A spatially explicit habitat selection model incorporating home range behaviour
JONATHAN R. RHODES,1,2,4 CLIVE A. MCALPINE,1'2 DANIEL LUNNEY,3 AND HUGH P. POSSINGHAM2
1School of Geography, Planning and Architecture, University of Queensland, Brisbane, QLD 4072, Australia
2The Ecology Centre, University of Queensland, Brisbane, QLD 4072, Australia
3New South Wales Department of Environment and Conservation, P.O. Box 1967, Hurstville, NSW 2220, Australia
Understanding habitat selection is of primary interest in theoretical and applied ecology. One approach is to infer habitat selection processes from differences in population densities between habitats using methods such as isodar and isoleg analysis. Another approach is to directly observe the movements of individuals. However, habitat selection models based on movement data often fail to adequately incorporate spatial processes. This is problematic if the probability of selecting a particular habitat is dependent upon its spatial context. This would occur, for example, where organisms exhibit home range behavior and the choice of habitat is dependent on its location relative to the home range. In this paper we present a spatially explicit habitat selection model for movement data that incorporates home range behavior as a spatial process. Our approach extends a previous model by formulating the probability of selecting a habitat as a function of its distance from the animal's current location and home range center. We demonstrate that these enhancements lead to more parsimonious models when applied to a koala radio tracking data set from eastern Australia. This approach could also be applied to modeling other spatial habitat selection processes, leading to more biologically meaningful models for a range of species and applications.