Long‐ and short‐term temporal variability in habitat selection of a top predator

• Main Authors: Alessia Uboni, Douglas W. Smith, Julie S. Mao, Daniel R. Stahler, John A. Vucetich
• Format: Article
• Language: English
• Published: Wiley 2015-04-01
• Series: Ecosphere
• Subjects: Canis lupus; interannual variability; long-term series; predator-prey systems; resource selection; seasonal habitat selection
• Online Access: https://doi.org/10.1890/ES14-00419.1

Considerable theory explains the importance of understanding temporal variation in ecological processes. Nevertheless, long‐term variability in habitat selection is rarely assessed or even acknowledged. We explored temporal variability in the habitat selection of a top‐predator, the wolf (Canis lupus), at two time scales: interannual and seasonal variability. To do this, we developed resource utilization functions to relate wolf habitat selection to environmental variables in different years and seasons. We used radiotelemetry data collected from a wolf population in Yellowstone National Park during a 10‐year period (1998–2007) and added a Year variable in the models to account for interannual variation in the studied processes. We also used a three‐year data set (nested within the 10‐year data set) to incorporate additional variables in the models and test for differences in short‐ and long‐term patterns of habitat selection. Wolves exhibited seasonal variation in habitat selection with respect to distance from roads, elevation, openness, and habitat type. Habitat selection was considerably more complicated during the winter compared to summer, when wolves only selected habitat based on distance from roads. We detected clear patterns of habitat selection in the three‐year data set that could not be detected in the 10‐year data set, despite the longer data set had more statistical power for pattern detection. This observation is likely the result of the longer data set being comprised of several shorter‐term and countervailing patterns. This explanation is also consistent with having detected significant year effects in the 10‐year data set. Insomuch as habitat selection is important to conservation and management, this research is significant for demonstrating the different impressions that can be given by short‐term and long‐term studies. It may be common for short‐term data sets to suggest patterns of habitat selection that do not prevail over longer periods of time.