Introducing a `stochastic movement simulator' for estimating habitat connectivity

Authored by Justin MJ Travis, Stephen C F Palmer, Aurelie Coulon

Date Published: 2011

DOI: 10.1111/j.2041-210x.2010.00073.x

Sponsors: European Union French Ministries

Platforms: C++

Model Documentation: Other Narrative Flow charts

Model Code URLs: Model code not found

Abstract

P>1. Estimating and improving landscape connectivity has become a key topic in conservation biology. While a range of connectivity indices exist and are widely used to inform spatial planning, they have potentially serious limitations. 2. We introduce a new method for modelling animals moving between habitat patches across a heterogeneous matrix. Our approach integrates features of least cost path analysis with stochastic movement modelling. Importantly, it relaxes assumptions that are implicit in the least cost path algorithm: our method does not assume omniscience nor does it assume that an individual has a planned destination when it leaves a habitat patch. The algorithm incorporates resistance values for matrix elements and parameters that relate to both perceptual range and to the degree of correlation in movement. By simulating sets of movements for individuals emigrating from habitat patches, relative connectivities between habitat patches are estimated. 3. Using an already published stylised landscape, we demonstrate that the connectivities estimated using our new method can differ considerably from those provided by structural methods and by least cost path analysis. Further, our results emphasise the sensitivity of the relative connectivities to an organism's perceptual range and the degree of correlation between movement steps. 4. We believe that using stochastic movement modelling can improve estimates of connectivity and also provide a method for determining how robust the indices derived from simpler methods are likely to be for different types of organisms.
Tags
Evolution behavior Landscape connectivity Dispersal Model Circuit-theory Gene flow Heterogeneous landscapes Random-walks Animal movements