916 SHORT COMMUNICATIONS

The Condor 89:9 16-919 0 The Cooper Ornithological Society 1987

PHILOPATRY, MATE AND NEST-SITE FIDELITY IN THE BROWN SKUAS OF ANVERS ISLAND, ANTARCTICA ’

DAVID F. PARMELEEAND PAMELAJ. PIETZ* Bell Museum, Universityof Minnesota, Minneapolis, MN 55455

Key words: Antarctic Peninsula; Anvers Island; were not difficult to sex.Nevertheless, we verified many Brown Skua; Catharacta lonnbergi; philopatry; mate of our designationsthrough necropsiesof banded in- jidelity; nest-site fidelity;food source. dividuals that died during the study. Knowing the sex of one paired bird allowed us to establishthe sexesof For nine austral summer seasons(October to March) former and subsequentmates of both pair members. from 1974-1975 through 1980-1981, and 1983-1984 It was not possibleto witnesseach fledging;therefore through 1984-1985, the authors and others observed our “fledgling” counts included birds that we believed two skua speciesthat breed sympatrically near U.S. had a good chanceof fledging,judging by the advanced Palmer Station (64”46’S, 64”03’W) on Anvers Island staaeof an individual’s iuvenal plumage. These birds near the Antarctic Peninsula. South Polar Skuas(Ca- provided the basis for determining the percent of re- tharacta maccormicki)nest here in the hundreds and turns. In this study philopatry refers to the return of forage primarily at sea. Brown Skuas (C. lonnbergi) chicksto their natal groundsas prebreedersor breeders. nest in much.smaller numbers and feed at AdClie Pen- Mate and nest-site fidelity refer to the return of adults guin (Pygoscelisadeliae) colonies. During the first sev- to former mates and nest sites, respectively, whether en seasons,our studies centered on behavioral and or not on natal grounds. ecologicalrelationships between the two species(Par- melee et al. 1978, 1979; Neilson 1983; Pietz 1984, RESULTS 1985; Parmelee 1985; Pietz 1986, 1987). During the From 1974-1975 to 1984-1985 we banded 72 Brown final two seasonsevery effort was made to locate pre- Skuaadults and 116 fledglings,including 64 adultsand viously banded adults and young within the principal 94 fledglingswithin the study area. In nine seasons,a study area and in outlying regions(Parmelee and Rim- minimum of 50 Brown Skuas (29 males and 21 fe- mer 1985, Parmelee et al. 1985). Becauseskuas mav males)occupied 18 territories in the study area. In nine live over 30 years(Ainley 1981) and do not breed until casesBrown Skua pairs occupied sites but apparently several years old, long-term studies are essential for produced no eggs.Of 75 nestingsin which eggswere understandingtheir population biology. In this paper found, one was artificially terminated and three pro- we summarize Brown Skua mate and nest-site fidelity ducedchicks whose fates were unknown; the remaining over an 1l-year period, and report returns of banded 7 1 nestingsproduced 96 fledglings,or 1.35 fledglings/ young to their natal areas. nest. One male and two female Brown Skuasfrom this samplewere involved in mixed-speciesmatings as well. STUDY AREA AND METHODS Pairs were composed of South Polar Skua males and The principal study area (Fig. 1) included a number of Brown Skuafemales in all but one case.Eleven of these smallmoss-and-lichen-covered peninsulas and off-shore pairings produced eggs,and 17 hybrid chicks fledged islets that were mostly snow-free in summer and pro- (1.5 fledglings/nest). vided nesting grounds for skuas and other seabirds Except for a few of the adults that formed mixed (Parmelee et al. 1977). Access to these sites was by pairs with South Polar Skuas,Brown Skuasinvariably outboard motor-driven rubber boats. Outlying areas nested on islands occupied by Adtlie Penguins (Fig. were visited by means of larger vesselsor helicopters. 1). On all other coastsof Anvers Island and throughout In each seasona concertedeffort was made to iden- much of Palmer Archipelago, both Brown Skuas and tifv Brown Skuasat 18 of the 19 nestina sites shown Ad&lie Penguinswere rare or absent.The only variance in Figure 1. Excluded were unbanded individuals that from this food link was noted on Brabant Island nestedon TorgersenIsland. Most adults were captured (64”08’S, 62”36’W) where, in the absenceof penguins, for banding in hand-held nets, and a few in cannon a pair of nesting Brown Skuasfed on other speciesof nets and baited traps. The majority were color-banded seabirds(Parmelee and Rimmer 1985). for convenient field identification. Within the study area, Brown Skua pairs numbered Female Brown Skuasaverage larger in size and weight from six to 12 per season for the 9-year period. An than males (Neilson 1983);thus, paired birds generally average of 8.6 pairs per year produced eggs.Activity was concentrated within the triangle formed by Tor- gersen (Site T-l), Litchfield (Sites L-l to L-l l), and ’ Received 5 November 1986. Final acceptance 1 Humble (Sites H-l to H-3) Islands (Fig. 1). Over nine June 1987. seasons,the triangle accountedfor more than 80% of 2 Present address:Northern Prairie Wildl. Res. Ct., all Brown Skuaadults, 79% oftheir fledglings,and 77% P.O. Box 2096, Jamestown, ND 58402. of their nests. The triangle also contained 84% of the SHORT COMMUNICATIONS 917

I I PALMER STATION, NTARCTCA,’ AND VC’ N’ TY’

i446’ S’ - + +

Anvers Island Lipps lslond Palm_er Station

L-ii - l-10 “‘ w””‘ Bonaparte PO/

j4O48S’ - +

P a& ’ e&=7- nautical mile Outcast Islands 0 kilometer

64-08’ W’ 64*04/W 64OOOW’ R I I

FIGURE 1. Palmer Study Area. Approximate nest sites of Brown Skuasare shown on Cormorant, Christine, Torgersen,Litchfield. and Humble Islands. AdClie Penguincolonies on these five islandsare indicated by parallel lines. nearly 13,800 penguin nests in the study area. Litch- a pair had successfullyraised one or more fledglings. field Island accommodated only about 580 penguin In at least eight instances,individuals returned to their nests (8-year mean), yet it had 62% of all Brown Skua previous mates despite reproductive failure in the pre- adults, 6 1% ofall fledglings,and 65% ofall nests.Litch- ceding season. field’s Brown Skuashunted mostly on TorgersenIsland It was not clear why someindividuals changedmates wherethere were over 8,700 penguinnests (5-year mean) or nest sites. According to Neilson (1983) different but few nesting skuasof either species. arrival times of the sexesfrom one seasonto the next The longest recorded continuous residency at one likely have a disruptive influence on mate and site site by the same Brown Skuawas 11 seasons,assuming fidelity. A more common factor leading to mate and this female was present during the two seasonsnot site changeswas probably failure of the previous mate checked. The longest recorded continuous residency to return to the territory. Of 50 banded adult breeders, by the same pair was five and likely seven seasons, we know that 30% died (nine from fowl cholera and after which the pair was collected by a parasitologist three as a result of collecting); another 40% failed to (Hoberg 1984). Continuousresidency by the samepair return to the breeding grounds. Inexplicably, twice as was 5 years at three sites when terminated by fowl many malesas femalesdisappeared or were found dead. cholera. The strengthof site tenacity is reflectedby the Another disruptive influence on mate and site fi- fact that individual Brown Skuas averaged only one delity was what we call absenteeism;i.e., former breed- territory during nine seasons(Table 1). Only five birds ers returned to the area but did not breed, or they were changedterritories; two females and one male moved not seen in the area for one or more seasonsand then to different sites on the same island, and one female returned. Two female Brown Skuasreturned to breed and one male moved to sites on adjacent islands. at their former nest sitesafter absencesof 1 and 2 years, In our sample each change of territory involved a respectively; a third returned to a different nest site change of mate. A total of 26 mate changesoccurred after a 2- to 4-year absence.Two males were absent (five involving South Polar Skuas), 21 without a ter- from the breeding grounds for 1 year, then returned ritory change. Reproductive failure rarely preceded but failed to breed again. Another eight males and changesof sites or mates. In 12 of 15 cases a mate seven females were present on the breeding grounds changeoccurred immediately following a year in which but failed to produce eggsfor one to two seasons.Of 918 SHORT COMMUNICATIONS

TABLE 1. Nine-year summary for nine male and 10 occupiedthe seasonfollowing the epidemic by the sur- female Brown Skuasbanded as breeding adults in 1974- viving female and a new mate, but the pair did not 1975. K + SD is calculatedper individuah rangesare produceeggs until 1 year later. Another site was vacant given in parentheses. for one seasonand then occupiedby South Polar Skuas. In 1983 a parasitologicalcollecting expedition (Hoberg Malt3 Females 1984) removed a breeding pair that had occupied the same territory since 1976-1977. South Polar Skuas No. years seen at Palm- 4.4 f 1.3 6.1 f 3.0 occupied the site during the following two seasons, er: (l-5) (l-9) while a new pair of Brown Skuasestablished a territory No. territories held: 1.1 If- 0.3 1.3 f 0.5 nearby. The site from which Hoberg removed a mixed (l-2) (l-2) pair remained vacant for at least two seasons.Clearly Maximum no. years oc- 4.2 i 1.3 5.1 * 2.6 the recovery period at Palmer can be slow. Three Brown Skuasbanded as chicks later returned cupied same site: (l-5) (l-9) as nesting adults (Parmelee et al. 1985), two others Minimum no. mates+ 1.3 * 0.5 1.9 f 0.9 returned as prebreeders. One of the breeding adults (l-2) (l-4) nestedabout 15 km from its natal site, another within Maximum no. years 3.4 f 1.4 3.6 ? 1.4 1 km, and one precisely at its natal nest scrape-an paired with same (l-5) (l-5) example ofperfect philopatry. So far, all returneeshave mate? been female. The youngestbreeding age recorded was No. seasonsoccupied a 4.0 i 1.5 5.5 i 2.8 6 years. site with a mate? (l-5) (l-9) Of 116 banded fledglings,only 4.3% returned during No. seasonshad eggs? 3.9 f 1.5 4.9 f 2.2 the study. This suggestsa wide dispersaland/or a high (l-5) (l-7) mortality of young following fledging. Wide dispersal of young is not supported by our data. Although we No. fledglingsproduced? 4.6 + 2.2 6.5 +- 3.4 searchednumerous coastal areas (interiors ice-capped) (l-7) (l-12) . throughoutthe Palmer Archipelago,we found no band- sAssumes unhanded mates of consecutiveseasons were the samein- ed Brown Skuas.Furthermore. there have been no re- dividual. bExcludes matings wtth South Polar Skuas. coveries of our banded Brown Skuas at other skua nestingsites in the Antarctic or sub-Antarctic, or from more distant areas. 16 cases in which an adult was present but did not breed, 10 involved birds whose mates did not return (three known dead, seven never seen again), four in- DISCUSS1oN volved two injured birds and their mates, and two Along the south coast of Anvers Island, Brown Skuas involved birds whose former mates had formed new have established a breeding ground that is stable in pairs. Of the 12 birds who had lost their previous sea- numbers of nest sites and breeding pairs. By monop- sons’ mate, eight did not have mates for at least 1 year; olizing the penguin colonies as a food source, Brown four found new mates but produced no eggs. Skuas of Anvers Island compete successfullyagainst Despite the number of mate changesobserved, mate the greater numbers of South Polar Skuas that nest fidelity was strong. Of the nine males banded the first close-by but forage almost exclusively at sea. Judging year of the study, 67% had only one mate during 9 by the large numbers of Brown Skua fledglings pro- years of observations, and the rest had two. Of 10 duced on Litchfield and Humble Islands, and the fact females banded the first year of the study, 30% had that thoseislands had the region’s largestconcentration one and 50% had two mates in 9 years, probably re- of nesting South Polar Skuas,we find little correlation flecting the greaterresidence time of females at Palmer between numbers of South Polar Skuas and fledging (Table 1). No male was seenin more than five breeding successin Brown Skuas.Triveloiece et al. (1980) also seasons,whereas four of 10 females were seen in all 9 witnessedhigh fledging success-ofBrown Skuasin the years(median of distribution of number of years sight- presenceof South Polar Skuas,although the latter were ed was higher for females; Smimov Tz = 5, P = 0.05). not nearly so numerous in that study area. The sex differencein return rate affectedsite tenacity During our study we observed the re-occupation of as well. The number of years a bird returned to the vacated territories following a severe epidemic that Palmer area obviously limited the number of years it decimated the Brown Skuas but had little effect on could return to the same territory (Table 1). Of 50 South Polar Skuas.Young (1972) documentedthe way banded Brown Skuas, 10 females returned to the same new skua pairs became establishedin a vacant area in territories long enough to successivelyacquire second a breedingcolony. He concludedthat the stabilityfound and, in two cases,third Brown Skua mates. Only five in territory boundariesfrom year to year arisesbecause males acquired secondmates while retaining their ter- territories becoming vacant are occupied by replace- ritories. ment pairs while neighborsare still inhibited from en- In February 1979 Brown Skuasat Palmer suffereda croachingon them. Our observationsindicate that un- fowl cholera epidemic (Parmelee et al. 1979) which der certain conditions the integrity of skua boundaries left severaltraditional nest sitesvacant for one or more may persist for one or more seasonseven when un- seasons.Three sites were empty at least two seasons. occupied. When next checked in 1983-1984, these sites were As missing mates were replaced and vacant teni- occupied by Brown or South Polar skuas, including tories filled, the total number of Brown Skuas occu- new birds and survivors of the epidemic. One site was pying established territories rose to 40% above the SHORT COMMUNICATIONS 919 minimum two birds per site (at least 50 adults at 18 phallus deliciosusOlsson 1893). Antarct. J. U.S. sites in 9 years). The fact that new birds entering the 19(5):159-160. area usedpreviously established territories suggests that NEILSON,D. R. 1983. Ecologicaland behavioral as- suchsites are somehow advantageous.Incoming skuas pectsof the sympatric breeding of the South Polar may be attracted to these sites becausetheir locations Skua (Cutharucta maccormicki) and the Brown provide proximity to ample food and potential mates. Skua (Catharacta lonnbergi) near the Antarctic During one seasonat Chatham Islands, New Zea- Peninsula. M.S.thesis, Univ. of Minnesota, Min- land, Young (1978) reported that five of 11 Brown Skua neapolis-St. Paul. territories contained three birds (trios). The formation PARMELEE,D. F. 1985. Polar adaptationsin the South of trios was rare at Anvers Island (four instances,re- Polar Skua (Catharacta maccormicki) and the sulting in one fledgling in 9 years), as in other parts of Brown Skua (Catharacta lonnbergi)of Anvers Is- the species range distant from the Chatham Islands land, Antarctica. Proc. XVIII Int. Omithol. Congr. (Young 1978). Becausetrios are common in the New (1982):520-529. Zealand area, one suspectsthat they are advantageous PARMELEE,D. F., AND C. C. RIMMER. 1985. Omi- under certain conditions. thologicalobservations at Brabant Island, Antarc- Male skuasgenerally arrive on the breedinggrounds tica. Br. Antarct. Surv. Bull. 67:7-12. before females (Neilson 1983) suggestingthat males PARMELEE,D. F., N. BERNSTEIN,AND D. R. NEILSON. establishterritories to which they attract mates. How- 1978. Impact of unfavorable ice conditions on ever, in our study many females returned to and re- bird productivity at Palmer Station during the mained at a nest site even when they had lost their 1977-78 field season.Antarct. J. U.S. 13(4):146-~ , mates. Most of these females later bred at the same 147. site with new mates, indicating that females can hold PARMELEE,D. F, W. R. FRASER,AND D. R. NEILSON. a territory and attract a male to it. 1977. Birds of the Palmer Station area. Antarct. The low percentof returning Brown Skuachicks sug- J. U.S. 12(1-2):14-21. gestsa high mortality of youngfollowing fledging.Great PARMELEE,D. F., S. J. MAXSON,AND N. BERNSTEIN. Skuas (C. skua) also suffer high mortality in the first 1979. Fowl choleraoutbreak among Brown Skuas. year (Fumess 1978). Only additional observationswill Antarct. J. U.S. 14(5):168-169. determine whether the current rate of returning birds PARMELEE,D. F., J. M. PARMELEE,AND M. FULLER. is sufficientto maintain population stability at Palmer. 1985. Ornithological investigations at Palmer Station:The first long-distancetracking of seabirds This study was financed by the National Science by satellites.Antarct. J. U.S. 19(5):162-163. Foundation Grant DPP82-13688. For field assistance PIETZ,P. J. 1984. Aspects of the behavioral ecology we are indebted to the following at or formerly from of sympatric South Polar and Brown skuas near the University of Minnesota: N. P. Bernstein, W. R. Palmer Station, Antarctica. Ph.D.diss., University Fraser, B. Glass, G-A. D. Maxson, S. J. Maxson, D. of Minnesota, Minneapolis-St. Paul. R. Neilson, J. M. Parmelee, C. C. Rimmer; also M. PIETZ, P. J. 1984. Long call displays of sympatric Bergaas,S. Elvig, T. English, C. Franke, and R. Payne South Polar and Brown skuas. Condor 87:3 16- for office and technicalassistance. We also are indebted 326. to the referees for helpful suggestionsand a critical PIETZ, P. J. 1986. Daily activity patterns of South review of the manuscript. M. Fuller, Patuxent Wildlife Polar and Brown skuasnear Palmer Station, Ant- Research Center, and G. Jonkel, Chief Bird Banding arctica. Auk 103:726-736. Laboratory, Laurel, Maryland, provided invaluable as- PIETZ, P. J. 1987. Feeding and nesting ecology of sistancewith the Antarctic banding program. Special sympatricSouth Polar and Brown skuas.Auk 104: thanks to G. A. Llano and F. Williamson, formerly of 617-627. the National Science Foundation, who assistedus in TRIVELPIECE,W., ANDN. J. VOLKMAN. 1982. Feeding various ways throughout the study. strategiesof sympatric South Polar SkuasCutha- racta maccormickiand Brown SkuasC. lijnnbergi. LITERATURE CITED Ibis 124:50-54. TRIVELPIECE,W., R. G. BUTLER,AND N. J. VOLKMAN. ATNLEY,D. G. 1981. Population studies of South 1980. Feeding territories of Brown Skuas Ca- Polar Skuas.Antarct. J. U.S. 16(5):148. tharacta lonnbergi.Auk 971669-676. FURNESS,R. 1978. Movements and mortality rates YOUNG,E. C. 1972. Territory establishmentand sta- of Great Skuasringed in Scotland. Bird Study 25: bility in McCormick’s Skua. Ibis 114:234-244. 229-238. YOUNG,E. C. 1978. Behaviouralecology of lonnbergi HOBERG,E. P. 1984. Trematode parasitesof marine skuasin relation to environment at the Chatham birds in Antarctica: the distribution of (Gymno- Islands, New Zealand. N.Z. J. Zool. 5401-416.