Ecology of Pacific white-sided dolphins (Lagenorhynchus obliquidens) in the coastal waters of British Columbia, Canada
MetadataShow full item record
The ecology of Pacific white-sided dolphins (Lagenorhynchus obliquidens) in the Broughton Archipelago, British Columbia (BC), Canada was explored through photo-identification, mark- recapture, acoustics, and sociality studies. New population parameters were estimated from photo-ID data for the first time in this species. Abundance was highly variable, ranging from 546 (95% CI: 293-1,018) to 2,889 (95% CI: 1,424-5,863), after accounting for the proportion (0.57; 95% CI: 0.55 - 0.60) of marked dolphins. A “match uncertainty” analysis showed that less strict matching criteria caused negative bias in abundance estimates and an apparent improvement in precision. Estimates of survival rate ranged from 0.907 (SE=0.03) to 0.989 (SE= 0.066). Robust design analyses revealed random temporary emigration movement at 0.14 (SE=0.318) annually and no movement seasonally. The study revealed new evidence for philopatry and sociality: some individuals were resighted over 19-year periods, and associated pairs more than a decade apart. Evidence was found for a high degree of sociality. The mean proportion of calves was estimated as 0.0597 (SE=0.0083, 95% CI: 0.045-0.079) per capita, translating to an average probability of pregnancy in adult females of 0.238 (95% CI: 0.180- 0.316) and an average interbirth interval of 4.2 years. Approximately 3.9% of dolphins bore injuries from killer whales, but only 0.5% showed evidence of interactions with fishing gear or propellers. Acoustic evidence for population structure was equivocal, but warrants additional, targeted research. Population viability analysis predicted an average rate of annual decline of - 0.122 (95% CI: -0.143 to -0.101), given a range of input values in a sensitivity test, over the next 50 years.
Thesis, PhD Doctor of Philosophy
Items in the St Andrews Research Repository are protected by copyright, with all rights reserved, unless otherwise indicated.