Foraging strategies of Adelie penguins at Bechervaise Island

A detailed multi-year analysis of the breeding biology, diet and foraging strategies of Adelie penguins (Pygoscelis adeliae) from the Bechervaise Island colony in East Antarctica was carried out in relation to gender and environmental conditions such as sea ice extent. Data on breeding success and foraging activities (location, trip duration, and diving behaviour) were collected during the 1994-95, 1995-96 and 1996-97 austral summers along with dietary samples. Analyses included data and samples from four previous seasons in addition to the new material. Several hundred penguins of known sex within the colony were individually identified by means of implanted electronic tags to allow automated detection of their travel to and from their nest sites. Multi-year data on the nest and mate fidelity and breeding success of these birds were available for analysis. The breeding success and variations in foraging strategies (trip duration, diet, dive depth, and foraging location) of tagged individuals were analysed in detail over a number of seasons to study the ways in which these variables change from year to year and to determine how the variations relate to environmental factors. Subcutaneously implanted transponders have proved to be a reliable means of identifying individual penguins, and the usefulness of these passive electronic tags was evaluated as the research project progressed. Survival of Adelie penguins carrying transponders over seven seasons was found to be equal to or better than that of birds with flipper bands, although not statistically significant on an annual basis. Survivorship of fledglings tagged as chicks was greater than that determined by previous researchers using banded populations of fledglings. Occasional problems associated with the use of implanted transponders were observed. The transponder removed from one bird had developed a slimy biofilm harbouring potentially pathogenic organisms incorporated at the time of implantation. If such contamination was to be common the long-term survival of groups of birds carrying implanted transponders could be lowered below that of unmarked populations. Migration of transponders away from the injection site into potentially hazardous locations, which might jeopardise survival in some individuals, was also demonstrated. Such risk factors could limit the use of implanted identification devices in long-lived or endangered species. However, introduction of bacteria can be minimised by careful injection techniques and cleansing of instruments and skin with iodine or alcohol. The choice of a suitable implantation site, such as midway down the back, from which transponders may migrate without impinging upon vital structures, is also important. It was concluded that transponders, when used with care, provide a viable alternative to flipper bands in demographic studies of penguins. Consistent sex differences in foraging trip duration, feeding locality and diet of breeding Adelie penguins were demonstrated at two widely separated locations over several breeding seasons. Differences in foraging behaviour were most pronounced during the guard stage of chick rearing. Female penguins made on average longer foraging trips than males, ranged greater distances more frequently and consumed larger quantities of krill (Euphausia superba). In contrast, males made shorter journeys to closer foraging grounds during the guard period and fed more extensively on fish throughout chick rearing. Mean guard stage foraging trip durations over four seasons at Bechervaise Island, East Antarctica and over two seasons at Edmonson Point, Ross Sea ranged between 31-73 hr for females and 25-36 hr for males. Ninety percent of males tracked from Bechervaise Island by satellite during the first three weeks post-hatch foraged within 20 km of the colony, while the majority (60%) of females travelled to the edge of the continental shelf 80--120 km from the colony to feed during this period. The overall composition of the Adelie penguin diet varied between seasons at both locations. However, krill comprised a greater proportion of the diet of female penguins than that of male birds at each colony during the guard period. Males on the other hand tended to eat greater amounts of fish and amphipods than did females. These sex differences, although consistent from year to year, were however not statistically significant. Analyses of the body masses of males and females departing on foraging trips of long and short duration (> and <40 hr respectively) showed that the departure weights of birds prior to long trips were significantly lighter than were those prior to short trips. Birds, particularly males, were heavier at the start of the guard stage than at the end, and both sexes gained weight slightly over the crèche period. The observed gender differences in trip duration, foraging location and diet are discussed in terms of energy requirements, intraspecific partitioning of foraging and territorial behaviour. The existence of a two-fold foraging strategy due to a trade-off between the allocation of food to chicks and the storage of parental body reserves is postulated. The relevance of such a foraging strategy to the breeding success of penguins in the Mawson region is discussed in relation to krill, fish and zooplankton distribution in the area. Variations in the diving patterns of 18 Adelie penguins rearing chicks were examined over two breeding seasons in relation to foraging trip duration, gender and change in fast ice extent. Mean dive depth, duration and percent bottom time overall were 18.5 m, 1.02 min and 26.4% respectively. Maximal depths reached on individual trips ranged from 14 to 112 m. Twenty-seven percent of all dives were to depths <5 m; 54% were <10 m; 68% <20 m and only 11% >50 m. No diurnal patterns of diving frequency or depth were apparent. Penguins making short trips (<40 hr) rarely dived deeper than 50 m, with 91% of all dives <40 m in depth. Most short trips took place during the guard stage when fast ice was present around the island. Long trips (>40 hr) occurred during both the guard and crèche stages, and showed a bimodal pattern of dive depths with the secondary peak occurring at 40-70 m. Mean dive depth, duration and percent bottom time all increased as the ice extent reduced. The birds, particularly males, showed a tendency to make shorter trips and perform shorter, shallower dives when fast ice was present than when there was open water. Male birds undertaking short trips showed significantly greater variability in mean dive depths and durations than did females on short trips or birds of either sex on long trips. Penguins of both sexes making long trips spent on average approximately fifty percent more time on the bottom than did those making short trips. Logistic regression analysis identified percentage of time on the bottom as the strongest indicator for discrimination between trips of long and short duration. No predictor variables were identified that could be used to discriminate between the trips of males and females. The extensive foraging range of Adelie penguins in the Mawson region enables birds to forage in three distinct oceanographic zones each dominated by different micronekton communities. Most birds ingested larger quantities of krill (E. superba) on long trips than short, while consumption of fish was more common on short trips. The relationship between diet composition and diving behaviour on trips of different duration is discussed. Changes in diving behaviour as environmental conditions alter may reflect the capacity of penguins to modify their foraging strategies in response to spatial and temporal variations in the distribution of their prey. The results of the studies described in this thesis provide some new insights into the foraging ecology of Adelie penguins in the Mawson region of East Antarctica. These may help facilitate the modification and improvement of existing ecosystem-based monitoring programs such as the Convention for the Conservation of Marine Living Resources (CCAMLR) Ecosystem Monitoring Program (CEMP), and also aid the interpretation of data from such programs.