Theses and Dissertations at Montana State University (MSU)
Permanent URI for this collectionhttps://scholarworks.montana.edu/handle/1/733
Browse
1 results
Search Results
Item Validating alternative methods of modeling wildlife corridors using relocation data from migrating elk and dispersing wolverines(Montana State University - Bozeman, College of Letters & Science, 2012) Rainey, Meredith McClure; Chairperson, Graduate Committee: Andrew J. Hansen; Andrew J. Hansen was a co-author of the article, 'What is the current state of our ability to model wildlife corridors? An overview of the development, application, and validation of connectivity models' submitted to the journal 'Conservation biology' which is contained within this thesis.; Andrew J. Hansen was a co-author of the article, 'A test of the ability of connectivity models to predict migration movements using GPS collar data from migrating elk' submitted to the journal 'Journal of wildlife management' which is contained within this thesis.; Andrew J. Hansen and Robert M. Inman were co-authors of the article, 'A test of the ability of connectivity models to predict dispersal movements using relocation data from dispersing wolverines' submitted to the journal 'Journal of wildlife management' which is contained within this thesis.; Andrew J. Hansen was a co-author of the article, 'Assessing the sensitivity of connectivity models to model parameterization' submitted to the journal 'Landscape ecology' which is contained within this thesis.Habitat loss and fragmentation increasingly impede wildlife movements that are essential for the long-term persistence of populations. Wildlife corridors facilitating movement have become a key component of conservation planning as a result. Models are increasingly used to identify likely corridors, but predictions are rarely validated against empirical data, resulting in uncertainty in whether corridors function as intended. This study asks: 1) What is the current state of our ability to predict corridors?, 2) How well do the most common models predict wildlife movement?, 3) Which method performs best?, and 4) How sensitive are models to parameterization decisions? I addressed these questions by synthesizing the corridor modeling literature, modeling migration paths of elk and dispersal paths of wolverines using the two most common corridor models (cost-distance and circuit theory models), validating predicted corridors against relocation data, and analyzing model sensitivity to parameterization. An additional question, 5) What drives habitat selection during long-distance migration and dispersal movements?, is also explored through these analyses. Key findings include: 1) Corridor modeling has progressed enormously in recent years, but model validation has been rare, 2) Corridor models are capable of offering valuable information, but their utility depends on conservation objectives, 3) Relative model performance depends on the match between model assumptions and focal species movement ecology, 4) Both models have similar sensitivity to model parameters, but spatial patterns of sensitivity are unpredictable for circuit theory models, and 5) The drivers of habitat selection during long-distance movements may be similar to those of primary habitat selection, but further research is needed. These studies constitute the most rigorous, comprehensive effort to validate corridor model predictions to date. They are expected to offer guidance for conservation scientists and practitioners as to reasonable expectations for corridor model performance, the importance of selecting methods appropriate to species movement ecology, and which decisions in the modeling process matter most. This work also highlights the need for further collection of data suitable for model validation, broader comparison of corridor models across diverse species and landscapes, and continued study of the ecology of long-distance movements.