Developing a census method based on sight counts to estimate rabbit (Oryctolagus cuniculus) numbers
A rabbit-census method, based on systematic counts conducted at night or at dawn and dusk, was developed and validated in terms of estimating the total number of rabbits present in a given area. Initially, models were developed under semi-natural conditions to describe the relationships between the numbers of rabbits counted and population size. Confidence limits were also calculated. The models were developed by comparing rabbit counts with the actual number of rabbits present, from a known population of animals. Only spotlight counts at night were considered reliable enough to estimate rabbit population size. During the autumn and winter months these represented ~60% of the population present. The model was subsequently validated, in two different exercises, following a series of field trials conducted under a variety of conditions on commercial farms. Initially, population estimates derived from the model were compared with those calculated using an alternative census technique. Population estimates, using the two techniques, were very similar at nine of the ten study sites. A second validation exercise was also conducted whereby the number of rabbits removed at each of the sites was compared with the difference between the mean pre- and post-removal spotlight counts. The results further supported the proposition that spotlight counts represent ~60% of the population present, with the difference between the two mean spotlight counts representing 61.2% (± 11.0, s.d.) of the number of rabbits removed. The census method therefore shows considerable promise as a means to estimate rabbit numbers under a range of agricultural conditions and therefore has the potential to predict accurately the economic costs of rabbit damage and also to gauge the effectiveness of various methods of rabbit control. W R02014 Sight count s t o es ti ma te ra bbi t number s D. W. P oo le et al .