE 1. Organochlorine Chemical Concentrations (ng g-1 ww) in Stomach Oils Collected from Adult Northern Fulmars from Two Arctic Coloniesf
n %lipid PCB 31/28 PCB 52 PCB 101 PCB 99 PCB 118 PCB 138 PCB 153 PCB 180 R-HCH HCB HE p,p′-DDE ΣPCBa ΣCBzc ΣCHLd ΣDDTe Σ4PCBb (% of ΣPCB) p,p′-DDE (% of ΣDDT)
St. George Is., AK
Cape Vera, NU
10 65.9 ( 4.1 3.3 ( 0.6 3.4 ( 0.8 5.3 ( 1.7 6.1 ( 1.9 5.8 ( 1.6 6.9 ( 2.4 10.9 ( 3.4 2.2 ( 0.8 10.6 ( 1.3 16.5 ( 2.3 7.8 ( 1.2 35.5 ( 12.1 72.6 ( 21.1 21.0 ( 2.8 20.9 ( 4.0 44.0 ( 14.3 35.0 ( 2.4 80.9 ( 2.7
2 83.1/80.0 1.1/1.2 0.4/0.2 0.5/0.4 1.9/2.0 3.0/3.9 3.2/3.8 6.4/10.0 2.5/4.3 12.3/9.4 6.1/9.4 5.1/7.1 8.5/11.3 28.2/35.0 9.3/12.6 9.8/10.1 11.3/13.9 53.4/62.8 75.2/81.2
a ΣPCBs ) sum of congeners 5/8, 18, 29, 31/28, 52, 49, 44, 66, 101, 99, 87, 110, 149, 118, 146, 153, 132, 105, 163, 138, 187, 183, 128, 156, 201, 157, 180, 170, 195, 194, 206, 209. b Σ4PCBs ) sum of congeners 118, 138, 153, 180. c ΣCBz ) sum of chlorobenzenes: 1,2,3-triCBz, 1,2,3,4-tetraCBz, pentaCBz, HCB. d ΣCHL ) sum of chlordanes: heptachlor, HE, R-chlordane (coelutes with e R-endosulfan), γ-chlordane. ΣDDT ) sum of dichloro-diphenyl-trichloroethanes: p,p′-DDT, o,p′-DDT, p,p′-DDD, p,p′-DDE. f Mean values are shown ( standard error.
precluded strong, intercolony comparisons. For example, four congeners (PCBs 118, 138, 153, and 180) on average represented 58% of ΣPCB in stomach oils collected from the Cape Vera colony, whereas these four comprised only 35% of ΣPCB in stomach oils from the St. George Island colony (PCBs 31/28, 52, 101, and 99 were also important at this site). For both colonies, more than 75% of ΣDDT was in the form of the metabolite p,p′-DDE. Stomach oils from fulmars at both colonies also contained similar, notable concentrations of R-HCH, heptachlor epoxide (a main component of ΣCHL), and HCB (a main component of ΣCBz). St. George Island stomach oils contained a mean of 66% lipids, and Cape Vera oils contained a mean of 82%. Wet and lipid weight (lw) concentrations of PCBs and OCPs in fulmar stomach oils as well as in typical fulmar prey items from the North American Arctic compiled from the literature are summarized in Table 2. On a wet weight basis, concentrations of organochlorine chemicals and chemical groups in stomach oils were on average 19-fold higher (ranging from 1- to 60-fold higher) than those reported for crustaceans and 9-fold higher (ranging from 1- to 29-fold higher) than those reported for fish. However, on a lipid weight basis the average concentrations of organochlorine chemicals in stomach oils were similar to those reported for crustaceans (stomach oils were between 0.1 and 4.8 times that of crustaceans) and 0.3 times those reported for fish (stomach oils were between 0.04 and 1.6 times that of fish).
Discussion Organochlorine concentrations (ww) in stomach oils were consistently higher than those reported for potential prey items of the fulmars (Table 2). However, since stomach oils are produced via the collection of lipids and associated
hydrophobic contaminants from prey over time in the proventriculus, wet weight concentrations of hydrophobic contaminants in the stomach oils are probably higher than those of the prey due to the higher lipid content and fugacity capacity (31) of the stomach oils. Indeed, the lipid weight concentrations of the prey were similar to stomach oil concentrations (Table 2). Given that stomach oils can be sampled noninvasively, we consider three implications of these results: (1) energy benefits versus contaminant exposure costs of stomach oils as a food source; (2) potential for diet analysis applications; and (3) potential for contaminants biomonitoring research. 1. Stomach Oils: Energy Benefits versus Contaminant Exposure Costs. In order to assess the implications of such high concentrations of contaminants in stomach oils on chick exposure, concentrations expressed per unit of energy should also be considered and are calculated by dividing the concentration (ww; Table 2) by the energy density of the dietary item (ED kJ g-1 ww). Stomach oils have a higher energy density than either crustaceans or fish, i.e., 9 and 8 times more, respectively (Table 2). Thus, to meet the energy demands of the chick, presumably less stomach oil needs to be consumed than whole prey items such as crustaceans or fish. Each unit of energy (kJ) obtained by a fulmar from a food source has an associated amount of contaminant, which is also transferred to the fulmar (Figure 2). Energy obtained from stomach oils was generally associated with higher contaminant exposure than either crustaceans or fish (up to 5 times more). The concentrations (ng kJ-1) in stomach oils from Cape Vera fulmars were not as high as those from St. George Island and ranged from 0.4 to 2.7 times those of prey. Thus, the consumption of stomach oils as an energy source for chicks can be associated with a higher contaminant exposure cost. However, since stomach oil concentrations are dependent on the amount of contaminants in prey items, a desirable approach is to measure the concentration of hydrophobic contaminants of whole prey items and to track how these concentrations change through the formation of stomach oils (below). Chick Exposure: Mass and Energy Balance. The field metabolic energy requirement of a 30 day old fulmar chick is approximately 800 kJ day-1 (32, 33). We calculated the ΣDDT exposure cost to a chick associated with obtaining its energy requirement for a single day (EChick ) 800 kJ) from each of four hypothetical diets: 1) stomach oils generated from crustaceans, 2) whole crustaceans, 3) stomach oils generated from fish, and 4) whole fish (Figure 3). The amount of ΣDDT consumed (N ng) in each diet was calculated as the product of the mass (M g ww) of the diet required and the concentration (C ng g-1 ww) of ΣDDT in the diet, where M was calculated as EChick divided by the energy density of the diet (ED, kJ g-1 ww). C for whole prey was l.3 ng g-1 ww for crustaceans, which is consistent with Mysis oculata and Pandalus sp. from Baffin Bay, and 2.6 ng g-1 ww for fish, which is consistent with Arctic cod from Baffin Bay and the Beaufort/Chukchi seas (Table 2). C for stomach oils was equal to the lipid weight (lw) concentration of ΣDDT in the whole prey, reflecting the assumption that in the proventriculus of the fulmar, the entire amount of hydrophobic contaminants in the whole prey are retained and concentrated along with the lipids from the prey during stomach oil production. The hypothetical mass of stomach oils required to meet the energy requirement of the chick was 20 g, compared with 191 g of whole crustaceans and 170 g of whole fish. The large differences in diet masses reflects the higher energy density of stomach oils and illustrates the use of stomach oils as a compact and readily transportable food source for fulmar parents and chicks. However, the amount of ΣDDT VOL. 44, NO. 17, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6871
7.3 (190.0) 6.9 (180.0)
0.8 (66.1)
Theragra chalcogramma (Walleye pollock)
Beaufort/Chukchi Seas North Pacific and Bering Sea
4.70c
a Refer to references for the constituents considered in the sum of chlorobenzenes (ΣCBz), chlordanes (ΣCHL), polychlorinated biphenyls (ΣPCB), and dichloro-diphenyl-trichloroethanes (ΣDDT). Either lw or ww concentrations were calculated from reported concentrations and % lipid. b Reported value for Parathemisto libellula. c Reported value for Arctic cod. d Values shown are for whole prey items and for stomach oils. The energy densities (ED kJ g-1 ww) of each food type are also shown.
30a 13.7 (354.0)
29a 2.6 (70.0) 2.8 (76.0) 2.0 (54.0)
1.2 (95.9) 3.7 (308.3)
2.5 (67.0)
26 28a 3.1 (258.7)
0.8 (13.3) 1.4 (28.0) 1.3 (26.4) 3.1 (140.9) 0.9 (13.5) 0.7 (14.6) 1.7 (33.7) 3.8 (172.7) 0.4 (6.8) 0.4 (7.0) 0.6 (13.2) 0.4 (19.5) 5.1 (81.0) 5.8 (116.0) 29.9 (609.2) 8.8 (400.0) 4.18 Baffin Bay Baffin Bay Baffin Bay Baffin Bay
Crustaceans Calanus hyperboreus Mysis oculata Pandalus sp. Parathemisto libellula (adult) Fish Boreogadus saida (Arctic cod)
Baffin Bay
6.1/9.4 (7.4/11.7) 12.3/9.4 (14.8/11.7) b
39.2 Northern Fulmar stomach oils
St. George Island, AK Cape Vera, NU
ED
2.6 (215.7)
26 27a 27a 27a 27a
this study 11.3/13.9 (13.6/17.4) 9.8/10.1 (11.8/12.6) 9.3/12.6 (11.2/15.8) 28.2/35.0 (33.9/43.8)
72.6 (106.0) 35.5 (51.0) 16.5 (25.4)
8.5/11.3 (10.2/14.1)
10 this study 44.0 (63.4) 20.9 (31.4) 21.0 (32.3)
ΣCHL ΣCBz ΣPCB p,p′-DDE HCB r-HCH
10.6 (15.9)
reference ΣDDT 9
location
TABLE 2. Concentrations in ng g-1 ww and (ng g-1 lw) of Organochlorine Chemicals in Food Sources for Northern Fulmars in the North American Arcticd 6872
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 17, 2010
transferred to chicks by consuming stomach oils made from crustaceans was 531 ng, as compared with 249 ng for whole crustaceans. Stomach oils made from fish contained 1434 ng as compared with 443 ng for whole fish, a 2- and 3-fold difference, respectively. Obtaining energy from whole prey items, when energy densities are based not just on lipid but also on protein and carbohydrates, results in lower contaminant exposure than obtaining energy from lipids only (as with stomach oils). Thus, chicks fed proportionally more stomach oils are predisposed to receiving a higher dose of hydrophobic contaminants, such as ΣDDT or other organochlorine contaminants, than are chicks fed whole prey. This has important implications for assessing contaminant fate, exposure, uptake, and transfer in Arctic seabirds and for petrels in general and should be considered in any examination of contaminant uptake and accumulation in petrels. For example, sympatrically nesting seabirds (i.e., seabirds with overlapping breeding ranges), such as thick-billed murres (Uria lomvia) that feed their chicks whole prey, probably provide a lower dose of contaminants to their chicks than do fulmars, even if they feed their young proportionally more fish (26). Similarly, fulmars that fly long distances to acquire food to provision their chicks (e.g., the Cape Vera colony (13)) possibly provide a higher proportion of their meal as stomach oil, owing to increased time for digestion, than fulmars which feed close to their colony. Therefore, colony-specific differences in contaminant burdens of chicks might be attributable in part to foraging distances and consequent dietary proportions of stomach oils. Thus, for petrels such as northern fulmars, stomach oils are an important and overlooked source of contaminants. 2. Contaminants in Stomach Oils and Diet Analysis. Given the dietary origin of petrel stomach oils (3-7), the concentrations of contaminants found in them are dependent on the concentrations in the prey items from which the oils were produced. Northern fulmars are opportunistic feeders and will consume a broad range of prey including zooplankton (especially crustaceans), squid, fish, and fisheries discards (offal) (12, 34). Since fulmar diet varies both spatially and seasonally (12, 35), concentrations of various contaminants in stomach oils should also vary through the year and among sampling locations. Differences in the contaminant signatures of stomach oils from fulmars at the St. George Island colony and the Cape Vera colony suggest that they were produced from prey items with different contaminant signatures, either attributable to the consumption of different prey items or to different marine contaminant signatures. Indeed, fulmars from the High Arctic consume a higher proportion of invertebrates than birds from the North Pacific, and fish consumed by High Arctic colonies are almost entirely Arctic cod (Boreogadus saida) whereas North Pacific fulmars consume a greater variety of fish including fisheries offal (12, 34, 36). Future research on contaminant signatures in stomach oils, when coupled with those from possible prey items, could provide complementary information to diet analysis research linking fatty acid signatures in prey to those in petrel stomach oils (19-22). 3. Biomonitoring. Our study reinforces the findings of Boersma (23) and suggests that stomach oils are essentially a collection of lipid and associated contaminants from prey items found across the foraging area of the seabird colony. Petrels are exclusively marine feeders (1, 2), and thus all contaminants accumulated in stomach oils are of marine origin. Contaminants in oceanic waters at low concentrations are accumulated by prey items and transferred to petrels. Petrels then retain the lipid fraction of the prey and the associated hydrophobic chemicals, thereby concentrating marine-derived contaminants to detectable levels. Thus, petrel stomach oils are a valuable biomonitor of the prevailing
FIGURE 2. Concentrations of organochlorine contaminants in stomach oils collected from northern fulmars from two Arctic colonies (this study) and fulmar prey items from the North American Arctic (27-30), expressed per unit of energy. Error bars indicate standard error.
FIGURE 3. Calculations of ΣDDT (N ng) delivered to a one month old fulmar chick in one day via four hypothetical diets: 1) stomach oils produced from crustaceans, 2) whole crustaceans, 3) stomach oils produced from fish, and 4) whole fish. Where the mass of the diet required (M g ww) ) EChick/ED, C ) the concentration in the diet (ng g-1 ww), and N ) M × C. lipid levels of hydrophobic contaminants in the marine environment and hence of the levels (i.e., fugacities and concentrations) in water and sediments. Since stomach oils can be sampled without killing the birds and they reflect prey lipids collected recently (22), novel research strategies for regional and temporal contaminant biomonitoring are possible and make stomach oils particularly suitable biomonitors. Three examples include the following: (1) petrel populations at risk or in decline can be sampled without further contributing to the decline; (2) the same individual can be sampled repeatedly and over many years to allow for the separation of variability in contaminant burdens attributable to individual birds from that attributable to regional levels; and (3) repeated sampling of the same individual or colony enables an assessment of temporal contaminant patterns in the marine environment, both short (e.g., seasonally) and long-term (e.g., from year to year). In combination with other techniques (e.g., fatty acid or stable isotope analyses), contaminants may be traced to specific prey items. Given such information, and consistent laboratory protocols, a novel long-term biomonitoring research goal
could be to compare stomach oil contaminant concentrations and signatures from different colonies across the Arctic in order to determine regional differences in marine contaminants as distinct from differences in diet. We recommend petrel stomach oils as a noninvasive and nondestructive alternative to traditional techniques of monitoring contaminants in the marine environment.
Acknowledgments The authors are grateful to the funding agencies that supported this work: the United States Geological Survey (USGS), Alaska Science Center, the Natural Sciences and Engineering Research Council Canada (NSERC) for a Canadian Graduate Scholarship awarded to K.L.F. and a Strategic Projects Grant to J.M.B., and Indian and Northern Affairs Canada (NSTP), Natural Resources Canada (PCSP), and Environment Canada (CWS). We also thank S. Hatch and A. Fontaine for field assistance. Any use of trade names is for descriptive purposes only and does not imply endorsement by the U.S. Government. The authors wish to thank VOL. 44, NO. 17, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6873
three anonymous reviewers for their constructive criticisms on an earlier draft of this manuscript.
Literature Cited (1) Warham, J. The petrels: their ecology and breeding systems; Academic Press: San Diego, 1990. (2) Warham, J. The behaviour, population ecology & physiology of the petrels; Academic Press: San Diego, 1996. (3) Cheah, C. C.; Hansen, I. A. Stomach oil and tissue lipids of the petrels Puffinus pacificus and Pterodroma macroptera. Int. J. Biochem. 1970a, 1, 203–208. (4) Cheah, C. C.; Hansen, I. A. Wax esters in the stomach oil of petrels. Int. J. Biochem. 1970b, 1, 198–202. (5) Clarke, A.; Prince, P. A. The origin of stomach oil in marine birds: analyses of the stomach oil from six species of subantarctic Procellariiform birds. J. Exp. Mar. Biol. Ecol. 1976, 23, 15–30. (6) Imber, M. J. The origin of petrel stomach oils - A review. Condor 1976, 78, 366–369. (7) Warham, J. The incidence, functions and ecological significance of petrel stomach oils. Proc. New Z. Ecol. Soc. 1977, 24, 84–93. (8) Place, A. R.; Stoyan, N. C.; Ricklefs, R. E.; Butler, R. G. Physiological basis of stomach oil formation in leach’s storm-petrel (Oceanodroma leucorhoa). Auk 1989, 106, 687–699. (9) Roby, D. D.; Brink, K. L.; Place, A. R. Relative passage rates of lipid and aqueous digesta in the formation of stomach oils. Auk 1989, 106, 303–313. (10) Warham, J.; Watts, R.; Dainty, R. J. The composition, energy content and function of the stomach oils of petrels (order, Procellariformes). J. Mar. Biol. Ecol. 1976, 23, 1–13. (11) Peters, R. H. The ecological implications of body size; Cambridge University Press: New York, 1983. (12) Hatch, S. A.; Nettleship, D. N. Northern fulmar (Fulmarus glacialis). In The birds of North America, no 361; Poole, A., Gill, F., Eds.; The Birds of North America Inc.: Philadelphia, 1998. (13) Mallory, M. L.; Forbes, M. R. Costly pre-laying behaviors and physiological expenditures by northern fulmars in the high Arctic. Ecoscience 2008, 15, 545–554. (14) Swennen, C. Observations on the effect of ejection of stomach oil by the fulmar Fulmarus glacialis on other birds. Ardea 1974, 62, 111–117. (15) Clarke, A. Contamination of peregrine falcons (Falco peregrinus) with fulmar stomach oil. J. Zool., London 1977, 181, 11–20. (16) Roby, D. D.; Taylor, J. R. E.; Place, A. R. Significance of stomach oil for reproduction in seabirds: An interspecies cross-fostering experiment. Auk 1997, 114, 725–736. (17) Lewis, R. W. Studies of glyceryl ethers of the stomach oil of Leach’s Petrel Oceanodroma leucorhoa. Comp. Biochem. Physiol. 1966, 19, 363–377. (18) Lewis, R. W. Studies on the stomach oils of marine animals II. Oils of some procellariiform birds. Comp. Biochem. Physiol. 1969, 31, 725–731. (19) Connan, M.; Cherel, Y.; Mayzaud, P. Lipids from stomach oil of procellariiform seabirds document the importance of myctophid fish in the Southern Ocean. Limnol. Oceanogr. 2007a, 52, 2445–2455. (20) Connan, M.; Cherel, Y; Mabille, G.; Mayzaud, P. Trophic relationships of white-chinned petrels from Crozet Islands: Combined stomach oil and conventional dietary analyses. Mar. Biol. 2007b, 152, 95–107.
6874
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 17, 2010
(21) Connan, M.; Mayzaud, P; Trouve, C.; Barbraud, C.; Cherel, Y. Interannual dietary changes and demographic consequences in breeding blue petrels from Kerguelen Islands. Mar. Ecol.: Prog. Ser. 2008, 373, 123–135. (22) Wang, S. W.; Iverson, S. J.; Springer, A. M.; Hatch, S. A. Fatty acid signatures of stomach oil and adipose tissue of northern fulmars (Fulmarus glacialis) in Alaska: implications for diet analysis of Procellariiform birds. J. Comp. Physiol., B 2007, 177, 893–903. (23) Boersma, P. D. Ingestion of petroleum by seabirds can serve as a monitor of water quality. Science 1986, 231, 373–376. (24) Bogan, J. A.; Bourne, W. R. P. Organochlorine levels in Atlantic seabirds. Nature 1972a, 240, 358. (25) Bogan, J. A.; Bourne, W. R. P. Polychlorinated biphenyls in North Atlantic seabirds. Mar. Pollut. Bull. 1972b, 3, 171–175. (26) Elliott, K. H.; Gaston, A. J. Mass-length relationships and energy content of fishes and invertebrates delivered to nestling thickbilled murres Uria lomvia in the Canadian Arctic, 1981-2007. Mar. Ornithol. 2008, 36, 25–34. (27) Fisk, A. T.; Hoekstra, P. F.; Gagnon, J.-M.; Duffe, J.; Norstrom, R. J.; Hobson, K. A.; Kwan, M.; Muir, D. C. G. Influence of habitat, trophic ecology and lipids on, and spatial trends of, organochlorine contaminants in Arctic marine invertebrates. Mar. Ecol.: Prog. Ser. 2003, 262, 201–214. (28) Fisk, A. T. OCs in Arctic cod from northern Baffin Bay. 2002. In Canadian Arctic Contaminants Assessment Report II (CACAR II): Contaminant levels, trends and effects in the biological environment; Indian and Northern Affairs Canada: Ottawa, 2003. (29) Hoekstra, P. F.; O’Hara, T. M.; Fisk, A. T.; Borgå, K.; Solomon, K. R.; Muir, D. C. G. Trophic transfer of persistent organochlorine contaminants (OCs) within an Arctic marine food web from the southern Beaufort-Chukchi Seas. Environ. Pollut. 2003, 124, 509–522. (30) Kawano, M.; Inoue, T; Wada, T.; Hldaka, H.; Tatsukawa, R. Bioconcentration and residue patterns of chlordane compounds in marine animals: invertebrates, fish, mammals, and seabirds. Environ. Sci. Technol. 1988, 22, 792–797. (31) Mackay, D. Multimedia Environmental Models: The Fugacity Approach, 2nd ed.; CRC Press LLC: Boca Raton, FL, 2001. (32) Hodum, P. J.; Weathers, W. W. Energetics of nestling growth and parental effort in Antarctic fulmarine petrels. J. Exper. Biol. 2003, 206, 2125–2133. (33) Phillips, R. A.; Hamer, K. C. Growth and provisioning strategies of Northern Fulmars Fulmarus glacialis. Ibis 2000, 142, 435– 445. (34) Byers, T.; Smith, A.; Mallory, M. L. Diet of black guillemots and northern fulmars breeding beside a High Arctic polynya. Polar Biol. 2010, 33, 457–467. (35) Mallory, M. L. The northern fulmar (Fulmarus glacialis) in Arctic Canada: ecology, threats, and what it tells us about marine environmental conditions. Environ. Rev. 2006, 14, 187–216. (36) Phillips, R. A.; Petersen, M. K.; Lilliendahl, K.; Solmundsson, J.; Hamer, K. C.; Camphuysen, C. J.; Zonfrillo, B. Diet of the northern fulmar Fulmarus glacialis: reliance on commercial fisheries. Mar. Biol. 1999, 135, 159–170.
ES1009983
