Abstract
Aspects of social structure, behavioural ecology, and population biology of Hector's dolphin (Cephalorhynchus hectori) were studied. Especially information on ranging behaviour and group composition of this species were scarce albeit important for its conservation. Fieldwork was carried out in seven study areas along the east and west coasts of the South Island of New Zealand between November 1993 and January 1998 from a small boat. Individual Hector's dolphin were sighted and resighted using photo-identification.
Habitat selection of Hector's dolphins was investigated using three environmental parameters: water depth, underwater clarity, and sea surface temperature. A resource selection function was applied to discern used from unused sites. The dolphins were found significantly more often in warm, shallow, and murky waters. The temperature effect, however, was regarded to be an artefact, as dolphins were suspected to have left the nearshore habitat in winter moving farther offshore in similar surface temperatures.
In spring and early summer, Hector's dolphin exploited a short-term food source preying on small migrating fish on the surface. White-fronted terns (Sterna striata) associated with dolphins profiting from the dolphins' foraging.
Summer home ranges of 32 adult individuals with more than ten sightings around Banks Peninsula between 1985 and 1997 were about 10-60 km long alongshore. Individuals showed high site-fidelity for up to twelve consecutive summers but were never resighted in these near-shore home ranges in winter. The general decrease in nearshore densities in winter was interpreted as a consequence of home ranges being shifted offshore in winter. The complete lack of observed movements over 60-600 km within the study period made any alongshore migration unlikely. The inshore-offshore movement was confirmed within Akaroa Harbour, Banks Peninsula, between November and April.
A Monte Carlo randomisation test was developed on data sets from two populations (Banks Peninsula and Porpoise Bay). Association patterns were described for dyads in three other populations (Kaikoura, Moeraki, and Jackson Bay). Most associations were weak and did not differ significantly from randomness. This confirmed earlier findings that Hector's dolphin have a "fission-fusion society".
A combined stereo-photogrammetry and underwater-video system was developed and used at close range to measure the length (age) and determine the gender of bow-riding individuals. The system could be used only in calm sea conditions and hence provided limited data. Gompertz growth functions were computed for males and females from a small set of measurements from dead individuals. The majority of bow-riding individuals proved to be immature or subadult.
Mark-recapture and a Cormack-Jolly-Seber model were used to estimate the mean annual adult survival rates for Hector's dolphin in the same four areas to be 70-100%. Mean calving interval and calf mortality were estimated to be 2.2 years and at least 36% within the first six month using data from all study areas.