Microtopoclimatic Effects on a Climate-Sensitive Habitat Specialist, the American Pika (Ochotona princeps)

There has been increasing acknowledgement that refugia at different scales facilitate the long-term survival of species and populations through climate oscillations. Species distributions and persistence are already affected by current climate change, and many taxa will become more spatially limited and less connected under further warming scenarios. Identifying likely microrefugia will improve our predictions of how species, communities, and ecosystems are likely to respond to climate change by providing a clearer understanding of likely demographic processes and connectivity. In this thesis, I considered suitable microhabitat in the face of current changing climates in the context of the persistence or development of microrefugia. The processes that drive microhabitat use by individuals likely also scale up to impact broader scale occupancy and connectivity patterns. Describing those fine-scale processes may therefore better predict how species will react to environmental change. To address these issues, we studied the ways in which fine-scale features of the terrain drive microhabitat use for the American pika (Ochotona princeps), a small lagomorph of western North America that has been cited as a likely climate and ecosystem change indicator species. The microtopoclimatic effect on habitat may drive patterns at the scales of individual habitat use, territory occupancy, patch occupancy, and regional population health. Though many studies have addressed some of the macroclimatic drivers of occupancy, we addressed fine scale processes and variation that likely interact with those broader factors. In the first chapter, we used radio telemetry to track pikas October-July in the Niwot Ridge Long Term Ecological Research site (NWT LTER) in Colorado’s Roosevelt National Forest and compared those data to records of summertime trapping success in the same area from 2008-2015. We used logistic regressions to test how terrain drives habitat use and how those effects vary seasonally. In the second chapter, we made predictions about suitable habitat across NWT using probability surfaces from Chapter 1 and supervised models made using remotely sensed data. We compared these predictions and discussed the importance of relevant habitat parameters and data at the appropriate scale to detect processes impacting species distribution models and larger predictions