Current research on landscape connectivity has used graph theory techniques to analyze and value discrete habitat patches within a larger network. While much of this research has used Euclidean distances between habitat patches, a growing number of studies have begun to use least-cost-path analysis to analyze functional distances between patches. An important step in this process is the creation of a cost surface, which assigns costs across the landscape. The values assigned are often based on the researcher’s best professional judgment; however the effects of the ratios of costs are not fully understood and therefore the effects that these assumptions have on model results is not known. We have compared the predicted functional distances between habitat patches as well as predicted values of each patch in a landscape context using various scales in our cost surface to better understand the effect these assumptions have on landscape connectivity models.
Results/Conclusions
A comparison of various cost surfaces used in determining functional distances between patches found that the assumptions used in creating the surface have the potential to alter model results. The changes in results vary by patch and by metric being measured. When analyzing landscape connectivity based on functional distances, it is important to consider the effects that the assumptions made in creating a cost path may play on model results.
