A multistate mark recapture analysis to estimate reproductive rate in the Steller sea lion (Eumetopias jubatus), an endangered species

Thumbnail Image



Journal Title

Journal ISSN

Volume Title


Montana State University - Bozeman, College of Letters & Science


The Steller sea lion is an endangered species whose reproductive rate estimates need to be updated. The species is divided into two populations: the endangered western population has declined over 80% from historical levels, while the threatened eastern population has been increasing at approximately 3% for the past three decades. The statistically most compelling reproductive rate estimates for this species are based on now out of date population dynamics, and hence are not applicable to current concerns. Extensive recent branding and resighting efforts by the Alaska Department of Fish and Game in Southeast Alaska make possible an updated estimation of eastern population Steller sea lion reproductive rates. However, the complexity of these data required a different statistical approach than is typically used to estimate reproduction in marked and resighted animals. I developed a novel statistical analysis, based upon a multistate mark recapture likelihood function, specifically to analyze the Southeast Alaska Steller sea lion data. The likelihood function estimates a reproductive rate when only adult females (not pups) are marked, female sightability is correlated with reproductive status, state classification uncertainty is present and the population is open to births during many of the resighting intervals. I apply this analysis to the Southeast Alaska Steller sea lion data and estimate a reproductive rate of 0.66 (0.55, 0.77). Not only does this provide a reproductive rate estimate for the eastern population, which is important for monitoring its health, but it also provides a basis for comparison to the endangered western population. Furthermore, the Alaska Department of Fish and Game continues to have an active branding and resighting program. The methods developed here can be applied to future data collected in either population.




Copyright (c) 2002-2022, LYRASIS. All rights reserved.