Standard release–recapture analysis, using Cormack–Jolly–Seber (CJS) models to estimate survival probabilities between hydroelectric facilities for juvenile Snake River fall chinook salmon (Oncorhynchus tschawytscha), ignore the possibility of individual fish overwintering and completing their migration in the year following tagging. These models do not utilize available capture history data from this second year and produce negatively biased estimates of survival probabilities. A new multinomial likelihood model was developed that results in biologically relevant, unbiased estimates of joint survival and overwintering probabilities using the full two years of capture history data. This model was applied to 1995 Snake River fall chinook hatchery releases to estimate the survival probability between a release site at Asotin, Washington (U.S.) and Lower Granite Dam. In the example presented here, overwintering is not a common physiological response and thus the use of CJS models (estimate of the survival probability = .4235; SE = .0162) did not result in an appreciably lower estimate for the survival probability than that calculated considering overwintering (estimate of the survival probability = .4360; SE = .0164) and obtained using the new multinomial model. However, analysis of releases in subsequent years may show greater impacts of overwintering on the estimates of survival probabilities.

Key Words
Estimable parameters; Fall chinook salmon; Overwintering; Release–recapture; Sufficient statistics; Survival estimation.

Alan B. Lowther is Research Assistant, and John R. Skalski is Professor of Biological Statistics, School of Fisheries, Box 358218, University of Washington, Seattle, WA 98195-8218 (E-mail: alan@cqs.washington.edu and jrs@fish.washington.edu).