ABSTRACT
Estimating population size has a long history, and is applicable in many different fields ranging from the size of “hidden” populations, such as drug users or disease carriers, to the estimation of trends in wildlife population size over a period of time. The topic was introduced in Section 2.2.4. We are here primarily concerned with the estimation of the size of a closed population, which is where there are no births, deaths, or migration beyond the study area in the population throughout the study period. We shall consider capture-recapture data in which animals are generally observed over a short time period, in order that the closure assumptions are satisfied. As usual, the capture-recapture data collected are generally presented in the form of the capture histories for each individual animal observed within the study (see Williams et al, 2002, Chapter 14, for example). The capture history of an individual is simply a sequence of 1’s and 0’s, corresponding to whether the individual was, or was not, observed at each capture event. Clearly, to estimate the population size we need to estimate the number of individuals with capture histories represented by a sequence of only 0’s, i.e. we wish to estimate the number of individuals that were not observed within the study period. We combine this with the number of individuals we do observe to obtain an estimate of the total population size.
