The Diet of the Antarctic Shag During Summer at Nelson Island, Antarctica ’

The Condor 100:112-118 0 The Cooper Ornithological Society 1998

THE DIET OF THE ANTARCTIC SHAG DURING SUMMER AT NELSON ISLAND, ANTARCTICA ’

MARCO FAVERO Universidad National de Mar de1 Plata, Fact&ad de Ciencias Exactas y Naturales, Departamento de Biologia, Funes 3250, (7600) Mar de1 Plata, Argentina, e-mail: [email protected]

RICARDO CASAUX Institute Antdrtico Argentino, Division Biologia, Cerrito 1248, (1010) Buenos Aires, Argentina

PATRICIA SILVA UC-Universidad National de Mar de1 Plata, Facultad de Ciencias Exactas y Naturales, Departamento de Ciencias Marinas, Funes 3350, (7600) Mar de1 Plata, Argentina

ESTEBAN BARRERA-ORO AND NESTOR CORIA Instituto Antdrtico Argentino, Division Biologia, Cerrito 1248, (1010) Buenos Aires, Argentina

Abstract. Analysis of 139 stomachcontents of the Antarctic Shag (Phalacrocorax bransjeldensis) collected at Nelson Island, Antarctica, showed that fish were the main component in the diet, followed by octopods, gammarids, euphausiids,and polychaetes.The fish No- tothenia coriiceps predominatedin frequency (56%) and in mass(49%), whereasHarpagtfer antarcticus was the most important by number (58%). The sizes of N. coriiceps and H. antarcticus taken by males were significantly larger than those caught by females. This dietary difference is probably due to differential prey selection related to shags’ sexual dimorphism, temporal prey availability, and/or different foraging depths/areas.Observations at nests showed that females mainly foraged in the morning, whereas males foraged in the evening. No significant differences were observed between the number of daily foraging trips made by males and females, averaging 1.7 during incubation,2.0 in early chick-rearing, and 4.6 during the late chick-rearing period. The estimated daily food intakes were 321, 315, and 758 g for females, and 421, 582, and 1,288 g for males during incubation, early and late chick-rearing, respectively. Key words: Antarctic Shag, Antarctica, diet, fish, Phalacrocoraxbransfieldensis, prey temporal variation, sexual variation in foraging.

INTRODUCTION regurgitations and pellets (Casaux et al. 1995a, The Antarctic Shag (Phalucrocorux bransjel- Veldkamp 1995). On the other hand, analysis of densis) is distributed in the Antarctic Peninsula stomach contents can stressthe birds, demands and South Shetland Islands (de1 Hoyo et al. more time in the field, and must be used with 1992), and feeds predominately on demersal caution when inferring meal sizes (Duffy and benthic fish (Casaux and Barrera-Oro 1993, Co- Jackson 1986). However, it gives better infor- ria et al. 1995) close inshore at depths to 116 m mation on the size and mass of prey items, and (Croxall et al. 1991). The diet of shagshas been permits inferences about the foraging behavior studied mainly by analysis of regurgitated casts and energy requirements of birds. Coria et al. (Green et al. 1990a, 1990b, Barrera-Oro and Ca- (1995) provided information on the diet of Ant- saux 1996). Although this method provides use- arctic Shags at the South Shetland Islands by ful information with minimal disturbance and analysis of stomach contents, but the data were sampling effort, the erosion by digestion and restricted to the middle chick-rearing period. loss of otoliths and other hard parts throughout There is a paucity of information on sexual the gastrointestinal tract could give biased re- variation in the diet of this species. Bernstein sults, a fact largely proved in trials with captive and Maxson (1984) at Anvers Island observed cormorants (Casaux et al. 1995b, Zijlstra and sexual differences in time budget and feeding/ Van Eerden 1995), or by comparison between diving behavior of Antarctic Shags, thus sug- gesting the possibility of a diet composedof dif- ’ ’ Received10 October1996. Accepted17 October ferent size and/or speciesof prey as reported for 1997. King Cormorants (Phalacrocorax albiventer) at

11121

114 MARCO FAVERO ET AL.

TABLE 1. The composition of the diet of Antarctic ticus and P. charcoti predominated in female Shags at Duthoit Point during summer 1994-1995, as diet, whereas T. newnesi was mainly eaten by observed throughout different stages of the breeding season (n = 139). Only stages where the food item males. The average size (? SE) of the N. cori- was found are shown. iceps and H. antarcticus specimens ingested by males (23.4 ? 0.7 and 7.7 + 0.2 cm, respec- % of tively) was significantly larger (t,6 = 7.0, P < samples Number (occur- of % 0.001; Lj3* = 2.5, P < 0.05, respectively) than ~~~C.5) P”Y number Mass (g) % mass those ingested by females (15.7 + 0.8 and 7.1 Fish ? 0.1 cm, respectively). No significant differ- Incubation 100.0 334 87.4 9,816 97.4 ences were observed between L. nudifrons (t,,, Early rearing 100.0 328 89.1 9,455 97.6 = 0.2) and T. newnesi (t,, = 0.3) foraged by Late rearing 100.0 350 97.8 10,423 98.4 males and females (Fig. 1). In addition, no sig- Octopods nificant differences were observed within sexes Incubation 12.5 12 3.1 220 2.2 in the comparison of fish sizes caught through- Early rearing 8.7 5 1.4 190 2.0 Late rearing 13.0 8 2.2 170 1.6 out the three sampling periods (one-way ANO- VA, P > 0.05 for both sexes). Polychaetes No significant differences were observed be- Early rearing 2.2 1 0.3 2 0.0 tween males and females in the number of for- Gammarids aging trips per day made during the three sam- Incubation 20.8 33 8.6 36 0.4 pling periods (t-test, P > 0.5 in all comparisons). Early rearing 21.7 28 7.6 29 0.3 Shags averaged 1.7 + 0.2 (n = 14) trips bird-’ Euphausiids day-’ during incubation, 2.0 -t 0.3 (n = 12) in Early rearing 10.9 6 1.6 6 0.1 early rearing, and 4.6 ? 0.5 (n = 10) in late Stones rearing (Kruskal-Wallis test, x2* = 23.2, P < Incubation 2.1 3 0.8 5 0.0 0.001). The foraging activity started from 06:OO to 07:15 and finished from 20:00 to 21:30; females mainly foraged in the morning (71% of A total of 1,019 fish were represented in the the trips), whereas males foraged in the evening samples,belonging to 8 species:Not&z&u COI- iiceps, Lepidonotothen nudzjrons, Harpagifer TABLE 2. Sexual differences in the diet of Antarctic antarcticus, Trematomus newnesi, Gobionototh- Shags at Duthoit Point during summer 1994-1995. en gibberzji-ens, Notothenia rossii, Pagothenia bernacchii, and Parachaenichthys charcoti. Females Males (n = 70) (n = 69) TOtal Nine fish prey remained unidentified. N. coriiceps was the most frequent (F% [occurrence] = Fish 56.1) and important in mass (48.5%), whereas % Frequency 100.0 100.0 Number 760 252 H. antarcticus predominated in number (57.6%) % Number 75.1 24.9 more so in females (560) than in males (23). Octopods The mean number of fish per sample was 7.4 % Frequency 21.4 1.4 11.5 (range l-34), and no significant differences Number 24 1 25 were observed between periods (F2,,35 = 0.1). % Number 96.0 4.0 2.3 The differences were significant in the compar- Polychaetes ison between sexes (t,,, = 7.0, P < 0.001) being % Frequency 1.4 0.0 0.7 11.1 ? 0.9 (n = 70) for females and 3.6 5 0.4 Number 1 0 1 (n = 69) for males. Seventy-seven stomachsam- % Number 100.0 0.0 0.1 ples (55%) were composed of a single fish spe- Gammariids cies, 41 of them were N. coriiceps. In 32 sam- % Frequency 15.4 4.3 14.4 ples (23%) a single specimen was found, 29 of Number 49 12 61 them were N. coriiceps. % Number 80.3 19.7 5.5 The size of several fish species, including N. Euphausiids coriiceps, H. antarcticus, N. nudifrons, and T. % Frequency 5.7 1.4 3.6 newnesi varied significantly throughout the sam- Number 1 6 % Number 8:.3 16.7 0.5 pling periods (all P < 0.05). Harpagifer anturc- DIET OF THE ANTARCTIC SHAG 115

H. antarcticus L. nudifons

100 80 -

i$t 60 R 402:: i E

N. coriiceps T. newnesi 100 80

&F 60 L 40 /

9-11 15-17 21-23 27-29 33-35 3-5 9-11 15-17 21-23 27-29 33-35 Totallength (cm) Totallength (on)

FIGURE 1. The percent frequency distribution (F%) by length of the fish prey speciescaught by males (solid bars) and females (hatched bars) of the Antarctic Shag at Duthoit Point during summer 1994-1995.

(44%). The duration of the trips differed signif- Coria et al. (1995), also at Duthot Point but icantly throughout the season, averaging 173, in a previous season,reported that the foraging 126, and 91 min during the incubation, early and activity of Antarctic Shagswas restrictedto day- late chick-rearing periods, respectively (F2,,, = light hours. In agreement with previous studies 6.0, P < 0.01). The time spent away averaged carried out in the Antarctic Peninsula (Bernstein 150 + 17 min (n = 28) in males and 129 5 17 and Maxson 1984) and South Georgia Islands min (n = 32) in females (tss = 0.9). (Kato et al. 1991, Wanless et al. 1995), females Using the mean mass of the stomachcontents mainly foraged in the morning, whereas males by sex during the three sampling periods, and were more active during the rest of the day. As the mean number of daily foraging trips, the av- reported by Kato et al. (1995) on Macquarie erage daily intake ranged from 421 to 1,288 g shags (P. purpuruscens) in sub-Antarctica, Ant- for males, and from 321 to 758 g for females. arctic Shag males brought back heavier loads Assuming that fish represents98% of the shags’ composed of a few large fish, whereas females diet and that fish provide an energy content of brought back lighter loads made up of a larger 3.97 kJ g-l (Croxall et al. 1985) and an assim- number of small prey items. ilation efficiency of 75% (Wiens 1984), we estimated an averaged intake of 1,941 and 1,383 SEXUAL DIFFERENCES IN THE DIET W day-l for males and females, respectively. Although the fish species ingested by both sexes were similar, some differences were observed; DISCUSSION males preyed on larger fish (e.g., N. coriiceps As reported in previous diet studies based upon and H. antarcticus) than did females (Fig. 1). the analysis of pellets (Casaux and Barrera-Oro These differences in prey choice could be relat- 1993, Barrera-Oro and Casaux 1996), our results ed to (1) sexual size dimorphism, (2) temporal indicated that fish were the main prey of the prey availability, given that males and females Antarctic Shag. Mollusks were the next most differentially distributed their foraging effort common prey, which agrees with other studies during the day, and/or (3) sexual differences in carried out in the same area (Casaux and Bar- diving depths or foraging areas. Hypothesis (1) rera-Or0 1993, Coria et al. 1995). is supportedby the fact that females, which are 116 MARCO FAVERO ET AL. up to 20% lighter than males, preyed mainly on findings of Bernstein and Maxson (1985) at An- smaller fish. However, the largest H. antarcticus vers Island. In contrast, Wanless et al. (1995) specimens, which were ingested by males, were found no evidence that the number of trips and smaller than the N. coriiceps specimenseaten by the time spent foraging increased with chick age females. In support of hypothesis (2), Casaux et at Bird Island. South Georgian Shags probably al. (1990) observed at Potter Cove (10 km from respond to the growing chicks’ energy demands Duthoit Point) that activity of adult N. coriiceps by increasing the mass of the food loads and/or was infrequent during the morning and increased foraging on more energetically rich prey. On the in the afternoon/evening, which coincided with other hand, some differences were reported in male foraging activity. Regarding hypothesis other cormorants regarding the number and du- (3), it has been suggestedfor cormorants that, ration of foraging bouts by males and females. due to the higher functional body oxygen stored, In the Bank Cormorant (Phalucrocorax neglec- larger species display deeper dives (Cooper tus), female foraging bouts are longer than that 1986). Therefore, sexual differences in diving of males, whereas males showed significantly depths and/or behavior are expected. In Potter more foraging bouts than did females (Cooper Cove (King George Island), larger N. coriiceps 1985b). However, our data indicate that both specimensinhabit deeper waters (R. Casaux, un- sexes distributed similarly their foraging effort, publ. data), and it is probable that at Duthoit at least considering the time spent away and the Point, the Antarctic Shag partitioned the use of number of foraging trips performed during the the foraging depths and/or areas with males incubation and chick-rearing period. preying in deeper waters than did females, thus Considering the food loads for each sex and diminishing intraspecific competition. that there were no differences between the sexes Sexual differences in diet also have been re- in the number of foraging trips, males brought a corded in other Antarctic birds (Volkman et al. greater amount of food to the nests. The energy 1980, Croxall and Lishman 1987). Clarke et al. intake differed by 24,46, and 42% between sex- (1995) suggestedthat such differences in Adelie es during incubation, early and late chick-rear- Penguins Pygoscelis adeliae may become im- ing, respectively, which was higher than the portant as a means of reducing intraspecific 15% estimated by Bernstein and Maxson (1985) competition and of maximizing the chances of at the Antarctic Peninsula by means of activity one member of the pair locating food for the budget. The difference in energy intake between chicks without excessive waste of overall search sexes also occurred during the incubation peri- effort. Despite substantial differences in the tim- od, indicating that it cannot be solely attributed ing and pattern of diving between sexesof South to a simple difference in chick provisioning. Georgia Shags at Bird Island (South Georgia Is- The energetic estimations for incubating lands), no definitive difference in diving depths (2,384 and 2,754 kJ bird-‘) and early rearing between male and female birds has been ob- (2,330 and 2,688 k.I bird-l) Antarctic Shags (fe- served (Kato et al. 1991). males and males, respectively) represent about half of the daily energy expenditures calculated CONSUMPTION RATES by means of time budget estimations of Bem- The daily food requirement for the Antarctic stein and Maxson (1985). The differences ob- Shag is reported to range from 600 to 1,209 g served between our values and those from the (Shaw 1984). Coria et al. (1995), while studying literature could be due to limitations of the the same species during the middle rearing pe- methodology used in this work: (a) biasesduring riod (chicks averaging 1,400 g and 22 days of this study in the estimation of the mean number age), obtained a maximum daily intake of 1,410 of foraging trips; when the maximum numbers g. The range of daily food intake estimated for of observed trips were tested, our values for in- each period in this study was greater than pre- cubating and early chick-rearing fall into the vious work. The variation through the chick range reported by Bernstein and Maxson (1985); rearing period can be explained by the increase and/or (b) biases in the estimation of the food in the chicks’ energy demand. Although food ingested per trip; Wanless et al. (1993) postulat- loads remain constant, adults respond to that de- ed that breeding Phalacrocorax aristotelis digest mand by increasing the number of foraging trips their own food before returning to the nest and and the time spent away, which is similar to the the contents of the loads were primarily for the DIET OF THE ANTARCTIC SHAG 117 chicks. During our study, adult shags were fre- ShetlandIslands, Antarctica. Polar Biol. 11:63- quently observed resting in the foraging area be- 72. CLARKE, J., K. KERRY, AND E. FRANCHI.1995. Differ- tween dives for periods longer than 1 hr. More- encesin the foraging strategiesof male and female over, the digested remains found in regurgita- Adelie Penguins.Comm. Conserv. Antarctic Mar. tions could contribute, at least partially, to adult Living Reiour. Ecosys. Monitoring Manage. requirements. On the other hand, our estimations Working Groun. Document 95/46. Siena. Italv. COOPER,J. 1985a.’ k note on the diet of Cape kor- for the third period (3,834 and 2,257 kJ day-’ morant Phalucrocoran capensis at Mercury Is- bird-’ for males and females, respectively) fit land, South West Africa, in November 1978. S. better with the energy expenditure for late rear- Afr. Mar. Sci. 3:129-130. ing shags (2,844 and 2,443 kJ day-l bird-’ for COOPER,J. 1985b. Biology of the Bank Cormorant, males and females, respectively) reported by part 3: foraging behaviour.Ostrich 56:8&95. COOPER, J. 1986. diving patternsof cormorants(Phal- Bernstein and Maxson (1985). acrocoracidae1.Ibis 128:562-569. These findings suggest that more detailed CORIA, N., R. C&AUX, M. FAVERO,AND I? SILVA. studieson the foraging behavior of the Antarctic 1995. Analysis of the stomach content in the Shag are needed. Further knowledge of the en- Blue-eyed Shag Phalacrocorax atriceps bransfel- ergetic cost of adults foraging for chicks, their densis, at Nelson Island, South Shetland Islands. Polar Biol. 15:349-352. meal size and feeding frequency should allow CROXALL,J. l?, AND G. S. LISHMAN. 1987. The food more precise estimates of total adult energy ex- and feeding ecology of penguins, p. 101-133. In penditure during the chick-rearing period, par- J. P Croxall [ed.], Seabirds, feeding ecology and ticularly taking into account sexual differences. role in marine ecosystems. Cambridge Univ. Press, Cambridge. ACKNOWLEDGMENTS CROXALL,J. I?, Y. NAITO, A. KATO, P ROTHERY,AND D. BRIGGS. 1991. Diving patterns and perfor- We thank C. Bellisio for his assistancein part of the mance in the Antarctic Blue-eyed Shag Phalacro- field work, and M. Rovetta for his logistic support.J. c0ra.xatriceps. J. Zool. 225: 177-199. Cooper made helpful commentson an earlier draft. We CROXALL,J. I?, P. A. PRINCE,AND C. RICKETTS. 1985. also thank S. J. Maxson, W. D. Koenig, and two anon- Relationshipsbetween prey life-cycles and the ex- ymous referees whose criticisms improved the manu- tent, nature and timing of seal and seabird pre- script. dation in the Scotia Sea. D. 516-533. In W. R. Siegfried, F?R. Condy, an’d R. M. Laws [eds.], LITERATURE CITED Antarctic nutrient cycles and food webs. Springer- Verlag, Berlin. BARRERA-ORO,E., AND R. J. CASAUX. 1996. Fish as DEL HOYO, J., A. ELLIOT, AND J. SARGATAL.1992. diet of the Blue-eyed Shag Phalacrocorax atriceps bransjeldensis at Halfmoon Island, South Handbook of the birds of the world. Vol. 1. Lynx Shetland Islands. Cybium 20:37-45. Editions, Barcelona. DUFFY,D. C., ANDS. JACKSON.1986. Diet studiesof BERNSTEIN,N. F?,AND S. J. MAXSON. 1984. Sexually distinct daily activity patternsof Blue-eyed Shags seabirds: a review of methods. Colonial Water- in Antarctica. Condor 86:151-156. birds 9:1-17. AND BERNSTEIN,N. P, AND S. J. MAXSON. 1985. Repro- GREEN,K., R. WILLIAMS, D. SLIP. 1990a. Diet of ductive energeticsof Blue-eyed Shags in Antarc- the Macquarie Island Cormorant Phalacrocorax tica. Wilson Bull. 97:450-462. atriceps purpurascens. Corella 14:53-55. CASAUX,R., AND E. BARRERA-ORO.1993. The diet of Green, K., R. Williams, E. Woehler, H. Burton, N. the Blue-eyed Shag, Phalacrocorax atriceps Gales, and R. Jones. 1990b. Diet of the Heard bransfieldensis feeding in the Bransfield Strait. Island Cormorant Phalacrocorax atriceps nivalis. Antarctic Sci. 5:335-338. Antarctic Sci. 2:139-141. CASAUX.R.. E. BARRERA-ORO,M. FAVERO,AND I? SIL- HECHT,T. 1987. A guide to the otoliths of Southern VA.’ 1995a. New correction factors for the quan- Ocean fishes. S. Afr. J. Antarct. Res. 17:1-87. tification of fish representedin pellets of the Blue- KATO,A., J. l? CROXALL,Y. WATANUKI,AND Y. NAITO. eyed Shag Phalacrocorax atriceps. Comm. Con- 1991. Diving patterns and performance in male serv. Antarctic Mar. Living Resour.Ecosys. Mon- and female Blue-eyed CormorantsPhalacrocorax itoring Manage. Working Group, Document atriceps at South Georgia. Marine Omithol. 19: 95/83:1-18, Siena, Italy. 117-129. CASAUX,R., M. FAVERO,E. BARRERA-ORO,AND I? SIL- KATO, A., I. NISHIUMI,AND Y. NAITO. 1995. Sexual VA. 1995b. Feeding trial on an Imperial Cormo- differences in the diet of King Cormorantsat Mac- rant Phalucrocorax atriceps: preliminary results quarie Island. Polar Biol. 16:75-77. on fish intake and otolith digestion. Marine Or- MARCHANT,S., AND P. H. HIGGINS. 1990. Handbook nithol. 23:7&82. of Australian, New Zealand and Antarctic birds. CASAUX, R., A. MAZZOT~A,AND E. BARRERA-ORO. Vol. 1. Oxford Univ. Press, Melbourne. 1990. Seasonalaspects of the biology and diet of SHAW,I? 1984. Factors affecting the breeding perfor- nearshoreNototheniid fish at Potter Cove, South mance of the Antarctic Blue-eyed Shag (Phala- 118 MARCO FAVERO ET AL.

crocorax atriceps bransjieldensis). Ph.D. diss., WANLESS,S., M. HARRIS,AND E RUSSELL.1993. Fac- Univ. Durham, England. tots influencing food-load sizes brought in by VELDKAMP,R. 1995. Diet of cormorants Phalacro- shagsPhalacrocorax aristotelis during chick rear- corm carbo sinensis at Wanneperveen,The Neth- ing. Ibis 135:19-24. erlands, with special reference to bream Abramis WIENS,J. A. 1984. Modeling the energy requirements brama. Ardea 83:143-155. of seabirdpopulations, p. 255-284. In G. C. Whit- VOLKMAN,N. J., P PRESLER,AND W. TRIVELPIECE. 1980. Diets of Pygoscelid penguins at Ring tow and H. Rahn [eds.], Seabird energetics. Ple- George Island, Antarctica. Condor 82:373-378. num Press, New York. WANLESS,S., M. HARRIS,AND J. MORRIS. 1995. Fac- ZIILSTRA,M, AND M. R. VAN EERDEN.1995. Pellet tors affecting daily activity budgets of South production and use of otoliths in determining the Georgian Shags during chick rearing at Bird Is- diet of cormorants Phalacrocorax carbo: trials land, South Georgia. Condor 97:550-558. with captive birds. Ardea 83: 123-130.