Methods for measuring connectivity often involve the use of a cost-surface layer, which provides a map of a region with quantitative measures of how difficult it is for an individual to move from point to point. To begin, a practitioner can use GIS tools to perform least-cost modeling, which estimates the shortest path between two points with maximum efficiency for a moving individual. The actual measure of connectivity can be estimated using either the length of the least-cost path, or the overall accumulated-cost, which takes into account both length and the underlying ecological cost incurred to travel it. While accumulated-cost is a more popular measure, there are still numerous studies using least-cost path length to determine connectivity (Zeller et al. 2012). Etherington and Holland use 1,000 simulated fractal landscapes to explore whether using the length of a least-cost path is as effective in measuring connectivity as using the accumulated-cost of a least-cost path, and come up with a resounding no. They show that least-cost path length is often no more effective than using straight Euclidean distance, both of which ignore the ecological costs of movement for an individual. This then misrepresents the degree of connectivity between two points, and provides unrealistic information about the landscape. Their strong conclusion is that using least-cost path length to assess connectivity should be restricted to use only as a visual tool, and replaced with accumulated-cost measure of the least-cost path, which is much more useful in capturing the cost component of the path and therefore better for measuring connectivity.
Etherington, T. R., and E. P. Holland. 2013. Least-cost path length versus accumulated-cost as connectivity measures. Landscape Ecology 28 (7):1223-1229.
Zeller, K. A., K. McGarigal, and A. R. Whiteley. 2012. Estimating landscape resistance to movement: a review. Landscape Ecology 27 (6):777–797.