Relative estimates of small-mammal populations in relation to the spatial pattern of trappability.

The principles of evaluating relative estimates is described on the basis of a study of Clethrionomys glareolus (Schreber) and Apodemus flavicollis (Melchior). The frequency distribution of the trappability of the two species conforms to a Poisson distribution with added zeros. These surplus zeros are considered to be inversely proportional to the habitat penetration. If unrestricted trapping capacity (or a good approximation of it) is applied at each trap site, the mean of the catches at the trap sites constitutes a rectified population density index. ‘Rectified’ means that a doubled index value always implies a doubled population density. The properties of the Poisson distribution makes it possible, however, to sketch a theoretical model of relative estimates in which only one single-catching trap is set at each trap site. The known relation between the mean and the zero class of the Poisson distribution (a non-linear asymptotic regression) may be used to calculate a rectified population density index from such presence/absence data. In this case the surplus zeros must be formulated as a population parameter. [ABSTRACT FROM AUTHOR]