Marine Pollution Bulletin Vol. 40, No. 2, pp. 152±164, 2000 Ó 2000 Elsevier Science Ltd. All rights reserved Printed in Great Britain PII: S0025-326X(99)00194-0 0025-326X/00 $ - see front matter Comparison of Pigeon Guillemot, Cepphus columba, Blood Parameters from Oiled and Unoiled Areas of Alaska Eight Years After the Exxon Valdez Oil Spill PAMELA E. SEISER *, LAWRENCE K. DUFFYà, A. DAVID MCGUIRE§, DANIEL D. ROBY , GREGORY H. GOLETàà and MICHAEL A. LITZOW§§ Department of Biology and Wildlife, University of Alaska Fairbanks, Fairbanks, AK 99775-6100, USA àInstitute of Arctic Biology, University of Alaska Fairbanks, Fairbanks, AK 99775-7000, USA §US Geological Survey, Alaska Cooperative Fish and Wildlife Research Unit, Fairbanks, AK 99775-7020, USA Oregon Cooperative Fish and Wildlife Research Unit, Department of Fisheries and Wildlife, 104 Nash Hall, Oregon State University, Corvallis, OR 97331-3803, USA ààUS Fish and Wildlife Service, 1011 E. Tudor Road, Anchorage, AK 99503, USA §§US Geological Survey, Alaska Biological Science Center, 1011 E. Tudor Road, Anchorage, AK 99503, USA In 1997, we compared the haematological and plasma Keywords: alcid; Cepphus columba; Exxon Valdez; hae- biochemical pro®les among populations of pigeon guille- matology; oil spill; pigeon guillemot; plasma biochemistry; mots, Cepphus columba, in areas oiled and not oiled by the Prince William Sound. 1989 Exxon Valdez oil spill (EVOS) that occurred in Prince William Sound (PWS), Alaska. Pigeon guillemot populations in PWS were injured by EVOS and have not returned to pre-spill levels. If oil contamination is limiting Introduction recovery of pigeon guillemots in PWS, then we expected that blood parameters of pigeon guillemots would dier Population estimates of pigeon guillemots, Cepphus co- between oiled and unoiled areas and that these dierences lumba, in Prince William Sound (PWS), Alaska, have would be consistent with either toxic responses or lower declined from 15 000 individuals in 1972±73 to approx- ®tness. We collected blood samples from chicks at ap- imately 3 000 individuals in the mid-1990s (Dwyer et al., proximately 20 and 30 days after hatching. Physiological 1976; Klosiewski and Laing, 1994; Agler and Kendall, changes associated with chick growth were noted in sev- 1997; Sanger and Cody, 1994). A large-scale regime in eral blood parameters. We found that only calcium and the Gulf of Alaska during the late 1970s (Piatt and mean cell volume were signi®cantly dierent between the Anderson, 1996) likely caused much of this decline, as chicks in oiled and unoiled areas. Despite these dier- high quality forage ®sh were more widely available in ences, blood biomarkers provided little evidence of con- the 1970s than in recent years (Hayes and Kuletz, 1997; tinuing oil injury to pigeon guillemot chicks, eight years Kuletz et al., 1997). Pigeon guillemot populations in after the EVOS. Preliminary data from adults indicated PWS were further impacted by the Exxon Valdez oil elevated aspartate aminotransferase activity in the adults spill (EVOS; Murphy et al., 1997), when the supertanker from the oiled area, which is consistent with hepatocellular Exxon Valdez ran aground on 24 March 1989 and injury. Because adults have greater opportunities for ex- spilled 42 million L of crude oil into PWS. Approxi- posure to residual oil than nestlings, we recommend mately 40% of this oil was deposited on the shorelines of studies that fully evaluate the health of adults residing in PWS (Galt et al., 1991). Between 100 000 to 375 000 oiled areas. Ó 2000 Elsevier Science Ltd. All rights birds died in the spill, of which 1500 to 3000 were pigeon reserved. guillemots (Piatt et al., 1990). Seven years after the spill, pigeon guillemots had not recovered to pre-spill num- bers (Agler and Kendall, 1997; Oakley and Kuletz, *Corresponding author. Tel.: +907-474-5472; fax: +907-474-6716. 1996). It is not clear to what extent demography, food E-mail address: [email protected] availability or the physiological eects of lingering oil 152 Volume 40/Number 2/February 2000 exposure may be constraining recovery of pigeon gu- oil contacting the egg shell. As little as 5 ll of Prudhoe illemots in PWS. Bay crude oil has been reported to cause embryo death Pigeon guillemots are vulnerable to oil spills because (Albers, 1977; Szaro et al., 1978a). Dosing studies of they use the near-shore habitat (King and Sanger, 1979; weathered crude oil on congeneric black guillemots, Piatt et al., 1990). They breed in small colonies along Cepphus grylle, suggest that oil ingestion may cause rocky coastlines, and roost on intertidal rocks. Guille- long-term physiological eects which could reduce a mots spend much of their time on the sea surface or young birdÕs ability to survive at sea (Peakall et al., diving for surface schooling ®sh, demersal ®sh and in- 1980). vertebrates associated with the intertidal and subtidal Payne et al. (1986) suggested that detecting simple zones. changes in a biochemical or physiological response in a The prey of pigeon guillemots are also susceptible to population may provide information on the presence of oil contamination. There is evidence of longer-term toxins. Haematological analyses (dierential cell counts) toxic eects of oil to ®sh populations when oil persists in may provide information about the immunological sta- their natal habitats (Murphy et al., 1999; Rice, 1999). tus of birds (Campbell, 1986a). Levels of plasma en- For example, Paci®c herring, Clupea pallasii, embryos zymes provide information on the function of organs, exposed to oil yielded more physically deformed larvae e.g. liver (Campbell, 1986a). Elevated levels of acute- than unoiled embryos (Kocan et al., 1996; Hose et al., phase protein haptoglobin indicate responses to exoge- 1996). Biomarkers of oil ingestion were noted in PWS nous toxins, bacterial or viral infections, and physical ®sh several years after EVOS. Walleye pollack, Theragra trauma (Silverman and LeGrys, 1987). Physiological chalcogramma, collected from oiled Naked Island in changes occurring during the chick growth period have 1990 and 1991, exhibited high levels of ¯uorescent aro- been suggested by many authors to in¯uence blood matic compounds in their bile (Collier et al., 1996). parameters (Wolf et al., 1985; Homan et al., 1985; Jewett et al. (1995) reported that demersal ®sh in the Kostlecka-Myrcha, 1987; Starck, 1998; Work, 1996; oiled eelgrass beds of Herring Bay, PWS, demonstrated Prichard et al., 1997). To prevent age-dependent varia- a high incidence of haemosiderosis lesions in the liver. tion from biasing assessments, haematological and Kelp greenling, Hexagrammos decogrammus, collected plasma biochemical pro®les should be repeated on in 1996 showed signi®cantly higher expression of P450 chicks at dierent stages of development. activity in oiled Herring Bay versus unoiled Jackpot Bay To make an accurate assessment of clinical tests, (Holland-Bartels et al., 1998). Research in the early reference values of healthy individuals are needed 1990s demonstrated that oil exposure had detrimental (Hawkey and Samour, 1988), but information on hae- eects on near-shore predators including river otters, matological and clinical chemistry on pigeon guillemots Lutra canadensis (Bowyer et al., 1994; 1995, Duy et al., or other alcids is limited (Newman et al., 1997; Newman 1993, 1994) and sea otters, Enhydra lutris (Loughlin and Zinkl, 1998; Prichard et al., 1997; Kostleck-Myrcha, et al., 1996). Whether residual oil from the EVOS 1987). We assume therefore that colonies in the unoiled aected pigeon guillemots required further evaluation. areas represent healthy populations. If oil contamina- Acute toxic eects of petroleum hydrocarbons are tion is limiting recovery of pigeon guillemots in PWS, we well known (Leighton, 1993), but the lingering eects of expected that blood chemistry and cell counts would chronic oil exposure have not been investigated fully in dier between oiled and unoiled areas and these dier- free ranging piscivorous birds (Fry and Lowenstine, ences should be consistent with either toxic responses or 1985). Leighton (1993) provided an extensive review of lower ®tness. In this study, we compare the haemato- avian studies of petroleum oil toxicity. Dosing experi- logical and plasma biochemical pro®les between pigeon ments have shown that the eects of oil ingestion in- guillemot populations in an oiled area of PWS and in clude: (1) lower hatch rate and altered yolk structure unoiled areas of PWS. (Grau et al., 1977; Szaro et al. 1978a); (2) reduced rate of growth (Szaro et al., 1978b; Peakall et al., 1982); (3) Methods and Materials slower development and reduced survivorship of chicks (Trivelpiece et al., 1984); (4) liver, kidney and intestine During summer 1997, measurements of growth and damage in long-term exposure (Khan and Ryan, 1991; blood samples from pigeon guillemot chicks were col- Patton and Dieter, 1980; Fry and Lowenstine, 1985); lected in areas oiled by the EVOS and in reference areas and (5) Heinz-body haemolytic anaemia associated with that were not oiled (Fig. 1). The oiled area we evaluated a substantial decrease in packed-cell volume (Leighton was Naked Island (60° 400 N, 147° 280 W) in central et al., 1983). PWS. The prevailing winds and currents during spring Because guillemot chicks remain in their natal burrow of 1989 deposited oil predominately on the east and until they ¯edge, oil contamination can occur through north-west shorelines of Naked Island (Galt et al., 1991; contact with the oiled feathers of an adult while in the Oakley and Kuletz, 1996). The combined colonies of egg or chick stage, or through ingestion of contaminated Jackpot Island (60° 190 N, 148° 110 W) and Icy Bay (60° ®sh (Leighton, 1993; Peakall et al., 1980). At nine days 140 N, 148° 170 W) in south-western PWS were not oiled of incubation, avian embryos are extremely sensitive to and represent the reference areas in this study. For 153 Marine Pollution Bulletin Fig.