A Composite Graph Theoretic Approach to Modeling Landscape Connectivity for Wildlife Movement in Western Canada

Abstract

Connectivity among the resource patches that provide the wildlife with essential habitat is critical to their survival, but highways and other anthropogenic developments commonly impede wildlife movement. The purpose of this study was to identify suitable locations for highway crossing structures for wildlife movement in a fragmented landscape. Functional connectivity was modeled using human footprint data over a regional landscape in western Canada. A graph-theoretic approach was employed to identify corridors, link-age zones, and the locations where wildlife species cross the highways. A betweenness centrality model was used to compute the shortest path, current flow, and network flow of movements across various landscape lattices. The shortest path model identified a set of geodesic paths to connect the resource patches, the current flow model identified a number of movement zones around the resource patches, and the network flow model identified linkage zones in the network. Finally, a composite of the outputs was used to identify suitable locations for highway crossing for maintaining wildlife movement on the landscape.