Diet and feeding ecology of the diving petrels Pelecanoides georgicus and P. urinatrix at South Georgia

 The diet of the diving petrels Pelecanoides georgicus and P. urinatrix was studied during 1986 (P. georgicus) and 1987 (both species) by lavaging adults as they returned to feed chicks on Bird Island, South Georgia. The diet of both species was dominated by crustaceans, in particular euphausiids (mainly Euphausia superba and some Thysanoessa), which contributed 47–76% of the biomass of crustaceans in the diet of P. georgicus, and copepods, which contributed 71% of the biomass of crustaceans in the diet of P. urinatrix. Calanoides acutus was the most numerous copepod in the diet of both species; however, Rhincalanus gigas was more common in P. urinatrix than in P. georgicus. The dominant amphipod in the diet of P. georgicus, Primno macropa, was absent from the diet of Pelecanoides urinatrix, in which Themisto gaudichaudii (rare in Pelecanoides georgicus) dominated. Dietary differences were maintained in the period (2 weeks of a total of 10 weeks) when both species were simultaneously rearing chicks. Knowledge of the prey species and of the diving abilities and foraging habits of diving petrels suggests that at South Georgia Pelecanoides urinatrix feeds closer inshore and dives deeper than Pelecarnoides georgicus. Received: 24 August 1995/Accepted: 10 February 1996     

Development of homeothermy in the diving petrels Pelecanoides urinatrix exsul and P. georgicus, and the Antarctic prion Pachyptila desolata

Body temperatures of South Georgia diving petrel (P. georgicus) chicks increased from about 37.5 degrees C at hatching to between 38.5 and 39 degrees C within two weeks. Temperatures of common diving petrel P. u. exsul chicks averaged 38.8 degrees C after two weeks of age. Burrow temperatures varied between 5 and 10 degrees C. Measurements of oxygen consumption and body temperature indicated that chicks achieve effective endothermy at 5 degrees C after 9 days in P. u. exsul, 5-6 days in P. georgicus, and 0 days in the Antarctic prion (Pachyptila desolata). The maximum mass-specific, cold-induced oxygen consumption of small chicks that we could measure with our apparatus (ca. 5-6 cc O2/g per hr) was achieved at 5-6 days in P. u. exsul, 3 days in P. georgicus, and 0 days in P. desolata. Mass-specific thermal conductance decreased with age and body size in all 3 species, but was highest in P. u. exsul and lowest in P. desolata. Conductance was similar at the age of effective endothermy in all 3 species (ca. 3 J/g per hr per degrees C). The period required for the development of endothermy is related to age-specific changes in both conductance and capacity for heat production and it closely parallels the length of the brooding period. It is suggested that the length of the period of thermal dependence of the chick is related to the distance between feeding areas and the nesting site.     

The breeding biology of the white-chinned petrel Procellaria aequinoctialis at South Georgia

Aspects of the breeding biology of the whitc-chinned petrel Procellaria aequinoctialis were studied at South Georgia. Laying, hatching and fledging dates, together with egg and adult measurements, rate of weight loss of eggs and incubating adults and data on the duration of incubation shifts were recorded. Chick growth in weight, tarsus length, culmen length and wing length from hatching to fledging was followed in detail and breeding success and causes of mortality documented. Comparisons are made with other populations of the species, other members of the genus Procellaria and other Procellariiformes. Despite being the largest burrow-nesting petrel in the world, P. aequinoctialis conforms well to the overall relationships between body weight and breeding parameters for Procellariiformes in general.     

Moult of the giant petrels Macronectes halli and M. giganteus at South Georgia

Moult scores were collected from colour-ringed individuals of known reproductive status of the two species of giant petrel, Macronectes halli and M. giganteus , at Bird Island, South Georgia between 1978–81.
Both species showed a substantial overlap between breeding and wing-moult, unlike most other Southern Ocean seabirds. Males started moult before females and both sexes of M. giganteus moulted at an earlier stage of the breeding cycle than M. halli , which breeds six weeks earlier than its congener.
Changes in moult rate during the breeding season are documented for both species, with Id. halli showing a rapid increase as the chick nears fledging. Male M. giganteus show a notably different pattern to the other three species-sex groups, starting moult much earlier (at egg-laying), with greater individual synchrony and usually suspending primary moult throughout the main chick growth period, whereas only two male M. halli and no females of either species suspended moult. Differences in pattern, timing and rate of moult are interpreted in terms of availability of food resources and the competing energy demands of other activities, especially chick-rearing.
Completion of primary moult could not be observed in the field but was estimated using data frcsm non-breeding birds and failed breeders; the latter started a rapid moult almost immediately they failed. In both sexes of both species moult is probably concluded at least by early winter.
The general pattern of moult in giant petrels at Bird Island is contrasted with that of other populations and species of Southern Ocean seabirds. It is suggested that the unusually extensive overlap between breeding and moult in giant petrels is a consequence of the very abundant and easily available summer food supplies (especially carrion) and the much diminished winter resources, favouring a completion of moult by the beginning of the winter.     

The food and feeding ecology of the giant petrels Macronectes halli and M. giganteus at South Georgia

Giant petrels are the chief scavenging seabirds in the Southern Ocean. Quantitatively analysed regurgitations from chicks of both Macronectes halli and M. giganteus at Bird Island, South Georgia throughout the chick rearing period consisted predominantly of adult Macaroni penguins Eudyptes chrysolophus. Significant quantities of burrowing petrels and krill were taken but the amount varied in different years. Only male M. halli fed regularly at fur seal carcases. Although 10 species were found in regurgitations, squid were unimportant in the diet. Fish are possibly more important during the winter.
Intersexual differences in diet were of greater significance than interspecific ones with males taking more carrion and females more crustaceans. Chicks were fed during both day and night and in both species males received significantly more food than females during the fledging period. M. giganteus chicks received more than M. halli at all ages.
Despite their extensive reliance on carrion both species take a variety of prey and this may be especially important during the winter when penguin and seal carrion is scarce.     

Annual variation in breeding biology of macaroni penguins, Eudyptes chvysolophus, at Bird Island, South Georgia

The breeding biology of the macaroni penguin, Eudyptes chrysolophus , was studied over four years, 1976 and 1986–88, at Bird Island, South Georgia. Birds were migratory, being absent during winter (May to September). Arrival at the colony was highly synchronous between years: 14–23 October, over a 7-year period. The pre-breeding, incubation and chick-brooding period was characterized by long fasts ashore, for 36 and 39 days in males and 41 days in females, alternating with long periods at sea. Within years egg-laying was highly synchronous: 95% of clutches initiated within 4–6 days. Arrival date and mean egg-laying date were later (by 3 days), and breeding population size lower (by 20%) in 1987, compared to other years. The incubation period was 35 days and comprised three long shifts, the first shared by the male and female, the second by the female and the third by the male. In 1986 and 1988 these were of 12, 12 and 9 days' duration, but in 1987 the first shift was significantly shorter: 9 days. Chicks creched at 23–25 days of age and fledged at 60 days of age. Neither chick age nor weight at creching or fledging varied between the years 1986–88. The breeding biology of macaroni penguins at Bird Island is compared with that of other Eudyptes penguins, and with the sympatric gentoo penguin, Pygoscelispupuu. There is little variation in breeding biology within the genus Eudyptes , except in the length of time spent at sea prior to the annual moult. This is much shorter at Bird Island, probably reflecting a greater food availability compared to other localities. Inter-annual variation in certain breeding parameters, e.g. laying date, breeding population size, is much greater in the gentoo penguin than in the macaroni penguin. The shorter breeding season, rearing of only one chick and proportionately lower chick fledging weight in macaroni penguins, may be linked to this species' migratory strategy.     

The use of acoustics to monitor burrow-nesting white-chinned petrels Procellaria aequinoctialis at Bird Island, South Georgia

Tape recordings of two types of vocalisations were used to assess burrow occupancy by white-chinned petrels (Procellaria aequinoctialis) at South Georgia. Birds responded to the rattle call on 74% of occasions and a wheezy call on 85%, with only 10% failing to respond to both calls played one after the other. Occupancy was determined throughout the pre-egg and incubation period and results confirmed what is known about patterns of occupancy by white-chinned petrels. Adults in burrows known to hatch a chick responded on average to 69% of recordings during incubation and less than 10% in only 4% of burrows. Significant inter-site and inter-annual variations in occupancy were recorded, suggesting that this method is sensitive to changes in the number of petrels returning to breed. A sampling protocol for providing population indices for monitoring purposes is recommended. Accepted: 6 March 2000     

The diet of Antarctic fur sealsArctocephalus gazella during the breeding season at South Georgia

 The diet of lactating female Antarctic fur seals Arctocephalus gazella at South Georgia was investigated during the pup-rearing period (January/March) of 1991–1994. Antarctic krill Euphausia superba was the main prey item, occurring in 88% of all scats (n=497), whereas fish occurred in 47% and squid in 5% of all scats. There was considerable intra- and inter-annual variation in the characteristics of krill taken by fur seals. The distribution of krill sizes taken suggests that fur seals are not actively selecting particular sizes of krill and, therefore, that the krill in the diet reflects the krill available around South Georgia. The absence of group 3 krill (44–48 mm in length) in the South Georgia area, as indicated by their absence in the diet of seals, is suggested as a possible reason for low availability of krill and the subsequent reproductive failure among krill predators. The frequency of occurrence of fish was much higher than in previous studies; the pattern of fish consumption showed a consistent seasonal pattern in 3 of the 4 years studied. Of the total number of the myctophid Protomyctophum choriodon, the most numerous fish taxon, 98% were taken between early February and the middle of March. Champsocephalus gunnari and Lepidonotothen larseni agg., which both feed on krill, dominated the fish component of the diet outside this period and together constituted 94% of the total estimated biomass of fish consumed. The intra- and inter-annual variability in the diet of Antarctic fur seals emphasise the need for diet studies to be conducted during the entire pup-rearing periods of several years. Received: 15 March 1995/Accepted: 13 May 1995     

Determination of the absolute visual threshold of a nocturnal seabird, the Common Diving Petrel Pelecanoides urinatrix

Many of the smaller burrowing petrels are active at their colonies during the night. How they find their nesting burrows in the open terrain of an oceanic island or in the even more visually-taxing gloom under a forest canopy (Grubb 1974 and references therein) is a feat which has long intrigued naturalists (Lockley 1942). There is experimental evidence that vision is important in burrow location by the Manx Shearwater Puffinus puffinus (Brooke 1978, James 1986), while olfactory homing by Leach's Petrel Oceanodroma leucorhoa has also been suggested by Grubb (1974).
Additionally it seems likely that a number of petrel species catch a significant fraction of their food at night (Imber 1973, Prince & Francis 1984). Many of the squid presumed to be caught then are not bioluminescent (Imber 1973, Clarke et al. 1981), so their detection presumably demands good night vision. In the context of these observations, the dearth of information on the visual abilities of petrels is remarkable.
The present study investigated the absolute visual threshold of the Common Diving Petrel Pelecanoides urinatrix. Probably because the species is small (c. 130 g) and so vulnerable to predators such as Sub-antarctic Skuas Catharacta antarctica , it is strictly nocturnal at breeding colonies (Thoresen 1969, Payne & Prince 1979). It nests in burrows that are often located under tussock grass Poa spp. Whether the birds catch their crustacean prey (Payne & Prince 1979) by day or night is not known.     

Breeding biology of the Light-mantled sooty albatross (Phoebetria palpebrata) at South Georgia

Aspects of the breeding biology of the Light-mantled sooty albatross Phoebetria palpebrata were studied at South Georgia. Its breeding distribution there is summarized and an annual breeding population of c. 5000 breeding pairs estiamted. Laying, hatching and fledgine dates are given, together with egg measurements and data on the duration of incubation and brooding shift. Chick growth in weight and wing length from hatching to fledging was followed in detail.
Breeding season adaptations of P. palpebrata are compared with those of the other three albatrosses at South Georgia. The slow chick growth of P. palpebrata results in a nestling period even longer than those of the two larger mollymawks and is attributable mainly to the lower frequency of feeding resulting from the highly pelagic existence of adult P. palpebrata .
Comparisons are made with other breeding populations and especially with that on Marion Island. This is basically very similar and differences in chick growth patters are suggested to relate to differences in diet and particulartly to the frequency with which chicks receive food.     

Fishery biology of the seven star flying squid Martialia hyadesi at South Georgia during winter

A scientific research fishing expedition targeting the oceanic/slope ommastrephid squid Martialia hyadesi was undertaken by a Korean-registered squid jigger in CCAMLR area 48.3, near South Georgia, in June 1996, providing the first opportunity to collect data on the fishery biology of this species during the austral winter. Fishing took place over a period of 8 days; a series of eight drifts was undertaken along an approximately east/west transect of about 200 nautical miles to the north and west of South Georgia, over depths ranging from 1,700 to 2,713?m. All fishing was to the south of the Antarctic Polar Front. Data were collected on sea surface temperature, catch per unit of effort, size, sex, maturity status and stomach contents of the catch and a sample of squid was aged by counting putative, daily microgrowth increments in the sectioned statolith. All squid were caught by jigs operating at depths from 80 to 100?m to the surface. Catch per unit of effort per drift varied between 1.0 and 21.9?kg?min^?1 and there was no by-catch. Greatest numbers of squid were caught at dusk and dawn. Mantle length fell in the range 220–350?mm (males) and 212–370?mm (females). Most males were sexually mature (Lipinski's stages IV–V) and most females were immature (stage II). The absence of mature females suggests that no spawning takes place in this area during the austral winter. The squid were up to 1?year of age and had hatched during the previous winter. They were apparently from the same cohort as had been sampled at the Antarctic Polar Front in February 1996. Myctophids were the major prey in the stomach contents and the squid Gonatus antarcticus was also important; crustaceans were relatively unimportant. The results suggest that concentrations of Martialia hyadesi are present in the vicinity of South Georgia, south of the Antarctic Polar Front, during the austral winter. The squid are actively feeding during the austral winter and are susceptible to jigging gear.     

The breeding biology of the gadfly petrels Pterodroma spp. of the Pitcairn Islands: characteristics, population sizes and controls

This paper reports the breeding biology and nesting seasons of the gadfly petrels which nest on the four islands of the Pitcairn group, Pitcairn, Henderson, Oeno and Ducie. The species currently breeding are Murphy's petrel Pterodroma ultima , Kermadec petrel P. neglecta , Herald petrel P. heraldica and Henderson petrel P. atrata. Of these, Murphy's petrel is the most numerous; an estimated 250000 pairs bred on Ducie, which is probably the major breeding station of the species. Novel basic breeding data for Murphy's petrel are presented. Incubation spells, averaging 19.3 days, are exceptionally long for a petrel. Phoenix petrel P. alba appears to have ceased to breed on the Pitcairn Islands since the 1922 surveys of the Whitney Expedition. Nesting success was low on Henderson Island during the study. For all four breeding species, less than 20% of eggs laid yielded fledglings. Failure occurred at the early chick stage and observations indicated that it was due to predation by Pacific rats Rattus exulans. Although rats are present on Ducie, predation was apparently less severe there. The situation on Oeno may be intermediate. I consider how the populations of Henderson are maintained in the face of this intense predation. The Murphy's petrel population may be sustained by immigration from Ducie while the Herald and Henderson petrel populations could be undergoing a long-term decline on Henderson. It is not clear how the Kermadec petrel population is maintained. The conservation implications of these findings are discussed.     

Diving behaviour and diet of the blue-eyed shag at South Georgia

Summary This paper describes a concurrent investigation of individual variation in diet, diving patterns and performance of blue-eyed shags Phalacrocorax atriceps breeding at South Georgia. Within one day individual shags exhibited one of three foraging strategies: short diving (4 birds, all dives 120 s) and mixed diving (15 birds, predominantly long but with a few short dives). The mean number of dives per day was significantly higher in shags that only made short dives (mean=172.0, SE=43.2) than birds with a mixed diving strategy (mean=40.5, SE=4.7) and birds that made only long dives (mean=30.8, SE=1.8). Diet was assessed using hard remains recovered from pellets regurgitated by the shags. Small nototheniid fish (c. 10 kJ per item) were by far the commonest prey but most pellets contained additional items. The frequency of pellets with additional items of higher energy value than nototheniid fish (10.c. 900 kJ per item), lower energy value (>1–10 kJ per item) and both higher and lower energy items was strikingly similar to the frequency of shags making long, short and both long and short dives respectively. Predicted aerobic dive limits suggested that during long dives, blue-eyed shags were probably sustained by anaerobic metabolism. Models of prey capture rates demonstrated that for both long and short diving, many items must be caught per dive when birds are feeding on prey at the lower end of the energy range. Predicted capture rates for the commonest recorded prey (small fish) differ markedly between the two diving strategies.     

The distribution and abundance of white-chinned petrels (Procellaria aequinoctialis) breeding at the sub-Antarctic Prince Edward Islands

The white-chinned petrel (Procellaria aequinoctialis) is the seabird most often killed on longlines in the Southern Ocean and is listed as vulnerable to extinction. We estimated the population breeding at the Prince Edward Islands, the last breeding site for the nominate subspecies that lacks a recent population estimate. White-chinned petrel burrows are largely confined to deep, muddy soils, usually on slopes below 200?m, but locally up to 420?m. After correcting for count bias, Marion Island has an estimated 29,900 nests (95?% CI 27,700–32,400). Burrow occupancy rates at the start of the incubation period were 65?% during one-off surveys, but repeat surveys found that at least 73?% of burrows were occupied and 87?% of burrows showed signs of occupancy. This suggests that there were roughly 24,000 occupied nests on Marion Island (95?% CI 20,000–28,000). A more cursory survey on Prince Edward Island yielded 14,700 burrows, suggesting that there are 9,000–15,000 occupied nests. The nominate subspecies of white-chinned petrel occupies approximately 974,200 nests (95?% CI 678,000–1,286,000), with the Prince Edward Islands, the third most important breeding site, after South Georgia and Kerguelen. Assuming that populations breeding at islands in the Atlantic and Indian Oceans winter in different regions, the impact of fishery bycatch is likely to have had a greater impact on the Indian Ocean population. The Marion Island survey provides a baseline against which future population changes can be assessed.     

Fish prey of the Wandering Albatross Diomedea exulans at South Georgia

Summary The fish diet (45% of total diet by weight) of Wandering Albatrosses rearing chicks at South Georgia during the austral winters of 1983 and 1984 was investigated using otoliths retrieved from regurgitations. These provide the first quantitative data for this species and for any albatross. By number of identified otoliths (32% could be identified only as ?Macrouridae and ?Moridae), Pseudochaenichthys georgianus (35%), Muraenolepis microps (33%) and Chaenocephalus aceratus (20%) predominated, with Notothenia gibberifrons, Pagothenia hansoni and Champsocephalus gunnari (together 12%) also present. Composition by weight (estimated from otolith length) of the main species was Pseudochaenichthys 51%, Muraenolepis 14%, Chaenocephalus 27%; if digestion and wear had reduced otoliths by 10% the values would be Pseudochaenichthys 54%, Chaenocephalus 25%, Muraenolepis 13%. Composition by weight (actual or corrected values) was almost identical between years but epipelagic fish were significantly more abundant in 1983 than 1984. All identified fish eaten by Wandering Albatrosses are common on the South Georgia continental shelf and most of them are caught in the commercial fishery there. However, two of the three main target species of this fishery in 1983–1984, Notothenia rossii and Champsocephalus were not, or rarely, caught by Wandering Albatrosses. It seems unlikely, therefore, that the albatrosses depend greatly on the fishery for acquisition of fish prey but how they catch several species, including Muraenolepis, which are mainly benthic in habit is unknown.