Off-campus UMass Amherst users: To download campus access dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.

Non-UMass Amherst users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Dissertations that have an embargo placed on them will not be available to anyone until the embargo expires.

Author ORCID Identifier



Open Access Dissertation

Document Type


Degree Name

Doctor of Philosophy (PhD)

Degree Program

Environmental Conservation

Year Degree Awarded


Month Degree Awarded


First Advisor

Kevin McGarigal

Second Advisor

Andrew Whiteley

Third Advisor

Daniel Sheldon

Fourth Advisor

Samuel Cushman

Subject Categories

Ecology and Evolutionary Biology | Population Biology | Terrestrial and Aquatic Ecology


The continued growth of human populations and associated development in many areas of the world is causing persistent fragmentation of natural habitats. In response, wildlife corridors are often promoted as essential for the conservation of wildlife species. Wildlife corridors allow for the movement of individuals between habitat patches and confer many benefits including the maintenance of metapopulations and metapopulation dynamics, the maintenance of seasonal migratory routes, genetic exchange, and the potential for individuals and populations to shift their ranges in response to climate change.

Wildlife corridors are modeled across a resistance-to-movement surface where resistance represents the willingness of an organism to cross a particular environment, the physiological cost of moving through a particular environment, or the reduction in survival for the organism moving through a particular environment. Resistance surfaces can be estimated using a wide variety of methods yet, to date, there has been no in-depth methodological comparison of these methods and their appropriateness for modeling connectivity.

My dissertation has two main objectives. The first was to determine the sensitivity of species-habitat models, resistance surfaces and corridors for pumas (Puma concolor) in southern California to six key factors: (1) data type used (point, step, or path data); (2) Statistical models employed; (3) Behavioral state of the individuals; (4) Spatial scale of analysis; (5) GPS collar acquisition interval; and (6) Thematic resolution and richness of the underlying geospatial layers. The second objective was to determine which combination of factors results in the most appropriate resistance surfaces for connectivity modeling.

I found that species-habitat models, resistance surfaces and corridors were extremely sensitive to all six of these factors – to the point where using one scale versus another or one data type versus another resulted in conflicting conclusions about habitat use and differences in the location of corridors. I recommend that, for modeling movement and corridors, path data be used in a context-dependent multi-scale modeling framework. I also recommend that many different geospatial layers at different thematic resolutions be examined to identify the most appropriate landscape definition for the species and study area of interest.