The at-sea behaviour of intermittently breeding adult female southern elephant seals from Marion Island

Dissertation (MSc (Zoology))--University of Pretoria, 2023.Southern elephant seals (SES), Mirounga leonina, are large capital breeding marine predators that have a circumpolar distribution in the Southern Ocean. Female SES typically give birth to a single pup each year, however, recent studies from subantarctic Marion Island and Macquarie Island have identified adult female SES that skip reproductive events. Previous studies on intermittently breeding elephant seals focused on theoretical reasons for this behaviour using demographic data. Few studies have empirically analysed the at-sea behaviour of female SES when they skip a breeding event, and how this behaviour may relate to body mass and environmental conditions. This dissertation focuses on the at-sea behaviour of intermittently breeding adult female SES from Marion Island and how this behaviour may be described relative to their foraging patterns, body condition and environmental variables. Capital breeders need to acquire sufficient energy prior to breeding to support maternal maintenance and lactation costs. Variation in maternal body mass depends on environmental conditions and individual heterogeneity in foraging ability. Females should allocate more energy to their offspring when environmental conditions create profitable foraging potential and should prioritise their own survival in sub-optimal foraging years. Body mass is a proxy for body condition, and intermittent breeding may be an adaptive reproductive trajectory where females in sub-optimal condition skip breeding for a year and spend more time foraging to improve their energy reserves. This has the potential to increase maternal and offspring fitness compared to continually breeding females that may breed in sub-optimal condition. Three multiparous female reproductive skippers were identified from the Marion Island SES breeding population that were repeatedly tracked during skipping and breeding years from 2007 and 2012 (GG335, n = 5 tracks; WW061, n = 3 tracks; YY039, n = 2 tracks). Each female skipped reproduction once, and each gave birth in the year following skipped reproduction. Each female was fitted with a satellite telemetry device that recorded at-sea location and their daily dive duration, dive depth, and dive frequency. All three females had tracking data for their skipped years, and other breeding years when data were opportunistically collected. Using generalized linear models in the programme R, I determined that, overall, skippers dived deeper, longer, and less often than breeders. Individual heterogeneity was observed amongst the skipper females where they each showed markedly different diving behaviour between their skipped and breeding years. Apart from individual heterogeneity, environmental conditions also influence the foraging potential and breeding probabilities of SES. I aimed to describe and compare the environmental variables, net primary productivity (NPP) and bathymetry, that the multiparous female SES from Marion Island encountered at-sea during skipping and non-skipping years. I further investigated whether foraging potential, as reflected by body mass, may have been associated with environmental conditions. All three female SES travelled close to areas of frontal-bathymetric profitability during their skipped years, where the confluence of bathymetry and ocean currents stimulates increased primary production and foraging potential. Good foraging potential and increased time at-sea were reflected in the body masses of females WW061 and YY039 that were heavier in the years following skipped reproduction. There was no body mass data for GG335 in the year after she skipped, however, she consistently changed her migration route from her skipped year onwards. Moreover, all three females hauled out to moult at least four weeks earlier than their breeding counterparts during the year that they skipped. This study highlights the importance of longitudinal data collection, long-term mark-recapture programmes, and individual heterogeneity within the Marion Island SES population. Although only a small subpopulation of SES was considered in this work, the findings of this study are relevant to other wide-ranging marine predators that may exhibit intermittent breeding. Small sample sizes are statistically challenging to analyse; however, the results can be valuable additions to broader scientific contexts such as life history theory and optimal foraging theory in capital breeders. This work offers theoretical and methodological insights that may aid further studies on movement and behavioural ecology, especially those concerned with small sample sizes and rarely observed animal behaviour.National Research Foundation of South Africa (Grant number: MND200418514570)Zoology and EntomologyMSc (Zoology)Unrestricte