Consequences of habitat fragmentation: connectivity lies in the eye of the beholder

This dissertation was motivated by the problem of pattern and scale in ecology. All chapters present models that aim at predicting species’ responses to habitat fragmentation. Chapters differ mainly in the nature of the responses being investigated: spatial variation in abundance, or dispersal. In each chapter, I illustrate how current models can be modified to incorporate species’ perception of the landscape. Three sources of bias have been examined here: interspecific variation in (i) ecological neighborhood, (ii) ecological generalization, and (iii) in the response to regional processes. I have deliberately moved away from traditional single-scale, patch-based measures of landscape connectivity. Great emphasis has been placed on the anthropogenic aspect of the landscape, and on the role of the landscape matrix. Habitat fragmentation is a common feature of most (if not all) biodiversity hotspots. I hope the tools shown here can serve as general approaches to study how species are differentially affected by habitat fragmentation, and to ultimately understand how disturbed landscapes can “filter” natural communities.