Document: BRE-3-65-12

Different metrics of connectivity respond differently to landscape structure.

GOODWIN, B.J.* ¹ and L.FAHRIG ²

Institute of Ecosystem Studies, Millbrook, NY 12545, USA ¹
Carleton University, Ottawa, Ontario, Canada ²

Abstract:
Landscape connectivity can be thought of as the influence of landscape structure on the ability of an organism to move through the landscape. To be able to use this idea it is necessary to have some metric of landscape connectivity (ability of an organism to move through a landscape) yet currently there is no common metric of connectivity. Using an individual-based, spatially explicit simulation model we investigated the response of six potential metrics of connectivity (mean transition probability, mean visits per individual, and mean immigration rates at both the patch and cell scales) to varying landscape structure (amount of habitat, number of habitat patches, size distribution of habitat patches, edge to area ratio of habitat, mean distance between habitat patches, variance of distance between habitat patches). The six metrics of connectivity were remarkably different in terms of the aspects of landscape structure they responded to (e.g., the mean transition probability for patches was primarily driven by habitat spacing while the mean number of new patches visited was primarily driven by the number of habitat patches). Furthermore, the relationships between the measures of connectivity were complex and dependent upon the state of the landscape. These results suggest that some metrics of landscape connectivity will respond to some aspects of landscape structure while other metrics will respond to other aspects of landscape structure. This will complicate both the applications of the concept of connectivity and comparisons between studies using different connectivity metrics.

Keywords: connectivity, movement, landscape structure, simulation model

This abstract is being presented at: 12:00 PM in session:
Oral Session #58: Landscape Ecology.