Aspects of the ecology of Weddell seals at the Vestfold Hills, Prydz Bay, East Antarctica

At first glance, the head of a Weddell seal, Leptonychotes weddelli, seems small for its body. This is due to the streamline profile for swimming and breathing through holes in the ice. Ice frozen, or 'fastened', to the shore (fast-ice) surrounds the coast of Antarctica and is the Weddell seal's habitat (Kooyman 1968). Weddell seal females pup for the first time, on average, at seven years and can live to 24 years producing pups approximately annually until near the end of that time (Green et al. 1993). The number and timing of pups is related to the environment which changes cyclically (Testa et al. 1991, White and Peterson 1996). Interannual variation is currently being investigated by monitoring the population of Weddell seals at Vestfold Hills, Prydz Bay, East Antarctica. A diurnal cycle in Weddell seal haul-out behaviour has been described at McMurdo Sound, Antarctica, but prior to this thesis regional and seasonal variations in haul-out patterns of Weddell seals were unknown. Knowledge of activity patterns was required to standardise survey methods and survey data. This study examined diurnal haul-out behaviour of Weddell seals at the Vestfold Hills between October 1994 and March 1995. Sequential counts of seals on the ice showed that seal abundance differed up to 95% between 0930 and 1930 hours. Fewer seals were on the ice in the morning compared to the afternoon. Maximum numbers of seals were on the ice at the warmest time of day. The diurnal cycle was less pronounced in the breeding season than in the moulting season. The findings indicated the importance of surveying Weddell seals after 1200 hours and before 1700 hours local time especially in the moulting season. Correction factors were given for month and time of day. The breeding habitat of Weddell seals was investigated for information relevant to estimating Weddell seal population size by stratified surveys. The distribution of seals in the breeding season (September - November) was described for Long Fjord which was probably the most concentrated and consistent breeding site of Weddell seals in Prydz Bay, East Antarctica. The study area was `40` `km^2` from the most east to most west breeding sites in the fjord. The 18 years of seal data included sightings of 3 205 adults and 2 319 pups. Seals were not randomly distributed. They mainly occupied one of four channels in the fjord. The location of breeding sites varied, probably due to interannual variation in ice-structure. Original methods to quantify ice-structure were used to investigate whether icestructure accounted for the non-random distribution of seals as well as interannual variation in location of breeding sites. Data were preliminary but they suggested that the distribution of seals was not determined by ice-structure. Other aspects of habitat were discussed to emphasise the types of data required to describe the components of the breeding habitat of Weddell seals. The diet of Weddell seals at the Vestfold Hills has been sampled every summer in the 1990's as part of the population monitoring program. The aim has been to describe interannual variation in diet concurrent with changes in population demographics. The study in this thesis was of variation in Weddell seal diet within years so that, if present, the variation can be controlled when making interannual comparisons of diet. In this study, Weddell seal faecal samples (n = 216) were collected from various areas at the Vestfold Hills. The collections were bi-monthly between October 1994 and February 1995. The most common prey by frequency of occurrence were benthic fish (71% ) and prawns (58% ). Cephalopods (12%) were a minor component of diet in the breeding season (October - December). Pelagic fish (7%) were a minor component of the diet outside the fjords and in the moulting season (February). For the first time of any Antarctic predator Trematomus scotti, a common benthic fish, was detected in the diet. There was significant variatiom in the frequency of occurrence of the dominant prey types, both within Long Fjord and between Long Fjord and other sites. This indicated that it was necessary to control for temporal and spatial variation in Weddell seal diet to achieve precise interannual comparisons. It is far from a trivial matter to determine changes in pinniped populations. The number of seals on top of the ice varies considerably, and if counts of seals are to be used as an index of population size then factors that affect the proportions of seals on the ice and in the water need to be controlled or corrected for. This thesis reports diurnal variation in numbers of seals and how the diurnal pattern changes seasonally (Ch. 2), but there are other variables that also need to be considered such as tidal and lunar cycles. The ultimate effect of changes in population parameters is a population increase or decrease. Surveys of population size need to be on a regional scale and Weddell seals are not randomly distributed (Ch. 3). It would be more efficient to know where seals are abundant and survey those areas intensely. At this stage, aspects of habitat that determine Weddell seal abundance are not known and more research is required to be able to predict the locations of concentrations of Weddell seals. Interannual variation in diet provides useful information for inferring the reasons for population fluctuations and modelling potential impacts of man. Methods for collecting samples to represent Weddell seal diet accurately are discussed in Chapter 4. This will assist comparisons of diet between years in showing real changes in the Prydz Bay ecosystem and the impact on Weddell seals as top predators in that ecosystem