CORRECTION OF BIAS DUE TO HETEROGENEOUS CAPTURE PROBABILITY IN CAPTURE-RECAPTURE STUDIES OF OPEN POPULATIONS

Heterogeneity of individual capture probabilities (pi) is a commonly acknowledged cause of bias in capture-recapture studies of open populations. We propose and evaluate an extension of the indirect method of Carothers (1979, Journal of Animal Ecology 48, 863-869). Carothers' method identifies by simulation and inverse prediction the parametric survival rate, the mean capture probability (E(p)), and CV(p) that could have given rise to an observed set of capture histories and biased capture-recapture estimates. We also extend Carothers' approach to estimate population size and evaluate it in relation to field data from a study of brushtail possums in New Zealand. If heterogeneity is the only source of bias and it is adequately modelled by a beta distribution, the method provides unbiased estimates of average population size, average survival rate, E(p), and CV(p) under the Cormack-Jolly-Seber model. Although the method assumes a constant population size, our results suggest that it is robust to temporal variation.