Catch Curve and Capture Recapture Models: A Bayesian Combined Approach
Year: 2008
Type: article
Abstract: When studying animal populations, one demographic parameter of interest is the annual survival rate.Methods for estimating survival rates of animal populations fall into two general categories: methods based on marked and on non-marked animals.Catch curve analysis falls into the latter category, is based on strong assumptions about population dynamics, and is much less expensive than marked animal methods.Capture-recapture methods, on the other hand, use marked animals and require strong assumptions about individual capture and survival probabilities.Chapman and Robson developed methods to analyze catch curve data in the sixties using geometric and multinomial distributions to obtain an estimate of the survival rate.The assumptions for the Chapman-Robson survival estimate are: a random sample of the age structure of the living population is obtained, that is, equal catchability for all fully vulnerable age classes, and the population has a stable age distribution, meaning that survival and fecundity rates have remained constant for a long period of time.A stable age distribution where the population size also remains constant is called stationary.Rarely, it is possible to collect ages-at-death data, a random sample from the natural deaths that occur in a population between two time periods.Udevitz and Ballachey augmented catch curve data with ages-at-death data in the late nineties, resulting in the ability to relax either the assumption of a stable age distribution or known population growth rate.Neither type of data involves marking animals.Marked animal methods include band recovery, telemetry, and capture-recapture data.We focus specifically on capture-recapture data, and the Cormack-Jolly-Seber (CJS) open population model.It is important to note that the CJS model estimate of survival is "apparent survival."In capture-recapture methods, mortality is confounded with emigration.With telemetered animals, the two processes can be separated.Tag-return models where
Author: Emily H. Griffith
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W164905337
