HABITAT-SPECIFIC MOVEMENT PARAMETERS ESTIMATED USING MARK–RECAPTURE DATA AND A DIFFUSION MODEL

Abstract

I describe a diffusion model aimed at the quantitative analysis of movement in heterogeneous landscapes. The model is based on classifying a landscape into a number of habitat types, which are assumed to differ from each other in terms of the movement behavior of the focal species. In addition to habitat-specific diffusion and mortality coefficients, the model accounts for edge-mediated behavior, meaning biased behavior close to boundaries between the habitat types. I illustrate the model with three examples. In the first example, I examine how the strength of edge-mediated behavior and the size of a habitat patch affect the probability that an individual will immigrate to a patch, the probability that an individual will emigrate from a patch, and the time that an individual is expected to spend in a patch. In the second example, I study how a dispersal corridor affects the probability that an individual will move through a landscape. In the third example, I estimate the movement parameters for a species of butterfly from mark–recapture data. In the butterfly example, I classify the landscape into habitat patches, other open areas, and forests. Edge-mediated behavior is found to have a highly significant effect on the general dispersal pattern of the butterfly: the model predicts that the density of butterflies inside habitat patches is >100 times the density of butterflies in adjacent areas.