Abstract:
Heaviside's dolphin (Cephalorhynchus heavisidii) is a poorly studied coastal delphinid with a limited inshore distribution off the west coast of southern Africa where it is sympatric with the similar sized dusky dolphin (Lagenorhynchus obscurus). It is exposed to an unknown level of bycatch particularly in near-shore set-net fisheries and is also potentially impacted by the growing boat-based whale watching industry in South Africa. In this thesis I describe the results of a study investigating the distribution, movements, behaviour and abundance of Heaviside's dolphins in the near-shore environment of the Western Cape of South Africa as a precursor to assessing its potential vulnerability to anthropogenic threats. Data were gathered using three different approaches in the field; diurnal shore based observations, boat based photo-ID surveys along ~390 km of coastline and satellite telemetry. Data were collected for dusky dolphins where feasible. Heaviside’s dolphins exhibited a strong pattern of resting inshore during daytime and foraging offshore at night that was presumed to be related to the movement of juvenile hake (Merluccius capensis) closer to the surface at night. In addition, despite near-shore observations failing to indicate feeding, dolphins were consistently found to be more abundant along regions of the coast which over the long term had higher levels of small hake available offshore. In contrast the near-shore distribution of dusky dolphins varied considerably between years possibly due to the very near-shore environment being at the edge of their habitat. A tendency for dusky dolphins to move offshore during upwelling conditions was observed from shore in St Helena Bay, the site of a strong predictable upwelling cell. However this was the only location in which very large groups of animals (50-500) were seen during coastal boat surveys suggesting this pattern may have been area specific and feeding strategies may vary throughout their range. Sympatric appears to be mediated by differences in overall range and prey type and size differentiation. Heaviside’s dolphins were found to have small home ranges and show a high degree of site fidelity over several years, at least during summer months. Dolphins fitted with satellitelinked transmitters used only limited home ranges (~876 to 1990 km2) which scaled positively with body size within the ~50 day tracking period, while photographically identified animals did not disperse significantly further than this over 3 years. The maximum coastwise displacement observed for an individual (88.4 km after 1 year) was considerably less than the 390 km length of the overall study area, and equivalent to the maximum coastwise movement seen during satellite-tagging (83.1 km). This suggests that measured home ranges may be stable over several years, although the number of resighted individuals (n = 76) was small and (as data collection was restricted to summer months) seasonal differences in movements or migrations cannot be ruled out. Associations among photographically identified animals did not differ from a random mixing of individuals, suggesting that this species has a fission-fusion type social structure at least over the short term. The abundance of Heaviside's dolphins was calculated from photo-ID mark-recapture data at three spatial scales using Chapman’s modified Petersen estimators. The proportion ofdistinctively marked individuals in this species is low (14 - 17%) reducing sample sizes and introducing a large extrapolation factor, both of which contribute to an increase in the variance of any resultant population estimate. Using resightings after one year, the total number of animals over the whole 390km study area was calculated to be 6345 (CV = 0.26, CI = 3573 – 11 267) while using same-season re-sightings the total number of animals using a 20km long section of coastline in the centre of the study area was calculated as 527 (CV = 0.35, CI = 272 – 1020). These estimates may be significantly biased downwards by the effects of heterogeneity in the capture probability of individuals which it was not possible to account for analytically due to small sample sizes. Sympatry of Heaviside's and dusky dolphins appears to be mediated by differences in the overall range and in the type and size of prey consumed. From a conservation point of view, our findings for Heaviside's dolphins are encouraging in that they indicate a relatively large population size, although their strong site fidelity does make them more vulnerable to localised impacts than a more widely ranging species like dusky dolphins and this needs to be considered during management of the population.