Get access

When is connectivity important? A case study of the spatial pattern of sudden oak death


A. M. Ellis, Dept of Geography and Earth Sciences, Univ. of North Carolina at Charlotte, 9201 University City Blvd., Charlotte, NC 28223, USA. Present address for AME: Dept of Entomology, Univ. of California-Davis, One Shields Ave., Davis, CA 95616-8584, USA. E-mail:


Although connectivity has been examined from many different angles and in many ecological disciplines, few studies have tested in which systems and under what conditions connectivity is important in determining ecological dynamics. Identifying general rules governing when connectivity is important is crucial not only for basic ecology, but also for our ability to manage natural systems, particularly as increasing fragmentation may change the degree to which connectivity influences ecological dynamics.

In this study, we used statistical regression, least-cost path analysis, and model selection techniques to test the relative importance of potential connectivity in determining the spatial pattern of sudden oak death, a tree disease that is killing millions of oak and tanoak trees along coastal forests of California and Oregon. We hypothesized that potential connectivity, in addition to environmental conditions, is important in determining the spatial distribution of sudden oak death, the importance of connectivity is more apparent when measured using biologically meaningful metrics that account for the effects of landscape structure on disease spread, and the relative importance of environmental variables and connectivity is approximately equal.

Results demonstrate that potential connectivity was important in determining the spatial pattern of sudden oak death, though it was relatively less important than environmental variables. Moreover, connectivity was important only when using biologically meaningful metrics as opposed to simple distance-based metrics that ignore landscape structure. These results demonstrate that connectivity can be important in systems not typically considered in connectivity studies – highlighting the importance of examining connectivity in a variety of different systems – and demonstrate that the manner in which connectivity is measured may govern our ability to detect its importance.