Recolonizing carnivores: Is cougar predation behaviorally mediated by bears? Cougar Killsite Data

Conservation and management efforts have resulted in population increases and range expansions for some apex predators, potentially changing trophic cascades and predatory behavior. Changes in sympatric carnivore and dominant scavenger populations provide opportunities to assess how carnivores affect one another. Cougars (Puma concolor) were the apex predator in the Great Basin of Nevada, USA, for over 80 years. Black bears (Ursus americanus) have recently recolonized the Great Basin and are known to heavily scavenge on cougar kills; however, competitive interactions between the two species in the Great Basin have yet to be examined. We investigated kill sites of 31 cougars between 2009 and 2017 across a range of bear densities to evaluate the impacts of sympatric, recolonizing bears on cougar foraging behavior. We modeled the variation in feeding bout duration (number of nights spent feeding on a prey item) and the proportion of primary prey, mule deer (Odocoileus hemionus), in cougar diets using mixed-effects models. We found that feeding bout duration was driven primarily by the size of the prey item being consumed, local bear density, and the presence of dependent kittens. The proportion of mule deer in cougar diet across all study areas declined over time, was lower for male cougars, increased with the presence of dependent kittens, and increased with higher bear densities. In sites with feral horses (Equus ferus), a novel large prey, cougar consumption of feral horses increased over time. Our results suggest that higher bear densities over time may reduce cougar feeding bout durations and may influence the prey selection trade-off for cougars when alternative, but more dangerous, large prey are available. Shifts in foraging behavior in multi-carnivore systems can have cascading effects on prey items. This study highlights the importance of measuring the impacts of sympatric apex predators and dominant scavengers on a shared resource base and, thus, provides a foundation for monitoring dynamic multi-predator/scavenger systems.

Methods
Field Methods:

We conducted kill-site investigations at clusters of cougar GPS points which were identified using the algorithm developed by Knopff et al. (2009), as likely to contain a cougar kill. We established the criteria for a kill site to be ≥2 GPS points within 200m, including at least one location obtained overnight. We prioritized visiting all clusters with a 25% or greater probability of containing a kill and then searched as many clusters with a probability <25% as the field crew could successfully visit. For each prey item located at each kill- site location, we identified the species, as well as sex and age where possible. The age of ungulates was determined using tooth eruption and wear. We documented signs of other predators or scavengers at the carcass location.