|
Abstract:
|
Mark-recapture studies are a familiar tool for linking environmental covariates to demographic rates such as population survival. Yet incorporating the effect of time-varying covariates over longer or variable length exposure histories has remained challenging. In particular, space-for-time Cormack-Jolly-Seber (CJS) models, commonly used to estimate survival and travel time of migrating anadromous fishes, have been unable to incorporate the effects of time-varying individual covariates when the time between detections is long or varies considerably within the population. We present results from a CJS, complete data likelihood analysis of mark-recapture data from juvenile Chinook salmon migrating out of the Sacramento River in California. Imputation of unobserved travel times allowed for estimating survival and travel time via a proportional hazards model. The effect of time-varying covariates such as river discharge could then be integrated over each individual’s exposure history. We believe this novel application could provide a valuable tool for insights into the relationship between environment, exposure history, and demographic vital rates in many other populations of interest.
|
