Estimating Animal Abundance with Capture Frequency Data

Abstract

I describe a new method for estimating the number of animals in a closed population with capture-recapture data under a heterogeneity model. The model is applied to capture frequency records of striped skunk (Mephitis mephitis), eastern chipmunk (Tamias striatus), eastern cottontail (Sylvilagus floridanus), and taxicab populations. I also report the results of a Monte Carlo simulation to assess the relative merits of the proposed moment estimator and the jackknife estimator. If many individuals are caught more than twice, the jackknife estimator is superior to the proposed moment estimator. However, when the mean capture probability is small, so that most captured animals are caught only 1-2 times in the samples, the moment estimator is usually less biased than the jackknife estimator. The mean coverage probability of the confidence interval associated with the moment estimator is also closer to the nominal level than that of the jackknife interval.