Language selection


Research Document - 2012/041

Population viability analysis for northern fur seals (Callorhinus ursinus) in Canada

By P.F. Olesiuk


A population viability analysis (PVA) was conducted for northern fur seals to assess risk of extirpation in Canada. The PVA is based on the diffusion model developed by Dennis et al. (1991) that projects populations along a trajectory based on recent trends with drift due to natural variability. The North Pacific fur seal population was considered a meta-population, with each breeding site representing a subpopulation (the smallest breeding site at San Miguel Island was excluded from the analysis). Recent trends for each subpopulation were estimated from pup counts using density independent or dependent models, and unexplained variation was assumed to represent natural variability (demographic stochasticity or environmental variability) after adjusting for measurement error. Models were fitted to the most recent 10-year, 30-year and 50-year time-series of pup counts. Monte Carlo techniques were used to project each subpopulation 200 years into the future, and risk of quasi-extinction estimated for the first 100 years. The threshold for quasi-extinction was set to an effective population size of 1,000 animals required to sustain genetic diversity, which for a polygamous species like the northern fur seal represents a total subpopulation size of 8,000 seals. I adopted a precautionary approach, and considered subpopulations to be endangered if there was a 1% chance of falling to quasi-extinction levels within the next 100 years, and threatened if there was a 20% chance of becoming endangered within the next 20 years. Tagging data were analyzed to determine the contribution of each subpopulation to abundance in Canada based on the propensity of animals from each breeding site to migrate along the west coast of North America. The threshold for quasi-extinction of the population and segment migrating through Canadian waters was set to the minimum viable population (MVP) size, which was 7,000 adults or 10,000 seals. The PVA indicated that all subpopulations are presently secure with little risk (0.00%-0.42%) of extirpation within the next century. The Pribilof Island subpopulation, the largest breeding site and source of the majority of animals migrating through Canadian waters, has been declining for the past 50 years. However, due to its large size, it is not at risk of extirpation within the next 100 years, but the probability of quasi-extinction will escalate if the declines persist beyond the next century. As the Pribilof Island subpopulation declines, animals from other breeding sites that are have been stable or increasing will play an increasingly important role in maintaining a MVP in Canada, and there appears to be no discernible risk of extirpation from Canadian waters. Currently, it is estimated that 75% of seals migrating through Canadian waters originate from the Pribilof Islands, compared with 97% when the pelagic research collections made in 1958-74. An estimated 15% of seals in Canada originate from the rapidly growing breeding site established on Bogoslof Island in 1980, and 8% from the Commander Islands where numbers have increased over the last 50 years but are now stable, and less than 1% from the growing but distant breeding sites in the Kuril Islands and Tuleny (Tyuleniy) Island. The assessment indicates that risk of extinction is a function of population size, trend and degree of natural variability, and that PVA provides a means of integrating such information. Population trend in itself is not a meaningful criterion for assessing risk, particularly for very large populations like the northern fur seal. Indeed, the northern fur is expected to continue to be the most abundant and widely distributed pinniped in the North Pacific.

Accessibility Notice

This document is available in PDF format. If the document is not accessible to you, please contact the Secretariat to obtain another appropriate format, such as regular print, large print, Braille or audio version.

Date modified: